Abstract 1774: Examination of variants of unknown significance (VUSs) and co-occurring mutations from comprehensive genomic profiling (CGP) results in a cross tumor model

Tincy Simon,Rob Oliver,Robyn Harrell,Paul Conkling,Emily Paul
DOI: https://doi.org/10.1158/1538-7445.am2024-1774
IF: 11.2
2024-04-04
Cancer Research
Abstract:Data from Comprehensive Genomic Profiling (CGP) generates both actionable and non-actionable results for oncology patients. While the actionable results are included in reports, the non-actionable are less often investigated. This study explored the results of both actionable and non-actionable mutations as well as the co-occurrence of mutations in a pan-tumor model.In this study, we explored data generated from 41 oncology practices where patients (+ 18yo) were tracked within the iKnowMed EHR and had received CGP as part of their care. CGP is defined as a large-scale (300+ genes) NGS-based assay. As part of this study, results between 2017 - present were collected across tumor types and stages.Assessment of the data pool demonstrated that a majority of the 44,769 profiles were generated from Stage IV patients (n=24,545) and NSCLC, colorectal, and breast cancers were the highest representing disease populations in the cohort. The tested gene targets aligned with the diseases that were most represented, with EGFR, PIKC3A, ERBB2, and KRAS genes as the most common genes to have documented results. Exploration of the actionable biomarkers showed the expected pathogenic mutations in the relevant disease context. When exploring the remaining reported variants, it was found that CGP results yield a significant (p<0.05) number of VUSs as well (n=23106). The complexity of mutations was also examined on an intra- and intergenic basis. Intragenic assessment for variants focused on complex mutations, qualified by 2+ mutations including multiple SNV, multiple indels, or SNVs and deletions in the same gene. Trends of co-occurring alterations were not obvious, with TP53 having the most reported occurrences of complex mutations (n=7491). Exploration moved to intergenic assessment and, while disease-bounded exploration did not reveal trends, a disease-agnostic approach revealed co-occurrences of mutations. The top 5 genes with co-existing mutations are TP53, KRAS, APC, PIK3CA, and ATM. Co-occurrence of alterations in any gene with tumor agnostic markers was also explored, notably TMB. The presence of mutation with TMB-high was similar to that of intergenic complex mutations and the same genes had high co-incidence levels. This study has demonstrated that CGP data provide insights beyond actionable biomarkers, and that a pan-tumor approach allows new trends to be observed. Real-world data bias for reported biomarkers skewed the results observed. Taking the results out of disease context allows us to visualize new possibilities for identifying variants implicated in disease processes. Future studies could allow the exploration of unreported data from raw files, overlaid with cellular pathways on complex patient profiles to identify exploitable targets. Inclusion of treatment and outcome data can demonstrate the influence of genetics on a patient journey. Citation Format: Tincy Simon, Rob Oliver, Robyn Harrell, Paul Conkling, Emily Paul. Examination of variants of unknown significance (VUSs) and co-occurring mutations from comprehensive genomic profiling (CGP) results in a cross tumor model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular s); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl) nr 1774.
oncology
What problem does this paper attempt to address?