The n e w e ng l a n d j o u r na l of m e dic i n e

original article

Infliximab, Azathioprine, or Combination

Therapy for Crohn’s Disease
Jean Frédéric Colombel, M.D., William J. Sandborn, M.D., Walter Reinisch, M.D.,
Gerassimos J. Mantzaris, M.D., Ph.D., Asher Kornbluth, M.D.,
Daniel Rachmilewitz, M.D., Simon Lichtiger, M.D., Geert D’Haens, M.D., Ph.D.,
Robert H. Diamond, M.D., Delma L. Broussard, M.D., Kezhen L. Tang, Ph.D.,
C. Janneke van der Woude, M.D., Ph.D., and Paul Rutgeerts, M.D., Ph.D.,
for the SONIC Study Group*

A bs t r ac t

Background
The comparative efficacy and safety of infliximab and azathioprine therapy alone From Hôpital Claude Huriez and Centre
or in combination for Crohn’s disease are unknown. d’Investigation Clinique, Centre Hospita-
lier Universitaire de Lille, Université Lille
Nord de France, Lille, France (J.F.C.);
Methods Mayo Clinic, Rochester, MN (W.J.S.); Uni-
In this randomized, double-blind trial, we evaluated the efficacy of infliximab mono- versity Hospital Vienna, ­Vienna (W.R.);
Evangelismos Hospital, Athens (G.J.M.);
therapy, azathioprine monotherapy, and the two drugs combined in 508 adults with Mount Sinai Medical Center, New York
moderate-to-severe Crohn’s disease who had not undergone previous immunosup- (A.K., S.L.); Shaare Zedek Medical Cen-
pressive or biologic therapy. Patients were randomly assigned to receive an intrave- ter, Jerusalem (D.R.); Imelda Ziekenhuis,
Bonheiden (G.D.), and University Hospi-
nous infusion of 5 mg of infliximab per kilogram of body weight at weeks 0, 2, and tal Gasthuisberg, Leuven (P.R.) — both
6 and then every 8 weeks plus daily oral placebo capsules; 2.5 mg of oral azathio- in Belgium; Centocor Ortho Biotech,
prine per kilogram daily plus a placebo infusion on the standard schedule; or com- Malvern, PA (R.H.D., D.L.B., K.L.T.); and
Erasmus Medical Center, Rotterdam, the
bination therapy with the two drugs. Patients received study medication through Netherlands (C.J.W.). Address reprint re-
week 30 and could continue in a blinded study extension through week 50. quests to Dr. Colombel at Hôpital Claude
Huriez, Centre Hospitalier Universitaire de
Lille, Rue Michel Polonovski, Lille, 59037
Results France, or at ­jfcolombel@[Link]; or
Of the 169 patients receiving combination therapy, 96 (56.8%) were in corticoste­ to Dr. Sandborn at the Mayo Clinic, 200
roid-free clinical remission at week 26 (the primary end point), as compared with First St. SW, Rochester, MN 55905, or at
[Link]@[Link].
75 of 169 patients (44.4%) receiving infliximab alone (P = 0.02) and 51 of 170 pa-
tients (30.0%) receiving azathioprine alone (P<0.001 for the comparison with combi- Drs. Colombel, Sandborn, and Rutgeerts
nation therapy and P = 0.006 for the comparison with infliximab). Similar nu- contributed equally to this article.
merical trends were found at week 50. At week 26, mucosal healing had occurred *Members of the Study of Biologic and
in 47 of 107 patients (43.9%) receiving combination therapy, as compared with 28 Immunomodulator Naive Patients in
of 93 patients (30.1%) receiving infliximab (P = 0.06) and 18 of 109 patients (16.5%) Crohn’s Disease (SONIC) Study Group
are listed in the Supplementary Appen-
receiving azathioprine (P<0.001 for the comparison with combination therapy and dix, available with the full text of this
P = 0.02 for the comparison with infliximab). Serious infections developed in 3.9% of article at [Link].
patients in the combination-therapy group, 4.9% of those in the infliximab group,
N Engl J Med 2010;362:1383-95.
and 5.6% of those in the azathioprine group. Copyright © 2010 Massachusetts Medical Society.

Conclusions
Patients with moderate-to-severe Crohn’s disease who were treated with infliximab
plus azathioprine or infliximab monotherapy were more likely to have a cortico­
steroid-free clinical remission than those receiving azathioprine monotherapy.
([Link] number, NCT00094458.)

n engl j med 362;15  [Link]  april 15, 2010 1383

The New England Journal of Medicine
Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

C
rohn’s disease is a chronic inflam- and remission as a score of less than 150.) Pa-
matory disorder of the gastrointestinal tients were either corticosteroid-dependent (with
tract that is defined by relapsing and re- a CDAI score of at least 220 points after reduc-
mitting episodes, with progression over time to tion of the corticosteroid dose), were being con-
complications of stricture, fistulas, or abscesses.1 sidered for a second course of systemic cortico­
Symptoms of mild-to-moderate disease are treat- steroids within 12 months, or had not had a
ed with mesalamine, budesonide, or systemic response to at least 4 weeks of either mesala-
corticosteroids.2,3 The therapeutic benefit of cor- mine (at a dose of ≥2.4 g per day) or budesonide
ticosteroids is frequently offset by side effects of (at a dose of ≥6 mg per day). None of the pa-
prolonged exposure.4 In addition, systemic corti- tients had undergone previous treatment with
costeroids and budesonide are not effective for azathioprine, 6-mercaptopurine, methotrexate, or
maintenance therapy.5-7 Azathioprine and 6-mer- an anti-TNF biologic agent.
captopurine are frequently prescribed for patients Patients with the short bowel syndrome, an
in whom first-line therapies fail — in particular, ostomy, a symptomatic stricture, an abscess, a
those who are dependent on or do not have a re- recent history of abdominal surgery (within the
sponse to systemic corticosteroids.2,3,8 Approxi- previous 6 months), a history of tuberculosis or
mately 40% of patients treated with azathioprine other granulomatous infection, a positive chest
remain in remission at 1 year.9 Infliximab and radiograph or tuberculin skin test with purified
other monoclonal antibodies targeting tumor protein derivative, a recent history of an oppor-
necrosis factor (TNF) have shown efficacy in in- tunistic infection (within the previous 6 months),
ducing and maintaining remission in patients active infection with hepatitis B or C, infection
with Crohn’s disease.10-12 Treatment guidelines with the human immunodeficiency virus, multi-
generally recommend initiating treatment with ple sclerosis, cancer, or a homozygous mutant or
first-line agents, including mesalamine and sys- heterozygous thiopurine methyltransferase pheno­
temic corticosteroids, followed by azathioprine, type were not eligible (see Table 1 in the Supple-
with anti-TNF therapies reserved for patients in mentary Appendix, available with the full text of
whom conventional therapies have failed.2,3,8 In this article at [Link]).
a multicenter trial, we compared the efficacy of
infliximab, azathioprine, and the two drugs com- Study Treatments and Randomization
bined for inducing and maintaining corticoste­ Patients were randomly assigned to receive intra-
roid-free clinical remission in patients with ac- venous infusions of infliximab (Remicade, Cen-
tive Crohn’s disease. tocor Ortho Biotech) at a dose of 5 mg per kilo-
gram of body weight plus daily oral placebo
Me thods capsules, oral azathioprine capsules at a daily
dose of 2.5 mg per kilogram plus placebo infu-
Study Design and Patients sions, or combination therapy with infliximab
The Study of Biologic and Immunomodulator and azathioprine. Infusions were administered at
Naive Patients in Crohn’s Disease (SONIC) was weeks 0, 2, and 6, and then every 8 weeks. Patients
a randomized, double-blind, 30-week trial, with were followed through week 30, when they were
a 20-week extension in which blinding was main- given the option of continuing to receive their as-
tained. The trial was conducted at 92 centers signed therapy, with blinding maintained, in a 20-
from March 2005 through November 2008. The week extension trial, with follow-up through week
institutional review board at each center approved 50. A follow-up telephone interview was conduct-
the protocol, and all patients provided written ed to collect reports of serious adverse events and
informed consent. related concomitant use of medications 4 weeks
Eligible patients were at least 21 years of age after a patient completed the study at week 30 or
and had had Crohn’s disease for at least 6 weeks, week 50 or withdrew from the trial.
with a score of 220 to 450 points on the Crohn’s Randomization was performed centrally with
Disease Activity Index (CDAI).13 This index con- the use of an adaptive randomization procedure
sists of eight factors, with each factor totaled stratified according to center, the duration of
after adjustment with a weighting factor. (Severe Crohn’s disease (<3 years or ≥3 years), and status
disease is defined as a score of more than 450, with respect to the systemic corticosteroid dose

