Obesity-associated genomic heterogeneity in breast tumors.

e12611 Background: Obesity is associated with increased breast cancer risk and worse outcomes; however, its molecular and ancestry-specific effects across breast cancer subtypes remain poorly defined. This is particularly relevant for aggressive subtypes that disproportionately affect women of African ancestry. We investigated the clinical, molecular, and survival associations of obesity across breast cancer subtypes, with a focus on basal-like disease. Methods: Breast cancer cohorts with available body mass index (BMI), molecular subtype, gene expression, clinical outcomes, and genetic ancestry data were analyzed. Patients were classified as normal weight, overweight, or obese. Associations between BMI and subtype distribution, disease-free survival (DFS), and vital status were assessed using chi-square and survival analyses. Genome-wide correlations between BMI and gene expression were performed, followed by pathway enrichment analyses. Subtype-specific analyses were conducted, including comparisons within Basal and Luminal-B tumors. Findings were examined across multiple cohorts, including Vanderbilt BEST and METABRIC. Results: Among 663 patients, 80.8% were overweight or obese, with obesity significantly enriched in basal-like and luminal B subtypes (p < 0.05). Basal-like and HER2-enriched tumors showed the highest obesity prevalence (70.3% and 62.5%, respectively). Overweight and obese tumors were associated with higher African ancestry, particularly West African ancestry, with the strongest enrichment observed in obese patients with basal-like disease. A total of 533 genes positively correlated with BMI were enriched in metabolic and DNA repair pathways, including glycine, serine, and threonine metabolism and homologous recombination. In contrast, 384 genes negatively correlated with BMI were enriched in ketone body and thiamine metabolism pathways. Among basal-like tumors, overweight and obese patients experienced significantly poorer DFS and exhibited distinct obesity-associated transcriptional profiles. These associations were not consistently observed across all subtypes or in the METABRIC cohort. Conclusions: Obesity and its association with poorer outcomes in hormone-receptor positive breast cancer is well known. However, here we show it is also associated with aggressive breast cancer subtypes which is associated with adverse outcomes in basal-like disease. This appears to be associated with distinct metabolic and DNA repair-related transcriptional programs, particularly among patients with higher African ancestry. These findings will be validated in independent cohorts using clinically annotated data from Caris Life Sciences Molecular Profiling to confirm robustness across diverse populations. Future analyses will also identify obesity-associated somatic mutations using whole-genome sequencing data and characterize genomic-wide alterations in obese breast cancers.