Abstract A43: Population attributable risk of postmenopausal breast cancer according to known and modifiable breast cancer risk factors

Abstract

Introduction: Multiple risk factors for postmenopausal breast cancer are well established. We aimed to determine the proportion of breast cancers in the population that are attributable to established risk factors, and to different combinations of risk factors, to better understand the proportion of breast cancer which could be prevented by changes in modifiable risk factors.

Methods: We used data from the Nurses' Health Study (NHS), an ongoing prospective study of 121,700 US nurses initiated in 1976. Information on breast cancer risk factors and breast cancer diagnosis is collected every two years through mailed questionnaires. Follow-up for this cohort of women through 2008 has been greater than 90%. The analysis was restricted to postmenopausal women. Relative risks and their 95% confidence intervals were estimated by Cox models stratified by calendar year and age in months at the time of each questionnaire return using the counting process data structure to facilitate the incorporation of time-varying covariates. We included the following established breast cancer risk factors in the analyses: age, age at menarche, height, age at first birth/parity, current adult BMI/weight change since age 18, alcohol consumption, physical activity, breast feeding, menopausal hormone therapy, family history of breast cancer, and prior benign breast disease. Full and partial population attributable risk percents (PARs) were calculated. We conducted analyses for total invasive breast cancer, estrogen receptor (ER)+, and ER- breast cancer.

Results: Over the course of follow-up, 8,421 postmenopausal invasive breast cancer cases developed over 2,424,778 person years between 1980 and 2008. As expected, earlier age at menarche, increasing weight gain since age 18, increasing alcohol consumption, current use of menopausal hormones, and being tall were all significantly positively associated with risk of breast cancer in multivariate models. Because age has a large influence on the risk of breast cancer we estimated the PAR both with and without age. When all of the risk factor variables except for age were include in the model the PAR for all invasive breast cancers was 0.62 (95% CI 0.46-0.75). Thus, if women could change to the lowest risk category for all of the variables considered, the number of invasive breast cancers would be reduced by 62%. In models that additionally included age, the PAR was 0.92 (95%CI 0.76-0.97) for total invasive and 0.95 (95%CI 0.75-0.99) for ER+ breast cancer and 0.86 (95%CI 0.21-0.98) for ER- breast cancer.

However, because many of the risk factors considered are not easily modifiable, we also examined more readily modifiable risk factors singly and in combinations. The risk factor with the largest PAR, was BMI/weight gain since age 18 (PAR=0.24(95%CI 0.13-0.34)). Reproductive risk factors such as age at first birth/parity (PAR=0.10 (95%CI 0.07-0.14)) and age at menarche (PAR=0.08 (95%CI 0.03-0.12)) also had modest impact on the PAR, both of which were higher for ER+ breast cancer than ER- breast cancer. When considering all of the modifiable risk factors together we found that changing one's risk factor profile to have the lowest weight gain, no alcohol consumption, high physical activity, breast feeding and no hormone use was associated with a PAR of 0.38 (95%CI 0.23-0.52) for overall invasive breast cancer.

Conclusion: The established breast cancer risk factors explain a large proportion of both ER+ and ER- postmenopausal breast cancers in the population. The most important modifiable risk factor for postmenopausal breast cancer prevention is maintaining a healthy weight.

Citation Format: Rulla M. Tamimi, Molin Wang, Mathew Pazzaris, Stephanie A. Smith-Warner, Walter C. Willett, A. Heather Eliassen, Donna Spiegelman, David J. Hunter. Population attributable risk of postmenopausal breast cancer according to known and modifiable breast cancer risk factors. [abstract]. In: Proceedings of the Thirteenth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2014 Sep 27-Oct 1; New Orleans, LA. Philadelphia (PA): AACR; Can Prev Res 2015;8(10 Suppl): Abstract nr A43.