V

SCIENCE
LESSONS
SCIENCE
LESSONS
What the Business of
Biotech Taught Me^bout
A.; .0-1
S OF HEALTH
Management V/H •&P„:

C- r* •>

C., ,

Gordon Binder
Philip Bashe

HARVARD BUSINESS PRESS

BOSTON, MASSACHUSETTS
Copyright 2008 Gordon Binder
All rights reserved
Printed in the United States of America
12 11 10 09 08 5 432 1

No part of this publication may be reproduced, stored in or introduced into a re¬

trieval system, or transmitted, in any form, or by any means (electronic, mechanical,
photocopying, recording, or otherwise), without the prior permission of the pub¬
lisher. Requests for permission should be directed to permissions@[Link],
or mailed to Permissions, Harvard Business School Publishing, 60 Harvard Way,
Boston, Massachusetts 02163.

Library of Congress Cataloging-in-Publication Data

Binder, Gordon M., 1935-
Science lessons : what the business of biotech taught me about
management / Gordon Binder, Philip Bashe.
p.; cm.
ISBN 978-1-59139-861-5 (alk. paper)
1. Binder, Gordon M., 1935— 2. Amgen Inc.—Management. 3.
Biotechnology industries—United States—Management. 4. Chief executive
officers—United States—Biography. I. Bashe, Philip. [Link].
[DNLM: 1. Binder, Gordon M., 1935— 2. Amgen Inc. 3.
Biotechnology—economics—Personal Narratives. 4.
Biotechnology—history—Personal Narratives. 5. History, 20th
Century—Personal Narratives. 6. Industry—Personal Narratives.
HD9999.B442 B612s 2008]
HD9999.B444A443 2008
338.7'6151092—dc22
[B] 2007038583

The paper used in this publication meets the minimum requirements of the Ameri¬
can National Standard for Information Sciences—Permanence of Paper for Printed
Library Materials, ANSI Z39.48-1992.
Dedicated to my wife, Adele

—Gordon Binder
Contents

Introduction 1

1 The Start of a Start-Up 17

2 Amgen Goes Public 39

3 The Business of Science Meets

the Science of Business 57

4 From IPO to EPO in Four Months 75

5 EPO on Trial 89

6 Partnerships Made in Heaven—

and That Other Place 107

7 Red Hot Summer 139

8 Nuclear Winter 153

9 How Amgen Built a Winning Team 177

viii Contents

10 How Amgen Kept Employees Committed 195

V
11 The Value of Ethics in Business 249

12 My Failed Retirement 269

Acknowledgments 219
Index 281
About the Authors 293
 Introduction

here’s what not to say when a senior partner of San Francisco’s

most prominent law firm calls to recommend that you interview for
the chief financial officer position at a fledgling outfit called Amgen.
“Amgen? What’s Amgen?”
“It’s a biotechnology company We represent them.”
I nearly blurted, “What’s biotechnology?”—not an unreasonable
question in [Link] biotech industry, barely a decade old, had yet to
market a single drug. My twenty years in the business world had been
spent almost exclusively at major corporations like Litton Industries
and Ford Motor Company. The comparatively small System Devel¬
opment Corporation (SDC), where I served as vice president of fi¬
nance for ten years, supported a workforce of roughly four thousand.
Now, at age forty-six, did I really want to get involved in a start¬
up operation with only a few dozen employees and no products even
remotely on the horizon? Probably not.
Except that I was newly unemployed, again, having had a com¬
pany sold out from under me for the second time in twelve months.
2 Science Lessons

In 1980 the board of directors of SDC’s principal owner, a founda¬

tion, decided to unload the firm, which designed information tech¬
nology and computer systems primarily for the U.S. military. The
board’s plan was to donate the proceeds to various universities and
other worthy institutions and then shut down the foundation. As
CFO, it fell to me to find a buyer and arrange the sale—rather like
being asked to conduct your own funeral.
I wound up negotiating a deal with Burroughs Corporation for
$100 million cash, one-third more than the board had projected. But
I had no interest in staying and being downscaled to a Burroughs di¬
vision controller. Nor were my wife, Adele, and I eager to uproot our
two young sons and move from Los Angeles to Detroit, the business
machine giant’s headquarters. We’d lived there during the first five
years of our marriage, when I was a unit controller at Ford. We had
nothing against Michigan, but it wasn’t for us, a couple of native
[Link], by closing the transaction with Burroughs—
handshakes all around—I was in effect laying myself off.
Then United Geophysical Corporation (UGC), the world’s lead¬
ing oil-exploration company, promptly hired me. Its parent com¬
pany, Bendix, had recently tried to sell the division but couldn’t find
anyone willing to meet its price. Then investment firm Lehman
Brothers stepped in to propose taking UGC public. Bendix, a manu¬
facturer of automotive and aircraft parts, agreed.
There was no better news as far as I was concerned. Finally I
would have the opportunity to realize a goal I’d set when I was a
graduate student at Flarvard Business School: to be chief financial of¬
ficer of a company listed on the New York Stock Exchange. In short
order, a board of directors was assembled and a preliminary draft of
the public-offering prospectus was written.
However, news of our impending initial public offering (IPO)
rekindled interest from one of the prospective suitors that had come
courting only months before. Suddenly the company doubled its
 Introduction 3

bid, matching Lehman’s IPO price. You can probably guess the rest:
Bendix took the money, and once more I found myself out of work.
But not for long. Unbeknown to me, Ed Huddleson, of Amgen’s
primary law firm, heard about what had happened. Huddleson, a
former trustee of the foundation that had owned System Develop¬
ment Corporation, apparently had been impressed by my handling
of the sale to Burroughs. He took it upon himself to play match¬
maker by pitching me to Amgen’s chief executive officer, George
Rathmann, who was seeking to hire his first CFO. I knew next to
nothing about the young biotech firm, but I greatly respected Hud-
dleson’s opinion. If he thought that Amgen and I would prove a
comfortable fit, that was good enough for me.
I drove out for an interview at company headquarters in Thousand
Oaks, California, about an hour west of Los Angeles. Nestled in the
scenic Conejo Valley, the developing city had to be one of the prettiest
in the state. It still is, its sparkling wide streets lined with stately oak
trees and rimmed by rolling foothills and rugged mountains.
In contrast, Amgen’s offices on Oak Terrace Lane were, to be
kind, on the plain side: eight thousand square feet of space in a lonely
single-story concrete building set in a mostly barren industrial park
just off the Ventura Freeway. There were three other tenants, one of
them the Continental Ministries traveling evangelical choir. I no¬
ticed their road-weary bus convalescing out front.
George Rathmann, a PhD in physical chemistry, exuded almost a
ministerial zeal as he talked animatedly about biotechnology’s vast
potential to create revolutionary new drugs. I’d read up on the na¬
scent field in preparation for the interview, so I was already intrigued.
But when George offered me the position following a second meet¬
ing, I explained that I needed to discuss the matter with my wife be¬
fore I could give him an answer.
A stall tactic for negotiating a higher salary? Only in part. Adele
and I had been married since almost the beginning of my career, and
4 Science Lessons

I had relied heavily on her advice many times. This was a decision
that needed to be made jointly. y
The two of us talked it over after dinner. The salary represented a
sizable cut from what I had been earning. We were comfortable
enough, though, to ride out the reduction in pay. Much of the com¬
pensation would take the form of stock options, a standard practice for
a promising but risky new venture. Those shares might one day turn
into the proverbial pot of gold at the end of the rainbow. Or we could
be left clutching a worthless piece of paper—as countless starry-eyed
victims of the dot-com collapse of the 2000s know well.
After we volleyed the pros and cons back and forth, Adele said,
“Gordon, I think you should do it. If it goes belly up, you can always
look for a new job.”
“And who knows?” she added. “Maybe it will turn into some¬
thing really big.”

REALLY, REALLY BIG

That it did, by any standard, albeit at the glacial pace typical of

biotech success stories. Nine years—no, that’s not a typo—passed be¬
fore Amgen brought its first therapeutic product to market: epoetin
alfa, sold under the brand name Epogen.
EPO, as it’s called for short, is a genetically engineered version of
the hormone erythropoietin (pronounced ee-rith-row-poy-i-tin). I ex¬
plain genetic engineering in depth later. (I know: you signed on for a
book about business practices, not [Link] can relax. These con¬
cepts aren’t difficult to understand, even if you dozed through high
school biology.) But for now, genetic engineering is a process for cre¬
ating and mass-producing copies of natural substances that activate or
suppress certain biological functions. Hormones, for instance, travel
the bloodstream, bearing chemical messages for various organs.
A healthy person’s kidneys release erythropoietin into the circula¬
tion for delivery to the spongy bone marrow inside our long bones.
 Introduction 5

Marrow is the factory for the three primary components of blood:

red cells, white cells, and platelet cells. Erythropoietin stimulates the
production of red cells, which carry out the vital task of transporting
oxygen from the lungs to every cell. Your body depends on oxygen
for energy.
More than four hundred thousand men, women, and children in
the United States suffer from irreversible kidney failure, or end-stage
renal disease (ESRD). Their damaged kidneys cannot manufacture
enough erythropoietin. Until the introduction of Epogen in 1989,
ESPJD patients regularly experienced debilitating exhaustion from
severe chronic anemia: a deficiency of red cells and consequently of
[Link] condition often closed the door on their ability to work
or to be as active as they’d once been.
Epoetin alfa returns anemics’ red-cell levels to a nearly normal
range within a matter of weeks. It is no exaggeration to say that the
drug has given millions of folks back their Eves and livelihoods.
By the time EPO received approval from the U.S. Food and Drug
Administration (FDA), the government agency charged with regu¬
lating all medicines, George Rathmann had retired and I was in my
first year as Amgen’s CEO. The company finished fiscal 1988 $8.1
million in the red. But with Epogen’s much-anticipated arrival, we
netted $19.1 million in 1989 and $34.3 million the next year, on
total revenues of $199.1 million and $381.2 million, respectively.
In 1991 Amgen launched its second human therapeutic: filgrastim
(Neupogen), EPO’s biologic cousin. Neupogen, too, exerts a stimula¬
tory effect on bone marrow, stepping up its output of a type of white
blood cell called neutrophils (noo-trow-fills). For cancer patients un¬
dergoing chemotherapy, this can spell the difference between life and
death. The toxic anticancer drugs destroy malignant cells, but one of
their adverse side effects is to decimate the population of healthy neu¬
trophils, the immune system’s foot soldiers against infection.
Should the blood’s concentration of neutrophils fall perilously
low, chemo may have to be postponed or the dose reduced. While
6 Science Lessons

the patient, her family, and her doctor wait anxiously for the white
count to rebound, the delay affords cancer cells lurking in the body
an opportunity to proliferate unchallenged. What’s more, men and
women whose immunity is compromised are susceptible to poten¬
tially fatal bacteria. By stabilizing neutrophil levels, Neupogen enables
cancer patients to receive the most effective dose of tumor-killing
medicine, on schedule.
Nearly two decades later, Epogen and Neupogen remain the
best-selling biologic agents in history. Each has earned additional
applications, known as indications in FDA parlance. EPO is now used
to treat anemia triggered by the human immunodeficiency virus
(HIV), cancer, chemo, or surgery, and Neupogen’s patient population
has expanded to encompass people with HIV, myelodysplastic syn¬
drome (MDS), and low neutrophil counts brought about by certain
medications.
Between 1988 and 2000, the year I retired, Amgen grew into the
largest and most successful biotechnology company in the world, far
outdistancing all but a handful of competitors. Our financials as¬
cended so sharply that you could strain your neck trying to follow
the trajectory on a line graph. Accordingly, I’ve confined them to
table I-1. Putting these figures in perspective, bear in mind that in my
final full year as CEO and chairman, when Amgen earned $1.1 bil¬
lion in net income and $3.6 billion in total revenues, the biotech in¬
dustry as a whole lost $5 billion, a reflection of the approximately
$500 million it costs to deliver a single biologic drug to patients.
Such consistent growth enabled Amgen to establish itself as one
of Wall Street’s top performers of the 1990s. The company debuted
on the NASDAQ stock exchange on June 17,1983. Considering that
Amgen didn’t have any products at the time, going public seemed
premature to some observers. And it was; an IPO wasn’t in the origi¬
nal timetable at all. But our other sources of capital had shriveled up
like foliage during Southern California’s dry season, leaving an initial
public offering our only option.
 Introduction 7

TABLE 1-1

Amgen financials, 1988-2000

Year Profit/Loss Revenue

1988 - $8.1 million $19.1 million

1989 + $19.1 million $199.1 million
1990 + $34.3 million $381.2 million
1991 + $97.9 million $682 million
1992 + $357.6 million $1.1 billion
1993 + $383.3 million $1.4 billion
1994 + $319.7 million $1.6 billion
1995 + $537.7 million $1.9 billion
1996 + $679.8 million $2.2 billion
1997 + $644.3 million $2.4 billion
1998 + $863.2 million $2.7 billion
1999 + $1.1 billion $3.3 billion
2000 + $1.1 billion $3.6 billion

The IPO tossed us a $43 million life preserver, enough to keep

Amgen afloat until our medical research started producing results. If
you’d purchased one hundred shares that day, at $18 per, by 2000
your original $1,800 investment would have been worth eighty
times that, roughly $150,000.
Twenty-five years have passed since I first pulled up to that non¬
descript white building on Oak Terrace Lane. It’s still there, by the
way, but now it’s dwarfed by more than forty other structures on
Amgen’s sprawling 100-acre campus. The staff, in Thousand Oaks
and at nearly sixty other facilities in the United States and overseas,
numbers close to seventeen thousand.
Naturally it’s a tremendous source of pride to have had a hand in
nurturing the first Fortune 500 company to emerge from an industry
that, like Amgen, was in its infancy. More than 300 million people
have benefited from the roughly 130 genetically engineered drugs
and vaccines licensed thus far for public use. But because the gestation
8 Science Lessons

period for new biotech medicines is long—it takes nine to twelve

years, on average, to run the gauntlet of scientifically rigorous tests
that must be passed in order to win IDA licensing—only now is this
revolution in health care coming of age. Currently, there are seven
hundred drugs at various stages of development, or “in the pipeline.”
Fully half the candidates are descendants of biotechnology. All told,
they target more than two hundred medical conditions, including
many that have proved resistant to conventional pharmaceuticals.

WHAT THIS BOOK IS—AND ISN’T

Writing a memoir seems to be as expected of retired CEOs as perpet¬

ual rounds of golf. (It might seem heretical, but for the record, I don’t
play golf; I’m a tennis buff.) I wasn’t interested in adding to the clutter
of books in which former corporate executives declaim one board-
room epic after another, as if regaling colleagues over a glass of sherry
at the hunt club. I’ve read only a few titles from the genre, usually
when I’m stranded in an airport or on a long flight. But I generally
come away with the impression that little effort is made to clarify for
readers lessons they can extrapolate and integrate into their own busi¬
nesses, divisions, or departments, however large or small.
Science Lessons is not primarily a company history, nor is it my life
story. My purpose in taking you behind the scenes is to highlight
real-life examples of the guiding principles and management tech¬
niques that contributed greatly to Amgen’s success. Amgen was and is
a unique place. We did things differently from most businesses, be¬
ginning with our decision to let research and development lead the
way. The strongest science determined which products we pursued,
as opposed to the conventional method of targeting a desirable mar¬
ket—say, diabetics or people disabled by painful rheumatoid arthri¬
tis—and then trying to come up with a drug or device that might
benefit them. You can be fairly successful that way, but will your
company establish itself as a true innovator? Probably not.
 Introduction 9

Although the story is set in the pioneering days of biotechnology,

I’m confident that you’ll find the advice in these pages to be both
practical and universal, helping you sharpen your managerial and
leadership skills no matter what field you’re in.

AMGEN’S SECRET WEAPON

From the beginning, Amgen was a magnet for gifted, innovative men
and women. How does an organization attract outstanding employ¬
ees? Not to sound glib, but it does that by being a place where good
people want to work. Five years in a row, magazines such as Fortune,
Working Mother, and Industry Week named Amgen one of America’s
best companies to work for.
Certainly we offered attractive salaries and benefits, and the stock
options made available to every Amgen employee no doubt induced
some folks to stay who otherwise might have sought opportunities
elsewhere. As numerous studies have established, however, pay and
perks aren’t what foster long-term employee loyalty. It’s something
more profound, something that speaks to the very soul of a company.
In 1994, as Amgen approached its fifteenth anniversary, we set
out to put into words the qualities that made it special. For years our
people had contended that much of the organization’s success could
be attributed to its culture. Because a company’s culture emerges
from its values, we interviewed hundreds of staff members in all areas
of Amgen to learn which values they believed constituted the core of
that culture.
Today it seems that every company under the sun (or under a
cloud) has a values statement. Some are written by the CEO, and oth¬
ers are concocted by the public relations or human resource depart¬
ment. Sometimes they’re written by consultants who don’t even work
there. More often than not, the statement doesn’t truly reflect the or¬
ganization’s values; it’s either a wish list of what the company aspires to
be or a PR tool for impressing customers, suppliers, and investors.
10 Science Lessons

Our project was as simple as it was unusual: define Amgen by

defining the values that had shaped it over the past decade and a half.
Personally, I’d hoped that we would end up with five values on the
list. Why five? No particular reason, other than it just seemed like a
number that would be easy to remember. In the end, there were ex¬
actly eight. Many people, both inside and outside the company, con¬
sidered the Amgen values our secret weapon (see “AmgenValues”).
As you will see throughout this book, operating on the basis of
clearly understood and accepted values has many advantages. When
these concepts permeate all levels of the company—helping guide
personal behavior, decision making, and daily interactions with col¬
leagues—your people are better equipped to manage themselves,
with less need for supervision. In a crisis, they’re more likely to know
instinctively the right thing to do, as compared with people workng
for a company whose values have not been articulated or are not
universally shared.
Here’s an example of what I mean: people suffering from end-
stage renal disease must have their blood cleansed through an outpa¬
tient procedure called hemodialysis three times a week. A dialysis
machine filters out waste products and other toxins and then returns
the blood to the patient by way of a tube inserted into a vein in the
arm. Without these treatments, the poisons build up in the circula¬
tion, eventually bringing about mass organ failure and death. For
most men and women on hemodialysis who use Epogen, the drug is
injected into the tube.
In the 1990s, a hurricane demolished a kidney dialysis center in
Florida. Most of its clients were elderly people, who would now have
to travel to other dialysis centers. The local Amgen representative,
hearing that many roads were impassable because of flooding and
downed trees, took it upon himself to hastily charter a bus. He called
the head of the center and told him to notify everyone to report for
dialysis as scheduled. Amgen would take them to and from treatment
 Introduction 11

Amgen Values
Following are the eight Amgen values, together with a brief clarification
of each, extracted verbatim from a pamphlet we gave to all employees.

Be Science Based
Our success depends on superior scientific innovation, integrity,
and continuous improvement in all aspects of our business
through the application of the scientific method. We see the
scientific method as a multistep process that includes designing
the right experiment, collecting and analyzing data, and rational
decision making. It is not subjective or emotional but rather a
logical, open, and rational process. Applying the scientific
method in all parts of the organization is expected and highly
valued.

Compete Intensely and Win

We compete against time, past performance, and industry rivals
to rapidly achieve high-quality results. Winning requires taking
risks. We cannot be lulled into complacency by previous achieve¬
ments. Though we compete intensely, we maintain high ethical
standards and demand integrity in our dealings with competitors,
customers, partners, and each other.

Create Value for Patients, Staff, and Stockholders

We provide value by focusing on the needs of patients. Amgen
creates a work environment that provides opportunities for staff
members to reach their full potential. We strive to provide stock¬
holders with superior long-term returns while balancing the
needs of patients, staff, and stockholders.
12 Science Lessons

Work in Teams ^
Our teams work quickly to move scientific breakthroughs from
the lab through the clinic to the marketplace and to support
other aspects of our business. Diverse teams working together
generate the best decisions for patients, staff, and stockholders.
Our team structure provides opportunities for Amgen staff to
impact the direction of the organization, to gain a broader per¬
spective about other functions within Amgen, and to reach their
full potential.

Collaborate, Communicate, and Build Consensus

Leaders at Amgen seek input and involve key stakeholders in im¬
portant decisions. In gathering input, strong leaders will welcome
diverse opinions, conflicting views, and open dialogue for serious
consideration. They will clearly communicate decisions and ra¬
tionale openly and in a timely manner. Once a decision is made,
the leader and members of the team will all be accountable for
the results and for implementing the decision rapidly.

Trust and Respect One Another

Every job at Amgen is important, and every Amgen staff member
is important. We attract diverse, capable, and committed people
and provide an environment that fosters inclusion, respect, and
individual responsibility and that values diversity. Trust is
strengthened through personal initiative and by obtaining quality
results rapidly.

Ensure Quality
Quality is a cornerstone of all of our activities. We seek the high-
est-quality information, decisions, and people. We produce high-
quality products and services. Quality is woven into the fabric of
everything we do.
 Introduction 13

Be Ethical
We are relentless in applying the highest ethical standards to our
products, services, and communications.

A number of principles evolved naturally from the eight values.

Among the most important of these are the following:

• Employees must have the freedom to make mistakes.

• A manager’s primary function is to help staff members do their

jobs; it is not a staff members primary function to help the
boss do his or her job.

• The people who do the work should help plan the work.

• Every rule, policy, and procedure has exceptions.

• Have fun! (Most of the time.)

at no cost. Only after he’d made all the arrangements did the rep have
time to tell his supervisor about the situation and his response. Under
the circumstances, there was no need for him to obtain permission in
advance; the Amgen values made the appropriate decision obvious.
An executive who was fairly new to the company heard the story
and remarked to me, aghast, “Boy, Amgen is really out of control! Can
you imagine a guy doing something like that without first getting his
boss’s approval? A sales rep has no authority to charter buses!”
I replied, “I can’t imagine anybody not chartering the bus first.”
He gave me a dumbfounded look. Then the light bulb flickered
on. He realized that Amgen was a different kind of place. Hewlett-
Packard, IBM, and a few other outstanding companies ran this way,
but they were a minority. Given the benefits—superior manage¬
ment, lower cost—it’s a mystery to me why more organizations
don’t adopt a values-driven approach.
14 Science Lessons

Given an expanding global economy, faster communication, and

an endless parade of technological advances, the business climate is
increasingly subject to volatility. Companies must keep watch on the
horizon and respond quickly and intelligently, or else they will fall
behind their competitors. It’s difficult to compete if your organiza¬
tion is so intricately structured that it lumbers along even in its high¬
est gear.
As Amgen grew exponentially, we constantly wrestled with the
same quandary that confronts most flourishing companies at some
point: how to remain nimble when you’re no longer a small start-up.
You do it by decentralizing power, of course, but also by establishing
an entrepreneurial culture that embraces change and encourages in¬
novation. For that to happen, management must empower its people
and then support them 100 percent, because staffers do not offer ideas
freely if they secretly believe they will be hung out to dry should their
promising project flop. In an industry such as biotechnology, failures
abound. Had Amgen not lived its principle “Employees must have the
freedom to make mistakes,” we would not have survived.
Although I’m not a fan of the New York Yankees, they epitomize
an organization self-governed in part by values. When I was growing
up in Alamogordo, New Mexico, the franchise was often compared
to industrial colossi like U.S. Steel and General Motors for the way
its members carried themselves, not to mention the businesslike man¬
ner in which the team dismantled opponents. From 1949 through
1964, only twice did the Bronx Bombers fail to take center stage in
the World Series.
Yankees managers, then and now, have rarely had to assume the
role of strict disciplinarians. It wasn’t that the players didn’t enjoy the
nightlife, as young professional athletes are prone to do. But if some¬
one’s carousing got out of hand to the point that it adversely affected
his performance, his teammates could be counted on to confront
him and growl, Don t mess with my World Series money!”This was
before athletes commanded higher salaries than most CEOs, so the
 Introduction 15

extra thousands in their Series checks were a windfall. Rarely did the
skipper have to intervene.
I’ve always been fascinated by the inner workings of companies.
When you think about it, a business is much like a living, breathing
entity, experiencing constant change. How are some corporations
able to achieve success year after year? Does the company mold its
personnel, or is it the other way around? I say it’s a bit of both.
When hiring, an organization with consistent internal values se¬
lects candidates that possess similar traits. However, it can also afford
to take chances now and then on gifted people who may not be com¬
pletely compatible with the corporation’s values. Returning to base¬
ball for a moment, look at the Yankees of the late 1990s and the early
2000s. Why was it that they were able to sign stars saddled with repu¬
tations as troublemakers, egotists, malingerers—you name it—and yet
almost from the start they adapted and became model citizens? It was
the combination of a proud tradition as the sport’s preeminent fran¬
chise, continued excellence on the field, the quiet but firm dugout
presence of manager Joe Torre, and the self-discipline that came from
a teamwide desire to be part of a winner.
I’d like to think of Amgen as being the New York Yankees of its
field—although if using that analogy means casting me as Yankees’
owner George Steinbrenner, I will have to reconsider.
■

-
CHAPTER

The Start of a Start-Up

if i asked you to reel off the names of the most successful execu¬
tives of the past twenty years and describe their individual traits,
probably no two profiles would be exactly [Link] extroverted and
the introverted, the micromanager and the delegator, the personable
and the borderhne misanthropic—any one of them can be ideally
suited to run a particular company or department. The success of
executives depends largely on external factors, such as the caliber of
the people around them and the stage of the organization’s develop¬
ment. As the old saying goes, horses for courses.
In the setting of a start-up operation, I can think of no one better
qualified than my predecessor, George Rathmann, first CEO of
Amgen. The group of venture capitalists and scientists who decided
to enter the biotechnology sweepstakes under the name Applied Mo¬
lecular Genetics—AMGen, as it was first known—certainly thought
he was the one.
But I’m getting ahead of myself. Let’s go back to the conception
of Amgen. In 1978 William K. Bowes Jr., a fifty-one-year-old former
18 Science Lessons

investment banker, stepped down from the board of Cetus Corpora¬

tion, the world’s first biotechnologycompany. Bowes had grown in¬
creasingly frustrated with the seven-year-old firm, feeling that it had
drifted off course. Over the next year or so, he began to consider what
he calls a “half-baked idea” to launch a similar enterprise. Some half-
baked idea.
The Bay Area native canvassed local universities for a scientist
who would be interested in collaborating with him. His search took
him to his alma mater, Stanford University, where a friend referred
him to Winston A. Salser, a molecular biologist and physicist at the
University of California at Los Angeles (UCLA) with a pedigree in
genetic engineering. Over drinks, the two discussed the climate in
the biotech industry and whether or not they should go forward.
Perhaps the young field was already too crowded. By the end of the
evening, though, they’d concluded that at least a few opportunities
remained and shook hands on their new venture.
Salser’s first task was to assemble a formidable scientific advisory
board composed of biologists from leading California universities,
while Bill reeled in investors. First to come on board were silver-
haired Franklin P. “Pitch” Johnson Jr. and Raymond Baddour. John¬
son, a former classmate of Bowes’s at both Stanford and Harvard
Business School, had founded a California investment firm called
Asset Management in 1965 and had been instrumental in assisting a
number of fledgling companies. Baddour, a professor of chemical en¬
gineering at the Massachusetts Institute of Technology, lent the new
venture the invaluable asset of academic credibility. He would sit on
Amgen’s board of directors for sixteen years; Pitch Johnson, for
twenty-seven. Both men imparted much valuable advice during my
time as Amgen’s CEO.
All told, Bowes coaxed six venture capitalists into putting up
roughly $81,000 apiece in seed money. For one of them, Moshe
Alafi, Amgen marked his third biotech undertaking. The prolific
 The Start of a Start-Up 19

Alah had been an early chairman of Cetus. Like Bill, he’d lost confi¬
dence in the company’s direction and left the board. In 1978 Alafi
had cofounded Biogen—only the third biotechnology firm in the
world and the first on the East Coast.
Applied Molecular Genetics was incorporated on April 8, 1980.
The next step: hire a chief executive. Winston Salser suggested Rath-
mann, then the head of Abbott Laboratories’ diagnostics division.
George, with a PhD in physical chemistry, had come to the pharma¬
ceutical giant from Litton Industries in 1975 and was instrumental in
transforming what had been a lackluster division into the industry
leader.
Ordinarily, Rathmann might not have been Salser’s first choice,
given that Abbott had no genetic-engineering program. But an Ab¬
bott colleague had piqued George’s interest in biotechnology, telling
him about the exciting advances being pioneered by a professor at
UCLA’s Molecular Biology Institute. The prof’s name? Winston
Salser. In a fortuitous coincidence, Rathmann wangled a six-month
sabbatical in Salser’s lab so that he could immerse himself in this new
science. How he got Abbott to agree to the leave of absence, I’ll
never know. Universities may do that for faculty, but companies?
Rarely.
George returned to Chicago awed by biotechnology’s potential. “I
could see that it was going to move ahead very rapidly,” he recalls. At
about the same time, Salser was telling his Amgen colleagues about the
dynamic Abbott executive his laboratory had recently hosted.
Bill Bowes placed the first of many calls offering Rathmann the
chief executive position. Amgen courted him throughout the spring
and summer. He was suitably impressed by the company’s power¬
house scientific advisory board and, most of all, by its commitment
to solid science. “I remember Bill emphasizing to me, ‘We’re out to
build a great company, not just a great stock,”’ says George. “Those
were convincing words.” Sincere, too. Bowes’s mother had been a
20 Science Lessons

[Link] him, biotechnology wasn’t merely another investment op¬

portunity; he was excited by the prospect of making a lasting contri¬
bution to medicine.
The main reason it took six months to pry Rathmann away from
Abbott was that he and his wife, Joy, felt the same way about the
prospect of moving to California that Adele and I felt about Michi¬
gan. They hated it. Except for George’s four years at New Jersey’s
Princeton University (where it wasn’t unusual to spot Albert Ein¬
stein or J. Robert Oppenheimer strolling across campus), he was a
born-and-bred Midwesterner: raised in Wisconsin, attended North¬
western University in Illinois, and logged twenty-one years at Min¬
nesota’s 3M, originally as a plastics researcher. The Rathmanns lived
with their five children in a magnificent lakeside home in a Chicago
suburb and were understandably reluctant to move halfway across
the country. Not to mention the fact that George was perfectly
happy at Abbott Labs.
Had he ultimately decided to stay there, he still would have influ¬
enced the course of biotechnology, but in a different way. To this day,
not one pharmaceutical company has created a biotech division, for
reasons I examine later. When George told Abbott’s CEO and presi¬
dent about Amgen’s proposal, they countered by offering him his
own biotech subsidiary.
Naturally, George was [Link] was only one hitch: “They
wanted to retain 52 percent of the stock,” he explains. “In the end, I
didn’t think it would be as much fun as Amgen, because Mother Ab¬
bott would always be in control. I really wanted to be part of a truly
independent effort.” In October 1980 Rathmann relocated to Ven¬
tura, California, to become Amgen’s chief executive officer and sole
employee. Abbott Laboratories never did pursue genetically engi¬
neered therapeutics in any meaningful way.
Although George had successfully run a division, he’d never been
in charge of an entire company. In fact, several Amgen board mem-
 The Start of a Start-Up 21

bers initially voiced reservations about hiring him for that reason.
“They’d also pointed out to me—quite clearly, too!—that someone
who’d worked at two huge companies might not be effective in a
small company,” he recalls, laughing.
What the venture capitalists didn’t fully appreciate was that Abbott
and 3M, both founded at the beginning of the twentieth century, op¬
erated somewhat unconventionally for such large corporations. In the
mid-1970s Abbott had decentralized its bureaucratic management
structure, granting some division heads far-reaching authority.
3M, best known at the time for its many tape goods and a fast¬
growing business manufacturing health-care products, fostered a
progressive multidisciplinary approach to management. “When you
got into a division at 3M,” George explains, “there were four key
people: the president, the vice president of research, the head of mar¬
keting, and the finance guy. As a member of the management team,
you were always discussing each of those aspects, which gave you an
opportunity to think like a general manager.”
At Amgen, Rathmann swiftly put to rest any lingering concerns
about his never having directed a company. Though first and fore¬
most a scientist, he displayed keen business instincts; and if he didn’t
know something, he learned it fast.
To me his greatest strengths were an infectious enthusiasm and a
prodigious power of persuasion. In the biotech world, George is
known as “Golden Throat.” The nickname, though well deserved,
doesn’t do him justice, invoking the slick patter of a traveling medi¬
cine show huckster or the Gatling gun delivery of a used-car dealer
straight out of central [Link] wasn’t George’s style at all.
To begin with, at six-feet-four and 250 well-muscled pounds, the
burly, bearded Rathmann was an imposing presence. And although
he is highly cerebral, he can tailor his explanations of genetic engi¬
neering to his audience, whether a conference room of molecular
biologists or a prospective investor with no scientific background.
22 Science Lessons

AMGEN’S ADVENTURES IN VENTURE

CAPITAL FUNDING

It’s no coincidence that U.S. manufacturers produced all of the

world’s first twenty biotech drugs and continue to dominate the
market, as they do in personal computers and software, semiconduc¬
tors, and many other fields. In the late 1970s and early 1980s, the
United States boasted the technology as well as a university system
for conducting research—with the help of government grants—and
for turning out future scientists. Most important, it possessed the en¬
terprising spirit that to the rest of the world defines U.S. business, if
not the United States itself. Neither the National Institutes of Health
nor the major pharmaceutical manufacturers gave birth to biotech¬
nology; it was comparatively tiny start-ups like Amgen, Genentech,
and Biogen. That’s unheard of in most other countries.
But then, our economic system was designed to encourage the
financial risk taking that fuels innovation. U.S. entrepreneurs held a
tremendous advantage over their foreign counterparts in that even
an untested company might one day have the opportunity to earn
millions of dollars by selling publicly traded stock. That possibility is
largely what entices venture capitalists to lay the foundations for new
businesses, when the risk of failure is greatest. In contrast, technolog¬
ically advanced countries like Japan and Germany typically did not
allow businesses to pursue this route until they had demonstrated
profitability and stability.
When George Rathmann began searching for funding in the fall of
1980, the U.S. venture-capital market was emerging from years in the
doldrums. The more favorable climate was stimulated, in part, by two
recent legislative and regulatory reforms. First, Congress hacked the
capital-gains tax rate from 50 percent to 28 percent, and then the U.S.
Department of Labor removed the Employee Retirement Income Se¬
curity Act as a barrier to venture investing. VC financing would soon
soar, from less than $600 million in 1980 to $4.6 billion in 1986.
 The Start of a Start-Up 23

The average first-round investment in biotechnology firms was

$1.1 million, the same as the average for all [Link] George set
his sights on $15 million. Although he’d never approached venture
capitalists before, he’d once served as a scientific consultant to a small
start-up launched by friends in [Link] experience, though
brief, left quite an impression. “I watched how venture capitalists op¬
erated,” he explains. “They always seemed to argue that you didn’t
need as much money as you thought you needed, and they’d hint
that there would be more money down the road. Except that when
you came back to them for a second or third round, they’d make it
very difficult. It creates this vicious cycle where the board members
hold complete control of your company. And the weaker you are, the
happier they are, because they’re going to get bargain-bin prices.”
Such circumstances force companies to perpetually chase the
carrot on a stick. A progressively smaller carrot, too, because of a stage
of VC financing called down rounds. Here, investors’ shares of equity,
or ownership interest in the corporation, come at a reduced price
compared with previous rounds. During flush times, a down round is
a scarlet letter that shouts failure and shoddy management. Though
more acceptable in a sluggish economy, it’s still considered a measure
of last resort. For many of those early-stage investors whose equity
has just been watered down like a cocktail at a cheap tavern, down
rounds are abominations.

Rathmann Makes the A Round

George Rathmann rejected the conventional wisdom that a new

business shouldn’t procure too much venture capital, lest its founders’
stakes become diluted to the point of irrelevance. “When Amgen
started, there were plenty of silver-tongued people saying to me,‘Fif¬
teen million dollars? You don’t need fifteen million dollars! You
should take four or five million. Then you can get the rest of it at a
premium someday, because you’ll have made some real progress.’”
24 Science Lessons

George’s skepticism about whether someday would ever materialize

proved to be well founded.
Because investor interest was light at first, Amgen wisely decided
to pursue institutional investors as well as [Link] first per¬
son approached, oddly enough, was Kirk Raab, corporate group vice
president at Abbott Laboratories, the company Rathmann had left
for Amgen. Raab, soon to be named Abbott’s president and chief fi¬
nancial officer, was a hard-charging, cigar-chewing salesman, not a
scientist. Yet he was intrigued by genetic engineering, whereas most
of his researchers harbored serious doubts about its value and safety.
By sheer dint of personality, Raab convinced a reluctant Abbott
board of directors to sink $5 million into the new endeavor. Good
thing they did, as much for their company’s sake as for Amgen’s: in
1990 Abbott sold its stock for $700 million, or 140 times its original
investment. By then Raab was happily indulging his fascination with
biotech as the CEO of Genentech, Amgen’s prime competitor.
We always suspected that a second factor motivated Abbott’s de¬
cision: the Illinois drugmaker secretly worried that this California
upstart might encroach on its lucrative diagnostic division. By be¬
coming Amgen’s major shareholder and acquiring a seat on our
board, the company could keep a hand in—and an eye on—things.
Abbott attorney Robert Weist hopped on a plane to hand-deliver
the $5 million check to his once and future colleague. (First, though,
he made a photocopy of it for his children to play with.) Within the
year, Weist joined Amgen as general counsel.
Next, Rathmann kindled interest from Tosco Corporation of
Concord, [Link] independent oil and energy company had
determined that the future of the oil industry lay in hard, dark sedi¬
mentary rocks called shale. An organic substance in the rocks can be
converted into medium-grade oil or gas. About three-fifths of the
world’s oil-shale resources happen to be found in the western
United States. Unfortunately, the process to extract and refine this
abundant fuel source is complicated and prohibitively expensive, as
 The Start of a Start-Up 25

several major oil companies discovered in the wake of the oil crises
of the 1970s.
Tosco put up $3.5 million, crossing its fingers that perhaps
biotechnology could create bacteria capable of drawing out the nec¬
essary oil-making materials. The overly simplified explanation of this
theory is that genetically engineered cells would ingest—but not di¬
gest—the shale’s oil-making components, which could then be re¬
trieved in the [Link] did not require Amgen to be involved
in researching and developing oil shale, and it didn’[Link] date, bioengi¬
neering has yet to yield any significant breakthroughs in oil shale.
Nevertheless, the company made a tidy profit on its investment.
Amgen probably benefited from the fact that its first round coin¬
cided with a milestone in the history of genetic engineering. On
October 14, 1980, Genentech became the first biotech company to
enter the stock market. Its initial public offering stunned Wall
Street—stunned everyone—by generating $35 million and tripling
its stock price on the first day of trading. The four-year-old firm’s
successful IPO demonstrated to venture capitalists that an investment
in biotech could achieve liquidity within the preferred timetable of
four to six years even if the company hadn’t made a product.
George, now with $8.5 million in hand, finally went after the
venture capitalists who’d been straddling the fence. Having Abbott
Laboratories in his corner proved to be a coup, because it validated
Amgen’s scientific goals and helped convince fence-sitters to declare
themselves. Pitch Johnson and Bill Bowes, two of the original in¬
vestors, contributed to the first round. Then the three of them went
before New Court Partners and dazzled it into investing $3 million.
New Court was something of an oddity: a cross between an old-
fashioned family-run partnership and a modern investment firm.
The $40 million fund was managed by the wealthy Rothschild clan
and underwritten by institutional investors.
On January 23, 1981, Rathmann closed the A round with a re¬
cord $19.4 million in financing. A quarter century later, that is still
26 Science Lessons

an astronomical sum, nearly four times the average investment of

$5.3 million. Making the feat more remarkable still, Amgen consisted
of only three people, including the bEO. Ordinarily, by the time of a
first round, a new company has already tried out a prototype of its
product in the lab. George, however, couldn’t point to any products
or patents, only a plan. No doubt the financiers who backed Amgen
were swayed as much by his personal magnetism and professional
stature as they were by the company’s business prospects.
Believe me, that Golden Throat nickname wasn’t hyperbolic in
the least. Not that his presentation to prospective investors didn’t re¬
quire some fine-tuning. At the very first session, someone piped up,
“What are sales going to be in five years?”
“Zip!'” he replied matter-of-factly. “There can’t possibly be any
sales in just five years.” The financiers zipped up their wallets.
Afterward, a perturbed Rathmann collared board member Bill
Bowes. “They decided not to invest, and they were good candidates.
How do you explain that?”
Bowes, always the diplomat, thought for a moment. “You shouldn’t
let it bother you, George,” he began. “But ... the next time someone
asks you what the projected sales are going to be in five years, you
might come up with a better answer than ‘zip.’”
“After that advice,” Rathmann recalls, laughing, “I still made it
clear that Amgen would not have any products to show for its first
five years. But I added that by then the company’s value would be
ten times what it was when we started, based on a couple of projects
that we believed could be launched within five years. Incorrectly, it
would turn out. We were looking more at a ten-year plan, which was
why it was so difficult to promise a significant rate of return.”

Building the Staff

With $19.4 million in the bank, Amgen now sought to build a re¬
search and development staff. Having as CEO an exceptional scien-
 The Start of a Start-Up 27

tist who’d managed scientists at two outstanding corporations for

more than twenty-five years certainly helped. Rathmann’s charm
and genuine passion for biotechnology enabled him to recruit out¬
standing young scientists who might otherwise have turned up their
noses at the company’s unspectacular digs in sleepy Thousand Oaks.
Kirby Alton, an early hire, remembers his first impressions upon
arriving for an interview in the summer of 1981. “There really wasn’t
anything there except for George”—sitting at a desk in a corner of
the lobbyless office—“and a picture of a one-story building,” says the
microbiologist from Georgia. “It was an artist’s rendering of what the
laboratory then under construction was going to look like. But
George Rathmann was such an impressive, unbelievably charismatic
man, I immediately felt that if anybody could build a biotech com¬
pany, he could. When George offered me a job as a research scientist
over dinner at a local Holiday Inn, I accepted before he’d even
spelled out the terms.” Alton would spend eighteen years at Amgen,
eventually ascending to senior vice president of development.
The company’s seemingly firm financial footing was also appeal¬
ing. “Nineteen million dollars? We thought that was going to last
forever!” he exclaims with a laugh.

LESS THAN TWO YEARS LATER . . .

Did I mention that when I joined Amgen as its first chief financial
officer in October 1982, the company was running out of money?
I knew that going in. During my first meeting with George
Rathmann, he had been more than candid about the firm’s need to
generate capital, and quickly. At the pace it was spending money,
Amgen could last perhaps through the following summer before its
bank balance hit zero.
In most industries, such news probably would have generated
panic. But in biotechnology, where the burn rate—the monthly cost
of doing business—is appreciably hotter than in other industries,
28 Science Lessons

hemorrhaging money was standard operating procedure. Probably

every biotech teetered on the brink of financial disaster at one time
or another, locked as they all were irf a long-distance race to discover
a marketable drug before their funding ran out. Even today, three in
four publicly traded biotech companies operate with less than eigh¬
teen months’ capital.
Before my arrival, Amgen’s finance department consisted of a
certified public accountant in private practice who showed up twice
a week just to keep a general ledger. It’s not unusual for a start-up
venture to wait a year or more before hiring a CFO. For one thing,
when you’re a small company that’s still years away from even con¬
templating going public, how will you attract someone who is suit¬
ably experienced and won’t be out of his depth when the enterprise
begins expanding?
Frequently, a fledgling corporation brings in a chief financial offi¬
cer at the very stage where Amgen now found itself: anxious to re¬
plenish its coffers. When George was selling me on joining the
company, he’d observed, “You know, you’d have a great advantage
over our other employees. You’d be betting on yourself to raise the
money we need to stay in business, whereas everybody else would be
betting their futures on a complete stranger.” No pressure there, right?
Projections that the funds raised would carry Amgen through
1984 underestimated by about half a biotech R&D’s insatiable ap¬
petite for money. At the end of 1982, our tank was already three-
quarters empty, with roughly $5 million left.
The fierce burn rate was a concern throughout the industry, and
not only at Amgen (see “Tick,Tick,Tick: Burn Rate”). In the 1980s,
the average cost of bringing a genetically engineered drug to market
ranged from $25 million to $75 million. A decade later, developing a
comparable product exceeded $100 million. By then, companies had
grown. Early projects may have been the work of a single research
scientist and a handful of lab assistants; now it was common for proj¬
ect teams to consist of fifty or more people. The expense of con-
 The Start of a Start-Up 29

Tick, Tick, Tick: Burn Rate

To calculate a business's burn rate, look at its most recent form 10-Q,
the quarterly financial report that all public companies must file with
the federal Securities and Exchange Commission (SEC) three times a
year. (No 10-Q follows the end of the fourth quarter, because that is
when a company submits its annual report, or 10-K.)
For an estimate of its annual net loss, you multiply the business’s
quarterly net loss by 4. Next, total the available cash. Then divide the
latter figure by the first number to see how long the company has
until its last nickel runs out. You can make a number of adjustments—
such as adding back noncash charges like depreciation—but the re¬
sults don’t change much.
Here’s a quick example, using the imaginary 10-Q of fictitious toy
retailer Woe Is Us:

• $8 million (quarterly net loss) x 4 = $32 million (annual net toss)

• Cash and other assets = $40 million

• 40 + 32 = 1.25 years, or 15 months, before bankruptcy looms

structing laboratory space nearly doubled, to more than $500 per

square foot. And a piece of equipment that used to run, say, $25,000
often exceeded $1 million.
Those figures cited for the average cost of developing a new drug
don’t take into account the largest expenditure of all: the millions
poured into promising agents that never pan out. According to the
Pharmaceutical Research and Manufacturers of America (PhRMA),
the trade association for the country’s leading pharmaceutical re¬
search and biotechnology companies, for every five thousand com¬
pounds that show promise in the laboratory and in animals, only five
30 Science Lessons

will advance to testing in human beings. Of those five hopefuls, only

one will go on to win the blessing of the U.S. Food and Drug Ad¬
ministration and land on your pharmacy shelf.
When you factor in the cost of failures, as researchers at Tufts
University did in a 2003 study of drug development costs, the total
expenditures of each federally sanctioned medication—prior to ap¬
proval—averaged $138 million in 1979, $318 million in 1991, and
$802 million in 2003. Those numbers are somewhat inflated, be¬
cause the researchers added interest for each year of research and
development. A more appropriate figure would be in the vicinity of
$500 million (without interest) per new medication—still an enor¬
mous outlay.

Drug Development Phases

In many businesses, the solution to the problem of high overhead is

simple: cut overhead. Speed up production. But drug manufacturers
don’t set their own standards; the government does. Drugmakers are
bound by the FDA drug-approval process, which requires extensive
preclinical testing in the laboratory and in animals, followed by three
phases of evaluations in patient volunteers. (In medicine, the word
clinical denotes human involvement, whether it refers to investiga¬
tional studies, diagnostic procedures, or treatments.) Later in this
book I’ll walk you through each stage, but briefly, a clinical trial is de¬
signed to answer various questions concerning a compound’s safety
and efficacy.
Table 1-1 outlines the three phases of clinical testing. The num¬
bers of patients and the study duration are for a highly efficient drug
company testing a best-case product—one that is safe and effective,
with no anticipated problems. Many investigations require even
more patients and take years longer.
Before clinical trials even begin, the company has invested
three to five years in research, drug design, and preclinical testing.
 The Start of a Start-Up 31

TABLE 1-1

The three phases of clinical trial testing

Phase of Attempts to answer Number of Duration

testing these questions patients of study

Phase 1 • At what dose is the drug safe, • 20 to 200 healthy • 6 months

without producing unwanted volunteers to 1 year
side effects?
• How should the drug be
administered (orally, by
injection, and so on)?

Phase II • Does the drug show evidence • 200 to 500 healthy • 1 to 2 years
of being effective? volunteers and
• What is the correct dosage? patients afflicted
with the disease the
drug is intended to
treat or prevent

Phase III • Does experimental data • 300 to 5,000 • 2 to 3 years

support substantial efficacy or more patients
with high statistical significance? with the same
• Does long-term usage lead to medical condition
any side effects?

Then tack on roughly one and a half years at the end of phase III
to prepare the new drug application (NDA)—which can be as long as
500,000 pages—and then wait a year while the agency reviews it.
All told, the path to commercial licensing takes eight to fourteen
years. In the interest of public safety, there’s no cutting corners, nor
should there be.
Now here’s the part about biotechnology—and drug develop¬
ment in general—that someone who thrives on predictability would
probably find difficult to accept: when trying to create life-changing
medications, especially for applications that haven’t been tackled be¬
fore, you can do everything “right”—couple the best science with
hard work and impeccable execution—and yet the project may im¬
plode at any point, without warning, leaving you with nothing.
32 Science Lessons

Pharmaceutical research and development presents the reverse

challenge of R&D at, say, an engineering-based company. It’s a more
difficult challenge, too. In engineering, little doubt exists that a de¬
sired product can be created; the uncertainty lies in what it will cost
and how many customers will buy it and at what price. In drug
R&D, you know in advance that there’s a need for the product and
that the health-care system will pay for it at a reasonable price. The
big question is, Can you do it? I believe that Amgen was much better
than its competitors at R&D decision making, where there’s a wide
gap between capable management and poor management. You’ll
learn more about R&D management in chapter 3.
I liken the process to a marathon. Failure can come immediately
out of the gate, during preclinical studies, or in a phase III clinical
trial, with the finish line almost in sight. Early failures are relatively
inexpensive, but late failures can be extraordinarily costly. Even after
FDA approval is granted, a drug may have to be withdrawn from the
market because of unanticipated side effects, and subsequent lawsuits
can tally in the billions of dollars.

TIME FORA SECOND ROUND—RIGHT?

There was a good reason Amgen was consuming capital at an alarm¬

ing rate. As Bob Weist, our general counsel, points out, “We were
making progress faster than was originally anticipated, which meant
that we had to hire more employees sooner.”
George Rathmann—keenly aware that companies like Genen-
tech had a five-year head start (not to mention millions of dollars
from its initial public offering)—wasted no time in trying to com¬
pete. “We had a lot of incentive to move rapidly,” he reflects. “Genen-
tech was gigantic compared to us. In fact, some people used to refer
to the biotech industry in general as Snow White and the Seven
Dwarfs Snow White, of course, being Genentech. The pervasive
feeling was that the rest of us were going to stay dwarfs forever unless
 The Start of a Start-Up 33

we worked very hard, followed a sound strategy, and were blessed

with some good luck.”
When the need to raise additional money became painfully ap¬
parent, in the fall of 1982, George naturally assumed that it was time
for a second round of venture capital. Biotech companies routinely
returned to the same well for financial sustenance twice, three times,
sometimes even a fourth time. A second round was typically ex¬
pected to bring in $5 million to $14 million.
Rathmann turned to Frederick Frank of Lehman Brothers. Frank,
a legendary name in the venture-capital and biotech worlds, has been
an investment banker since 1958. After eleven years at Smith Barney
(now Salomon Smith Barney), he moved to Lehman Brothers and has
been there ever since.
George wanted to know how to initiate a successful second
round.
“Oh, it’s not hard,” Fred said confidently. “You’ve got a great
thing going here, George. Just ask your investors to offer one dollar
for every four dollars they put up last time. Do that, and you’ll have
raised around five million, which should give the company breathing
room for another couple of years.”
Collaring the original financiers couldn’t have been simpler, be¬
cause four out of the five major VCs sat on our board: Pitch John¬
son’s Asset Management; Bill Bowes’s company, U.S. Ventures; Kirk
Raab of Abbott Laboratories; and New Court Partners’ representa¬
tive, Jim [Link] Corporation, not wanting to incur the liabilities
that come with being a board member, had declined a seat. Of the
four, Blair was the most vocal in expressing his concern about
Amgen’s burn rate.
Although Rathmann hadn’t yet officially been declared chairman,
he essentially functioned as one and therefore conducted board meet¬
ings. “We’re going to run out of money in about eleven months,” he
announced at the November meeting. Fie then recounted Frank’s
one-dollar-for-every-four-dollars strategy.
34 Science Lessons

According to George, “The room went dead quiet. Not a single

venture capitalist had any appetite i^r putting more money into the
company. I thought,‘Holy smoke!’ It was extremely scary.”
Now what? Try elsewhere? Not likely. If no first-round investors
saw fit to reinvest, why would a newcomer? That was a tough sell
then and a tough sell today. Scarier still was the board’s ominous di¬
rective that it wanted to see a contingency plan at the first meeting
of 1983. To us, contingency plan could mean only one thing: laying off
[Link] was virtually no one else to let go. Except for a cou¬
ple of secretaries (George’s doubled as head of human resources), the
staff of fifty belonged almost entirely to research and development.
Rathmann summoned the four senior officers to his office: along
with me, he called in Bob Weist, research director Dan Vapnek, and
Noel Stebbing, an Englishman who’d defected from Genentech to
become our vice president of scientific affairs.
“I will not accept any layoffs,” George said firmly. “Not one. It
would be the kiss of death.” He explained that at one time, while he
was at Abbott, budget cuts forced all divisions, including his, to dis¬
miss employees. Rathmann reluctantly had to let some scientists go.
The blow to morale was devastating. “Factory workers get laid off,
not scientists/” Top researchers shunned the drugmaker’s job offers
for years afterward, and Rathmann vowed then that he would never
again pink-slip R&D personnel.
He proposed that if the board insisted on staff cuts, all five of us
would resign immediately. Now, this was hardly a room full of youth¬
ful radicals. But being part of a company that develops medications
for enhancing or prolonging human life tends to revive one’s ideal¬
ism. We all nodded in agreement.
However, there was still the matter of presenting a contingency
plan to the directors. “What if we take the company public?” I said. I’d
been privately discussing the prospect with every senior investment
banker I knew—all two of them: Lehman’s Fred Frank and Bob Hotz
of Smith Barney. Frank had pulled the strings for the second IPO
 The Start of a Start-Up 35

by a biotechnology company, in March 1981, when the Bay Area

firm Cetus Corporation raked in an unbelievable $107 million—still
an all-time industry record. Both men said essentially the same thing:
an IPO was possible, but only if the climate for IPOs improved.

How an IPO Works

The process of a company’s going public makes for a fascinating

study in human psychology. It’s remarkable how investors, normally
an analytical lot, can get swept up in often groundless euphoria over
a business or field and lose all sense of reason.
Under the Securities Act of 1933, every company conducting a
public offering is required to file a registration statement that dis¬
closes all pertinent information about its business, finances, and so
on. Beyond that, it is strictly a case of buyer beware, as was demon¬
strated in gruesome fashion by the e-commerce feeding frenzy of
the late 1990s and the subsequent indigestion brought on by the swift
collapse of many dot-coms. Regardless of the industry involved, the
story hne usually plays out pretty much the same.
The cycle gets rolling when one strong company dares to test the
waters. It boasts genuine prospects, solid management, and powerful
investment bankers, and its public offering is reasonably successful.
Next, another upper-tier business follows suit and then maybe an¬
other, forcing the IPO window of opportunity wide open. Eventu¬
ally, all the big fish in what is a rather small pond get snapped up, and
subsequent public offerings feature progressively weaker companies.
Investors, intoxicated by the rapidly rising prices of the recent past,
don’t catch on until they hook some scrawny minnows represented
by third-tier investment banks. When enough casualties have piled
up, the cycle groans to a halt.
Here’s an example: after Cetus Corporation’s spectacularly success¬
ful IPO in March 1981, speculation ran wild about how many mil¬
lions a Washington biotech firm called Genetic Systems Corporation
36 Science Lessons

would fetch in its offering, scheduled for April. One hundred ten mil¬
lion? One hundred fifty? More? v
It wasn’t even close. Genetic Systems hobbled away with $6 mil¬
lion, and a May public offering by Ribi Immunochem of Montana
grossed a microscopic $1.8 [Link] proceeds probably didn’t last
the summer. And with that—thunk!—the window slammed shut. Its
reverberations lasted a year, which passed without any activity.

An IPO for Amgen?

