1097021

License

IJC
International Journal of Cancer

Risk factors for breast cancer among Filipino women in Manila

Lorna J. Gibson1, Clarisse He´ry2, Nicolas Mitton3, Abigail Gines-Bautista4, D Maxwell Parkin5, Corazon
Ngelangel6 and Paola Pisani7
1
Cancer Research UK Epidemiology and Genetics Group, London School of Hygiene and Tropical Medicine, London, UK
2 International Agency for Research on Cancer, Lyon,
France
3 Registre du Cancer de l’Ise`re, Meylan, France 4
WHO Regional Office for the Western Pacific, Manila,
Philippines
5
Clinical Trial Service Unit & Epidemiological Studies Unit, University of Oxford, Oxford, United Kingdom 6
Department of Clinical Epidemiology, College of Medicine, University of the Philippines, Manila
7 Cancer Epidemiology Unit, University of Turin, Turin, Italy

Age-adjusted incidence rates of breast cancer vary greatly worldwide with highest rates found in the typically ‘westernised’ countries of
North America and Europe. Much lower rates are observed in Asian and African populations but an exception to this has been reported
for the Manila Cancer Registry in the Philippines. The reason for this high rate is unknown but may be associated with the change in
lifestyle that has occurred in urban Manila since the 1960s. In 1995, a randomised controlled trial was set up in Manila to evaluate the
feasibility of a screening intervention by clinical breast examination as an alternative to mammography. The cohort of 151,168 women
was followed-up to 2001 for cancer incidence and a nested case-control study carried out. This aimed to evaluate the increase in breast
cancer risk associated with known risk factors. Increased risks were seen for a high level of education (OR 5 1.9 95%CI 1.1–3.3 for Epide
education stopped at 13 versus <13 years), nulliparity (OR 5 5.0 95% CI 2.5–10.0 for nulliparity versus five or more children), and late
age at first birth (OR 5 3.3 95% CI 1.3–8.3 for age 30 versus <20 years). We found no association with excess body weight, height, use of
exogenous hormones or alcohol consumption. From this study, the recognised ‘‘classical’’ risk factors do not fully explain the high breast
cancer incidence in Metro Manila, especially when compared to other urban Asian populations. We conclude that it is too simplistic to
ascribe the high risk to ‘westernisation’.

When this study was carried out, around 1995, age-adjusted United Kingdom, Fax: 0207 580 6897, E-mail:
incidence rates of breast cancer varied more than 10-fold [Link]@[Link]
worldwide.1 Generally, the highest rates are found in the
typically ‘westernised’ countries of North America and Int. J. Cancer: 126, 515–521 (2010) VC 2009 UICC
Europe, whilst much lower rates are observed in Asian and Registry in the Philippines where, at 55.1 per 100,000, the age
African populations.2 standardised (world) incidence rate (ASR) truncated at age 74,
An exception to the normally low breast cancer incidence over the period 1993–972 when our study was carried out,
in Asian women has been reported for the Manila Cancer was similar to the 52.0 per 100,000 rate seen in the UK in
Key words: epidemiology, breast cancer, risk factors, Philippines 1983-87, before the introduction of screening.3
Abbreviations: ASR: age standardised incidence rate; CBE: clinical The reason for this high breast cancer rate in Manila is
breast examination; OR: odds ratio; CI: confidence interval; FFTP: unknown but may be associated with the change in lifestyle
first fullterm pregnancy; BMI: body mass index; PEM: protein energy that has occurred in urban Manila since the 1960s. As
malnutrition; OC: oral contraceptive; SES: socioeconomic status countries become more ‘westernised’, the patterns of cancer
Grant sponsor: US Army Medical Research and Material incidence change. ‘Westernisation’ comprises the decline in
Command; Grant number: DAMD17-94-J-4327; Grant sponsor: fertility rates, change in diet and decreased physical activity.
Cancer Research UK programme; Grant number: CR-UK-C150/ Singapore, for example, underwent a dramatic transformation
A5660 in the 30 years following World War II, and this transition was
DOI: 10.1002/ijc.24769 accompanied by changes in the rates of cancer, 4,5 including
History: Received 26 Mar 2009; Accepted 6 Jul 2009; Online 22 Jul
breast cancer, that has been attributed to the major shift in
2009
reproductive and other lifestyle patterns.6 The decline in
Correspondence to: Lorna J. Gibson, Cancer Research United
fertility rates has been the major factor in the increase in
Kingdom Epidemiology and Genetics Group, London School of
breast cancer incidence in affluent countries. Given this, the
Hygiene and Tropical Medicine, Keppel Street, London, WC1E 7HT,
 1097021

