Cancer Causesand Control 2, 31 - 36

Parental age at birth and risk of breast cancer in daughters: a prospective study among US women

G r a h a m A. C o l d i t z , W a l t e r C. W i l l e t t , Meir J . S t a m p f e r , C h a r l e s H . H e n n e k e n s , B e r n a r d R o s n e r , a n d F r a n k E. S p e i z e r

(Received 29 October 1990; accepted in revised form 15 November 1990) We examined the relation between parental a g e at birth and risk of breast cancer among daughters in a population of 118,309 US women who were 30 to 55 years of age in 1976 and without prior diagnosis of cancer. During 1,140,239 person-years of follow-up, we documented 1,799 incident cases of breast cancer in this population. After adjusting for established breast cancer risk factors, we observed only a weak and nonsignificant trend in risk of breast cancer with increasing maternal age at birth and no relation for paternal age. After adjusting for other risk factors, the chi trend was 1.10, P = 0.27 for increasing maternal age at birth. Daughters born to mothers 30 to 34 years of age had an age-adjusted relative risk of breast cancer of 1.11 (95% confidence interval: 0.89, 1.37) compared to daughters born to mothers less than 20 years of age. The weak posidve trend in risk with increasing maternal age was present among both pre- and postmenopausal women. These findings suggest that there is little or no association between maternal age and risk of breast cancer, and that paternal age is not related to risk of breast cancer.

Key words: Breast cancer, maternal age, paternal age, prospective study, USA.

Introduction Intra-uterine exposure to radiation causes leukemia and other tumors in children I and intra-uterine exposure to DES causes vaginal adenocarcinoma in adolescent women, 2 demonstrating that prenatal environmental factors may contribute to carcinogenesis. Several investigators have described a modestly increased risk (up to 25 percent) of breast cancer among women born to older mothers, 3'4'5'6'7 although others 8'9 have failed to find such an association. Recent analysis of data from the Cancer and Steroid Hormones Study 1° has again drawn attention to this relation. Trichopoulos 11 has outlined mechanisms that are consistent with such an association, including prenatal exposure to high elevels of endogenous estrogens among older pregnant women that may increase the risk of gonadal germ cell tumors. 12

Moreover, total blood concentrations of estrogens in pregnancy are significantly lower (by 30 percent) in younger women, aged less than 20 years, than in older women 20 to 29 years of age. 13 Germ cell mutations are, however, more strongly related to paternal age than to maternal age. 14'15 In all but three of the previous studies, investigators have limited their inquiry to the association with maternal age. Henderson 5 observed an association for maternal age but not for paternal age in a study that included 308 cases. Janerich et al 6 reported a positive relation for both maternal and paternal age at birth, but had limited power to assess the independent effects of maternal and paternal age in an analysis that was based on 801 cases of breast cancer. Le Marchand et al 7 observed that the mean maternal age of cases was

Authors are with the Department of Medicine, Brigham and Women's Hospital (GC, CH, FS, MS, IUVY)and Harvard Medical School (GC, CH, FS), and the Departments of Epidemiology (MS, WgU), Nutrition (W~V), and Biostatistics (BRJ, HarvardSchool of Public Health. Address correspondence to Dr Colditz at the Channing Laboratory, 180 LongwoodAvenue, Boston, MA 02115 5899, USA. This research was supported by PublicHealth Servicegrants CA40935 and CA40356from the National CancerInstitute, Nationallnstitutes of I-tealth, Department of Health and Human Services. © 1991 Rapid Communications of Oxford Ltd

31

G. A. Colditz et al not significantly greater than that of controls (26.9 years cf26.5 years, P < 0.43), and the mean paternal age for cases likewise was not significantly greater for cases than for controls (mean age 31.8 years for cases and 31.4 years for controls, P = 0.55). To address the independent relations of maternal and paternal age at birth, we undertook an analysis of prospectively collected data from the Nurses' Health Study.

Methods The Nurses' Health Study cohort was established in 1976, when 121,700 female registered nurses, 30 to 55 yeats of age, completed a mailed questionnaire that included items about known or suspected risk factors for cancer and cardiovascular diseases. Baseline information included: previous cancer history, age at menarche, parity, age at first pregnancy lasting six months or more, age at menopause, history of benign breast disease, maternal history of breast cancer, and sororal history of breast cancer. In addition, we included questions on height, weight, current and past smoking habits, postmenopausal use of hormones, history of use of oral contraceptive agents, and the year of birth for the participant's mother and father. Every two years, follow-up questionnaires have been mailed to the women to bring up to date the information on risk factors and to ascertain whether major medical events have occurred.

Data on parental age In 1976, women reported the year of birth for their mothers and fathers. We used this year of birth and the year of birth of the participant to calculate the maternal age and paternal age at the birth of the daughter participating in the study. We used five-year age categories for parental age at birth: 39.

