Abstract 1176: Quantitative integration of PIK3CA hotspots and paralogy identifies occult oncogenic mutants in PIK3CD

Abstract

Abstract The majority of somatic variants identified by tumor genomic profiling are low frequency alterations of unknown biologic and clinical significance. To facilitate the systematic identification of likely oncogenic versus benign somatic mutations, there has been a growing interest in harnessing additional computationally amenable metrics such as sequence paralogy. Here, we developed a novel paralogy-directed computational methodology called Enriched Paralogy to identify Occult Cancer Hotspots (EPOCH), which calculates the significance of mutational recurrence at aligned positions in evolutionarily related genes to prioritize likely oncogenic mutations for in depth functional characterization. We validated the utility of EPOCH using sequencing data from a cohort of 52,676 tumor/normal pairs, with a focus on identifying likely oncogenic mutations in the well characterized PIK3CA (p110α) class 1a PI3K isoform and the lesser characterized PIK3CD (p110δ) isoform. p110δ mutants nominated as likely oncogenic by EPOCH induced AKT pathway activation, enhanced anchorage-independent colony formation, and demonstrated allele-specific dependencies on p85 regulatory protein binding. The potency of PI3K inhibitors against p110δ mutants was at best equal to or less than wildtype p110δ. In sum, we present the novel algorithm EPOCH that integrates paralogy to extend the utility of current recurrence-based precision oncology predictions and to automate the functional interpretation of less common mutations. Our work identifies novel oncogenic and targetable p110δ mutants and supports the inclusion of PIK3CD in targeted sequencing panels despite the relative rarity of mutations in this gene. The therapeutic studies herein also provide the rationale for the development and clinical testing of isoform- or mutant-specific p110δ inhibitors for patients with rare, yet druggable, oncogenic p110δ mutations. Citation Format: Arijh Elzein, Alexander N. Gorelick, Ino de Bruijn, Barry S. Taylor, Aphrothiti J. Hanrahan, David B. Solit. Quantitative integration of PIK3CA hotspots and paralogy identifies occult oncogenic mutants in PIK3CD [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1176.