Abstract 2773: The Gundersen Precision Oncology Cohort: a biospecimen resource for mutation characterization and target discovery

Abstract Background: The generation and widespread sharing of large somatic mutation datasets has revolutionized our understanding of genetic mechanisms driving cancer initiation and progression. For well-understood mutations, this has created the opportunity for individualized therapeutic intervention in a wide range of malignancies. Beyond these well-characterized mutations, there exists a wide range of alterations that remain poorly understood and which require some kind of functional genomics analysis to determine whether such patients are eligible for the targeted therapy used for more typical mutations in that gene, whether these alterations are passenger mutations that don’t affect throughput on the associated signaling pathway, or reflect another more nuanced phenomenon. Whether using ectopic overexpression in cell lines or generation of genetically modified mice, addressing this problem at scale using functional genomics approaches is challenging. A key limitation is the absence of the original tissue context in which that variant co-evolved with other genetic alterations as well as data on the clinical response of the patient to successive lines of therapy. This can be addressed, in part, by the banking of tissue from patient cohorts who have had comprehensive genomic profiling as part of their routine clinical care. For this purpose, we have established the Gundersen Precision Oncology Cohort. Methods: The Gundersen Precision Oncology Cohort is a retrospective collection of tissue specimens, NGS reports and associated clinical data from deceased patients who received cancer care at the Gundersen Health System. The cohort is updated twice per year and banked tissue specimens are available for collaborative research studies. High resolution digital pathology whole slide images are available for artificial intelligence and machine learning studies. Results: At November 26, 2024 there were data from 908 patients (913 tumors) in the cohort, with specimens tested between 2015 and 2024. 55 tumor types were represented, of which lung (n = 225), colorectal (n = 101), pancreas (n = 69), esophageal (n = 62) and breast (n = 56) were most common. The most common diagnostic companies were STRATA Oncology (n = 533) and Tempus (n = 332). Most patients had testing at a single timepoint, but longitudinal testing is available for 40 patients. We present data from some of the studied subgroups, including atypical point mutations in BRAF as well as fusions affecting BRAF and ROS1. Conclusions: This cohort, available for collaborative studies via the Gundersen Cancer Biobank, illustrates the value of banking well-annotated biospecimens to address outstanding questions in oncology. Citation Format: Peter Feiszt, Paraic A. Kenny. The Gundersen Precision Oncology Cohort: a biospecimen resource for mutation characterization and target discovery [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2773.