Identification o f breast cancer eases On each questionnaire, we inquired whether breast cancer had been diagnosed and, if so, the date of diagnosis. All women who reported breast cancer (or the next of kin for decedents) were contacted for permission to review the relevant hospital records and confirm the self-repotted diagnosis. Pathology reports were obtained for 93 percent of the cases, and were reviewed by physicians blinded to exposure information reported by the participants. The records for all but six of these cases confirmed the self-report of breast cancer. With the exclusion of these six cases and the inclusion of newly diagnosed fatal cases identified through the National Death Index up to June 1986, we identified 1,799 incident cases between the return of the 1976 question32

naire and 1June 1986. Although pathology reports could not be obtained for seven percent of these cases, we based our analysis on all incident breast cancers because the rate of accuracy of self-reporting was extremely high (99.6 percent). 16

Population for analysis The endpoint for this analysis was the occurrence of incident breast cancer. We therefore excluded from the analysis all women who reported breast or other cancer (excluding nonmelanoma skin cancer) on the 1976 questionnaire. This left a total cohort of 118,309 women eligible for follow-up. During the 10 years of follow-up, from the return of the 1976 questionnaire to 1 June 1986, we accrued 1,140,239 person-years. Follow-up of the cohort for the identification of nonfatal breast cancer, by questionnaire and telephone interview, was 90 percent complete based on the person-years accrued up to the last returned questionnaire compared to total person-years. For fatal breast cancer, follow-up was more than 98 percent complete. To examine separately the association among pre- and postmenopausal women, we classified a woman as postmenopausal from the time she returned a questionnaire on which she reported natural menopause or hysterectomy with bilateral oophorectomy. In this cohort, self-report of menopause is highly reproducible and valid with regard to details of the extent of ovarian surgery. 17 In addition, we classified women reporting hysterectomy without bilateral oophorectomy as postmenopausal at the age when natural menopause had occurred in 90 percent of the cohort (54 years for current cigarette smokers and 56 years for nonsmokers). The report of menopause was updated every two years, and the postmenopausal population increased to include women who became postmenopausal and remained free from cancer.

Statistical analysis The primary analysis used incidence rates with personmonths of follow-up as the denominator. For each participant, person-months were allocated according to the 1976 exposure variables and were updated according to information on follow-up questionnaires. For women who developed breast cancer or died, person-months were assigned according to the covariate status reported on the most recently completed questionnaire, but follow-up terminated with the diagnosis of breast cancer or death. Women who reported a diagnosis of cancer other than nonmelanoma skin cancer on the 1978, 1980, or 1984 questionnaire were excluded from subsequent follow-up. If no questionnaire was returned for a follow-up cycle, the most recently recorded covariate data were used for the subsequent follow-up interval.

Breast cancer and parental age Accumulated person-time and newly diagnosed cases of breast cancer were allocated to the status of each variable reported at the beginning of each interval. Agespecific incidence rates (with five-year age intervals) were calculated by dividing the number of breast cancer cases by the person-time of follow-up. We used relative risk (RR) with 95 percent confidence interval (CI) as a measure of association, defined as the incidence of breast cancer among women in a category of parental age at birth divided by the corresponding rate among women who had maternal age less than 20. Additional stratified analyses were carried out to control for risk factors and to explore the possibility that the effect of maternal age might be modified by parity or other factors. We used the Mantel test for linear trend to examine the relation between parental age and risk of breast cancer, TM and proportional-hazards models to adjust for multiple risk factors simultaneously. 19 We recorded alcohol intake for the first time in 1980; 20 hence, analyses including data on alcohol intake were limited to women with breast cancer diagnosed between the return of the 1980 questionnaire and 1 June 1986. To test for interactions between maternal age, paternal age, and breast cancer risk factors, we fitted a continuous interaction term to the proportional hazards model.

Results In 1976, information on maternal year of birth was provided by 97.9 percent of the cohort, and paternal year of birth was provided by 93.5 percent. Compared to women whose maternal age at their birth was less than 20, we observed that risk of breast cancer was positively but only slightly related to maternal age; the trend was of borderline statistical significance (chi trend = 1.90, P = 0.057) (Table 1). Paternal age also tended to be positively related to risk of breast cancer (chi trend = 1.80, P = 0.072). A 15-year increase in maternal age (30 to 34 compared to less than 20) was associated with a RR of 1.16 (CI: 0.93, 1.44), and for a similar increase in paternal age the relative risk was 1.02 (CI: 0.68, 1.54). Because the epidemiology of breast cancer varies by menopausal status, we assessed the association with parental age separately for pre- and postmenopausal women. In both categories, the same pattern emerged, with a slight nonsignificant trend for an association with increasing maternal age, and no association with paternal age (Table 2). We further examined these associations within strata of parity, age at menarche, age at first birth, family history of breast cancer, history of benign breast disease, and body mass index (BMI), and observed little change in the associations across strata of these breast cancer risk factors (Table 3). However, we did note an inverse

Table 1. Parental age at birth and age-adjusted relative risk (RR) and 95 percent confidence interval (CI) of breast cancer among daughters: Nurses' Health Study, 1976 to 1986 Age

Cases

Person-years

RR (CI)