516 Breast cancer risk factors in filipino women

observed high rate of breast cancer in Filipino women, variables. Crude ORs were adjusted only for age and
especially amongst the young, is unexpected. municipality. All P values are two-sided.
The purpose of this study was to evaluate, in a population
characterised by high fertility rates, the increase in breast
cancer risk associated with some of the known risk factors
that are determinants of the high incidence in Western Results
countries. The mean age of the 123 cases and 978 controls was 47.9 and
Material and Methods 47.8 years, respectively.
The study is a case-control design, nested within a large Lifestyle factors
intervention cohort recruited in the Manila resident In this study population, 18 cases (14.6%) and 142 controls
population. In 1995, a randomised controlled trial was set up (16.3%) were overweight or obese, defined as having a body
in twelve municipalities of Metro Manila, comprising 202 mass index (BMI) of 25 Kg/m2 or greater. Of these, only three
health centres, in areas covered by the Philippine Cancer (2.4%) cases and 13 controls (1.3%) were obese (BMI 30
Society— Manila (Manila—PCS) and the Department of Kg/m2 or greater). Once adjusted for education and parity, we
Health—Rizal (Rizal—DOH) Cancer registries. The aim of the observed no relationship between BMI and breast cancer risk
trial was to evaluate the feasibility of a screening intervention either overall or by menopausal status (Table 1). The adjusted
by clinical breast examination (CBE) as an alternative to ORs were 1.1 (95% CI 0.3–4.1) in premenopausal women and
mammography and assess whether this led to a subsequent 0.8 (95% CI 0.3–2.2) in postmenopausal women, not
reduction in mortality.7 During the period 1995–1997, 151,168 statistically significant.
eligible women aged 35–64 years were interviewed and The mean height of cases and controls was 155.5 cm and
offered CBE. Of those interviewed, 138,392 (91.5%) accepted 155.2 cm, respectively. There were only eight women (one
the examination. Young women were included because of the case) over 168 cm, approximately 50600. We observed no
high number of breast cancer cases below age 50. 2 The increase in risk with increased height in 5 cm increments.
participants were interviewed and examined by trained Alcohol consumption was rare in these women with 93.5% of
nurses at their place of residence or local health centre. cases and 89.8% of controls being non-drinkers. The
Questions covered demographic and socio-economic details, prevalence of tobacco smoking was also low, and the
reproductive history, self-reported weight and height, family proportion of women in each group who had never smoked
history of breast and ovarian cancer, and consumption of was similar (89%). Neither smoking nor drinking was
alcohol and tobacco. The cohort was followed-up to 2001 by associated with the disease.
the two cancer registries for cancer incidence and vital status Reported family income was low by Western standards,
of all cases. The case-finding procedures of both registries ranging from 400 to 80,000 pesos per month (at the time of
were enhanced to reduce the time to registration, and cases the study, the exchange rate was US$1 ¼ 30 pesos) with per
identified during the follow-up period linked to the cohort. 7 capita income (family income divided by the number of
cohabitants) varying from 50 to 30,000 pesos. This variable
This nested case-control study included 123 incident breast
was not associated with the risk of breast cancer. Conversely,
cancer cases identified in the intervention cohort of 151,168
Epidemiology educational level, which can also be used as a proxy for SES,
women to the end of 2001, 32 of which had been detected
was a significant predictor of risk. Cases were more educated
through the screening process. For each cancer case, eight
than controls (Table 1) with 18.7% completing a tertiary
controls were chosen at random from the intervention
education compared with 14.1% controls. The risk of breast
cohort, to match cases by municipality of residence, date of
cancer almost doubled for women who had received a
CBE (6 3 months), and age (6 3 years); 978 controls were
tertiary education compared with those receiving only
selected. In 2001, an attempt was made to contact cases and
minimal education (OR 1.8, 95% CI 1.0–3.6). Breast cancer risk
controls to assess vital status and, if alive, re-interview them
also increased for those who continued education after 13
using the same questionnaire. Of these, by case and control
years of age (OR 1.9, 95% CI 1.1–3.3).
status, respectively, 29% (36) and 49% (480) had moved and
could not be traced; 42% (52) and 2% (20) had died; 7% (8)
and 12% (114) refused the interview; the remaining 22% (27) Reproductive factors
and 37% (359), were re-interviewed. This second interview Cases were more likely to be nulliparous (25.2% vs. 13.5%),
was used to determine the reproducibility of the have their first full-term pregnancy (FFTP) at age 30 or over
questionnaire and the information from the two (16.3% vs. 7.2%) and have an age at menarche below 13 years
questionnaires was pooled. The proportion of records having (30.9% vs. 25.9%) though the mean age at menarche was
missing information, however, remained high. similar (13.5 vs. 13.7 years) (Table 1). There was a non-
We used unconditional logistic regression in STATA 10 statistically significant decrease in breast cancer risk with a
(Stata Corporation, College Station, TX) to estimate odds later age at menarche (p ¼ 0.49). Age at FFTP and parity were
ratios (ORs) and 95% confidence intervals (CI) for individual the only variables strongly associated with breast cancer.
When analysed separately, the ORs declined steadily to 0.2