1384 n engl j med 362;15  [Link]  april 15, 2010

The New England Journal of Medicine

Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 Infliximab, Azathioprine, or Both for Crohn’s Disease

(the equivalent of <20 mg or ≥20 mg of prednisone healing at week 26 among those who had ulcer-
daily). ations at baseline, the rate of any remission, re-
Oral mesalamine was continued at a stable sponse-70, response-100, the IBDQ score, the cor-
dose. Systemic corticosteroids could be initiated ticosteroid dose at each data-collection time point
(for patients not receiving them at baseline) with (weeks 0, 2, 6, 10, 18, 26, 34, 42, and 50), and the
the dose maintained, increased, or decreased change in the CRP level from baseline to week 26.
until week 14 (maximum allowed dose, 40 mg
per day). After week 14, the dose was tapered at Study Oversight
a rate of at least 5 mg per week. Budesonide The study was jointly designed by members of
could be maintained or decreased until week 14 the SONIC executive committee of academic in-
(maximum dose, 9 mg per day). After week 14, vestigators and researchers employed by Cento-
budesonide was tapered at a rate of 3 mg every cor Ortho Biotech, one of the trial sponsors. Data
2 weeks to a dose of 6 mg per day or less. were collected and analyzed by Quintiles. The two
lead academic authors wrote the first draft of the
Evaluation of Efficacy and Safety manuscript. The academic authors and represen-
Scores on the CDAI and the Inflammatory Bowel tatives of the sponsor made the decision to sub-
Disease Questionnaire (IBDQ)14 were determined mit the manuscript for publication. The academic
at weeks 0, 2, 6, 10, 18, 26, 34, 42, and 50. Ileo- authors vouch for the veracity and completeness
colonoscopy was performed at baseline and again of the data and data analyses.
at week 26 in patients who had mucosal ulcers at
the baseline examination. All colonoscopies were Statistical Analysis
videotaped with the use of a standard protocol For the primary end point, the rate of cortico­
and interpreted by a single reviewer, who was steroid-free clinical remission at week 26, it was
unaware of study-group assignments and the estimated that 500 patients would be needed to
timing of the procedure (i.e., at baseline or week provide a power of 94% in order to detect a dif-
26). Clinical remission was defined as an abso- ference in remission rates of 20% between the
lute CDAI score of less than 150 points.13,15 Corti- combination-therapy group and the azathioprine
costeroid-free clinical remission was defined as group, on the assumption that the rate of remis-
clinical remission in patients who had not re- sion would be 60% in the combination-therapy
ceived budesonide at a daily dose of more than group and 40% in the azathioprine group.
6 mg or systemic corticosteroids for at least To control for a type I error of 0.05 or less,
3 weeks. Response-70 and response-100 were de- the primary end-point analyses were conducted
fined as reductions from baseline in the CDAI in a prespecified, sequential manner, in which
score of at least 70 and 100 points, respective- the azathioprine group was compared with the
ly.13,15 Mucosal healing was defined as the ab- combination-therapy group first at a 0.05 (two-
sence of mucosal ulceration at week 26 in pa- sided) significance level. The azathioprine and
tients who had confirmed mucosal ulceration at infliximab groups were then compared at a 0.05
baseline. (two-sided) significance level only if the first
Monitoring for adverse events and use of con- comparison was significant. If the first compari-
comitant medications was performed through son was not significant, then the second com-
week 54. Blood samples were collected for test- parison was to be considered not significant.
ing at weeks 0 and 26 for levels of C-reactive Given the large number of prespecified second-
protein (CRP) and at weeks 0, 30, and 46 for the ary efficacy variables that were evaluated at mul-
presence of antibodies to infliximab.16 tiple time points during the study, the P values
for all secondary efficacy variables should be
Primary and Secondary End Points considered nominal, since no adjustments were
The primary efficacy end point was the rate of made for multiple comparisons.
corticosteroid-free clinical remission at week 26; Demographic and baseline characteristics were
the rates of corticosteroid-free clinical remission compared with the use of the chi-square test or
at other time points were secondary efficacy end Fisher’s exact test for categorical variables and
points. Additional secondary efficacy end points with analysis of variance on a van der Waerden
included the proportion of patients with mucosal normal-scores test for continuous variables. A

n engl j med 362;15  [Link]  april 15, 2010 1385

The New England Journal of Medicine
Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Demographic and Clinical Characteristics of the Patients.*

Combination
Azathioprine Infliximab Therapy All Patients
Characteristic (N = 170) (N = 169) (N = 169) (N = 508) P Value†
Male sex — no. (%) 90 (52.9) 84 (49.7) 88 (52.1) 262 (51.6) 0.83
White race — no. (%)‡ 147 (91.3) 146 (93.0) 142 (94.0) 435 (92.8) 0.22
Median age — yr§ 35.0 35.0 34.0 34.0 0.95
Median body weight — kg 69.6 68.9 72.0 70.2 0.45
Median disease duration — yr 2.4 2.2 2.2 2.3 0.60
Median C-reactive protein — mg/dl¶ 1.0 1.1 1.0 1.1 0.40
Crohn’s Disease Activity Index score‖ 287.2±52.9 284.8±62.1 289.9±55.0 287.3±56.7 0.59
Gastrointestinal area involved — no./
total no. (%)
Ileum or colon 170/170 (100.0) 163/169 (96.4) 167/169 (98.8) 500/508 (98.4)
Ileum only 68/170 (40.0) 54/163 (33.1) 54/167 (32.3) 176/500 (35.2) 0.34
Colon only 33/170 (19.4) 45/163 (27.6) 40/167 (24.0) 118/500 (23.6)
Ileum and colon 69/170 (40.6) 64/163 (39.3) 73/167 (43.7) 206/500 (41.2)
Proximal gastrointestinal tract 7/170 (4.1) 12/169 (7.1) 16/169 (9.5) 35/508 (6.9) 0.15
Systemic corticosteroids — no. (%)
Any type, according to daily dose**
0 130 (76.5) 117 (69.2) 122 (72.2) 369 (72.6) 0.59
<20 mg 14 (8.2) 19 (11.2) 14 (8.3) 47 (9.3)
≥20 mg 26 (15.3) 33 (19.5) 33 (19.5) 92 (18.1)
Budesonide — no. (%) 25 (14.7) 28 (16.6) 19 (11.2) 72 (14.2) 0.36
5-Aminosalicylic compounds — no. (%) 104 (61.2) 87 (51.5) 85 (50.3) 276 (54.3) 0.09

* Plus–minus values are means ±SD.