Maternal a 39

104 448 551 385 200 73

68,383 1.0 (reference) 309,126 1.02 (0.82, 1.26) 346,015 1.12 (1.04, 1.38) 231,535 1.16 (0.93, 1.44) 117,337 1.17 (0.92, 1.48) 44,157 1.08 (0.80, 1.46) X, trend = 1.90 (P = 0.057)

Paternal b < 20 2 0 - 24 25 - 29 30 - 34 35 - 39 >39

24 223 479 445 289 219

14,654 1.0 (reference) 156,590 0.91 (0.60, 1.39) 318,886 0.97 (0.64, 1.46) 277,262 1.02 (0.68, 1.54) 168,262 1.09 (0.72, 1.65) 131,105 1.02 (0.67, 1.55) X, trend = 1.80 (P = 0.072)

alnformation for age of mother was missing for 2,478 women among whom 38 cases of breast cancer were diagnosed during 23,686 personyears of follow-up. bInformation for age of father was missing for 7,681 women among whom 120 cases of breast cancer were diagnosed during 73,570 personyears of follow-up.

Table 2. Parental age at birth and age-adjusted relative risk of breast cancer among daughters for pre- and postmenopausal breast cancer Age Maternal 39 x, trend

(p) Paternal 39 X, trend

(P)

Premenopausal

Postmenopausal

1.0 0.94 1.09 1.19 1.06 0.97

1.0 0.97 1.07 0.97 1.23 1.20

1.35

1.44

(0.18)

(0.15)

1.0 0.94 0.97 0.91 1.03 1.08

1.0 1.10 1.03 1.36 1.33 1.16

1.09

1.68

(0.28)

(0.09)

association for maternal age among women with a positive family history of breast cancer among sisters or mother. Among those with a family history of breast cancer, women with maternal age 35 to 39 had a RR of 33

G. A. Co/ditz et al Table 3. Parental age and relative risk of breast cancer according to status of breast cancer risk factors __

Maternal age 39

Parity Nulliparous Parous

1.0 1.0

1.0 1.0

1.3 1.1

1.2 1.2

1.6 1.2

1.1 1.1

Age at menarche (yrs) ~12 13 >13

1.0 1.0 1.0

1.0 1.3 0.7

1.2 1,2 0,9

1.2 1.4 1.0

1.2 1.5 0.9

1.2 1.1 0.9

Age at first birth (yrs) 29

1,0 1.0 1.0

1,1 0.8 1.4

1,3 0.8 1.3

1.4 0.7 1.6

1.3 0.8 1.6

Familybistoryof bre~t cancer Yes No

1.0 1.0

0.7 1.1

0,9 1.1

0.7 1.2

History of benign breast disease Yes No

1.0 1.0

1.7 0.9

1.8 1.0

Body mass index (kg/m 2) 39

P interaction

0.59

1.0 1.0

0.7 0.9

0.5 0.9

0.8 1.0

0.9 1.0

0.8 1.0

0.38

0.92

1.0 1.0 1.0

0.9 1.1 0.6

1.0 1.2 0.7

1,0 1.3 0.8

1.1 1,3 0.7

1.0 1.2 0.8

0.28

1.2 0.8 1.3

0.50

1.0 1.0 1.0

1.5 0.6 0.7

1,7 0.6 0.7

1.9 0,6 0.7

1.6 0.8 0.8

1.6 0.7 0.8

0.64

0.6 1.2

0.6 1.1

0,34

1.0 1.0

0,8 0,9

0.6 1.0

0.7 1.0

0.5 1.2

0.4 1.1

0,92

1,9 1.1

2.2 1.1

0.9 1,1

0.60

1.0 1.0

1.8 0.8

1.7 0.9

2,0 0.9

2.1 1.0

2,2 0.9

0.17

1.2 1.2 1.1 1.1

1.3 1.1 1.1 0.8

1.0 1.1 1.3 0.6

0.92

1.0 1.0 1.0 1.0

1.6 1.5 0.8 0.7

1.7 1.7 0.7 0.8

2.0 1.7 0.8 0,7

2.1 1.6 0.8 0,8

1.9 1.9 0.8 0.7

0.92

0.61 compared to women with maternal age less than 20, the corresponding RR for those without a family history was 1.22. A similar variation in the association for paternal age was also seen across strata of family history. Maternal age appeared to be more strongly related to risk of breast cancer among nulliparous women, those with a history of benign breast disease, and in leaner women, and paternal age was somewhat weaker among nulliparous women and stronger among women with a history of benign breast disease and leaner women. However, when we tested these interactions in a proportional hazards model, none attained statistical significance. When we controlled for all risk factors simultaneously, the multivariate-adjusted RRs for maternal age were uniformly reduced from the ageadjusted results (Table 4) as was the test for trend (chi trend = 1.10, P = 0.27). Using the beta coefficient from the regression with maternal age as a continuous variable, the adjusted RR for the association between a differential of 15 years in maternal age at birth and risk of breast 34

Paternal age

P

Parental age at birth and risk of breast cancer in daughters: a prospective study among US women.

We examined the relation between parental age at birth and risk of breast cancer among daughters in a population of 118,309 US women who were 30 to 55...
458KB Sizes 0 Downloads 0 Views