Greed springs eternal, however, amnesia sets in, and before long the
sequence revs up all over again. That’s where we stood at the begin¬
ning of 1983: three respectable biotech public offerings in a row over
the previous six months, a favorable sign. Biogen was due to step up
to the plate next, in March.
Two other recent events boded well for the biotechnology indus¬
try in general: at the end of October, Genentech received FDA ap¬
proval to market the world’s first genetically engineered drug,
Humulin, a synthetic form of the human hormone insulin. Men and
women with type 1 diabetes inject themselves with Humulin to help
prevent dangerously high levels of glucose, the body’s principal fuel
source, from building up in the bloodstream.
Then, four days into the new year, Congress passed the Orphan
Drug Act of 1983, a law granting financial incentives to drug manu¬
facturers to develop medications for treating less-common health
problems. Given the enormous cost, drugmakers could expect to
lose money on products aimed at these disorders, which the bill de¬
fined as any condition affecting fewer than two hundred thousand
Americans. Consequently, whole groups of patients were being neg¬
lected, including those with such familiar disorders as muscular dys¬
trophy, cystic fibrosis,Tourette’s syndrome, and hemophilia. In all, six
thousand ailments met the criteria for rare-disease status.
 The Start of a Start-Up 37

Makers of conventional drugs probably let out a mild hooray, but

for the biotechnology field the new legislation was a godsend, be¬
cause most of its early candidates were intended to take on orphan
diseases.
With some trepidation, George Rathmann convened the Amgen
board for the January meeting. The directors were eager to see the
contingency plan they’d requested. They expected to hear about
massive layoffs and other cost-cutting measures. Instead, I calmly in¬
formed them that our contingency plan was to enter the IPO on-
deck circle and take Amgen public.
Had we been in a Warner Bros, cartoon, you would have heard
crickets chirping. The board members looked at one another, per¬
plexed. One of them finally broke the silence: “How can you possi¬
bly think of having a public offering?!” he spluttered. “You have
nothing to brag about, no products even remotely approaching the
marketplace!”
When I explained that the esteemed Fred Frank of Lehman
Brothers felt that it was possible, the board members softened a bit.
To their credit, they agreed to discuss the proposal again at the
March meeting, with three provisos: that an investment banker had
been lined up, that they were supplied with a specific plan and time
line, and that the dreary economic climate had improved.
Luckily for us, the IPO market continued to recover, with Bio¬
gen commanding an exceptional $57.5 million the day it debuted on
the NASDAQ exchange. The directors gave us the go-ahead we
were looking for. They never did force us to implement those staff
cuts; nor did they ever find out that Amgen’s five senior managers
had been fully prepared to walk if they had.
 k
V
.

'

'
CHAPTER

Amgen Goes Public

it is NO exaggeration to say that Amgen’s entire future hinged

on a successful initial public offering. Fred Frank wanted to know
how much capital we hoped to raise. “As much as possible!” I told
him, recalling a nugget of wisdom dispensed by Bill Bowes. “I’ve
heard of lots of companies that ran out of money,” he once said, “but
I’ve never heard of a company that ran out of stock certificates.”
Anything less than, say, $25 million would be too little to sustain us
for a significant time or to impress Wall Street.
I’m going to walk you through Amgen’s IPO. Maybe this isn’t
relevant to your career at the moment. But at some point, you might
be a senior manager at a company about to embark on this route, and
you’ll be expected to pitch in. Even if you’re not directly involved in
the process, you certainly want to have a basic understanding of
what’s going on. After all, the outcome of a company’s public offer¬
ing affects every person who works there.
The freewheeling late 1990s painted a surrealistic picture of the IPO.
Today’s business-school graduates, lacking the historical perspective
40 Science Lessons

that comes with age, may not fully appreciate what an aberration those
years were. No doubt many of them have read the seductive stories
about self-promoting entrepreneurs charming investors out of millions
of dollars and are crossing their fingers that history will repeat itself.
I wouldn’t count on it, certainly not soon. In 1999, at the height
of Wall Street’s giddy infatuation with e-commerce and, to a lesser
extent, the technology sector, 457 IPOs raised $69.2 billion in fi¬
nancing. During less hallucinatory times, a business usually was six or
seven years old before it took the giant step of selling shares on a rec¬
ognized stock exchange. You can subtract a few years if the firm is
part of an emerging industry, as Amgen was. But in the mass delirium
of the high-tech bubble, online media company [Link] com¬
manded $27.9 million after four profitless years, and online pet-sup¬
plies store [Link] was still in the terrible twos when it raked in
$82.5 million. Who could blame young people for starting to believe
that an IPO bonanza was virtually one’s birthright just for showing
up at work?
On Friday, April 14, 2000, a 600-point free fall in the Dow Jones
average abruptly ended the longest peacetime bull market in U.S. his¬
tory. The NASDAQ composite index, which had been slipping since
March, shed 355.61 points, its largest single-day descent on record. It
was as though investors had sobered up after a long bender and took a
close look at some of the companies they’d bedded during the past
few years. When IPO activity is unusually high, as it was then, neither
investors nor the investment banks overseeing the process (the under¬
writers) have adequate time to fully research the organizations and
their management teams. Flaws, some of them significant, get over¬
looked: too much flab, too little going on upstairs, and so on.
In the less-forgiving light ofWall Street’s new dawn—and minus
the heat of the moment—hot prospects can turn out to perform dis¬
mally in the boardroom. More than five hundred e-commerce retail¬
ers were tossed on the corporate bone pile in 2000 and 2001. There
 Amgen Goes Public 41

was plenty of blame to go around for the dot-com debacle, starting

with inexperienced executives who couldn’t master even the rudi¬
ments of running a business—or were openly contemptuous of
them. Fingers also must be pointed at the underwriting syndicates
for championing these subpar companies, and at besotted investors
for throwing money at them. Presumably everyone learned a lesson,
because the next three years produced only 266 initial public offer¬
ings, about half the total in 1999 alone.

NO TIME TO LOSE

A successful initial public offering typically takes at least six months,

and that’s assuming the stars are all in alignment: a strong stock mar¬
ket and an organization with its house in order. Yet Amgen pushed its
IPO through in half that time.
Why the rush? Simple: since Wall Street is subject to outside forces,
you have no way of forecasting whether today’s sunny conditions
won’t darken a month from now, closing the IPO window. Believe me,
I know. In 1972, my second year as chief financial officer at System
Development Corporation (SDC), the board of directors concluded
that the time was right to take the company public. The gradual with¬
drawal of U.S. troops from Vietnam was nearing completion, and that
was expected to jump-start the economy. An investment banking syn¬
dicate was promptly pulled together, the appropriate documents were
filed with the Securities and Exchange Commission, and so on, until
we were only two days away from the effective date: the day of an initial
public offering.
Then, just before Christmas, President Richard Nixon ordered
U.S. B-52s to resume bombing the North Vietnamese capital of
Hanoi, ending a four-year [Link] turnaround in U.S. military policy
sent Wall Street into a spin. Smith Barney, SDC’s lead underwriter,
was willing to go ahead with the offering; however, it cautioned that
42 Science Lessons

any undertaken at that time probably wouldn’t be very [Link]

board wisely decided to wait. And wait.
Although the resumption of boifrbing helped hasten the war’s
end in a matter of weeks, the United States entered its worst stock-
market decline in more than thirty-five years. When six months
passed with no revival in sight (it would take nearly two years just for
the market to bottom out), SDC reassessed its strategy and elected to
sell the company outright.
Mindful of how a military action halfway across the globe had
torpedoed SDC’s IPO forty-eight hours before showtime, I informed
Amgen’s lead investment bank, Smith Barney, that we intended to
condense six months of preparation into three. When one ot the out¬
side attorneys handling the public offering read the schedule we drew
up, he rumbled, “This is impossible!” His law firm was already shep¬
herding five other IPOs the day it began working on ours.
“Sorry,” I told him. “We’re going to meet these dates, so you’re
going to have to meet them too. In fact, we’re going to set a Guin¬
ness world record for the fastest initial public offering ever.” As often
happens when a challenge is dismissed as impossible, all deadlines
were met, on both ends.
You don’t want to sacrifice performance for speed, though, be¬
cause filing inaccurate or incomplete documentation with the SEC
constitutes a serious offense. Also, an unrealistically narrow timetable
could lead to missed deadlines, forcing the IPO to be postponed.
Will the window still be open by the time you get back on the cal¬
endar? Maybe, maybe not. Companies don’t always appreciate the
importance of timing when pursuing a public offering and go about
their business as if they have all the time in the world. Even some
veteran underwriters proceed at a leisurely pace unless management
cracks the whip.
If your organization has been behaving like a public entity long be¬
fore it registers with the SEC, you stand a good chance of sailing
through the process with a minimum of ulcer-inducing moments. For
 Amgen Goes Public 43

instance, as soon as a company resolves to cross over from private to

public, it should get right to work on the following, if it hasn’t already:

• Hire a chieffinancial officer. Preferably, your CFO should be

experienced in IPOs.

• Develop relationships with seasoned underwriters, attorneys, and ac¬

countants. Lehman Brothers’ Fred Frank generously dispensed
sage advice even though we threw in our lot with Smith Bar¬
ney. The interest from the latter came about, in part, because
I’d known its head of corporate financing, Bob Hotz, for
years. Coincidentally, Hotz had been the underwriter for the
ill-fated System Development Corporation public offering;
the two of us had worked together closely and developed a
lasting mutual trust and admiration. Otherwise, there’s no
telling whether Bob could have persuaded his colleagues to
take on an unproved company like Amgen in such an uncer¬
tain IPO market.

• Employ an outside accounting firm to audit the company for at least

the past three years. A business that neglects to get an inde¬
pendent audit rolling well in advance of filing with the SEC
will not only incur needless expenses but also may be shocked
to discover that company performance has been falling short
of expectations. If the clock is ticking down, though, there’s
no time for improving matters.

• If necessary, give your board of directors a makeover. Under the

Sarbanes-Oxley Act of 2002, an organization’s board must be
genuinely independent. Members may not accept any fees
from the company or one of its subsidiaries for, say, consult¬
ing. Amgen, from its inception, always practiced exceptional
corporate governance. For example, we wouldn’t admit to
the board anyone from a company that did business with
Amgen, whether a bank, a law firm, or an advertising agency.
44 Science Lessons

Realistically, are people who profit from an association with

a company truly going to speak their minds or perhaps take
issue with a position held by the CEO?
If the chief executive is secure enough to welcome oppos¬
ing viewpoints—a trait of great leaders and consequently of
great companies—it shouldn’t be a concern. But the mere ap¬
pearance of a possible conflict of interest is enough to under¬
mine your business, and not only with federal regulators. Put
yourself in an investor’s shoes: would you feel confident pur¬
chasing stock in an organization whose board was stacked
with the founder’s relatives and old college roommates? I
wouldn’t.
A distinguished, autonomous board lends a company pres¬
tige. I can tell you that Amgen’s initial public offering un¬
questionably benefited from the presence of well-respected
academics and businessmen like Raymond Baddour and New
Court Partners’Jim Blair. These men were offered more seats
than an expectant mother on a crowded bus. Investors could
therefore infer that Amgen must have something going for it;
why else would folks of this caliber sit on its board?

• Make sure that company policies concerning executive loans and

stock-option plans conform to SEC regulations. Sarbanes-Oxley
prohibits publicly traded businesses from making certain per¬
sonal loans to directors and executive officers. Before the of¬
fering, management should consult its legal department to
learn which loan arrangements will be off-limits after the
company arrives on Wall Street, and bring its practices into
line.

Conducting business like a public company long before you be¬

come one instills invaluable discipline in managing earnings. It also
affords you some breathing room should a problem crop up—which,
in accordance with Murphy’s Law, you can pretty much bank on.
 Amgen Goes Public 45

Amgen’s IPO pursuit was nearly derailed before we’d even left the
station, and believe it or not, the hand throwing the switch belonged
to one of our own directors.
Abbott Laboratories, you’ll recall, bankrolled roughly one-quar¬
ter of our first venture-capital round in 1980, more than any other
shareholder. At the time, the two companies signed a letter of agree¬
ment granting Abbott the right of first refusal on some Amgen prod¬
ucts. The letter also stipulated that a definitive legal document
spelling out the details would be drawn up at a later date. Well, it
never was.
Now, two years later, we were on the eve of filing with the SEC,
and Abbott’s president, Kirk Raab—an Amgen board member!—said
that he would sue if we didn’t honor the letter. As per Abbott’s inter¬
pretation, this entitled it to license any Amgen product merely by
matching whatever terms we’d negotiated with another company.
Our position was simple: no! Amgen had been spending its own
money on R&D and would continue to year after year. It was never
the intent of the original agreement that Abbott could cherry-pick
whichever products it fancied, including those not even conceived at
the time of the signing. This is what we proposed: “If you want to
exercise your right of first refusal, we’ll give you one opportunity
now. Here is a list of everything our scientists are working on. Li¬
cense what you want, but you have no claim to them after that, or
anything else.”
Abbott responded by making good on its threat and taking
Amgen to court. The pharmaceutical manufacturer probably figured
it had us over a barrel, assuming that we couldn’t go forward with the
IPO until the matter was settled. However, I convinced Smith Bar¬
ney to hold the stock offering despite the dispute. Furthermore, the
underwriters would attend our March 1983 board meeting and state
their decision in person. I suggested to George Rathmann that we
address the pending lawsuit at the end, after the discussion about
whether to finalize our plans for the IPO.
46 Science Lessons

To be fair, Kirk Raab had always been supportive of Amgen, and

certainly he found himself in an awkward position. Glancing at the
written agenda, he protested: “I think this agenda is backward. How
can we discuss the public offering until we’ve discussed the Abbott-
Amgen lawsuit?You can’t conduct the IPO without a settlement.”
That’s when the investment bankers made their cameo appear¬
ance, effectively kicking Abbott’s legs out from under it. The show¬
down was resolved amicably a few days later and on very favorable
terms for Amgen. However, it was an eye-opening lesson that corpo¬
rate partners don’t always act like partners.
With that crisis behind us—and it was a true make-or-break mo¬
ment in the company’s history—we were free to set our sights on
Wall Street.

ON WITH THE SHOW

The action really ratchets up during the final two to four weeks,
when senior management barnstorms across the country, speaking to
roomfuls of financial analysts, brokers, and potential institutional in¬
vestors assembled by the underwriter. The series of presentations, fit¬
tingly dubbed the “road show” and sometimes the “dog and pony
show,” epitomizes capitalism in its rawest, purest state.
A company’s representatives have approximately thirty minutes
to promote interest in their stock offering—no small feat. Today
most dog and pony shows incorporate PowerPoint visuals—a tech¬
nological leap from the Kodak Carousel 35-millimeter slide projec¬
tor that George Rathmann and I toted in 1983. Usually the CEO
commands the spotlight, but we wanted to emphasize Amgen’s man¬
agerial depth. When the attendees bombarded us with questions,
George fielded the ones related to science (for more, see “A Brief Bi¬
ology Lesson: Recombinant DNA”) and I handled the queries about
our finances.
 Amgen Goes Public 47

A Brief Biology Lesson:

Recombinant DNA
The chemical deoxyribonucleic acid (DNA), which makes up the more
than thirty thousand different genes in the human body, is analogous
to a software program designed to create and sustain human life.
Everything about us, from eye color to perhaps a preordained suscep¬
tibility to a health problem, is encoded at conception by the DNA in
our cells.
Recombinant, essentially a synonym for “genetically engineered,”
refers to the scientific method of transferring DNA from one type of
organism into another and then growing these synthesized cells in
large numbers.

“How long is it going to take to get your first product on the

market?”
“It’s hard to say. Probably at least five years from now.”
“But you’ll run out of money before then. What will you do in
that case?”
“We’ll either strike up a licensing deal with a pharmaceutical
company or have another public offering—whatever the circum¬
stances dictate.”
Because Wall Street knew next to nothing about genetic engi¬
neering, the questions about Amgen’s product development were
unsophisticated by today’s standards. (Not long after, investment
firms began hiring PhDs and MDs as securities analysts and handing
them the biotech beat.) So someone might pipe up, “If your drugs
can be administered by injection only, aren’t you worried about what
48 Science Lessons

will happen when the Big Pharma companies come out with pills to
treat the same medical conditions?”
“No,” Rathmann would reply, ‘ Decause it’s scientifically impossi¬
ble to make a pill that does what recombinant protein drugs can do.”
If pressed, he’d explain that conventional oral medicines rely on small
molecules, which work by blocking biological processes in the body,
such as inflammation. The therapeutic proteins that defined medical
biotechnology for its first twenty years are extremely large molecules.
They exert the opposite effect, stimulating specific biological activi¬
ties. In the introduction, I described how epoetin alfa, Amgen’s ge¬
netically engineered version of the human hormone erythropoietin
(also called EPO), combats anemia by stimulating production of red
blood cells. It takes a large molecule to bind to the erythropoietin re¬
ceptor on the surface of a cell in the bone marrow so that it can signal
the cell to mature into a healthy red cell.
Unfortunately, the digestive juices in our gastrointestinal tract
easily dismantle large proteins, and that is why these agents must be
injected directly into the blood. Small-molecule drugs survive the
tour through the GI system, but no one has ever developed an oral
medication that can activate a receptor.
Did every attendee walk away truly understanding the essence of
biotechnology? No. But to be honest, they were there to assess the
caliber of Amgen’s management team as much as (if not more than)
our science or business plan. Investors have two timeworn sayings:
Bet the jockey, not the horse” and “The ‘A’ management team with
the ‘B’ idea is better than the reverse.”

The Benefit of Our Experience

Much of Amgen’s success in raising capital can be attributed to the

fact that every one of our senior managers had worked for large cor¬
porations. As a result, even when Amgen consisted of only fifty em¬
ployees, we had the organizational discipline of a far bigger company,
 Amgen Goes Public 49

with salary grades, annual performance reviews, monthly reports, and

budgets that were taken seriously All the things that start-ups rarely
do, we did; to us, it was second nature.
What also set Amgen apart from our competitors was George
Rathmann’s background. As Fred Frank points out, “At many biotech
companies, the ideas for the science came from academic institu¬
tions, and the people who founded the company were academics,
with no industry experience. George, on the other hand, had exten¬
sive industry experience from his time at 3M and Abbott Labs, and
that clearly resonated with investors.”
An IPO’s success rides heavily on your poise and credibility
while espousing your company’s virtues at each stop of the road
show, particularly during the question-and-answer session afterward.
So it’s vital that you come well prepared. Shortly before Rathmann
and I flew to our first scheduled meeting, the underwriting syndicate
asked whether we would give our presentation to a group of its peo¬
ple. Once there, we realized that the audience was window dressing;
the bankers had papered the house, as they say on Broadway, just to
see how well we did. Afterward they critiqued everything from our
slides to certain turns of phrase. Underhanded? You bet. But I was
glad they’d tricked us into an impromptu rehearsal.
Bob Hotz encouraged us to sum up Amgen’s story in thirty sec¬
onds, as if we were pitching a movie idea to a Hollywood producer.
We needed something short but compelling that a broker could re¬
peat to his clients. We decided to position ourselves as a second-
generation biotech company committed to improving upon nature.
Let me explain. First-generation companies like Genentech were
trying to clone replicas of existing DNA molecules. One advantage
Amgen had over its competition was that a member of our scientific
advisory board—chemist Marvin Caruthers of the University of
Colorado at Boulder—had been instrumental in devising a revolu¬
tionary technique for producing DNA molecules from scratch, so to
speak. In theory, this process would allow us to go beyond nature and
50 Science Lessons

readily design a better molecule than evolution had created. At the

time, no other biotech firm possessed this capability.

On the Road

The hit-and-run itinerary of a road show is exhausting but exhilarat¬

ing. You’re generally out for several days two weeks in a row and
home on weekends, because nobody turns out for a presentation on
a Saturday or Sunday. Or a Monday or Friday. That leaves Tuesday,
Wednesday, and Thursday—but not later than three or four in the af¬
ternoon. As a result, you’re limited to about two performances a day.
Common sense dictates that your schedule take you in the direc¬
tion of the key city; for us that was New York. Presumably by that
time you’ve sharpened your presentation and know which questions
to expect. It’s the same logic of having a theater company open in
New Haven, Connecticut, to work out the kinks before bringing the
show to the Great White Way.
One of our first stops was a luncheon in San Francisco for three
hundred investors. It goes without saying that when you’re trying to
stir up interest in your company, it’s helpful not to keep would-be
buyers waiting. Around the time we should have been landing, the
pilot announced that San Francisco International Airport was fogged
in, forcing us into a holding pattern.
We still had plenty of time, so none of us was worried. But our
composure changed in a hurry when the pilot’s voice came over the
intercom again to inform us that the plane was running low on fuel
and would have to make an unscheduled pit stop in San Jose, fifty
miles south. “Once we’re back in the air, we’ll still probably have to
keep circling for at least another hour,” he added. “Maybe more.”
It was quickly decided that our group would squeeze into a taxi
to complete the rest of the trip. When we touched down in San Jose,
however, the plane sat on the tarmac, waiting to be refueled. Now
George Rathmann and I began looking at our watches obsessively.
 Amgen Goes Public 51

I approached a flight attendant. “We really need to get off this

plane. Otherwise we re going to be late for a very important meeting.”
“You can’t get off,” she huffed. “There aren’t any stairs.”
Just then the handle of the cabin door began to turn, seemingly
by itself. The door slowly creaked open to reveal an airline mechanic
standing atop a portable stair ramp. He looked as surprised to see us
congregated by the door as we were to see him.
In a scene right out of an old Keystone Kops movie, the four of
us eyed one another and then looked back at the flight attendant.
“Thank you very much!” we chorused. With that, we nudged aside
the puzzled mechanic and sped down the stairs and into the passen¬
ger terminal. An hour and a half later a Yellow cab deposited us at the
door of the hotel with minutes to spare. I don’t remember how
much the taxi fare came to, but whatever it was, it was a bargain.
Thank goodness we’d listened to our underwriter’s strict orders to
pack lightly and bring only carry-on luggage. Never check your bag;
the schedules are too tight to have to deal with a lost suitcase.
In all, we visited six cities. While you’re center stage, the invest¬
ment bank’s sales force buzzes about the room gauging the level of
interest in the offering. No hard sell is allowed: securities law pro¬
hibits stockbrokers from accepting orders, and investors’ verbal indi¬
cations of interest are nonbinding. By the end of the tour, though,
the underwriters should have a pretty accurate reading of the num¬
ber of shares they can expect to sell and at what price.

HOW MANY SHARES AND FOR HOW MUCH?

According to Bob Hotz, the demand for Amgen’s forthcoming IPO

appeared extremely strong. “We’re oversubscribed. We could either
sell more shares or sell the same number of shares at a higher price.
Which option do you want us to do?”
“Whichever will get us the most money,” I said.
“Then sell more shares.”
52 Science Lessons

End of discussion. Most companies would have preferred the

other, less diluting route, but we didn’t know when the next oppor¬
tunity to raise funds would come klong.
Even with the public offering days away, you still don’t know the
exact selling price per share. On the registration statement to the
SEC, your underwriter provides a ballpark figure—say, $14 to $16—
based primarily on the company’s value. However, because biotech
firms were expected to dwell in the red for years, traditional formu¬
las for estimating worth, such as book value, were obsolete. At the
time of Amgen’s IPO, says Fred Frank, “there were very few analysts
and no precise, well-accepted models of valuation. In fact, they could
be rather idiotic. Some early analysts valued companies at $1 million
per PhD; others based their valuations on $1 million per employee. It
took a while before they got down to discounted cash-flow models,”
a perennial Wall Street favorite.
For a major investment firm, the ideal starting point falls between
$10 and $25 per share, with a stock split—or, more commonly, a re¬
verse stock split (explained in a moment)—-undertaken shortly be¬
fore the offering. In the end, however, a security is worth only what
people will pay for it. Therefore, investor feedback figures promi¬
nently in determining the final price, which is fine-tuned based on
prevailing market conditions as well as the sometimes irrational buy¬
ing practices peculiar to Wall Street.
For example, there’s an unspoken maxim that many investors
won’t buy fewer than one hundred shares of a stock. If you set the
share price too high, they may consider the stock overpriced and by¬
pass your offering. But if you set it too low—$10 being the Mason-
Dixon line of respectability—your offering risks being perceived as a
second-rate investment. You know how sometimes you’ll be shop¬
ping, and you find the item you’ve been looking for at a dirt-cheap
price? At first you’re delighted: what a bargain! Then a voice of doubt
starts whispering, If it’s so inexpensive, the quality must not be very good.
The same outlook may dissuade investors from purchasing a stock.
 Amgen Goes Public 53

Therefore, a company seeking to reap $25 million from its IPO

would be better off selling 1.67 million shares at $15 each than 2.5
million shares at $10. With a $10 IPO, if the stock price drops even
slightly, it’s now in single digits and treading perilously close to the
$5 minimum for maintaining respectability.
According to Fred Frank, the Sigmund Freud of investor psy¬
chology, a successful public offering is usually oversubscribed by a
ratio of about 3 to 1. He explains: “The way it works is that if an in¬
stitutional investor calls me up and says he is interested in one hun¬
dred thousand shares, and I agree to sell him that amount, his
reaction is likely to be, ‘You must not have much demand if you’re
going to give me everything that I want!’As the underwriter, I want
to allocate to all interested parties a modest proportion of what
they’ve asked for.”The hope is that they’ll satisfy their hunger by fill¬
ing up on additional shares after the IPO, and at a richer price.

BACK AT THE RANCH

An investment firm typically trims 10 percent to 20 percent off the

stock’s estimated value to provide an incentive for institutional buyers
and to reward investors with a quick if modest appreciation. In addition,
underpricing, or “leaving something on the table,” serves as a buffer
against a weak post-IPO market and helps protect the underwriter in
case a dissatisfied investor complains that the issue was overpriced.
Amgen’s securities would have sold at $9 per share, which was
lower than we’d wanted. To double the price, we executed a reverse
stock split: an employee with options for one thousand shares would
now be entitled to five hundred shares. Some of our scientists grum¬
bled that they must have been cheated somehow; they didn’t like the
idea that their number of shares had been reduced, even though the
monetary value remained exactly the same. But overall, the atmos¬
phere in Thousand Oaks was surprisingly calm during the weeks
leading up to Friday,June 17,1983.
54 Science Lessons

Throughout the winter and spring, we’d regularly kept Amgen’s

people—now numbering one hundred or so—apprised of where
the company stood financially and the status of the IPO. Researcher
Kirby Alton recalls, “Everyone knew that Amgen was running out of
money, but I guess we were all young and naive enough to assume
that it would all work out. So we just kept pressing on, which, when
you think about it, was an amazing thing. It’s really a credit to senior
management that they were able to keep everyone focused on the
work while all of this behind-the-scenes activity was going on.”

THE BIG DAY

In a firm-commitment agreement, whereby the underwriting syndicate

buys all available shares regardless of its ability to sell them to investors,
the effective date of the public offering is more or less anticlimactic.
We knew going in that Amgen would be issuing 2.35 million shares
at $18, for a total of $42.3 million in proceeds. Nevertheless, that Fri¬
day contained more drama than an episode of TV’s Dallas.
I’d asked Smith Barney whether I could be present on the trading
floor of its New York headquarters throughout the offering. I had no
ulterior motive; I just thought it would be an interesting experience.
Was it ever.
At eight in the morning, an hour before NASDAQ’s opening bell
sounded, sell orders started trickling in from some major institu¬
tional investors. Possibly they’d felt that the stock wasn’t going to go
up on its first [Link] protect themselves against a short-term loss—it
couldn’t get any shorter—they could unload it to the syndicate,
which was obligated to buy back the shares.
On Wall Street, word spreads faster than a White House leak.
After a few more calls like that, tempers started flaring, because the
traders could see that we were off to a rocky start. “Who’s the son-
of-a-bitch salesman that sold these shares to this guy?” one trader
 Amgen Goes Public 55

railed.“Make sure he doesn’t get a commission on ’em!”By the time

the market opened, sell orders were pouring in. Meanwhile, I was
sitting there, watching the commotion around me and thinking,
“What the hell happened? This whole thing is a disaster!”
Far from it, actually. On the negative side, the price per share
dipped to $16.75 by the end of the day. Our syndicate struggled to
maintain the offering price, but it proved to be impossible. At one
point, though, the price had plummeted to less than $10, so we had
to be thankful that it recovered in the afternoon. Normally, investors
hope to see a gain of approximately 15 percent on the first day of
trading; needless to say, we disappointed them.
Still, we achieved what we’d set out to do: put Amgen back on
solid ground financially for some time to come. Our $42.3 million in
proceeds might seem like small potatoes compared with the sums
raised in the heady 1990s, when the average IPO fetched $67 million
and made instant millionaires out of mailroom clerks (or so the
breathless media coverage had us believe). At the time, however, it
ranked as the industry’s third-highest public offering.
I flew home to California satisfied with how everything had
turned out. The accomplishment would shine brighter still in a year
or so, when it became clear that we’d slipped under a closing IPO
window by a whisker. That summer, public offerings by Immunex (a
biotech firm that Amgen acquired many years later) and Chiron de¬
livered only $16.5 million and $17 million, respectively, while in
November the Immunomedics company’s paltry $2.5 million IPO
signaled the end. The window would not budge open again for two
and a half years.
Oh yes: Remember those five companies whose public offerings
were already in progress when we were just getting started on ours?
Amgen beat them all. In fact, a couple of the procrastinators didn’t
make it to Wall Street at all before the window came down like a
guillotine, cutting them off from millions of dollars in financing.
.

'
CHAPTER
3
The Business of Science Meets
the Science of Business

with amgen’s finances stabilized for the foreseeable future, we

returned our attention to the business of science, minus the distrac¬
tions of the past several months.
When you have an organization staffed almost exclusively by sci¬
entists, as Amgen was for its first seven years, perhaps it’s only natural
that its business approach would be unorthodox. It’s fair to say that at
many companies, if not most, sales and marketing dominate corpo¬
rate strategizing; the scientific or creative end may be behind the
wheel, but ultimately the sales-and-marketing people commandeer
the road map, barking out directions from the passenger seat.
Not so in the field of biotechnology—and certainly not at Amgen,
where even the company’s location was chosen in part to attract first-
rate scientists. Our Thousand Oaks headquarters sat more or less
equidistant from the three principal research centers in Southern
58 Science Lessons

California: the University of California at Los Angeles (UCLA), the

University of California at Santa Barbara (UCSB), and the Cahfornia
Institute of Technology (CalTech^, in Pasadena.
The scientific mind-set permeated Amgen’s operations to create
a highly efficient science-based business model. Many companies
now incorporate elements of science in the workplace, although
they may not be conscious of it. What is test marketing if not an ex¬
periment conducted to gauge, say, consumers’ response to a planned
product? Rarely, though, do businesses rely on science to help shape
decision-making procedures and internal policies. We did. I’ll give
you an example of what I mean.
This is jumping ahead a bit, but in 1992, my fourth year as CEO,
Amgen was struck by twin tragedies. First, our manager of sales and
marketing, Paul Dawson, died of a heart attack. Immensely talented
and popular throughout the company, Paul had been hired as our
first vice president of sales and marketing in 1987, when Epogen
looked to be a year away from winning FDA approval. He built
Amgen’s sales and marketing from the foundation up. Even more
impressive:You know how some managers shine during a business’s
upward climb but seem to lose their footing once large-scale success
arrives, bringing with it exponential growth and change? Paul kept
pace the whole time, and we had expected him to lead Amgen sales
and marketing for years to come.
We were all still reeling from the shocking news when Thomas
Beard, vice president of sales, also suffered a heart attack. He sur¬
vived, thankfully, but his health was impaired to the degree that he
could not continue in such a demanding position.
We could have tapped an outsider to succeed Paul Dawson, but
that would have required months of searching. What’s more, Amgen
was in a crucial period, having only recently launched its second
product, filgrastim (Neupogen); we couldn’t chance hiring an un¬
known and then discover that the person wasn’t going to work out.
 The Business of Science Meets the Science of Business 59

Fortunately, one of the best pure managers I’d ever known had
been with the company since 1981. He was a scientist, though, and
hadn’t shown any interest in business, much less sales and marketing.
Dennis Fenton had come to Amgen as a thirty-year-old bench re¬
searcher. A refugee from conventional pharma, he’d spent four frus¬
trating years at Pfizer, in sleepy Groton, Connecticut (“Rotten
Groton,’’ he jokingly called it)—not the most exciting environment
for an avid surfer with a budding interest in genetic engineering.
“Pfizer was going in that direction at a glacial pace,” Fenton re¬
calls. “One day I saw a tiny ad in the Simi Valley News: ‘Applied Mo¬
lecular Genetics Seeks Scientists.’ Pretty general. I thought, ‘Hmmm,
I might fall into that category.’The address was in Newbury Park,
California; I’d once eaten at a famous hamburger joint there called
Du-Par’s while on vacation. That was enough to encourage me to
send in my resume. And lo and behold, George Rathmann invited
me out for an interview and hired me.”
At the time of Paul Dawson’s death, Dennis was serving as vice
president of operations in charge of process development, as well as
building and overseeing our manufacturing [Link] thought of
entering sales and marketing had never grazed his mind. That would
explain his dumbfounded expression the day I called him into my
office and asked, “Dennis, have you ever considered the possibility of
running sales and marketing?”
“Gordon?” he said, waving at me as if to jolt me back to reality.
“It’s me—[Link]’re talking to me about sales and marketing.”
“I know.”
He might not have envisioned himself in that capacity, but I be¬
lieved he was exactly what Amgen needed, for a couple of reasons.
First, although the sales and marketing department was in good
shape in many ways, it was not well organized. I knew that Dennis
would quickly tighten things up and get everyone singing from the
same page. He would promote the right people and transfer others,
60 Science Lessons

establish more internal structure, devise plans and teach people to

follow them—in short, create a smooth-running [Link],
after a couple of years, he could ^return to operations and leave a
well-disciplined sales and marketing department as his legacy.
Thinking back to that afternoon in my office, Dennis says, “Gor¬
don proceeded to sell me on the idea. He spoke enthusiastically
about how much I could learn, and how it would be fun, exciting,
different. I didn’t have any doubts about that. My big fear was that I
would fail and wind up having to leave this company that I loved.”
I sensed his misgivings. “You’re not going to fail,” I told him. “I
wouldn’t offer you the job if you weren’t able to do [Link] me; just
try it. If it turns out to be a mistake, it’s my decision, my fault, not
[Link]’ll go back to your old job or another one within the com¬
pany.” Regarding his other major concern about switching to sales—
“But Gordon, I don’t even own a suit!”—that could be rectified
easily enough.
Dennis needed reassurance that he would succeed. Who doesn’t
from time to time? Although assuming the reins of sales and market¬
ing was “the single hardest thing I ever did in my life,” Dennis says,
he shepherded Amgen through the crisis brilliantly. Being a PhD in
microbiology, he also implemented the scientific approach to prob¬
lem solving, as might be expected.
He started by speaking at length with the staff—listening more
than talking, as a good manager should—soliciting their opinions
about how the existing systems might be improved.
I should explain here that pharmaceutical sales isn’t like hawking
vacuum cleaners door to door. Our clients are physicians, and a fast-
talking salesperson with a slick sales pitch would do more harm than
good. What doctors respond to is accurate information anchored in
science, and hold the hype. Because they’re busy, they seldom have
time to read the latest studies in medical journals. Amgen’s sales and
marketing people may not be MDs or PhDs, but they know their
 The Business of Science Meets the Science of Business 61

product inside and out; more like educators than salespeople, their
primary function is to teach health-care professionals how to use the
company’s medications most effectively.
As reps of a two-product company, our salespeople split their
time between visiting dialysis facilities to talk to nephrologists (doc¬
tors who specialize in diagnosing and treating kidney disorders)
about Epogen and then dropping in at oncologists’ offices to discuss
new developments regarding Neupogen in cancer [Link]’s a
tremendous amount of highly technical information to absorb. One
of the biggest challenges, according to the staff, was determining
how much time to devote to each product.
One day at a sales and marketing meeting, Dennis posed this
question: “Have you ever considered dividing the sales force so that
each of you can focus on one product?”
From the incredulous response, you would have thought he’d
proposed that from now on all Amgen reps rollerblade to their ap¬
pointed rounds.
“No, never!”
“It’s a bad idea.”
“The salesmen wouldn’t go for it.”
“Other companies don’t do that.”
“This way is more fun.”
“That might be true,” Denms replied evenly. “But we’re a science-
based company. Why not run an experiment? Let’s take two districts
at random and have each of their people sell only one product for a
while—say, six months. Then we’ll see which way works best. If it
doesn’t pan out, we’ll go back to the way things were. But at least
we’ll know the true answer instead of speculating.”
Everyone agreed, even though the one-product-per-salesperson
approach flew in the face of conventional wisdom. (The industry
standard was three to four products per rep.) The results of Fenton’s
experiment surprised everyone. Our salespeople discovered that they
62 Science Lessons

preferred representing a single drug after all. Not only that, but they
were more effective, and that led us to reconfigure the entire national
sales force.

THE SCIENTIFIC METHOD

Designing an experiment to help resolve a business problem is sim¬

ple, at least compared with structuring a clinical trial for testing a
new drug’s safety and effectiveness. But how frequently do compa¬
nies do it? Not very often. They may try out new ideas, but few actu¬
ally conduct formal experiments in the way that scientists would:

Step 1: define the question you hope to answer. As DanVapnek,

Amgen’s longtime head of research, used to say,“If we knew the
results we would get from an experiment, we wouldn’t run the
experiment.” That’s what an experiment is: the creation of a
controlled environment in which to evaluate a hypothesis. Our
question was straightforward: will the sales force feel more com¬
fortable and be more productive representing one product rather
than two?

Step 2: determine the particulars of conducting the experiment. As a

general guideline, the larger the sample size (the number of par¬
ticipants), the greater the accuracy, but the smaller the sample
size, the lower the cost. In testing new medications, for example,
studies of fewer than twenty patients are considered preliminary
or suggestive at best.
Our informal sales study was analogous to an investigational
drug trial in that it compared the current standard (having all
salespeople promote both Epogen and Neupogen) with the
untested method (having them promote one drug).The first
group is referred to as the control group, while the volunteers as¬
signed to the experimental arm belong to the study group.
 The Business of Science Meets the Science of Business 63

The trickiest part of designing any study is to minimize fac¬

tors that could cloud the issue. Although you can never entirely
eliminate bias, assigning participants to either the study group or
the control group through a random process—such as a lottery
system or picking names out of a hat—reduces the chance that
the study’s coordinators might subconsciously create groups that
are not evenly matched.
In addition to randomizing subjects, the other major consid¬
eration for ensuring an objective measurement is to even the
playing field as much as possible. If, let’s say, a clinical trial pits a
new chemotherapy drug for early-stage colorectal cancer against
the established intervention, both groups should have equal num¬
bers of patients with similar characteristics. The researchers would
want to enroll men and women with the same severity, or stage,
of disease; the same treatment history (surgery to remove the
tumor and no previous history of chemotherapy); and the same
current protocol (postsurgical administration of the experimental
or standard drug regimen; no radiation therapy). In this way, if a
significant benefit emerges, you can be reasonably confident that
it is not a mirage but is attributable to the investigational agent—
the only major difference between the two groups. For our ex¬
periment, it was essential that the drug reps in each group had
similar numbers of clients and similarly sized territories.

Step 3: Determine how you will measure the results. For an informal
study such as this one, you might use the following tools:

• A simple survey rating the sales staff’s satisfaction on a scale

of 1 to 5

• A comparison of actual sales versus sales from the prior

period

• A comparison of actual sales with the sales quotas

64 Science Lessons

LISTEN TO YOUR DATA

After you’ve compiled and analyzed the evidence, heed what it tells
[Link] might be difficult if you consider yourself an intuitive busi¬
ness executive and prefer to go with your gut. Maybe all this reliance
on data for making decisions sounds duller than sitting next to the
boss at the annual managers’ weekend retreat. Maybe you think it
would stunt creativity and the open exchange of ideas.
To the contrary: the scientific approach fosters ffee thinking.
When you allow the results of a quality study to guide decision mak¬
ing, it depersonalizes the issue, taking it out of the realm of “good
idea,” “bad idea,” and, most poisonously, “his idea” or “her idea.”
Look at it this way: the idea won enough support to warrant being
put to the test. It may still be valid—or not. The important thing is
that you’re evaluating the idea and not the person who proposed it.
You tried the experiment, and it didn’t work; now discuss what the
results can teach you and move forward.
As Dennis Fenton points out, “Most companies run away from
failed ideas, as though they exude a stench. If you’re science based,
analyze what went wrong. Was the design of the experiment flawed?
Did the experiment prove that the hypothesis was correct, or incor¬
rect, or did it leave us without an answer? Another question to ask is.
Was the experiment conducted properly? Regardless of the results,
your company will get progressively smarter if it uses the scientific
method.”
When ideas become too closely associated with individuals, peo¬
ple start digging in their heels. In reality, they’re no longer defending
their idea; they’re defending themselves and perhaps defending or
attacking others. Using the scientific method makes it easier to de¬
couple what’s best for the organization from what’s best for the in¬
dividual. Another benefit: experimentation allows you to arrive at
decisions based on a sample. To test Dennis’s hypothesis, Amgen
didn’t have to overhaul its entire sales force—only two districts.
 The Business of Science Meets the Science of Business 65

If you act on the results of your experiment, you’ll make the right
decision far more often than not. Doesn’t every company do that?
No. How many times have you seen someone embrace data that sup¬
ports a preconceived conclusion while ignoring data that contradicts
it? If too many people in an organization become overly invested in
their own ideas, the company will have fewer ideas to consider, and it
will take forever to evaluate the ones that are put on the table.

SCIENCE DRIVES PRODUCTION

That Amgen’s culture was science driven from the top down was
most apparent in the resolve to let research and development dictate
our product [Link] runs counter to the way things are done at con¬
ventional pharmaceutical companies. Typically, a desirable patient
population is [Link] the firm’s scientists are set out like dogs
on a scent to see whether they can dig up a new way to treat the con¬
dition. It doesn’t have to be groundbreaking, only enough of an ad¬
vancement to muscle its way into, say, the $1 billion migraine drug
market or the $3.5 billion osteoarthritis drug [Link] notable ex¬
ceptions are 3M and Abbott Laboratories, where, not coincidentally,
George Rathmann had spent his career before coming to Amgen.
Biotechnology firms were trying to develop landmark drugs that
either were the first of their kind, like Epogen and Neupogen, or that
marked a significant advance over what was available. Coming up
with a slightly better product and trying to win the competitive bat¬
tle through superior sales and marketing might have been acceptable
for big pharmaceutical companies, with their large and highly effec¬
tive sales forces, but any biotech firm following that strategy was cer¬
tain to fail.
To develop revolutionary drugs, we had to go wherever the best
science led us, whether the destination happened to be a huge market
or one of more modest [Link] entire biotech industry operated on a
measure of faith that recombinant drugs would mark an unparalleled
66 Science Lessons

advance in medicine and would benefit from strong patent protec¬

tions, so that even if the numbers were small there would be scant
competition. We could become the large (genetically engineered)
fish in the small pond and, with sohie luck, perhaps the largest fish in
the ocean.
For example, Epogen’s original consumer base was composed of
men and women forced onto kidney dialysis by end-stage renal dis¬
ease (ESRD), a patient population that barely exceeded one hundred
thousand the year the drug entered the market. At the time, nephrol¬
ogists had nothing to offer the nine in ten dialysis patients who suf¬
fered from chronic anemia so severe that many of them didn’t have
the energy to get out of bed. Therefore, you can imagine the de¬
mand for a drug that restored their red-blood-cell counts to near
normal. Since that time, the number of people on hemodialysis has
more than quadrupled; in addition, Epogen has acquired FDA ap¬
proval for other uses, including chronic anemia brought on by AIDS,
cancer, and cancer treatment.
If not for the passage of the Orphan DrugAct of 1983,it’s question¬
able whether the pharmaceutical industry could afford to serve small
patient populations such as the ESRD community. The cost of devel¬
oping a new drug is the same whether the number of potential users
totals in the millions or only a few thousand. Coming up with treat¬
ments for rare disorders is almost always a money-losing proposition.
In the United States, the drug manufacturers themselves foot
most of the bill for medical research. Public law 97-414, spearheaded
by Rep. Henry Waxman (D-CA), offered seven years’ patent exclu¬
sivity to drugmakers that were willing to invest in discovering new
treatments for any of the six thousand or so orphan diseases: those
that affect fewer than two hundred thousand people.
This seven-year exclusivity took effect the day the drug was li¬
censed for commercial use. Non-orphan medications are limited to
twenty years of patent protection from the day the patent application
 The Business of Science Meets the Science of Business 67

is filed. (Until 1995, the term of a patent was seventeen years from the
date it was issued.) But it can easily take ten years or more for a prod¬
uct to surmount all the hurdles on the road to FDA approval, so in
reality the standard patent protection often turned out to be about
the same as the seven years provided by the Orphan Drug Act.
The bill quickly achieved its intended effect: from 1983 to 2007
more than 280 conventional drugs and biologies for orphan diseases
have been introduced as opposed to only 10 in the previous decade.
Japan admired the legislation so much that it adopted its own, but for
10 years.

AMGEN’S FOUR-PRONG STRATEGY

FOR MANAGING RISK

Epogen was one of several products on the drawing board at the time
of Amgen’s IPO. From the start, George Rathmann had wisely sent
his researchers scouting in four directions: human health care, human
diagnostic testing, animal health care, and specialty chemicals. Initially
the four-prong strategy didn’t sit well with some board members, in
particular Jim Blair, who wanted the company to be more market
driven and to concentrate on a single product or a few products at
most. Blair’s objection to building a multifaceted product portfolio
was difficult to understand, because Genentech, Biogen, and most of
our other chief competitors had adopted that strategy. Tensions some¬
times flared at board meetings, where for three years George was
barred from stating his opinion that Amgen should aspire to be more
like a pharmaceutical company than an R&D boutique.
Not many people understood what we and the other biotech
firms were doing. With no existing data to dictate which areas we
should pursue, it made sense to dabble in each and decide which
ones looked the most promising after some facts were known. We
understood from the outset that we couldn’t afford to spend too
68 Science Lessons

much time and money on multiple areas; those that didn’t pan out
would be dropped early. Our R&D decisions would be based on our
own scientific evidence. V
Ultimately the tug-of-war went Rathmann’s way, as it should
have all along. There was never any disagreement that our long-term
success hinged on developing drugs designed to treat people. But be¬
cause income from this area was a good decade away, clearly we
needed to pursue other avenues of revenue.
Most of them, such as animal therapeutics, turned out to be dead
ends. That one surprised many of us. The entire biotech industry had
been banking on genetically engineered animal medications as quick
moneymakers, on the assumption that the government approval
process would be far shorter than it was for human drugs. What no
one took into consideration was the endless red tape involved in
having to obtain a license not only from the FDA but also from the
U.S. Department of Agriculture (USDA). More than a dozen human
products reached the marketplace before the first biotech animal
drug was approved.
Amgen’s forays into this field yielded little except three great
yarns.
Arbor Acres, a poultry breeder, invested in our chicken growth
hormone [Link] study had been farmed out, so to speak, to local
chicken growers. Although virtually everyone believed that the sub¬
stance would make the birds grow faster, it didn’t. Our disappoint¬
ment was compounded by the fact that one of the farmers died of a
heart attack while feeding the fowl right there in the farmyard, and
apparently his body went undiscovered for two days. Naturally the
National Enquirer couldn’t pass up the story: “Giant Genetically Engi¬
neered Chickens Peck Farmer to Death!” trumpeted its headline. Not
exactly the kind of publicity that a company goes chasing after.
A few years later, the SmithKline pharmaceutical company
funded work on a porcine growth hormone that it hoped would
produce leaner pigs. It did. Scientist Burt Ensley, another of Amgen’s
 The Business of Science Meets the Science of Business 69

earliest hires, recalls, “The animals had almost no fat on them. But
because of that, the concern was that they probably wouldn’t taste as
good. So, to find out, we ate the experiment.”
Befitting a company of scientists, we turned one of our Friday
barbecues into a randomized single-blinded study, better known as a
taste test. People’s plates were piled high with either porcine somat-
ropin (PST) center-cut pork chops or “control” pork chops, although
no one knew which was which. This being a tightly controlled trial,
each piece was cooked to the same internal temperature and cut into
similarly sized pieces.
After dinner, everyone filled out a detailed survey pertaining to
taste, juiciness, texture, and so on. The ultralean meat compared fa¬
vorably with the ordinary fatty pork in every category. Unfortu¬
nately, the growth hormone turned out to be a pig in a poke,
because 2 percent of the animals developed peptic ulcers. That mor¬
tality rate was considered too high.
In addition, we were unable to develop a long-term sustained-
release version of the hormone, which would have required only a
single treatment every thirty days. Farmers weren’t interested in a
product that had to be administered more frequently than that. And
so the project had to be scuttled. (Strangely enough, the porcine
growth hormone did make salmon grow faster.)
Not long afterward, Amgen decided to close the barnyard door
on animal therapeutics. The only salable product to emerge from the
program was a bovine growth hormone, licensed to the Upjohn
Company, that made cows produce more milk. We decided to show¬
case the success in our annual report and put a photograph of a beau¬
tiful dairy cow on the cover.
There weren’t any cows to be found within a country mile of
Thousand Oaks, so we arranged to fly to Kalamazoo, Michigan, and
take a picture of one of Upjohn s animals. In preparation for the shoot,
the Upjohn people cleaned their dairy facility until it was as spotless as
an operating [Link] was going to be a great photograph.
70 Science Lessons

When it was time to bring out the star, however, the poor animal
slipped on the wet floor, landing spread-eagled. She lay there for
thirty minutes while Upjohn employees pushed and tugged. They
were visibly embarrassed; we bit our cheeks not to laugh out loud.
The annual report cover was not graced by a dairy cow after all. Nor
did our partnership with Upjohn ever get on its feet. For some rea¬
son, the pharmaceutical giant decided against selling the bovine
growth hormone. The product is on the market but is sold by an¬
other company.

OUR FORAY INTO SPECIALTY CHEMICALS

In the specialty-chemicals field, Burt Ensley could claim responsibil¬

ity for one of our early successes. As sometimes happens in science, it
was the result of a happy accident. “You can use bacteria to make a
number of products in the laboratory,“ he explains. “For instance,
benzene is a fairly nasty chemical.” Carcinogenic, too. “But exposing
it to bacteria can turn it into lots of interesting things, like catechol, a
substance used in photo developer, dyes, pigments, and tanning
agents.”
Ensley notes that one product his group was working on was
naphthalene. “There’s a bacterium, Pseudomonas, that grows on naph¬
thalene and causes it to decompose into something useful called
alpha-naphthol. But the bacterium wouldn’t quit there; it would
proceed to eat the alpha-naphthol. We wanted to try stopping it.
How to do that was not well known back then. We extracted the
genes from this bacteria and cloned them in a sample of E. coli bac¬
teria. When we did that, the bacteria started turning bright blue.
Now, we had absolutely no idea why this should have happened.”
Eventually another scientist guessed that perhaps we were inad¬
vertently making indigo, a plant-derived dye that is used to color
blue jeans. That turned out to be right. When we compared the
 The Business of Science Meets the Science of Business 71

product to a small amount of indigo, we discovered that the two ma¬

terials were identical. Apparently the bacteria were converting some¬
thing in the E. coli into indigo.
“So, voila/” says Ensley. “Just like that, we were out of the naph-
thol business and into the indigo business.” His team’s unexpected
discovery landed on the front page of the prestigious international
journal Science.
What made for intriguing science did not translate into a profit,
however, but not because there wasn’t interest in a substitute for the
dye. Ultimately, we learned that there were only three major manu¬
facturers of indigo in the world: one in the United States, one in Eu¬
rope, and one in Japan. They didn’t seem to compete much in one
another’s home market, and although the selling price was $12 per
pound, our analysis showed that each of the three might have manu¬
facturing costs as little as $4 per pound. Given our process, we couldn’t
hope to go that low. We suspected that if we ever became a serious
competitor, the $12 price would drop very far, very fast.
The lone financial success from specialty chemicals came in laun¬
dry detergents, of all things. Procter & Gamble, Lever Brothers, and
their competitors were concerned that the stain-removing enzymes
they advertised prominently tended to degrade in less than a month.
Amgen scientists fashioned a biotech-based enzyme that was hardy
enough to stand up to the harsh environment of a liquid detergent.
Genentech also came up with a suitable enzyme.
Detergent enzymes couldn’t create huge opportunities for
biotech companies, however, and we didn’t have any other product
candidates. Genentech off-loaded its enzyme to Corning, while we
licensed ours to Kodak, which had been bankrolling the specialty-
chemicals program. And with that, the biotech industry acknowl¬
edged another blind alley. At least the diagnostics program, funded
by Abbott, was chugging along quietly on the strength of a few lab¬
oratory test kits.
72 Science Lessons

AN EXCEPTION TO THE RULE

The only program in which Amg^en put a substantial amount of its

own money was the big one—human therapeutics—where we had
five contenders: recombinant alpha-interferon and gamma-interferon,
a hepatitis C vaccine, a skin-growth factor (epidermal growth factor,
or EGF), and erythropoietin. None of the five was a sure thing, but
you’d have to name erythropoietin the runt of the litter. In the sum¬
mer of 1983, nearly two years after scientist Fu-Kuen Lin and his
small team began working on the project, they still had not been able
to identify the gene responsible for the hormone that stimulated red-
blood-cell production. Like every other biotech enterprise, we were
desperate for one product to shine so brightly in the laboratory that it
would support the entire company. Adding to the pressure in the case
of erythropoietin, other firms were hot on our heels. In a race like
this, where finishing first brings exclusive patent protection, there is
no such thing as second place. It’s all or nothing.
By now, you probably have a good sense of how devoted George
Rathmann was to his scientists. But he also knew that scientists, with
their innate enthusiasm and their sometimes obsessive nature, would
go on trying to develop a product for eternity unless a time limit was
imposed. What made Amgen unique among biotech companies was
that projects were expected to show some progress within a pre¬
scribed time, or else they had to be reevaluated. In the aftermath of
our successful IPO, it looked as if erythropoietin might have reached
the end of the line.
Based on the evidence, terminating the project increasingly ap¬
peared to be the logical decision. For all the efforts to prevent infor¬
mation leaks, word had spread throughout the industry that no
company was close to pinpointing the gene. Perhaps it couldn’t be
done; after all, scientists had been searching in vain for EPO since the
1930s. When one of our competitors bailed out to preserve its lim-
 The Business of Science Meets the Science of Business 73

ited resources, Dan Vapnek and Noel Stebbing strongly recom¬

mended that we do the same.
George wouldn’t hear of it. “We can’t quit,” he said adamantly.
“These scientists have been working on this project day and night,
seven days a week for more than two years! They still believe they
can do it. We can’t just pull the rug out from under a team of people
who are that positive and who are working that hard. We have to as¬
sume that they know what they’re doing and let ’em keep going.”
From a scientific standpoint, Vapnek and Stebbing were absolutely
correct. However, George astutely judged that despite Amgen’s dedi¬
cation to data-driven decision making, this was one instance when
the people factor took precedence. It also illustrated an important
Amgen principle: “Every rule, policy, and procedure has exceptions.”
CHAPTER

From IPO to EPO

in Four Months

fu-kuen LIN, the son of an herbal-medicine doctor, emigrated

from his native Taiwan to the United States in 1967 at the age of
twenty-six to study plant pathology. After he graduated from the
University of Illinois in 1971, his postdoctoral work took him on a
winding course through the Midwest, back to Taiwan for a couple of
years, and then through the southern United States. While at the
Medical University of South Carolina, he began conducting experi¬
ments in genetic engineering.
In 1981 Lin answered the same help-wanted ad as Dennis Fen¬
ton—this one appearing in the pages of Science magazine—and
became Amgen’s seventh scientist. He was offered a choice of assign¬
ments. It is a testament to his adventurous nature that the soft-
spoken biologist asked to work on cloning the elusive erythropoietin
gene. Of our five human-therapeutic projects then under way, EPO
76 Science Lessons

was the only one for which the difficult step of identifying the gene
had yet to be accomplished.
At least six other companies, including Biogen and Genetics In¬
stitute (GI), were in pursuit of the^erythropoietin gene, one of the
Holy Grails of the nascent biotechnology industry. Amgen held a
slight edge in that sixty-year-old Eugene Goldwasser had agreed to
serve as a consultant.
Goldwasser, a biologist at the University of Chicago, discovered
the hormones existence in 1977 after more than twenty often frus¬
trating years of research. As far back as 1906, scientists had suspected
that a substance in the blood instructed the bone marrow to churn
out replacements for expired red blood cells at two million to three
million per second. But with more than two hundred proteins in the
circulation, no one could find the biochemical messenger named
after erythropoiesis, the Latin word for red-cell production.
In the late 1950s, Goldwasser and his team of researchers deter¬
mined that the kidneys were the body’s source of [Link] figured
this out through process of elimination, surgically removing different
organs from laboratory rats and waiting to see whether their red-cell
counts dropped precipitously. As suspected, the animals whose kid¬
neys were removed developed severe anemia.
The Brooklyn-born Goldwasser speculated correctly that if you
induced anemia in an animal by injecting it with an agent that de¬
stroyed red blood cells, the kidneys would attempt to correct the im¬
balance by secreting additional EPO into the circulation. Where he
erred was in assuming that a large surplus of erythropoietin would
turn up in the blood, which hopefully could then be isolated and in¬
jected into anemic rats as a cure. The “enriched” blood samples he
withdrew contained only trace amounts of the hormone, too little to
be isolated using the available technologies and equipment.
Goldwasser might have remained stranded had not a Japanese re¬
searcher read a paper Goldwasser had published. By this time, 1973,
the growing suspicion was that excess EPO showed up more in
 From IPO to EPO in Four Months 77

urine than in blood. Certainly urine contained smaller amounts of

other proteins, and that would make separating EPO easier. The re¬
searcher, [Link] Miyake, offered to collect urine samples from pa¬
tients who suffered from a rare blood disorder called aplastic anemia.
The idea was that their defective marrow produced insufficient
numbers of red cells; consequently, their kidneys produced additional
EPO, thereby increasing the level in their urine.
Eighteen months later, Miyake arrived in Chicago bearing a
shoebox-sized package of dried urine. Goldwasser was able to cull
from it a small amount of purified erythropoietin. When the speci¬
men was injected into the anemic lab rats, it stimulated their bone
marrow to produce red cells, just as Goldwasser had theorized twenty
years before.
A little EPO goes a long way in a mouse; human kidneys produce
the hormone in minuscule amounts. It took 675 gallons of urine,
courtesy of Miyake, to yield a mere 8 milligrams of pure erythropoi¬
etin. Clearly, not enough EPO could be extracted from urine sam¬
ples. Goldwasser briefly considered cultivating kidney cells and
trying to manipulate them in the laboratory to generate the hor¬
mone, but despite some initial promise, nothing came of this process.
He knew he didn’t have the resources to pursue the next major
step on his own and started looking for a pharmaceutical company to
partner with. In retrospect, it’s hard to believe that Goldwasser was un¬
able to interest Parke-Davis in his research; and Abbott Laboratories,
where George Rathmann worked (but in diagnostics, not therapeu¬
tics), rejected the biologist’s solicitations more than once. When
Goldwasser told his employer, the University of Chicago, about his
patentable discovery, the administrators there didn’t bother to return
his phone call—shades of the Mayo Clinic’s failure to patent cortisone
in the 1940s and subsequently losing millions in income.
Biogen approached Goldwasser before Amgen did. He turned
down Biogen, unimpressed with the caliber of the company and its
scientific advisory board. You have to wonder why, because most
78 Science Lessons

observers at the time would have rated Biogen one of the industry’s
top firms. In any event, Amgen and its star-studded scientific advi¬
sory board convinced him to cast his lot with us. That was an early
example of how recruiting the be$t people brings rewards in addi¬
tion to its obvious advantages. It can help do the following:

• Draw to the organization other talented professionals who

want to play on a winning team

• Attract desirable licensing opportunities

• Interest the most successful companies in forming business

relationships

Certainly Amgen’s people were one of the keys to its sustained

success over the years.