† P values are for all comparisons among the three groups. P values for all categorical variables are based on a two-sid-
ed chi-square test. P values for continuous variables are based on analysis of variance on the van der Waerden nor-
mal scores.
‡ Race was self-reported. Data regarding race were not collected for 39 patients in France: 9 in the azathioprine group,
12 in the infliximab group, and 18 in the combination-therapy group.
§ The age range was 18 to 79 years in the azathioprine group, 18 to 80 in the infliximab group, and 19 to 68 in the com-
bination-therapy group. On March 27, 2007, after hepatosplenic T-cell lymphoma had been reported in adolescents
and very young adults receiving combination therapy with azathioprine and infliximab, the protocol was amended to
increase the minimum eligible age from 18 to 21 years.
¶ Data for C-reactive protein were missing for one patient each in the azathioprine and infliximab groups and for four
patients in the combination-therapy group.
‖ This index consists of eight factors, with each factor totaled after adjustment with a weighting factor ranging from
1 to 30, for a total possible score of approximately 600. Severe disease is defined as a total score of more than 450,
and remission as a score of less than 150.
** Patients were stratified according to whether they were receiving a daily prednisone-equivalent dose of less than
20 mg (including those not taking corticosteroids at baseline) or 20 mg or more.

two-sided Cochran–Mantel–Haenszel chi-square mal scores, with adjustment for the duration of
test, stratified according to the duration of Crohn’s disease and status with respect to the
Crohn’s disease and status with respect to corti- corticosteroid dose at baseline. Descriptive sta-
costeroid dose at baseline, was used to compare tistics summarize systemic corticosteroid use.
remission, response, and mucosal healing. Chang- To evaluate the consistency of the treatment
es in IBDQ scores were compared with the use effect on corticosteroid-free clinical remission
of analysis of variance on van der Waerden nor- among the azathioprine, infliximab, and combi-

1386 n engl j med 362;15  [Link]  april 15, 2010

The New England Journal of Medicine

Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 Infliximab, Azathioprine, or Both for Crohn’s Disease

817 Patients were assessed for eligibility

309 Were excluded

266 Were not eligible
29 Withdrew consent
8 Were lost to follow-up
3 Had other reasons
2 Had an adverse event
1 Died

508 Underwent randomization

170 Were assigned to receive 169 Were assigned to receive 169 Were assigned to receive
azathioprine capsules daily infliximab infusions at wk 0, 2, infliximab infusions at wk 0, 2,
plus placebo infusions at wk 6, 14, and 22 plus placebo 6, 14, and 22 plus azathioprine
0, 2, 6, 14, and 22 capsules daily capsules daily
8 Received combination therapy 3 Received combination therapy 1 Underwent randomization
and were assessed for safety and were assessed for safety but was not treated
on the basis of treatment on the basis of treatment 179 Were treated (including 11
received received patients who were inadver-
1 Underwent randomization 3 Underwent randomization tently given combination
but was not treated but were not treated therapy)
161 Were treated as assigned 163 Were treated as assigned

84 Discontinued the study 48 Discontinued the study

58 Discontinued the study
3 Were not eligible 2 Were not eligible
8 Were not eligible
18 Withdrew consent 7 Withdrew consent
9 Withdrew consent
38 Had an adverse event 28 Had an adverse event
20 Had an adverse event
5 Were lost to follow-up 2 Were lost to follow-up
5 Were lost to follow-up
19 Had other reasons 9 Had other reasons
16 Had other reasons
1 Died

86 Completed 30-wk trial 111 Completed 30-wk trial 121 Completed 30-wk trial

75 Were enrolled in the study 97 Were enrolled in the study 108 Were enrolled in the study
extension and received extension and received extension and received
azathioprine capsules daily infliximab infusions at wk 30, infliximab infusions at wk 30,
through wk 50 plus placebo 38, and 46 plus placebo 38, and 46 plus azathioprine
infusions at wk 30, 38, and 46 capsules daily through wk 50 capsules daily through wk 50

8 Discontinued extension trial 18 Discontinued extension trial

12 Discontinued extension trial
2 Withdrew consent 4 Withdrew consent
9 Had an adverse event
3 Had an adverse event 7 Had an adverse event
1 Was lost to follow-up
2 Were lost to follow-up 4 Were lost to follow-up
2 Had other reasons
1 Had other reasons 3 Had other reasons

67 Completed extension trial 85 Completed extension trial 90 Completed extension trial

through wk 50 through wk 50 through wk 50

Figure 1. Enrollment and Outcomes through Week 50.

nation-therapy groups, we performed 11 pre- calculated on the basis of logistic regression. We

AUTHOR: Colombel RETAKE: 1st
specified subgroup analyses of demographic and also performed post hoc subgroup 2nd analyses of
baseline disease characteristics (Fig. 1 in1 ofthe
FIGURE: 2 the rates of corticosteroid-free
3rd clinical remission
Revised
Supplementary Appendix). The odds ARTIST:
ratios MRL
were according to baseline mucosal
SIZE
lesion status and
7 col
TYPE: Line Combo 4-C H/T 39p6
AUTHOR, PLEASE NOTE: 1387
n engl j med 362;15  [Link]  april 15, 2010
Figure has been redrawn and type has been reset.
Please check carefully.
The New England Journal of Medicine
Downloaded from [Link] on October
JOB: 36207 5, 2015. For personal use only. No other2-18-10
ISSUE: uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 2. Efficacy Results.*

