B Rosner’s research while affiliated with University of Massachusetts System and other places

The aim of this study was to evaluate the reproducibility and validity of a 61-item semiquantitative food frequency questionnaire used in a large prospective study among women. This form W88 administered twice to 173 particIpants at an interval of approximate'y one year (1980-1981), and four one-week diet records for each subject were collected during that period. Intraclass correlation coeffidents for nutrient intakes estimated by the one-week diet records (range = 0.41 for total Vitamin A without supplements to 0.79 for Vitamin B, with supplements) were similar to those computed from the questionnaire (range = 0.49 for total Vitamin A without 8upplements to 0.71 for sucrose), indicating that these methods were generally comparable with respect to reproducibility. With the exception of sucrose and total carbohydrate, nutilent Intakes from the diet records tended to correlate more strongly with those computed from the questionn&re after adjustment for total caloric Intake. Correlation coefficients between the mean calorie-adjusted intakes from the four one-week diet records and those from the questionna ire completed after the diet records ranged from 0.36 for Vitamin A without supplements to 0.75 for vitamin C with supplements. Overall, 48% of 8ubjects n the lowest quintile of calorie-Adjusted intake computed from the diet records were also in the lowest questionnaire quintile, and 74% were in the lowest one of two questionnaire quintiles. SImilarly, 49% of those in the highest diet record quintile were also in the highest questionnaire quintile, and 77% were In the highest one or two questionnafre quintiles. These data indicate that a simple self-administered dietary questionnaire can provide u8efuI information about indMdual nutrient intakes over a one-year penod.

Obesity is well established as a cause of postmenopausal breast cancer incidence and mortality. In contrast, adiposity in early life is inversely related to breast cancer incidence. To better integrate understanding of these relations, we assess adiposity in childhood, in late adolescence, and in adult years, as well as change in weight in relation to total invasive breast cancer and subtype defined by receptor status. The Nurses' Health Study cohort was established in 1976 when 121,701 female US registered nurses ages 30-55 responded to a mailed questionnaire about risk factors for breast cancer including reproductive factors, hormone use, anthropometric variables, benign breast disease (BBD), and family history of breast cancer. The risk factors have been updated by repeat questionnaires every 2 years. We followed a cohort of 77,232 women from 1980 to 2006 (1,408,188 person-years), with routinely updated risk factor information, documenting 4,254 incident cases of invasive breast cancer. ER and PR status were obtained from pathology reports and medical records. A total of 2,065 ER+/PR+ tumors, 604 ER-/PR- tumors, 520 ER+/PR- tumors were identified among women with complete information on breast cancer risk factors. Weight at age 18 was inversely related to incidence of pre and postmenopausal breast cancer. The relative risk (RR) per 50lb weight difference at age 18 was 0.80 (95% CI = 0.73, 0.88). However, the inverse association is completely explained by weight at age 10. Weight at age 10 was more strongly inversely related to ER-PR- breast cancer, RR per 50lb difference in weight at age 10 = 0.48; 95% CI 0.32 – 0.74. Weight gain from 10 to 18 was not related to risk. After controlling for weight at age 18, weight loss of 10lb or more was significantly related to lower risk of breast cancer (RR=0.80; 0.66, 0.95) overall. The association was stronger for premenopausal breast cancer (RR=0.51; 0.30, 0.88). Long-term weight change was positively related to total incident breast cancer risk and most clearly to postmenopausal cancer (RR per 50lb weight gain since age 18 =1.26; 1.19-1.34). Weight gain during premenopausal years (weight at menopause minus weight at age 18), and after menopause, were both directly related to increased risk. Weight gain of 30 or more pounds during premenopause increased risk of postmenopausal breast cancer compared to less than 10lb weight change (RR=1.25; 1.14-1.38) and weight gain of 30 or more pounds after menopause carried similar increase in relative risk (RR=1.24; 1.10-1.41). These results were unchanged after control for weight at age 10, and were stronger for ER+PR+ breast cancer. The associations for long-term weight gain and postmenopausal breast cancer are stronger for never vs. ever users of hormone therapy, but are significantly positively associated in both groups. In conclusion, adiposity in childhood has a protective lifelong relation to breast cancer risk most clearly seen for ER-PR- disease. There are deleterious effects of long-term weight gain both pre- and post-menopause. Weight loss in premenopausal years significantly reduces risk of breast cancer. Weight change can importantly modify breast cancer risk. Citation Format: Colditz G, Eliassen H, Toriola A, Hankinson S, Willett W, Rosner B. Weight change across the life-course and breast cancer risk among pre and postmenopausal women. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P1-07-05.