A NEW OPTION: GENETIC ENGINEERING

By the dawn of the 1980s, a new possibility had opened up for mak¬
ing EPO a practical treatment for anemic patients. That the science
was so advanced is astounding when you consider that only twenty-
five years had passed since James Watson and Francis Crick, a pair of
researchers at England’s Cambridge University, had determined the
structure of DNA, the biochemical that encodes the genetic infor¬
mation of all living organisms.
To appreciate this remarkable manmade process for unlocking
the body’s natural secrets, it helps to first aquaint yourself with some
basic cell biology. High school biology students have been learning it
for years. I promise to keep this short and simple, but you may pick
up enough to help your children with their homework.
Your body is made up of roughly 100 trillion cells. All of them,
with the exception of red blood cells, contain the complete genetic
blueprint for creating and sustaining human life, referred to as the
human genome. Imagine that you’re looking at a single cell magnified
100,000 times through a powerful microscope. Within its membra-
 From IPO to EPO in Four Months 79

nous wall lies the command center known as the nucleus, home to
twenty-three matched pairs of threadlike chromosomes; you inherit
half of each pair from your mother, and half from your father.
A chromosome holds about 1,000 genes, on average, although
the number varies—from chromosome 1, the largest (with an esti¬
mated 2,100 to 2,800 genes) to chromosome 21, the smallest (with
200 to 300 genes). [Link] is simply a short, tightly coiled segment of
DNA. Every cell is packed with about six feet of DNA, which meas¬
ures only about one-millionth of an inch wide.
The DNA molecule (as first described in 1953 by Watson, an
American biologist, and the British physicist Crick), resembles a lad¬
der that has been twisted into a corkscrew shape—the famed double
helix (see figure 4-1). Its long rails consist of sugar and phosphate,

FIGURE 4-1

The ladder of life

Sugar-phosphate backbone

Base pair

Nitrogenous base

Source: Illustration courtesy of the National Institutes of Health.

80 Science Lessons

and each rung is constructed from a pair of similar chemical bases—

either adenine and thymine, or guanine and cytosine —bonded together
by hydrogen (see figure 4-2). Typically, a gene is made up of several
thousand bases along the length of a DNA strand.
The DNA in each cell functions as a set of instructions, and it
programs its cell to produce proteins. These complex molecules, com¬
posed of a chain of smaller building blocks called amino acids, partici¬
pate in the structure, function, and regulation of your body’s cells,
tissues, and organs.
You can think of DNA as a language having only four letters—
A, T, G, and C—each corresponding to one of the chemical bases:

A = adenine

T = thymine

G = guanine

C = cytosine

All proteins are built from a combination of twenty amino acids.

The order of As,Ts, Gs, and Cs spells out which acids will be assem¬
bled in which sequence to form a given protein. DNA also governs
how much of the protein will be manufactured by your cells, as well
as when, why, and where in your body.
A four-letter alphabet might seem limited. But don’t forget that
computers can perform highly complex tasks by using a language
consisting of only two characters—the numbers 0 and 1—some¬
times called binary code. Our biological language is further simpli¬
fied by the requirement that each “word” contain exactly three
letters. For example, the three-letter word, or codon, UGG translates
into an order for the amino acid tryptophan; two three-letter
words—UUU and UUC—are both codons for the amino acid
phenylalanine. Other amino acids are associated with as many as
three, four, or even six different codons.
Source: Illustration courtesy of the National Institutes of Health.
82 Science Lessons

Why does nature contain this redundancy? Is it only a meaning¬

less artifact of evolution? The answer is that different organisms pre¬
fer different codons. Here’s why this is important for manufacturing
biological products: if, say, yeast is used as the production system for a
human protein, the production yields of that protein are greater if
you construct a gene using only the codons preferred by yeast rather
than the codons in the human gene. Many people believe that differ¬
ent organisms use different codons for the same amino acids to en¬
hance the ability of the body’s immune system to distinguish friends
from foes.
A gene, then, can be thought of as a sentence made up exclusively
of these three-letter words. Mathematically, with four letters and
with three-letter words, there are sixty-four possible combinations of
codons, in addition to three stop codons for denoting the end of a
gene, much like a period at the end of a sentence.

RECOMBINANT DNA TECHNOLOGY:

CUT AND PASTE

The basic process of using recombinant technology to manufacture a

therapeutic protein is as follows: you locate the gene (genetic in¬
structions) for producing the desired protein; you cut out this gene
or manufacture it in the laboratory; then you insert the gene into a
cell that will mass-produce the protein coded by the gene. In
essence, you turn the cell into a tiny protein factory.
In April 2003, fifty years after Watson and Crick’s landmark discov¬
ery, an international consortium of scientists finished sequencing the
approximately three billion base pairs that make up the approximately
thirty thousand genes in the human genome. This blueprint for hu¬
manity is more than 99 percent the same in every person on earth.
As of the mid-1970s, however, only a few hundred genes had been
mapped to their exact locations on particular chromosomes. The se-
 From IPO to EPO in Four Months 83

quence of base chemicals on one side of the DNA ladder complements

the other: adenine complements thymine, and guanine complements
cytosine. For example, if one strand reads AGCGT, we know that the
other reads TCGCA, and vice versa. Therefore, it’s not necessary to
identify the counterpart of any one sequence. Nevertheless, pinpoint¬
ing the code for a specific gene was a time-consuming, painstaking
process.
In 1978 molecular biologist Dr. Herbert Boyer, cofounder of
Genentech, created the first recombinant gene: a genetically engi¬
neered version of the DNA that produces the human hormone in¬
sulin, located on chromosome 11. Insulin—like erythropoietin, an
essential hormone—enables our cells to absorb the simple sugar glu¬
cose, the body’s principal source of energy. Following a meal, glucose
builds up rapidly in the blood. The pancreas, detecting this, floods the
circulation with insulin. People with diabetes make too little of the
hormone or none at all, causing excess glucose to accumulate in the
blood—a state known as hyperglycemia. Most of the 14.6 million diag¬
nosed diabetics in the United States manage to keep their blood-
sugar levels within a safe range through dietary restrictions alone or
through diet plus insulin injections or other medications. Repeated
episodes of dangerously elevated blood glucose can eventually dam¬
age multiple body systems, leading to severe complications such as
cardiovascular disease, blindness, and nerve disorders.
The method for isolating a targeted gene is difficult, laborious,
and complicated. Here it is in executive summary form: Boyer ex¬
amined human pancreatic cells to determine the genetic code for
insulin, 110 amino acids long. The next step—combining the in¬
sulin DNA and that of another organism—had been pioneered five
years earlier by Boyer and scientist Stanley Cohen. First, a delivery
vehicle is fashioned from a small ring of bacterial DNA called a
plasmid. Plasmids can be moved from one cell to another even when
the recipient cell belongs to a different species. This makes them
84 Science Lessons

ideal for smuggling DNA, like a microscopic Trojan horse, into a

target cell.
A portion of DNA is cut out of^the plasmid (imagine an O made
into a Cj, not with a snipping tool but with restriction enzymes. Still
other enzymes are used to splice the insulin gene into the opened
ends of the plasmid. Then the plasmid-gene combination is sealed
inside a single-celled organism capable of rapidly repheating itself,
such as bacteria, yeast, or mammalian cells. In Boyer’s time, all this
was done by hand. Now automated equipment can make thousands
of pieces of DNA within hours, but then, it was a major project to
create only one.
Boyer chose Escherichia coli—yes, the same E. coli that, if you
ingest its harmful strain in tainted hamburger or apple juice, can
induce fatal damage to your kidneys and digestive system. But the
E. coli bacterium used in recombinant engineering is extremely
weak and safe. (For Boyer and Cohen’s historic first effort at cloning,
in 1973, they placed a plasmid containing genetic material from an
African clawed toad inside the bacterium.) Each time the cell di¬
vides, its imported DNA instructions for manufacturing human in¬
sulin are passed down to the next generation. Before long, you have
billions of cells harboring the insulin gene.
Animal insulin, for injection beneath the skin, had been available
to diabetics since the early 1920s and was extracted in large amounts
from pig pancreases when the animals were slaughtered. It was used
to treat diabetes with good success, although few patients were aware
that they were injecting themselves with pig [Link] insulin hor¬
mone in pigs is so similar to human insulin that it can be used in
people, but it is different enough that some patients developed reac¬
tions to it. In addition to causing painful inflammation at the injec¬
tion site, this compromised the treatment’s effectiveness. Humulin,
the first drug produced through biotechnology, was hailed as a sig¬
nificant improvement; it is identical to human insulin, so the body’s
defense system pays no attention to it.
 From IPO to EPO in Four Months 85

LIN’S PREDICAMENT

When Genentech’s Humulin went on the market to great fanfare in

late 1982, Fu-Kuen Lin had been struggling for more than a year to
locate the EPO gene without success and with little cause for opti¬
mism. The chief problem that he and his research associate, Chi¬
ld wei Lin (no relation), faced was that they had little of the natural
product to work with. As George Rathmann explains, “It’s always
easier to sequence a molecule if you have lots of it. Unfortunately,
Gene Goldwasser was sending us only a half milligram of EPO at a
time, which just wasn’t enough.”
Somehow, word of our experiments reached The Dow Chemical
Company, which proposed becoming Amgen’s partner. What could
the giant corporation bring to the collaboration? Tens of thousands
of gallons of human urine partially purified to contain high concen¬
trations of erythropoietin. It seems that Dow had contracted with
the Italian army to collect urine from some fifty thousand soldiers.
A sample was sent to us, but despite the company’s claims, the
EPO levels turned out to be no greater than that of normal urine.
The people at Dow Chemical were so embarrassed that we never
heard from them [Link] new technology into a success¬
ful business is never simple, quick, or without numerous failures.
In the summer of 1983, while Amgen was still riding the eupho¬
ria of its initial public offering, Dan Vapnek argued that Amgen
should discontinue the EPO project and devote our limited re¬
sources to other endeavors. He wasn’t the only scientist to feel this
way. Based solely on science, we probably should have stopped. But
there was another important aspect of this decision: the human fac¬
tor. And that strongly supported continuing.
“Fu-Kuen’s lab bay was right next to mine,” Kirby Alton recalls
with a laugh. “Sometimes I would come in at weird hours because of
an experiment that I was working on, and there was never a time
when either he or Chi-Hwei was not there. I’m talking about the
86 Science Lessons

first four years of the company! They worked very, very hard, all of
the time.”
George Rathmann understood firsthand the gambles inherent in
any kind of research. Scientists are like anglers, casting their fishing
lines for answers in a vast sea of data. Failures far outweigh successes,
and it’s common for years of experimentation to pass without a bite.
A chief executive who wasn’t accustomed to the incremental pace
and unpredictability of drug development might have grown impa¬
tient and scuttled the program prematurely. Rathmann couldn’t
bring himself to fold the program, mostly out of respect for Lin’s
dedication.
He did, however, issue an ultimatum. “Nobody was questioning
how good Fu-Kuen was or how hard he worked, but the program
was not moving,” he reflects. “I told everybody involved that if we
didn’t see significant progress within sixty days, we’d have to get out
of the erythropoietin business—not as a threat but as a statement of
fact.” He continues, “Sixty days was an arbitrary number picked out
of thin air. I’d encountered situations like this before at 3M and at
Abbott, and setting a deadline usually had very beneficial results. I
knew it would get things in gear that otherwise would have been
taken for granted.”
The number 1 priority was to appeal to Eugene Goldwasser for
more erythropoietin. “He was very reluctant to part with it, which
was understandable,” says George, “because as long as he had the
world’s only supply of EPO, people would beat a path to his door.
But our people went to see Gene and insisted that he give us more
material, because what we had wasn’t enough to do the job. Now, if
they’d thought the project had another year, they probably wouldn’t
have done that as quickly as they did. It turned out that he had much
more than we thought.”
The extra supply provided Fu-Kuen Lin with additional infor¬
mation regarding a portion of EPO’s amino-acid sequence, enough
for him to design more than one hundred special probes that could
 From IPO to EPO in Four Months 87

then be tested on a vast number of gene fragments. Sensing a break¬

through, other scientists began pitching in, [Link] probes consisted
of short single strands ot DNA that might match the sequence for
EPO. When the correct probe came upon its complementary gene
segment, it formed a pair, thereby providing a clue to a part of EPO’s
DNA sequence.
Within a few weeks, in October, Lin isolated the EPO gene, on
chromosome 7. He elected to clone it using cells derived from the
ovary of a Chinese hamster (CHO). One reason for this novel ap¬
proach was that these CHO cells were especially responsive to treat¬
ment with the drug methotrexate, and that improved the chances
that the cloned gene would successfully express, or switch on, its pre¬
programmed protein. Another reason was that Genentech had al¬
ready convinced the U.S. FDA that CHO cells were suitable for
manufacturing biotech drugs; this would facilitate faster government
approval for EPO.
At that month’s employee meeting, which by now had to be held
at the nearby Holiday Inn to accommodate all 120 of them, Rath-
mann asked Fu-Kuen to stand up so that he could acknowledge his
achievement in front of the whole company. Not surprisingly, the bi¬
ologist wasn’t there; he was back at the lab, engrossed in his work. It
took fifteen minutes for someone to fetch him. George couldn’t
contain himself, though, and blurted out the extraordinary news.
When Lin walked through the door, the room erupted in cheers and
a standing ovation.
We were on our way ... or so it seemed.
.

I*.

'
CHAPTER

EPO on Trial

brilliant research, although an achievement in itself, does

not necessarily lead to an effective, salable product. Given the long
odds against Amgen’s developing a successful medication from re¬
combinant erythropoietin, Fu-Kuen Lin’s scientific milestone could
turn out to be a millstone, siphoning millions of dollars and hours
from the company. That’s the nature of the biotech industry.
For a 2003 study, researchers at Tufts University evaluated sixty-
eight medicinal compounds that entered into human trials to see
how they ultimately fared. (Six were genetically engineered; the rest,
conventional pharmaceuticals.) You can see in table 5-1 that about
two in three agents test well enough in phase I investigational studies
to warrant proceeding with phase II, but only one-third of phase II
clinical trials deliver results worthy of moving on to phase III.
To put in perspective the excruciatingly slow tempo of progress
in biotechnology, Amgen was nearly four years old when it finally
unlocked EPO’s molecular secrets. It would take two more years be¬
fore we could begin testing the hormone in patients, and then an-
90 Science Lessons

TABLE 5-1

The cost of each phase of drug testing

V Likelihood of moving to
Phase Cost the next phase of clinical trials

Phase I clinical trial $15.2 million 71 percent

Phase II clinical trial $23.5 million 31.4 percent

Phase III clinical trial $86.3 million

Source: “The price of innovation: New estimates of drug development costs,” Tufts Center for the
Study of Drug Development, Tufts University; William E. Simon Graduate School of Business Adminis¬
tration, University of Rochester; Department of Economics, Duke University, 2003.

Note: All figures represent the mean.

other three and a half years until Epogen finally reached the market¬
place. This was equivalent to a new world speed record.
Knowing that we had a long way to go and that the enterprise
could collapse at any point tempered some of the excitement over
Lin’s discovery. Don’t get me wrong: the whole company was
thrilled, but we were also cautious. Nor did EPO’s leapfrogging to
the front of the line diminish the commitment to our other prospec¬
tive human therapeutics and the diagnostics program. However, the
relatively poor showing of our work in specialty chemicals and ani¬
mal pharmaceuticals led us to discontinue them soon thereafter so
that we could concentrate our financial and human resources on
therapeutics and diagnostics.

THE DRUG APPROVAL PROCESS

The United States adheres to the world’s most rigorous guidelines

for ensuring the safety and efficacy of medicines, but it wasn’t always
 EPO on Trial 91

that way. As is true of consumer protection in general, government

regulations to safeguard the American people have been fortified
layer by layer over time—often in the aftermath of a tragedy.
Under the Food and Drugs Act of 1906, the first major piece of
federal legislation pertaining to pharmaceuticals, manufacturers were
required to demonstrate only the strength and purity of their prod¬
ucts. In 1937 the S. E. Massengill Company of Bristol, Tennessee,
marketed an elixir of its popular antibacterial medication sulfanil¬
amide, then available in tablet and powder forms. Hundreds of men,
women, and children across the country fell deathly ill after ingesting
the raspberry-flavored liquid, which turned out to contain a lethal
chemical solvent used in antifreeze.
All told, 107 people died. The death toll would have been signifi¬
cantly higher were it not for the quick action of the U.S. Food and
Drug Administration (known previously as the Food, Drug, and In¬
secticide Administration and, before that, the Bureau of Chemistry).
Two hundred forty gallons of sulfanilamide had been distributed;
with remarkable speed, the FDA tracked down and recovered all ex¬
cept about six gallons.
S. E. Massengill denied any culpability. Legally, the company
could not be held accountable, for pharmaceutical firms weren’t ob¬
ligated to test their products for toxicity. But chief chemist and phar¬
macist Harold Cole Watkins, who’d formulated the elixir, committed
[Link] tragedy prompted Congress to pass the Food, Drug, and
Cosmetic Act of 1938. Among the law’s provisions was a requirement
that new medications be proved safe before they could be marketed.
It took a near catastrophe twenty years later to spur the next
landmark drug regulation. In 1958 a nonprescription morning-sick¬
ness aid called thalidomide appeared on drugstore shelves in West
Germany and eventually throughout Europe. Two years later, evi¬
dence began surfacing that thalidomide could cause severe birth de¬
fects. By the time the William S. Merrill Company withdrew the
92 Science Lessons

drug in 1961, thousands of so-called thalidomide babies, whose

mothers had taken the drug during their pregnancies, exhibited
thalidomide’s disfiguring effects.
The United States was spared^a similar epidemic only because
Merrill s application to sell thalidomide in the United States landed
on the desk of FDA scientist Frances Kelsey. Ironically, she’d been in¬
volved with the recall of liquid sulfanilamide more than twenty years
earlier. While reviewing the materials submitted by Merrill, she
questioned some of the findings and requested further information,
postponing approval. During the delay, the story about thalidomide-
induced deformities broke internationally. But as the Washington Post
noted in an investigational report on July 15, 1962, it was only
Kelsey’s diligence (and perseverance in the face of criticism within
the agency) that kept the drug from passing muster. Three thousand
American women received thalidomide anyway, because free sam¬
ples had found their way to some thirteen hundred U.S. doctors.
Congress responded swiftly by unanimously amending the Food,
Drug, and Cosmetic Act. For the first time, pharmaceutical compa¬
nies would have to prove to the FDA’s satisfaction “through adequate
and well-controlled studies” that a new product was both safe and ef¬
fective. What’s more, manufacturers were required to advise the gov¬
ernment of all adverse reactions to a drug and to make sure its
advertisements carried comprehensive information for physicians
about unwanted side effects as well as benefits.
The Drug Amendments Act of 1962 was beefed up four years
later. Originally, the new provisions allowed any medication intro¬
duced before 1938—when only product safety had to be estab¬
lished—to remain on the market as long as there was no proof that it
didn’t work. But in 1966 the FDA assigned the National Research
Council of the National Academy of Sciences the monumental task
of assessing the effectiveness of each of those roughly four thousand
drugs. “The Drug Efficacy Study,” as it was named, took three years.
 EPO on Trial 93

Preclinical Studies

Before a pharmaceutical agent can be evaluated in people, it must

show impressive results in the laboratory and in animals. Most folks
probably aren’t aware that the FDA requires animal testing for all po¬
tential medications. Because a compound’s effects generally differ
from one species to the next, it is typically given to more than one
type of animal. For much the same reason, the results of animal stud¬
ies can’t necessarily be extrapolated to human beings, but they pro¬
vide us with essential, if preliminary, information: how the body
absorbs the drug; the types and severity of toxicity; and the time the
drug takes to be excreted.
Just so you know, all animal-research facihties must adhere to
stringent government standards for the ethical and humane treat¬
ment of the animals used in testing. Animals in a well-run research
center live in a cleaner environment and eat healthier diets than do
most people.
According to the Tufts Center for the Study of Drug Develop¬
ment, this preclinical phase consumes $5.2 million and three and a half
years, on average. After the company has analyzed and compiled its an¬
imal test findings, it develops a plan for a clinical trial: a study incorpo¬
rating human volunteers. The information is submitted, along with an
investigational new drug application (IND), to the FDA. Experts there
review the laboratory and animal research. Unless the FDA rejects the
application within thirty days, the clinical trial may proceed.
Under certain conditions, the FDA may allow investigators to re¬
cruit patients instead of healthy subjects for a phase I trial, as in the
case of Epogen. One reason was to expedite the process, because
there was no existing treatment for dialysis-related [Link] other
reflected a safety concern: epoetin alfa was intended solely for people
with depleted red counts. If a patient’s red-cell level was normal, tak¬
ing the drug could actually be harmful.
94 Science Lessons

Patient Studies: Knuckle-Cracking Time

Epogen’s phase I—phase II clinical trial commenced at Seattle’s Uni¬

versity ofWashington Medical Center in December 1985 under the
direction of Dr. David Dale, a prolific investigator in the area of ane¬
mia and other secondary blood disorders brought about by other
conditions.
Large university hospitals, traditionally steeped in research, tend
to be the settings for most early-stage medical [Link]’re prefer¬
able to smaller nonacademic facilities for several reasons. Impeccable
record keeping and patient scheduling are imperative. If the protocol
calls for daily injections of the experimental compound, that’s what
it must be. No exceptions. Big centers have the staffs to make it hap¬
pen on time. They routinely keep accurate records, and, more to the
point, they consider research a priority.
They also have large patient populations. Compliance by physi¬
cians and patients is likely to be better, and costs will be lower, if a
study takes place at a single institution where all the participants are
treated by one doctor, or perhaps two or three, as opposed to having
twenty-five subjects under the care of twenty-five doctors. (How¬
ever, phase III studies, which involve hundreds more subjects—
sometimes thousands more—must necessarily be carried out at
multiple centers.)
Regardless of the hospital’s size, it cannot administer an experi¬
mental drug to patients without the approval of an institutional review
board (IRB), a local body appointed by the FDA to authorize clinical
research. The board, composed of scientists, doctors, and nurses not
affiliated with the study, critiques the trial’s safety in advance and ei¬
ther grants approval or demands changes before giving the go-ahead.
Most major university research centers have IRBs in house, but if
not, they can use the services of another IRB.
During my eighteen years at Amgen, we had some drug candi¬
dates that dazzled in the lab only to fizzle in clinical trials. That’s
 EPO on Trial 95

something no one can foresee. However, we never had a product re¬

buffed by the FDA because of questionable science, as has happened
to many other companies. Probably the most memorable instance of
this in recent years involved the New York biotech company Im-
Clone and its highly touted colorectal cancer drug cetuximab (brand
name Erbitux), a genetically engineered version of an antibody that
interferes with cancer-cell growth.
In 2001, ImClone appeared to be months away from receiving a
license for Erbitux, which had been featured on 60 Minutes and
graced the cover of BusinessWeek. Based on early-phase clinical trials,
the drug looked as if it would be such a winner that pharmaceutical
giant Bristol-Myers Squibb agreed to pay $1 billion for the rights to
sell Erbitux domestically and in Canada while letting ImClone keep
60 percent of the profits.
But on December 28, 2001, the company sheepishly disclosed in
a press release that the FDA had rejected its application, citing serious
flaws in the phase III study. For one thing, the trial was supposed to
demonstrate Erbitux’s effectiveness against metastatic colorectal can¬
cer (denoting that the tumor has spread to another organ or organs)
in patients whose disease had not responded to previous treatment
with an approved chemotherapy drug. Yet half the men and women
enrolled in the trial did not meet that criterion. What’s more, said the
agency, the data concerning safety was incomplete and riddled with
discrepancies and inconsistencies. ImClone lamented publicly that its
having to run more studies would delay the product’s launch by
months. People in the industry knew better: if Erbitux ever joined
the ranks of commercially licensed anticancer agents—by no means
a certainty—it would probably take a year or two. We weren’t sur¬
prised that the drug didn’t go on the market until 2004. Although it’s
not the landmark therapy it was hoped to be, there is no doubt that
cetuximab has helped many patients.
The ImClone story is best remembered for the subsequent scan¬
dal, when it came to light that its chief executive, far from being
96 Science Lessons

“stunned” by the FDA’s rejection, as he’d claimed, had apparently

gotten wind of the decision days in advance and used that time to
quickly unload millions of dollars’ worth of company stock. Not
only that, but he advised family meiVibers and friends to do the same.
Martha Stewart, businesswoman and TV/magazine personality (and
a former stockbroker), heard of the sales from her broker and
promptly sold her ImClone stock. Both he and Stewart went to
prison. Although it’s seen by many as a cautionary tale of corporate
greed and deception, and it is, at the root of the whole sorry saga was
nothing more than inept design and execution of a clinical trial.
That never would have happened at Amgen. Our patient studies
were meticulously conceived and carried out, largely because we
insisted on having scientists, rather than physicians, devise the ex¬
periments. Why? With all due respect to MDs, medical schools typ¬
ically do not teach students how to design an experiment or interpret
its results. PhDs are thoroughly trained in both. Doctors partici¬
pated in our studies, of course, but in advisory roles. For example,
the physicians might review the design of an upcoming trial and
point out that in the real world the patients would probably be tak¬
ing a second medication in addition to the one being tested. That’s
something the PhDs wouldn’t have known. So in some studies we’d
add the other drug to the protocol for a more realistic, meaningful
experiment.
Amgen’s criteria for choosing principal investigators sometimes
frustrated our sales and marketing staff, who tended to favor highly
prestigious doctors—opinion leaders who can give speeches and get
people to listen to them. My feeling, and that of R&D, was that a
study’s positive findings carried far more weight than the chief re¬
searcher’s name recognition. As CEO I instituted what I believe was
a fair rule: our people in R&D would select the physicians—after all,
they were responsible for conducting clinical trials—but they had to
give serious consideration to recommendations made by sales and
marketing.
 EPO on Trial 97

The fact that U.S. drug manufacturers provide nearly all the
funding for such studies and therefore hire the doctors who run
them touches on a bit of a systemwide dilemma. It can be argued—
and plenty of people have—that this arrangement creates an inher¬
ent conflict of interest for the physicians involved and could
conceivably taint the findings. The reality is that few universities are
endowed with a sufficient budget to finance wholly independent
drug-evaluation studies.
I can tell you, though, that fears of professional bias are largely
unfounded. Confirmed cynics may sneer, but the vast majority of
medical researchers are highly ethical and impervious to any implied
pressure to tinker with the facts. Their allegiance is to high-quality
science and to patients, and not to the company bankrolling the
study. Besides, no amount of spurious interpretation can alter the
data, which is scrutinized by medical science professionals at the sci¬
entific journals that publish clinical trial results, not to mention the
FDA. (For the record, one of the most notorious cases of clinical trial
fraud in U.S. history occurred under the direction of the federal Na¬
tional Institutes of Health; in 1993 a Montreal researcher participat¬
ing in a large multicenter NIH study comparing two types of
breast-cancer surgery was found to have falsified results.)
Is there room for improvement? Of course. A 2006 report in the
American Medical Association’s journal, JAMA, examined the find¬
ings of more than two hundred randomized patient studies of car¬
diovascular medicines between 2000 and 2005. Here’s what the two
authors, from Boston’s Harvard Medical School and Brigham and
Women’s Hospital, found: about 40 percent of the trials financed by
not-for-profit organizations favored the experimental therapy over
the standard therapy. When the source of funding was a for-profit or¬
ganization, such as a pharmaceutical manufacturer, the drug being
tested proved superior in 66 percent of the studies; and in studies
jointly underwritten by both types of organizations, the newer treat¬
ment came out on top 54.5 percent of the time. Is this proof of
98 Science Lessons

investigator bias? No. What I draw from these numbers is that a phar¬
maceutical company knows more about its drug than any outsider;
therefore, it’s able to design more accurate trials. Shoddy data and
faulty conclusions stem mainly from incompetence and carelessness.
In any event, the debate is likely to go on forever.

A surprise from the fda. Certain medicines lend themselves

to testing a hypothesis more readily than others. For instance, what if
you were assessing the effectiveness of a medication intended to treat
pain and stiffness from rheumatoid arthritis? No diagnostic proce¬
dure can measure the degree to which an arthritis sufferer hurts or
the pain’s characteristics (sharp, dull, tingling, and so on). So the ver¬
dict might be based on the volunteers’ oral or written responses to
questions such as, “On a scale of 1 to 10, rate the intensity of your
pain both before and after you took the medication.”
With anemia due to kidney failure, however, the answer to
whether Epogen worked could be assessed from a single number.
Hematocrit, as it’s called, expresses in percentage form the volume of
red blood cells contained in blood. When you have your annual
physical, the hematocrit test is routinely performed as part of the
complete blood count (CBC), with the blood drawn from a vein in
your arm or the back of your hand. The average range for a man is 42
percent to 54 percent; for a woman, 38 to 46 percent.
Regular kidney dialysis does a remarkable job of removing im¬
purities from the blood in the absence of functioning kidneys. Un¬
fortunately, the treatment has no effect on red-cell production. Until
Epogen, the only therapy available to people with dialysis-related
anemia was to undergo transfusions of red blood cells. This was a
temporary solution and fraught with complications: repeated trans¬
fusions may bring about a potentially serious disorder known as iron
overload, where too much of the mineral accumulates in the body;
and they can transmit infectious diseases such as HIV and hepatitis B
 EPO on Trial 99

and [Link] procedure can also mobilize the body’s immune system
into churning out antibodies against the foreign red blood [Link]
could preclude a patient from one day receiving a kidney transplant,
the one full-fledged cure for ESRD.
The average dialysis patients hematocrit is about one-half to
one-third that of a healthy adult. “Testing erythropoietin seemed as if
it would be really straightforward,” says Kirby Alton, then the leader
of our interferon projects. “You’re anemic, your hematocrit is low,
you’re given EPO, and your hematocrit goes up. That is a very
doable, successful experiment.”
He was shocked, then, when the FDA objected to using hemat¬
ocrit as the study’s medical end point—the general outcome that the
protocol is supposed to evaluate, such as severe side effects, disease
progression despite treatment, or death. “We’d gone to Washington
to meet with the FDA and get approval for our study design,” says
Kirby. The first thing out of their mouths was,‘Improved hematocrit
is not a clinical benefit. What’s the clinical benefit?’John [Adamson]
nearly fell out of his chair.” John Adamson was to be one of the
study’s authors, along with his University of Washington colleague
Dr. Joseph [Link] two had performed important research on
EPO in the late 1970s and early 1980s, working with sheep.
A higher hematocrit didn’t confer a clinical benefit? Was the
FDA kidding? If your percentage of red blood cells is increased, your
tissues are imbued with more life-sustaining [Link] energy re¬
turns. Indeed, as would be shown only months into the study, re¬
combinant erythropoietin transformed patients’ lives.
Well, prove it, said the FDA.
“So,” says Alton, “we ended up having to use the subjects’ de¬
pendence on blood transfusions as the study’s end point.” In other
words, the goal was to demonstrate that raising the patients’ hemat¬
ocrits would reduce the number of transfusions they needed to be
able to function.
100 Science Lessons

OF mice and men. Although recombinant EPO had worked

extremely well in mice, it could still fail in people. Many human bio¬
logical systems are controlled by a feedback loop and sometimes in¬
clude the equivalent of an accelerator and a brake. In mice, EPO
acted as a gas pedal, stimulating the bone marrow to manufacture red
blood cells. But what if ESRD sufferers became anemic not due to a
defective accelerator but because they had a faulty braking mecha¬
nism that somehow interfered with red-cell production or prevented
the cells from maturing normally? If that proved to be the case,
flooring the gas pedal might merely prompt greater resistance from
the [Link] was one big question we needed to answer.
Another question concerned the effect of increasing the hemat¬
ocrit of people on dialysis. In some medical circles, it was believed
that ESRD patients were supposed to have abnormally low hemat¬
ocrits; that the body reduced the volume of red cells for good rea¬
son, intentionally thinning the blood. No one knew for certain
whether that was true. A third source of worry was whether the drug
would trigger an immune response, with the body attacking the ge¬
netically engineered hormone.
Every one of those doubts was erased well before the end of the
trial. Data from animal studies helps us to estimate the therapeutic
dose in people: a dosage strong enough to induce a response (hope¬
fully) but not so high that it produces bothersome side effects (again,
hopefully).The twenty-five participants were injected with Epogen
at their thrice-weekly hemodialysis sessions. Of the eighteen men
and women administered adequate doses, eleven soon saw their
hematocrit levels increase dramatically—and with that, a boost in
energy—and twelve patients who had regularly required transfusions
no longer needed them. These results occurred at a dosage so low
that not one negative side effect was observed. In fact, we once cal¬
culated that an aspirin-sized amount of Epogen’s active ingredient
would be enough to treat someone with ESRD-related anemia for
forty years.
 EPO on Trial 101

We did have one scare, though, when patient number 6 arrived at

the dialysis center looking as red as a beet. It so alarmed a physician
associated with the trial that he immediately alerted the FDA instead
of contacting us or the lead investigator. Was EPO instigating a dan¬
gerous rise in blood pressure? Was it an unanticipated inflammatory
reaction? We couldn’t imagine. For two days, everyone at Amgen
seemed to wear the same fretful expression. Then the cause of the
change in the man’s complexion was revealed: he’d gone to the
beach and suffered a nasty sunburn. Now it was the reporting doc¬
tor’s turn to be red faced.
The study, published in the January 8, 1987, edition of the New
England Journal of Medicine, concluded, “These results demonstrate
that recombinant human erythropoietin is effective, can eliminate
the need for transfusions with their risks of immunologic sensitiza¬
tion, infection, and iron overload, and can restore the hematocrit to
normal in many patients with the anemia of end-stage renal disease.”

Phase III: Prospective, Randomized, Controlled, and Double Blind

One more crucial step remained: a large-scale phase III prospective

randomized controlled double-blind clinical trial, the type univer¬
sally preferred by researchers and drugmakers alike because it deliv¬
ers the purest, most convincing information for proving a definite
link, or causal relationship, between a particular treatment and a
health condition. The best way to define the lengthy term is to de¬
construct it word by word:

• Prospective: any research project that studies one or more

groups of patients from the present time until some point in
the future. (A retrospective study analyzes data that was col¬
lected in the past.)

• Controlled: a comparison study in which participants are as¬

signed to one of two or more [Link] study group receives
102 Science Lessons

the investigational treatment, and the control group is given a

lookalike inert substance (the placebo). If merely prescribing a
placebo could jeopardize the health of the volunteers, the
current standard therapy is u^sed as a [Link] are even
sham surgeries, simulated acupuncture treatments, and so on
for evaluating the effectiveness of new procedures.

• Randomized: If we re comparing one treatment to another or

to a placebo, we need to have some idea of their relative
value. The purpose of randomization is to make sure that the
control group and the treatment group are comparable. In this
way, if we introduce an investigational therapy in one group
and not in the other, and if the study group gets better com¬
pared with the other, there’s a high probability that the effect
shown is due to the therapy.
Many factors can distort a study’s outcome, beginning
with the selection of patients. Employing a random process
for distributing subjects to the study group and the control
group precludes the possibility that someone involved in the
trial will subconsciously stock the experimental arm with
“ringers”—patients who, for whatever reason, would respond
well, thus resulting in more favorable results.

• Double blind: Surely someone can tell whether she’s been

given a placebo, right? Not if she’s part of a blind study. In a
single-blind trial, the participants don’t know whether they’re
in the control group or the study group until the experiment
is completed; a double-blind trial also keeps the research staff
in the dark. Why do researchers take this precaution? It’s be¬
cause a doctor might inadvertently reveal which protocol a
patient is on, through facial expressions, body language, or
tone of voice. Keeping the subjects’ group allocations confi¬
dential also prevents investigators’ personal biases from influ¬
encing the results.
 EPO on Trial 103

Epogen’s phase III clinical trial, carried out at the University of

Alabama Medical Center, recruited 333 anemic dialysis patients.
Within twelve weeks, all except 2.6 percent of the control group ex¬
hibited improved red-cell levels, and not one required a transfu¬
sion—powerful findings by any [Link] drug was so effective,
in fact, that the researchers had a difficult time trying to disguise
which patients were in which group. As an Amgen scientist observed
pointedly to one of the frustrated doctors, “Why, I can stand here
halfway across the room and tell you which patients are on EPO and
which are on placebo.”
“How can you do that?”
He said, “Well, look at ’em! Some of them are pink and some of
them aren’t.” The men and women taking Epogen had healthy com¬
plexions and felt good, but the people in the control group were
pale. Pretty soon, many of the volunteers had figured out who was
receiving what. The drug was that effective.
The Internet has complicated the process of conducting blind
studies, because some patients open an online chat room and tell the
world how they’re feeling and swap information, trying to deduce
which group they’re in. Usually it’s only a matter of curiosity. How¬
ever, at times a person will determine that he’s in the placebo group
and quit the trial; it’s one of the unforeseeable negative aspects of
access to information through the Internet. Clinical studies must
enlist more people than in the past to compensate for the higher
dropout rate.
Incidentally, the FDA automatically classifies dropouts as nonre¬
sponders—people for whom the experimental therapy did not work.
Once, in a weight-loss trial that we were conducting, the volunteer
who’d lost more pounds than anyone else was tragically killed in a
car accident before completing the study. Under FDA rules, he was
reported as having dropped out and therefore was considered a non¬
responder, when in fact his response to the drug had exceeded
everyone’s expectations.
104 Science Lessons

THE SIMI VALLEY HOSTAGES

After the third and final phase of human trials, the drug company
carefully analyzes all of its scientific data and then submits it to the
FDA as a new drug application (NDA) or a biologic license application
(BLA). An average NDA or BLA runs more than 100,000 pages.
Based on the spectacular results of our patient studies, there was no
question that Epogen would receive the FDA’s blessing and become
Amgen’s first commercial product, so we were understandably eager
to speed the process on our end in any way we could.
In August 1987, a team of about twelve staffers began the arduous
task of preparing the documentation. The job probably should have
taken about 280 days; our employees resolved to get it done in only
90 days, working in shifts virtually around the clock.
One of them was Dennis Fenton. “We didn’t have a lot of office
space where we could work undisturbed,” he recalls, “so we rented a
block of ten rooms at the Posada Royale Quality Inn in Simi Valley,”
about ten miles away. The team became known as the Sinn Valley
Hostages.
By this time, we’d licensed some of the rights to Epogen to
Ortho Pharmaceuticals (you’ll learn more about that in chapter 6).
When the people at Ortho heard that we planned to complete the
BLA within ninety days, they told us it was impossible. “It was quite
an inspiration to us,” says Fenton. “We felt, ‘We’ll show those guys!
We’re going to get it done in ninety days, like we said we would.’
And we did.”
Well, almost. It took ninety-three days, to be exact, from August 3
to October 30. Mike Downing, our clinical affairs director, painted
“The Simi Valley Hostages Are Free! 93 Days!” on a bedsheet and
hung it over the entrance of building 1. The paperwork was com¬
piled into 140 volumes, which formed a stack nine feet high.
“In a weird way, it was kind of fun,” Dennis says. “It built a lot of
close relationships among the people who were involved. For a num-
 EPO on Trial 105

ber of years, we used to have reunions back at the same hotel.” This
wasn’t the last time that the people who worked at Amgen displayed
an extraordinary dedication and commitment to getting the job
done. How do you find employees like that? And how do you keep
them? In chapter 9,1 tell you.
V
'
CHAPTER

Partnerships Made in
Heaven—and That Other Place

one day in the fall of 1988, George Rathmann called me into his
office and told me that he was stepping down. Before I’d even di¬
gested that news, which came as a shock, he shook my hand and said,
“Congratulations, [Link] board has elected you CEO.”
At the time, Amgen was only months away from earning FDA ap¬
proval to market Epogen, and our second recombinant hematopoietic
growth factor, filgrastim (Neupogen), had recently tested well in a
phase I/II clinical trial for cancer patients undergoing chemotherapy.
The question on many people s minds, inside and outside the com¬
pany, was, “Why on earth is George retiring now?” After an arduous
eight-year trek up the mountain, why not savor the view from the
top, at least for a little while?
“It was a combination of things,” explains George. “I’d been
working extremely hard, and my wife decided that she would much
rather have a poor husband than be a wealthy widow. But also, I felt
108 Science Lessons

it was a good time for Amgen to bring in a new CEO, so that he

would have the momentum from taking Epogen across the finish
line, instead of coming in afterward.’*) He adds with a laugh, “It turned
out to be a brilliant call.” (It was so brilliant, in fact, that twelve years
later I followed his precedent, timing my retirement just as the
pipeline was about to deliver four new products. Neither of us be¬
lieves in changing horses in midstream.)
Still, the announcement shook up everyone.“We all loved George,”
says Kirby Alton. “I spent six hours in his office one night trying to
change his mind.” I’m sure, too, that some folks questioned the wis¬
dom of leaving a biotech firm in the hands of a “finance guy” instead
of, say, a molecular biologist. After six years as Amgen’s CFO, however,
I’d become fairly fluent in the science. I didn’t always have the an¬
swers, but I’d learned enough to ask our researchers most of the right
questions about ongoing projects. Although we tend to think of sci¬
entists as reserved and dispassionate, nothing could be further from
the truth; their eyes lit up whenever they discussed their work.
I’d always been fascinated by science, having grown up in Ala¬
mogordo, New Mexico, a major government site for testing and de¬
veloping guided missiles, pilotless aircraft, and other research programs.
In my home town it wasn’t out of the ordinary to see a missile streak
toward a radio-controlled airplane overhead and knock it out of the
sky. For a teenager, this was thrilling stuff. I made up my mind then
that I was going to become an engineer and, by God, help put a man
on the moon. But I got sidetracked.
After graduating from Purdue University with a BS in electrical
engineering in 1957, I served three years as an officer on the USS
Intrepid, now a floating pierside museum in Manhattan. Even during
peacetime, aircraft carriers spent six days a week at sea because navy
pilots had to practice flight maneuvers constantly. One thing about
being in the middle of the ocean is that you have plenty of time on
your hands. Between reading Fortune magazine and a pile of books
about business, I became increasingly drawn to commerce and less
 Partnerships Made in Heaven—and That Other Place 109

interested in electrical engineering. As soon as my tour of duty

ended, I enrolled in the Harvard Business School MBA program.
In Amgen’s early years, almost every employee held a doctorate in
one science or another, and with biotechnology still in an embryonic
stage, it was central to the company’s success that its first chief execu¬
tive happened to be a scientist himself. That was no longer true in
1988. Given the makeup of the company, Amgen would always be in¬
extricably anchored in research no matter who sat in the CEO’s chair.
And by the time I handed the reins to Kevin Sharer, a background in
science was even less a prerequisite for a biotech chief exec.
The board of directors could have hired a veteran CEO from
outside the company, but the members knew that it would involve a
great deal of change, and they wanted Amgen to continue as it was.
I’ll let Dennis Fenton describe the mood around Amgen after Rath-
mann made his decision known: “At first, there was a sense of panic
over George’s leaving. But then we thought,‘Wait, it’s OK—-it’s Gor¬
don. Always very reasonable, always willing to listen. We’re going to
be all right.’ With an outside person, though, everyone would have
worried that the whole place was going to change.”
By announcing his retirement and his successor at the same time,
Rathmann enabled Amgen to avoid a runoff for the job. That’s one
of the worst things an organization can do. It turns candidates who
should be working together into competitors, and that can poison
the atmosphere by forcing lower-level employees to choose sides. It
all but guarantees that every one of the losing contenders will end up
leaving the company.
Amgen tried a different approach: under our new structure, George
would remain chairman of the board for two years, and Harry Hixon,
VP of product development, was elevated to chief operating officer
and president. Hixon had come to Amgen in 1985 from Abbott Lab¬
oratories, where at one time he’d reported to Rathmann.
Although this arrangement was intended to ensure a seamless
transition, to some extent it strained the organization at the seams.
110 Science Lessons

George Rathmann divvied up the company, starting with research

and development, which had always been a single entity. Now devel¬
opment was handed to Hixon, aloq.g with manufacturing and mar¬
keting, and I oversaw research, finance, all staff functions—and
Harry. It’s hard for me to be impartial, so here are Kirby Alton’s ob¬
servations on this shaky period in Amgen’s history: “George probably
restructured the organization this way in order to give Gordon and
Harry enough control that both would stay. But it created a lot of
friction. As the new vice president of development, I was kind of
caught in the middle, because development basically fits in between
research and manufacturing.”
Hixon has been quoted as saying that he and I are two very dif¬
ferent people with dramatically different management styles, and on
this point we’re in full agreement. To be honest, the board of direc¬
tors probably sensed from the outset that the shotgun marriage
might be headed for trouble. Both of us were told bluntly that if we
didn’t try to make this sort of coalition government concept work,
an experienced CEO would be imported from another company,
and we would both have to leave. As professionals, of course we had
every intention of working together. But we were like two porcu¬
pines mating: how do they do it? Very carefully.
Although George Rathmann had been named Gold Medalist
biotechnology CEO of the Year in 1987 and 1988 by The Wall Street
Transcript, following in his footsteps wasn’t as daunting as you might
think. Amgen was in great shape, with great people, great product
candidates, and momentum. It also helped that with Epogen’s ap¬
proval drawing near, the company was changing. George had thrived
on the vitality of a smaller organization, and the larger it became, the
less fun it was for him. (He intended to try his hand at a life of
leisure, but within a year and a half he became the chief executive of¬
ficer, chairman, and president of ICOS, a biotech company in subur¬
ban Seattle. After eight years there, Rathmann quickly cut short a
 Partnerships Made in Heaven—and That Other Place 111

second retirement to assume all three positions at a California

biotech firm called Hyseq.)
As for me, I couldn’t wait for Amgen to grow. I’d worked for large
firms before and felt that my managerial skills were even better
suited to a company of that size.
One classic mistake that executives often make following a pro¬
motion or a change in companies is to try to imprint their style too
soon—sometimes when no change is warranted at all. For example,
Kevin Sharer freely admits to being overly aggressive when he ar¬
rived at MCI Communications Corporation from General Electric:

At GE, there was a presumption that a new general manager

would make an impact on the business quickly, and that if you
can run one business, you can run any business. When I went
to MCI as president, I was forty years old and, in some ways, a
relatively young and unsophisticated senior executive. I prob¬
ably felt my oats more than was fully prudent. After just a
month or two on the job, I went to the CEO and told him, “I
think you’ve all misconceptualized this business, and we need
to fundamentally reorganize.” The fact that I was right didn’t
matter. I hadn’t built sufficient internal credibility.
At the senior executive level, management is more nu-
anced, and you have to carefully weigh the organization’s
readiness for change. Obviously, if you believe it’s a burning
platform issue and the very existence of the organization is
threatened without the change, then you’re more likely to act
in a somewhat unilateral way. But if it’s change that could
happen three months from now, or six months from now, it’s a
lot better to prepare the ground first than to simply exercise
your power to effect change.
I realized my mistake and learned from it, so that when I
came to Amgen three years later, I consciously waited to
112 Science Lessons

become an insider before making any dramatic moves. The

changes were subtle. People would notice that things were
different, but they couldn’t poisjt to any single decision that
caused it.

As for me, I certainly had no intention of instituting immediate

wholesale changes as chief executive officer. There was no call for it.
Besides, my time was monopolized almost entirely by two crises that
could have capsized the company.

PATENTLY ABSURD:THE CASE AGAINST

GENETICS INSTITUTE

Only three months into my tenure as CEO, Amgen was embroiled

in a pair of lawsuits with two other drug companies. The first involved
a patent-infringement suit that we’d filed against Massachusetts bio¬
tech firm Genetics Institute (GI) and its Japanese partner, Chugai
Pharmaceutical Co., in 1987.
Patent litigation is so commonplace in the biotech field that it gets
factored in to the overall cost of bringing a drug to market. And what
a cost. A no-stone-unturned dispute can drag on for years and easily
chew through $20 million to $30 million on each [Link] you spend
it without flinching because it’s your only way to protect your re¬
search and development achievements. If Amgen had owned billions
of dollars in gold bars, we’d have locked them away in an under¬
ground vault and posted armed guards around the building to prevent
theft. However, our assets were intellectual property—intangible
human knowledge and ideas (see “What Is and Isn’t Patentable”). So
we hired patent attorneys to preserve the rights granted to us by the
U.S. Patent andTrademark Office (USPTO).
A U.S. product patent entitles the patent holder to prevent any¬
one else from making, using, or selling its invention or discovery for
twenty years from the date the patent application was filed. Because
 Partnerships Made in Heaven—and That Other Place 113

What Is and Isn't Patentable

Over the years, the U.S. Patent and Trade Office has gradually clarified
the ground rules for patenting genetically engineered products and
processes. Here’s a rundown of the key guidelines:

• A naturally occurring gene or protein as it exists in the body

(an important distinction) is not eligible for patent protection.
However, once the substance has been isolated outside of the
body—and has been shown to have at least one credible use in
therapeutics, diagnostics, or some other application—it can be
patented.

• New uses for a product are also patentable, even if the gene or
protein itself has been patented by someone else.

it typically takes the USPTO three years to issue a patent, in reality

the protection lasts roughly seventeen years. When the patent ex¬
pires, all companies are free to make, use, and sell the discovery or in¬
vention. In addition to patent protection, under the Orphan Drug
Act, Epogen was already guaranteed seven years of market exclusiv¬
ity, to start as soon as it received FDA approval.
Amgen submitted its patent application on December 13, 1983.
As of mid-1987, we were still waiting. The patent office operates a
bit like a Turkish bazaar. Invariably, it initially turns down most
[Link] applicant then supplies additional documentation and re¬
states its case. Frequently the USPTO rejects claims a second time,
forcing the company to submit even more evidence. In the end, the
patent office and the company may agree on a narrower patent. The
haggling can go on for years.
114 Science Lessons

Another problem was that the USPTO staff consisted almost en¬
tirely of engineers and chemists, and not molecular biologists. Con¬
sequently, the agency didn’t really Jcnow how to process claims on
genetically engineered drugs. Applications started to pile up, and it
took longer and longer for patents to be issued.
Out of frustration, our trade association, the Biotechnology In¬
dustry Organization (BIO), offered to train USPTO employees in
the science of biotechnology at no cost—anything to speed up the
process. The patent office gratefully accepted, and the program was
highly successful in helping reduce the backlog of applications.
To be fair to the USPTO, the complexities regarding ownership
were far from settled. To some degree, they still aren’t. What exactly
are you declaring as yours? The gene sequence? The amino-acid se¬
quence? The process for making the drug? The drug formulation? A
method of using the drug? All of the above? The patent ultimately is¬
sued to Genetics Institute claimed only the purification of EPO
from human urine, whereas our first patent covered the gene se¬
quence of human EPO. We had filed for other claims as well, includ¬
ing one covering the amino-acid sequence and several related to
manufacturing, but these were still pending.
On June 30,1987, Genetics Institute announced that it had been
awarded the first patent for EPO, completely blindsiding us. Amgen
had filed its application more than a year before GI. Also, as our gen¬
eral counsel, Robert Weist, explains, “Usually you’ll know what your
competition is up to because publications in foreign countries will
contain news about other companies filing patent applications there.
In the case of Genetics Institute’s patent on erythropoietin, there was
no such early warning. For whatever reason, they hadn’t filed outside
the U.S., so our normal patent surveillance didn’t reveal anything
about what they were doing here in the States.” At the time, U.S.
patent applications remained confidential until the patent was as¬
signed, whereas other countries publish patent applications a year or
more after they’re filed but before the patent is [Link] USPTO
 Partnerships Made in Heaven—and That Other Place 115

has since joined the rest of the world in publishing patent applica¬
tions prior to granting patents.
Weist continues:

Not only that, but after we’d filed our basic patent applica¬
tions, I’d written letters to all the other companies that had
been trying to clone the gene for EPO and put them on no¬
tice that Amgen had no intention of licensing the patent to
anyone; we were going to manufacture and distribute Epogen
ourselves or with a partner. We received some interesting re¬
sponses. 3M, one of the world’s most ethical organizations—
but also highly aggressive at filing on any and all inventions
and pressuring the USPTO to issue stronger patents—-sent a
very gracious letter of congratulations and said that it would
pursue other projects.
Genetics Institute’s position was altogether different. Basi¬
cally, they said that Amgen might have been first, but they
were working in this area, too, and they saw no reason why
the field wasn’t big enough for two players. They planned to
proceed with their own recombinant erythropoietin product.