Int. J. Cancer: 126, 515–521 (2010) VC 2009 UICC

 1097021

(95% CI 0.1–0.4) for five or more pregnancies compared to 30 or over. Compared to an early pregnancy, nulliparous
none. Among parous women, the ORs increased with women had a risk similar to that of women aged over 30 at
increasing age at FFTP (p for trend <0.001) reaching an OR of FFTP (crude OR 4.4, 95% CI 2.0–9.8). When parity and age
5.4 (95% CI 2.3–12.8) in those who had their first child at age
Table 1. ORs for breast cancer, adjusted for age and municipality (*); adjusted for age, municipality, parity, age at FFTP, education ( y)
Case Control
(n5 123) (n5 978) Crude* Adjusted†
Variable
n % n % OR 95%CI p value* OR 95%CI p value†

Age stopped education <13 34 27.6 430 44.0 1.0 1.0

13þ 71 57.7 521 53.3 1.9 1.2–3.0 ph < 0.001 1.9 1.1–3.3 ph ¼ 0.020

missing 18 14.6 27 2.8 9.6 4.7–19.6

Level of education minimal 39 31.7 482 49.3 1.0 1.0

vocational/high 43 331 33.8 1.7 1.0–2.8 1.8 1.0–3.0

35.0 school
tertiary 23 18.7 138 14.1 2.2 1.2–4.0 pt < 0.001 1.8 1.0–3.6 pt ¼ 0.049

missing 18 14.6 27 2.8 9.2 4.6–18.6

Age at menarche mean [SD] 13.5 [1.8] 13.7 [1.7]

