Abstract
BACKGROUND: The influence of age at menarche (AAM) and age at natural menopause (ANM) on breast cancer (BC) risk in BRCA1 and BRCA2 germline pathogenic variant (PV) carriers is uncertain. Observational studies are prone to bias and have limited statistical power. Mendelian randomization (MR) minimises bias and may be used to examine causal effects. METHODS: Two-sample and age-specific MR analyses for BC were performed. For AAM, two-sample multivariable MR and mediation analyses to account for the confounding effect of body mass index (BMI), were undertaken. RESULTS: Genetic scores for ANM and AAM predicted the respective traits in PV carriers. Inverse-variance weighted hazard ratios (HR) for genetically predicted ANM per-year were HR = 0.99 (95%CI:0.97-1.01, p = 0.45) and HR = 1.04 (95% CI:1.01-1.06, p = 0.003) for BRCA1 and BRCA2 PV carriers, respectively. After adjusting for genetic associations with BMI, AAM per-year on BC risk were HR = 0.90 (95%CI:0.83-0.98, p = 0.01) and HR = 0.95 (95%CI:0.86-1.04, p = 0.26) for BRCA1 and BRCA2 carriers, respectively, consistent with a protective effect of later AAM. DISCUSSION: MR analyses support causal associations between ANM and BC risk in BRCA2, but not BRCA1, and between AAM and BC risk in BRCA1 and BRCA2 PV carriers. These results may aid risk prediction models and genetic counselling of carriers.