Bob wasn’t overly worried, though. Neither was George Rath-

mann, .who over the years had learned to analyze patents about as
well as any patent lawyer, in part because one of his brothers was a
patent attorney. According to Weist, “We felt comfortable with the
position that Amgen was indeed first, we’d made a significant discov¬
ery, and we’d filed the patent applications properly. We were confi¬
dent that would be recognized in the courts if it came to that.” But
considering that the patenting of genes, proteins, and so on was unfa¬
miliar territory, nothing was certain.
For some technical reason, our application and GI’s application
had been reviewed by different sections of the patent office. Appar¬
ently the section that handled its application took fewer coffee
116 Science Lessons

breaks; how else can we explain why GI was awarded its patent first?
We perfected a slow burn throughout the summer, knowing that the
moment we received our patent a*n the DNA used to produce
Epogen, we would sue GI in federal court for patent infringement.
That day came on October 27,1987. As expected, Genetics Institute
immediately countersued, claiming that our patent was invalid.
A drug company with a potential blockbuster on its hands often
feels like the besieged sheriff of a dusty frontier town. One desperado
after another comes gunning for him on the off chance that they
might outduel him and be free to plunder the town bank. Generic-
drug manufacturers often do this. Rather than wait until an existing
patent expires, they try to prove it invalid. If they’re successful, it’s
often a double victory: not only do they get to market early, but also
they usually receive a head start on their generic competitors as the
sole producer for six months.
Another strategy is to pressure the patent holder to enter into a
settlement. Predictably, at one point GI offered to cross-license
Epogen with us in exchange for dropping its suit. Under this
arrangement, each company would be free to sell EPO. We refused.
Genetics Institute didn’t even start working in earnest on EPO until
well after Fu-Kuen Lin’s successful cloning of human erythropoi¬
etin, so why should we share the rewards from our groundbreak¬
ing—and immensely costly—research?
The USPTO froze all further work on applications related to
Epogen until the lawsuits were settled. As a result, throughout 1988
two companies were laying claim to recombinant human erythro¬
poietin, each with its own patent, and with still other patents pend¬
ing. Wall Street investors weren’t sure what or whom to believe.
Genetics Institute had a policy of licensing its drugs to large
pharmaceutical firms instead of marketing them itself. Thanks to a
loophole in a federal trade law, the Tokyo-based Chugai company
could help GI conquer the U.S. market in a way that no company
based in the United States could. Our patent prevented competitors
 Partnerships Made in Heaven—and That Other Place 117

Two Landmark Laws That Shaped

Biotech History
Nineteen eighty was a watershed year for biotechnology-related leg¬
islation. Without the Bayh-Dole Act and the Supreme Court ruling in
the case of Diamond v. Chakrabarty, it’s doubtful that the industry
would have grown as rapidly as it did.
In the early 1970s, a General Electric biochemist named Ananda
Chakrabarty genetically modified a Pseudomonas bacterium so that it
would break down and consume crude oil; the compound was designed
for use in the event of an oil spill. The patent office denied his claim, as¬
serting that “as living things, microbes are not patentable subject mat¬
ter.” After eight years of legal appeals, the case landed in the U.S.
Supreme Court. In June 1980 the Court narrowly upheld Chakrabarty’s
patent by a 5-4 vote, stating that the oil-eating bacterium was man¬
made and not a product of nature. With biotech companies now as¬
sured of protection for their discoveries and inventions, research
surged.
Six months later Congress passed the Patent and Trademark Act
Amendments of 1980, cosponsored by senators Birch Bayh (D-IN) and
Bob Dole (R-KS). This law, too, stimulated research by permitting small
businesses and universities to own the titles to inventions resulting
from federally funded research programs.
Until then, those patents had belonged to the government, which
handed out nonexclusive licenses to anyone and everyone. With no
patent protection, few companies were willing to invest in and de¬
velop these discoveries; of 28,000 government-controlled patents,
only about 1 in 20 were licensed to industry. Today approximately
2,000 patents are awarded to universities each year, as compared with
perhaps 250 annually before the Bayh-Dole Act took effect in 1981.
118 Science Lessons

from manufacturing Epogen in the United States, but it did not

apply to those that produced the drug outside U.S. borders. Remark¬
ably, this oversight—corrected manyvyears later—created the absurd
situation wherein a company such as Chugai could circumvent a
U.S. patent simply by making EPO offshore, freely using the pro¬
tected technology, and then exporting the product to America. For a
time the U.S. patent system actually favored foreign-based businesses
over their U.S.-based counterparts.
Four days after our patent was issued, Gabe Schmergel, GI’s pres¬
ident, publicly stated that his company intended to ship Chugai the
genetically engineered cells needed to make EPO. We complained
vigorously to the U.S. International Trade Commission. ITC Admin¬
istrative Law Judge Sydney Harris agreed with our contention that
Chugai would be guilty of patent infringement. Nevertheless, she
dismissed our initial complaint on the grounds that, for technical
reasons, the commission lacked jurisdiction to prevent the importa¬
tion of EPO. Although that decision was overturned on appeal in
federal court, no action was taken; the judge concluded that the mat¬
ter should be remedied by Congress in the form of new legislation,
and not by the legal system.
The U.S. patent system is designed to grant patents to the first
person to invent, and not the first person to [Link] Genetics Insti¬
tute could contest Amgen’s patent on any of four grounds:

• We did it first. This was GI’s contention.

• The patent is not valid because the invention was “obvious. ” The
challenger must demonstrate that prior work in the field ren¬
ders the invention an inevitable development rather than a
true innovation. GI claimed this, too.

• The patent is not valid because the invention was covered by “prior
art. ” Once something has been published or otherwise re¬
vealed publicly, it cannot be patented. Many otherwise
 Partnerships Made in Heaven—and That Other Place 119

patentable university inventions cannot be protected for this

reason.

• The patent is not enforceable because the applicant misinformed or

withheld information from the patent office. As far as I know,
patent applications are the only documents submitted to the
government in which the applicant is required to disclose
every reason it can think of that the application should be re¬
jected. The onus is on the patent seeker to make all relevant
details known to the patent office, including information that
might reflect poorly on its application. Failing to reveal the
whole truth is considered “fraud on the patent office.”
Strangely, this action doesn’t void your patent but makes it
unenforceable. Of course, there’s always the chance that the
data you supply could derail your obtaining patent protection
in the first place.

One frequent mistake that patent filers make is to pursue the

broadest possible protection in the belief that casting a wide net will
provide the most security. In fact, the reverse is true. I learned this lit¬
tle-understood principle when I was at Litton Industries in the early
1960s.
At the time, there was a British inventor who’d been awarded a
patent on any and all printed circuit boards, believe it or not. He had
thirty companies paying him sizable royalties. IBM and Litton,
though, held out and refused to pay royalties. Litton’s strategy was to
stall and hope that some other company would dispute the patent,
pay the necessary legal fees, and win.
Thank you, IBM. A judge threw out the patent on the grounds
that it was too broad. For years, those thirty companies had been
paying money for an invalid patent.
You see, patents constitute a covenant between inventors and so¬
ciety. Society says, “We will give you exclusivity for a period of time
to produce and sell your product. In return, you will reveal to the
120 Science Lessons

world what you’ve come up with.” Society does not want one per¬
son to wield too much control, though, and rightly so. At some rung
on the legal system ladder, a patent that is too broad will get knocked
down. The ideal patent should be broad enough to prevent other
people from infringing on the claimant’s innovation, and no broader.
It’s wise, though, to apply for multiple patents with multiple
claims; theoretically, as long as you possess one ironclad claim, you
win. Adding others, even if they’re weak, can help reinforce the
power of the first, because judges and juries are likely to feel that a
product embodying more than one invention and protected by more
than one patent is more deserving of exclusivity.

AMGEN AND GENETICS

INSTITUTE GO TO TRIAL

The federal district judge who was set to try Amgen v. Genetics Insti¬
tute in Boston was mired in a complex, high-profile organized-crime
trial with more than a dozen defendants. He indicated that, at most,
he’d be able to devote one day a week to our case—and two hundred
[Link] expedite matters, both sides agreed with his recommenda¬
tion to go before Patti Sarris, then a magistrate. (In Massachusetts, a
magistrate is a local judicial official, and not an actual judge, who
presides over a case under a judge’s supervision. Most states do not
have magistrates.)
The trial opened in January 1989. If you sit in a courtroom long
enough, you’ll conclude that big-time law firms, prominent attorneys,
and major corporations have no compunction about putting forth the
most preposterous positions imaginable. My all-time favorite was ut¬
tered in a different patent dispute involving Epogen. The opposing
company had run afoul of an Amgen patent on using mammalian cells
to manufacture EPO and a second Amgen patent on using vertebrate
cells. Our adversary used human cells to produce EPO; Amgen, as I’ve
mentioned, relied on Chinese hamster ovary cells.
 Partnerships Made in Heaven—and That Other Place 121

Any high school student who stayed awake during biology class
knows that humans are both vertebrates and [Link] opposing
counsel got up in court and with a straight face argued that, in the
context of this patent, humans were neither vertebrates nor mam¬
mals. Amgen won the case.
Genetics Institute’s countersuit contained several equally auda¬
cious claims, not the least of which was its allegation that one of its
scientists had come up with Epogen first. During the trial, Magis¬
trate Sarris determined the exact date on which Amgen’s Fu-Kuen
Lin had invented the drug.“We’ve won!” I thought. GI’s “inventor”
wasn’t yet an employee of the company at that time. This seemingly
insurmountable evidence did not rattle the witness in the least. “I
invented EPO during my job interview at Genetics Institute,” he
testified under oath. Next, the person who’d allegedly conducted
the interview took the stand to attest that, yes, the scientist had in¬
vented Epogen in between asking about GI’s benefits package and
vacation policy.
Admittedly, this account does fall within the realm of legal possi¬
bility. In patent law, the date of invention can be the date that the in¬
ventor “conceived” the idea. However, for the invention to be
patentable, the inventor must exercise reasonable diligence in reduc¬
ing the invention to practice.
All scientists keep hardbound laboratory notebooks in which
they record each day’s findings. A second staff member then counter¬
signs the pages to prevent someone from backdating information.
These notebooks are introduced as evidence in patent-infringement
cases. The Genetics Institute scientist produced no such notebook
dated before ours. Nor did the company document that it had met
the requirement of due diligence.
I was convinced that the case would go Amgen’s way and that
therefore we should fight it to the end. Attorney Steve Odre, head of
our patent department, agreed, but few others did. Every Wall Street
analyst I knew disapproved of our aggressive position and urged us to
122 Science Lessons

accept GI’s offer to cross-license our respective patents. Even some of

our board members were leaning in that direction.
I fully understood their concern^and caution. Whenever a busi¬
ness litigates, it assumes a degree of risk, no matter how sturdy its
case. New evidence can be introduced that undermines your posi¬
tion. And judges are human and make mistakes. What if the magis¬
trate declared both Amgen’s and GI’s patents invalid? Then any
company would be free to manufacture EPO. One thing was certain:
the lawsuit was depressing the price of Amgen stock.
The ongoing dispute produced our first incident of corporate es¬
pionage. An Amgen researcher secretly mailed Genetics Institute
several pages of documents related to Epogen. His accompanying
note said, “This is just a free sample” and promised much more im¬
portant information for sale. It ended, “Where can we meet?”
To GI’s credit, it alerted us. George Rathmann immediately
called the Thousand Oaks police department to report the theft. He
was astonished when the officer manning the desk brushed him off.
“We really aren’t equipped to do much about white-collar crime. If
somebody robs a bank, we know what to do, but we can’t help you
very much. Sorry about that.” Out of frustration, he called the FBI.
At first, the person he spoke to was equally dismissive. “We’re really
busy and have limited resources. Because there’s so much white-col¬
lar crime, we don’t investigate any theft under one million dollars.”
“Fine,” snapped George. “This is worth hundreds of millions of
dollars.” With that, the bureau came on board. One of its agents spent
enough time at Amgen to learn the language of [Link]
he called the suspect, pretending to be from Genetics Institute, and
arranged to meet him at a local restaurant. By the time the thief ar¬
rived, hidden cameras had been installed, and every waiter and pa¬
tron had been replaced by a member of the FBI in disguise. It didn’t
take long for the cameras to capture money changing hands, and
then papers. Soon the suspect was in handcuffs and hearing his Mi-
 Partnerships Made in Heaven—and That Other Place 123

randa rights. He went to jail for two [Link], his theft didn’t
harm our case at all.
The second week in January 1989, Magistrate Sarris upheld
Amgen’s patent but at the same time preserved GI’s [Link] deci¬
sion wasn’t the complete victory we had hoped for; that would come
two years later in federal appeals court. However, it appeared to clear
the way for Epogen’s coming-out party.
Any relief we felt was quickly extinguished, though. On January
23, Ortho Pharmaceuticals, a subsidiary of Johnson & Johnson, filed
a court action requesting an injunction to block the launch of
Epogen. At Amgen, reaction to the news was one of shock. After all,
Johnson & Johnson was our business partner.

AMGEN’S SEARCH FOR PARTNERSHIPS

Amgen’s original business plan had called for enlisting partners to

market our first three products, an established practice that started
with Humulin, the first biotech drug. Genentech produced the syn¬
thetic insulin, but it was sold by pharmaceutical heavyweight EH
Lilly and Company, which boasted major distribution capabilities
and, most important, plenty of money.
We recognized that even though Amgen would probably be a
decade old when its first product was ready, we still would not be
large enough to market pharmaceuticals on our own. Seattle biotech
firm Immunex, founded the same year as Amgen, discovered the dis¬
advantage of being small the hard way in the late 1990s. Immunex de¬
veloped the highly effective arthritis medication Enbrel, only to find
that its modest production plant couldn’t keep up with consumer de¬
mand. Of the more than one million Americans who suffer from se¬
vere rheumatoid arthritis, only seventy-five thousand could get their
hands on the drug. Many more went onto a waiting list for Enbrel.
Business Week estimated that the production lag cost Immunex at least
124 Science Lessons

$200 million in profits in 2000 alone. A year later, the company

agreed to be bought by Amgen for $16 billion, in the largest biotech
acquisition in history.
From the start, we were convinced that we had only a few years
in which to become a full-fledged biotech pharmaceutical firm. If
we failed, we’d be dog meat or, at best, somebody’s acquisition (see
“The Fermenta [Near] Fiasco”). Surely, we thought, the big pharma
corporations would eventually catch on to the promise of biological
medicines and take over the industry. Much to our surprise—and
satisfaction—they waited for more than a decade before leaving the
starting gate, an action that gave the biotech industry the opportu¬
nity to develop.
Amgen’s strategy for geographic growth was as follows: We
would license out the distribution of our first product in each of
three markets—-the U.S., Europe, and Japan. After that, the company
was to be running at full speed and capable of doing everything itself
in all major countries. Unlike many start-ups, Amgen actually ad¬
hered to its global strategy; history shows that it was a good plan.
After Fu-Kuen Lin cloned the EPO gene in late 1983, it seemed
a ripe opportunity to seek out partners. For an up-and-coming
company, partnerships also help establish credibility with stock in¬
vestors and provide healthy infusions of cash.
Although we weren’t even a year removed from our $43 million
initial public offering, Amgen never stopped trying to raise money.
When your future success turns on research and development, there’s
no such thing as enough. Many companies worry about dilution of
their stock holdings and sell fewer shares than they could. Amgen
didn’t concern itself with dilution. The way we saw it, raising twice
as much money would enable us to hire twice as many scientists,
conduct twice as much research, and develop twice as many prod¬
ucts. There really wouldn’t be any dilution.
Furthermore, given the poor showing of our stock, along with
most other biotech stocks, we couldn’t be certain that Amgen would
 Partnerships Made in Heaven—and That Other Place 125

The Fermenta (Near) Fiasco

Numerous companies launched during the first wave of biotech firms
were eventually swallowed by other firms. To date, only one potential
buyer has ever contacted Amgen.
In 1985, Refaat el-Sayed, a thirty-eight-year-old Egyptian business¬
man by way of Sweden, came calling. In only four years, the charis¬
matic el-Sayed had amassed the second-largest fortune in Sweden by
buying up money-hemorrhaging drug companies, starting with one
called Fermenta. Swedish national TV named him the country’s man
of the year.
George Rathmann was intrigued by Fermenta’s proposition and in¬
vited el-Sayed to an Amgen board meeting to explain what he had in
mind. The directors were almost universally turned off by the man, in
part because he brought with him an executive from another Swedish
company to sing his praises. It seemed odd. Neither I nor the other
senior managers were in favor of collaborating with el-Sayed, so noth¬
ing came of it.
It was a good thing, too. A few months later, el-Sayed was exposed
as having fraudulently misrepresented himself as a microbiologist
holding multiple PhDs. The accusations, which he admitted were true,
sank a proposed merger with Volvo, Sweden’s largest company, and
forced el-Sayed to resign as Fermenta president. By 1987 he had filed
for bankruptcy, with debts of $60 million.

be able to conduct a second public sale very soon. Conventional wis¬

dom says that bankers or investors won’t accept a second offering at a
price below the IPO. To them, it’s a bad sign, indicating that the
company is heading in the wrong direction. Amgen had a long way
to go to climb back to $18 per share.
126 Science Lessons

We were always extremely disciplined financially. Certain things

were out of our hands, such as unanticipated lawsuits from partners
and competitors or an industrywide (drop in stock prices. But what¬
ever we could control, we did, like making sure that our plans for ac¬
quiring new buildings and personnel were in line with our financial
capabilities. Three fiscal years in a row, Amgen showed a profit—not
bad for a company without any products. Our aim was to get our
stock price up and earn at least one penny per share each quarter, be¬
cause there’s magic in being in the black. Why one cent? We wanted
to spend the extra money on R&D, and earning two or three cents
per quarter wouldn’t have been any more beneficial than a one-cent
[Link] made sense to me then, and it makes sense now.
Our search for a corporate partner started at home. Much to our
shock, not a single U.S. pharmaceutical firm showed interest. Perhaps
we shouldn’t have been mystified; if you’ll recall, before our initial
public offering, Abbott Laboratories, one of Amgen’s original in¬
vestors, had the opportunity to be involved in the Epogen project.
CEO and chairman Bob Schoellhorn turned it down. He’d been in¬
fluenced by Abbott’s chief chemist, who apparently didn’t think much
of drugs based on large proteins. As we would discover, that bias was
not unique to Abbott; in fact, it dominated traditional pharma.
One company’s representative informed us that his bosses were
passing on Epogen because the opportunity was too small; their mar¬
ket research department predicted that sales would never eclipse $50
million per year. (For the record, the drug generates $10 billion in
annual revenue. Some market research!)
Next, George Rathmann and DanVapnek went to Japan several
times to search for an overseas partner there, while Noel Stebbing
and Phil Whitcome, our director of strategic planning, made multiple
trips to Europe. All four came home astonished that no one else
seemed to share our unshakable faith in biotechnology-based phar¬
maceuticals. Not one of the five products we had in the clinic or
headed for the clinic elicited so much as a nibble.
 Partnerships Made in Heaven—and That Other Place 127

“Surprised is an understatement,” George Rathmann recalls,

chuckling at the memory.

We couldn’t crack the European market, because they were

all looking to buy cheap. Whatever your figure was, they
wanted to pay significantly less. For instance, we met with
Hoechst Marion Roussel [now part of Sanofi-Aventis Phar¬
maceuticals] about partnering with us on [Link] asked
how much we wanted for it.
I said, “Well, we started with the idea that we’d get $10
million plus a 10 percent royalty on all sales.” They offered
$6.5 million and a 6.5 percent royalty, then added that Hoechst
would expect to credit the milestones that it paid against fu¬
ture royalties.

Then there was the fellow from a British pharmaceutical firm

who flew across the pond for a meeting about the possibility of his
company’s licensing Amgen products. About halfway through, he
grumbled, “Wait a minute! Are you telling me that all of these drugs
are injectable?”
As mentioned earlier, they had to be injected, either into a vein
or a muscle, or subcutaneously, just beneath the skin. As large pro¬
teins, they couldn’t be administered orally, because if you swallowed
them, the enzymes in your digestive tract would chop them up and
digest them. Oral drugs, or small-molecule drugs, as we called them,
were designed to resist the enzymes so that the stomach or intestines
could absorb them and send them along to the bloodstream.
The man stood up and snapped, “Well! I wish you had told me.
You could have saved me a trip across the ocean!” And with that, he
stomped out of the room; he didn’t even stay for the rest of the meet¬
ing. He knew that injectable medicines generally had much lower
sales than those taken by mouth; therefore, his company wasn’t about
to license any of our offerings, including Epogen.
128 Science Lessons

As for Japan, its once blossoming market for finding licensing

partners had been effectively trampled by Genentech, our San Fran¬
cisco rival. From 1981 through 1983, nearly twenty U.S. biotech
companies had entered into collaborations with Japanese companies,
Genentech being one of the most active. Rathmann explains:

Genentech had cut several deals with Japanese companies for

huge sums of money, like $20 million for one of its recombi¬
nant interferons. These large payments were just for the rights
to sell its products in Japan, a market that would be hard for an
American company to enter anyway.
So we thought that we’d have no trouble getting $20 mil¬
lion, too. That was the benchmark. We met with a small Japa¬
nese pharmaceutical firm and were pitching the chief executive
on making a significant equity investment, when he suddenly
became highly animated and started ranting in Japanese,
pounding his hand on the conference table. Afterward, we
learned what he had said: “Twenty million dollars? Amgen
thinks it’s worth twenty million dollars for something that
doesn’t even resemble a product yet? It’s not worth it!” Ironi¬
cally, Amgen was further along with some of its programs than
Genentech was when it signed that $20 million deal, but by
1984 that rose had lost its bloom.

AMGEN’S PARTNERSHIP WITH KIRIN

As it turned out, that wasn’t entirely the case. Out of nowhere, we re¬
ceived a phone call from Japan’s Kirin Brewery Co. It seemed that an
employee there had the job of poring through the U.S. and German
press and clipping articles of interest for her superiors. Why U.S. and
German? Because those were the two foreign languages she knew. A
news item about Amgen’s discovery of EPO caught her attention,
 Partnerships Made in Heaven—and That Other Place 129

and she forwarded it to her boss. Kirin, looking to enter the biotech
field, had been trying to develop human recombinant EPO but
without any success.
Kirin faced the kind of financial problems that any business
would love to have. It was highly profitable and generated cash faster
than it could use it. To make things “worse,” it was too successful in
its primary business, which was beer. Japan’s federal antitrust system
is less formal than that of the United States; when Kirin’s share of the
nation’s beer market approached 60 percent, the government, instead
of filing a lawsuit, quietly discouraged the company from getting any
bigger in that line of business. Brewery executives decided to funnel
some of their excess cash into marketing hard liquor in a joint ven¬
ture with Seagram. The new enterprise was a failure in that it be¬
came a highly profitable cash cow right away. Next, Kirin entered
the soft-drink market and “failed” again. Money continued to flutter
down upon Kirin like cherry blossoms in the spring.
What’s a poor company to do? The answer, Kirin decided, was to
start a business that would consume huge amounts of yen and take
many years to creep into the black. That’s how Kirin’s pharmaceuti¬
cal division was born. When the people there learned that Amgen
scientists had discovered the elusive EPO gene, they made the fateful
overseas phone call.
Soon afterward, a contingent of executives flew to Thousand
Oaks to serenade us, and in no time at all, the two companies were
almost falling into each other’s arms. We had much in common, such
as fermentation know-how. Kirin also understood the importance of
R&D. It was probably the only beer maker in the world to boast a
bona fide science-based R&D department. To most beer manufac¬
turers, brewing was an art.
Before inking a deal, however, the folks from Kirin wanted to be
certain that the cultures of our respective companies and countries
were harmonious. None of Kirin’s people knew how to play golf
very well, and, probably somewhat to their surprise, neither did we.
130 Science Lessons

Nevertheless, about twelve of us went out on the links together to

watch one another miss easy putts and hook drives into the [Link]
took a traditional communal bath together. Fortunately, Rathmann,
Weist, and I all loved Japanese food and were reasonably adept with
chopsticks. I did, however, draw the line at eating raw chicken.
Once we learned such arcane Japanese protocols as which person
gets on the elevator first, last, and so on, we started being less formal
with one another. By then, the Kirin executives had seen that the
heads of Amgen had demonstrated respect for their customs. The
Kirin people, in turn, showed a willingness to learn from us and re¬
spect our customs as well. As one of their executives put it succinctly,
“We cannot become Americans, and you cannot become Japanese.
So we both have to move toward the middle, or else this is not going
to work.” It seemed clear to everyone that compromise would not be
a problem for this partnership.
During the subsequent negotiations, we were pleasantly sur¬
prised to discover that, as per Japanese tradition, deals are conducted
with a minimum number of attorneys. Contracts are unusually short,
too. The Japanese principle is that a business relationship should al¬
ways be fair to both sides. Consequently, if the terms of the original
agreement cease to be fair to one side, because of changing circum¬
stances or whatever, the terms must be amended accordingly. It is not
exactly a Western concept, but a wise one.
A Japanese contract contains phrases such as, “The division of the
cost of the advertising program will be negotiated later.”
Later?
The rationale is this: we don’t know enough right now to make a
fair allocation. So we’ll figure it out when the time comes.
For example, one of Amgen’s contractual obligations to Kirin was
to manufacture EPO for all preclinical and clinical testing—as
you’ve seen, a time-consuming process. A few years later, one of our
scientists came up with a method to produce everything we needed
 Partnerships Made in Heaven—and That Other Place 131

for a fraction of the $2 million earmarked for this purpose. Thinking

like true Americans, we fretted that we would have to give back the
considerable difference. It seemed a bit unfair to be penalized for
having made a scientific breakthrough. When we raised the issue
with our partner, the head representative looked right at us and said,
“You take the full $2 million .’’Just like that.
What he didn’t say was that there was now an invisible IOU—a
rather large one—sitting on the table, and at some point in the future
it would have to be repaid. In other words, the reverse situation
would arise eventually, and we’d be expected to reciprocate with
equal generosity. The system works, but only if both parties are mu¬
tually trustworthy and flexible like Amgen and Kirin.
A joint venture was struck in May. Kirin obtained the rights to
sell Epogen in Japan; we retained full control of the United States.
Europe was to be shared. Our partner put up $12 million; Amgen
chipped in $4 million, largely as a sign of good faith, along with $8
million in EPO-related patents and technology. The influx of licens¬
ing money helped us step up the pace of research and development
on Epogen. If we’d collaborated with a U.S. company instead of
Kirin, I firmly believe that Amgen would have wound up having to
split the U.S. market, at the very least.
We threw a celebration dinner for both companies at the exclu¬
sive Jonathan Town Club, a century-old establishment housed in a
magnificent Italian Renaissance-style building in the heart of down¬
town Los Angeles. I stopped in early to see how the preparations
were going, as did one of Kirin’s regional sales managers. To our dis¬
may, we discovered that the club didn’t have a drop of Kirin beer.
Fortunately, the sales manager was able to call a nearby hotel and
convince the head bartender to send over several cases right away.
They arrived just before our guests did. I took it as a good omen, and
to this day the collaboration between Amgen and Kirin stands as a
model of cooperation.
132 Science Lessons

THE AMGEN PARTNERSHIP

WITH JOHNSON & JOHNSON
v
The same cannot be said about our alliance with Johnson & John¬
son, which soured almost from the outset. George Rathmann bro¬
kered the September 30, 1985, product licensing agreement, and by
his own admission, selecting Johnson & Johnson as a partner remains
the one serious blunder Amgen has made. We sold its Ortho division
the rights to sell EPO everywhere in the world except Japan and
China, leaving us the one hundred thousand American kidney dialy¬
sis patients suffering from chronic anemia.
On August 19, 1988, the two companies signed an agreement to
work together on three major drugs, one of them EPO. We would
sell Epogen; Johnson & Johnson would sell its own brand of epoetin
alfa, Procrit, which Amgen would manufacture. Less than six months
later, Johnson & Johnson and Ortho ambushed us with the request
for an injunction in the U.S. District Court of Delaware. According
to Johnson & Johnson, Procrit would suffer “irreparable harm” if
Epogen hit the market first. We failed to see how, considering that
our drug was intended exclusively for the dialysis market, and Procrit
had a lock on every market except dialysis. Johnson & Johnson asked
the court to postpone Amgen’s planned launch, tentatively set for
late March or early April, indefinitely. Bob Weist took the phone call
from our partner informing him of its intentions:

It was a Friday afternoon. Our good friends at Johnson &

Johnson told me that if we didn’t do this, that, and the other
thing, they were prepared to take immediate legal action.
They were basically placing a gun to our heads, especially
since Gordon and George Rathmann were both out of the
office. George was home recuperating from back surgery,
while Gordon had just left the hospital following surgery to
remove a melanoma, a life-threatening form of skin cancer.
 Partnerships Made in Heaven—and That Other Place 133

I asked if we could have until Monday to respond, but

they didn’t find that satisfactory.

The lawsuit was filed the same day.

It was hard to believe that the same company whose corporate
credo begins “Our first responsibility is to the doctors, nurses and pa¬
tients, to mothers and fathers, and all others who use our products and
services” would pursue a legal outcome that, if successful, would tem¬
porarily deny more than one hundred thousand dialysis patients the
only effective treatment for their condition. To this day, contempt for
Amgen’s former partner runs so deep that many employees proudly
proclaim their homes to be 100 percent “J&J free.” Considering that
Johnson & Johnson and its many businesses sell more than one thou¬
sand products, from Band-Aids to Tylenol, that’s no small feat.
The hearing, in Wilmington, Delaware, was scheduled for March
17—St. Patrick’s Day—giving our attorneys a scant six weeks to
mount a defense, whereas Johnson & Johnson’s move probably had
been in the works for months, maybe even years. This opening salvo
of litigation hurled so much shrapnel into the air that debris would
still be crashing to the ground as recently as 2003. Let me whittle the
basic complaint down to its core.
Ortho contended that Amgen had breached the recently signed
agreement because we had not worked hard enough to help it get
Procrit to the market. Our position was that, to the contrary, we’d
met every one of our contractual obligations and then some. If any¬
thing, Ortho was delinquent in doing what it was supposed to in
order to move its drug along toward licensure—a clear violation of
the contract. As it turned out, Procrit did not receive FDA approval
until late 1990, eighteen months after Epogen did. Even we didn’t
realize that Johnson & Johnson was so far behind.
Amgen submitted a counterclaim requesting that the license
agreement be terminated. We’d learned that our partner’s clinical tri¬
als of Procrit were recruiting patients with chronic renal failure, the
134 Science Lessons

one territory that we believed contractually belonged to us. We be¬

lieved that Johnson & Johnson was supposed to be testing its brand
of EPO only in nondialysis patients, such as people with cancer
whose red-cell counts had fallen as a result of chemotherapy.
Many people were highly critical of our deal with Johnson &
Johnson, though no one more so than we. What I regret aren’t so
much the terms but the fact that we didn’t do a thorough job of re¬
searching Johnson & Johnson’s history when it came to partnerships
(see “20/20 Hindsight”). You could argue that we gave away too
much, and maybe we did. But at the time of the joint venture, in
1985, we needed cash flow to get the necessary clinical trials of
Epogen under way. Teaming up with Johnson & Johnson, a highly
respected company, brought in a quick $6 million, enabling us to
achieve our goal of showing a profit for the first time ($548,000 on
$23.4 million in revenues), for the 1985—1986 fiscal year. Ironically,
in 1986 our stock rebounded sufficiently that we were able to exe¬
cute our second public offering after all, raising $35 million, and a
third financing the next year brought in $120 million.
The part that rankled me most was Johnson & Johnson’s drag¬
ging this matter into federal court. In negotiating the pact between
our two companies, we’d agreed to take all disputes to arbitration as
an alternative to expensive, protracted lawsuits with endless appeals.
We’d even specified a neutral city, Chicago, where the hearings
would take place.
Another advantage of arbitration is that details of the case are
kept confidential. That can be crucial if one or both companies are
publicly owned. Every time a document is filed with the court, it’s
out there for everyone to see—security analysts, business reporters,
the general public, and your company’s customers and vendors—and
that tends to keep the litigation perpetually in the news.
Our dispute taught me a valuable lesson that most attorneys
know but few business executives realize: under some circumstances,
 Partnerships Made in Heaven—and That Other Place 135

20/20 Hindsight
One of George Rathmann’s famous sayings was, “Success is the ability
to survive your mistakes.” Had those of us in management done a
more thorough job of researching Johnson & Johnson’s previous part¬
nerships, I believe we would have uncovered enough evidence of
problems that we would have chosen a different business partner.
When I speak in front of an audience of businesspeople about
partnerships, I ask, “How many of you have hired a senior executive in
the past year?” A bunch of hands go up. “How many of you checked
references on the person you hired?” About the same number of
hands go up.
“Now, how many of you have signed a major corporate partner¬
ship deal in the past year?” Lots of hands. “How many of you checked
references on your corporate partner?” No hands—but a lot of sheep¬
ish expressions. You would think that all companies research prospec¬
tive partners as thoroughly as they do potential new hires. But they
don’t. We didn’t, either.
One lesson we learned from our happy experience with Kirin
Breweries was that the most important factor in any partnership is
ethics. If your partner is untrustworthy, there’s no contract binding
enough to protect you. But at the time we entered into our early
partnerships, we didn’t give much thought to how ethical the com¬
panies were. We were more focused on the size of their sales force,
the brilliance of their marketing, and so on. We learned rather
late—and the hard way—that ethics should have been at the top of
the list.
136 Science Lessons

an arbitration clause cannot prevent a party to a contract from by¬

passing arbitration and taking its case to court.
What was a skirmish for Johnsoti & Johnson was a life-and-death
struggle for Amgen. That may have worked to our advantage. I can
remember Rathmann’s being contacted at one point by a business¬
man whose company had been burned by Johnson & Johnson in an¬
other partnership deal. He despised the large corporation so much
that he offered to help us however he could.
“Chin up!” he said.“You’re going to win.” How could he be sure?
“Because,” he said,“when you’ve driving to work in the morning, this
is all that’s going to be on your mind. You’ll think about it when
you’re driving home, eating dinner, and in the shower. They won’t be.
This is no big deal for [Link]’s why you’ll find a way to win.”
It sounds like a pep talk delivered by a football coach at halftime.
But the fellow was right. In April 1989 the hearing judge dismissed
Johnson & Johnson’s lawsuit, saying that it belonged in arbitration and
not in district court, just as the partnership agreement stipulated. Be¬
cause an arbitrator does not have the authority to issue an injunction,
Ortho’s bid to block Epogen’s arrival dissolved like a sugar cube in
the rain. Although the legal maneuvering succeeded in delaying the
launch by two months, for Amgen the judge’s decision removed the
final obstacle to getting our first drug to patients and their doctors.

EPO GOES TO MARKET AT LONG LAST

Correction: there was still one hurdle left to clear, albeit a small one.
The Food and Drug Administration notified us that it was going to
license Epogen for use effective June 1, 1989, a Thursday. The FDA
also must approve all patient and physician information that is to ap¬
pear on the box it comes in, as well as the package insert—a folded
sheet of paper crammed with essential details concerning dosage,
side effects, and so on.
 Partnerships Made in Heaven—and That Other Place 137

Typically, pharmaceutical companies take about thirty days from

the approval date to introduce a new product. It’s a hectic four
weeks, with a national sales meeting to arrange, sales training to im¬
plement, and sales aids to print, among still other preparations.
Amgen couldn’t afford a leisurely approach; we needed Epogen to
start earning money right away.
“We are going to have this product ready to sell within twenty-
four hours of approval,” I told everyone. “That’s the goal. Now let’s
figure out what we have to do to make it happen.” Looking back, I
have to laugh at how naive we all were. We went ahead with the plan,
blissfully unaware that it bordered on the impossible.
In hopes of saving time, we asked the FDA to give us the go-
ahead on the language for the box and the package insert a few days
in advance. In that way, we could start printing. Sorry, came the reply,
can’t do it. However, the agency informed us off the record that it
was going to make us change some words—although which ones, it
couldn’t say.
We took a chance and had the vial labels, boxes, and package in¬
serts printed exactly as we’d submitted them, in case the FDA
changed its mind and approved the wording. Then, to cover our¬
selves, we paid a local printer to reserve an idle press for us—with its
crew standing by—so that printing could commence as soon as we
knew what had to be [Link] went on for days.
Finally, the FDA revealed the required rewording responsible for
all this tension and commotion: an a had to be changed to a the. Or
was it a the to an a? I can’t remember. At the time, though, with
twenty-four hours to go before the launch of our first product, it was
a critical change.
The warehouse at Thousand Oaks was piled high with tens of
thousands of boxes of Epogen. Every one of them had the wrong
word printed on it. Before we could ship the drug, we had to remove
each vial from its box by hand and place it in the newly printed box.
138 Science Lessons

Amgen scientists and laboratory personnel volunteered to stay

through the night to get this done. In fact, so many of our family of
650 offered to help that we had to turn people away.
Pizza was brought in, and the partylike atmosphere continued
until Thursday morning. The warehouse echoed with laughter and
lots of kidding around. When the sun rose, sure enough, we were
ready to ship the product. We had met our goal. Thirty days was the
industry standard. But for us, even forty-eight hours would have been
considered a failure. It was a matter of pride. Everyone departed into
the sunlight, looking forward to a well-deserved product launch.
After eight years and $300 million, we’d done it.
CHAPTER

Red Hot Summer

the launch of epogen in the summer of 1989 changed every¬

thing. After three consecutive years of posting a profit, ramping up
for the big debut had left Amgen $8.1 million in the red for fiscal
year 1988—1989. The drug earned $20 million its first day on sale—
exactly the cost of building 6, our first large-scale production facility,
built expressly for producing EPO. When construction started in
1986, the $20 million expenditure had seemed like a huge gamble,
especially because it represented about 80 percent of the company’s
liquid assets. But the plant would manufacture $8 billion worth of
epoetin alfa before being mothballed in 2003.
Fortune named Epogen the number 1 product of 1989, a year in
which our profits more than tripled, to $19.1 million. In 1990 they
nearly doubled, to $34.3 million. Total sales for both years: $580.3
million. Quite an impact for a product targeted to a market that
dozens of other companies had scorned as too small. Epogen’s im¬
pact on the lives of chronically anemic kidney-dialysis patients,
though, was incalculable.
140 Science Lessons

On August 25, 1990, Amgen entered the NASDAQ exchange,

which sent our stock rocketing to $149 per share following a two-
for-one split. If you’d purchased o»ne thousand shares when it had
bottomed out five years before, your $3,750 investment would have
grown to roughly $285,000. Meanwhile, Kirin was generating addi¬
tional income, having received clearance to distribute Epogen in
Japan. Then, on the first business day of 1991, our other partner,
Johnson & Johnson, finally won a license from the well-established
FDA to sell its Procrit brand (manufactured by Amgen) in the United
States. The approved indication was for managing anemia related to
the antiviral drug zidovudine (ZVD), then the first line of defense
for people infected with the HIV virus.
The most dramatic developments surrounding Epogen, however,
were taking place in Washington, D.C. On January 15,1991, the FDA
rejected Genetics Institute’s petition that it overturn Amgen’s seven
years of exclusive rights to Epogen under the Orphan Drug Act and
allow GI to sell its version of epoetin alfa, to be called [Link]
decision left our foe from Massachusetts pinning its hopes on the
ongoing patent-infringement case, which, after two years of split de¬
cisions, now rested in the hands of a U.S. Appeals Court.
In March the three-judge panel upheld Amgen’s patent on
Epogen, effectively ending GI’s chances of getting Marogen on the
market in the United States, at least until the last claim of the original
patent expires in 2013. The New York Times called the decision “a
stunning victory,” noting that Gl/Chugai had been expected to
make significant inroads into Amgen’s market—as much as 20 per¬
cent, according to some industry analysts. Paradoxically, GI’s at¬
tempts at infringement kept Marogen out of the marketplace longer,
because if not for its lawsuit, all of Amgen’s patents would have ex¬
pired sooner.
Naturally the news was greeted ecstatically at Amgen, which had
recently undergone a radical transformation. Chairman of the Board
George Rathmann no longer reported to building 10 daily as he had
 Red Hot Summer 141

since resigning as CEO. In December 1990 George severed his re¬

maining ties to the company. It marked the end of an era, and all 850
staff members attended his retirement dinner to wish their beloved
leader well.
A few weeks after I assumed the chairmanship, Harry Hixon left
Amgen. Our two years of working in tandem had been difficult,
though by no means unbearable. It may have been a positive experi¬
ence for me, giving me time to learn a lot of what a CEO needs to
know. However, by 1991 it was apparent that if the dual setup con¬
tinued much longer, it would start to become divisive, especially
with Rathmann the peacemaker now gone. It wasn’t Hixon’s fault,
and it wasn’t my fault. We had a lot of respect for each other and still
do, but the situation just wasn’t workable.
I felt so strongly about it that I went to the board and told the di¬
rectors that they should choose one of us to run the company, and
soon. Otherwise we’d reach the point where people had to start
choosing sides, and that, of course, could damage the company irre¬
versibly. The board decided that I should be the one to stay.
My daily routine changed drastically. We decided to hold off re¬
placing Harry for a while, so all the people who’d reported to him
now reported to me in addition to those already under my supervi¬
sion. I had to get more deeply involved in areas such as sales and
marketing, which previously had been managed by Hixon.
The biggest change of all was that on February 21,1991, Amgen
became a two-drug company.

SOMETHING NEU: NEUPOGEN

In a sense, Amgen had been a two-drug company ever since Fu-

Kuen Lin successfully cloned human erythropoietin: from then on,
there was Epogen, and there was everything else. Call it classic R&D
management. When you have one high-priority product, decisions
regarding the allocation of resources become simple. Basically, the
142 Science Lessons

Epogen team got whatever it needed, whether human resources,

equipment, or Rathmann’s time.
The Neupogen project, which Crystallized in 1986, was spear¬
headed by Amgen molecular biologist Dr. Lawrence Souza. Neu¬
pogen began life known as G-CSF, which was short for granulocyte
colony-stimulating factor. Like epoetin alfa, G-CSF is a naturally oc¬
curring protein. But instead of stimulating the bone marrow to
churn out red blood cells, G-CSF hikes production of a type of
white blood cells known as neutrophils, which are the immune sys¬
tem’s principal defenders against bacterial infections.
One hallmark of cancer is frenzied, uncontrolled cell division.
The toxic chemotherapy drugs used to treat cancer destroy many of
these cells. Unfortunately, other cells in the body—healthy ones—
exhibit rapid growth, too. These include the cells that make up the
delicate lining of the digestive tract, from the inside of the mouth
down to the intestines; hair; red blood cells; and neutrophils. Most
anticancer agents cannot distinguish fast-dividing benign cells from
malignant ones, and that is why chemotherapy often, although not
always, gives rise to side effects such as sores in the mouth and throat,
hair loss, and abnormally low levels of neutrophils, a condition called
neutropenia (see “Different Grades of Neutropenia”). Consequently,
for a week or so following drug treatment, patients may become
highly susceptible to infection until the neutrophil level recovers.
The harm from neutrophil depletion is twofold: not only is an
immunosuppressed person less equipped to fight bacteria, but if her
absolute neutrophil count does not rebound in time for the next
drug cycle, her oncologist may be alarmed enough to postpone ther¬
apy, reduce her drug dosage, or both. In a national survey of cancer
patients, nearly half had at least one chemotherapy treatment delayed
because of severe neutropenia. The interruption could conceivably
allow the cancer to progress.
Souza, a graduate of the University of California, was collaborat¬
ing with scientists at New York’s Memorial Sloan-Kettering Cancer
 Red Hot Summer 143

Different Grades of Neutropenia

The blood test used to diagnose neutropenia is called an absolute
neutrophil count, or ANC. Without going into detail about what the
numbers mean, a normal ANC ranges from 1,800 to 8,000. A person
with an ANC of 1,000 to 1,800 is said to have mild neutropenia, with
low odds of developing an infection. Therefore, most patients in this
range do not have their next cycle of chemotherapy delayed or their
dosage lowered. Moderate neutropenia (500 to 1,000) carries a mod¬
erate risk of infection, while severe neutropenia (499 or less) places
patients at high risk of infection.

Center. By the late 1990s, Amgen had about three hundred such
arrangements with researchers around the country. Individual scien¬
tists initiated the process, and Larry’s collaboration with MSKCC
turned out to be a major part of Amgen’s ultimate success. He even¬
tually succeeded DanVapnek as head of research.
Memorial Sloan-Kettering possessed what I called “Mrs. Mur¬
phy’s Chowder”: a mixture of about two hundred proteins. But it
didn’t have the technology to separate them. Amgen did. The cancer
center suspected that this protein hodgepodge held stem-cell growth
factor. Today you hear a lot about stem cells, the undeveloped “seed”
cells that mature into red cells, white cells, or platelets. Stem cells are
found in the bone marrow as well as in the peripheral blood vessels.
At that time, few people, including scientists, knew much about them.
As it happened, the sample contained not stem-cell growth fac¬
tor but a much more valuable growth factor: granulocyte colony-
stimulating factor. Souza and his team pursued this lead vigorously
and discovered the human gene that produces G-CSF, located on
chromosome 17. Once isolated, the gene was cloned using the same
144 Science Lessons

process as for human EPO, but with E. coli bacteria as the repheating
organism instead of Chinese hamster ovary cells.
Memorial Sloan-Kettermg had filed a weak patent, not knowing
what it actually had. Therefore, said my general counsel, Amgen
was legally free to proceed on its own, without paying a royalty to
MSKCC. That didn’t seem ethical to me; without Sloan-Kettering,
we wouldn’t have stumbled across filgrastim (Neupogen’s generic
name, pronounced fill-grass-tim).We negotiated a license with a mod¬
est royalty.
Months after Amgen’s discovery, our erstwhile nemesis Chugai—
which, despite everything, I considered to be Japan’s best pharma¬
ceutical company—isolated and cloned filgrastim independently.
Suddenly there were three patents on file vying for the rights to the
protein. Chugai indicated that it intended to litigate.
Frankly, we had Chugai over the proverbial barrel: as the licensee,
Amgen held the original Sloan-Kettering patent, so if a judge or ar¬
bitrator chose the victor based on which organization had applied
first, we would win. What’s more, the Amgen patent, though filed
after the Japanese drugmaker’s, was much more complete, so if the
judge based the decision on that criterion, again we would win.
During our previous legal tussle, over Epogen, I’d become
friendly with Chugai’s CEO, whom I admired very much. Realisti¬
cally, he knew that Chugai had little chance of beating us. We
reached an agreement allowing both companies to sell Neupogen in
Europe, with Amgen retaining exclusive rights to the U.S. market.
Negotiating instead of htigating saved both sides many millions in
legal fees and resulted in a fair outcome for everyone. Without the
mutual respect between the two companies, that probably wouldn’t
have been possible.
The initial human trials for Neupogen took place at Memorial
Sloan-Kettering Cancer Center and at the Ludwig Institute for
Cancer Research in Melbourne, Australia. We didn’t have any em¬
ployees Down Under at the time, but the world’s greatest experts on
 Red Hot Summer 145

colony-stimulating factors were there, so it seemed a logical choice

for a science-driven organization. Typically, drug manufacturers
choose one medical center over another to run their clinical trials
because it is conveniently located or charges less. We wanted to work
with the leading authorities wherever they might be located and
whatever the cost.
Dr. George Morstyn, head of the Ludwig Institute’s clinical pro¬
gram, stood out as exceptionally talented. He was also unusual in
that he was both a PhD and an [Link] recognized right away that
he would make a terrific addition to Amgen. After a year of wooing
by Kirby Alton—and many thirty-hour plane trips—in 1991 George
joined the company as vice president of medical and clinical affairs.
Designing the patient trials for Neupogen was trickier than it had
been for [Link] couldn’t attempt to prove that the drug would
help people with cancer live longer, because that would entail a pro¬
hibitively expensive and time-consuming study. Merely demonstrat¬
ing an increase in the number of circulating neutrophils—-the
objective of the early investigations—wouldn’t be ambitious enough
to satisfy the FDA when it came time for phase III.
“We had to show that Neupogen could decrease the incidence of
infection associated with chemotherapy,” explains Kirby. “The prob¬
lem was, it can be hard to document infection in cancer patients
using the standard laboratory tests. After negotiating with the FDA,
we came up with an acceptable end point: febrile neutropenia.” Suc¬
cess would be measured by a significant reduction in the number of
chemo patients who developed neutropenia accompanied by fever.
In a phase III trial carried out at various centers, 210 men and
women with lung cancer were randomly assigned to two groups. For
ten days during each three-week chemotherapy cycle, patients re¬
ceived either Neupogen or a placebo. Because it was a double-blind
study, neither they nor the investigators knew whether they were
being injected with the active ingredient or a harmless [Link]
results were impressive, as shown in table 7-1.
146 Science Lessons

TABLE 7-1

Results of Neupogen tests

v
Study group Control group
Result (Neupogen) (placebo)

Percentage of patients who developed 40 percent 76 percent

at least one infection following
chemotherapy ,

Percentage of patients who needed to 52 percent 69 percent

be hospitalized due to infection

Percentage of patients who required 38 percent 60 percent

treatment with intravenous antibiotics

Percentage of patients who developed 57 percent 77 percent

severe neutropenia over the entire
course of chemotherapy

Average number of days patients were First cycle: 2 days First cycle: 6 days
neutropenic All cycles: 1 day All cycles: 3 days

For neutropenic patients, average First cycle ANC: 72 First cycle ANC: 38
severity of neutropenia as measured All cycles ANC: 403 All cycles ANC: 161
by absolute neutrophil count (ANC);
the higher the number, the better

In the late 1980s, there were perhaps fifty chemotherapeutic drugs,

some of them more immunosuppressive than others. Because different
forms of cancer call for different agents, our Neupogen trials covered a
range of malignancies, including breast, ovarian, kidney, and neuroblas¬
toma, a type of cancer that affects the nervous system in children.
What we didn’t want was for the FDA to approve the use of Neu¬
pogen only for some cancers or only with certain anticancer drugs.
On February 21,1991, word came that we’d received the broader in¬
dication after all: for any chemotherapy that suppresses bone marrow,
the body’s blood-cell factory.
 Red Hot Summer 147

A NEW BREED WALKS THE HALLS OF

AMGEN: SALES REPS

Neupogen gave Amgen its second grand slam in a row. The drug
racked up $233 million in sales its first year—the fourth-highest
debut in the history of the pharmaceutical industry (Epogen being
second)-—and was named product of the year for 1991 by Fortune.
Our profits nearly tripled from the year before, to $97.9 million. In
1992, the year that we leaped into the Fortune 500 and the S&P 500,
Amgen rang up $1.1 billion in total sales, for a profit of $357.6 mil¬
lion. We were now an international company, too, with marketing
and sales operations in Europe, Australia, and China.
For the next few years, our focus shifted, as it had to, toward es¬
tablishing our two products. Epogen, to an extent, sold itself. One
key difference between Epogen and Neupogen was that no accept¬
able alternative existed for Epogen’s target: managing chronic ane¬
mia in people on kidney dialysis. A low red-cell count didn’t rally on
its own; patients would need transfusions, which if repeated too
often could trigger complications in the future.
In contrast, not all oncologists saw the urgency in reversing neu¬
tropenia in cancer patients, even though this condition could in¬
crease the chances of infection and possibly force chemotherapy to
be postponed or the dose reduced. It was our sales staff’s job to bring
to their attention studies concluding that men and women with
cancer fared better when full chemotherapy dosing was maintained
consistently.
Another difference between Neupogen and Epogen was that
dialysis patients could feel EPO working; the drug was infused into
their dialyzing machine, or they injected it under their skin, and soon
they had much more energy and were able to function again. In con¬
trast, Neupogen didn’t make patients feel better; in fact, it allowed
their oncologists to hit them harder with the cancer-killing drugs.
148 Science Lessons

Many chemo patients don’t develop severe infections anyway. So

if a physician treats a large number preventively, or prophylactically,
some of those men and women woVild have done just as well with¬
out Neupogen. There’s no way to know which ones stayed healthy
because of the drug. It’s no different from chemotherapy itself, which
for many cancers is an adjuvant, or add-on, treatment; after surgery or
radiation therapy to remove the tumor or as much of it as possible,
the systemic drugs are given to hunt down rogue malignant cells that
may be lurking in the body. Most patients, though, when faced with
a life-threatening illness such as cancer, are willing to take advantage
of any treatment that may improve their chances, no matter how
small, of staying alive or beating the disease.
If you were a sales representative for Neupogen, you’d want on¬
cologists to know that the drug halved the risk of infection associ¬
ated with chemotherapy and greatly reduced the need for inpatient
antibiotics. That’s a substantial benefit, because the last place an im¬
munocompromised person wants to be is in a virtual petri dish like
a hospital, where every year about two million Americans acquire
clinically significant infections. Also, by cutting down on hospital¬
izations, Neupogen more than paid for itself, a winning feature
from the standpoint of health-maintenance organizations and insur¬
ance companies.
Most folks probably assume that physicians are routinely brought
up to speed on the latest medicines, their side effects, the results of
clinical trials, and so on. But that’s not necessarily the case. Once
doctors graduate from medical school, there is no formal system for
continuing instruction; unlike, say, accountants or schoolteachers,
physicians need not meet an educational requirement to retain their
licenses. Not only do drug companies fund the bulk of medical re¬
search and development in the United States, but their sales repre¬
sentatives serve as most physicians’ primary source of information
about new products for patients.
 Red Hot Summer 149

Pharmaceutical sales is something of a misnomer, at least at a science-

based company like Amgen. A sales rep isn’t selling anything; it’s not
as though he walks out of a sales call with a purchase order. There is
no purchase order. What we call sales mainly entails imparting infor¬
mation. And by that I don’t mean strictly promotional literature; the
sales reps give doctors information from new studies in leading med¬
ical journals that might help them better determine whether a patient
could benefit from a particular drug. They also provide information
about new techniques for administering medication and tips for
minimizing side effects.
It is not an easy job. Much of a sales rep’s time is spent behind the
wheel and sitting in doctors’ offices waiting for an audience, all for
four or five calls a day lasting about ten minutes each. From the drug
company’s point of view, it couldn’t be less efficient, but it has to
be efficient for the doctor, who may be slotting a sales rep between
appointments.
Physicians often depend on the sales rep to summarize journal
articles that they don’t have time to read in depth. One aspect of
conventional sales applies here: the object is to solve problems, the
most effective sales approach of all. You get the customer to describe
the problem, and then you explain how your product or service can
help solve the [Link] end.
A stereotypical “salesperson” straight out of central casting, with
the quick smile and I-can-sell-snow-to-Eskimos attitude, doesn’t last
in this field. Many of the people who gravitate toward pharmaceutical
sales took premed, biology, or chemistry in college and see themselves
as part of the health-care system and contributing to improving pa¬
tients’ lives. Maybe not all reps feel this way, but most of them take
pride in their jobs.
It’s always more fun to sell a brand-new product; doctors are hap¬
pier to see you and full of questions, and you often get to go to the
front of the line, ahead of reps from other [Link], hiring
150 Science Lessons

Amgen’s sales staff, beginning in 1988, turned out to be easy. The

word was out that Epogen was going to be a blockbuster medica¬
tion; it didn’t just pave over a symptbm or two but effectively treated
the problem of anemia.
Our partner Johnson & Johnson probably didn’t believe that
Amgen would go through with forming its own sales division. In fact,
that had been a bone of contention going back to the original licens¬
ing agreement in 1985. We had to insist that our company would
handle its own sales and marketing, or we’d find a different partner.
Although Johnson & Johnson went ahead with the deal, I suspect
that executives there thought that when the time arrived, we’d come
to our senses and hand over that responsibility, something Johnson &
Johnson would have been more than happy to oversee. What it and
most other members of Big Pharma may not have realized was that
Amgen didn’t need an army of salespeople as they did. Taking our
cue from Genentech, the only other biotech firm that had assembled
its own sales and marketing department, for dialysis patients we
could cover Epogen’s entire national territory with about forty-five
people.
How? Well, we weren’t pitching to every general practitioner in
the United States, only to specialists in diseases of the kidneys
(nephrologists) and the blood and bone marrow (hematologists).
Our representatives didn’t necessarily have to visit them at their of¬
fices; they called on the dialysis centers. A second burst of hirings
took place before Neupogen’s introduction, because our salespeople
would be dividing their time between nephrologists and oncologists.
But the forty-odd figure was adequate for the launch of Epogen.
Everyone in the industry agreed that we’d scored a coup by lur¬
ing Paul Dawson away from pharmaceutical company G. D. Searle to
head up sales and marketing. Paul’s goal was to have the staff in place
three to six months before FDA approval. Because of the delay cre¬
ated by the ongoing legal skirmish with Johnson & Johnson, it
turned out to be closer to a year. So our salespeople went out into
 Red Hot Summer 151

the field extraordinarily well trained to anticipate any question a

physician might have. We even sent them for multiday training at
Ohio’s prestigious Cleveland Clinic, where they donned white coats
and accompanied the doctors on their rounds.
Similarly, before the debut of Neupogen, we arranged for a simi¬
lar learning opportunity at M. D. Anderson Cancer Center in Hous¬
ton. No one balked at the extra education; if there’s one thing that a
sales representative hates, it’s having to say “I don’t know” to a doc¬
tor’s question. They really liked being the people with the answers.
At dialysis centers, we expanded our sales calls to include the
nephrology nurses, because often physicians aren’t there at the facil¬
ity; the nurses are tending the patients. They were full of questions
about Epogen. Ultimately, we hired nurse trainers, whose sole job
was to work with the nurses. It turned out to be an excellent policy,
one that Amgen continues to this day. If we could get in to see the
hospital pharmacist, we’d educate her, too, because she was also an
important decision maker.
One of our best innovations was to establish an 800 number pro¬
gram for answering questions about Amgen products from doctors,
nurses, and pharmacists twenty-four hours a day, in addition to an
800 number strictly for patients. Questions might be something
along the lines of, “I gave Epogen to a patient of mine, and he exhib¬
ited [a certain symptom]. Could this possibly be a side effect of the
drug? Was this reported in any of the clinical trials?” If it took waking
up an Amgen doctor in the middle of the night to get the answer,
that’s what happened. Most pharmaceutical companies maintain
toll-free numbers, but the folks who call them are often disappointed
by the responses they receive, especially if they phone at night or
over the weekend.
I went to the call center once to watch it operate, and it was really
something. The person taking the call, usually a pharmacist or nurse,
would prepare an e-mail to the sales rep from that territory, informing
him of the doctor’s questions, explaining how they were answered,
152 Science Lessons

and so on. Consequently, when the salesperson went to see that doc¬
tor the next day, it was as if he’d been in on the conversation.
Including nurses and pharmacists was unorthodox at the time, and
very different from what many traditional pharmaceutical companies
did. But our feeling was that we owed it to the patients to establish a
relationship with all of their medical professionals, because they were
part of the supply chain—the people who delivered Amgen products
to our customers.
CHAPTER
8
Nuclear Winter

with epogen and neupogen solidly established, Amgen returned

much of its attention and resources to developing new products. In
1993 we funneled $255 million into R&D, a fourfold increase over
the budget at the time of Epogens introduction. Another $324 mil¬
lion and $452 million followed in 1994 and 1995, respectively.
The world, it seemed, was waiting for Amgen to complete a hat
trick of three blockbusters within only four years, the blink of an eye
in the time line of biotechnology. When 1993,1994,1995, and then
1996 went by without a new drug approval, people began to won¬
der. Had Amgen merely been lucky twice? Had the company lost its
golden touch? Why was the pipeline empty? The whispers were par¬
ticularly loud among naysayers on Wall Street—this, even though our
profits nearly doubled, from $383.3 million in 1993 to $679.8 mil¬
lion in 1996, the first year total revenues exceeded $2 billion.
Such expectations ignored fundamental realities of biotechnol¬
ogy, such as the fact that it typically takes nine to twelve years to de¬
velop a drug. Nor was it a secret that Amgen had temporarily
154 Science Lessons

diverted a good deal of research funding to our two big products. I

didn’t expect this decision to make our scientists happy, and it hadn’t.
The day I took over as CEO, in 19B8, head of research Dan Vapnek
came into my office and said intently, “Gordon, we’re going to have
to start spending some money again on new-product research.” He
was right, and his sense of urgency was understandable. But for the
next few years, we simply were not able to devote much capital to
that part of the company. It was time to work as hard as we could on
Epogen and Neupogen. If we were successful, we reasoned, before
long Amgen would have plenty of money to invest in research. That
is exactly what happened.
Some observers failed to appreciate the inherent uncertainties of
science. Amgen’s mid-1990s pipeline, far from having dried up,
brimmed with a number of promising early-stage product candidates.
Following are profiles of the winners and some of the also-rans.