<13 38 30.9 253 25.9 1.0 1.0

13þ 84 68.3 714 73.0 0.8 0.5–1.2 ph ¼ 0.235 0.8 0.5–1.4 ph ¼ 0.492

missing 1 0.8 11 1.1

Parity 0 31 25.2 132 13.5 1.0 1.0

1–2 30 24.4 171 17.5 0.7 0.4–1.3 0.4 0.2–1.0

3–4 34 27.6 287 29.3 0.5 0.3–0.9 0.4 0.2–0.9

5þ 19 15.4 311 31.8 0.2 0.1–0.4 pt < 0.001 0.2 0.1–0.4 pt < 0.001

missing 9 7.3 77 7.9 0.5 1.3–8.6

Age at first birth <20 9 7.3 159 16.3 1.0 1.0

20–24 27 22.0 307 31.4 1.6 0.7–3.5 1.3 0.6–2.9

25–29 17 13.8 178 18.2 1.7 0.7–4.1 1.2 0.5–2.9

30þ 20 16.3 70 7.2 5.5 2.3–12.9 pt < 0.001 3.3 1.3–8.3 pt ¼ 0.032

nulliparous 31 25.2 132 13.5 4.6 2.1–10.3 – –

missing 19 15.4 132 13.5 3.3 1.1–8.4

Menopausal status pre 78 63.4 632 64.6 1.0 1.0

post 37 30.1 290 29.7 1.1 0.5–2.0 ph ¼ 0.771 1.3 0.6–2.8 ph ¼ 0.593

missing 8 6.5 56 5.7

OC use never 103 83.7 831 85.0 1.0 1.0

518 Breast cancer risk factors in filipino women
ever 16 13.0 125 12.8 1.0 0.6–1.8 ph ¼ 0.948 0.8 0.4–1.8 ph ¼ 0.595

missing 4 3.3 22 2.2

BMI <25 92 74.8 722 79.0 1.0 1.0

25 18 14.6 142 16.3 1.0 0.6–1.6 ph ¼ 0.898 1.0 0.5–1.7 ph ¼ 0.945

missing 13 10.6 114 4.7 2.6 1.3–5.2 1.0 0.4–2.7

Premenopausal <25 58 74.4 511 80.9 1.0 1.0

25 12 15.4 93 14.7 1.1 0.6–2.2 ph ¼ 0.656 1.1 0.5–2.2 ph ¼ 0.817

missing 8 10.3 28 4.4 3.3 1.3–8.1 1.1 0.3–4.1

Postmenopausal <25 30 81.1 217 74.8 1.0 1.0

25 6 16.2 57 19.7 0.8 0.3–2.1 ph ¼ 0.601 0.8 0.3–2.2 ph ¼ 0.531

missing 1 2.7 16 5.5 –

Smoker never 110 89.4 878 89.8 1.0 1.0

ever 13 10.6 87 8.9 1.2 0.6–2.3 ph ¼ 0.571 1.3 0.6–2.9 ph ¼ 0.486

missing 0 0.0 13 1.3

Drinker no 115 93.5 878 89.8 1.0 1.0

yes 8 6.5 83 8.5 0.7 0.3–1.6 ph ¼ 0.387 0.5 0.2–1.5 ph ¼ 0.191

missing 0 0.0 17 1.7

ph, p-value for heterogeneity; pt, p-value for linear trend.

Table 2. ORs for breast cancer by parity and age at first full-term shows the ORs for combinations of parity and age at FFTP,
pregnancy compared with nulliparous adjusted for age, municipality and adjusted for age, municipality and education, with nulliparous
education
as the reference group. We observed a strong protective
Age at first live birth effect of five or more pregnancies at all levels of age at FFTP
although the risk reduction was largest among women who
Parity <20 20–29 301 had their first child before age 20 (OR 0.1, 95% CI 0.04–0.5).
Nulliparous* As the questionnaire only asked about current
breastfeeding, we were not able to assess breast feeding as a
1–3 OR 0.5 0.3 0.8 risk factor in our analysis. There was no relationship between
95%CIs 0.2–1.5 0.1–0.7 0.4–1.5 increased breast cancer risk and age at last childbirth (p ¼
0.30). There was a non-significant increase in risk (OR 1.5,
4–5 OR 0.3 1.1 0.2 0.8–3.0) with any preterm pregnancies (abortions or
95%CIs 0.03–2.0 0.5–2.3 0.1–0.9 miscarriages) versus none, but this question was only
answered by half the participants. Oral contraceptive (OC) use
5þ OR 0.1 0.2 0.4 was not widespread amongst women in the study with only
95%CIs 0.04–0.5 0.1–0.4 0.1–1.1 13% reporting a positive history of use. Among women who
had used contraception, nearly half reported OC use. There
*Reference category.
was no evidence to support an increased risk with OC use.
Slightly more cases were premenopausal (57.7% vs. 63.6% for
at FFTP were adjusted for each other, the OR associated with controls) and there was some indication of an increased risk
age at FFTP of 30 or over was substantially reduced (OR 3.3, of breast cancer in those women who reported being
95% CI 1.3–8.3), while the protection due to parity was postmenopausal for less than five years (OR 2.6, 1.1–6.3).
unaffected and women with five or more children had one Of the total intervention cohort of 138,392 women, 3000
fifth the risk of nulliparous women. The risk decreased with (2%) reported a positive family history of breast or ovarian
increasing numbers of births (p for trend <0.001). Table 2 cancer. The proportion was similar (2.4%) in this nested case-