P Value for P Value for

P Value for Combination Combination Combination
Azathioprine Infliximab Infliximab vs. Therapy Therapy vs. Therapy vs.
Variable (N = 170) (N = 169) Azathioprine† (N = 169) Infliximab† Azathioprine†
Patients with corticosteroid-free clinical
remission — no. (%)
Week 6 24 (14.1) 50 (29.6) <0.001 55 (32.5) 0.55 <0.001
Week 10 41 (24.1) 63 (37.3) 0.006 79 (46.7) 0.07 <0.001
Week 18 44 (25.9) 72 (42.6) <0.001 89 (52.7) 0.06 <0.001
Week 26 51 (30.0) 75 (44.4) 0.006 96 (56.8) 0.02 <0.001
Week 34
All patients‡ 45 (26.5) 62 (36.7) 0.04 73 (43.2) 0.22 0.001
Patients with week 26 status carried 52 (30.6) 70 (41.4) 0.03 89 (52.7) 0.03 <0.001
forward through week 50§
Patients entering trial extension¶ 45 (60.0) 62 (63.9) 0.41 73 (67.6) 0.56 0.20
Week 42
All patients‡ 44 (25.9) 67 (39.6) 0.007 77 (45.6) 0.27 <0.001
Patients with week 26 status carried 51 (30.0) 75 (44.4) 0.006 93 (55.0) 0.04 <0.001
forward through week 50§
Patients entering trial extension¶ 44 (58.7) 67 (69.1) 0.13 77 (71.3) 0.67 0.06
Week 50
All patients‡ 41 (24.1) 59 (34.9) 0.03 78 (46.2) 0.04 <0.001
Patients with week 26 status carried 48 (28.2) 67 (39.6) 0.03 94 (55.6) 0.002 <0.001
forward through week 50§
Patients entering trial extension¶ 41 (54.7) 59 (60.8) 0.32 78 (72.2) 0.07 0.01
Patients who received systemic cortico­
steroid during the main study‖
No. of patients (%) 60 (35.3) 60 (35.5) 58 (34.3)
Baseline
No. of patients 40 52 47
Mean dose (mg/day) 23.8±12.6 24.8±16.2 ND 24.9±12.8 ND ND
Week 2
No. of patients 48 50 49
Mean dose (mg/day) 22.9±12.5 21.2±11.9 ND 22.8±11.9 ND ND
Week 6
No. of patients 53 52 51
Mean dose (mg/day) 18.6±11.6 17.7±11.0 ND 18.3±11.6 ND ND
Week 10
No. of patients 56 56 52
Mean dose (mg/day) 16.2±11.2 15.7±14.9 ND 15.0±11.1 ND ND
Week 18
No. of patients 59 57 56
Mean dose (mg/day) 13.5±10.9 13.2±17.2 ND 11.6±10.9 ND ND

Week 26
No. of patients 60 60 58
Mean dose (mg/day) 11.6±10.3 11.0±16.0 ND 9.4±10.1 ND ND

1388 n engl j med 362;15  [Link]  april 15, 2010

The New England Journal of Medicine

Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 Infliximab, Azathioprine, or Both for Crohn’s Disease

Table 2. (Continued.)

P Value for P Value for

P Value for Combination Combination Combination
Azathioprine Infliximab Infliximab vs. Therapy Therapy vs. Therapy vs.
Variable (N = 170) (N = 169) Azathioprine† (N = 169) Infliximab† Azathioprine†
Extension-study patients who received
systemic corticosteroid during the
main study or study extension‖
No. of patients (%) 22 (29.3) 41 (42.3) 36 (33.3)
Week 34
No. of patients 22 37 36
Mean dose (mg/day) 7.4±6.3 5.8±5.5 ND 6.4±5.7 ND ND
Week 42
No. of patients 22 38 36
Mean dose (mg/day) 6.3±5.3 4.7±4.6 ND 5.7±5.8 ND ND
Week 50
No. of patients 22 41 36
Mean dose (mg/day) 5.7±5.0 3.9±3.9 ND 5.0±5.4 ND ND
Mucosal healing
Baseline lesions — no. (%)** 115 (67.6) 99 (58.6) 0.08 111 (65.7) 0.18 0.70
Patients included in week 26 analysis — 109 (64.1) 93 (55.0) 0.09 107 (63.3) 0.12 0.88
no. (%)††
Mucosal healing — no. (%) 18 (16.5) 28 (30.1) 0.02 47 (43.9) 0.06 <0.001

* Plus–minus values are means ±SD.

† P values were calculated by means of the Cochran–Mantel–Haenszel test, stratified according to the duration of Crohn’s disease and the dose
of systemic corticosteroid at baseline (the equivalent of 0 to <20 mg or ≥20 mg of prednisone daily). ND denotes not done; descriptive statis-
tics were preplanned, and no post hoc statistical tests were performed because mean doses were similar among the three study groups.
‡ This category includes all patients who underwent randomization, with the assumption that patients who did not enter the trial extension
did not reach the end point through week 50.
§ This category includes all patients who underwent randomization, with week 26 status carried forward through week 50 for patients who
did not enter the trial extension.
¶ Results for weeks 34, 42, and 50 are based on 75 patients in the azathioprine group, 97 in the infliximab group, and 108 in the combina-
tion-therapy group who entered the trial extension.
‖ The systemic corticosteroid dose for a study week was calculated as the cumulative dose that a patient received up to a given study week
divided by the number of days that the patient participated in the study up to that study week.
** Sixteen patients (four in the combination-therapy group, six in the infliximab group, and six in the azathioprine group) were excluded from
the analysis of mucosal healing at week 26 because they underwent endoscopy before or after week 26.
†† Patients with lesions at baseline who did not undergo endoscopy at week 26 or who had results that could not be evaluated were assumed
to have a lesion. These patients included 50 of 109 (45.9%) in the azathioprine group, 29 of 93 (31.2%) in the infliximab group, and 31 of
107 (29.0%) in the combination-therapy group.

CRP level and the presence or absence of anti- who entered the extension trial. To perform ex-
bodies to infliximab. ploratory analyses of outcomes for all patients
Efficacy analyses were performed according through week 50, we assumed that the clinical
to the intention-to-treat principle and included end point at week 50 was not reached by patients
all patients who had undergone randomization, who did not enter the trial extension, and we
with analyses performed according to the ran- used the week 26 clinical end-point status car-
domized study-group assignments, except for ried forward through week 50 for those who did
mucosal healing, which was analyzed according not enter the trial extension.
to prespecified per-protocol methods. Patients All patients receiving at least one dose of a
who required surgery for Crohn’s disease or study drug (administered orally or by infusion)
withdrew from the study were not considered to were included in the safety analysis, according
be in remission. After week 30, prespecified to the study drug that was actually received, with
analyses were based on data only for patients safety comparisons performed with the use of

n engl j med 362;15  [Link]  april 15, 2010 1389

The New England Journal of Medicine
Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

teristics were similar in the three groups (Table 1,

A Corticosteroid-free Clinical Remission at Wk 26
P<0.001
and Table 2 in the Supplementary Appendix). A
100
total of 318 patients completed the 30-week trial,
80 P=0.02 of whom 280 entered the extension trial (Fig. 1).

