Abstract 6129: Emerging disparities in the clinical actionability landscape for patients with inferred African ancestry

Abstract

Abstract The past decade has seen a surge in the approval and availability of precision oncology therapies targeting distinct genomic alterations. We sought to determine whether these advances have benefitted patients with cancer from various ancestral backgrounds equally over time. We therefore annotated all genomic alterations identified in a pan-cancer cohort of 55,970 solid tumor samples sequenced with the MSK-IMPACT clinical assay using incremental versions of the precision oncology knowledge base OncoKB and computed the fraction of samples with OncoKB Level 1 alterations (biomarkers included in the FDA drug label) per year based on the annual cumulative FDA-approved targeted therapies. In 2010, 4.4% of patients with inferred African ancestry (AFR), and 2.4% of patients with inferred non-Ashkenazi Jewish European ancestry (EUR) had Level 1 alterations, largely explained by differences in cancer type prevalence between these groups. However, by mid-2023, AFR had a significantly lower percent of Level 1 alterations (25.3%) compared to EUR (31.2%), and this difference remained significant after adjusting for cancer type and disease status (q-value = 2.6e-08). Next, we compared AFR versus EUR percentages of Level 1 alterations over time in breast and colorectal (CRC) patient samples, two cancers in which racial disparities of cancer incidence and outcomes are well-documented. We observed that prior to 2019, the fraction of breast cancer patient samples with a Level 1 alteration was similar among AFR (13%) and EUR (12.2%), primarily due to comparable rates of ERBB2 amplifications (12.4% in AFR, 11.7% in EUR). However, following the breast cancer-specific FDA approvals of alpelisib+fulvestrant for PIK3CA-mutant disease in 2019 and elacestrant for ESR1-mutant disease in 2023, a significant shift in Level 1 actionability was observed between EUR (45%) and AFR (34%) attributed in part to the paucity of PIK3CA (26.3% vs. 36.2% in EUR) and ESR1 (4.1% vs. 7.5% in EUR) oncogenic alterations in AFR. We further observed AFR with CRC to have lower prevalence of MSI-H (5.8% vs.10.7% in EUR), TMB (12.5% vs. 18.8% in EUR) and BRAF V600E (5.1% vs. 9% in EUR). Correspondingly, in 2023, 16.4% of AFR with CRC have Level 1 alterations compared to 25.4% of EUR. Moreover, while AFR have higher rates of KRAS mutations (57.2% vs. 42.6% in EUR), they have a lower prevalence of the standard care biomarker KRAS G12C (3.4% vs. 7% of KRAS-mutant CRC in EUR) further exacerbating the disparities in CRC clinical actionability. Taken together, here we demonstrate that a significant disparity exists in the clinical actionability landscape of AFR with cancer compared to EUR, largely due to differences in relative mutational frequencies of Level 1 genomic alterations. Citation Format: Kanika Arora, Sarah P. Suehnholz, Hongxin Zhang, Ritika Kundra, Subhiksha Nandakumar, Moriah Nissan, A. Rose Brannon, Chaitanya Bandlamudi, Marc Ladanyi, Alexander Drilon, David B. Solit, Nikolaus Schultz, Michael F. Berger, Debyani Chakravarty. Emerging disparities in the clinical actionability landscape for patients with inferred African ancestry [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6129.