BIG CATCHES

That many of our potential products would not survive clinical trials
was to be expected. A bigger story was that six of our potential med¬
ications did make it onto pharmacy shelves. Like a pack of presiden¬
tial hopefuls at the start of primary season, only the strong survived.

Interferon Alfaeon-1 (Infergen)

Infergen was one of Amgen’s original five therapeutics to get as far as

patient studies, and the only one (besides Epogen) to go all the way. Its
FDA license came in 1997, breaking our six-year [Link]-II gamma-
interferon also belonged to that fivesome, but it did not impress as a
remedy for viral infections and was the first to be abandoned.
Infergen, a genetically engineered synthetic version of type I in¬
terferon, is used to treat chronic viral hepatitis C in adults who are
also suffering from other diseases of the liver. When the body’s im-
 Nuclear Winter 155

mune system detects the presence of an invading virus, it releases

natural interferons into the circulation to fight the HCV infection.
Infergen sold disappointingly in the United States—other interfer¬
ons got there first—and its U.S. rights were peddled to another com¬
pany (see “20/20 Hindsight”).
But Infergen did find a welcome home at Amgen Japan. Our
Japanese subsidiary had been frozen out of both Epogen and Neu-
pogen by our preexisting licensing deal with Kirin Brewery. Mean¬
while, several of the products under development for the U.S. market
weren’t suitable for Japan. Year after year, Amgen Japan pressed ahead
with development but had nothing to sell; in my view, it was threat¬
ening to become a serious morale issue.
There are several subtypes of hepatitis C. It so happened that In¬
fergen was most effective against the type most prevalent in Japan.
Yamanouchi Pharmaceutical Company, today the country’s second-
largest drugmaker (under the name Astrellas Pharma, following a

20/20 Hindsight
Looking back, I can see that Infergen was a long shot that didn’t pay
off, in part because we had to quit working on it for several years
while we were building up Epogen and Neupogen. By the time Amgen
got into the market, Schering-Plough was already there with its inter¬
feron alfa-2b (brand name Intron A). It was Schering’s top product, and
it put its best people on it, whereas interferon would never be num¬
ber 1 for us. We just couldn’t compete with Schering’s A team.
Yet, given the identical circumstances again, I would make the
same decision to go ahead with developing Infergen. In the end, it
could have turned out to be highly superior or far worse, and it wasn’t
overly expensive for us to find out which.
156 Science Lessons

2005 merger with Fujisawa Pharmaceutical Company), was willing

to partner with Amgen Japan. Several competitors had staked their
claims to the market long before ottr arrival, but the drug has built
up a 16 percent share and continues to grow.

Ancestim (Stemgen)

One aggressive approach to combating certain cancers is to adminis¬

ter extremely high doses of chemotherapy followed by peripheral-
blood stem-cell rescue. It’s hoped that the toxic dose will eradicate the
tumor cells. But high doses inevitably kill the bone marrow. Stem¬
cell rescue, an alternative to bone marrow transplantation (BMT),
entails using a special apheresis machine to “skim” stem cells (or pro¬
genitor cells) from the circulation and then return them to the pa¬
tient’s bloodstream to replenish the impaired marrow.
The trouble is that our bodies make few peripheral-blood stem
cells (PBSCs); if a patient’s supply is further depleted by harsh chemo
drugs, collecting stem cells can be like panning for gold. Amgen’s ge¬
netically engineered human stem-cell growth factor (Stemgen) stimu¬
lates stem-cell production. In Stemgen’s phase III investigational study,
the subcutaneous drug was administered in conjunction with Neu-
pogen to roughly one hundred women with breast cancer. Another
one hundred or so breast cancer patients received Neupogen alone.
The women on the combination protocol required less apheresis
to accumulate the necessary volume of transfusable stem cells (four
treatments compared with six). In addition, significantly more of them
(63 percent versus 47 percent) reached the targeted goal, and their
bodies produced two to three times as many stem cells as did the bod¬
ies of participants who’d been given Neupogen only. Therefore, more
of them were able to undergo high-dose chemotherapy with PBSC,
which in the United States is now performed as frequently as bone
marrow transplantation. In addition to being less expensive, the newer
procedure doesn’t require general anesthesia. Furthermore, patients’
 Nuclear Winter 157

levels of red, white, and platelet cells seem to return to normal faster
than they do following BMT.
Unfortunately, in the eyes of many physicians and insurance com¬
panies, these advantages didn’t justify the cost of Stemgen, which also
produced some adverse side effects. Because the company had more-
promising product candidates competing for limited R&D resources,
we dropped the drug as a human therapeutic. However, it has found a
use for growing stem cells outside the patient (in vitro) and is em¬
ployed in many stem-cell research projects.

Cinacalcet (Sensipar)

Once Amgen grew large enough, we could afford to license other

companies’ discoveries instead of the other way around. One of
those was cinacalcet, our first small-molecule drug. The significance
of that probably eludes the average person suffering from secondary
hyperparathyroidism; all these patients need to know is that Sensipar
comes in tablet form to be taken orally.
By the mid-1990s, the biotech sector began to understand that
most people’s bodies contain ample amoun ts of most proteins. Clin¬
ically significant deficiencies, such as neutropenia and anemia, are
uncommon. Given the limited number of therapeutic proteins, we
realized that we needed to expand into small-molecule drugs. A
number of our contemporaries had reached the same conclusion: in
1996, about half the new medications manufactured by biotech
firms were derived from small molecules; today that proportion is
about 70 percent.
The question was where to begin. The Big Pharma companies
had been working with small molecules for decades; Amgen had a lot
of catching up to do. Assessing the effectiveness of small-molecule
agents tends to be trickier because they mostly inhibit, or block,
processes of the body, whereas proteins promote, or stimulate, biolog¬
ical activities. I felt confident that our scientists’ expertise in molecular
158 Science Lessons

biology would get Amgen off to a fast start, but our development
staff’s lack of experience in small-molecule drugs would probably
cause mistakes and delays. Thereforb, we needed to adopt a small-
molecule product candidate that was already undergoing human test¬
ing; it almost didn’t matter which one. “I don’t care if it fails,” I said.
“Let’s just give our development people something to work with and
learn from.”
Amgen’s scientists went off in search of a small-molecule thera¬
peutic, not really knowing what to look for. A New Jersey company
called NPS Pharmaceuticals licensed to us a compound designed to
treat the hormonal disorder secondary hyperparathyroidism, and we
set to work on it.
Secondary hyperparathyroidism usually occurs as a frequent
complication of chronic renal failure. The parathyroids, four pea¬
sized glands located in the neck, secrete parathyroid hormone
(PTH). Hormones act as chemical messengers. PTH’s job is to regu¬
late the amounts of calcium and phosphorus in bones. The kidneys,
too, balance calcium levels. When someone’s kidneys fail, the body
no longer absorbs the mineral efficiently from food. Consequently,
the concentration of calcium in the bloodstream falls.
Calcium sensors on the parathyroid glands read the dropping
level, become alarmed, and secrete excess [Link] restock the circu¬
lation, calcium is leached from the bones, in what becomes an end¬
less feedback loop. Eventually, the bones may lose too much calcium
and weaken.
The drug we were working on, cinacalcet, acted directly on the
parathyroids’ calcium sensors to break the cycle of hyperparathy¬
roidism. First, though, we had to correct a problem with the mole¬
cule. NPS hadn’t performed metabolism studies to see how the body
rid itself of the drug. If it had, it would have realized that about one
in one hundred people lacks the gene to produce the enzyme pri¬
marily responsible for metabolizing cinacalcet. For those few, the
medication would have cleared too slowly from their systems, causing
 Nuclear Winter 159

a serious overdose. Our scientists ingeniously redesigned the molecule

so that it was metabolized by a different enzyme and yet retained its
therapeutic effect.
Sensipar, approved in 2004, has sold exceptionally well. Because we
were already familiar with the dialysis community and its physicians,
the drug turned out to be an excellent choice for Amgen’s entree into
small-molecule pharmaceuticals—and an unexpected success.

Anakinra (Kineret)

Licensing an outsider’s discovery is one strategy for building a prod¬

uct portfolio; buying the company lock, stock, and high-pressure liq¬
uid chromatograph is another. In 1994, with a whopping $700
million in cash on hand, Amgen purchased twelve-year-old Syner-
gen Inc. for $254 million. It was our first acquisition.
Only a year earlier, the Boulder, Colorado, biotech appeared to
be en route to success with a recombinant drug for treating systemic
bacterial infections {sepsis), which take one hundred thousand lives
each [Link] compound, Antril, performed well in early clinical tri¬
als, reducing death rates by as much as 28 percent over placebo. Two
other companies—Centocor and Xoma—had recently seen their
promising sepsis drugs go up in smoke. However, Synergen’s em¬
ployed a different mechanism of action, interfering with the body’s
immune response, whereas Centocor’s and Xoma’s agents more or
less tried to apprehend toxins in the circulation like a pair of arrest¬
ing police officers.
In a replay of the unpredictability of drug development, Antril
faltered in phase III, testing barely better than placebo. It wasn’t Syn¬
ergen’s first flameout; another medication, for treating the debilitating
foot ulcers that diabetics frequently develop, advanced to phase III,
only to flop. News of Antril’s failure sent Synergen stock plummeting
from $40 to $13 and change per share. By the time we stepped in to
buy the company, it had laid off more than half its employees.
160 Science Lessons

Probably makes you wonder why Amgen would even consider

acquiring the company, doesn’t it? Actually, we had been interested
in some of its other projects, such Vs a compound for treating the
painful joint inflammation associated with rheumatoid arthritis (RA).
The drug, anakinra (Kineret), became Amgen’s fourth approved human
therapeutic in 2001.
At the time we bought Synergen, though, anakinra was perform¬
ing so disappointingly in early-stage clinical trials that we stopped
accepting additional patient volunteers. From a purely scientific po¬
sition, we should have halted the study then and there, but for con¬
tractual reasons, and from an obligation to the patients already
enrolled, we let it continue. It’s a good thing we did, because the
long-term data was encouraging enough to warrant further testing.
Rheumatoid arthritis, which affects more than 2.1 million men
and women in the United States, descends from a dysfunction of im¬
munity. As with other autoimmune disorders, for some reason the
body’s defenses mistakenly attack the person’s own tissue—in this
case, the joints. Anakinra controls inflammation and slows destruc¬
tion by blocking the protein interleukin-1, one of the immune sys¬
tem’s weapons against infection.
Less than a year after its approval, Kineret was eclipsed by another
RA drug. Etanercept (Enbrel) was the prize of Amgen’s acquisition
of Immunex, the Seattle-based biotech that had introduced Enbrel
in 1998. At the time, it was the first genetically engineered RA med¬
ication. Though somewhat similar to anakinra, etanercept belongs to
a different family of biologic anti-inflammatories called tumor
necrosis factor (TNF) inhibitors. Adding etanercept to methotrexate,
the drug of choice in RA management, produces superior results
(compared with the use of either drug by itself) in relieving pain,
tenderness, and swelling; slowing joint erosion; and enabling patients
to go about their daily activities in greater comfort. Enbrel’s acquisi¬
tion nudged Kineret to the sidelines, but that’s the kind of “problem”
most pharmaceutical companies would love to have.
 Nuclear Winter 161

THE ONES THAT GOT AWAY

As for the prospects that slipped off the hook before we could reel
them in, they had looked just as promising as the big catches. But
that’s science and medicine. You ride the experiment as far as it will
take you, and most of the time it’s not far enough.
One of the disappointments, megakaryocyte growth and devel¬
opment factor (MGDF), would have been the third gem in Amgen’s
crown of hematopoietic growth factors. Epogen jump-started the
marrow’s production of red cells, Neupogen did the same for white
cells, and MGDF was expected to stimulate production of the
blood’s other primary component: platelets, the disc-shaped cells that
converge on wound sites and stanch bleeding. Cancer patients on
heavy doses of chemotherapy may need platelet transfusions because
their suppressed bone marrow makes fewer platelets. Although trans¬
fusions are generally safe, there’s always a slight risk of acquiring an
infectious disease.
Clinical trials got under way in 1995, but our onetime nemesis
Genetics Institute beat us to the market with interleukin-11
(Neumega) two years later. However, that’s not what caused us to
discontinue MGDF. In 1998 a handful of the volunteers in the phase
III studies developed neutralizing antibodies; their immune systems
perceived the drug as an intruder and deactivated it. Worse still, the
antibodies were destroying whatever platelet growth factor their
bodies made. We discontinued development.
Another experimental agent identified with four letters, this one
called BDNF (brain-derived neurotrophic factor), also progressed all
the way to phase III studies before fizzling. Developed in collabora¬
tion with Regeneron Pharmaceuticals, BDNF was intended to slow
the devastating degenerative disease amyotrophic lateral sclerosis, also
known as Lou Gehrig’s disease. Although BDNF had performed ex¬
tremely well in the lab and in preclinical trials, it struck out when
tested head-to-head against a placebo in humans.
162 Science Lessons

A small protein called epidermal growth factor (EGF) was ini¬

tially considered a likely godsend for people with stubborn bedsores
and other skin wounds, and not only by Amgen. In the mid-1990s,
more than half the R&D partnerships in the biotech industry were
sold by Paine Webber. One day the firm came to us with an offer: tis¬
sue growth factors represented an untapped product opportunity.
Paine Webber wanted to enter this area by creating a limited partner¬
ship for R&D with a star biotech company—Amgen was its first
choice—to develop tissue-healing drugs. Its idea was that the part¬
nership would raise so much money that no other company would
want to compete in this field.
We weren’t enthralled by the idea of pursuing EGF, but it wasn’t
unrealistic, either. There are substances that the body uses to speed
tissue growth; we were working on one ourselves. Ultimately we
agreed to the deal, a bit reluctantly. While finalizing the terms, Paine
Webber came back and said that we’d have to chip in some royalties
from Neupogen. “After all,” the rep said, “that’s a sure thing; the rest
of this is speculative.” So we agreed, more reluctantly this time, and
the partnership was on.
As promised, money for research was lavished on Amgen. But the
much-anticipated breakthroughs never came. In animals the com¬
pounds made wounds heal faster, but in human trials, the benefit
wasn’t substantial enough to make the drug a commercial success. To
this day, no one has been able to make a wound-healing growth fac¬
tor work, for various reasons.
One of the studies inadvertently pointed up the impact of sub¬
standard medical care on people’s [Link] conceived a clinical trial
for hospital patients suffering from intractable wounds that had resis¬
ted every available treatment. The reason for the hospital setting was
that the patients would receive constant attention from the nursing
staff, so we could be assured that the ulcers, abscesses, and so on were
being tended to properly. Remember that in an investigational study,
you try to weed out all variables that could skew the findings.
 Nuclear Winter 163

All the patients in the trial were afforded first-rate wound care;
additionally, half were put on our growth factor. The result? Many of
the patients in both groups healed. This was highly embarrassing to
the doctors and nurses. The trial proved that many of these patients
had not been adequately cared for in the past and that the excellent
nursing care required by the clinical-trial protocol was far more ef¬
fective than the product under investigation.
Perhaps the biggest disappointment of all was the weight-loss bi¬
ologic leptin. In 1994 researchers at Rockefeller University, in New
York City, discovered and isolated the gene for a hormone that plays
a major role in regulating body weight. It does this by acting on the
part of the brain known as the hypothalamus to govern food intake
and calorie expenditure. Rockefeller University named it leptin, a
derivation of the Greek word leptos, meaning “thin.” Mice and rats
injected with the genetically engineered protein ate two-fifths less
food during the experiment; they also lost 30 percent of their body
weight in a matter of weeks. Given that some seven in ten Americans
can be classified as overweight, this was exciting news. When Amgen
licensed the research for $20 million—with the university to receive
an additional $70 million if we should successfully turn the protein
into a salable drug—the New York Times gushed, “It could turn out
that Amgen has bought a license to print money.” Not quite. We had
bought a license to spend money.
In the large patient trial that we sponsored, 30 percent to 45 per¬
cent of the people who took leptin for thirty days lost a modest four
and a half pounds or more. But 19 percent of the control group,
given an inactive substance, achieved the same results through diet
and weight counseling. This wasn’t the first time that a potential
medication worked well in mice but not in people.
Leptin is still being studied, but not as a cure for obesity. Based on
recent patient trials, it may be applicable to two conditions that involve
the hypothalamus: hypothalamic amenorrhea (the unexplained absence
of menstrual periods in menstruating women) and lipodystrophy.
164 Science Lessons

People with the latter disease lose their body fat and subsequently
develop metabolic disorders such as insulin resistance and hyper¬
glycemia. The hormone may also prove helpful for men and women
with type 2 diabetes.

TAKING STOCK

A total of six drugs from this period would eventually win FDA ap¬
proval. But as of 1995, the year that Amgen celebrated its fifteenth
birthday with a party for its four thousand employees on the lawn
behind building 34, we were still looking for a blockbuster product
to join Epogen and Neupogen. It was a serious cause for concern,
not only on Wall Street but also within the company.
At one point, the board of directors appointed what it called a
strategy committee. Before each board meeting, the members met
with the researchers for a few hours to try to figure out whether we
knew what we were doing. They didn’t say we did, but neither did
they say we didn’t.
I never lost confidence that Amgen was on the right path. We
were simply feeling the pinch of the gap in the pipeline caused by
the temporary lack of resources when everything was devoted to
Epogen and Neupogen. The organization wasn’t doing anything
“wrong.” We had our fair share of company, too. Starting in about
1992, the biotechnology sector slid into several years of doldrums.
“Nuclear winter,” business scribes came to call it. A number of com¬
panies had failed to come through with much-lauded technologies,
and before long disenchanted investors were being seduced by
promises of the untold riches to be mined in cyberspace.
It was one of those times when a CEO truly earns his pay. You
become a corporate cheerleader, constantly offering encouragement
within the company and correcting fallacies on the outside. I often
reminded everyone that each year management carefully reviewed
 Nuclear Winter 165

the portfolio of products under development, and we were con¬

vinced that our research was producing viable drug candidates. At
the same time, we were actively seeking licensing opportunities. In
short, Amgen was doing everything it should be doing. We just
needed to stay the course and keep on believing that everything was
going to work out.
The annual portfolio review was a painstaking process. For three
days, top management pored over thick books of information about
Amgen drugs that were either in human trials or about to be: the
market, competition, patents, potential manufacturing issues and
costs, the effect on the sales force—pretty much [Link] rated
the projects A, B, or C, with those rated A (high priority) receiving
the most funding and resources. Some projects moved up in ranking
from one year to the next, while others were demoted or halted alto¬
gether. In addition to these yearly summits, management met less
formally once a month to keep abreast of recent developments.
It’s not as though there was internal dissension about the direc¬
tion of R&D. However, I would occasionally be asked, “If plan A
fails, what’s plan B?”
I thought long and hard before answering, “There is no plan B.”
Plan A was the right approach, and we were going to keep work¬
ing it until it succeeded. If it didn’t, only then would we change di¬
rection. Some folks never believed me, thinking that I secretly had a
plan B up my sleeve all along but wouldn’t say so.
My only fear was that the stock price would drop so low that the
company might be in danger of a takeover, the fate of numerous
biotech firms in the 1990s. At times, names were bandied about in
the financial news. Bristol-Myers Squibb was one such party, I re¬
member. Another was Swiss pharmaceutical giant Roche Holding
Ltd., which had merged with Genentech in 1990. What if Amgen
were acquired and the mother company destroyed our corporate
culture? We all worried about that.
166 Science Lessons

20/20 Hindsight
Here are some thoughts from Kevin Sharer about Amgen’s (and my)
performance during this period:

Gordon took some unfair criticism about the productivity of

Amgen’s pipeline in the mid- to late nineties. We had three big
drugs—GDNF, MGDF, and leptin—and if any one of them had
hit, he would have looked like a genius. It wasn't that we
weren’t swinging for the fences; it was science. Sometimes it
just doesn’t work.
Acquiring Synergen was a bold move, I thought. The other
thing Gordon did was to scale up the company so that it was
ready for its next phase when I got there, and he delivered spec¬
tacular results in terms of financials and share price. He also
groomed a successor that is stilt in the job eight years later.
In my book, that’s a pretty clean sweep for a CEO.

Investment bankers often pitch takeovers. Every couple of years,

Morgan Stanley, Goldman Sachs, and others probably showcased
Amgen to all the pharmaceutical behemoths. But when they seri¬
ously analyzed the possibility, the conclusion was always that the
stock was too expensive. We figured that as long as our price-to-
earnings ratio remained high enough, no one could acquire us, be¬
cause it would be too dilutive to their own earnings per share.
Simply put, the best defense against a takeover is to be success¬
ful—very successful—and to make sure that your stock price keeps
pace. Not once did anyone formally approach Amgen about an ac¬
quisition or merger. It became a familiar refrain for investment
bankers to say to us, “If you ever want to sell the company, let us
 Nuclear Winter 167

know!” To which we’d reply, “Don’t worry, we will!” Then we’d

laugh. Our goal was to create the world’s best biotechnology com¬
pany. If Amgen had been acquired, the senior managers would have
felt like failures.

“THE MEMO”

Not deviating from our game plan was one struggle. The other was
maintaining the fragile dynamic between the science and sales sides
of the company. Bear in mind that Amgen consisted almost exclu¬
sively of researchers for its first seven years. When Paul Dawson
began assembling his sales team in 1988, it was as though an alien
species had descended on the campus. Sales and marketing people
are naturally extroverted; that was one difference. More significantly,
our new hires came from Big Pharma, where, typically, science takes
a backseat to sales and marketing.
“Culture shock” is how researcher Burt Ensley describes the re¬
action among the scientists at this intrusion on their domain. “There
was a certain amount of animosity,” he says, conceding that not all of
it was fair. “But the people in marketing tended to act as though they
were your boss. So the disdain was [Link]’s what made it really
interesting: they thought you were a bunch of egghead jerks, and you
thought they were a bunch of lowlifes.”
Dennis Fenton agrees. “There was a complete lack of under¬
standing between the two groups, which I shared, too, until I became
part of sales and marketing.” One of my reasons for shifting Dennis
from vice president of operations to manager of sales and marketing
was to help bridge that gap. “We scientists didn’t really appreciate
how difficult it was to get the products to the patients,” he reflects.
“Maybe it doesn’t entail the same degree of difficulty as creating
Epogen, but it doesn’t just happen on its own. And without smart
people working together to get it done, Epogen wouldn’t have been
the success that it became.”
168 Science Lessons

I’d seen this indigenous tension between the two tribes when I’d
interned at Procter & Gamble while in college. Most of the time,
P&G was a well-managed company. But occasionally the factory
manager resembled the unfortunate kid at the end of the hne in the
children’s game (I’m probably dating myself here) crack the whip—
getting jerked around by the marketing department. Someone in
marketing would decide on the spur of the moment to have a two-
for-one sale on Crest toothpaste, and suddenly the factory would be
bombarded with orders it couldn’t meet. Manufacturing didn’t
know whether marketing could have hatched this plan earlier or
whether a competitor’s unexpected moves forced the last-minute
changes. All the factory manager knew was that no one had given
him sufficient notice so that an inventory could be built up. Not sur¬
prisingly, some animosity simmered between the two divisions.
At Amgen, an effort was always made to encourage our different
departments to learn about one another and to try to understand how
their actions would affect the rest of the organization. Some compa¬
nies, though, give sales and marketing carte blanche, with R&D and
manufacturing ordered to cater to marketing’s every whim. They
waste a lot of money that way, not to mention stir up resentment.
In part through Fenton’s efforts, mutual respect was built over time.
“We now have a real commercialization process,” he says, “where re¬
search, development, sales, and marketing are all part of a continuum
to the patient. It’s like a relay race. If two of the parties don’t cooper¬
ate and the baton gets dropped, it’s both their fault.”
Things weren’t quite that civilized in 1995. As sales and market¬
ing staffers flooded into Amgen, they started vying for the company’s
soul. It grew serious enough that shortly after our fifteenth birthday
celebration, I drafted a strong statement for everyone. It has since en¬
tered Amgen lore as “The Memo.” Essentially, it said that unlike in
Big Pharma, at Amgen scientists were and would always be the kings
of the hill. Furthermore, although we’d briefly considered diversify¬
ing into over-the-counter medicines—a strategy that would have
 Nuclear Winter 169

transformed Amgen into more of a traditional pharmaceutical com¬

pany, in which marketing often directs research—the memo offered
this reassurance: “Amgen has reached a broad consensus to continue
to follow our science-driven breakthrough-product strategy.” Every¬
one should get over it and get back to work.
Soon thereafter, we formulated the Amgen values. In case anyone
doubted our commitment, topping the list: “Be science based.”

VARIATIONS ON A THEME:
ARANESP AND NEULASTA

Amgen didn’t have to look far for its next two big [Link] were
right under our noses.
In science, there’s always more to learn, especially as technology
advances. By the standards of the mid-1990s, the technology that ex¬
isted when Fu-Kuen Lin discovered the gene for human erythropoi¬
etin was fairly primitive. A team directed by researcher Steve Elliott
was delving deeper into Epogen to better understand how it worked.
Meanwhile, another team, headed by JeffBrowneJoan Egrie, and
Tom Strickland, was studying the drug’s carbohydrate makeup. Let
me explain.
Unbeknown to either Amgen or the FDA, the Epogen that we
sold was actually a mixture of five EPOs. Each had the same amino-
acid backbone but with different amounts of carbohydrate attached
to it. Initially our EPO contained more than five; the outliers were
cut off during the purification process. When we first discovered that
we were selling a combination of EPOs, we didn’t understand the
significance of the differing amounts of carbohydrate.
Later it was discovered that if you separated the various versions,
each one exerted the same stimulatory effect on bone marrow, but to
varying degrees: the protein with the least carbohydrate was least ef¬
fective, and the one with the most carbohydrate was most effective.
The team wanted to explore what would happen if the epoetin alfa
170 Science Lessons

molecule was reengineered to contain more carbohydrates than would

be produced by the human EPO gene. In other words, they wanted to
improve upon nature. That was something we’d shied away from gen¬
erally, because when you do that you run the risk that patients’ im¬
mune systems will form antibodies against the substance. All sorts of
things can go wrong, so it wasn’t considered a sound investment.
Carbohydrates connect like branches of a tree at specific sites on
the protein. To alter the amount of carbohydrate, therefore, you must
add one or more sites, a process that entails altering the sequence of
amino acids. Epogen, in its present form, was extraordinarily good;
more than 99 percent of patients responded to it, with virtually no
adverse side effects. The only benefit of this reconfigured epoetin alfa
was that because of its increased molecular weight, it would stay in
the body longer and thus a person could take it less often.
On the surface, it didn’t seem worth pursuing. Men and women
on a self-administered blood-cleansing therapy called peritoneal
dialysis (PD) would find a longer-lasting red-cell stimulant far more
convenient because they had to inject themselves with EPO every
day; however, PD patients made up only one-fifth of all people with
end-stage renal disease. Amgen’s only other market for Epogen was
hemodialysis patients in the United States, and most of them had the
medication administered directly into the rubber tubing of the dia-
lyzer during their thrice-weekly treatments, an arrangement that was
fully satisfactory.
In Europe, Johnson & Johnson’s territory, doctors frequently in¬
jected EPO subcutaneously during hemodialysis, believing that this
procedure was more effective and [Link] drug would also be
ideal for cancer patients who developed anemia due to chemotherapy.
But that, too, was Johnson & Johnson’s area, for which we received
10 percent in royalties..
Dan Vapnek, representing the majority of Amgen researchers,
summarily dismissed the proposal of developing a more active,
longer-lasting Epogen. I respected his views and understood why he
 Nuclear Winter 171

might feel that way. But despite listening carefully, I didn’t hear com¬
pelling reasons for Amgen not to at least explore the possibility. What
I heard from several of our executives was mostly personal bias of¬
fered as science-based logic.
One unspoken explanation for the skepticism was that many re¬
searchers regarded prolonging a drug in the bloodstream as “low-
grade” science. Making a wholly new discovery was “genuine” science
in a real-men-don’t-eat-quiche sort of way. But for patients who had
to inject Epogen several times a week, being able to go a whole week
without a single needle stick would improve the quality of their lives
in a very real way. Some folks can’t bear to inject themselves, and
doctors’ offices are closed on weekends. A long-lasting EPO would
enable them to go on vacation without having to take along the
medication, which had to be refrigerated.
Internal politics, something Amgen usually managed to sidestep,
also was a factor. Members of the research division were privately
miffed that the scientists responsible for these intriguing findings be¬
longed to development. What’s more, they’d “defected” from R to D.
All in all, many smart people were against long-acting Epogen for
reasons that weren’t very good. High science, low science—what’s
the difference if the drug is effective, helps patients, and makes
money? I certainly wasn’t concerned with who came up with the
idea or which department got credit. There were only two valid rea¬
sons for not adding darbepoetin alfa to our portfolio:

• We didn’t yet know how long people on dialysis would be

able to go between injections; perhaps the practical benefit for
most of them would turn out to be minimal.

• Antibody formation might prevent the protein from working

in some patients.

If either of these issues didn’t go Amgen’s way, we’d have wasted

millions of dollars.
172 Science Lessons

One potential obstacle was crossed off the list of concerns when
we considered that carbohydrate acts as a sort of shield against the
immune system; the more carbohydrate you add to a protein, the
lower the chances that antibodies would be mobilized against it. I
also spent a lot of time discussing the merits of our legal position
with the law department. Johnson & Johnson could be expected to
contend that its rights to the original Epogen extended to this drug
as [Link] were confident that a neutral third party would agree that
the licensing agreement between our two companies did not entitle
Johnson & Johnson to any stake in the potential new product. To
proceed with darbepoetin would cost us perhaps $5 million to $10
million in legal fees, along with many times that amount for R&D.
If we won on the legal issue, the rewards would far outweigh the
expense. Even if we lost, the benefit to our dialysis patients and to
Amgen would probably be great enough to justify the expenditure.
However, that’s something we wouldn’t know until we’d spent the
money and completed clinical testing.
After giving the matter much thought, I felt strongly that this was
a wager the company should make. Another factor in my decision
was that someday the patent on EPO was going to expire. Any supe¬
riority at all, no matter how small, can loom large in a head-to-head
marketing battle. It was a long way off, but a CEO is paid to think
further into the future than other members of the company. The
higher you go up the organizational pyramid, the longer your time
horizon should be.
All these considerations made it a fairly easy decision. Kevin
Sharer, who’d joined us as president in 1992, was also 100 percent
onboard. A lower-level executive probably wouldn’t bet a long shot
like that, but it’s a CEO’s privilege—unless the board of directors says
otherwise. It was one of the rare times that I opted to spend some
executive capital and rule against the group consensus, but I knew I
was in a stronger position than Amgen’s scientific leadership to judge
the financial rewards and risks and assess how the arbitration was
 Nuclear Winter 173

likely to turn out. In announcing that we would go ahead with dar-

bepoetin, I made it clear that I would take full responsibility for the
decision.
When early attempts to design a new red-cell stimulator failed, it
added to critics’ views that this was a waste of funding. About one
hundred molecules with varying amounts of carbohydrate were
made, and in each case, the additional carbohydrate blocked its ac¬
tion. It took a few years, but in the end, the scientists found one that
worked. Far from being low-end science, this was pioneering work.
Adding two carbohydrate chains gave darbepoetin a substantially
longer half-life (the time it takes the body to eliminate half the given
dose of a drug), an approach that no company had tried before. It
could be manufactured fairly easily, too.
Clinical trials of the drug, called Aranesp, commenced in 1996.
Meanwhile, other Amgen scientists were perfecting a way to pro¬
long Neupogen’s effect. Unlike with Epogen, we’d always suspected
that filgrastim could be spun off into a beneficial second-generation
medication, but not by adding carbohydrates. Instead we used a
method called pegylation.
The peg in pegylation stands for polyethylene glycol (PEG).
Adding this chemical compound to a molecule extends the life of
the drug, because the patient’s kidneys must metabolize the PEG first
before they start working on the protein itself. The advantage of this
new version over the original Neupogen was stunning. A typical
cycle of chemotherapy lasts two weeks. For people considered to be
at risk for neutropenia and fever, that meant ten to fourteen days of
daily subcutaneous injections. To spare patients the extra injections
and reduce the cost, oncologists sometimes tried to get by with
fewer doses, but that exposed patients to potential infection. Pegfil-
grastim (Neulasta), we found, required only one injection per
chemotherapy cycle. Patient trials also revealed a self-regulation
feature that hadn’t been seen in animal studies or with Neupogen:
the Neulasta drug remains in the circulation during the time the
174 Science Lessons

person is neutropenic. But when the neutrophil level recovers, Neu-

lasta clears rapidly from the bloodstream, so that any side effects are
short-lived. V

ANOTHER FACE-OFF WITH

JOHNSON & JOHNSON

As in every legal dispute between companies, no one at either

Amgen or Johnson & Johnson, including the lawyers, could predict
how this latest chapter would turn [Link] issue came down to this:
was Aranesp an improvement on the product licensed to Johnson &
Johnson, or was it a new product? If it was found to be the former,
our partner would have the same rights to darbepoetin that it had to
EPO. If the ruling went the other way, Johnson & Johnson would
have no rights to the new drug. The contract wasn’t crystal clear—as
is frequently the case—but we believed that its language strongly fa¬
vored our position.
Johnson & Johnson certainly couldn’t claim that Amgen had de¬
liberately intended to deceive it. Even many of our brightest scien¬
tists hadn’t wanted to pursue a long-acting red-cell stimulating
[Link]’s how tmobvious a step it was. Our position was that we’d
invented darbepoetin. Why should Johnson & Johnson be allowed to
sell it, especially considering that Amgen had funded it 100 percent?
That decision had been intentional on my part, in anticipation of a
possible courtroom clash. I didn’t want to have the issue further
complicated by our spending even one penny of Johnson & John¬
son’s money on our new product.
The hearings took place in Chicago, the same site as our previous
encounter with Johnson & Johnson. Back then, I’d decided that we
should have a local law firm fight with us. Bob Weist, our general
counsel, had once practiced patent law in the Windy City. I asked
him to go back there, talk to people he knew, and get a recommen-
 Nuclear Winter 175

dation for the best litigation attorney in the city. I didn’t care what
firm the attorney was with.
Everyone Weist spoke to said the same thing: get William J. Harte.
It turned out that Harte was a private plaintiff’s-type attorney and a
litigation legend. Among other cases, he’d handled a famous class-ac¬
tion suit against General Motors in which GM was sued for sticking
Chevrolet engines into its Oldsmobiles without telling anyone.
When Bob went to see him, Harte’s first words were, “Mr. Weist, are
you sure you’re in the right place? I don’t represent people like you; I
sue people like [Link] must be some mistake.”
It turned out that Harte’s wife had cancer, and it irked him that
the world’s fifth-largest pharmaceutical manufacturer would try to
interfere with the availability of a beneficial drug. He agreed to join
the team. Harte was an interesting study in contradictions: a pillar of
the local Catholic church who possessed a vocabulary that could
make a navy seaman blush. It was a lot of fun to work with him, and
we became good friends.
The arbitration panel’s decision, handed down just before Christ¬
mas 1998, granted Amgen exclusive rights to darbepoetin, by then in
phase III clinical [Link] result blew open the U.S. market for us,
in that as soon as Aranesp gained FDA approval, we could sell to
Americans suffering from chronic anemia due to causes other than
kidney disease. We could also enter Europe for the first time. It was
the sweetest victory in our long war with Johnson & Johnson.
Here’s something I didn’t tell anyone until after the hearings
ended. Had we lost arbitration, Aranesp would have held little value
for Amgen, at least not compared with the potential income it repre¬
sented to Johnson & Johnson. Under those circumstances, I’d
planned to inform Johnson & Johnson that it wasn’t worthwhile for
Amgen to put the new drug on the market. We would tell Johnson &
Johnson what we knew and teach its people how to make small
quantities in the lab; we didn’t have a high-volume production
176 Science Lessons

process, and there was no incentive for us to develop one. We would

be saying, in effect, “Be our guest—go right ahead.”
As I saw it, a stunned Johnson &yJohnson would have replied that
it didn’t know how to produce the drug. Exactly. We would then
offer to do it, but at a high enough price that Amgen would turn a
profit. I never shared the details with anyone, for fear that somehow
Johnson & Johnson might find out. That was another crucial factor
in the decision to go forward with Aranesp: the knowledge that even
under the worst-case scenario, the decision posed little long-term fi¬
nancial risk.
A few of the people who had initially opposed Aranesp conceded
that they were glad I’d persisted with it. Others probably thought it
but didn’t say so. It was a tremendous source of pride for me, espe¬
cially after Aranesp reached the market in 2001, followed only
months later by Neulasta. In their first year on sale, the two medica¬
tions generated a combined $715 million in the United States alone.
CHAPTER

How Amgen Built

a Winning Team

BACK in THE introduction, I posed the rhetorical question,

How does an organization attract outstanding employees? Answer:
by devoting time, thought, and effort to becoming the kind of place
where talented and motivated men and women want to work. And
how do you build and maintain that kind of company? By hiring
quality people. It’s simultaneously the catch-22 and yin and yang of
the business world.

HIRING SMART

Like any successful organization, Amgen was fortunate to sustain an

exceptional workforce. With so many scientists onboard, the staff was
unusually well educated. But other biotechs and science-driven firms
had just as many PhDs and MDs walking the hallways as Amgen did.
178 Science Lessons

Few of them, though, matched our researchers (and our “civilian”

personnel) in their willingness to go the extra mile, as exemplified by
the self-proclaimed Simi Valley Hostages. Clearly there’s more to
building a winning team than stockpiling individuals who have glit¬
tering credentials. Here are some of the ways that we managed to
hire smart.

THE 360-DEGREE JOB INTERVIEW

You hear many business leaders bemoan the supposed dearth of capa¬
ble employees in the United States. It could be that many employers
are casting their lines in the wrong fishing holes or using the wrong
bait. The bigger problem, I’m willing to bet, is that if they actually
hooked a high-achieving job applicant, they’d unwittingly toss her
[Link]’s because many of the desirable qualities that a boss should
look for—such as resourcefulness, ethics, and adaptability—aren’t
found on a [Link] is why, even early on, Amgen typically con¬
ducted ten or twelve interviews by seven to ten people to get a sense
of how well a candidate would fit the organization and vice versa.
I hope you didn’t just spill your caffe latte. “Ten to twelve inter¬
views? For one person? We don’t have time for that!”
Granted, it is time-consuming to spend part of a day—in rare in¬
stances, two days—compiling detailed profiles of the finalists for a
position. But it is efficient, and not a waste of time, when you com¬
pare it to the weeks or months you would squander in training a new
hire (pulling other staffers away from their jobs), only to discover that
he isn’t going to work out. That’s in addition to the time, cost, and
headache of repeating the process and either retraining the person
for another assignment or, regrettably, letting him go.
We usually interviewed five or six people per job opening. Then
we narrowed the list to two and asked them back. Each candidate
might have an audience with some of the same interviewers as be¬
fore, while meeting other Amgen staffers for the first time.
 How Amgen Built a Winning Team 179

Corporations such as IBM, [Link], and Motorola have

adopted the practice of peer interviewing, wherein the folks who
would work alongside the candidate sit down with her and ask ques¬
tions of their own. It’s a great idea, except that it addresses only two
sides of the work triangle. What if you’re hiring a middle-level man¬
ager? She will not only be reporting to supervisors and interacting
with other middle managers but also overseeing junior workers. At
Amgen, even subordinates got face time with the candidates and
weighed in with their assessments afterward. Their feedback wasn’t
the primary factor in our offering the job to applicant A or applicant
B, but it was taken into serious consideration.
Letting folks interview their prospective boss pays a dividend: if
management winds up hiring the preferred applicant, workers are
obligated to try their best to make the relationship work, because
they own the decision in part. They can’t complain that the new su¬
perior was foisted on them. Ownership is important. During my
summer job at Procter & Gamble, I observed its system of hiring fac¬
tory workers. The personnel department sent candidates to be inter¬
viewed by the foremen, who ultimately did the hiring. Pretty
progressive for 1956.1 haven’t forgotten.
From Amgen’s science-based point of view, having staffers repre¬
senting various tiers of the company interview job applicants was
only another experiment, one intended to provide a deeper, broader,
multidimensional understanding of an unanswered question: is this
person compatible with our organization? In the laboratory, we might
expose a chemical to various substances to gauge its reaction; in much
the same way, observing how someone handles questions from a vari¬
ety of perspectives can be extremely revealing. Any setting will suf¬
fice, although one advantage of convening a small panel rather than
conducting a series of one-to-one interviews is that you avoid repeti¬
tive questioning.
One response always raised a red flag, at least in my [Link] was
when a candidate expressed surprise that the men and women who
180 Science Lessons

would be working under him were included in the interview. Some

seemed deeply [Link] was good to know.
But when I explained the logic of our interview policy, even the
skeptics usually came to see it in a different light.“It’s so important to
us to hire the right person,” I would say, “that we feel this is time well
spent. Plus, an interview is a two-way street. We want to give you an
opportunity to make the right decision for your future. And the more
people that you meet from Amgen, the better you’ll get to know our
company.”
It usually left a favorable impression. What’s more, interviewees
almost always came away dazzled by the caliber of our staff.

BE HONEST ABOUT THE JOB

Some organizations try too hard to [Link] set their sights on a

hot prospect and work overtime selling him on the organization,
even if it means revealing only selective information to paint a rosy
[Link]’s a big mistake. A job interview should be a mutual ex¬
ploration of whether the two parties would make a harmonious
match, with full disclosure on both sides.
When George Rathmann interviewed me for the position of
chief financial officer, he made no attempt to sugarcoat the serious¬
ness of Amgen’s looming financial crisis. Nor should he have. The
company needed a CFO who would be up to the challenge of ob¬
taining private or public funding, and I needed a vivid picture of
what lay ahead so that I could hit the ground running.
Do you know the number 1 reason that people leave their jobs
within six months? It’s feeling blindsided by unrealistic expectations,
about either the duties of the job itself, the company, or their role. To
be sure, it’s up to job applicants to ask questions until they’re satisfied
that they understand what will be expected of them. But some com¬
panies make pie-in-the-sky promises—about future promotions, im-
 How Amgen Built a Winning Team 181

proved facilities and equipment, and so on—that they know aren’t

likely to happen anytime soon.
Never try to snow [Link] might succeed and be sorry you
did! When the hard realization settles in that your new hire was misled,
whether it was a bald-faced lie or acts of omission, you have lost her
for [Link] sense of betrayal burns too intensely to be extinguished.
United Parcel Service, alarmed that half of its part-time ware¬
house workers were quitting the company, discovered that many of
its interviewers were leaving candidates with the false impression
that a step up to full-time status was just around the corner. In reality,
it could take years, if it happened at all. After UPS corrected the lapse
in communication, its turnover rate plummeted to only 6 percent.
Other techniques for giving potential employees a glimpse of what
they can reasonably expect include escorting them on guided tours
of the workplace or producing videotape or DVD presentations with
realistic portrayals of everyday situations.
Always tell applicants the whole truth. In that way, if you should
make the mistake of offering the job to someone who isn’t right for
the position, hopefully he’ll take himself out of the running—spar¬
ing you from reaching the same conclusion after the fact. It’s like
self-selection, but in reverse.

VOICE YOUR VALUES DURING

THE INTERVIEW

Companies, like people, have personalities. In the same way that

modifying DNA changes an organism’s genetic structure, whenever
you add someone new to the payroll you risk altering the psycho¬
logical makeup of the organization, especially if it’s a small company
or [Link] can help ensure that the chemistry will be com¬
patible rather than combustible by espousing your company values
prominently during interviews.
182 Science Lessons

Ed Garnett was Amgen’s head of human resources from 1994 to

2002. He recalls, “When interviewing applicants, I used to tell them,
‘If you read our list of values, and Vhey don’t match your personal
values, you shouldn’t work here. Because you won’t be happy, and we
won’t be [Link] Amgen values reflect what life is like here.’”The
interview report form used by our HR team required them to assess
a candidate on each of the values: ethics, trust and respect, and so on.
Ed continues:

By the time the person arrived at my desk, they’d passed muster

regarding their technical skills and whatnot. So our focus was
on assessing how well he or she would click with our culture.
One of my favorite techniques was to ask, “How do you think
you would react under the following circumstances?” and pre¬
sent them with a situation that they might face.
Their answers would tell us a lot about whether or not
they were likely to blend in. I might describe a scenario
where a staffer showed poor judgment. If the response was, “I
would call the person in and put him on probation!” well, that
wouldn’t work at Amgen. For one thing, we didn’t discipline
people for making honest mistakes.
They couldn’t have known that, naturally. But it gave us a
glimpse into their managing style. We would try to weed out
the more parochial-type administrators and, in general, any¬
one who was too set in his ways. We looked for people who
were collaborative and entrepreneurial, which was more the
way that Amgen did business.

I used to stress to applicants that they shouldn’t take the job if

they were uncomfortable with change, because Amgen was moving
rapidly in many different directions. It was a vital piece of informa¬
tion. In the past, we had unwisely hired a few scientists from conven¬
tional pharmaceutical companies, where one year is just like the last.
 How Amgen Built a Winning Team 183

After a while, people can develop an almost civil service attitude. It

wasn’t that they were bad scientists, but they felt comfortable only in
a regimented environment; they didn’t want to be pushed out onto
the cutting edge of science. The world of biotechnology was so new,
though, that the rules hadn’t been written, and this made them ex¬
tremely ill at ease.
For instance, when testing drugs in clinical trials, our researchers
designed their own studies. Inevitably, a recent arrival from Big
Pharma would ask, “Where’s the Amgen clinical trials manual?”
Apparently the big drug manufacturers’ product-development de¬
partments had a primer on how to formulate a one-size-fits-all in¬
vestigational study.
“What are you talking about?” we would say. “We don’t have
5?
one.
The scientist would chuckle nervously. “Come on, stop kidding
around. Where’s the manual?”
“We don’t have one.”
“Every company has one! You can’t not have one!”
Our people would explain that because every product is differ¬
ent, every clinical trial is different and must be customized accord¬
ingly. Some newcomers were put offby this; they had forgotten how
to think for themselves and were so far out of their depth that they
didn’t know what to do next. After all, the one thing that every bu¬
reaucracy does is to substitute rules and regulations for thinking. We
wanted to substitute thinking for rules and regulations.
We learned that for the Amgen approach to succeed, we needed
extremely talented, adventurous people. Fortunately, those types would
be attracted to a company like ours anyway, whereas those who are
afraid to go out on a limb would look for an organization that oper¬
ated by the [Link] was fine, because we didn’t want them.
Establishing values as a centerpiece of your employment criteria
may not eliminate weak hires, but it will certainly trim the number
of poor picks. Think about why people get fired. It’s usually not that
184 Science Lessons

they lack the necessary skills. More likely, they’ve alienated their
coworkers, or they’re ineffective communicators, or they’re not team
players. Whatever the reasons, rightly or wrongly, they’re out of step
with the rest of the organization. This principle cuts both ways: at a
poorly managed company, really good employees tend not to last
very long because their work ethic, integrity, and so on conflict with
the prevailing office culture.
Occasionally you come across an exceptional talent who does
not play well with others in the corporate sandbox. You’re aware of
this because her track record precedes her; or perhaps her combative
demeanor throughout the interview makes it evident that this is a
person who’s probably going to step on some toes. In short, she’s the
reason that on the fourth or fifth day, I forget which, God created
managers. Should you hire this prickly personality, or should you let
her be a thorn in the side of some other organization?
It depends on your needs at that moment. In business, decisions
should be weighed using the twin scales of benefit and risk. You de¬
termine which is greater. Let’s say that you run a struggling opera
company, and the only thing that will rescue your production of
Puccini’s Tosca is to sign up for the lead role a certain celebrated so¬
prano with notoriously erratic behavior and a volcanic temper.
There’s no question what you should do: get La Diva now. You’ll
enforce company rules and minimize star treatment as best you can.
Often, if a prima donna’s presence strengthens the organization, ben¬
efiting everyone, your other employees may be willing to accept the
inevitable double standard. But if her talent fades or her conduct
starts to poison the atmosphere, then everyone’s tolerance, including
management’s, may grow thin. After all, even the great Maria Callas
was fired by the Metropolitan Opera.
However, a company can abide and willingly give special treat¬
ment to only a few superstars.
Prodigious talent isn’t always enough to offset poor work habits
or a disagreeable personality, but I discovered the hard way that talent
 How Amgen Built a Winning Team 185

almost always trumps experience. It took me years to learn that ex¬

perience alone isn’t a reliable barometer of someone’s expertise; it
could mean only that he has taken a lifetime to perfect mediocrity.
The more salient point is that an employer should look beyond
today’s need and not merely fill a slot. Companies change, and job
descriptions change; a person with ability is more likely to grow and
adapt.
So don't reflexively toss aside an intriguing resume just because
the candidate’s previous duties may not conform exactly to your
current requirements or because he’s worked in a different industry.
True talent often renders such arbitrary boundaries meaningless. To
spot and reel in gifted employees requires probing beneath the sur¬
face of a curriculum vitae and extrapolating which past experiences
are transferable. Many skills transcend job titles and can be applied in
different arenas. You also need to educate yourself about industries
you’re not familiar with, because a practice that might be unaccept¬
able in your organization might be the status quo elsewhere.
For example, when my jet charter business, Prime Jet, was first
hiring pilots, I noticed from perusing resumes that some of the appli¬
cants had bounced around quite a bit—never an encouraging sign.
Then I learned why: many companies that employ pilots are under¬
capitalized. All it takes is one slow month or a major repair, and they
vanish hke a flight entering the Bermuda triangle. In the jet charter
business, it happens to a startling degree. Amgen wouldn’t have even
considered someone with, say, five pit stops in nine years. But this
wasn’t Amgen, and it wasn’t the biotechnology field.

TINKER WITH THE JOB DESCRIPTION

While you’re at it, you can apply the Amgen principle “The people
who do the work should help plan the work” (see the Introduction)
by soliciting feedback from coworkers. How can the position be tai¬
lored to best reflect the organization’s needs at this time?
186 Science Lessons

Perhaps the departing manager was ideal for the job when you
hired her three years ago. But the company may have changed since
then, or the departments role within the organization may have
changed, and now the skills called for are dramatically different. Once
you’ve hired a replacement, fine-tune the job descriptions to maxi¬
mize his abilities.

HIRE PEOPLE FROM TOP COMPANIES

Why should you go after people who have worked for top compa¬
nies? It’s because they’ve been surrounded by talented colleagues—
that’s what makes a company great—and probably have picked up
good work habits. One professional football team has stated publicly
that it gives strong preference in the annual college draft to players
from colleges with highly successful football programs and excellent
coaches.

