Genome-wide association and Mendelian randomization analyses of body mass index and breast cancer risk in women of African ancestry.

10598 Background: The relationship between adiposity and breast cancer risk is complex and incompletely understood, particularly among women of African ancestry. We investigated the association between genetically predicted and excess body mass index (BMI) and breast cancer risk in the African Ancestry Breast Cancer Genetic Consortium (AABCG) using complementary genetic and epidemiologic approaches. Methods: We conducted a genome-wide association study (GWAS) of BMI in AABCG. Independent BMI-associated variants were then used as instrumental variables in one-sample Mendelian randomization (MR) analyses, using two-stage residual inclusion with logistic regression to estimate the causal effect of BMI on breast cancer risk. To explore the joint influence of genetic and non-genetic contributions to BMI, we constructed a BMI polygenic risk score and defined the BMI polygenic-measured gap (BMI-PGM) as the difference between the percentile of observed BMI and the percentile of genetically predicted BMI. BMI-PGM was categorized into three groups based on quartile distribution: discordantly low (≤25 th percentile), concordant (25 th -75 th percentile), and discordantly high (≥75 th percentile). The association between BMI-PGM and breast cancer risk was evaluated. Results: We identified thirteen loci that were associated with BMI at the genome-wide significance level. Of these, seven loci had not been reported in prior GWAS of BMI. In the MR analyses (n = 31,522), genetically predicted BMI was inversely associated with overall breast cancer risk: odds ratio (OR) = 0.92 per 5 kg/m 2 increase, 95% confidence interval (CI) 0.86-0.99, p = 0.028. In stratified analyses, there was a stronger inverse relationship observed for estrogen receptor (ER) negative compared with ER positive breast cancer. Inverse associations with similar effect sizes were observed in analyses stratified by menopausal status. Compared with women whose observed BMI was discordantly low relative to their genetically predicted BMI, those with concordant BMI had a modestly increased risk of breast cancer (OR = 1.08, 95% CI 1.02-1.14, p = 0.0074), while women with discordantly high BMI had a 12% higher risk of breast cancer (OR = 1.12, 95% CI 1.05-1.19, p = 0.00049). Conclusions: Genetically predicted higher BMI was associated with lower breast cancer risk, whereas BMI in excess of genetic predisposition was associated with increased risk. These opposing associations highlight a distinction between genetically mediated body size versus excess weight that likely reflects adverse metabolic processes. Together, these results underscore the roles of both genetic susceptibility and modifiable influences on adiposity in breast cancer etiology and highlight the importance of maintaining a healthy weight to reduce breast cancer risk, despite variability in genetic predisposition to adiposity across the population.