Risk Factors by Molecular Subtypes of Breast Cancer: a Pooled Analysis of Nine Cohorts

Etiological differences between molecular subtypes of breast cancer, defined by estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2), have been examined in previous studies but results are inconsistent, most likely due to small numbers of rarer subtypes. In a recent systematic review of these studies, only six of the possible 44 associations had consistent findings, primarily with the most common luminal A subtype (ER+ or PR+/HER2-). To provide larger numbers from prospective studies for these associations, we utilized a harmonized dataset of nine cohort studies (Cancer Prevention Study-II Nutrition Cohort, Melbourne Cancer Cohort Study, the National Cancer Institute-American Association of Retired Persons cohort study, Nurses’ Health Study, Nurses’ Health Study-2, Prostate, Lung, Colorectal, and Ovarian Cancer Screening cohort, Swedish Mammographic Cohort, Swedish Women’s Lifestyle and Health Study, and Women’s Health Initiative) that had case data on ER, PR, and HER2 from medical records or state tumor registry records. Multivariate, joint Cox proportional hazard regression models were used to calculate hazard ratios (HR) and 95% confidence intervals (CI), and to compare associations across subtypes. Among 616,109 study participants, 11,861 invasive breast cancer cases were diagnosed, including 8,777 luminal A, 1,388 luminal B (ER+ or PR+/ HER2+), 532 HER2-enhancing (ER-/PR-/HER2+), and 1,164 triple negative (ER-/PR-/HER2-) subtypes. The mean age at baseline of the study participants was 54.8 (SD=12.2) years for cases and 54.2 (SD=14.4) years for controls, and the majority self-described as white. We found that the associations of race, benign breast disease, parity, number of live births, age at first birth, age at menopause, alcohol consumption, and smoking initiation relative to first birth were statistically-significantly different for at least one of the molecular subtypes compared to the associations for the ER+ or PR+/HER2- subtype (p-value for tumor homogeneity Citation Format: Mia M. Gaudet, Gretchen Gierach, Brian Carter, Juhua Luo, Roger Milne, Elisabete Weiderpass, Graham Giles, Wendy Chen, Rulla Tamimi, Heather Eliassen, Diane Feskanich, Alicja Wolk, Hans-Olov Adami, Karen Margolis, Susan Gapstur, Montserrat Garcia-Closas, Louise Brinton. Risk factors by molecular subtypes of breast cancer: a pooled analysis of nine cohorts [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2278. doi:10.1158/1538-7445.AM2017-2278