USE THE WORLD’S BEST RECRUITERS

Your company’s most effective ambassadors of goodwill are already

on your payroll. Approximately one in three hirings depends on
word of mouth. If you foster an atmosphere where people generally
feel productive, valued, challenged, energized, and proud of their or¬
ganization, they’ll tell their friends and former colleagues, “Hey, we
have an opening for an accounts payable manager. This is a great
place to work. Why don’t you send over your resume?”
Have you ever heard the saying “Turkeys don’t fly with eagles”?
We tend to associate with people who share our values, including
our attitudes toward work. So when a good employee recommends
your company to others, they’re likely to be quality candidates with
compatible values. Talent becomes self-perpetuating, because tal¬
ented people want to play for a winning team alongside other ac¬
complished men and women.
 How Amgen Built a Winning Team 187

Every month I used to present “An Evening with Gordon

Binder.” It sounds like a cabaret act, but it was a presentation to our
new employees. I would tell them, “A recent study showed that for
the past one thousand jobs filled here at Amgen, we had fifty thou¬
sand applicants. Each of you is one out of fifty. On average, forty-
nine people wanted your job and didn’t get it; you did. So you are
part of an elite group. Now that you’re here, it’s your job to keep this
going and to keep Amgen the kind of place where the best people
want to work. We expect you to recruit people just like you.”
As a result, Amgen rarely had to advertise or use an outside re¬
cruitment firm—for example, when we needed a superspecialist or
when our expansion was so rapid that the employee referral system
couldn’t keep up with the demand. You can imagine how much
money that saved over the years.
I’m seeing the same pattern at Prime Jet, where we treat our pi¬
lots and other staff better than do most of our competitors. Less than
two years after starting the company with one aircraft, we purchased
a second Gulfstream IV. The jet became available unexpectedly when
an order was canceled, so we didn’t have a hiring plan in place.
I worried aloud that perhaps we wouldn’t be able to hire the six
pilots we needed fast enough. Then my general manager smiled and
opened a desk drawer. “We can start with these,” he said, pointing to
more than one hundred resumes that people had sent us unsolicited.
They were good people, too: we promptly interviewed the ten who
looked best on paper and hired four of them. It was possible because
the staff had put out the word that Prime Jet is a great place to work.
Two-thirds of our new employees have come to us that way.
Some organizations, such as computer networking giant Cisco
Systems, have instituted programs that give cash rewards to employ¬
ees for [Link]’s not a wise idea, because it implies that helping
recruit people is not part of everyone’s job but is optional. Everyone
in the organization should naturally consider it his responsibility to
encourage people he feels would be assets to join him as coworkers.
188 Science Lessons

EXPLORE UNCONVENTIONAL
TALENT POOLS

As you build your exceptional workforce, don’t overlook these sources

of talent.

Baby Boomers

The generation born between 1946 and 1964 will almost certainly
remake retirement in its own image, either by refusing to retire at all
or by choosing to work part-time between yoga classes and skydiv¬
ing. Given that the subsequent generation, the so-called Generation
X, is only about half the size, the United States is facing an imminent
shortage of men and women in the thirty-five- to forty-five-year-
old age bracket. Boomers, generally well educated and health con¬
scious, should become increasingly attractive to business. Firms have
already rolled out the welcome mat by offering benefits to part-time
workers.

The Armed Services

The military might be the richest repository of reliable workers in the

United [Link] of Amgen’s three chief executive officers served in
the U.S. Navy: my successor (Kevin Sharer) and me. One of my
CFOs was a navy man, too.
Yet veterans are generally an untapped resource, to the extent that
many seeking to enter the private sector face out-and-out discrimi¬
nation. That’s too bad but not surprising when you consider that
since the draft was ended in 1973, surprisingly few civilians have
much personal contact with veterans—especially women under, say,
forty, who make up a sizable proportion of HR departments. As a re¬
sult, misconceptions abound about former military men and women.
 How Amgen Built a Winning Team 189

I have been privileged to meet many senior admirals and gener¬

als, including the second-ranking army and air force generals at a
breakfast at the Pentagon. Almost without exception, they were ex¬
tremely impressive people—-more impressive, to tell you the truth,
than the average senior business executives I encounter.
Probably the most deeply ingrained stereotype is that veterans are
rigid automatons unable to think for themselves. It’s true that being
in the armed forces teaches you respect for authority—try finding a
boss who wouldn’t welcome that!—but the typical servicemember
makes as many pivotal decisions daily as any company [Link]
U.S. Army’s longtime slogan “An army of one,” recently retired, is
truer than most folks realize.
Kevin and I have often talked over the years about what a valu¬
able experience our navy service was and how it has benefited us in
business. I may be guilty of generalizing here, but people who come
out of the military tend to exhibit many of the traits that a manager
should have: leadership, responsibility, self-sufficiency, a belief in
teamwork, and a drive to get the mission done, all grounded by a
moral center. I particularly like the fact that military veterans rarely
make excuses.
Some of the most successful companies in the United States have
made a concerted effort to recruit veterans, among them The Home
Depot, Adolph Coors Company, General Motors, 7-Eleven, Ameri¬
can Express, Hershey, and Johnson & Johnson. That’s an impressive
list, isn’t it? Maybe you should consider following their lead.

ENSURE A SMOOTH TRANSITION

Smart hiring doesn’t end with a hearty handshake and “Welcome

aboard!” Far from it. Being the new person at work is no less anxiety
inducing than being the new kid at school. It can take months to
undo an awkward start. It is imperative that new employees not be
190 Science Lessons

thrown into the deep end to sink or swim while the boss turns his
back. If they sink, they may carry others with them.
The damage isn’t confined to staff morale and productivity. Ac¬
cording to one estimate, a midlevel senior manager must work for
roughly six months before his new employer begins recouping its fi¬
nancial investment, which includes salary, training, recruitment, and
possibly relocation expenses. If he quits, you might have to tack on
severance, too.
Increasingly, organizations are turning to a “buddy system” to
help new employees navigate their unfamiliar surroundings. Natu¬
rally, this positive trend has spawned a new buzzword in HR circles:
onboarding. (It sounds like surfer-dude speak to me, but maybe that’s
because I live in Southern California.) The buddy is a colleague who
makes it her business to introduce the new hire, answer questions
and explain the idiosyncratic protocols peculiar to all organizations,
show him where to obtain supplies and equipment, and generally
offer encouragement. Not insignificantly, she also asks the newcomer
to join her for lunch.
Amgen never had a buddy program per se, for the same reason
that we didn’t go around handing out cash rewards for referrals. Our
philosophy was that every member of the team should help everyone
else, period, especially if someone was struggling. But any system,
formal or informal, that helps folks get acclimated is a good thing.
Upper management should receive the same consideration. To
stumble out of the gate at that level can have disastrous consequences,
shredding credibility forever. Some executives’ oversized egos may
not let them accept special support from their new organization, be¬
lieving that it would undermine their authority. But if it leads to im¬
proved communication and a smoother transition for everyone, why
not welcome it?
In a survey of more than one hundred new senior executives,
three in five claimed to be dissatisfied with their companies’ efforts
to help them get settled. When Amgen’s partner Johnson & Johnson
 How Amgen Built a Winning Team 191

implemented an onboarding program for incoming corporate lead¬

ers, the response was overwhelmingly positive. Afterward, 125 of the
participants were asked to put a figure on how much the extra assis¬
tance improved their performance during their initial time on the
[Link] pegged it at an average of 30 percent to 40 percent.
After thirty days, it’s a good idea for the newbie and his superiors to
sit down and review any concerns on either end. If he’s working out
well, tell him how pleased you are; it’ll go a long way toward alleviating
any uncertainty he might have. A follow-up session at sixty or ninety
days—or, better still, at both landmarks—certainly couldn’t hurt.

LOOK INWARD TO MOVE FORWARD

An article I read recently quoted a human resource administrator as

saying that three out of five employees possess all the basic skills nec¬
essary to fill every job in a typical company. I wouldn’t go that far,
but I know that flashy outside applicants are often hired instead of
highly competent internal candidates.
Frequently, the most qualified candidates for a position are right
under your nose. And this is something that can be proved. A report
published by the School of Hotel Administration at Cornell Univer¬
sity’s Center for Hospitality Research analyzed twenty studies of em¬
ployee productivity covering a wide range of occupations: welders,
insurance salespeople, bank tellers, teachers—even research scientists.
The authors of “How to Compare Apples to Oranges” concluded
that interviewing applicants from outside a company isn’t nearly as
reliable a predictor of future competence as performance reviews of
in-house employees. Not only that, but most of the time the steady if
unexciting person already in the fold is a better choice than the ex¬
ternal candidate who wows everyone in interviews.
That shouldn’t be surprising. As I’ve said, most terminations stem
from the worker’s incompatibility with the company’s culture, not an
inability to do assigned tasks. Someone who has worked in the same
192 Science Lessons

firm for a while has demonstrated that she knows how to get things
done within the organization, eliminating a major concern. Increas¬
ingly, businesses seem to be recognizing this, as nearly half now let
their current workers apply first for new job opportunities. Only
then do they advertise the openings.
Early in Amgen’s history, we did a lot of promoting internally be¬
cause of financial necessity. Although we never had a single layoff
during the mid-1980s, we instituted a hiring freeze once for about
twelve months because we couldn’t raise the necessary money at that
[Link] was before Epogen had entered clinical trials. Kirby Alton
remembers how it was:

When we needed something done, we’d look around the

company and say, “Okay, you do it.”That’s more or less how I
ended up as director of therapeutic-product development,
which entailed finding doctors to run our Epogen studies and
going to Washington to deal with the U.S. Food and Drug
Administration—in short, getting EPO on the market. One
day George Rathmann called me into his office and told me
he’d hke me to take on the position.
I said, “Why me?” I was a molecular biologist five years
out of grad school.
He said,“I’ll take my chances with you.”
I just decided, “Well, this is what we have to do.” So we did
it.

I mentioned earlier that the tragic death of our head of sales and
marketing forced me to replace him, and the man I chose was vice
president of operations Dennis Fenton. He may have been a scientist,
but I had every confidence that he would excel despite having zero
experience in sales and [Link] years later, I made our director
of logistics the head of human resources. That was Ed Garnett, whom
you met earlier in this chapter. Ed was smart, he had tremendous peo-
 How Amgen Built a Winning Team 193

pie skills, and, not coincidentally, he had trained for several years in a
seminary, something that made him an extraordinarily good listener
and counselor. He remembers well the day I offered him the job:

Gordon called me into his office and said he wanted me to be

his HR guy. I laughed uproariously. “What are you doing
talking to me?” I asked.“I’m an operations guy!”
But he insisted. He looked at me and said, “You’re very
good with people. You understand us and our culture. You’re
patient, and people respect you. You could work for me as a
vice president.”
When I heard “vice president,” I thought, “You know, I
probably could learn this; I guess it wouldn’t be so tough.”
And that’s what I did.
Most companies and most CEOs wouldn’t do that. They
would worry, “Gee, if I make this unorthodox move, and it
fails, I’m really opening myself up to criticism:‘How could he
be so stupid as to appoint the logistics guy to head HR?”’

The truth was, I had the advantage of having seen this kind of
move work before. While at System Development Corporation, I’d
watched a talented sales and marketing executive go on to become
the company’s head of human resources, and be highly successful.
Harry Gray, my first boss, joined Litton Industries as a public rela¬
tions man with a master’s degree in journalism. He eventually be¬
came chief executive officer of United Technologies.
Cultivating talent from within is a major component of em¬
ployee retention, a topic explored in depth in the next chapter. It’s
not only a matter of bold leadership; as manager, you must encourage
people to step outside their comfort zones and convince them they
will succeed. Just as importantly, you must reassure them that if things
don’t pan out, you’ll absorb the blame for the decision. If manage¬
ment isn’t willing to risk its neck, why should they?
 ,

'
CHAPTER
10
How Amgen Kept
Employees Committed

the FACT that the annual turnover rate in the United States ap¬
proaches 40 percent should tell us that not enough employers are pay¬
ing attention to the breeze being stirred by the revolving door. Nor are
they in touch with their people’s professional aspirations and what
they look for in a job. In my experience, strong leaders are almost
without exception born teachers. Throughout my career, I’ve had the
good fortune to learn from some of the best, such as Amgen’s George
Rathmann, SDC’s George Mueller, and Harry Gray of Litton Indus¬
tries. And as an executive, I’ve always found mentoring younger col¬
leagues to be one of the most rewarding aspects of running a company.
Today more than ever, a boss’s ability to nurture high-caliber peo¬
ple is perhaps the most effective, cost-efficient way to prepare an or¬
ganization for the future. In a large survey (conducted by respected
consulting firm McKinsey & Company) of executives from seventy-
seven companies, three in four lamented a lack of available talent. The
196 Science Lessons

problem seems likely to deepen in light of two trends. One is the im¬
pending shortage of workers in general. The U.S. Department of
Labor predicts that by 2010 at least^ten million positions will go un¬
filled. And that doesn’t take into account the mass exodus of older
workers, beginning in 2011, as the baby boomer generation reaches
retirement age and starts trading in its power suits for Bermuda shorts.
At the same time, there will be fewer skilled men and women to
replace them. The growth rate of the U.S. population has been de¬
clining since the 1990s and is expected to continue that way. In addi¬
tion, we’re entering a slowdown in the share of workers possessing at
least some education and training beyond high school, further shrink¬
ing the pool of qualified managerial candidates.
The other major factor affecting staffing is the public’s fractured
trust in corporations following a decade of deplorable corporate
scandals and rampant downsizing. Sadly, the concept of mutual loy¬
alty between employer and employee is fading from the American
picture. Every other year, Indianapohs-based research company Walker
Information assesses the state of worker commitment. Its 2001 and
2003 reports are sobering. Even in the midst of an anemic job market,
approximately two in three employees polled said that they planned
to leave their employers within two years.
The U.S. Department of Labor has put the cost of losing a
worker at one-third the annual salary of a new hire. When a manage¬
ment-level employee departs, an organization can expect to forfeit
more than his yearly pay in recruiting and training a replacement.
That’s not to mention the ding to staff morale and the disruption to
ongoing projects.
You’ve seen Amgen’s financial numbers during my twelve years
as CEO (see the Introduction). Also extremely satisfying to me was
our 5 percent turnover rate in personnel—half the industry average
and, at the time, less than one-third the median figure for U.S. com¬
panies overall. Of course, high retention can breed complacency in
some people. (It may seem counterintuitive, but aggressively weed-
 How Amgen Kept Employees Committed 197

ing out the least-desirable workers is one of the secrets to maintain¬

ing a low turnover rate, for reasons I discuss later.) The goal is to cre¬
ate an environment that encourages valuable employees not only to
stay but also to stay motivated; a growing body of research substantiates
something that most bosses knew intuitively long before the first man¬
agement consultant came calling: an energized, committed workforce
improves the bottom line. Conventional business wisdom (not al¬
ways a ticket to smart business practices) decrees that the way to keep
valued employees is to occasionally throw them a bone, in the form
of a raise, a company car, and so forth. Certainly we all want to be
compensated fairly for our skills, and a competitive salary is indeed
powerful bait for luring workers to a company. But that’s not what
discourages professional wanderlust.
Likewise, no amount of perks can compensate for a company’s
deficiencies. I learned this from the military, of all places, when I was
treasurer and vice president of finance for System Development
Corporation. After the Vietnam War ended in 1973, the decision was
made to discontinue the draft. The last time the country had con¬
verted to an all-volunteer military, following World War II, enlist¬
ment plunged, forcing the government to quickly reinstate selective
service. A quarter-century later, the Defense Department was under¬
standably worried that history might repeat itself. An uneasy peace
may finally have been brokered in Southeastern Asia, but Cold War
tensions between the United States and the Soviet Union still sim¬
mered. A way had to be found to make an all-volunteer force work.
So the army introduced a number of costly policies intended to
attract and keep recruits once their hitches ended. One of them ac¬
tually called for permitting beer in the barracks. In classic bureau¬
cratic fashion, before long more than two hundred such benefits had
been implemented. Unfortunately the military had no idea which
ones, if any, truly worked.
SDC primarily designed information technology and computer
systems, but we also had a psychology department. The Pentagon
198 Science Lessons

awarded our scientists a multimillion-dollar contract to evaluate the

program’s [Link] of servicemen and women were in¬
terviewed at length about their arpay experience, supplemented by
questionnaires and focus groups. Interestingly, they rarely complained
of wanting more this or better that. Their primary concerns had to do
with negatives, especially the seemingly trifling, irritating rules and
regulations that, according to the soldiers, could make military life un¬
bearable. Ultimately, we helped the army discover that the most effective
and economical way to maintain troop levels wasn’t to offer one in¬
centive after another but to eliminate the negatives as much as possible.
The same principle is just as applicable to a division of a corpora¬
tion as it is to a division of soldiers. Amgen strove to be a company
where the most talented scientists would want to put down roots.
Many organizations unwittingly undermine their personnel by clut¬
tering the path to progress with roadblocks, whether it’s superfluous
paperwork or a labyrinthian corporate structure that makes accom¬
plishing even the simplest task an exercise in head-banging frustra¬
tion. Amgen offered generous salary and benefits, but so did plenty of
biotech firms. However, George Rathmann instinctively understood
what the army spent millions of dollars to discover. At least once a
month top management used to sit down with some of our re¬
searchers and ask them flat out, “How can we help you? Is anything
getting in the way of your work?” As a rule, if any administrative red
tape was found to be tying up research and development, it was un¬
ceremoniously snipped. These steps cost us little or no money, yet the
scientists were as appreciative as if they’d received 10 percent raises.
Similarly, at Prime Jet, we sit around asking, “Is anything bother¬
ing the pilots?” It’s usually some little thing that’s dealt with easily.
Here’s one: we have an agency that books the hotel rooms for our pi¬
lots wherever they may be around the world. One time, according to
the pilots, the hotel booked for them was so filthy that they went out
and found another one and paid for it with their own money—then
grumbled about it.
 How Amgen Kept Employees Committed 199

“Okay,” we said, “new rule: if you don’t like the hotel room the
agency books for you, get a different one, and the company will re¬
imburse you. But report it, so we can make sure they never book that
place for us again. Pay for another room yourself or be forced to stay
in a bad room? What could be more ridiculous?” The fact that some¬
body showed enough concern to make sure it wouldn’t happen
again told the pilots a lot about our company.
Now, in less enlightened sectors of the business world, where
managers still bark out orders, proposing such a simple approach to
lifting morale and productivity might prompt an outbreak of head
scratching. No carrot? No stick? Won’t work. Granted, it does run
counter to the lessons learned in many business books and business
schools. If you take into account what really drives human nature,
however, the logic becomes clear.

What Do Employees Want?

Pretty Much the Same Things That You Want

That’s what several large surveys of American employees have

determined.1
The researchers may have phrased their questions differently, but
basically they all asked the same thing: What do you want from your
company that would make you more inclined to settle down to¬
gether? Among the responses heard most frequently:

• The opportunity to contribute ideas and input on decisions

• A collaborative work environment

1. See “The Walker Loyalty Report: Loyalty in the Workplace,” [Link]

com/what/loyaltyreports/studies/employee07,Walker Information, Inc., 2007; “Work¬
ing Today: Exploring Employees’Emotional Connections to Their Jobs,’’Towers Per¬
rin 2003 Talent Report; “Working Today: Understanding What Drives Employee
Engagement,” Towers Perrin 2003 Talent Report; Nancy Glube, “Retention Tools for
Turbulent Times,” Society for Human Resource Management,January 1998.
200 Science Lessons

• Challenging work

• Feelings of accomplishment and satisfaction

• Training and career-advancement opportunities

• Equipment, information, and other resources to get work done

• Management demonstrates concern for employees’ well-being

• Respect for a balance between work and home/family

• Pride in the company

• To feel appreciated, both in terms of comphments and finan¬

cial compensation

Using real-life examples of our corporate values in action, this

chapter presents my thinking on how we kept employees engaged
and committed. I think you’ll see why Amgen consistently landed on
Fortunes annual lists of “America’s Most Admired Companies” and
“The 100 Best Companies to Work For”—and how we achieved
our low turnover rate.

CREATE A COLLABORATIVE WORK

ENVIRONMENT

Ever been to Japan? If you walk past a Tokyo department store ten
minutes after closing, through the windows you’ll see the employees
in each department gathered in a circle to review the day’s events and
prepare for the next day. They, and not the supervisors, do most of
the talking. The underlying philosophy, which Amgen adopted to a
greater extent than did most U.S. companies, is that the men and
women who do the work should help plan the work. After all, they
intimately understand the job at hand and have a realistic perspective
 How Amgen Kept Employees Committed 201

on how much time and resources a task will take. We used to invite
the future occupants of a new office or plant to give us feedback on
the blueprints before we finahzed the design.
Now, some would call that going overboard. I respectfully dis¬
agree, since those people were going to have to work there. After
we’d built several new buildings and learned from our mistakes, we
were able to go from ground breaking to occupancy in twelve
months for a multistory office structure, and fifteen months for a
laboratory.
Another consideration in soliciting employees’ input is so that
they take ownership. If you’ve contributed to a decision, you’re
probably going to try harder to make it work than if orders are
handed down from on high; especially if the demands seem imprac¬
tical or unreasonable.
This bottom-up style of management is gradually infiltrating
American companies-—startup ventures in particular. However,
there’s still a heavy reliance on the traditional top-down approach,
where the manager sets the course for the team, who are expected to
carry it out. A study by the Work in America Institute, a professional
organization for human-resource managers, headquartered in Scars-
dale, New York, reported that fewer than 1 in 10 U.S. employees
enjoy a truly participative environment at work.
Amgen was a place where ideas flowed freely, and people were
empowered to make decisions with minimal bureaucratic interfer¬
ence. You couldn’t run a biotechnology company full of scientists
any other way. Most of them came from academic laboratories,
which afforded them plenty of autonomy. In contrast, commercial
labs can be stifling, especially at conventional pharmaceutical firms,
where marketability and profit potential often take precedence over
scientific merit.
The freedom to decide how work should be done wasn’t lim¬
ited to R&D. I remember one time, in the 1990s, the people in our
202 Science Lessons

shipping department came to me with a proposal. One of Amgen’s

eight values was to ensure quality. With that in mind, they wanted to
ship every order within twenty-foUr hours.
I told them I appreciated their dedication, but it just wasn’t possi¬
ble. The products required refrigeration, and although they were
packed in reusable gel-pack ice, they couldn’t be sent on a Friday and
sit around over the weekend. Mondays had to be busy shipping days.
“We can’t afford the extra staff on Monday and have half the people
in on Friday with no work to do,” I explained. “It’s too expensive.”
And distributors routinely carry plenty of inventory, so it wouldn’t
be important to them anyway.
The shipping department people thought about it some more,
came back, and asked, “If we ship everything within twenty-four
hours with no additional staff, then would it be OK?”
“I don’t know how in the world you’re going to do that,” I said,
“but sure.”
On their own, they tinkered with their hours. Instead of working
eight hours a day five days a week, they began to work ten hours on
Mondays. Fridays, reduced to six hours, were spent getting ready for
the big push on Mondays. Although you couldn’t take the medica¬
tions out of the refrigerators and put them in the shipping boxes on
Friday, a lot of prep work could still be done.
The change in procedure benefited distributors all year but really
proved its value each December. Large pharmaceutical companies
often shut down for all of Christmas week, forcing their distributors
to order extra product to cover patients’ needs during the holidays.
As you can imagine, storing all that extra inventory was expensive.
Amgen, in contrast, was able to continue shipping on schedule. Our
distributors loved us for it and showed their gratitude by carrying
smaller inventories of our products and charging our customers a
lower [Link] this day, no one else in the pharmaceutical industry
does that. And the innovation can be credited to the folks in ship-
 How Amgen Kept Employees Committed 203

ping. All management did was listen to them and let them imple¬
ment their ideas.

Give People the Right to Be Wrong

Whenever you propose an idea, regardless of your standing in the

corporate hierarchy, you’re going out on a limb. People have to know
they have the right to be wrong on occasion, or else they’ll stop tak¬
ing risks.
For that matter, how many great ideas progress from conception
to implementation without a misstep or two? Very few, I imagine—
especially in the uncertain world of biotechnology, where disap¬
pointments far outnumber successes. As Dennis Fenton, now Amgen’s
executive vice president of operations, says, “In an open environ¬
ment like ours, failure is regarded as part of the process.”
A company’s staffers also must feel secure that their superiors
aren’t going to abandon them or scapegoat them if a plan fails. Oth¬
erwise, they won’t volunteer ideas. Or, worse, they might try to con¬
ceal mistakes from higher-ups.

Work in Teams

Amgen’s dependence on working in teams and reaching a group

consensus all but eliminated that concern. I’ve described how our
science-based culture—letting data from our experiments guide
us—depersonalized decision making. Along similar lines, when
everyone on a team takes responsibility for a decision, no one person
can be singled out for finger-pointing if things go awry; the focus is
more on problem solving than fixing blame.
A consensus-driven approach helps facilitate sound decisions by
minimizing interference from clashing egos, infighting, and the like.
204 Science Lessons

It isn’t that it doesn’t pose its own unique set of challenges, starting
with the common misconception that consensus is synonymous with
“unanimous support.” V
“It was never supposed to mean ‘unanimous,’” says Dennis Fen¬
ton. “What it means is that you participate in the discussion and give
your input, which the team leader is obligated to consider. Then the
team decides what to do and is accountable for executing whatever
the plan is.” The first time Dennis attended a meeting at Amgen after
his arrival in 1981, “it was a revelation to me that here I could actu¬
ally be heard,” he says. “At Pfizer the attitude was, ‘Just do what
you’re told; we don’t care what you think.”’
A team can reach a consensus even though one or two members
may harbor [Link] bottom line is, Can everybody live with
the decision? If so, then they must all work to support it fully—be¬
cause if the project fails, the excuse that “I knew this was going to
fail” isn’t going to fly.
The team leader’s job is to guide the discussion at meetings. Not
to tell members of the team what to do or to necessarily make a de¬
cision. More often than not, a clear consensus emerges from the data
presented. If you de-emphasize position or level of authority, every¬
body in the room is just a member of the team. This encourages even
the lowest-level person (who often has more detailed knowledge
than anyone else in the room) to speak up, as opposed to meetings
where the boss makes it clear that his view will predominate. Some
attendees are going to think, “Why should I bother giving my opin¬
ion? I’ll just get in trouble.”
When explaining the team approach during our monthly orien¬
tations for new hires, I pointed to a chart titled “Who Manages
You?” (see figure 10-1). The upper half showed the distribution of
authority at most companies, and below that, at Amgen, using differ¬
ent-sized circles. In the typical organization, “Your boss” was the
largest circle, followed by “You” and, last in significance, “Peers.”
 How Amgen Kept Employees Committed 205

FIGURE 10-1

Employee independence at Amgen

Who manages you?

Typical
(^Peers^j
company

Amgen

You can see from the chart that in our company, employees were
granted a good deal of independence. Your peers supervised you
more than the boss did. “That’s what working in teams means here,”
I concluded, usually to a room of pleasantly surprised faces—and
some skeptics. It didn’t take long, though, before they discovered
that the “Who Manages You?” chart accurately depicted the Amgen
way of doing things. Sometimes, however, the team leader has to ex¬
ercise authority, mostly in the interest of keeping the group from
getting mired in interminable discussions. If we didn’t come to an
agreement within a reasonable amount of time, I might look at my
watch and say, “Well, we’ve spent an hour on this and heard a lot of
facts and viewpoints. We haven’t agreed on either plan A or plan B.
But we’ve got to move on, so I’ll arbitrarily say that we’re going to
go with ‘A.’”
Everybody accepted it as a boss’s prerogative. I didn’t do it very
often, mind you; that would be dictatorial, which definitely was not
the Amgen way. The decision came with a tacit caveat, too, that if
206 Science Lessons

time demonstrated we’d gone down the wrong road, we’d reconvene
and either correct our course or change direction completely. That’s
another advantage of deciding by consensus: Since the action isn’t
associated with one person—the Entire team owns it equally—it’s
easier for members to concede that things aren’t working out. No
need for saving face, an understandable human response that usually
has unfortunate consequences in the workplace. Also, if the decision
was science based, everyone would readily accept a change if new
data happened to invalidate the previous decision.

Make Your Meetings Matter

One of the biggest complaints in most companies is that meetings

eat up too much time and don’t always accomplish very much. In
keeping with Amgen’s determination to trim red tape, we did some¬
thing about that by instituting performance reviews of meetings.
First, we took a hard look at our regularly scheduled meetings,
whether they were daily, weekly, or monthly. Very often those are the
first meetings to outlive their usefulness, because circumstances usu¬
ally change. Ask yourself, Is this meeting still constructive? Sometimes
you conclude that the same information can be conveyed more effi¬
ciently via e-mail or memo, enabling you to schedule fewer meetings.
If you weed out even a few sessions, you’ve freed up time on every¬
one’s calendar.
We decided that we would hold only one meeting on any deci¬
sion that needed to be made, with occasional exceptions. We didn’t
want to discuss the decision, adjourn, and then reconvene the next
week. An agenda was set days in advance. The idea was that anyone
with relevant information should attend. In other words, speak now
or forever hold your peace. Consequently, our meetings tended to be
large, a practice that violates most companies’ rules. You would have
key people sitting at the conference table, with junior staffers and
 How Amgen Kept Employees Committed 207

various experts sitting behind them. They rarely spoke unless some¬
one at the table asked them to contribute.
When you make it clear that you’re hell-bent on reaching a deci¬
sion before everyone files out the door, people focus. I remember
one meeting that went on for about an hour without consumma¬
tion. Finally, an irritated voice spoke up,“We’ve been here an hour. If
we don’t make a decision real soon, I’m leaving!” It forced everyone
to get down to business.
Sometimes there isn’t sufficient data on hand for making a deci¬
sion, in which event you postpone it until fresh information becomes
available. Otherwise, what happens? The same data gets restated, and
people cling more tightly to their positions. Managers shouldn’t let
themselves feel pressured to go ahead with a decision they believe is
reckless. Sometimes, though, “We need more information” is an ex¬
cuse to delay pulling the trigger. If you wait long enough, the data
will always be better. The question is, Is it reliable enough now to
plunge ahead? Furthermore, no decision is irreparable; it can always
be corrected. But falling into a pattern of indecision is deadly, because
time truly is money.
In another break from convention, we did not defer meetings just
because “Bill” or “Nancy” couldn’t make it. Someone is always sick, on
vacation, or at a conference. I used to encourage regular attendees at
our weekly management meetings to miss a session or two. Their di¬
rect reports were to take their place, but not to merely jot down notes.
In a true baptism by fire, they were expected to serve as a proxy: to
have studied the agenda and to add to the discussion as if they held
their boss’s [Link] would even take part in the decision making.
After a while, people adapted to this disciplined style of meeting.
You show up prepared, bring all the experts you need, and stick to
the subject at hand, and no going off on arias just to delight in the
sound of your own voice. Believe it or not, it can be done, and can be
done consistently.
208 Science Lessons

Emphasize the Value of Collaboration

George Rathmann believed in the old saying that two heads are better
than one—or, in the case of most of our teams, six to eight heads. Let¬
ting colleagues volley an idea back and forth usually improves upon
the original faster than one person’s [Link]’re also quicker
to recognize whether a concept is fatally flawed and should be junked.
Collaboration was so deeply ingrained in Amgen’s culture that
we designed our new buildings with areas that promoted employee
discussions, such as clusters of seats in a sunny atriumlike setting. To
me, a building should function as a piece of equipment, facilitating
whatever an organization is trying to do. For instance, we put the
product-development team leaders on the same floor as my office in
our headquarters building. In part, the purpose was practical—close
proximity fosters collaboration—but it also communicated through¬
out Amgen that product development was at the heart of our com¬
pany and that the product team leaders were important, even though
they weren’t senior executives.
Product-development teams were responsible for everything re¬
lated to shepherding a new product to market, and other types of
teams handled other tasks. Every team had a designated leader, and
every team leader had a boss, who was expected to supervise the
leader with a light touch (see “Tips for Teams”).The members were
given veto power over who could be on the team, a policy that few
companies adopt.
Amgen, however, considered this practice essential. Let’s say the
team needed a statistician. It would be assigned one by the head of
the statistics department. This person was under the day-to-day su¬
pervision of the team leader, although he could consult his boss for
help with statistical or technical issues. If the team requested some¬
one else, for any reason at all, the statistician returned to his “home”
department, and the head of statistics had to provide a replacement.
 How Amgen Kept Employees Committed 209

Tips for Teams

• Groups of five to nine people tend to be most effective.

Adding two or three members has been shown to aid produc¬
tivity; if you go higher than that, however, scheduling conflicts
and other problems begin to surface.

• They say that familiarity breeds contempt, but teams whose

members stay together for a long time actually perform bet¬
ter. One exception, oddly enough, is teams in research and de¬
velopment. A study published in the journal Administrative
Science Quarterly tracked fifty project groups at a large cor¬
poration. It showed that R&D team productivity appears to
peak after about three years; therefore, it’s wise to add or sub¬
stitute new faces around that time.

Similarly, teams were allowed to replace members at will. Bear in

mind that a project might go on for years. So “Andrea” might be a
perfect fit for phase 1 but less than ideal for phase 2. It wasn’t neces¬
sarily a mark against her, because a group s needs naturally evolve
over time.

Encourage a Free-Ranging Perspective

You’ve heard of executives who say, “My door is always open.” At

Amgen, our meetings were always open to anyone in the company,
and that helped us sidestep all sorts of petty office politics (see “Every¬
body Gets to Play”). If someone tried complaining to me that So-and-
So hadn’t invited her to a meeting, I’d reply, “Why didn’t you go
210 Science Lessons

anyway?” End of discussion. As with any open-door policy, it was rare

to find people at a meeting who really didn’t belong there.
Nonteam members were encouraged to voice their opinions. If
our general counsel had an idea for something that he felt R&D
should consider, the research people had to listen to it. They didn’t
have to accept the advice or even explain the reasons for rejecting it,
but they did have to listen sincerely. By the same token, if one of our
researchers made a suggestion related to an ongoing lawsuit, the legal
department was obligated to hear her out, because sometimes an
outsider can inject a fresh perspective.
Ed Garnett, who came to Amgen as purchasing manager in 1986,
remembers the first operating-committee meeting he attended after
being promoted from director of logistics to head of human resources:

It was intimidating, being surrounded by all of these ex¬

tremely bright scientists. I just sat there and listened. After¬
ward, Kirby Alton, our senior vice president of development,
put a friendly arm around my shoulder and said quietly, “Ed,
we have a rule that you’re free to offer your opinion, even if
you don’t completely understand the subject. We need your
input.”
I said, “But Kirby, you were talking about drug develop¬
ment. I’m the new HR guy.”
“Yes, but I’m sure that you have an opinion, and we want
to hear the human resources department’s perspective on
whatever it is we’re discussing.”
I thought,“Whoa!” That was a big shock to me.

This may sound like democracy run amok. Certainly if your peo¬
ple are poor listeners and undisciplined thinkers, meetings could sink
beneath the weight of rambling thoughts and pointless debate.
For Amgen’s policy to work, you also need a manager who is
comfortable delegating to others—a skill that doesn’t come naturally
 How Amgen Kept Employees Committed 211

Everybody Gets to Play

You know you’re not distributing the workload evenly if any of the
following apply:

• You always feel overwhelmed, but your team does not

• Deadlines are missed.

• Your workers are so accustomed to your second-guessing

them midway through projects and taking over that they never
seem to finish anything on their own.

• Morale is low, in part because your subordinates feel that no

one is helping them grow, and they resent it. How are they
ever going to advance their careers?

to some people. If you’re a person who thrives on the challenge of

being in charge, chances are you have a talent for keeping multiple
plates spinning at once, boundless enthusiasm, and the tenacity to see
projects through from beginning to end. Maybe you’re so good, you
could practically do everything yourself.
Resist the urge. A boss who resists deputizing others for key tasks
may like to rationalize that he takes on so much work out of dedica¬
tion to the company. But his motivation is as much self-serving as it
is altruistic. If it’s really the case that “no one else can do the job as
well as I can,” as the nondelegator tells himself to justify hogging the
spotlight, then he hasn’t done a good job of hiring or of developing
people.
He’s letting down the organization in other ways. How will his
staff ever learn to handle important projects if he doesn’t give them
the opportunity? What would happen if he were to miss work for a
212 Science Lessons

stretch due to illness or injury? The reluctant delegator should start

by assigning less-skilled workers nonessential tasks. Then, as they
earn your trust, reward them with itiore important responsibilities. A
word of advice: don’t let all the grunt work fall on one person—do
your best to spread it around.
Fortunately, micromanagement was never an issue for me. Right,
Ed Garnett?

If Gordon had one fault as an executive, it was his tendency to

micromanage. I told him that as part of his 360-degree per¬
formance review. “You need to work on being more of a
macromanager. Be the CEO, and hold your people account¬
able for results.” But some of this was just part of his nature.
For instance, he had to get involved in going over the
blueprints for the new buildings! I’d walk into an office, and
he’d be down on the floor with the engineers, with the blue¬
prints spread out before them. I said,“Gordon, you’re just too
involved with the building program.”
He stopped me.‘Tm not disagreeing with you, but I really
like looking at the plans. Besides, if we make a mistake design¬
ing a building, it’s usually irreversible.” After that, if anyone
complained, “Gordon’s micromanaging building 32. What
should I do about it?” I’d just laugh and say, “It’s his hobby, he
loves it, so get over it.”

As should be evident, curbing the impulse to do too much may

be something many managers wrestle with throughout their careers.

Save the Competition for the Competition

Some bosses, perhaps equating calm with complacency, believe in

fanning the competitive flames between employees or units as a form
 How Amgen Kept Employees Committed 213

of motivation. Ford Motor Company was like that during my five

years there. In fact, it was one of the reasons I left.
I will always be grateful for the opportunity to work there; the
company was a world leader in finance and control, and I learned a
great deal about finance and other things that were extremely valuable
later. However, I also learned some practices to avoid. Ford divisions
were so cutthroat toward one another that before interdepartmental
meetings, each faction plotted how to get the rest of the organization to

20/20 Hindsight
Between 1990 and 2000, Amgen’s staff ballooned from fewer than one
thousand to more than seven thousand. At times our policies didn’t
keep pace. The team system had to be tweaked, because it had be¬
come unwieldy.
According to Kirby Alton, “In the eighties, working in teams and
building consensus gave us real power. The important thing was peo¬
ple learning and being willing to do something different. If somebody
dropped the ball, you had to pick it up. But as you get bigger and more
successful, it can become chaotic, especially when you have all these
new people who are used to many different management styles. It ul¬
timately led to a perception that Amgen was slow to make decisions.
That’s when we had to make some changes.”
Dennis Fenton adds, “Our teams used to be fairly small. All of a
sudden at meetings you’d have two guys from quality control, plus
two guys from manufacturing, plus five guys from somewhere else,
until before you knew it, there were sixty people in the room. A team
of that size is going to slow down everything. So we scaled back to
one representative from each department, plus a project manager,
and off you went. It was like a return to the way we used to do things.
214 Science Lessons

adopt its position. Compromise, even if it would benefit the company,

was seen as defeat, because the objective was to “win the meeting.”
At Ford, when the person representing the department came
back from the meeting, his manager asked, “Did you win?” Nothing
else mattered. To me, that kind of internecine battling is harmful to
an organization, undermining the whole concept of teamwork.
When there’s a great deal of internal turmoil, too much energy gets
spent licking the boss’s boots or stabbing colleagues in the back—
everyday occurrences at some companies. Fortunately, such behavior
was rare at Amgen, especially during its first seven years, when the
company consisted almost exclusively of researchers.
But when we started to build a sales force in 1987, in anticipation
of Epogen’s winning FDA approval, I sensed at times an almost con¬
genital rivalry between the sales and marketing side and the science
side. There are various permutations of this strained relationship in
other businesses, like in the entertainment industry, where the “cre¬
ative types” may be dismissive of the “suits” or “bean counters,” and
vice [Link] can chalk this up in part to an inherent difference in
personalities. Folks who gravitate toward sales tend to be extro¬
verted, whereas scientists are more introspective—not all of them, of
course, but many of them.

Respect and Trust One Another

When employees are set against one another, they may hoard ideas
and information instead of sharing them. I’ll never forget the re¬
searcher who came to Amgen from another leading biotech firm. In
one of his first projects for us, he was having difficulty purifying a
particular molecule. Nothing seemed to work. His team was getting
a little frustrated, because this was holding everybody up.
One day he came into a product-development meeting all smiles.
“I did it!” he announced. “I tried another method and purified the
molecule.”
 How Amgen Kept Employees Committed 215

“How did you do it?” someone asked.

“I’m not going to tell you.”
The room turned quiet. I was CFO then and happened to be at
the meeting. “Oh, boy” I thought, “he’s gone.” Sure enough, he left
Amgen only weeks later. More accurately, the company spit him out
as if ridding itself of a sour taste. Refusing to reveal to your fellow
workers how you did something was not the response of a team
player. Later I leaned of a subplot to this story: It seems that the scien¬
tist had tried to purify the molecule using what he considered to be
the superior methods of his previous employer, but nothing worked.
It was only after he tried one of Amgen’s purification techniques that
he succeeded. I guess he was a little mortified by this and didn’t want
to admit it. In any event, he never would have fit in with that attitude.
It’s essential for management to cultivate mutual trust and respect
among the departments of the company. The best way is to make
people aware of what their coworkers do and how each department
supports the others. When people understand other aspects of the
company, they make better, more well informed decisions. Abbott
Laboratories, for example, expected every scientist to spend one day
each month in the field with sales staff and customers.
Amgen encouraged our researchers to do the same thing. At first
some of our R&D personnel didn’t see the need for a sales and mar¬
keting department to make physicians aware of Epogen. That in¬
cluded Dan Vapnek, the longtime head of research. Vapnek once
famously said, somewhat kiddingly, “All we need is an 800 number,
and people will call in asking for EPO. It’s so good, it’ll sell itself.”
When Epogen was launched, we hadn’t yet found a salesperson
for Ohio. Sales were strong everywhere except that one state. Then
Paul Dawson, director of sales and marketing, expanded the sur¬
rounding territories temporarily so that Ohio was covered, and
Epogen use there soared almost overnight. Dawson couldn’t resist
showing the sales figures to Vapnek. “I’m sure you’ll agree this is pretty
persuasive evidence that sales and marketing is useful,” he said.
216 Science Lessons

“Well,” Dan conceded,“you’ve convinced me that salespeople are

useful. But I still don’t know what these marketing people are for.”
His face betrayed a faint smile. In f^ct,Vapnek developed a newfound
respect for the sales department, a feeling that only deepened after he
accompanied a salesperson on a call to a kidney specialist. He did this
on his own initiative; he didn’t tell me about it until afterward. Later,
he organized a two-day off-site meeting for outside speakers to teach
the entire research staff about finance. The scientists, much to their
surprise, found it fascinating to learn about balance sheets, profit-
and-loss statements, present values (compound interest in reverse),
and general ledgers.

Eliminate Unnecessary Hierarchies

If I hire you and pay your salary, we’re in a hierarchy. We can give it a
benign-sounding name, but it is what it is. Still, a hierarchical system
doesn’t have to be draconian.
I can imagine first-time visitors to Amgen driving around the
parking lot wondering whether perhaps they’d made a wrong turn
off the Ventura Freeway, because our headquarters had no assigned
parking spaces. As far as I was concerned, early arrivals deserved the
closest spots. If the last car pulling into the lot happened to be mine,
then I should have the longest walk to the door. This might seem to
be a trivial matter, but the policy set a tone of equality and discour¬
aged the prestige-based corporate hierarchies that can get in the way
of true teamwork. As I used to stress to new employees at orientation
meetings, “There’s no ‘us’ and ‘them’ here; we’re all employees.” One
of my favorite stories is the time that a staff member said something
about working at Amgen. A colleague quickly corrected him: “We
don’t work at Amgen,” he said, “we are Amgen.”
Businesses would do well to ditch any practice that creates artifi¬
cial boundaries between employees, even seemingly innocuous indi¬
vidual honors like “Employee of the Month.” You’re probably
 How Amgen Kept Employees Committed 217

wondering what possible harm could come from hanging one em¬
ployee s picture on the wall for thirty days. Here it is: if you have one
employee of the month, you have five or fifty or five hundred “Non¬
employees of the Month.” Only one person benefits, at the potential
expense of many bruised egos or grumblings of unfairness. So why
do it? Furthermore, recognition of quality work shouldn’t be rele¬
gated to a monthly contest limited to one employee. It should be dis¬
pensed as warranted.
The military uses stripes or bars on a uniform to denote rank. In
the corporate world, distinctions in seniority may be signified with a
key to the executive washroom or some other token. This always
struck me as needlessly divisive, and it reminds me of one of my first
job interviews, with Standard Oil Company of New Jersey—later to
become Exxon—in 1957. Fresh out of Purdue University, I wanted
to investigate potential future employers before entering the navy.
The interviewer who escorted me around Standard’s midtown
Manhattan headquarters told me a story about an executive there
who’d been promoted to the level just below the one where you
were rewarded with drapes for your office. His wife, an interior dec¬
orator, nevertheless picked out a set of drapes for him and had them
installed.
“Now, of course,” the interviewer said conspiratorially, “when he
came to work the next morning, the drapes were gone. And when he
asked what happened to the drapes, we said,‘What drapes?”’
I thought, “No way am I working for this chicken-s**t outfit; I
don’t care what they offer me.” Looking back, I can’t understand
why the interviewer felt compelled to share this story, and proudly,
too. Maybe he thought I aspired to have my own office drapes some¬
day, and telling me this would help recruit me. All it did was con¬
vince me that Standard Oil wasn’t the right company for me.
In the 1990s, Amgen bought an empty building that sat in the
middle of our campus. It had once belonged to one of the top word-
processor companies in the world back when terminals were still
218 Science Lessons

dumb and computer chips were all centralized. Raytheon, the appli¬
ance-electronics-aircraft giant, bought the company, but it went out
ofbusiness. ^
When I toured the building, it consisted mostly of drab cubicles.
But then we came upon palatial executive offices with handsome
wood paneling, along with a private kitchen and dining room—all
for a handful of officers. “This is too nice,” I said. “We’ll have to bull¬
doze it.”We [Link] could have predicted the word-processor com¬
pany’s demise just from the glaring disparity between the extravagant
executive suite and the area where everybody else worked.
Many businesspeople get caught up in trappings of success that
ultimately are meaningless. “I’ve worked hard for a lot of years,” they
think, “so I deserve this.” I’m not impressed with that stuff. Never
have been. Maybe it’s because I grew up in very modest surround¬
ings as a child. Whatever the reason, I never cared about having a big,
fancy office, or the other accouterments that seem to matter so
much to many corporate leaders. Maybe a better way to say it is that
other things held more importance for me. I like promotions as
much as the next guy, but I knew enough not to confuse the reality
with the symbol.
Executives at other companies didn’t always understand or, for
that matter, approve of Amgen’s relative frugality and informality. Be¬
ginning in 1992,1 spent eight years on the board of the Pharmaceu¬
tical Research and Manufacturers of America, including one year as
chairman. After I became the first representative of the biotech in¬
dustry elected to the PhRMA board, one of the members, a fellow
CEO, cornered me in the hallway and said, almost angrily, “Well! I
certainly hope that now-'you’ll get yourself an airplane.”
Far more gratifying was having an Amgen employee come up to
me in the parking garage one time. The company had become too
large for me to know everyone’s name. He introduced himself and
said, “You know one of the things I really like about Amgen? The
 How Amgen Kept Employees Committed 219

fact that you don’t have a reserved parking space.” I’d wondered if
anybody even noticed.

HELP EMPLOYEES EXCEL

THROUGH TRAINING

Most good workers—and that describes the majority of profession¬

als—genuinely want to do well at their jobs and see their company
prosper. In Amgen’s view, the fundamental duty of management was
to furnish employees with the equipment and manpower they
needed and to clear away any administrative underbrush that might
trip them up. If this were a football analogy, your staff would be the
ball carrier, aiming for daylight, while you’d be an offensive lineman,
blocking would-be tacklers.
In other words, a smart manager recognizes that he works for his
employees more than they work for him. The relationship is a cycle,
not a one-way street. And that should extend all the way up the cor¬
porate ladder.
High-quality equipment and facilities and the freedom to make
decisions make up two sides of the employee-growth triangle (see fig¬
ure 10-2).Without continued training and education, your best people
will seek opportunities at firms that will invest in their future. Even
more troubling, your less-talented personnel will stay—and stagnate.
Amgen did a better job of cultivating its talent after Kevin Sharer
joined us as president and chief operating officer in 1992. He’d spent
five years working at General Electric before becoming president of
the business markets division of MCI. We more or less adapted GE’s
formal method of reviewing the organization’s talent pool and outlin¬
ing career paths for our best people. Kevin and I used to sit down with
the managers and go over their rosters one by one, hour after hour.
Among the questions we hoped to answer were, Who were the
most promotable candidates, and in which areas of the company?
What training would they require? Which jobs would provide the
220 Science Lessons

FIGURE 10-2

The employee-growth triangle

v

Continued training
and educational opportunities

development opportunities they needed? These points were dis¬

cussed individually with staffers as part of their performance reviews;
most of them appreciated knowing exactly where they stood in the
organization and what their prospects might be.
Lets say that a manager is considering a top sales rep named
Frank as a future sales manager. During the performance review, the
boss says, “Frank, you already have many of the skills that a sales man¬
ager needs. But there are two areas where you could benefit from
more experience. We re going to give you a special assignment that
will help you develop those skills.” You’re not promising a promo¬
tion (never promise something that you may not be able to deliver);
rather, you’re promising that the person will progress professionally
so that he will be promotable, either at your company or elsewhere.
People who know they’re being trained for greater things won’t be
going anywhere anytime soon.
These discussions Kevin and I had with the managers often took a
full day for each department in the company, but it was time well spent.
Not only did it benefit our employees, but also it gave Sharer and me an
overview of the people who might become Amgen’s future stars.
 How Amgen Kept Employees Committed 221

Small businesses, which employ more than half the U.S. workforce,
are often derelict in career development. Among other reasons, they
often lack the necessary infrastructure and human resource personnel.
Or perhaps the owners worry that their people will hone their skills
only to jump to larger companies. That’s a possibility, of course, but it’s
a shortsighted way to run a business, especially when technological ad¬
vances can shake up entire industries seemingly overnight.
Prime Jet is one of those smaller businesses. We started with only
three employees, and as of this writing, we’re up to thirty-nine. Our
retention rate rivals that of Amgen. But we recently lost two pilots, in
part because of my belief and my general manager’s belief that an
employer has an obligation to advance workers’ careers. As I see it, it’s
fundamental to the employer-employee contract.
Let me explain. Private copilots earn about $40,000 a year, com¬
pared with $100,000 for captains. So, obviously, there’s a lot of incen¬
tive for copilots to obtain a captain’s position. During interviews, we
promise that if they come to work for Prime Jet, after two years we
will get them certified by the Federal Aviation Administration as cap¬
tains. Not only that, but if we don’t have openings for captains and if
they want to find positions elsewhere, we’ll call around the industry to
help them, even though it will lead to their departure. That’s almost
unheard of; in fact, in an effort to keep pilots, companies often surrep¬
titiously try to interfere with their copilots’ getting certified as captains.
Not only is Prime Jet’s policy the right thing to do, but also it’s
good business. A report from the U.S. Department of Education
found that increasing education and training improved productivity
by 11 percent to 20 percent. Other studies associate education and
training with higher profit margins and stock performance.
There’s another benefit, too: when an organization develops a rep¬
utation as a place where people can build careers, the word spreads.
You should look at training as an investment that strengthens both re¬
cruitment and retention. IBM, which spends approximately $1 billion
annually on employee education, recognized this early in its history.
222 Science Lessons

In 1915, only four years after incorporating, the organization estab¬

lished a separate education department for instructing workers on the
use of its products (such as the mechanical key punch, the vertical
sorter, and the tabulator). According to an internal study, 4 in 5 IBM
employees who are afforded learning and developmental opportuni¬
ties stay with the company for at least three years.
Training used to be a headache logistically: you had to hire quali¬
fied instructors, and seminars often took place off-site, taking em¬
ployees away from their work and forcing companies to incur travel
costs. Increasingly, employers are turning to e-learning, either over
the Internet or through a private computer network (intranet). One
great advantage of e-learning is that it doesn't have to disrupt the
normal workday; your people can access programs at their own pace
whenever their schedules allow. Another option popular with em¬
ployees is tuition reimbursement for college courses, although only
about one-third of U.S. businesses offer it.
Amgen paid for one university course per semester for all em¬
ployees, who were entitled to take one class per year unrelated to
their jobs. As is typical in Southern California, English was a second
language for many of our lowest-paid workers; all of them were eli¬
gible for free after-hours instruction in English.

HELP YOUR EMPLOYEES BALANCE THE

DEMANDS OF WORK AND FAMILY

George Rathmann and I were the “old men” of Amgen. Except for us
and a handful of others, it was a youthful organization, with most of the
researchers in their late twenties or early thirties.“Everyone in the com¬
pany, it seemed, had babies within a year,” Burt Ensley says, laughing.
“We used to joke that there must have been something in the water.”
When I joined Amgen in 1982, many of the fifty or so employees
were new to Thousand Oaks. As often happens at start-up compa¬
nies—where people work long hours in the quest of a common
 How Amgen Kept Employees Committed 223

goal, not to mention survival—’’there was a lot of camaraderie,” re¬

flects Kirby Alton. “We all got to be really good friends.”
George, himsell a transplant, promoted the family atmosphere.
Once a month, everyone fell out into parking lot 2 (now a volleyball
court) for what became known as “fermentation seminars.” It’s more
fun than it sounds like. Fermentation, employed to mass-produce
bacteria for cloning, is also essential to processing beer and cheese.
Pizza-and-beer parties is what they were; and spouses and children
also attended.
Genentech, located in Sacramento, had its own version of out¬
door gatherings known as ho-hos. I’m not sure whether the casual at¬
mosphere at both companies was an outgrowth of the laid-back
California lifestyle or of biotechnology. It was probably a bit of both.
We added events such as chili cook-offs, Easter egg hunts, and Hal¬
loween haunted houses, always with families included.
In preparation for our first national sales conference, in 1989,
Paul Dawson, head of sales and marketing, came into my office. “I’m
going to ask you something very unusual,” he said, “and if you agree
to it, I promise I’ll never ask you for this again. We’d like to invite the
spouses and kids to the sales meeting.”
No company did this, but as Paul explained, the sales staff was still
new to Amgen. Most of the reps were out in the field all day, driving
around and talking to doctors about Epogen. Their offices consisted
of the fax machines in their homes, and that made it hard for them to
truly feel a part of the company. Inviting the families, he said, would
be a welcome gesture. How could I say no?
It worked out better than anyone could have imagined. My wife
took it upon herself to plan day and evening events for everyone.
During the afternoon, buses took all the children to Disneyland,
while the spouses participated in their own activities. Everyone had a
wonderful time. Pharma reps change jobs pretty often, mainly be¬
cause everyone wants to be in on the launch of a great new product
instead of promoting the same thing year after year. One woman was
224 Science Lessons

overheard saying to her sales rep husband, “You’re not thinking about
changing jobs again soon and leaving Amgen, are you?”
She may have been half-kidding. But the point is that bosses
rarely consider the impact of their employees’ spouses and children
on the decision of whether to stay with a company. People often
make these choices as a family, just as I consulted Adele before ac¬
cepting George Rathmann’s offer to come to work for Amgen. Not
to be overlooked, the fact that my wife organized everything for the
families, instead of some impersonal event-coordinating company,
meant a lot to our people.
By the way, Paul didn’t have to ask a second time about making
our national sales meetings family affairs. It became an annual tradi¬
tion—and, years later, also for our international management meet¬
ings. (Some of the kids became pen pals with children from other
countries.) We wanted our employees’ spouses to know that Amgen
appreciated all that they did for the company; this was our way of
saying thank you.

Have Fun—Most of the Time

I’m sure that this policy was a factor in Amgen’s high retention rate.
For instance, the entire time that I was CEO, not a single country
manager left Amgen to work for a competitor, even though they
were actively pursued. It wasn’t cheap to fly hundreds of family
members to California and arrange a weekend of fun in the sun. But
if you calculated the huge cost of having to replace sales staff, I am
sure that it paid for itself and then some.

The Most Coveted Commodity Mime

For a study of Americans’ attitudes about work, researchers at the

John J. Heldrich Center for Workforce Development at Rutgers
University and the Center for Survey Research and Analysis at the
 How Amgen Kept Employees Committed 225

University of Connecticut surveyed one thousand employed adults.