Int. J. Cancer: 126, 515–521 (2010) VC 2009 UICC

 1097021
License

control study although 25% cases and 17% controls did not We found no association between BMI and breast cancer
answer this question. Previous benign breast disease was risk in either pre- or postmenopausal women, albeit the latter
reported by 28 women and could not be evaluated. group was based on small numbers. A recent meta-analysis
has shown that, whereas in Western countries, BMI is
Discussion inversely associated with premenopausal breast cancer risk
We report on the relationship between reproductive and but positively associated with postmenopausal breast cancer,
lifestyle factors and breast cancer risk in an urban population in the Asia-Pacific region increased BMI is positively
of South East Asia. The recognised ‘‘classical’’ risk factors do associated with both pre- and postmenopausal breast
not explain this high incidence in comparison to other Asian cancer.21
populations and, in particular, other urban populations such Gibson et al.
as Bombay and Shanghai. Women in Manila still have fertility
rates that are greater than other Asian populations, including
Japan, where breast cancer rates are significantly lower. We confirm the protective effect of full-term pregnancies,
Cancer is the third leading cause of morbidity and mortality which was greater the earlier age at FFTP, These results are
in the Philippines.8 In 2002, there were an estimated 13,000 similar to those found in other populations although the
new cases annually,9 of which 80% had regional involvement strength of the association appeared somewhat stronger than
at presentation,7 leading to around 7600 deaths each year. in Western populations where relative risks for five or more
Breast carcinoma is the most common malignancy in women, children are half of that for nulliparous women. 6,22–24 None of
and the second most common after lung cancer in the the other reproductive factors investigated had a statistically
population, constituting roughly 13% of all cancer significant association with the disease. A previous study
cases.8 carried out in 199425 showed similar results to our study. As
An unusually high incidence of breast cancer in comparison parity and age at first birth have been shown to be risk factors
to rates in other Asian regions, has been observed. The rate in in this Filipino population, it is of interest to consider possible
Manila is considerably in excess of the more typical figures reasons for both the high rate compared to other Asian
seen in other Asian cities for the same period: Osaka populations and the increasing trend.
Prefecture 27.8; Bangkok 25.2, Hanoi 23.3 (aged to 65); Seoul Although fertility in the Philippines has been falling since Epide
21.9; Shanghai 26.7; Hong Kong 34.3 per 100,000. Incidence is the 1950s,26,27 the total fertility rate (TFR) in the 1990s was, at
particularly high in the younger age groups, approaching or 3.5, still higher than the TFRs of other Asian countries with
even exceeding UK rates (22.1 cancers per 100,000 women lower breast cancer rates. For example Malaysia, Indonesia,
aged 30–34 in Manila compared with 22.0 in England and Vietnam, Thailand and Singapore had TFRs of 3.2, 2.8., 2.3, 2.0
Wales, 11.8 in Osaka, 13.8 in Hanoi, 15.2 in Bangkok, 14.7 in and 1.7, respectively.26 There is also wide regional variation
Shanghai and 19.3 in Hong Kong. within the Philippines. Using the 1991–1993 specific fertility
Of the factors investigated, only parity, age at FFTP and rates, the overall TFR was 4.1 but this varied between 2.8 in
educational level were associated with breast cancer risk in Manila to 5.9 in rural Bicol28 and had fallen to 2.5 in Manila by
Epidemiology
this population. The main limitation of our study was the 1998; this is still higher than in European populations.
relatively high proportion of missing data in some of the Two major determinants of fertility are age at marriage
exposures that are normally quite reliably reported, e.g., and contraceptive practices. There has been little change in
miscarriages. A certain level of misclassification may have age at first marriage and, in this aspect, the Philippines differs
reduced the strength of the estimated associations but it is from most other Asian societies. Age at marriage is rather late
unlikely that this caused bias as the interviews were but, as it has changed little over time (23.4, 24.5 and 23.8
conducted before the onset of the disease. The strengths of years in 1968, 1978 and 1988, respectively), it is now lower
the study are the absence of differential misclassification bias than in some neighbouring Asian countries where the age at
and a high degree of representativeness of the study base marriage has risen steadily over the same period. 26 In the
since the cohort was recruited in a population-based survey. same time period, there has been a threefold increase in
In economically advanced countries, breast cancer risk is contraceptive use although use of hormonal contraception
known to increase with early menarche, late menopause, low remains a rather uncommon practice; according to the 1995
parity and delayed first pregnancy.10 Other census survey, only 11.2% of the female population were
reproductiverelated factors associated with the disease users.29 It is interesting to note that there is also a relatively
include the use of OCs11 and hormone replacement therapy,12 high proportion of women in the Philippines who remain
and breastfeeding.13 Lifestyle factors influencing breast cancer childless at the end of their reproductive years (estimated to
risk include excess body weight in postmenopausal women, 14 be 9–10% compared to 3–4% in other Asian societies). 26
diet15 and alcohol consumption.16 Certain benign breast Nulliparous women represented 13% of our cohort of Manila
diseases,17 a family history of breast cancer,18 height,19 and residents.
SES20 are also risk factors. Breast cancer is more common in more affluent societies
as well as in women of higher SES level within populations. 30