P=0.006 Primary End Point

Patients (%)
60 56.8

44.4 At week 26, a total of 96 of the 169 patients

40 (56.8%) receiving combination therapy, 75 of the
30.0
169 patients (44.4%) receiving infliximab, and 51
20
51/170 75/169 96/169
of the 170 patients (30.0%) receiving azathio-
prine were in corticosteroid-free clinical remis-
0 sion (P = 0.006 for the comparison of infliximab
he e

he b

op –
vs. azathioprine, P<0.001 for the comparison of
py

py

e
ot rin

hi b
ot ma

rin
at ma
ra

ra
on op

on xi

Az flixi
combination therapy vs. azathioprine, and P = 0.02
M nfli
M thi
a

In
I
Az

for the comparison of combination therapy vs.

infliximab) (Table 2 and Fig. 2A). The efficacy of
B Mucosal Healing at Wk 26
combination therapy and infliximab was generally
100 P<0.001
consistent among subgroups defined according
80
to demographic and baseline disease characteris-
P=0.06 tics (Fig. 1A, 1B, and 1C in the Supplementary
Appendix).
Patients (%)

60
P=0.02
43.9
40 Secondary End Points and Exploratory
30.1 Analyses
20 16.5 At baseline, mucosal ulcerations were detected in
18/109 28/93 47/107 325 patients: 111 of 169 patients (65.7%) in the
0 combination therapy group, 99 of 169 patients
he e

he b

op –
py

py

(58.6%) in the infliximab group, and 115 of 170

ot rin

hi b
ot ma

rin
at ma
ra

ra
on op

on xi

Az flixi

patients (67.6%) in the azathioprine group. In 93

M nfli
M thi
a

In
I
Az

patients, no ulcerations were found on ileocolonos-

Figure 2. Patients with Corticosteroid-free Clinical copy, and in 90 patients, either the findings were
­RemissionColombel
AUTHOR: RETAKE:
(Panel A) and Mucosal Healing (Panel B) 1st judged to be inconclusive by the central reader be-
2nd
atFIGURE:
Week 26.
2 of 2 3rd cause of poor colon preparation or poor videotape
ARTIST: MRL
Revised technique or the procedure was not performed. At
SIZE
3 col week 26, mucosal healing had occurred in 47 of
TYPE: Line Combo 4-C H/T
Fisher’s exact test. Patients with at least one
se- 16p6 107 patients (43.9%) receiving combination ther-
rum sample that wasPLEASE
AUTHOR, collected
NOTE: after the first apy, in 28 of 93 patients (30.1%) receiving inflix-
Figure has been redrawn and type has been reset.
infliximab infusion Pleasewere included in the analy-
check carefully. imab, and in 18 of 109 patients (16.5%) receiving
sis of trough levels. Infliximab levels were com- azathioprine (P = 0.02 for infliximab vs. azathio-
JOB: 36215 ISSUE: 4-15-10
pared by means of analysis of variance on van prine, P<0.001 for combination therapy vs. aza-
der Waerden normal scores, with adjustment for thioprine, and P = 0.06 for combination therapy
the duration of Crohn’s disease and status with vs. infliximab) (Table 2 and Fig. 2B).
respect to corticosteroid dose at baseline. For secondary end points at weeks 34, 42, and
50, the between-group differences also favored
R e sult s combination therapy and infliximab monother-
apy over azathioprine (Table 2, and Table 3 in
Patients the Supplementary Appendix). Results of explor-
Of the 508 patients who underwent randomiza- atory analyses that included all randomized pa-
tion, 170 were assigned to receive azathioprine tients at week 50 were consistent with those at
monotherapy, 169 to receive infliximab mono- week 26, with combination therapy and inflix-
therapy, and 169 to receive combination therapy imab monotherapy providing a significantly
with the two drugs. The baseline disease charac- greater benefit than azathioprine monotherapy,

1390 n engl j med 362;15  [Link]  april 15, 2010

The New England Journal of Medicine

Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 Infliximab, Azathioprine, or Both for Crohn’s Disease

and combination therapy providing a signifi- fliximab in the serum) than among patients with
cantly greater benefit than infliximab mono- negative or positive results (Fig. 4A and 4B in the
therapy (Table 2, and Tables 4 and 5 and Fig. 2A Supplementary Appendix). Median trough levels
and 2B in the Supplementary Appendix). of serum infliximab at week 30 were 1.6 μg per
At baseline (6 days before the first dose of milliliter for patients in the infliximab group
study medication was administered), budesonide and 3.5 μg per milliliter for those in the combi-
or a systemic corticosteroid was being adminis- nation-therapy group (P<0.001) (Fig. 5A in the
tered in 66 of 169 patients (39.1%) receiving Supplementary Appendix). Rates of corticoste­
combination therapy, 80 of 169 patients (47.3%) roid-free clinical remission were greater among
receiving infliximab, and 65 of 170 patients patients with increased trough levels of serum
(38.2%) receiving azathioprine (Table 1). During infliximab but were still high among patients with
the main study, systemic corticosteroid therapy lower trough levels (Fig. 5B in the Supplementary
was initiated in 11 patients receiving combina- Appendix). The findings were similar for serum in-
tion therapy, 8 patients receiving infliximab, and fliximab trough levels and rates of corticosteroid-
20 patients receiving azathioprine. Mean doses free clinical remission at week 46 (Fig. 6A and 6B
of systemic corticosteroids through week 50 are in the Supplementary Appendix).
summarized in Table 2.
The rates of corticosteroid-free clinical remis- Discussion
sion at week 26 in both the combination-therapy
group and the infliximab group, as compared In this study, treatment with infliximab-based
with the azathioprine group, were greater among strategies, as compared with azathioprine, re-
subgroups of patients with higher baseline CRP sulted in significantly higher rates of corticoste­
levels (0.8 mg per deciliter or more), baseline roid-free clinical remission and mucosal healing
mucosal lesions, and both higher baseline CRP at week 26 among patients with active Crohn’s
levels and mucosal lesions (Fig. 3A, 3B, and 3C disease who did not have a response to first-line
in the Supplementary Appendix). therapy. The greatest efficacy was observed with
combination therapy.
Safety The absolute rate of corticosteroid-free clini-
Through week 50, the incidence of adverse events cal remission among patients treated with aza-
was generally similar among the three groups thioprine (30% at week 26) was generally similar
(Table 3). Infusion reactions occurred in 9 of 179 to such rates that have been reported in previous
patients (5.0%) in the combination-therapy group, studies that used the CDAI to measure efficacy:
in 27 of 163 patients (16.6%) in the infliximab 36% at week 17,5 27% at week 16,17 29% at week
group, and in 9 of 161 patients (5.6%) in the aza- 24,18 and 38% at month 7.19 In contrast, several
thioprine group. other studies that either had very small sam-
In one patient receiving combination therapy, ples20,21 or used other, nonvalidated instruments
tuberculosis developed approximately 3 months to measure efficacy22 have reported higher rates
after a negative tuberculin skin test and chest of treatment success with azathioprine. The
radiograph. The patient recovered after receiving time to the onset of action for azathioprine is
antituberculosis treatment. Colon cancer devel- estimated at 8 to 12 weeks.17,22 For this reason,
oped in two patients receiving azathioprine. An- we permitted the initiation or continuation of
other patient receiving azathioprine died from systemic corticosteroid use until week 14. The
sepsis after colectomy. proportion of patients receiving corticosteroids
during the trial was similar among the three
Antibodies to Infliximab and Infliximab Levels study groups, and the magnitude of the differ-
Antibodies to infliximab were detected at week ence in the end points among the groups was
30 in 1 of 116 patients (0.9%) receiving combina- relatively greater among patients who had re-
tion therapy and 15 of 103 patients (14.6%) re- ceived corticosteroids at baseline. It is possible
ceiving infliximab. The rates of corticosteroid- that by excluding patients who had a heterozy-
free clinical remission at weeks 26 and 50 were gous thiopurine methyltransferase phenotype we
higher among patients with inconclusive results excluded patients who were more likely to have
of antibody tests (indicating the presence of in­ a response to azathioprine.23 However, such pa-

n engl j med 362;15  [Link]  april 15, 2010 1391

The New England Journal of Medicine
Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 3. Safety Data at Week 54.*