The most pressing concern, hands down, was maintaining a healthy
equilibrium between career and family.
When I graduated from Harvard Business School in 1962, women
made up only one-third of the U.S. workforce. Today it’s about half.
With increasing numbers of single parents, and with married couples
sharing more family responsibilities, the worlds of work and home
have come to overlap. It has become harder for people to compart¬
mentalize their private and professional lives.
Amgen took a holistic view of its employees. Our obligation to
them didn’t end when they left the office; it was up to us to help
people better juggle the demands at home and at [Link] that in
mind, the company provided conveniences right there on campus,
such as a pickup-and-delivery dry-cleaning service, a film-develop¬
ment service, an automated teller machine, a card and gift shop, and a
state-of-the-art fitness center. Employees could order take-home
dinners from the cafeteria for pickup after five o’clock. We always
tried to add amenities that would help ease stress and reduce the
amount of time that people had to spend running errands.
“We used to have a saying: ‘The worst day at Amgen is better than
the best days elsewhere,”’ recalls HR director Ed Garnett. “At one
point, we came up with the crazy idea of having a car wash in a park¬
ing [Link] operating committee put its foot down and said ab¬
solutely not. Gordon, though, was intrigued. ‘You mean I can drop off
my car, and it’ll be ready when I come out of work?’ I said,‘Yeah, for
$4.50.’ He asked,‘How soon can it be finishedP’The car wash was such
a big hit that if you didn’t get your car there by nine o’clock in the
morning, sorry, you’d have to come back and try again the next day.”
The centerpiece of our commitment to enhancing employees’
quality7 of life was our child-care center, as far as I know still the
largest of any corporation in the country. My wife deserves the credit.
What happened was that people were beginning to ask for child care,
a service provided by only one in eight employers. I dismissed the
226 Science Lessons

idea at first, thinking, “We all raised our kids on our own; why can’t
they do it?” One night I made the mistake of articulating that opin¬
ion to Adele. '•
“Gordon, get real!” she exclaimed, rolling her eyes. “The world
has changed.” Discussing it with her made me see things differently,
and when I broached the possibility with Ed Garnett, he became ex¬
cited. “The employees would love it if we added child care,” he said.
“It would help us with recruiting and retention, too.”
So we looked into it. Adele toured an excellent child-care center
at a government agency in Washington, D.C., reinforcing her convic¬
tion that we should start our own. “Camp Amgen,” managed by a
Massachusetts company called Bright Horizons Family Solutions,
opened in 1992. Eight years later we doubled its size to thirty-two
classrooms and sixteen playgrounds on 44,856 square feet, large
enough for 432 children.
Employees paid 10 percent to 15 percent below the market rate,
and Amgen provided roughly an equal amount. Having a first-class
child-care center on the premises played a part in reducing turnover.
According to a study conducted by Boston’s Simmons College, one-
third of workers have thought about changing jobs because of child¬
care issues. Of those, 85 percent said they would consider staying if
on-site child care was made available.
Another benefit is a sizable drop in absenteeism, which has been
estimated to cost employers between $650 and $1,000 per employee
annually. Nearly half of all parents miss at least two days a year be¬
cause of lapses in child care, such as a babysitter’s illness. Child-re¬
lated problems are also responsible for parents arriving late or leaving
early fifteen times per year, on average. Family-friendly benefits may
entitle companies to federal and state tax breaks as well, so investing
in a child-care facility can actually save a business money.
There’s been a lot of clamor lately for alternative work schedules:
telecommuting, flextime, job sharing, and the [Link] may be well
suited for organizations that don’t run on teamwork; nearly half of all
 How Amgen Kept Employees Committed 227

telecommuters claim that they’re more productive working from

home. But none of these practices would have succeeded at Amgen,
which relied heavily on teamwork and peer pressure. We explained
this to our people, and, although we didn’t offer these family-friendly
practices, Amgen consistently made the lists of best companies to
work for.

BE GENEROUS WITH COMPENSATION

A company’s compensation system influences retention rates like the

moon affects the tides, although it does not necessarily follow that
the firm that pays the highest salaries has the lowest turnover. Em¬
ployees are concerned about fairness, too, in comparison both to
their colleagues and to workers in the rest of the industry.
The pay policy you choose can motivate your people to work to¬
gether, or it can sow dissension. Individual incentives are appropriate
for efforts that don’t depend on teamwork, such as sales. On the
other hand, if your business relies on collaboration, then profit-shar¬
ing plans and stock options can help encourage workers to view one
another as business partners.
Amgen combined both types of compensation arrangements. I
offered our senior management group two choices: all employees
could receive one-size-fits-all cost-of-living increases and promotion
raises, or we could award higher salaries to the highest performers and
lower salaries to the poorest performers, based on the company’s per¬
formance review system. The latter option is time-consuming and
sometimes controversial, but it helps organizations keep their best
[Link] managers chose it by an overwhelming margin.
We used a matrix system, as shown in figure 10-3. (The numbers in
this example are for illustration and are not the actual numbers.) One
dimension was the person’s salary position within the salary range for
her pay grade, by quartile. The higher she rose within the range, the
smaller her raise would be. The other dimension was performance.
228 Science Lessons

FIGURE 10-3

Amgen salary grade performance matrix

v
Percent increase in salary

4 4 percent 5 percent 6 percent 7 percent

3 3 percent 5 percent 6 percent 6 percent

Performance

2 2 percent 3 percent 4 percent 4 percent

1 0 percent 2 percent 3 percent 3 percent

4th 3rd 2nd 1st

Quartile

The best performers near the bottom of the salary range received the
largest increases, and the raises were smallest for the poorest per¬
formers near the top of the range.
The bigger attraction, though, was stock options, which granted
employees the right to buy a certain number of shares at a specified
price. Options, offered by one in seven public businesses, are mutu¬
ally beneficial because they allow cash-hungry start-ups to entice
well-qualified people at lower salaries in exchange for the chance to
make a lot of money if the company is a success. The number of op¬
tions awarded to each person varied, even among those in the same
pay grade. Our criteria included (1) past contributions to the com¬
pany, (2) current value to the company, (3) future potential, and (4)
attractiveness to competitors.
In a policy that was unorthodox for its time outside the biotech
world, every Amgen employee—not only senior executives—received
stock [Link] Ledder, respected longtime CEO and chairman of
 How Amgen Kept Employees Committed 229

Abbott Laboratories, remembers thinking that our decision to in¬

clude the rank and file was a bad idea.“I didn’t see how giving stock
options to the janitorial staff would help Amgen,” says Ted, who
joined our board of directors in 1988. “But it worked, and it’s been
done more and more since then.”
Companywide visions of a sudden windfall, though it may never
come to pass, can be a powerful agent for getting people to pull to¬
gether. And your employees may be more willing to stay with the
organization while keeping their fingers crossed that the stock value
will appreciate. There’s a good reason that options are referred to as
“golden handcuffs.”
If staffers aren’t educated about stock options, however, their ex¬
pectations may be unrealistic, and the plan could backfire. Amgen
probably could have done a better job of informing our employees
before the June 1983 initial public offering. People can get so swept
up in the euphoria surrounding an IPO that they start thinking they
will be instant millionaires. Not quite. First, the options can’t be ex¬
ercised for some time, perhaps years, and there’s no telling what
might happen to the stock between now and then. Amgen’s IPO
came in at $18 per share. By the third quarter of 1985, the price had
plummeted to an all-time low of $3.75.
It was disappointing. But as we reassured everyone, stocks go up
and stocks go down; in the long run, the best companies have the
best stock prices. A business’s stock price doesn’t necessarily reflect its
financial health. Amgen was doing fine, meeting its research targets
and still flush with cash from the public offering.
“The market is cyclical,” we told our people, “and the prices will
come back. Don’t worry.”
Two years later, the price had climbed back, but only to $6 per
share. As predicted, however, when we got Epogen into clinical trials,
the stock stirred. By 1987, still two years before the drug received
FDA approval, Amgen’s price per share stood at about $40.
230 Science Lessons

BE GENEROUS WITH PRAISE,

TIME, AND ATTENTION
V
George Rathmann and I used to joke that our researchers were so
consumed by their work that all we needed to do was to stay the hell
out of their way. As I pointed out earlier, this entailed doing our best
to eliminate needless red tape that might impede their progress. Ulti¬
mately, no amount of stirring speeches will lift employee morale like
building a winning organization. If you don’t provide an efficient,
energizing work environment, your words will ring hollow.
Still, even the most self-directed, independent person seeks
recognition for a job well done. It’s human nature. Even your suppli¬
ers appreciate recognition (see “Reach Out to Your Suppliers,Too”).
It doesn’t cost anything to hand out praise, and yet many managers
insist on trying to prod subordinates with threats—loss of a privilege,
probation, perhaps termination—if a goal is not met. This may
“work” in the short term, but it does not inspire people to do their
best over the long haul.
Feeling underappreciated in the workplace gnaws at people. They
don’t resign in a huff; instead, they sulk or simmer, which can be more

Reach Out to Your Suppliers, Too

They may not be on the payroll, but your suppliers contribute to the
overall success of your company. When suppliers do a good job, tell
them. Word circulates among subcontractors about which companies
are a pleasure to do business with and which ones are a source of mi¬
graines. Establishing the former reputation may expand your pool of
suppliers; it may even give you an edge during contract negotiations.
 How Amgen Kept Employees Committed 231

destructive to your organization. They’re likely to grumble, “Why

should I bother working hard? The boss doesn‘t notice anyway.”
In the words ofWarren Bennis, a distinguished business professor
and prolific author, “Good leaders make people feel that they’re at
the very heart of things, not at the periphery. Everyone feels that he
or she makes a difference to the success of the organization. When
that happens, people feel centered, and that gives their work mean¬
ing.” All it takes is for the manager to touch base regularly with
workers and offer thoughtful feedback. Even a supervisor’s construc¬
tive criticism is preferable to toiling in a vacuum. But try to look
harder for opportunities to compliment—a 3-to-l ratio is about
right. If you can’t find any, that’s a different kind of problem.
I don’t think I’ve ever seen a corporate leader more revered than
George Rathmann, particularly among the Amgen scientists. Science
can be lonely work, so George always made himself available, whether
to offer advice or simply words of encouragement. One time, he and
I were meeting with several businessmen from another company
when suddenly the door to Rathmann’s office flew open and two
researchers burst into the room. One of them said, “We did it,
George! We did it!” I can’t recall exactly what it was they had done,
but suffice it to say they’d solved a problem in the laboratory.
The visitors looked up, scowling, as if to say, “What the hell is
going on? This is the CEO’s office. Employees can’t just come barg¬
ing in here like that.”
But Rathmann politely excused himself. “I really need to take a
few minutes to hear this,” he said, motioning for the two scientists to
follow him into the hallway. It was obviously important to them that
they have his ear right then, and George understood because he was
one of them. He’d been enamored of chemistry since the age of
twelve, when he traveled from Milwaukee to Indianapolis to visit an
older brother-in-law who worked in medical research at Eli Lilly
and Company. “Although I just got a superficial look at the labs
232 Science Lessons

there,” Rathmann recalls, “I was absolutely dazzled.” He decided al¬

most on the spot to pursue a career in medical research.
George laughs when reminded of the interrupted meeting from
nearly twenty-five years ago. “Scientists like challenges, but they
need support, too. It’s up to you, the supervisor, or boss, or whatever
you call yourself, to fall in love with the projects that your people are
working on. Otherwise, how are you going to convince them that
they can conquer the world?”

LEARN FROMYOUR EMPLOYEES

I’m not suggesting that Amgen scientists were without complaint,

but most were pretty content. Especially if they’d come from Big
Pharma, where senior management usually treated research as a nec¬
essary evil. They didn’t understand it—it wasn’t sales and market¬
ing—and didn’t particularly want to. I’m sure that the scientists
sensed the executives’ lack of interest.
At Amgen, every Friday a different department of research held
an informal poster [Link] scientists stood next to poster displays
and explained what they were working on to anyone who was inter¬
ested. Attendance was more or less mandatory for upper research
management. I went to most of them, which isn’t something that a
chief executive usually does. I didn’t always understand the science,
but it was vital for me to mill about and eat pizza with the others be¬
cause it was clearly important to the researchers. It also meant a lot to
them that those responsible for the project, and not only their bosses,
got to talk about it.
I leaned a great deal from these informal sessions. Sometimes
something that I read on one of the posters helped clarify a point
under discussion at a meeting a month later. It took no more than an
hour and a half per week. In twelve years as CEO, I often spent
ninety minutes doing other things that weren’t nearly as productive.
 How Amgen Kept Employees Committed 233

When supervisors fail to get out from behind their desks and in¬
teract with people from different levels of the company, the ramifica¬
tions are as detrimental to them as to the rank and file. Not only do
the administrators become isolated from their staff, but also their de¬
cisions suffer from the limited input they get.
One danger, particularly in upper management, is that the people
reporting to you tailor the news to what they think you want to
hear. Most folks want to make the boss happy, for obvious reasons, so
they may gloss over troublesome details and embellish aspects that
reflect favorably on the company (and maybe on themselves). The
result is a distorted picture of the organization.
Once, at Amgen, we’d recently hired an influx of new sales and
marketing people from Big Pharma, where this sort of thing went on
all the time. Word got back to me that before our monthly marketing
meetings, they were getting together to secretly “Binderize” their
presentations. I suppose I should have been flattered; I was not. It was
a rerun of what I’d experienced at Ford Motor Company in the
1960s. I informed them that this was not the Amgen way. “We’ll have
one meeting, and you’ll tell me what I need to hear, not what you
think I want to hear.”
To help myself stay tuned in, I used to rely on the chief financial
officer as my alternative channel. Typically, the CFO is involved with
every part of the company and can provide an all-encompassing per¬
spective. A talented head of human resources can be another excellent
information source. People aren’t always comfortable approaching the
CEO, especially if they’re several levels down in the corporate hierar¬
chy. In Amgen’s case, they used to tell Ed Garnett things that they
wanted me to know. That was fine.
Ed, being the good guy he is, would check out the veracity of the
information before passing it on to me. He might say,“I’ve heard this
complaint from a number of people; it may be worth your looking
into.” Or he’d let me know that someone had come to him with a
234 Science Lessons

grievance, but it wasn’t anything that required my attention. He was

an extremely valuable third channel.
Sometimes employees are relpctant to report the truth to their
supervisors because they don’t want to get their heads bitten off. If
you’re consistently in the dark about news that’s circulating on the
company grapevine, ask yourself, Is it because you’re frightening
the messengers? A yes answer requires a change in behavior right
away. You can’t manage an organization based on half-truths. Says
Kevin Sharer, “I always felt safe telling Gordon what I thought,
even if I didn’t agree with him.” As far as I’m concerned, there’s no
other way.
Employees deserve to hear the truth from management, too—
within reason, of course. In 1991, just when it seemed that things
could not be going better for Amgen, a federal arbitrator in our
never-ending battle with Johnson & Johnson ordered us to pay $149
million in damages. Johnson & Johnson had claimed that Amgen was
delinquent in helping it to get its human erythropoietin on the mar¬
ket, allegedly costing the company millions in sales. We’d countered
that the fifth-largest pharmaceutical outfit in the world shouldn’t
have needed our assistance in the first place. It wasn’t our fault that
Johnson & Johnson conducted business at such a leisurely pace, at
least according to Amgen standards.
The day in September that the court handed down its decision
was devastating. One hundred forty-nine million dollars was more
money than we had! This could have conceivably spelled the end of
our company. Many employees’ first thought was that management
must have mishandled the situation. How could Johnson & Johnson
possibly deserve that much money from us? Morale was lower than
hell, and I thought,“Oh, boy, how do I deal with this problem?”
The answer was to face our people and be forthright. I immedi¬
ately announced that at one o’clock I would take the stage in our au¬
ditorium to explain what had happened and to field questions until
 How Amgen Kept Employees Committed 235

Other Ways to Build Morale

• Help your staffers draw the connection between their duties
and the company’s success. Twenty-five thousand men and
women at seventeen large companies participated in a Towers
Perrin poll about intracompany communication. The survey
takers found that, overall, the employees were keenly inter¬
ested in understanding their roles in their organizations and in
learning how they could contribute more to helping the busi¬
ness grow.

• Get into the habit of saying "we” instead of "I.” There’s only so
much credit to go around. If the big boss monopolizes the
limelight, it is demoralizing for the other folks who work hard,
too. As per the Chinese proverb, “The best leader, the people
do not notice. When the best leader’s work is done, the peo¬
ple say, ‘We did it ourselves.'”
With that in mind, share credit freely. In 1994 Amgen be¬
came the first biotech company to win the National Medal of
Technology, the highest honor awarded by the president of
the United States to American innovators. Usually it was given
to individual scientists and engineers; only two corporations,
DuPont and Bell Labs, received the award prior to Amgen. Two
of our staff members, winners of a companywide lottery, ac¬
companied me to a private Oval Office ceremony with Presi¬
dent Bill Clinton and to a public-awards ceremony conducted
by Vice President At Gore. I accepted the award “on behalf of
Amgen’s thirty-two hundred outstanding men and women.” It
was a proud day for every member of the organization, which
had its own celebration back in Thousand Oaks.
236 Science Lessons

the last employee left. No vice presidents, no attorneys, just me, an¬
swering to the best of my abilities.
The biggest concern, not surprisingly, was that we might not be
able to fund research and development. Bankruptcy, I told them can¬
didly, was not out of the question. A few folks worried aloud whether
or not Amgen knew how to wage legal warfare, and that we’d lose
again on appeal. Some took it personally.
Their attitude was, “We’re a great company! How dare the judge
criticize us!” Still others were just plain ticked off and needed to
vent.
Well, by three o’clock, the place was half empty, and by six, the
last stragglers were heading out the door. Dealing honestly and
openly with the crisis had healed the fractured morale in just half a
day—a really painful half a day! Fortunately, the following year, be¬
fore Amgen had to pay any of the $149 million to Johnson & John¬
son, we were awarded $90 million in damages from them. We also
had a lot more money by then.

DON’T BE AFRAID TO BE HUMAN

Counseling subordinates one-on-one is a part of leadership. Ed Gar¬

nett remembers something I said to him when he was suffering a cri¬
sis of confidence a year into his tenure as head of human resources:

HR was a lot more difficult than I thought it would be. We

were growing by leaps and bounds and hiring like crazy. One
day I went into Gordon’s office, and without closing the door
behind me blurted out, “Gordon, I think you need to recon¬
sider me as your HR man. I’m not sure that I’m the right guy
for the job anymore.”
Gordon stood up, walked over and closed the door, and
leaned against it. “Ed,” he said, “every morning when I’m
 How Amgen Kept Employees Committed 237

shaving, I look in the mirror and think, ‘They got the wrong
guy for the CEO job.”’
I was stunned, because he always seemed so confident.
“You’re serious?”
“Every morning.” Then he went on to reassure me that I
had his complete confidence. It was probably the biggest
morale booster I could have received, and I went back to
work feeling much better.

To be honest, I don’t remember that exchange. It goes to show

the weight that a supervisor’s words can carry, so be sure to speak
thoughtfully, never glibly.

Let People Know Where They Stand

George Bernard Shaw, the always quotable Irish playwright, once

said famously, “The greatest problem with communication is the as¬
sumption that it has taken place.” In many organizations, it’s not just
the quality of communication that’s lacking, but the quantity.
Communication between a supervisor and the supervised should
be a continuous process, culminating in the annual review. At some
firms, employees may go the whole year without meaningful feed¬
back about their performance or concerns they might have until the
formal sit-down with management. That’s simply inadequate. For a
worker, few things stir up anxiety more than not knowing where he
stands. And if he’s fallen into some poor work habits, a boss’s failure to
address them promptly is only going to make them harder to correct.
There are various forums for conveying information to your
staff. Surveys find that employees generally prefer print material and
e-mails to voice mail and Web casts, probably because they can read
the former at their leisure. Such methods should be regarded as com¬
plementary to, not a substitute for, checking in personally with your
238 Science Lessons

[Link] higher your perch on the corporate ladder, the more im¬
perative it is to regularly touch back down and get a feel for what’s
going on. A lesson I learned in Harvard Business School—“You have
to roll up your sleeves and get dirty”—still holds true.

INCLUDEVALUES IN
THE PERFORMANCE REVIEW

If you’ve been engaged in an ongoing dialogue with your employ¬

ees, their year-end performance evaluations should be anticlimactic.
No one should be surprised by the results.
Often, though, formal employee evaluations are thought of in the
same terms as a grade school report card. Instead of your mother or
father yelling at you, your boss does the honors. I view the evaluation
more as an opportunity for education. Only after praising the em¬
ployee’s strengths and highlighting the ways in which she’s helped
the company do I discuss areas that can stand [Link] to¬
gether we brainstorm ways that management can do a better job of
helping her do a better job. Perhaps we need to provide additional
training, juggle her responsibilities, or maybe transfer her to a differ¬
ent department. The review isn’t the end product, it’s the first step
toward a solution and a milestone in the person’s development.
Over the years, we tried various methods of employee evaluation
and were never completely satisfied with any of them. There are,
however, two suggestions I recommend highly. The first is to incor¬
porate your company’s values prominently in the review. As I’ve
mentioned, when people get fired it’s usually because they’re out of
step with the organization’s culture. Amgen employees received con¬
structive feedback if there was a value that they needed to work on.
Rarely was it more than one or two. This helped make the perfor¬
mance review more concrete. Do you work well in a team or not?
Maybe “Joe’s” immediate supervisor feels that he has a bad habit of
steamrolling colleagues during meetings, even though the company
 How Amgen Kept Employees Committed 239

credo plainly states that people must “collaborate, communicate, and

build consensus.”What can be done about that?
Instead of Joe’s boss stating bluntly, “You talk too much at meet¬
ings,” she might frame it this way: “Joe, you could be a more effec¬
tive participant in meetings if you made fewer points but made each
one more strongly. Because by bringing up so many issues, you’re di¬
luting the impact of everything you say. Now, the next two or three
meetings that you go to, I want you to listen very carefully and ob¬
serve somebody who offers only one or two points, and how closely
people pay attention to that person. Then take note of someone
who’s always talking, and you’ll see it for yourself; you don’t have to
take my word for it.”
How well a manager’s comments are received often lies in the
presentation. Most folks will listen to advice like that—delivered
firmly but clearly intended to help, without subjecting them to
ridicule or embarrassment.
And don’t forget to have your own performance reviewed (see
“A Review for You,Too”).

DEAL WITH HABITUAL UNDERACHIEVERS

One thing that you’ll never find anywhere in the business world is a
boss who fires an employee and then says, “I’m really glad I didn’t
let him go sooner!” We always wait too long. Getting rid of a chron¬
ically poor performer is the only fair thing to do for your hard¬
working employees. Otherwise, not only are they forced to pick up
the slack, but also the deadwood is blocking someone’s opportunity
for advancement.
What’s more, a single employee’s feeble effort can subvert pro¬
ductivity across the [Link] rest of the staff sees what’s going on. If
management turns a blind eye, maybe it’s not paying attention to the
people who excel at their [Link], why bother trying? Failure
to take action may ultimately drive model employees, those with
240 Science Lessons

A Review fpr You, Too

A manager who always deflects criticism from himself does not
command respect. Without feedback on your performance, how are
you going to grow in the job? Senior executives can easily lose per¬
spective on themselves—it’s not as though lower-level employees
often collar them in the hallway and tactfully suggest ways for them
to become better bosses.
Amgen offers all supervisory-level managers a voluntary 360-
degree review, arranged through the human resources department. As
the name implies, colleagues on all sides—subordinates, peers, superi¬
ors—fill out an anonymous performance survey. Another way is for
HR personnel to interview people throughout the company.
Perhaps I have an unusually thick skin, but I volunteered for a 360-
degree review every year. What’s especially valuable is the three tiers
of input. Some people deal effectively with their peers but come on
too strong with subordinates, or vice versa. In my experience, those
doing the critiquing took it seriously; no one was looking to knife the
boss in the back.
The job of conducting my review fell to Ed Garnett. When I asked
to hear the first critique, he didn’t know how to react. “I said to Gor¬
don, ‘This is very scary for me,”’ he recalls. ‘‘Most CEOs don’t ask for
performance reviews. They measure their performance by their salary
increases and the proxy statement. Gordon said, ‘I understand that,
but I really want some honest feedback.’ So I interviewed people
around him, the board members, people below him. Then I summa¬
rized it in my words—for instance, ‘Three people at the senior-man¬
agement level thought this . . . ’ so that none of the sources was
revealed.”
 How Amgen Kept Employees Committed 241

pride and ambition, from the company in search of an organization

that values hard work and diligence.
Studies suggest that in the average company about one in seven
workers can be considered complacent. The signs are usually trans¬
parent: a noticeable change in attitude and behavior, an uptick in ab¬
sences and lateness, and so on. The solution may be less obvious.
Termination is always an option. But unless the person is hopelessly
inept or his infractions egregious, it’s best to consider the less drastic
route of a transfer within the company.
When a staff member wasn’t meeting expectations, I had one
question for him. “Are you having fun?” If you hate your job, I can
guarantee that you’re not performing well. I’d go on to explain that
just because you don’t like your current position, it doesn’t mean that
you have to leave Amgen. There are many jobs in various parts of the
company—and even in other locations. “So how about going to
human resources and seeing about a possible transfer?”
We weren’t looking to reshuffle a poor employee like a bad card
so that he merely became some other supervisor’s problem. With file
cabinets full of resumes, Amgen had no need to do that. Still, trans¬
ferring a worker is less expensive than recruiting someone new; it
may be worthwhile to see whether a change in duties—or scenery—
can rejuvenate him.
I didn’t keep score, but over the years I’d say our success rate fol¬
lowing transfers was about fifty-fifty. It usually took no more than six
months to determine whether the person was taking advantage of
his second chance. If he couldn’t be fired up, then unfortunately, fir¬
ing was the inevitable next step.

Handling Resignations

A staff member’s resignation might come as a disappointment, but it

shouldn’t come as a [Link] it does, observes Ed Garnett,“that’s
242 Science Lessons

a sign of poor management, because it indicates the supervisor was

not in tune with his people.” If you can learn the reasons a person is
leaving, through a confidential exk survey or an exit interview, it can
only be beneficial, although employees are not required to oblige.
Some human resource professionals favor taking a survey, con¬
tending that because it is confidential, it’s more likely to yield honest
answers. One advantage of the exit interview, however, is that you
may be able to salvage the situation (assuming that you want to retain
the person). She might raise issues that management wasn’t aware of;
perhaps they can be redressed to everyone’s satisfaction, and the em¬
ployee can be persuaded to stay. If that’s your preferred outcome,
hold the exit interview as soon as possible after notice has been
given. Don’t wait until the person has grown completely comfort¬
able with the decision and colleagues are busy planning the menu for
the farewell party.

BE A COMPANY YOUR PEOPLE

CAN BE PROUD OF

A worthwhile mission helps a company retain employees. We’d all

like to believe that our lives have meaning, and what we do for a liv¬
ing constitutes a large part of who we are. Amgen was extremely for¬
tunate in this regard. As Kevin Sharer, who succeeded me as CEO in
2000, puts it, “We really are trying to cure cancer.”
Working at Amgen was a source of pride for our people. We had a
tradition that any noteworthy achievement by a team or group enti¬
tled the members to announce it on a T-shirt. It became a joke that
employees would never have to buy another T-shirt, even well into
retirement.
The staff members liked to wear their Amgen T-shirts around
Thousand Oaks, because strangers often came up to them and said, “I
just want to thank you. Your product did so much for my aunt.” Or
uncle. Or brother. Who wouldn’t like to hear that about her company?
 How Amgen Kept Employees Committed 243

Employees also felt good because Amgen behaved like a model

corporate citizen. In addition to bringing thousands of employees and
visitors to Thousand Oaks, the company pitched in however it could.
One time we bought a building that we were eventually going to tear
down. An employee, on his own initiative, got the idea to contact the
local fire department and invite its personnel to use the structure for
practice, an offer that was gladly accepted. They could spray all the
water they wanted and chop holes in the walls to their hearts’ con¬
tent. When the firemen pulled up outside the building for the first
drill, they were as giddy as kids on Christmas morning.
A similar offer was made to the Thousand Oaks police department
regarding a small bank that we’d purchased for the property. Until the
building was to be demolished, someone figured that perhaps the de¬
partment’s SWAT team would like to use it to simulate bank rob¬
beries. They were thrilled. Nearby residents were less than thrilled,
however, when they were jolted awake at two o’clock in the morning
by the sounds of explosives detonating and gunfire. I’ll bet the police
never received an offer like that again, certainly not from Amgen.
In 1991, the city planned a huge parade to honor the men and
women veterans of Operation Desert Storm. On the eve of the
event, someone came into my office and said that Thousand Oaks
was going to have to cancel its plans, because the city council hadn’t
raised enough money to cover the cost of cleanup. Seventy-five
thousand spectators were expected to attend.
“How much do they need?” I asked.
“Seven thousand dollars.”
“You know what to do,” I said. “Take care of it.” The tribute to
our soldiers went on as scheduled. “Amgen Saves Parade,” trumpeted
the local newspaper’s eight-column headline. How much money
would a company have to spend to win the immeasurable goodwill
that this small gesture brought us? A lot more than seven thousand
dollars, that’s for sure. And not that we did it for this reason, but not
long afterward Amgen went before the city council to seek approval
244 Science Lessons

for its master building plan. A public hearing was to be held at which
townspeople were invited to have their say. To ensure a balance of
viewpoints, residents would fill o&t cards stating their position, and
five or six names on either side would be called to the microphone.
At the hearing, the mayor of Thousand Oaks thumbed through
the cards and started to laugh. “I guess there’s probably no point in
anybody speaking today,” he said. “Every single card is for approval;
there aren’t any against.”

Spread the Good News

If your business makes, say, wheelbarrows and not life-enhancing

drugs, the level of corporate esteem might not be as great, but there
can still be considerable pride in being a manufacturer of quality
wheelbarrows sold at fair prices. Or maybe you manage an excellent
division within a weak company; that too can be a source of pride.
No matter what products or services you provide, your organization
should take advantage of opportunities to remind employees of the
good it does.
For example, we used to have patients who’d benefited from
Amgen medications come to speak to our staff. Public relations isn’t
only external; you can use your company newsletter to relay impor¬
tant developments. Increasingly, companies use electronic newslet¬
ters, which can be timely, have no space limitations, and eliminate
the cost of printing and distributing. Remember, though, that there’s
a fine line between instilling pride and making empty boasts.

EMBRACE CHANGE TO KEEP

YOUR COMPANY VIBRANT

I’ve mentioned that I spent a summer interning at Procter & Gam¬

ble’s corporate office in Cincinnati. One aspect of the company has
 How Amgen Kept Employees Committed 245

stayed with me: its emphasis on making change the status quo.
“There’s always a better way” was the Procter & Gamble philosophy,
and it backed it up with action. Senior managers were expected to
spend three-fourths of their time working on keeping things run¬
ning as usual. But the other one-fourth was to be devoted to new
ideas. It was made very clear, even to us summer interns, that innova¬
tive managers were the ones who got promoted.
I saw the virtue of this system. Incremental improvements added
up to making a product cheaper, easier to use, more effective, better.
At Amgen, we always looked for ways to refine existing products. By
trying to gain a better understanding of Epogen and Neupogen, we
learned things that let us develop their cousins Aranesp and Neulasta.
For another example, we later changed Epogen s packaging for rea¬
sons of patient safety. The drug vials are all the same size, with the
same amount of liquid; only the concentration is different. To help
ensure that a care provider didn’t accidentally administer the wrong
dose, we instituted a color-coded system with a different-colored
stopper for the 5,000-unit vial, the 10,000-unit vial, and so on.
It’s more fun to work in an atmosphere where change is embraced
and creativity encouraged (except when it comes to accounting; ex¬
cessive creativity there can create problems). At most companies, peo¬
ple are conditioned to fear change. If the new approach fails, someone’s
head will land on the chopping block. Management must accept that
innovation involves an element of risk, or else its employees will ap¬
proach their jobs like an athlete playing defensively—afraid to make a
mistake—instead of making things happen. Teams that play not to
lose usually lose, and those that play to win are more likely to win.
George Mueller, president of System Development Corporation,
where I served as CFO from 1971 to 1981, taught me a lot about the
importance of taking calculated risks. Before I tell you about
George, whose last name is pronounced “Miller,” let me give you a
little background on this fascinating nonprofit company.
246 Science Lessons

SDC evolved out of RAND Corporation, a think tank that’s had

a hand in furthering such revolutionary technologies as computing,
missiles, and space satellites. SDC\ consulted for the navy and the
army, but its number 1 customer was the U.S. Air Force. Competitors
began complaining bitterly to Congress that SDC held an uniair ad¬
vantage. The company, worried that steps might be taken to put it
out of business, converted to a for-profit enterprise in 1969. Under
California law, this is no simple process. Legally, a nonprofit corpora¬
tion belongs to the state’s citizens. Therefore, you must first form a
foundation, which then owns the new company and its stock.
Maybe because SDC had spent twelve years as a nonprofit, the
people at the top did not know how to turn a profit. That’s when
they brought aboard George Mueller, who’d headed the National
Aeronautics and Space Administration’s manned spaceflight program
from 1963 to 1969. If not for Mueller, a narrow-shouldered, bespec¬
tacled man with a PhD in physics, the United States would not have
met President John F. Kennedy’s 1961 pledge to put a man on the
moon by the end of the decade.
Throughout his distinguished career, Mueller had consistendy
followed his natural impulse to achieve goals that conventional wis¬
dom dismissed as impossible. For example, only weeks into his tenure
as NASA administrator, Mueller instituted tighter controls, begin¬
ning with a landmark decision to implement “all-up” testing of the
Saturn V rocket that would propel U.S. astronauts beyond the bonds
of gravity.
The Saturn V consisted of three rocket-powered motors stacked
one on top of the other. Under existing NASA protocols, each stage
was to be tested independently. Only after the stage 1 rocket proved
itself airworthy would a live second stage be added, and then a live
third stage. To George, this plan seemed needlessly time-consuming
and costly and, what’s more, stemmed from a presumption of failure.
He proposed road-testing the fully assembled rocket with the com-
 How Amgen Kept Employees Committed 247

plete Apollo spacecraft on top. The first full-scale Saturn V lifted off
from Cape Kennedy on November 9,1967, and all three stages per¬
formed flawlessly.
Similarly, the Apollo program’s original itinerary called for sixteen
manned or unmanned trial flights before the first moon landing was
attempted. Mueller considered this plan overly cautious and telescoped
it by more than half. The streamlined plan enabled Apollo 11 astronauts
Neil Armstrong and Buzz Aldrin to plant the U.S. flag in the moon’s
powdery surface on July 20,1969, months ahead of schedule.
Mission accomplished, George left NASA to become a vice pres¬
ident at defense contractor General Dynamics [Link], at
age fifty-three, he moved on to System Development Corporation,
where I watched him exhibit the same bold yet steadfast leadership.
Mueller didn’t merely think “out of the box,” as per the latest bro¬
mide; his worldview dismissed the very idea of [Link] was only
the correct answer to whatever problem he was trying to solve.
Mueller philosophized that in business, as in life, many folks fall
into the habit of saying that something can’t be done, when what
they really mean is only that it hasn’t been done before. That alone
justifies trying it. On numerous occasions, I came to him with elab¬
orate explanations of why something the company wanted to do
wasn’t feasible—say, because of state regulations.
What I saw as legitimate reasons were merely excuses to George.
“Well,” he’d say thoughtfully, “I’d like for you to go back and see if
there isn’t room within those regulations to support what we want to
do. If there isn’t, then find a way to get the regulations changed.”
Sometimes I used to leave his office thinking, “What the hell is
wrong with this guy? I gave him a clear explanation of why this can’t
be done, and he’s telling me to do it anyway.” During the ten years
that we worked together, I kept a mental scorecard, and you know
what? At least 50 percent of the time, he was right. I’d given up too
quickly. Most people do.
248 Science Lessons

MAINTAIN THE ENERGY OF A START-UP

LONG AFTER YOU’VE STARTED
V
There’s something special about the energy and passion of a start-up
company. George Rathmann describes it as a place where most of the
people “demonstrate exuberance. In many cases, though, businesses
emphasize rigidity and structure more than they inspire new ideas.”
George worked at 3M for more than twenty years. There, he says, “it
was a risk to be exuberant. There were behavioral restraints that ad¬
vised you to be cautious and not get too excited about plans that you
might not be able to deliver. But at Amgen, there was no doubt that
the sky was the limit. That kind of atmosphere is very attractive.”
Dennis Fenton thought so, too. When he came to interview at the
small company in 1981, he notes, “the energy level was indescribably
different from my experience at Pfizer. Everyone I encountered had
the belief that you could do something profound with science and
move knowledge forward, which was why I decided to throw my lot
in with George and these scientists and see what happened.”
The trick is to not lose that entrepreneurial spirit as your business
grows. We were very conscious of this and spent a great deal of time
working on preserving our culture. One way is to work in teams, to
mimic the dynamics of a lean company. All the Amgen values and
principles that I’ve described here—allowing employees to help plan
their work, encouraging collaboration, and so on—are just as essential.
They’re part of a larger whole, with each one reinforcing the others.
CHAPTER

The Value of Ethics in Business

in 2000, my final year as CEO ofAmgen, BusinessWeek named me

one of its top twenty-five managers in the world. Within a year or
two, several of my fellow honorees resigned in disgrace from compa¬
nies such as Enron, MCI WorldCom, Inc., and Martha Stewart Om¬
nimedia. A few wound up trading their pinstripes for prison stripes.
The ethics scandals of the past few years have undermined public
confidence in U.S. business and have helped fuel the undercurrent of
cynicism that pervades our culture. I’ve heard some of my contem¬
poraries remark ruefully that they’re almost ashamed to admit they’re
senior executives. What disillusions us most is that the actions of a
corrupt few tarnish the image of corporate leaders, the vast majority
of whom are hard working and abide by the law.

PRESSURES ON BUSINESS

As for what can be done to prevent such abuses of power and the pub¬
lic trust, we need to rethink the current system. Of course individuals
250 Science Lessons

are responsible for their own conduct. However, three fundamental

pressures have contributed to destabilizing the ethical foundation ol
corporate America. y

Changes in Compensation Practices

Over the past two decades, changes in executive compensation prac¬

tices helped spawn the corporate obsession with puffing up stock
prices, a strategy that may have helped bring about some unfortunate
actions in the executive suites of Global Crossing, Tyco Industries,
and others. Ironically, some of the blame rests with shareholders, who
complained that many executives accomplished little but were paid
as generously as if they accomplished a lot. Congress agreed and
passed a law eliminating the tax deductibility of executive compen¬
sation in excess of $1 million that was not “performance based."
Business responded by instituting incentive pay programs heady
weighted toward stock options and “performance” bonuses as a way
to hold managers more accountable. The trend grew throughout the
1980s and 1990s. According to consulting firm Towers Perrin, base
pay for a CEO at a mid-sized U.S. company currently averages 27
percent of total compensation. Compare this to similarly sized busi¬
nesses in England and France, where base pay makes up 43 percent
and 40 percent, respectively, of a CEO’s total income, and incentives,
only 35 percent and 41 percent. Companies might not have granted
so many stock options if the new tax laws hadn’t given them a de¬
duction every time an option was exercised.
Having more than two-thirds of an executive’s remuneration
hinge on the company’s reported financial results seems to invite
trouble. I’m referring not only to the flagrant offenses that made the
evening news—wherein some corporations spun the illusion of
great profitability through accounting sleights of hand—but also to
business decisions that, though legal, pumped up the stock price in
the short term at the expense of long-term growth. Given the inher-
 The Value of Ethics in Business 251

ent conflict of interest between what is good for the company and
what is good for the executive, none of this should have come as a
surprise. Most executives were steadfast enough ethically to resist
temptation, but many weren’t.
Similarly, when directors’ compensation is heavily option ori¬
ented, it can compromise their effectiveness in policing the executive
leadership. Think about it: they’re supposed to prevent mismanage¬
ment, but when fraud is committed on their watch, they’re essentially
being paid a percentage of the take. Accordingly, some companies
have done away with incentives for directors and now pay them
straight salaries. More businesses should follow their example.

Analysts’ Conflicting Interests

Fortune magazine estimated the cost of a Wall Street soothsayer at $1

million a year, for which he brings in $500 million in investment
banking business and trading revenues. If those figures are to be
believed, we have had another inherently perilous set of circum¬
stances. Analysts understood that the more business they procured
for their firms, the bigger their bonuses; that undercut their ability
and incentive to make sound recommendations with investors’ in¬
terests in mind.
Unfortunately, the cure for this problem has substantially reduced
analysts’ compensation, because they aren’t worth as much to their
firms if they’re not permitted to generate business. The result has
been an overall decline in the quality of analysts, as many of Wall
Street’s best have moved on to other, better-paying lines of work.

The Credibility GAAP

The U.S. Securities and Exchange Commission has had a hand in

this misadventure as well. Formerly, a company could publicize only
one type of earnings per share, which its auditors would [Link]
252 Science Lessons

SEC did not allow proforma earnings, operating earnings, or recur¬

rent earnings, for which some of the costs are left out.
Then the SEC began allowing businesses not only to report fi¬
nancial statements based on the set of standards known as generally
accepted accounting principles (GAAP) but also to declare an addi¬
tional set of numbers. In theory, these other numbers represent the
business more accurately than GAAP numbers, which often obscure
more than they reveal. An auditor’s job went from helping turn out
accurate financial statements to determining whether the company
was in compliance with GAAP. Fortunately, the two goals are not
mutually exclusive. An unscrupulous but creative business executive
can write a misleading financial statement that nevertheless receives
the auditor’s statement of approval. From what I’ve read about MCI
WorldCom, one year senior management simply took the number it
needed to meet analysts’ expectations and worked backward, trans¬
ferring the exact amount from expenses to capital for each quarter. I
never imagined that large, publicly held companies would dare to be
so blatant about it.
Maybe we should add a provision that if a financial statement is
found to be false or misleading, the auditor should refuse to certify it,
whether or not it meets GAAP criteria. The Universal Code of Mil¬
itary Justice, the legal system of the U.S. armed forces, has an article
that has worked well for many years: “conduct unbecoming an offi¬
cer and a gentleman.” There’s no definition for this charge—it’s often
explained as “You’ll know it when you see it”—but a violation is
grounds for a dishonorable discharge.

A Perfect Storm

The combined effect of these oversights created a situation wherein

the natural forces would inevitably lead to disaster. It was only a mat¬
ter of time. I’m a big believer in natural forces. If they align to facili-
 The Value of Ethics in Business 253

tate unethical behavior, we’ll see more unethical behavior; if they

align to discourage dishonest behavior, we’ll see more honorable
conduct. How many executives would have sanctioned phony ac¬
counting if none of their income was tied to the stock price? How
many auditors would have looked the other way if they made their
money primarily on auditing fees and not on consulting, as was gen¬
erally the case?
Maybe the tragic sight of formerly powerful corporate leaders
being escorted out of their grand buildings in handcuffs and paraded
before the media will prove to have a sobering effect on those who
might otherwise be tempted to cheat the system.

CREATING A CULTURE OF ETHICS AND

HONESTY: IT ALL STARTS WITH YOU

I’ve never understood people who think they must he, cheat, and steal
to get ahead; I guess that comes from the example set for me by my fa¬
ther, who was probably the most honorable person I’ve ever known.
Charles Binder was a schoolteacher and later a school principal
and superintendent. He hadn’t planned on teaching for a iiving. My
dad graduated from the University of Illinois with a master’s degree in
chemistry. In the early 1930s, most folks didn’t seek opportunities
very far from home, even if they were well educated. But not long be¬
fore graduation, he spotted a notice on a bulletin board: “Teach Sci¬
ence for a Year in New Mexico.” The idea of living in the Wild West
had a lot of appeal for a young single fellow. Not that New Mexico
was exactly the Wild West, but to someone who grew up in Waterloo,
Illinois, outside St. Louis, it must have seemed close enough.
In those days, you didn’t need a teaching degree, only a thorough
knowledge of the subject matter. Off went my father, figuring he
would spend a year in New Mexico and then return home to marry
his high school sweetheart, Rodella—or Rody, as everyone called
254 Science Lessons

her. But my father loved New Mexico. He also discovered that he

loved teaching and loved kids. He was still fascinated by science, but
here was a way to do both. He ancfRody, my mother, also a teacher,
married and moved permanently to Alamogordo.
My father was extraordinarily principled. I'm not sure how he
got that way, but he set an example for other people, especially me. It
wasn’t something that we talked about much; he lived it, and that was
far more eloquent. He was a phenomenal teacher. This I know, be¬
cause he was my chemistry and physics teacher. In a town the size of
Alamogordo—with perhaps twenty thousand people when I was in
high school—there was only one chemistry and physics teacher.
He put in many extra hours to help students. One young man
was a gifted athlete. He remarked to my dad that he wanted to be¬
come a javelin thrower. My father found a book about the sport and
worked one-on-one with the boy after school so that he could enter
the state track meet. Later, Dad became principal of a school in
nearby Cloudcroft, an even smaller town. The school, which taught
grades 1 through 12, didn’t have a formal library or a librarian, only a
bunch of books piled haphazardly in a room. Principal Binder took
it upon himself to organize all the books, apply the Dewey Decimal
System numbers on their spines, and create a comprehensive card
catalog—all on weekends. My father provided a shining example of
how an authority figure’s good deeds can set a positive tone that per¬
meates an organization.
His influence has guided me throughout my career, even when I
was starting out. I’ll tell you a story. Almost every job I’ve ever had
has come through word of mouth, as when Amgen counsel Ed Hud-
dleson recommended me to George Rathmann because he was fa¬
miliar with my work at System Development Corporation. Not one
help-wanted ad that I answered ever turned up a damn thing. In
1962, with graduation from Harvard Business School looming, I
sought one of my professors, Myles Mace, for career advice. My
dream job, I explained, was to become right-hand man to an execu-
 The Value of Ethics in Business 255

tive at a company involved in mergers and acquisitions. It seemed

extremely exciting to me. Where would he recommend I apply?
“There’s only one company,” he replied. “Litton Industries.” Mace
had firsthand knowledge of the famously successful conglomerate,
having been an original board member. In 1955 the company’s co¬
founder and chairman, the colorful Charles B. “Tex” Thornton, of¬
fered him a position as vice president and general manager of the
electronics equipment division. Harvard gave its blessing; the school
typically encouraged faculty members to immerse themselves in the
real world every few years. Mace spent three years at Litton, increas¬
ing annual sales from $3 million to more than $80 million. Then he
returned to teaching and sitting on Litton’s board.
Now he called out to his secretary, “Suzi, get me Tex Thornton on
the phone!” As it happened, the husky, ruddy-faced Litton chairman
was in his Beverly Hills office and took the call. I heard only the pro¬
fessor’s side of the conversation. He told Thornton, who had been the
leader of Ford Motor Company’s storied “Whiz Kids” in the late
1940s, that he had this bright young guy sitting across from him, an
electrical engineering undergrad, and that Thornton should hire him.
“Okay,” said Thornton, “have him come out for an interview.”
Litton’s highly decentralized corporate structure, so familiar
today, was novel for its time. There were five senior vice presidents,
each of whom was allotted one finance person, one secretary, and
one assistant fresh out of business school. I wound up being inter¬
viewed by Harry J. Gray, a World War II army hero who’d been with
Litton since 1954. He was vice president of the components group.
Ultimately Gray left Litton to become chairman, president, and chief
executive officer of United Technologies, another multibillion-dollar
company, in the 1970s and 1980s.
“If I offer you this assistant’s job and you take it,” he said, “how
long do you think you should keep it?”
I thought,“I don’t [Link] an odd question.” For some reason,
my mind settled on two years. We shook hands, and that was that.
256 Science Lessons

Harry Gray had the best people skills of anyone I’ve ever met, in
or out of business. Because he was a born salesman, and 1 am not—I
tend to be introspective and am vjiot a natural glad-hander, even
now—I spent a lot of time studying his interactions with his stall and
asking him [Link] Georgia native, who was forty-three when
I met him, was the kind of forward-thinking executive who hired
good people and then invested in them by grooming them for ex¬
panding responsibilities. He believed that a senior manager's job was
to help his subordinates do their work, and not the other way
around, a philosophy I carried with me after leaving Litton.
In 1964, right around my two-year anniversary as his assistant,
Harry called me into his office and reminded me of our very first
conversation. “Do you still think two years is a sufficient amount of
time in this job?” he asked. I said I did, and he agreed with me. After
investigating other opportunities within Litton, I decided to move
on to Ford, where I’d served an internship.
Gray lived next door to Keenan Wynn, versatile character actor
and son of vaudeville and film clown Ed Wynn. The younger Wynn
was an avid motorcyclist—he raced anything that moved, on land or
water—and had been after Gray to go riding with him. Harrv had
never been on a motorcycle and didn’t want to start now, but he al¬
lowed the actor to talk him into it.
The bike skidded on an icy patch, and slammed into a guard rail.
One of Harry’s legs was wedged between the ground and the cycle;
the other between the cycle and the rail. He wound up with com¬
pound fractures of both legs and faced several months in traction in
a hospital.
Arrangements were made for another Litton executive to take
Gray’s place while he recuperated. However, his division managers
complained that bringing in someone else wasn't going to work.
“Gordon,” he said to me one day, “I know you’re about to leave us
for Ford. But it looks like I’m going to have to keep running my part
 The Value of Ethics in Business 257

of the company from my hospital bed. If you could possibly stay until
I get out of here, I would really appreciate that.”
Despite the fifteen-year difference in our ages, Harry had be¬
come a close friend as well as a great teacher and an inspiration. I felt
that I owed it to him to help him through this crisis. So I called my
future boss at Ford and told him I wouldn’t be able to start for three
months. Although my new employer objected, I held firm. “I’m
sorry,” I said. “I promised my boss I would do this. If that doesn’t
work for you and you need to get someone else for the job, then so
be it.” They backed down and said they’d wait for me.
Conveniently,Tex Thornton’s partner in Litton, Roy L. Ash, had
donated the wing of the hospital where Gray was staying. We com¬
mandeered the room next door, replaced the bed with a desk, moved
his secretary in there, and conducted business as usual, despite
Harry’s being in traction. Even if my staying on had cost me the po¬
sition with Ford, it was the right thing to do. I strongly believe that
doing the right thing, in addition to being its own reward, frequently
pays [Link] was certainly true in this instance.
In contrast, the unethical business executive usually considers
only the external effect of her misdeeds—in other words, how her
deceptions benefit her company. What she usually overlooks is that
corruption at the top seeds a culture of dishonesty that extends
throughout an organization until it rots from within.
When employees work at a company where crooked practices are
not only tolerated but tacitly sanctioned—or flagrantly encouraged—
do you think they trust management not to turn on them at some
point? By the same token, when management encourages its people
to cheat customers, suppliers, and others, doesn’t it stand to reason
that some of them are going to scam the company, too, whether it’s
pocketing office supplies, filing false expense reports, or embezzling
large sums? According to a survey of 12,750 U.S. workers at all job
levels and in all major industries, fewer than half the men and women
 :-',i Ltsssrts

rotied said that they trusted their senior managers. In such an envi-
30C0SBL die rank and die often develop a do-it-to-management-
VhuMnan; '.^ement-iloei-it-tn-UTti attitude. The study, conducted
h. Noca! consulting tirm Watson Wvatt Worldwide, also showed that
companies with high levels ot trust outpertormed companies with
low trust levels by 186 percent.
Over rime, self-selection comes into play. Ethical people don’t
wart to remain at a place where Iving. cheating, and stealing .ire
commonplace, whereas those whose consciences don't nag them
wv tare rust tine—and stay a long time.
In that same Watson W yatt survey, one in four people claimed to
have known of or suspected an ethical violation in the past two
years: however, only about half reported the [Link] reasons usu¬
ally given tor falling to notify supervisors of misconduct are a tear of
retaliation as well as a lack of faith that the organization will respond.
Now hold up a mirror to the same scene. In a company where
[Link] are entrenched in the culture and practiced consistently, with
r.o double standards based on hierarchy, staff members are more
Mkdtv to expose unethical practices. First, they do so because they
know that management will support them; second, they take pride in
their company and therefore have a personal stake in protecting its
reputation; and third, they're probably the kind of men and women
who genuinely resent dishonest dealings. Shady characters, on the
other hand, usually get weeded out quickly at good organizations;
sometimes they voluntarily prune themselves from the pavroll be¬
cause they can see that their "skills." such as thev .ire. won't benefit
them there.
T hen tolks have worked for you long enough, thev come to un¬
derstand instinctively what's acceptable and what isn't. The organiza¬
tion's ethical guidelines inform their decision making, and thev don’t
always need a supervisor to clarify it tor them. One time Amgen
rnred a semor executive. A few weeks later. 1 ran into Ed Garnett,
read of human resources, who told me that he'd had to tire the man.
 The Value of Ethics in Business 259

“Fire him? He hadn’t even been here for a month. What

happened?”
Ed said, “We discovered that he’d falsified his resume in a major
way, so I knew I didn’t need to check with you first. I went ahead
and let him go.”
We’d already informed the other candidates that the position had
been filled, and to be honest, the abrupt departure created a mess for
a while. Ed was right, though; he didn’t need to check with me first.
The misrepresentation almost certainly wasn’t an isolated incident
but indicative of the character of the employee-turned-ex-em-
ployee. It would have revealed itself again in the future, perhaps with
far more serious consequences. What if he made false statements to
the IRS or the FDA? There was no question about it; he was out.
Sometimes it took a while for new hires who’d come from other
companies to realize that Amgen was serious about its commitment
to ethics. If you follow professional bicycle racing at all, you probably
know that there have been reports of cyclists injecting themselves
with EPO to improve their endurance. The increase in red blood
cells brings extra oxygen to the body’s cells, including muscle tissue.
A doping scandal marred the 1998 Tour de France, when teams from
the Netherlands and Spain had to drop out because some of their
members were abusing EPO. The practice is not only illegal but also
dangerous: when a person who is not anemic takes Epogen, the in¬
flux of red cells thickens the blood, and that could heighten the risk
of suffering a heart attack or stroke.
I believe that companies have an ethical obligation to do the right
thing. An awful lot of people in the pharmaceutical business would
probably disagree; in their view, it’s up to somebody else to regulate
abuses of medications. Amgen did whatever it could to discourage
athletes from misusing Epogen. We worked with the National Colle¬
giate Athletic Association, the National Olympic Committee, and the
International Olympic Committee. At our expense, we had an educa¬
tional booklet written and sent to high school coaches. To help get
260 Science Lessons

the word out, we invited the media to an all-day information seminar

in New York City, with gold-medal winners as some of the presenters.
In addition, during my talk tc\new salespeople on their first day
of training, I always stressed that Amgen actively worked to prevent
EPO abuse. We didn’t want those sales. And I could see from the
smirks on some of the faces that not everyone believed me. “Oh, I
get it; this is for public consumption, right? Maybe you don't want
me selling like that as long as I’m making my quota, but just take a
month where we’re coming up short, and I’ll bet you ...” I had to
drive the point home more forcefully, warning,“I am not kidding. If
we catch anyone encouraging illegal use, they’ll be fired. Don't think
we’re not serious; we’re deadly serious.” By the end of my chat, most
of the skepticism in the room had been expunged.
I also told each class to always tell doctors the truth about each
product, warts and all, even if it meant we sold less. The truth was,
we’d sell more. A salesperson is nothing if he loses credibility with a
doctor. Goodwill is vital for any business, but particularly a company
selling health-care products.
Researcher Burt Ensley, who left Amgen in 1989 after eight
years, went on to found several companies of his own. Amgen, he
says, “had a higher level of integrity than anything I’ve seen before or
since. I really learned that there and use it. Contracts were to be hon¬
ored. You kept your [Link] is an unusually high standard, as I dis¬
covered after leaving.”