520 Breast cancer risk factors in filipino women
In our study, when income was used as a proxy for SES, it was Metro Manila, especially among those of higher SES,
not associated with risk, but when educational level was used resembles that of cities in the US, and so the effect of citizens
as a proxy, there was a significant increase in risk. Breast from rural areas moving to Manila is similar to migration to
cancer incidence is rising in many countries including those in the US. The US colonised the Philippines from 1899 to 1940
Asia where previously risk was low. 1 In populations of south and is still a considerable influence today in many aspects of
and east Asia, increases in rates range from 1 to 3.6% per life. There have been found to be modest but positive
year5,31–33 and are often more marked in younger generations associations between consumption of refined carbohydrates
of women.1,34,35 In many countries in Asia, for example Korea, and breast cancer risk.42 There is a plethora of US-style fast
Taiwan, Singapore, society has changed markedly in the past food outlets, high-calorie soft drinks are widely consumed and
30 years due to rapid economic devel- so the diet of many inhabitants of Metro Manila may
resemble that of the US rather than rural
Philippines.
Int. J. Cancer: 126, 515–521 (2010) VC 2009 UICC
Which components of diet affect breast cancer risk and
519
how they do this remain poorly understood. The most
authoritative and recent review of the subject identified only
excess body weight and lack of physical exercise, or energy
opment and the corresponding increase in standard of living.
balance, as clear determinants of the disease.43 In this
In western countries, increasing SES is a risk factor for breast
population, obesity was rare and overweight uncommon;
cancer.30 Although urbanisation has been rapid with 52% of
however incidence rates of breast cancer were already high
the population living in cities in 1996 compared to 33% in
when the study was conducted. Our results confirmed that
1970,36 the Philippines have not experienced the same
excess body weight was not a major determinant of risk in
economic boom as many other countries in their region.
this population. Use of hormonal treatment for menopausal
Therefore the incidence of breast cancer cannot be explained
symptoms, and use of other exogenous hormones were
by improved SES.
negligible. In addition, 91% of Filipinas living in urban areas
In Filipino women in Hawaii, a study examining trends of
were lifelong alcohol abstainers (90% in our study).44
breast cancer incidence and risk factor prevalence over 25
years observed changes in reproductive behaviour and Our work shows that it is simplistic to ascribe increasing
increasing obesity.37 It is well-established that when Chinese, rates of breast cancer in low- and middle-income countries to
Japanese, or Filipino women migrate to the US, their breast the ‘westernisation’ of lifestyles. This fails to recognise that
cancer risk rises with each subsequent generation until it there are already substantial differences in breast cancer risk
approaches that of Caucasians living in the US, indicating that in those populations which are not obviously explained by
exposures rather than genetics are involved.38 Incidence rates differences in those habits that define ‘westernisation’. For
of Filipinos in the US39 and California40 are rather higher than this reason, aetiological research in Asian and other
those of other Asian-origin groups, despite the fact that most middleincome countries offers a powerful opportunity to test
are relatively recent immigrants to the US. 41 One possible hypotheses developed in high risk populations and thus could
hypothesis is that the lifestyle with respect to eating habits in provide much needed new ideas.
Acknowledgements C150/A5660 and would like to thank Julian Peto and Isabel dos Santos
The study was funded by the US Army Medical Research and Material Silva for allowing her time to work on this paper, as well as their Command,
grant number DAMD17-94-J-4327. Lorna Gibson is invaluable comments. The authors thank the staff of the Manila and supported by Cancer Research
UK programme grant number CR-UK- Rizal Cancer registries.
References HP. Breast cancer in Singapore: trends in Mortality and Prevalence Worldwide ed.,
incidence 1968–1992. Int J Epidemiol vol. 2004. Lyon 2004: IARCPress, 2002.
1. Bray F, McCarron P, Parkin DM. The 1996; 25: 40–5. 10. MacMahon B. Epidemiology and the
changing global patterns of female breast 6. Sim X, Ali RA, Wedren S, Goh DL, Tan CS, causes of breast cancer. Int J Cancer 2006;
cancer incidence and mortality. Breast Reilly M, Hall P, Chia KS. Ethnic differences 118:2373–8.
Cancer Res 2004;6:229–39. in the time trend of female breast cancer 11. CGHFBC. Breast cancer and hormonal
2. IARC. Cancer incidence in five continents. incidence: Singapore, 1968– 2002. BMC contraceptives: collaborative reanalysis of
Volume VIII. IARC Sci Publ 2002:1–781. Cancer 2006;6:261. individual data on 53 297 women with
3. IARC. Cancer Incidence in Five Continents. 7. Pisani P, Parkin DM, Ngelangel C, Esteban breast cancer and 100 239 women without
Age-standardized incidence rates, four- D, Gibson L, Munson M, Reyes MG, breast cancer from 54 epidemiological
digit rubrics, and agestandardized and Laudico A. Outcome of screening by [Link] Group on Hormonal
cumulative incidence rates, three-digit clinical examination of the breast in a trial Factors in Breast Cancer. Lancet 1996;347:
rubrics. IARC Sci Publ 1992:871–1011. in the Philippines. Int J Cancer 2006;118: 1713–27.
4. Lee HP, Duffy SW, Day NE, 149–54. 12. IARC. Combined Estrogen-Progestogen
Shanmugaratnam K. Recent trends in 8. PCS, Philippine Cancer Facts and Estimates, Contraceptives and combine
cancer incidence among Singapore 2005. EstrogenProgestogen Menopausal
Chinese. Int J Cancer 1988;42: 159–66. 9. Ferlay J, Bray F, Pisani P, Parkin D. Therapy IARC
5. Seow A, Duffy SW, McGee MA, Lee J, Lee GLOBOCAN 2002: Cancer Incidence, Monographs on the Evaluation of