P Value for P Value for

P Value for Combination Combination Combination
Azathioprine Infliximab Infliximab vs. Therapy Therapy vs. Therapy vs.
Variable (N = 161) (N = 163) Azathioprine† (N = 179) Infliximab† Azathioprine†
Mean duration of treatment in main study 21.1 24.1 24.9
— wk
Mean duration of treatment in study exten- 18.9 18.8 18.7
sion — wk‡
Mean duration of post-treatment follow-up 5.1 5.4 5.3
— wk
Total patient-years of follow-up — no. 108.1 126.9 142.2
Patients with any adverse event — no. (%) 144 (89.4) 145 (89.0) 1.00 161 (89.9) 0.86 1.00
Adverse events occurring in >10% of any
study group — no. (%)
Nausea 52 (32.3) 36 (22.1) 45 (25.1)
Abdominal pain 26 (16.1) 41 (25.2) 31 (17.3)
Worsening of Crohn’s disease 27 (16.8) 30 (18.4) 19 (10.6)
Vomiting 28 (17.4) 23 (14.1) 15 (8.4)
Diarrhea 13 (8.1) 20 (12.3) 14 (7.8)
Fatigue 25 (15.5) 24 (14.7) 26 (14.5)
Pyrexia 18 (11.2) 16 (9.8) 16 (8.9)
Arthralgia 18 (11.2) 32 (19.6) 21 (11.7)
Headache 20 (12.4) 27 (16.6) 23 (12.8)
Nasopharyngitis 20 (12.4) 15 (9.2) 21 (11.7)
Adverse events leading to discontinuation of 42 (26.1) 29 (17.8) 0.08 37 (20.7) 0.58 0.25
a study drug — no. (%)
Any serious adverse event — no. (%) 43 (26.7) 39 (23.9) 0.61 27 (15.1) 0.04 0.01
Infection — no. (%)
Any 73 (45.3) 75 (46.0) 0.91 75 (41.9) 0.45 0.58
Serious 9 (5.6) 8 (4.9) 0.81 7 (3.9) 0.79 0.61
Adverse event of interest — no. (%)
Colon carcinoma§ 2 (1.2) 0 0
Sepsis 1 (0.6) 0 0
Tuberculosis 0 0 1 (0.6)
Infusions
No. of patients with infusion reaction — 9 (5.6) 27 (16.6) 0.002 9 (5.0) <0.001 1.00
no. (%)¶
No. of infusions
Mean no. per patient 5.4 6.1 6.1
Total no. 862 990 1097
With infusion reaction 10 45 11

* Excluded from the safety analyses were five patients who underwent randomization but did not receive a study drug (one patient in the aza-
thioprine group, three in the infliximab group, and one in the combination-therapy group). The safety population for the combination-thera-
py group included 11 patients who were assigned to one of the monotherapy groups but inadvertently were given at least one dose of both
active oral and intravenous therapy (8 patients in the azathioprine group and 3 in the infliximab group).
† P values were calculated with the use of Fisher’s exact test.
‡ Analyses of results for the trial extension included 75 patients in the azathioprine group, 97 in the infliximab group, and 108 in the combina-
tion-therapy group.
§ The two patients underwent colectomy. Although colonoscopy was performed at baseline, the protocol did not require that biopsy samples
be obtained.
¶ Infusion reactions were defined as any adverse event that occurred during or within 1 hour after the infusion of a study drug.

1392 n engl j med 362;15  [Link]  april 15, 2010

The New England Journal of Medicine

Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 Infliximab, Azathioprine, or Both for Crohn’s Disease

tients are also more likely to be intolerant to therapy was superior to infliximab monotherapy.
azathioprine,24,25 and we believe it is unlikely However, the increased risks of rare but serious
that the net effect of these competing forces had toxic effects associated with combination therapy
an important effect on the outcome of the study. must also be considered. The concomitant use of
Subgroup analyses of earlier clinical trials of corticosteroids as a third immunosuppressive
infliximab that included patients who did not agent may further increase the relative risk of seri-
have a response to azathioprine have not shown ous and opportunistic infections.30 Ultimately,
greater efficacy during a period of 6 to 12 the choice of infliximab monotherapy or combi-
months among patients receiving combination nation therapy in patients who have not received
therapy with infliximab and azathioprine, as such therapy previously is an individualized ben-
compared with patients receiving infliximab efit–risk decision. Although the greater efficacy
monotherapy.26 In addition, a randomized trial of combination therapy in our study may have been
of azathioprine withdrawal in this patient popu- due in part to suppression of immunogenicity, it
lation did not show a clinical benefit from con- is likely that the enhanced benefit was primarily
tinued use of azathioprine,27 and toxic effects of due to the additive effects of two effective drugs.
azathioprine combined with anti-TNF biologic The two drugs have also been shown to share
agents have recently been reported.28,29 Our trial mechanisms of action, such as apoptosis.31,32
did not address the question of whether combi- In conclusion, infliximab monotherapy and
nation therapy was superior to infliximab mono- combination therapy with infliximab plus azathi-
therapy after failure of azathioprine. The bene- oprine, as compared with azathioprine alone, re-
fits that we found for combination therapy may sulted in significantly higher rates of cortico­
not extend to patients in whom azathioprine has steroid-free clinical remission among patients with
already failed. moderate-to-severe Crohn’s disease. Combination
Previously, no single variable has consistently therapy was the most effective treatment.
predicted a response to infliximab. In post hoc
analyses, we found that patients with objective Supported by research grants from Centocor Ortho Biotech
and Schering-Plough.
evidence of inflammation (i.e., a high CRP level Dr. Colombel reports receiving consulting or advisory board
or observed mucosal lesions) had the best clini- fees from Abbott Laboratories, ActoGeniX, AstraZeneca, Bayer-
cal results with infliximab. In patients with a Schering Pharma, Biogen Idec, Boehringer Ingelheim, Bristol-
Myers Squibb, Cellerix, ChemoCentryx, Centocor Ortho Biotech,
normal CRP level or no endoscopic lesions, no Cosmo Technologies, Danone France, Elan Pharmaceuticals,
significant differences were observed among the Genentech, Giuliani SPA, Given Imaging, GlaxoSmithKline,
three study groups. It is possible that prospec- Merck, Millennium Pharmaceuticals, Neovacs, Ocera Therapeu-
tics, Otsuka America Pharmaceuticals, PDL Biopharma, Pfizer,
tive studies will show that measurement of CRP RiboVacs Biotech, Schering-Plough, Shire Pharmaceutical, Synta
and endoscopy can identify patients who are Pharmaceutical, Teva Pharmaceuticals, Therakos, UCB Pharma,
most likely to have a greater response to inflix- and Wyeth, lecture fees from Abbott Laboratories, Centocor
Ortho Biotech, Elan, Falk Pharma, Ferring Pharmaceuticals,
imab monotherapy or combination therapy than Given Imaging, Otsuka America Pharmaceuticals, PDL Bio­
to azathioprine monotherapy. pharma, Schering-Plough, Shire Pharmaceuticals, and UCB
The overall incidence of adverse events was Pharma, grant support from Abbott Laboratories, Centocor
Ortho Biotech, Synta Pharma, Otsuka America Pharmaceuticals,
similar among the three groups. However, infu- Bristol-Myers Squibb, PDL Biopharma, Chiltem, AstraZeneca,
sion reactions occurred less frequently among pa- Pfizer, Teva Pharmaceuticals, Lesaffre, Giuliani SPA, Danisco,
tients receiving combination therapy than among Ocera Therapeutics, Danone France, Roquette, Mapi Naxis, Dys-
phar, Ferring Pharmaceuticals, Schering-Plough, and UCB
those receiving infliximab monotherapy. Serious Pharma, and having an equity interest in Intestinal Biotech De-
infections (including tuberculosis) developed in velopment; Dr. Sandborn, receiving consulting or advisory
3.9% of patients in the combination-therapy group, board fees from Abbott Laboratories (fees paid to the Mayo
Clinic), ActoGeniX, AGI Therapeutics, Alba Therapeutics, Albi­
4.9% of those in the infliximab group, and 5.6% reo, AM-Pharma, Amgen, Ardea Biosciences, Aspreva Pharma-
of those in the azathioprine group. Evidence sug- ceuticals, Astellas Pharma, Athersys, Atlantic Healthcare Limit-
gests that the combination of azathioprine and ed, Axcan Pharma, BioBalance, Bristol-Myers Squibb, Celgene,
Celek Pharmaceuticals, Cellerix, Centocor Ortho Biotech (fees
anti-TNF biologic agents increases the relative paid to the Mayo Clinic), Cerimon Pharmaceutical, ChemoCen-
risk of serious and opportunistic infections and tryx, CombinatoRx, CoMentis, Cosmo Technologies, Cytokine
hepatosplenic T-cell lymphoma.28,29 Our data Pharmasciences, Eagle Pharmaceuticals, Eisai Medical Research,
Elan Pharmaceuticals, Enteromedics, Enzo Therapeutics, Ferring
showed that for patients who had not received Pharmaceuticals, Flexion Therapeutics, Funxional Therapeutics,
previous therapy with azathioprine, combination Genentech, Genzyme, Given Imaging, GlaxoSmithKline, Human