PLAYING POLITICS BUT PLAYING FAIR

At some point during your career, it may fall to you to do business

with local, state, or federal legislators in order to have a say in public
policy. It might be an issue close to home, such as appearing before
the city planning board to seek a change in zoning restrictions. Or
you might find yourself addressing Congress as it deliberates a bill
that would affect your industry.
 The Value of Ethics in Business 261

Because the biotech and pharmaceutical industries are subject to

much federal legislation, Amgen felt that it needed to establish a
presence in Washington, D.C. In 1993 we hired Peter [Link], for¬
mer U.S. ambassador to Canada, as vice president of government and
public relations. Most of Pete’s long political career had been spent as
press secretary to Republican lawmakers—Sen. Robert Griffin (R-
MI), Sen. Jacob Javits (R-NY), and Vice President George H. W.
Bush—as well as communications director and chief spokesman for
the National Republican Committee. He was highly respected on
Capitol Hill and gained us entree to many key legislators.
When Amgen started, we knew next to nothing about lobbying.
Because of a corrupt minority, the practice has been maligned as un¬
savory, and that is unfortunate. At its core, lobbying enables industries
as well as groups of concerned citizens who come together in special
interest groups to educate lawmakers about various issues. In a democ¬
racy, they’re free to advocate a position; the government officials they
approach are free to listen to them or disregard them, find value in
their viewpoint or reject it.
Circumstances forced us to set up the Washington office. You’ll
recall Amgen’s patent dispute with Massachusetts biotech firm Ge¬
netics Institute and its Japanese partner Chugai over Epogen, dating
back to [Link] was no dispute that we held exclusive rights to
the CHO cells used for cloning the human erythropoietin gene.
Amazingly, a gap in patent law made it possible for Chugai/GI to
duplicate the process that we’d spent years and millions of dollars
developing, manufacture its own epoetin alfa overseas, and export it
to the United States. When we filed a complaint with the Interna¬
tional Trade Commission, an ITC judge ruled that the matter fell
outside its jurisdiction. The dispute called for a legislative solution,
she said.
In 1995 Sen. Orrin Hatch (R-UT) and Sen. Edward Kennedy (D-
MA) cosponsored the Biotechnology Patient Protection Act, which
finally closed the loophole. Steven Odre, Amgen vice president and
262 Science Lessons

associate general counsel, appeared before the House Judiciary Com¬

mittee to explain that under the existing patent and trademark laws—
drafted long before the inceptidn of biotechnology—any foreign
competitor was tree to trample all over U.S. companies’ patent protec¬
tion. Congress approved the measure in the fall.
That same year, 1 spoke before the House Commerce Committee
about the need for reforming the U.S. Food and Drug Administration.
I was representing Amgen, of course, but primarily the Biotechnology
Industry Organization. At the time, I chaired BIO’s Committee on
FDA Reform, in addition to serving as a board member. We were
there to lend support to the Food and Drug Administration Modern¬
ization Act, a bill introduced by Sen. Jim Jeffords (R-VT). Its seven
cosponsors came from both sides of the aisle.
This legislation was to extend an earlier measure, the Prescription
Drug User Fee Act of 1992, which had been enacted out of hair¬
pulling frustration with the amount of time it took the understaffed
FDA to evaluate the findings of drug trials. When a manufacturer be¬
lieves that its patient studies have yielded sufficient evidence of safety
and efficacy, it submits either a new drug application (NDA) or a bi¬
ological license application (BLA) for consideration.
By law, the agency is allowed ten to twelve months to review the
results and judge whether or not the product should be licensed for
commercial use. or six months if it looks to be of significant benefit
and is therefore designated a priority. But as of 1986, it was taking an
average of thirty months—twice as long as the FDA’s European coun¬
terparts. Amgen once had to wait nearly three years for clearance to
add a new use to Neupogen's list of indications. What’s more, during
that time we were prohibited from telling physicians how this new
application could help some of their patients.
In 1992 the pharmaceutical industry took the unusual step of
lobbying Congress for legislation that would impose a user’s fee
upon the companies themselves to help underwrite the cost of FDA
reviews. Over the next five years, this brought an additional $329
 The Value of Ethics in Business 263

million to the agency, which hired seven hundred new employees

and streamlined its operations. By 1997 the NDA and BLA review
time had been slashed to about thirteen months, a blessing particu¬
larly for people with incurable diseases such as AIDS. As for the drug
companies, the $329 million in user fees amounted to a fraction of
the sales revenue that they had been losing because of the delay in
getting products [Link] about a win-win-win outcome.
Now the Prescription Drug User Fee Act was about to expire.
The FDA Modernization Act would not only reauthorize the previ¬
ous bill for another five years but also make further improvements,
such as increasing patient access to experimental drugs and accelerat¬
ing review of important new medications. Congress approved U.S.
Public Law 105-115 by a nearly unanimous vote, and when Presi¬
dent Clinton signed it into law on November 21, 1997, I was pres¬
ent, proud to have spearheaded the effort on behalf of both the
biotechnology and pharmaceutical trade associations.
The lessons we learned in Washington also apply to dealing with
local government. Lesson number 1 is to not be greedy. Ask for too
much, and you’re likely to end up with nothing. PhRMA, the Phar¬
maceutical Research and Manufacturers of America trade associa¬
tion, had joined BIO in making suggestions for speeding the drug
review process. Some of its members were pushing for a time frame
that was unreasonably short, to the extent that it could compromise
patient safety. My concern was that if we made outrageous de¬
mands, the FDA wouldn’t join hands with us. We needed to present
a united front; otherwise, the modernization act stood little chance
of passing. I imposed a rule that our recommendations had to be
good for patients and the country, and not beneficial only for our
industry. Anything that failed to meet these criteria was to be tossed
overboard.
I said to Pete Teeley, “We’ve got to find some senior person at the
Food and Drug Administration who could go through our proposals
and tell us if there’s anything goofy in here, so we can omit it before
264 Science Lessons

it goes before Congress.” A certain person came to mind. I made the

call and explained why we’d like to meet with her.
“It’s a good idea,” she agreed.‘(But I can’t do it, because they keep
logs of everybody we meet with in our offices, and I’m not author¬
ized to have this kind of meeting.”
“It really would make for a better bill,” I countered. “Isn’t there
some way that we could get together?”
We agreed to meet for a drink in a hotel cocktail lounge. Pete
and I went over the entire proposal with her. By the time we fin¬
ished, she said approvingly,“I think everything in here makes sense.”
Among the legislators present at the Congressional hearing into
the proposed law were two perennial opponents of the drug industry:
Rep. Henry Waxman (D-CA) and Rep. Pete Stark (D-CA). An ad¬
ministrator from the FDA was on the stand.
Rep. Waxman, aware that the pharmaceutical and biotechnology
companies (and not the FDA) had been first to sound the call for the
new law, asked him somewhat skeptically, “Does the FDA believe
that the new legislation is really needed?”
“Yes, Congressman,” he replied, “we do.”
Waxman’s face registered surprise. He asked no further questions
and raised no objection to the bill. In the past, even presidential com¬
missions hadn’t been able to get FDA reform legislation through
Congress. But this bill passed easily. Not only were industry and its
regulatory agency both in agreement about what needed to be done,
but also the measure lent itself to bipartisan support. Who could be
against improving the FDA? I have been involved in countless leg¬
islative battles, and it is not unusual for members of Congress to op¬
pose biotech on one issue and support it on another.
Lesson number 2 is to always tell the truth, no matter what, even
if the facts might undermine what you’re trying to accomplish.
After several years of sticking steadfastly to that policy, Amgen’s
Washington office started receiving calls from congressional staffers
asking us to verify the accuracy of information they’d received from
 The Value of Ethics in Business 265

other pharmaceutical companies. That’s how much they trusted us

by then.
Lesson number 3: meet with any member of a lawmaker’s office.
Typically, if a company expects to set up a meeting with a member of
Congress, it must send its CEO. But things happen in Washington,
and sometimes the congressman or senator must cancel at the last
minute. The CEO is already in Washington anyway, but often he’ll
refuse to talk with the legislator’s chief of staff, as if dealing with a
staff person was beneath him. This stuff goes on all the time.
In many cases, it’s more important to talk to the chief of staff than
to the congressman or senator. Amgen made a rule that we would
take whatever appointment we could get with whomever we could
get. The more junior the person was, the more she appreciated our
willingness to go ahead with the meeting, and that often worked to
our advantage in the future. It’s no different from the age-old advice
of befriending the front-desk receptionist when you’re trying to set
up a job interview with the boss.
After the FDA Modernization Act passed, Amgen decided that we
should thank the many unelected folks on Capitol Hill who’d had a
hand in it. So we invited more than one hundred congressional staffers
to a celebration dinner—no senior people, only staffers. At first, some
of them thought that the invitation was intended for their bosses and
had landed on their desks by mistake. No one invites D.C. staffers to a
dinner out of appreciation for their hard work. But it struck us as the
right thing to do. And what did it cost us? It was a drop in the bucket
compared to the goodwill it generated. More importantly, many un¬
sung heroes finally received some much-deserved recognition.
The overall experience left me with an unfashionably positive
feeling about Washington, and not only because the bill passed. De¬
spite what many folks probably think, if you have the right cause and
go about advocating it in the proper way, you don’t need to reach
into your pocket for a large political campaign contribution. In fact,
when Amgen was small, we succeeded without making any at all.
266 Science Lessons

Until I started meeting legislators, I never appreciated how con¬

scientious most of them are. They work extraordinarily long hours
for not very much money. Cyniss may scoff, but, by and large, most
politicians do their best to make the country better in the face of
many constraints, such as having to raise money for campaigns, pres¬
sure to toe the party line, and so on.
I even found myself admiring Henry Waxman, the bane of the
pharmaceutical industry—proof that politics does indeed make
strange bedfellows. And although I rarely agreed with Sen. Edward
Kennedy, one of the industry’s most outspoken critics, there’s no ques¬
tion that he genuinely has patients’ interests at heart. Whenever we re¬
quested a meeting, the senator sat down with us and listened carefully.
Two years later, in 1999, he and Amgen worked closely on trying to
drum up support for providing prescription drug benefits to Medicare
recipients. The federal health-care program, established in 1965, had
not kept pace with advances in medicine. All manner of costly surger¬
ies remained covered, but no medications, even though prescription
drugs had become the primary remedy for many health problems.
Kennedy proposed that the additional funds could be siphoned
from the Taxpayer Refund and Relief Act of 1999. For a while, it
looked as if we might have enough Senate votes. But ultimately, the
amendment went down to defeat, 45 to 55. All except two of the
nays came from the Republican side, and the yeas were exclusively
Democrat, with one exception.
The loss, though disappointing, did not dampen my faith in the
system. (Several years later, a Medicare drug benefit became law.)
With cynicism toward politicians rampant these days, most folks
probably think that the only way to get anything done in Washing¬
ton is through Public Action Committees (PACs) making large cam¬
paign contributions to mercenary representatives more interested in
getting reelected than on doing something good for the country. I’m
not naive; I know that it happens.
 The Value of Ethics in Business 267

It’s not the only way, however. You can take the high road and
succeed, as Amgen did. Our success is proof that there are still honest,
conscientious people in our government. If they were as incompe¬
tent or corrupt as skeptics would have you believe, Amgen’s serve-
the-country-first, tell-the-truth position would not have worked to
our advantage.

“WE HAVE TO DO THIS TRIAL”

“When in doubt, do the right thing”—so maintained Roy Ash, co¬

founder and president of Litton Industries. That was Amgen’s philos¬
ophy, too. My favorite story about Amgen best encapsulates its view
of itself and its place in the world.
One time in the 1990s, a bunch of us were discussing the possi¬
bility of testing Epogen in infants. Pediatricians had been using the
drug for a rare type of anemia seen exclusively in babies, even though
it had been tested only in adults. Conducting clinical trials with in¬
fants might yield more specific information that could be highly
useful to physicians, particularly in terms of dosages and schedules.
However, every logical argument pointed to not going ahead
with the study. It was not required by the FDA, it would be costly,
and it could be risky.
“No mother is going to give us permission to enroll her baby in a
clinical trial.”
“Hardly any children have this form of anemia anyway.”
“They’re already getting Epogen, and it’s helping.”
“To do a clinical trial isn’t going to help sales.”
Finally, DanVapnek, the head of research, spoke up. “We have to do
this trial,” he said intensely. Everybody else looked at him as if to say,
Haven’t you been listening? “It’s our product. Those infants need this
product. And their pediatricians need to have the information to use
our product properly in those infants. So we have to do this trial.”
268 Science Lessons

I looked around the room. No one said a word. Decision made,

meeting over. Everyone stood up and filed out. From a commercial
standpoint, it was not the wisest choice. But we had an ethical obliga¬
tion, we were in the business of health care, and therefore we should
just do it.
I was never prouder of Amgen than at that moment.
CHAPTER

My Failed Retirement

the final chapter OF an executive’s tour of duty can’t be written

until after he’s moved on, because one of his hardest jobs—particularly
if he’s the chief executive officer—is to select and install a successor.
When Kevin Sharer joined Amgen as chief operating officer and
president in 1992, retirement wasn’t even on the horizon for me. Our
instructions to the executive search firm we hired were to find some¬
one who could eventually be a candidate for the CEO position. It
wasn’t preordained that the new COO would ascend to the top, but he
or she should at least be capable of being considered for the job.
Some chief execs handpick a virtual clone of themselves as their
number two. Although that might seem appropriate for a biotech¬
nology firm, I believe it’s a mistake for any company in any field.
One of the things I liked best about Kevin was that although we
shared a common background—both engineers, both naval officers
(Sharer aboard a nuclear submarine)—he was different from me. I
liked the science side best, whereas he gravitated toward sales and
marketing. That made for a good fit.
270 Science Lessons

Not long after Kevin’s arrival, there was little doubt that someday
the board would approve him as my [Link] only question was
when. I’ll be perfectly honest: I lbved running Amgen and wasn’t in
any hurry to retire. However, when word gets out that someone is in
line for the number 1 job, he becomes fair game for other companies
on the hunt for a CEO.
By 1999, Kevin was probably getting the proverbial seven-year
itch, and I knew it. One day Ed Garnett walked into my office and
tossed his keys onto my desk.
“What’s this?” I said.
“Are you going to walk in here one day and just throw your keys
on the desk and say ‘I’m leaving today—good luck’?”
“Of course not.”
“Well,” he said, “don’t you think you ought to talk to people you
really trust about working out a transition program?”
“Yeah, probably,” I said.“But I’m dreading it.”
I decided to announce my retirement plans publicly on Decem¬
ber 7, 1999—that it happened to be Pearl Harbor Day was purely
coincidental. The coming year seemed ideal for the transfer of lead¬
ership. I’d be turning sixty-five in September, but more significantly,
Amgen’s pipeline would be turning out three new drugs in the near
future: anakinra (Kineret) for rheumatoid arthritis, along with our
second-generation colony-stimulating factors darbepoetin alfa
(Aranesp) and pegfilgrastim (Neulasta). My stepping down as CEO
on May 11,2000, at our annual shareholders’ meeting, would get the
Sharer era off to a brisk start, just as George Rathmann had done to
my benefit. Since my leaving Amgen, launches of new medications
have become even more involved and expensive, so today you might
initiate the changing of the guard at least two years in advance.
I chose to do one thing very differently than George had: to limit
my service as chairman of the board to six months after my retire¬
ment as CEO, and part-time at that. A new CEO can benefit greatly
from the previous chief executive’s experience during the first few
 My Failed Retirement 271

months. But over time, the predecessor becomes less informed, and
his usefulness fades rather quickly.
Moreover, I knew that I would be uncomfortable continuing to
stay at Amgen for too long. You come to work, but you almost hope
that no one knocks on your door, because if they do you’re afraid
you’ll be tempted to meddle. And if no one asks for your advice,
you’re not contributing, so why are you there? Six months seemed
right; whatever hadn’t transitioned by then probably wasn’t going to
transition successfully anyway.
“ There is a time for everyone to retire,” I told Amgen’s employees
in my farewell speech, as Adele, along with my sons Todd and Brant,
looked on from the audience. “My time is now.” The smile on my
face, I said, could be explained by everything we had accomplished
together over the past twelve years. Kevin was inheriting a company
with a market value of more than $60 billion; nearly seven thousand
employees in eighteen countries; a campus of more than thirty
buildings; and a research and development budget of $845 million, or
27 percent of its sales revenue—almost twice the average percentage
for the top ten pharmaceutical companies in the United States. And
for the third year in a row, Amgen had just been named one of the
one hundred best companies to work for by Fortune.
At the same time, I added, my heart was heavy “because I will
miss Amgen so much and miss all of you in so many ways.
“So long, good-bye, and God bless you all.”

ONE FINAL BATTLE

Days later I was sitting in Boston’s magnificent ten-story brick-and-

granite U.S. courthouse, perched on the city’s historic harbor.
Our latest patient-infringement lawsuit revolved again around
Epogen. The offender this time was a small biotech firm from
Cambridge, Massachusetts, called Transkaryotic Therapies (TKT). Also
named in the suit was its partner, European pharmaceutical behemoth
272 Science Lessons

Aventis, formerly known as Hoechst Marion Roussel. It was our

contention that TKT, founded by a Harvard researcher who special¬
ized in gene therapy, had violated eighteen claims spread across five
patents on EPO and the process for manufacturing it.
Briefly, the company used a different method called gene activa¬
tion (GA) to manufacture its own EPO. Gene activation doesn’t rely
on genetic engineering, in that none of the DNA used by a cell to
make EPO has been inserted into the cell; it simply rearranges the
existing DNA. Although TKT could be congratulated on a clever
scientific achievement, it didn’t mark an advance over recombinant
engineering, only a different route to the same destination. Even if it
had been superior, that didn’t entitle the company to profit from
Amgen’s patented product.
Having the trial set in Boston was inconvenient but necessary.
We wanted this nuisance over with. If you sue a company outside its
hometown, it might request a change of venue. By suing in Boston,
we had precluded that action. A second consideration was that we
hoped the case would be tried by Judge William Young. He’d
presided over our legal clash versus Genetics Institute/Chugai years
before and had impressed us with his fairness and discipline. In addi¬
tion, his familiarity with Epogen and Amgen’s patents could be ex¬
pected to bring about a speedier decision. Indeed, he was appointed
to hear the case.
A month before the trial commenced. Judge Young granted a
summary judgment in our favor, contending that TKT/Aventis’s
EPO infringed patent number 5,955,422: “a pharmaceutical com¬
position comprising a therapeutically effective amount of human
erythropoietin and a pharmaceutically acceptable diluent, adjuvant,
or carrier, wherein said erythropoietin is purified from mammalian
cells grown in culture.” The other claims v'ould be decided in court.
ForTKT to win the case, it would have to comince the judge that
for each of our other four patents, the patent was either (1) not valid,
(2) not infringed, or (3) unenforceable. In every patent case, there is a
 My Failed Retirement 273

legal presumption that the U.S. Patent and Trademark Office is com¬
petent, and that places a heavy burden on proving invalidity.
Almost every stock analyst gave TKT little to no chance of
pulling off a sweep. Yet during the twenty-three-day trial, stretched
over four months, more than one hundred business reporters, Wall
Streeters, and patent attorneys packed the courtroom—including
patent counsels for other biotech firms—because of the case’s far-
reaching implications. If TKT triumphed, it might open the flood¬
gates for any company to copy and market virtually any genetically
engineered medication; one that another firm had spent hundreds of
millions of dollars and many years to develop.
We considered TKT’s stand laughable. Nevertheless, we took it se¬
riously, thanks in part to a valuable lesson I had learned while at Har¬
vard Business School: even if an opponent’s actions make no sense to
you, they apparently make sense to him. Therefore, you must consider
the other party’s thought process. Does he know something you don’t?
Why weren’t the people at TKT saying to themselves, “We don’t have
a prayer of winning this lawsuit.”
Here’s my guess as to what was going through the minds of TKT
officials: if you estimate that defending yourself in a lawsuit like this
would cost about $20 million to $25 million but a win would net
you more than $1 billion, it’s worth doing even if the likelihood of
prevailing is only 5 percent. Think about it: If money weren’t an ob¬
ject, wouldn’t you bet $20 million for a possible payout of $1 billion,
with your chances being 1 in 20? In fact, it wouldn’t be farfetched to
say that successfully capsizing Amgen’s Epogen patents could be
worth as much as $5 billion to TKT. So even at 1-in-100 odds, its a
reasonable gamble.
Other factors probably contributed to TKT’s decision to forge
ahead: money wasn’t an object, because wealthy partner Aventis was
footing the legal bill. And in the short term, at least, the publicity from
the trial coverage would undoubtedly boost the firm’s stock. That is
exactly what happened in June, when Judge Young let TKT/ Aventis
274 Science Lessons

off the hook on patent number 5,618,698, probably the weakest of

the four. In essence, he’d been persuaded that TKT s technique for
producing EPO was sufficiendy different from Amgen’s. At the same
time, he denied the defendant’s motion for noninffingement of
patent number 5,547,[Link] Therapies had hinted to Wall
Street that a win over Amgen would free it to produce knock-offs
of nearly every genetically engineered medication. So the victory,
though slight, sent tremors through the stock exchange. Our price
dipped by roughly 5 percent, and our adversary’s leaped 35 percent.
The mood of this trial contrasted sharply with the atmosphere at
the Genetics Institute/Chugai arbitration hearing. At that one, GI’s
president, Gabe Schmergel, and I frequently walked to and from the
courthouse together, chatting amiably. Although each of us disagreed
with the other’s legal position, we remained friends and continued to
have great respect for each other’s company.
However, we had nothing but contempt for Transkaryotic Thera¬
pies, and the animosity pervaded the courtroom. During this time,
my TKT counterpart and I both testified at a congressional hearing.
At the end, as everyone was gathering papers and other belongings, I
said icily to him,“I don’t understand why you would want to be the
CEO of a company whose entire research effort is devoted to
knocking off other companies’ patents.” I’m not proud of the re¬
mark, which was out of character for me, but that’s how irritated I
was by the whole business. He angrily denied that was the case, and
we both stalked away.
The Amgen-TKT trial concluded on September 8, four days
after my sixty-fifth birthday. Judge Young had advised both sides that
his decision would take several months. Shortly after I left Amgen for
good on the first of the year, the court handed down a 244-page rul¬
ing. Transkaryotic Therapies, which needed to bat 1,000, lost two of
the three remaining patent disputes. The court found the company
guilty of infringing all three claims related to patent number
5,621,080 and four out of five claims of patent number 5,756,349.
 My Failed Retirement 275

“In order to avoid infringing a product claim,” the judge wrote,

“a competitor must not make that product regardless of whether the
process used to do so differs in some way from the process or
processes described in the patent. If indeed the same product is ulti¬
mately obtained, it matters not that in order to do so the competitor
tweaked the process in some matter.” In other words, you cannot
skirt a product patent simply by using a different method to make it.
The ruling, later upheld on appeal, was widely regarded as good not
only for Amgen but also for the biotech industry at large.

MY LEGACY

When assessing the legacy of any business leader and assigning credit
or blame, you must take into account the company he took over as
well as the one he left behind. That’s especially true in the biotech¬
nology field, with its lead time of many years. I’m proud of every¬
thing that Amgen accomplished under my stewardship—and of my
earlier contributions as CFO—but as for summing up my twelve
years as chief executive officer, I prefer to let some of the people who
worked under me have a crack at it.

Dennis Fenton (at Amgen from 1981 to present): To follow the

founder is incredibly hard, no matter where you are. George
Rathmann was beloved. Gordon stepped into the CEO job
and did it right. He didn’t declare wholesale changes. He’s a
very patient and thoughtful person. He knew where he
wanted to go, and he didn’t mind taking his time to get there.
So you’ve got to give Gordon a lot of credit for that. His per¬
sonality was really what the company needed at the time, too.

Kevin Sharer (at Amgen from 1992 to present): Although I got to

Amgen a few years after Gordon become CEO, he encoun¬
tered three gigantic obstacles that he very skillfully overcame.
First of all, he was put in a shotgun marriage with Harry
276 Science Lessons

Hixon. Number two, he followed George Rathmann, this leg¬

endary guy who everybody loves. And third, it couldn’t have
been easy having George hanging around as company chair¬
man. That’s a tough situation for anybody. To Gordon’s enor¬
mous credit, he won over the company, he worked with Hixon
long enough to become the unquestioned number one, and he
delivered the results. He did a magnificent job. Just magnificent.
Let me tell you what Gordon is really good at: He’s a good
decision maker, deliberate and thoughtful, never rash. There
was no question that as the leader of Amgen, he was fully in¬
vested in the mission of the company. It was always about
Amgen, not Gordon as some kind of personality.
He was willing to take risks. He would bet the company if
he had to and was enormously courageous in doing that.
Finally, he was probably at his best when things were at
their worst. He was a leader who, when things got difficult, he
would get constructive, not destructive. Thats hard to do.

Ed Garnett (at Amgen from 1986 to 2002): Gordon had huge

shoes to fill, because George Rathmann is a legend. But so is
Gordon, although in his typically humble way he’d be the first
to deny it. If you look at the history of any organization, there’s
usually the founder, followed by the builder, and then the
changer. That certainly is true of Amgen, with George being
the founder, Gordon the builder, and Kevin the changer. Enor¬
mous growth took place on his watch. Would it have happened
under someone else? Probably, but not to the same extent, and
it wouldn’t have been the same organization.

MY FAILED RETIREMENT

My wife, who has been a crucial source of encouragement through¬

out my career, and whom I can always count on for a frank opinion,
 My Failed Retirement 277

told me flat out that I would flunk retirement. My sons seconded

their mother’s prediction. I knew it myself but wasn’t sure what my
next step should be.
Adele and I attended Amgen’s annual meeting and then took a
long vacation aboard a private boat that sailed around Majorca and
Menorca, a pair of enchanting islands off the southeastern coast of
Spain in the deep-blue waters of the Mediterranean, another of
Adele’s great ideas. For three weeks, we did absolutely nothing, and it
was marvelous and strange at the same time. (Let me qualify that:
every morning the boat’s three-man crew deposited me ashore so
that I could go jogging.)
Other than that, we just sailed around. Sometimes we’d fire up
the engines, depending on the wind. It was relaxing, a perfect way to
unwind after many years of hard work. I’d never taken a vacation that
long in my life, and I highly recommend it to all retirees. The trip
also gave me time to think about what I might want to do next.
After eighteen years, it seemed sensible to remain in biotech¬
nology in some form and not have to learn a new industry at age
sixty-five. However, becoming chief executive of another biotech
firm—or any company—was definitely not on my list.
To get a biotech firm off the ground requires a triumvirate of en¬
trepreneurs, private investors, and venture capitalists. I liked the idea
of moving to the venture capital side. Joining an existing VC firm
wasn’t an option: none of any significance as far as biotechnology was
concerned existed in the Los Angeles area, which had perennially
trailed Boston, San Francisco, and San Diego as a hub of biomedical
activity. The answer, therefore, was to start one in this under-served
part of the country. In 2001, the year I founded Coastview Capital,
L.A. County start-ups in the fields of biotechnology, medical devices,
and health services took in $65 million in venture capital funding,
compared with $681 million for new firms in San Diego.
Coastview invested in biotechnology firms at all stages of devel¬
opment. What we looked for in a prospective start-up calls to mind
278 Science Lessons

Amgen in its early years: innovative, groundbreaking research; well-

qualified management, skilled both in business and in science; and
ambitions for products or technologies that could be expected to af¬
fect patients’ lives significantly.
Given those criteria, most business proposals fall short, especially
on the science end. Sometimes a field just isn’t ready. For example,
gene therapy is an exciting concept, but it is so far from becoming a re¬
ality commercially that I don’t think venture capitalists should invest in
it now. Other liabilities, from a VC perspective, include the following:

• The science is too old or has been attempted without success

in the past.

• A comparable existing therapy is so effective that it probably

couldn’t be improved to any meaningful degree.

• The potential product’s market is too limited to justify the ex¬

pense of research and development.

• The idea is off-the-wall crazy.

Examples of each have crossed my desk at one time or another.

Sometimes you can judge the caliber of the science by the creden¬
tials of the people involved. The likelihood that third-rate scientists
will produce something of significance wasn’t something we were
willing to bet on. Similarly, we scrutinized the track records of a new
company’s managers. Where did they go to school? Who was their
thesis adviser, and how well respected? How successful were their
previous business ventures? And so on.
In 2002, in the aftermath of the 9/11 terrorist attacks, I started
my jet aircraft charter company, Prime Jet. It is extremely successful.
One reason is that it operates on the same values and principles that
underpinned Amgen’s success, demonstrating their [Link]
worked at Amgen. They are working at Prime Jet. And they can work
for you and your company too, no matter what business you’re in.
Acknowledgments

Gordon Binder: Thanks to my wife, Adele, for your love and for mak¬
ing our marriage such a happy [Link] also for your wise counsel
and unwavering support throughout my business life. Without you, this
book wouldn’t exist.

Thanks also to my sons, Brant and Todd, who are a great source of
pride to me and my wife. Both have MBAs from leading universities,
and I am proud to say that our relationship has reached the crossover
point, with my sons now providing more counsel and advice to me
than I am able to give them.

Thanks, too, to my invaluable assistant Lorraine Shulman.

Finally, I am grateful to Adele, Brant, and Todd for accepting without

complaint the long hours, extensive travel, and disruptive job changes
that a successful business career requires.

Philip Bashe:Thanks to Robert and Rachelle Bashe, the late Evelyn

Bashe, Justin Bashe, and, as always, Patty Romanowski Bashe, for
everything.

Together we’d like to thank our talented editor at Harvard Business

School Press, Jacqueline Murphy, for her expert guidance and support;
editorial director Hollis Heimbouch; associate editor Astrid Sandoval;
280 Acknowledgments

copy editor Betsy Hardinger; senior production editor Marcy Barnes-

Henrie;jacket designer Stephani Finks;jacket copy writer Laurie
Ardito; and literary agent James Levine and everyone else at Levine-
Greenberg Literary Agency, Inc.

Thanks also go to the following, at Amgen, Inc., and elsewhere, for shar¬
ing their memories and insights: alphabetically, they are Kirby Alton,
Burt D. Ensley, Dennis Fenton, Frederick Frank, Edward F. Garnett,
Ted Ledder, George Rathmann, Kevin Sharer, Daniel Vapnek, and
Bob Weist.

Finally, very special thanks to Peter Teeley for suggesting that the two
of us collaborate on a book. Good advice!

A NOTE ON QUOTED MATERIAL

Most of the people quoted in Science Lessons kindly agreed to share

their perspectives in interviews that were conducted in California or
over the telephone in late 2004 and early [Link] again to my
fellow Amgen CEOs George Rathmann and Kevin Sharer, as well as
Dennis Fenton, Kirby Alton, Ed Garnett, Burt Ensley, Ted Ledder,
Bob Weist, and Fred Frank.
Other quotations have been reconstructed as faithfully as possible
from memory: things that were either said directly to me or recounted
to me by another source. Any errors contained in these quotes are my
own.
Index

Abbott Laboratories, 19,24,44—45,71, capital, success in raising, 25—26,

77,126 48-50
acquisitions, 123—124, 159,160 collaboration with FDA, 263—264
Adamson, John, 99 compensation philosophy, 227—229
adenine, 80 conception of, 17—18
[Link], 18—19 concerns about a takeover, 165—167
ALS (amyotrophic lateral sclerosis), 161 criteria for choosing principal inves¬
Alton, Kirby tigators, 96
on Binder/Hixon joint control, 110 employees and (see employees)
first impressions of Amgen, 27 Epogen and (see Epogen)
involvement in recruitment, 145 ethics in business and (see values and
reaction to FDA’s testing criteria, 99 ethics in business)
reaction to Rathmann’s retirement federal regulatory lobbying, 261—263
announcement, 108 financials (1988—2000), 7t
on scientists’ dedication, 85 focus on recruiting scientists, 57—58
team work at Amgen and, 192,210, hiring of CFO, 28 (see also Binder,
213,223 Gordon)
Amgen hiring of first CEO, 17,19—21 (see
A round funding total, 25—26 also Rathmann, George)
animal therapeutics attempts, 68—70, initial corporate investors, 24-25
90 IPO (see Amgen IPO)
anti-doping efforts, 259—260 legacy, 7—8
basis of success, 14 lessons learned from partnerships,
burn rate of early funding, 28 135
business model (see science-based lessons learned while dealing with
business model) Washington, 263—265
business strategy in mid-1990s, maintaining the energy of a start-up,
164-165,166 248
282 Index

Amgen, continued final proceeds, 54—55

multifaceted product portfolio ap¬ proposal, 36—37
proach, 67 rationale for accelerated schedule,
NASDAQ listing, 140 41-42
Neupogen and (see Neupogen) readiness for, 43—44
patent-infringement suit from J&J, road shows, 46—48,50—51
132-136,234-235 success in raising capital, 6—7,48—50
patent-infringement suit with GI (see valuation, 53
patent-infringement suit from Amgen Japan, 155
GI) amyotrophic lateral sclerosis (ALS), 161
patent-infringement suit with TKT, anakinra (Kineret), 159—161
271-275 analysts’ compensation, 251
portfolio expansion through acquisi¬ ancestium (Stemgen), 156—157
tions, 123—124,159,160 anemia, 5,6,48,66,94
portfolio expansion through licens¬ animal therapeutics, 68—70,90
ing, 157,159 anti-doping efforts, 259—260
portfolio review process, 165 Antril, 159
R&D (see R&D at Amgen) apheresis, 156
Rathmann’s retirement, 107—108, Apollo program, 247
109,140-141 Applied Molecular Genetics. See
revenue in 1996,153 Amgen
search for partnerships (see business Aranesp (darbepoetin)
partnerships) business basis for pursuing, 170—173
search for second round funding, results of lawsuit with J&J, 174—176
32-34 arbitration versus lawsuits, 134,136
specialty-chemicals attempts, 70—71, Arbor Acres, 68
90 armed services as a source of talent,
staff building, 26—27 188-189
stock price, 229 Ash, Roy L., 257,267
support of change and innovation, Astrellas Pharma. See Yamanouchi
244-247 Pharmaceutical Company
threat of layoffs in 1982,34 autoimmune disorders, 160
transition to Rathmann’s exit, Aventis, 272
109-110
values importance (see values and baby boomers as a source of talent, 188
ethics in business) Baddour, Raymond, 18,44
VC funding, 22—23,25 bases, 80
view of itself and its place, 267 Bayh-Dole Act, 117
Wall Street’s expectations in mid- BDNF (brain-derived neurotropic fac¬
1990s, 153-154 tor), 161
Amgen IPO Beard, Thomas, 58
Abbott lawsuit threat, 45-46 Bendix, 2
date, 53 Bennis, Warren, 231
 Index 283

Binder, Adele, 2,3,4,223,224,226, biotechnology

276-277 capital burn rate, 27—30
Binder, Brant, 271 cost of bringing a drug to market,
Binder, Charles, 253—254 28-30
Binder, Gordon drug approval process (see drug ap¬
Amgen’s financial health at his exit, proval process)
6,271 drug development phases, 30—31
belief in few hierarchies, 216—219 favorable economic chmate in U.S.,
on counseling subordinates, 236—237 22
decision to join Amgen, 3—4 first drug produced using, 84
decision to press with Epogen modi¬ industry status in mid-1990s, 164
fication, 170—173 legislation to protect patents,
faith in intentions of federal workers, 261-262
266-267 new FDA user’s fees impact, 262—263
father’s influence on, 253—254 past IPOs, 35,36,37
hiring by Amgen, 28 patent protection for, 117
hiring by Litton, 255 pharmaceutical R&D challenges, 32
on instituting change, 112 recombinant DNA, 47,82—84
introduction to biotech, 1 recombinant drugs, 66
on leaders as teachers and mentors, risks in drug development, 31,32
195-196 science-driven basis of research,
long-term plan for Aranesp, 175—176 65-67
performance reviews, 240 tempo of progress, 89—90
post-retirement activities, 276—278 Biotechnology Industry Organization
pre-Amgen education and jobs, (BIO), 114,262
108-109 Biotechnology Patient Protection Act,
reflective comments from co-workers, 261-262
275-276 BLA (biologic license application), 104,
relationship with employees, 262
234-236 Blair, Jim, 33,44,67
on retirement transition, 270—271 bone marrow transplantation (BMT),
selection as CEO, 107—108,141 156
tendency to micromanage, 212 bovine growth hormone project, 69—70
TKT trial reaction, 274 Bowes,William [Link]., 17—18,19—20,25,
work history, 1—3 33,39
on working with Hixon, 110 Boyer, Herbert, 83,84
Binder, Rodella, 253 brain-derived neurotropic factor
Binder, Todd, 271 (BDNF), 161
BIO (Biotechnology Industry Organi¬ Bright Horizons Family Solutions, 226
zation), 114,262 Bristol-Myers Squibb, 95
Biogen, 19,37,67,76,77 Browne, Jeff, 169
biologic license application (BLA), 104, Bureau of Chemistry. See Food and

262 Drug Administration

284 Index

burn rate in biotechnology, 27-30 codons, 80,82

Burroughs Corporation, 2 Cohen, Stanley, 83,84
business partnerships colorectal cancer treatment, 95
fiscal plan, 126 V company philosophies
Immunex and, 123-124 communication principles, 237—238
Kirin Brewery Co. and, 128-131, on compensation, 227—229
140 culture and (see culture at Amgen)
lack of interested companies, 126 sharing research results, 232—233
overseas, 126—128 staying connected to employees,
strategy for geographic growth, 124 233-236
telling employees the truth, 234,236
capital-gains tax rate, 22 values and (see values and ethics in
Caruthers, Marvin, 49 business)
Centocor, 159 compensation philosophy, 227—229,
Cetus Corporation, 35,36 250-251
cetuximab (Erbitux),95 Crick, Francis, 78,79
Chakrabarty, Ananda, 117 culture at Amgen
chicken growth hormone project, 68 business ethics and, 258—260
child-care facility, 226 creating a culture of ethics, 253—254
Chinese hamster ovary cells (CHO), dedication of employees, 10,13,104,
87,120 137-138,201-202,227-229,
Chiron, 55 243-244
chromosomes, 79 fostering respect and trust, 214—216
Chugai Pharmaceutical Co., 112,116, hierarchy control, 216—219
144 living the company values, 242—244
cinacalcet (Sensipar), 157—159 values-based approach to business,
Cisco Systems, 187 9-13
clinical trials view of itself and its place, 267
cost of drug phases, 89—90 cytosine, 80
drug approval process (see drug ap¬
proval process) Dale, David, 94
Epogen phase I trial, 93 darbepoetin (Aranesp)
Epogen phase II trial, 94,100—101 business basis for pursuing, 170-173
Epogen phase III trial results, 103 lawsuit from Johnson & Johnson,
fraud in U.S., 97 174-176
the Internet and patient dropout Dawson, Paul, 58,150,167,215,223
rate, 103 detergent enzymes, 71
ofNeupogen, 144—145,146t diabetes and treatments, 83—84
occasional inadvertent findings, directors’ compensation, 251
162-163 Dow Chemical Company, 85
phases of, 30—31,101—102 Downing, Mike, 104
typical study design, 62-63 down rounds, 23
Coastview Capital, 277-278 Drug Amendments Act (1962), 92
 Index 285

drug approval process. See also clinical offering praise, time, and attention,
trials 230-232
choice of principal investigators, on-site amenities, 225—226
96-98 rewards from recruiting the best
consequences of a poorly designed people, 78
trial example, 95—96 Employee Retirement Income Security
drug development phases, 30—31 Act, 22
events contributing to passage of leg¬ employees
islation, 91-92 company philosophies (see company
FDA’s restrictions on Epogen trials, philosophies)
98-99 culture at Amgen and (see culture at
phase III components, 101—102 Amgen)
prechnical studies, 93 dedication to company and mission,
structure and needed approvals, 94 10,13,104,137-138,201-202,
Drug Efficacy Study, 92 243-244
executives willingness to learn new
EGF (epidermal growth factor), 162 culture, 130
Egrie,Joan, 169 hiring process (see hiring at Amgen)
e-learning, 222 performance reviews (see perform¬
Elliott, Steve, 169 ance reviews)
employee retention. See also employees retention of (see employee retention)
alternative work schedules and, reviews to identify promising em¬
226-227 ployees, 219—220
Amgen turnover rate, 196—197,224 rewards from recruiting the best
appointment of Fenton, 59—60 people, 78
benefits offered, 221—222 team building process (see hiring at
company gatherings, 222—224 Amgen)
company philosophies (see company Enbrel (etanercept), 123-124,160
philosophies) end-stage renal disease (ESRD), 5,66,
compensation and, 227-229 98-99
cost of losing a worker, 196 Ensley, Burt, 68-69,70,167,222,260
education allowances, 222 environment at Amgen
eliminating negatives versus adding allowance for errors, 203
incentives, 197—199 atmosphere of trust and respect,
employee growth triangle, 220 214-216
environment at Amgen (see environ¬ avoiding micro managers, 210—212
ment at Amgen) commitment to team work,
historical rate, 224,227 204-206,213
labor pool in early 2000s, 196 discouragement of intracompany
leaders as teachers and mentors, competition, 212—214
195-196 encouragement of free-ranging per¬
list of what employees want, spective, 209—210
199-200 meeting discipline, 206-207,213
286 Index

environment at Amgen, continued Eschbach, Joseph, 99

philosophy of collaboration, 200-202 ESRD (end-stage renal disease), 5,66,
training as a business investment, 98-99
221-222 etanercept (Enbrel), 123—124,160
unnecessary hierarchies elimination, ethics in business. See values and ethics
216-219 in business
value of collaboration, 208—209 executive compensation, 250-251
epidermal growth factor (EGF), 162
Epogen (epoetin alfa) FDA. See Food and Drug
Abbott’s dismissal of, 126 Administration
anti-doping efforts by Amgen, Fenton, Dennis
259-260 belief in science-based business
court’s upholding of patent, 140 model, 60,61,64
decision to do an infant clinical trial, on Binder’s tenure, 275
267-268 on the culture at Amgen, 203
drug approval process (see drug ap¬ EPO trial documentation work,
proval process) 104
early sales results, 139 first impressions of Amgen, 248
efficacy measure, 98 on Rathmann’s retirement, 109
EPO gene identification, 72—73, on team work, 204
76-78 on tensions between scientists and
FDA’s restrictions on trials, 98—99 sales force, 167
launch of, 4-6,136—139 work as sales and marketing head,
lawsuits from Johnson & Johnson, 59-62
132-136,174—176,234-235 Fermenta, 125
modification proposal, 169—173 filgrastim, 5—6,107,144. See also
original consumer base, 66 Neupogen
patent-infringement suit (see patent- firm-commitment agreement, 54
infringement suit from GI) Food, Drug, and Cosmetic Act (1938),
phase I trial, 93 91
phase II trial, 94,100—101 Food and Drug Administration (FDA),
phase III trial results, 103 91
purpose, 48 collaboration with Amgen, 263—264
synthetic EPO success, 5 drug-approval process (see drug ap¬
trial documentation preparation, 104 proval process)
Erbitux (cetuximab), 95 Modernization Act, 263,265
erythropoietin (EPO), 4—5 Neupogen approval, 146
cloning of the isolated gene, 87 restrictions on Epogen trials, 98—99
efforts to secure more lab samples, user’s fees proposal, 262—263
85-87 Food and Drugs Act (1906), 91
push to discover EPO gene, 72-73, Ford Motor Company, 213—214
76-78 Frank, Frederick, 33,34,39,43,49,52,
synthetic EPO (see Epogen) 53
 Index 287

Garnett, Ed Gray, Harry J„ 193,255-257

on being counseled by Binder, guanine, 80
236-237
on Binder’s tenure, 276 Harris, Sydney, 118
culture of ethics and, 258—259 Harte,William J., 175
environment at Amgen, 225,233—234 Hatch, Orrin, 261
on Gordon’s micromanagement ten¬ hematocrit test, 98
dencies, 212 hepatitis C, 154,155
performance reviews for Binder, 240 hiring at Amgen
on resignations, 241—242 baby boomers as a source of talent,
retirement talk, 270 188
team work at Amgen and, 192—193, choosing talent over experience,
210 184-185
on values and interviews, 182 hiring from within, 191—193
G.D. Searle, 150 hiring pool, 186
Genentech honesty about the job, 180-181
competition with Amgen, 24 job descriptions and, 185—186
employee gatherings, 223 job interview process, 178—180
Humulin launch, 85 military as a source of talent,
industry leadership, 32 188-189
IPO success, 25 transitioning efforts for new employ¬
Japanese collaborations, 128 ees, 189—191
multifaceted product portfolio ap¬ using employees as recruiters,
proach, 67 186-187
sales and marketing and, 150 values used in hiring criteria,
specialty-chemicals market and, 71 181-185
use of CHO, 87 HIV virus, 140
genes, 79 Hixon, Harry, 109,110,141
genetic engineering, 4 Hoechst Marion Roussel, 127,272. See
codons, 80,82 also Aventis
DNA, 79-80,81f Hotz, Bob, 34,43,49,51
human genome, 78—79 Huddleson, Ed, 3,254
Genetics Institute (GI) human genome, 82
EPO gene and, 76 Humulin, 36,84, 85
Neumega and, 161 hyperparathyroidism, secondary, 157,
patent-infringement suit (see patent- 158
infringement suit from GI) hypothalamic amenorrhea, 163
Genetic Systems Corporation, 36
Goldwasser, Eugene, 76—78,85,86 IBM, 119,221-222
goodwill, 243—244,260,265 ImClone, 95—96
granulocyte colony-stimulating factor Immunex, 55,123—124,160
(G-CSF), 142,143. See also Immunomedics, 55
Neupogen indigo dye, 70—71
288 Index

Infergen (interferon alfacon-1), 154—155 Ledder,Ted, 228-229

institutional review board (IRB), 94 leptin (weight-loss biologic), 163
insulin, 83—84 Lever Brothers, 71
intellectual property, 112—115. See also ^ licensing, 132,157,159
patents for drugs Lin, Chi-Hwei, 85
interleukin-11 (Neumega), 161 Lin, Fu-Kuen, 72,75,85,86,87
Intron A, 155 hpodystrophy, 163—164
investigational new drug application Litton Industries, 119,255—256
(IND), 93 lobbying, 263—265
IPOs (initial public offerings) Lou Gehrig’s disease, 161
Amgen’s (see Amgen IPO) Ludwig Institute for Cancer Research,
basis of success, 49 144,145
during dot-com peak, 40-41,55
Genentech’s success, 25 Mace, Myles, 254—255
importance of timing, 41—42 Marogen, 140
preparation work, 43—44 M. D. Anderson Cancer Center, 151
scheduling challenges, 50—51 “The Memo,” 168-169
valuation considerations, 52—53 Memorial Sloan-Kettering Cancer
workings of, 35—36 Center, 142,143,144
IRB (institutional review board), 94 MGDF (megakaryocyte growth and
development factor), 161
JAMA, 97 micromanagement avoidance, 210—212
Jeffords, Jim, 262 military as a source of talent, 188—189
Johnson, Franklin P.“Pitch”Jr., 18,25, Miyake,Takaji, 77
33 Morstyn, George, 145
Johnson & Johnson Mueller, George, 245—247,246
arbitration results, 234—235
contentions over Aranesp, 174—176 naphthalene, 70
lawsuits against Amgen, 132—136 NASDAQ, 140
sales and marketing contentions National Aeronautics and Space Ad¬
with, 150 ministration (NASA), 246
National Medal ofTechnology, 235
Kelsey, Frances, 92 National Research Council, 92
Kennedy, Edward, 261,266 NDA (new drug apphcation), 31,104,
kidney dialysis, 98 262
Kineret (anakinra), 159-161 Neulasta (pegfilgrastim), 173-174,176
Kirin Brewery Co., 128-131,140 Neumega (interleukin-11), 161
Kodak, 71 Neupogen (filgrastim)
challenges in selling, 147—148
lawsuit from Genetics Institute. See clinical benefits, 148
patent-infringement suit from GI FDA approval, 146
lawsuits from Johnson & Johnson, human trials, 107,144-145,146t
132-136,174-176,234-235 launch of, 4—6
 Index 289

licensing and patent positioning, 144 judges decision, 123

research into a long-lasting version, legal date of invention, 121
173-174 patent process in U.S., 112-115
research on G-CSF, 142-144 relevant EPO patents, 114
sales results, 147 start of trial, 120
neutropenia, 142,143 strategies typically used, 116
neutrophils, 142 patents for drugs
New Court Partners, 25 exclusivity for orphan disease re¬
new drug application (NDA), 31,104, search, 66—67
262 grounds for contesting a patent,
New York Yankees, 14—15 118-119,272-273
nonresponders in a trial, 103 legislation to protect patents, 261—262
NPS Pharmaceuticals, 158 patent protection, 117
nucleus, 79 process in U.S., 112-115
strategies typically used to challenge,
Odre, Steve, 121,262 116
onboarding, 190 peer interviewing, 179
orphan diseases, 66 pegfilgrastim (Neulasta), 173—174,176
Orphan Drug Act (1983), 36,66,67, pegylation, 173
113,140 performance reviews
Ortho Pharmaceuticals, 104,132 for executives, 240
to identify promising employees,
Paine Webber, 162 219-220
parade donation to Thousand Oaks, 243 including values in, 238—239
parathyroid hormone (PTH), 158 resignations and, 241-242
Parke-Davis, 77 underachievers and, 239,241—242
partnerships. See business partnerships peritoneal dialysis (PD), 170
Patent and Trademark Act Amendments [Link], 40
(1980), 117 pharmaceutical industry
patent-infringement suit from GI market-driven basis of research, 65—67
attempt to overturn Amgen’s patent R&D challenges, 32
rights, 140 sales and marketing (see pharmaceu¬
awarding of GI’s patent, 115—116 tical sales)
corporate espionage incident, Pharmaceutical Research and Manu¬
122-123 facturers of America (PhRMA),
GI’s countersuit, 121 29,218,263
GI’s plan to use a foreign company pharmaceutical sales
loophole, 116,118 focus on educating practitioners,
grounds for contesting a patent, 148-149
118- 119,272-273 hiring of reps at Amgen, 149—150
impact on stock price, 122 process of, 60-61
inventor/society covenant of patents, typical sales versus manufacturing
119- 120 tensions, 168
290 Index

plasmids, 83 search for partnerships, 126—128

platelets, 161 search for second round funding,
porcine growth hormone project, 68—69 32-34
prescription drug benefit legislation, search forVC funding, 22—23,25
266 securing of corporate investors,
Prescription Drug User Fee Act (1992), 24-25
262,263 on success, 135
Prime Jet, 185,187,198-199,221,278 support of his scientists, 34,72—73,
Procrit, 132,133,140 86,231-232
Procter & Gamble, 71,168,244—245 Rathmann, Joy, 20
product-development teams, 208 recombinant DNA technology
proteins, 80 Epogen and (see Epogen)
Pseudomonas, 70 explained, 47
PTH (parathyroid hormone), 158 first recombinant gene creation, 83
Public law 97-414,66-67 recombinant drugs, 66
targeted gene isolation method,
R&D at Amgen. See also Epogen; 83-84
Neupogen therapeutic protein manufacture
abandoned projects, 161—164 process, 82
funding in early 1990s, 153 Regeneron Pharmaceuticals, 161
Infergen, 154—155 reverse stock split, 53
Kineret, 159—161 rheumatoid arthritis (RA), 160
mid-1990s pipeline drugs, 154—161 Ribi Immunochem, 36
proposal to modify Epogen, 169—173 Rockefeller University, 163
Sensipar, 157—159
small-molecule research, 157-159 sales and marketing
Stemgen, 156—157 annual family sales meeting, 223—224
RA (rheumatoid arthritis), 160 challenges in selling Neupogen,
Raab, Kirk, 24, 33,44,46 147- 148
Rathmann, George, 5 focus on educating practitioners,
credibility of, 49 148- 149
hiring by Amgen, 17,19-21,109 gaining of respect from scientists,
job offer to Gordon, 3 215-216
on maintaining the energy of a start¬ hiring of sales reps, 149—150
up, 248 loss of key personnel, 58—59
passion for Amgen, 27 sales force establishment, 150—151
persona, 21 scientific-method applied to sales
post-Amgen activities, 110—111 trial, 62—63
promotion of a family atmosphere, scope of services, 151—152
223 shift to one-product per rep, 60—62
retirement, 140-141 tensions between scientists and,
retirement announcement, 107—108, 167-169
109 Salser,Winston A., 18,19
 Index 291

Sarbanes-Oxley Act (2002), 43,44 stop codons, 82

Sarris, Patti, 120 Strickland, Tom, 169
Saturn V, 246 sulfanilamide, 91
el-Sayed, Refaat, 125 suppliers support from Amgen, 230
Schering-Plough, 155 Synergen Inc., 159,166
Schmergel, Gabe, 118,274 System Development Corporation
Schoellhorn, Bob, 126 (SDC), 1,41,193,245-247
science-based business model
benefits of objective analysis, 64—65 Teeley, Peter B., 261,263
companies’ tendency to ignore con¬ thalidomide, 91—92
tradictory data, 65 [Link], 40
scientific-method applied to sales Thornton, Charles B.“Tex,” 255
trial, 62—63 3M, 21,115
secondary hyperparathyroidism, 157,158 TNF (tumor necrosis factor), 160
S.E. Massengill Company, 91 Tosco Corporation, 24—25,33
Sensipar (cinacalcet), 157—159 Tour de France, 259—260
sepsis, 159 Transkaryotic Therapies (TKT),
Sharer, Kevin, 109,188 271-275
on Amgen’s performance in mid- Tufts University, 89, 93
1990s, 166 tumor necrosis factor (TNF), 160
on Binder’s tenure, 275—276
decision to press with Epogen modi¬ underwriters, 53,54
fication, 170—173 United Geophysical Corporation
hiring by Amgen, 269—270 (UGC), 2
on instituting change, 111—112 Universal Code of Military Justice, 252
pride in Amgen, 242 University of Alabama Medical Center,
relationship with Binder, 234 103
reviews to identify promising em¬ University ofWashington Medical
ployees, 219 Center, 94
Shaw, George Bernard, 237 Upjohn Company, 69—70
Simi Valley Hostages, 104 U.S. Department of Labor. 196
Smith Barney, 54 U.S. Patent and Trademark Office
SmithKline, 68 (USPTO), 112-115
Souza, Lawrence, 142,143 U.S. Public Law 105-115,263
specialty-chemicals, 70—71,90
Standard Oil Company, 216 values and ethics in business
Stark, Pete, 264 compensation policies and, 250—251
Stebbing, Noel, 34,73,126 credibility of GAAP and, 251—252
stem-cell rescue, 156 culture of ethics at Amgen, 253—254,

stem-cell research, 143 258-260

Stemgen (ancestium), 156—157 dishonesty in companies, 257—258

Stewart, Martha, 96 effects of regulatory oversights,

stock options, 228—229 252-253

292 Index

values and ethics in business, continued Walker Information, 196

employee retention and, 242—244 Watkins, Harold Cole, 91
lessons learned from partnerships, 135 Watson, James, 78,79
values-based approach to business, Watson Wyatt Worldwide, 258
9-13,242-244,267-268 Waxman, Henry, 66,264,266
values described, 11—13 weight-loss biologic (leptin), 163
values used in hiring criteria, 181—185 Weist, Robert, 24,32,34,114,115,132,
view of itself and its place, 267 174
Vapnek, Dan, 34,143 Whitcome, Phil, 126
desire to discontinue EPO project, William S. Merrill Company, 91—92
73,85 Work in America Institute, 201
dislike of Epogen modification plan, Wynn, Keenan, 256
170
failure to see a need for sales division, Xoma, 159
215
infant trial recommendation, 267 Yamanouchi Pharmaceutical Company,
science-based business model belief, 62 155
search for partnerships, 126 Young,William, 272
venture capital (VC) funding
Amgen’s A round funding total, 25—26 zidovudine (ZDV), 140
mode of operation, 25
removal of congressional barriers,
22-23
About the Authors

Gordon M. Binder is the managing director of Coastview Capi¬

tal, LLC, a Los Angeles venture capital firm he founded in 2001.
From 1982 to 1988, he served as chief financial officer of Amgen, the
world’s largest and most successful biotechnology company, and then
became its chief executive officer (1988 through 2000) and chair¬
man of the board (1990 through 2000) until his retirement. He is
also the founder of Prime Jet, a jet-charter company.
Binder earned a bachelor’s degree in electrical engineering from
Purdue University in 1957 and an MBA from Harvard Business
School in 1962, where he was a Baker Scholar. From 1957 to 1960, he
served aboard the USS Intrepid as a U.S. Navy officer. He and his wife
Adele reside in the Holmby Hills section of Los Angeles. They are the
parents of two grown sons, Brant and Todd.

Philip Bashe has written or cowritten seventeen titles, spanning

several genres, including health and self-help, parenting, sports, biog¬
raphy, autobiography, and popular culture. He lives in Baldwin, New
York, with his wife author Patricia Romanowski Bashe and their
son Justin.
.
 • UBRAHy

NMH Amazing Research.

Amazing Help.

[Link]

10 Center Drive
Bethesda, MD 20892-1150
301-496-1080
 NIH Bf

3 1496 00995 1040