Int. J. Cancer: 126, 515–521 (2010) VC 2009 UICC

 1097021
License

Carcinogenic Risks to Humans, vol. 91 observational studies. Lancet 2008;371: Hildesheim A, Nomura AM, West DW,
Lyon: WHO-IARC, 2007:528. 569–78. Wu-Williams AH, Kolonel LN, Horn-Ross PL,
13. CGHFBC. Breast cancer and 22. MacMahon B, Cole P, Lin TM, Lowe CR, Rosenthal JF, Hyer MB. Migration
breastfeeding: collaborative reanalysis of Mirra AP, Ravnihar B, Salber EJ, Valaoras Gibson et al.
individual data from 47 epidemiological VG, Yuasa S. Age at first birth and breast
studies in 30 countries, including 50302 cancer risk. Bull World Health Organ 1970;
women with breast cancer and 96973 43:209–21.
patterns and breast cancer risk in
women without the disease. Lancet 2002; 23. Gao YT, Shu XO, Dai Q, Potter JD, Brinton
AsianAmerican women. J Natl Cancer Inst
360:187–95. LA, Wen W, Sellers TA, Kushi LH, Ruan Z,
1993; 85:1819–27.
14. Lahmann PH, Hoffmann K, Allen N, van Gils Bostick RM, Jin F, Zheng W. Association of
36. US. Population Trends Philippines. In: US
CH, Khaw KT, Tehard B, Berrino F, menstrual and reproductive factors with
Department of Commerce EaSA, ed., vol.
Tjonneland A, Bigaard J, Olsen A, Overvad breast cancer risk: results from the
PPT/92-11RV2: Bureau of the Census,
K, Clavel-Chapelon F, et al. Body size and Shanghai Breast Cancer Study. Int J Cancer
1996.
breast cancer risk: findings from the 2000;87:295–300.
37. Maskarinec G, Zhang Y, Takata Y, Pagano
European Prospective Investigation into 24. Yoo KY, Kim Y, Park SK, Kang D.
I, Shumay DM, Goodman MT, Le
Cancer And Nutrition (EPIC). Int J Cancer Lifestyle, genetic susceptibility and future
Marchand L, Nomura AM, Wilkens LR,
2004;111: 762–71. trends of breast cancer in Korea. Asian Pac
Kolonel LN. Trends of breast cancer
15. Linos E, Holmes MD, Willett WC. Diet and J Cancer Prev 2006;7:679–82.
incidence and risk factor prevalence over
breast cancer. Curr Oncol Rep 2007;9: 31– 25. Ngelangel C, Lacaya L, Cordero C, Laudico
25 years. Breast Cancer Res Treat 2006;98:
41. A. Risk Factors for Breast Cancer Among
45–55.
16. Hamajima N, Hirose K, Tajima K, Rohan T, Filipino Women. Phil J Internal Medicine
38. Ziegler RG, Hoover RN, Nomura AM, West
Calle EE, Heath CW, Jr, Coates RJ, Liff JM, 1994;32:231–6.
DW, Wu AH, Pike MC, Lake AJ,
Talamini R, Chantarakul N, Koetsawang S, 26. Costello M, Casterline J, Fertility decline in
Horn-Ross PL, Kolonel LN, Siiteri PK,
Rachawat D, et al. Alcohol, tobacco and the Philippines: Current status, future
Fraumeni JF, Jr. Relative weight, weight
breast cancer–collaborative reanalysis of prospects. United Nations, Department of
change, height, and breast cancer risk in
individual data from 53 Economic and Social Affairs, Population
Asian-American women. J Natl Cancer Inst
epidemiological studies, including 58,515 Division, 2002.
1996;88:650–60.
women with breast cancer and 95,067 27. Philippines.1998 National Demographic
39. Miller B, Kolonel L, Bernstein L, Young J,
women without the disease. Br J Cancer and Health Survey. In: Office NS, ed. NDHS
Jr, Swanson G, West D, Key C, Liff J,
