Abstract C083: Meta-analysis reveals differences in somatic alterations by genetic ancestry across common cancers

Abstract

Abstract Genetic similarity of populations (or genetic ancestry) contributes to cancer driving alterations. We performed a meta-analysis leveraging two real-world cohorts of targeted gene panel sequencing, comprising 275,605 tumor samples, to investigate ancestry-associated somatic alterations across 14 common cancers. We focused on five continental ancestries – European (EUR), African (AFR), East Asian (EAS), South Asian (SAS), and admixed American (AMR). Genetic ancestry was inferred using single nucleotide polymorphism markers from captured regions of gene panel sequencing in each cohort. The combined cohort from Memorial Sloan Kettering – Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) and Foundation Medicine (FMI) contains a substantial number (45,131) of non-EUR samples. Within each cohort, we performed multivariate logistic regression to test association of each cancer gene alteration with proportion of genetic ancestry as a continuous variable, controlling for relevant covariates (age, sex, gene panel version, and histological subtypes) for every combination of ancestry and cancer type. Our analysis focused on oncogenic or likely oncogenic alterations spanning mutations, copy number alterations, and structural rearrangements for each gene. We then aggregated the logistic regression results from the two cohorts using a fixed effects model of meta-analysis weighted by cohort sample size. We found 444 significant associations involving 116 unique genes with consistent logistic regression coefficients between cohorts and False Discovery Rate (FDR)-adjusted p-value < 0.1. These genes were involved in cancer-specific and cross-cancer associations. TP53, KRAS, and TERT were the top three recurrent ancestry-associated genes respectively. TERT promoter mutations were depleted in patients of AFR or EAS ancestry, but enriched in EUR ancestry in bladder urothelial carcinoma, glioblastoma, cutaneous melanoma, and hepatocellular carcinoma. Additionally, TP53 mutations were enriched in AFR and EAS ancestry across multiple cancers, consistent with previous reports, but depleted in prostate cancer in both AFR and EAS ancestries. Among the novel cancer-specific associations with FDA-approved targeted therapies, we found that CDK12 mutations in prostate adenocarcinoma were enriched in EAS ancestry while ERBB2 mutations in lung adenocarcinoma were enriched in AMR ancestry. We also found novel potential drug targets/biomarkers associated with AFR ancestry, such as enrichment of BAP1 mutations in head and neck squamous cell carcinoma and depletion of FGFR2 mutations in uterine endometrioid carcinoma. Overall, our work identified ancestry-associated somatic alteration differences of cancer genes in a combined cohort with a large number of non-European samples. This work underscores the value of studying under-represented populations to replicate or reveal new potential targets for precision oncology. Citation Format: Setor Amuzu, Amy Xie, Xuechun Bai, Kelly R. Pekala, David Ma, Tomin Perea-Chamblee, Kanika Aurora, Garrett Frampton, Michael Berger, Nikolaus Schultz, Justin Y. Newberg, Jian Carrot-Zhang. Meta-analysis reveals differences in somatic alterations by genetic ancestry across common cancers [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr C083.