Background: Male pattern baldness is positively associated with androgens as well as insulin-like growth factor 1 (IGF-1) and insulin, all of which are implicated in pathogenesis of colorectal neoplasia. Methods: From 1992 through 2010, we prospectively followed participants in the Health Professionals Follow-Up Study. Hair pattern at age 45 years was assessed at baseline with five image categories (no baldness, frontal-only baldness, frontal-plus-mild-vertex baldness, frontal-plus-moderate-vertex baldness, and frontal-plus-severe-vertex baldness). Cancer analysis included 32 782 men and used Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Restricted to men who underwent at least one endoscopy over the study period, adenoma analysis included 29 770 men and used logistic regressions for clustered data to estimate odds ratios (ORs) and 95% CIs. Results: Over the mean follow-up of 15.6 years, 710 cases of colorectal cancer (478 for colon, 152 for rectum, and 80 unknown site) developed. Significantly increased risks associated with frontal-only baldness and frontal-plus-mild-vertex baldness relative to no baldness were observed for colon cancer with respective HR being 1.29 (95% CI, 1.03-1.62) and 1.31 (95% CI, 1.01-1.70). Over the 19-year study period, 3526 cases of colorectal adenoma were detected. Evidence for an increased risk of colorectal adenoma relative to no baldness was significant with frontal-only baldness (OR, 1.16; 95% CI, 1.06-1.26) and borderline insignificant with frontal-plus-severe-vertex baldness (OR, 1.14; 95% CI, 0.98-1.33). Conclusions: Subtypes of male pattern baldness at age 45 years were positively associated with colorectal neoplasia. Future studies are warranted to confirm our results and to determine the predictive value of male pattern baldness to identify those at high risk for colorectal neoplasia.

Few prospective studies have investigated the relationship between pre-diagnostic obesity, smoking and prostate cancer (PCa) survival by timing of measurement, by age at diagnosis, and evaluated the interaction between obesity and smoking. We conducted a multinational survival analysis among 10,106 PCa cases (1,007 PCa deaths and 2,893 total deaths) from eight cohorts with an average of 8.2 years of follow up. Hazard ratio (HR) of PCa death was estimated using Cox proportional hazard model, adjusting for age, alcohol intake, diabetes status, cohort and duration between baseline and diagnosis and subsequently adjusted for tumor stage and grade. Higher prediagnostic BMI was related to higher risk of PCa death. With each 5 kg/m2 increase in BMI, the multivariate HR for PCa death was 1.08 (95% CI, 1.02-1.14) among overall participants (p-trend = 0.01) and 1.33 (95% CI, 1.18-1.51) among never or former smokers (p-trend < 0.001). This positive trend for PCa mortality was mainly observed among men with BMI measured more than 5 years before diagnosis, and among those age >65 years old at diagnosis. Compared with never smokers, current smokers had significantly elevated risk of PCa death, with a HR of 1.92 (95% CI, 1.52-2.43) regardless of the time of measurement, age at diagnosis and BMI. After further adjusting for tumor stage and grade, the association between BMI, smoking and PCa death was attenuated but remained statistical significant. In this consortium study of eight large cohorts, smoking and overweight/obesity before diagnosis were significant predictors for subsequent PCa-specific mortality. Smoking significantly modifies the association of BMI and PCa-specific mortality. ©2015 American Association for Cancer Research.

10518 Background: Although current use of hormone replacement therapy (HRT) is associated with a greater risk of breast cancer than past use, the effects of longer term past use of HRT are not well quantified. Methods: We examined the relationship between past HRT use and invasive breast cancer risk within the Nurses’ Health Study, a prospective cohort of 121,700 registered nurses aged 30–55 in 1976 who update information on cancer risk factors and outcomes through biennial questionnaires. For this analysis, the follow-up period was 1980 through 2004 and only included person-time for postmenopausal women. Status of HRT use (never, past, or current), type of HRT (oral unopposed estrogen (E alone) or combination oral estrogen + progesterone (E+P)), and duration of HRT use were assessed every 2 years and defined prospectively. Proportional hazards models controlled for age, body mass index, parity, age at first birth, family history of breast cancer, benign breast disease, alcohol consumption, type of menopause, and ages at menarche and menopause. Results: During 1,388,368 person-years of follow-up among postmenopausal women, invasive breast cancer was diagnosed among 1,554 women who had never used HRT, 459 past users of E+P and 527 past users of E alone. Regardless of duration of use, past use of E alone was not associated with breast cancer risk. Although past use of E+P for less than 10 years was not associated with breast cancer risk (RR (95% CI) 1.01 (0.88–1.17) for < 5 years and 0.95 (0.74–1.21) for 5–9.9 years), past use of E+P for greater than 10 years was associated with an increased risk of breast cancer (RR 1.53 (1.09–2.16) for 10–14.9 years and 1.48 (0.87–2.51) for 15 or more years). Results were similar regardless of time since last use, although power was limited for this subgroup analysis. Conclusions: Past use of E+P for greater than 10 years was associated with a persistent elevation in breast cancer risk. The relationship between duration of past use and breast cancer risk did not appear linear, but suggested a possible threshold effect. No significant financial relationships to disclose.