n engl j med 362;15  [Link]  april 15, 2010 1393

The New England Journal of Medicine
Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Genome Sciences, Hutchison Medipharma, Ironwood Pharma- Centocor Ortho Biotech, Shire Pharmaceuticals, Prometheus
ceuticals, KaloBios Pharmaceuticals, Merck Research Laborato- Laboratories, and UCB Pharma, lecture fees from Abbott Labo-
ries, MerckSerono, Millennium Pharmaceuticals, Nisshin Kyorin ratories, Procter & Gamble, Prometheus Laboratories, and Shire
Pharmaceutical, Novo Nordisk, Ocera Therapeutics, Pfizer, Pharmaceuticals, and grant support from Abbott Laboratories,
Procter & Gamble (fees paid to the Mayo Clinic), Prometheus Bristol-Myers Pharmaceuticals, Celgene, Centocor Ortho Biotech,
Laboratories, Purgenesis Technologies, Salient Pharmaceuticals, Osiris Pharmaceuticals, Procter & Gamble, and UCB Pharma;
Salix Pharmaceuticals, Santarus, Schering-Plough, Shire Phar- Dr. D’Haens, receiving consulting fees from Centocor Ortho
maceuticals (fees paid to the Mayo Clinic), Sigmoid Pharma, Biotech, Schering-Plough, UCB Pharma, and Abbott and lecture
Sirtris Pharmaceuticals, S.L.A. Pharma, Teva Pharmaceuticals, fees from Schering-Plough, Abbott Laboratories, and UCB
Tillotts Pharma, Tioga Pharmaceuticals, UCB Pharma (fees paid Pharma; Drs. Diamond, Broussard, and Tang, being employed
to the Mayo Clinic), Vascular Biogenics, Ventech, Viamet Phar- by Centocor Ortho Biotech and having an equity interest in
maceuticals, and Wyeth, and research support from Abbott Johnson & Johnson; Dr. van der Woude, receiving consulting or
Laboratories, Bristol-Myers Squibb, Celltech, Centocor Ortho advisory board fees from Schering-Plough and Abbott Laborato-
Biotech, Genentech, Millennium Pharmaceuticals, Novartis, ries, lecture fees from Ferring, Tramedico, Schering-Plough, and
Otsuka America Pharmaceuticals, PDL Biopharma, Pfizer, Abbott, and grant support from Ely Broad Foundation, Erasmus
Procter & Gamble, Robarts Research Institute, Shire Pharma- Medical Center, Schering-Plough, and Abbott Laboratories; and
ceuticals, and UCB Pharma; Dr. Reinisch, receiving consulting Dr. Rutgeerts, receiving consulting or advisory board fees from
or advisory board fees from Abbott Laboratories, Centocor Centocor Ortho Biotech, Schering-Plough, UCB Pharma, Abbott,
Ortho Biotech, Schering-Plough, and Genentech and lecture fees Millennium, Genetech, NovImmune, ChemoCentryx, Glaxo­
from Abbott Laboratories, Otsuka America Pharmaceuticals, SmithKline, and Italifarmako, lecture fees from Centocor Ortho
and Schering-Plough; Dr. Mantzaris, receiving advisory board Biotech, Schering-Plough, UCB Pharma, and Abbott, and re-
fees from Centocor Ortho Biotech, Schering-Plough, Abbott Im- search support from Centocor Ortho Biotech, Schering-Plough,
munology, Schering-Plough Hellas, and Abbott Hellas and lec- UCB Pharma, and Abbott. No other potential conflict of interest
ture fees from Ferring International, Schering-Plough Hellas, relevant to this article was reported.
and Abbott Hellas; Dr. Kornbluth, receiving consulting or advi- We thank James Barrett and Mary Whitman, employees of the
sory board fees from Abbott Laboratories, Elan-Biogen, Cento- Medical Affairs Publication Group of Centocor Ortho Biotech,
cor Ortho Biotech, and UCB Pharma, lecture fees from Abbott for their editorial and writing support; and David Krupa and
Laboratories and UCB Pharma, and grant support from Cento- Alex Brink, employees of Quintiles, and Ronald Hegedus and
cor Ortho Biotech and Abbott Laboratories; Dr. Lichtiger, receiv- David Zelinger, employees of Centocor Ortho Biotech, for their
ing consulting or advisory board fees from Abbott Laboratories, statistical support.

