Pattern of molecular testing among US veterans with gastrointestinal cancers: A single center experience.

Abstract

e13663 Background: In response to the rising global burden of cancers, the Veterans Health Administration's (VHA) National Precision Oncology Program (NPOP) supports the use of molecular testing (MT) among veterans. The Oklahoma City (OKC) VHA is serving patients with cancer in both rural and urban areas by implementing the standard practice of precision oncology testing across our diverse patient population. Identification of molecular biomarkers can facilitate treatment selection and trial access for veterans. Here we report real-world practices of physicians caring for veterans with gastrointestinal (GI) cancers namely gastroesophageal (GE), pancreatobiliary (PB) and colon cancers at OKC VHA and determine the degree to which molecular data is being utilized to prognosticate or alter clinical management. Methods: Patients presenting to the OKC VHA with metastatic GE, PB and colon cancer from June 2020 to September 2022 were included in the study. Data was collected retrospectively, including patient demographics, clinical characteristics, mutational data, actionability (OncoKB), and treatment recommendations. The primary outcome was the percentage of tumor sequencing performed, and actionable alteration annotated by the OncoKB database therapeutic evidence level. Results: 101 veterans were diagnosed with metastatic GI cancers (GE 18.8%, PB 28.7% and colon 52.4%), of which 82 samples (81.2%) were submitted for MT. The cohort was male-dominant (98%) with a median age of 69.5 years, and 46.5% of the veterans were from rural Oklahoma. Time from test order date to test result availability was 11.7 days. 25.7% (n=26) had at least one likely oncogenic mutation according to the OncoKB database, and 12.8% of them (n=13) were changed to FDA-approved targeted medications. The most common aberrant genes were TP53, KRAS, HER2, and MSI-H. Statistically significant differences were seen in the level of evidence by cancer group. In multivariate logistic regression, patients with colon cancer were most likely to receive MT (OR, 45.5 95% CI 2.27, 27.0 for colon over PB; OR 14.7 95% CI 2.27, 29.6 for GE over PB). Provider level (fellow vs faculty) and distance from the VHA (urban vs rural) were not associated with MT. Conclusions: Our study indicates clinical implementation of precision oncology is feasible across the VHA healthcare system, including rural veterans. Further analysis of this large cohort of patients will help to address gaps in MT for veterans with GE and PB cancers. The study also highlights the importance of MT to guide clinical trial enrollment and identify investigational drug opportunities among veterans.[Table: see text]