2002;87:1234–45. Manila, 1998.
Glover C, Alexander G. Racial/Ethnic
17. Hartmann LC, Sellers TA, Frost MH, Lingle 28. Westley S, Palmore J, Retherford R,
Patterns of Cancer in the United States Epide
WL, Degnim AC, Ghosh K, Vierkant RA, Explaining Regional Fertility Variations in
1988–1992. In: Institute NC, ed., vol. 96–
Maloney SD, Pankratz VS, Hillman DW, the Philippines. East-West Center Program
4104 Bethesda, MD: NIH Pub No, 1996.
Suman VJ, Johnson J, et al. Benign breast on Population, 1996.
40. Curado M, Edwards B, Shin H, Storm H,
disease and the risk of breast cancer. N 29. [Link]. Family Planning
Engl J Med 2005;353:229–37. Ferlay J, Heanue M, Boyle P. Cancer
Survey, 1995–1997 National Statistics
18. CGHFBC. Familial breast cancer: Office Manila, 1997. Incidence in Five Continents Vol IX IARC
collaborative reanalysis of individual data Scientific Publications Lyon: IARC No 160,
30. IARC. Social Inequalities and Cancered, vol.
from 52 epidemiological studies including 2007.
138. Lyon: IARC Press, 1997.
58,209 women with breast cancer and 41. Hedeen AN, White E, Taylor V. Ethnicity
31. Yeole BB, Jayant K, Jussawalla DJ. Trends in
101,986 women without the disease. and birthplace in relation to tumor size
breast cancer incidence in greater
Lancet 2001;358:1389–99. Bombay: an epidemiological assessment. 521
Epidemiology
19. Rinaldi S, Key TJ, Peeters PH, Lahmann PH, Bull World Health Organ 1990;68:245–9.
Lukanova A, Dossus L, Biessy C, 32. Jin F, Devesa SS, Chow WH, Zheng W, Ji BT,
Vineis P, Sacerdote C, Berrino F, Panico S, Fraumeni JF, Jr, Gao YT. Cancer incidence and stage in Asian American women with
Tumino R, et al. Anthropometric trends in urban Shanghai, 1972– 1994: an breast cancer. Am J Public Health 1999;89:
measures, endogenous sex steroids and update. Int J Cancer 1999;83: 435–40. 1248–52.
breast cancer risk in postmenopausal 33. Leung GM, Thach TQ, Lam TH, Hedley AJ, 42. Augustin LS, Dal Maso L, La Vecchia
women: a study within the EPIC cohort. Int Foo W, Fielding R, Yip PS, Lau EM, Wong C, Parpinel M, Negri E, Vaccarella S,
J Cancer 2006; 118:2832–9. CM. Trends in breast cancer incidence in Kendall CW, Jenkins DJ, Francesch S.
20. Colditz G, Baer H, Tamini [Link] Cancer. Hong Kong between 1973 and 1999: an Dietary glycemic index and glycemic load,
In: Fraumeni Sa. Cancer Epidemiology and age-period-cohort analysis. Br J Cancer and breast cancer risk: a casecontrol
Prevention, Third ed. New York: Oxford 2002;87:982–8. study. Ann Oncol 2001;12: 1533–8.
University Press, 2006:995–1012. 34. Parkin DM, Bray FI, Devesa SS. Cancer 43. WCRF/AICR. Food, Nutrition, Physical
21. Renehan AG, Tyson M, Egger M, Heller RF, burden in the year 2000. The global Activity and the Prevention of Cancer: a
Zwahlen M. Body-mass index and picture. Eur J Cancer 2001;37 Suppl 8: S4– Global Perspective. In: Report WAE, ed.
incidence of cancer: a systematic review 66. [Link] 2007.
and meta-analysis of prospective 35. Ziegler RG, Hoover RN, Pike MC, 44. [Link]/infobase. World Health
Survey, Philippines 2003.
Int. J. Cancer: 126, 515–521 (2010) VC 2009 UICC