References
1. Cosnes J, Cattan S, Blain A, et al. disease. Gastroenterology 2006;130:940- disease. Gastroenterology 2002;122:512-
Long-term evolution of disease behavior 87. 30.
of Crohn’s disease. Inflamm Bowel Dis 9. Pearson DC, May GR, Fick GH, 16. Hanauer SB, Wagner CL, Bala M, et al.
2002;8:244-50. Sutherland LR. Azathioprine and 6-mer- Incidence and importance of antibody re-
2. Travis SP, Stange EF, Lémann M, et al. captopurine in Crohn disease: a meta- sponses to infliximab after maintenance
European evidence based consensus on the analysis. Ann Intern Med 1995;123:132- or episodic treatment in Crohn’s disease.
diagnosis and management of Crohn’s 42. Clin Gastroenterol Hepatol 2004;2:542-53.
disease: current management. Gut 2006; 10. Targan SR, Hanauer SB, van Deventer 17. Sandborn WJ, Tremaine WJ, Wolf DC,
55:Suppl 1:i16-i35. SJ, et al. A short-term study of chimeric et al. Lack of effect of intravenous admin-
3. Lichtenstein GR, Hanauer SB, Sand- monoclonal antibody cA2 to tumor necro- istration on time to respond to azathio-
born WJ. Management of Crohn’s disease sis factor alpha for Crohn’s disease. prine for steroid-treated Crohn’s disease.
in adults. Am J Gastroenterol 2009;104: N Engl J Med 1997;337:1029-35. Gastroenterology 1999;117:527-35.
465-83. 11. Hanauer SB, Feagan BG, Lichtenstein 18. Lémann M, Mary J-Y, Duclos B, et al.
4. Faubion WA Jr, Loftus EV Jr, Harmsen GR, et al. Maintenance infliximab for Infliximab plus azathioprine for steroid-
WS, Zinsmeister AR, Sandborn WJ. The Crohn’s disease: the ACCENT 1 ran- dependent Crohn’s disease patients: a ran-
natural history of corticosteroid therapy domised trial. Lancet 2002;359:1541-9. domized placebo-controlled trial. Gastro-
for inflammatory bowel disease: a popu- 12. Peyrin-Biroulet L, Deltenre P, de Suray enterology 2006;130:1054-61.
lation-based study. Gastroenterology 2001; N, Branche J, Sandborn WJ, Colombel JF. 19. Reinisch W, Panés J, Lémann M, et al.
121:255-60. Efficacy and safety of tumor necrosis A multicenter, randomized, double-blind
5. Summers RW, Switz DM, Sessions JT ­a ntagonists in Crohn’s disease: a meta- trial of everolimus versus azathioprine
Jr, et al. National Cooperative Crohn’s analysis of placebo-controlled trials. Clin and placebo to maintain steroid-induced
Disease Study: results of drug treatment. Gastroenterol Hepatol 2008;6:644-53. remission in patients with moderate-to-
Gastroenterology 1979;77:847-69. 13. Best WR, Becktel JM, Singleton JW, severe active Crohn’s disease. Am J Gas-
6. Greenberg GR, Feagan BG, Martin F, Kern F Jr. Development of a Crohn’s dis- troenterol 2008;103:2284-92.
et al. Oral budesonide for active Crohn’s ease activity index. Gastroenterology 1976; 20. Candy S, Wright J, Gerber M, Adams
disease. N Engl J Med 1994;331:836-41. 70:439-44. G, Gerig M, Goodman R. A controlled
7. Idem. Oral budesonide as maintenance 14. Irvine EJ, Feagan B, Rochon J, et al. double blind study of azathioprine in the
treatment for Crohn’s disease: a placebo- Quality of life: a valid and reliable mea- management of Crohn’s disease. Gut 1995;
controlled, dose-ranging study. Gastroen- sure of therapeutic efficacy in the treat- 37:674-8.
terology 1996;110:45-51. ment of inflammatory bowel disease. 21. Markowitz J, Grancher K, Kohn N,
8. Lichtenstein GR, Abreu MT, Cohen R, Gastroenterology 1994;106:287-96. Lesser M, Daum F. A multicenter trial of
Tremaine W. American Gastroenterologi- 15. Sandborn WJ, Feagan BG, Hanauer 6-mercaptopurine and prednisone in chil-
cal Association Institute technical review SB, et al. A review of activity indices and dren with newly diagnosed Crohn’s dis-
on corticosteroids, immunomodulators, efficacy endpoints for clinical trials of ease. Gastroenterology 2000;119:895-902.
and infliximab in inflammatory bowel medical therapy in adults with Crohn’s 22. Present DH, Korelitz BI, Wisch N,

1394 n engl j med 362;15  [Link]  april 15, 2010

The New England Journal of Medicine

Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.
 Infliximab, Azathioprine, or Both for Crohn’s Disease

Glass JL, Sachar DB, Pasternack BS. Treat- ner CL, et al. Clinical trial: benefits and imab use in young patients treated for
ment of Crohn’s disease with 6-mercapto­ risks of immunomodulators and mainte- inflammatory bowel disease. J Pediatr
purine: a long-term, randomized, double- nance infliximab for IBD-subgroup analy- Gastroenterol Nutr 2007;44:265-7.
blind study. N Engl J Med 1980;302:981-7. ses across four randomized trials. Ali- 30. Lichtenstein GR, Feagan BG, Cohen
23. Dubinsky MC, Lamothe S, Yang HY, et ment Pharmacol Ther 2009;30:210-26. RD, et al. Serious infections and mortality
al. Pharmacogenomics and metabolite 27. Van Assche G, Magdelaine-Beuzelin in association with therapies for Crohn’s
measurement for 6-mercaptopurine ther- C, D’Haens G, et al. Withdrawal of im- disease: TREAT registry. Clin Gastroen-
apy in inflammatory bowel disease. Gas- munosuppression in Crohn’s disease terol Hepatol 2006;4:621-30.
troenterology 2000;118:705-13. treated with scheduled infliximab main- 31. Tiede I, Fritz G, Strand S, et al. CD28-
24. Black AJ, McLeod HL, Powrie RH, et tenance: a randomized trial. Gastroenter- dependent Rac1 activation is the molecu-
al. Thiopurine methyltransferase geno- ology 2008;134:1861-8. lar target of azathioprine in primary hu-
type predicts therapy-limiting severe tox- 28. Marehbian J, Arrighi HM, Hass S, Tian man CD4+ T lymphocytes. J Clin Invest
icity from azathioprine. Ann Intern Med H, Sandborn WJ. Adverse events associated 2003;111:1133-45.
1998;129:716-8. with common therapy regimens for mod- 32. ten Hove T, van Montfrans C, Pep-
25. Lewis JD, Gelfand JM, Troxel AB, et al. erate-to-severe Crohn’s disease. Am J Gas- pelenbosch MP, van Deventer SJ. Inflix-
Immunosuppressant medications and troenterol 2009;104:2524-33. imab treatment induces apoptosis of
mortality in inflammatory bowel disease. 29. Mackey AC, Green L, Liang LC, lamina propria T lymphocytes in Crohn’s
Am J Gastroenterol 2008;103:1428-35. Dinndorf P, Avigan M. Hepatosplenic disease. Gut 2002;50:206-11.
26. Lichtenstein GR, Diamond RH, Wag- T cell lymphoma associated with inflix- Copyright © 2010 Massachusetts Medical Society.

receive immediate notification when

a journal article is released early

To be notified when an article is released early

on the Web and to receive the table of contents
of the Journal by e-mail every Wednesday evening,
sign up through our Web site at
[Link].

n engl j med 362;15  [Link]  april 15, 2010 1395

The New England Journal of Medicine
Downloaded from [Link] on October 5, 2015. For personal use only. No other uses without permission.
Copyright © 2010 Massachusetts Medical Society. All rights reserved.