Implementation of pan-cancer universal germline testing in an ethnically diverse and rural community oncology practice.

Abstract

10569 Background: Universal germline genetic testing (GGT) is increasingly utilized in precision cancer care. There is limited data on this approach in patients (pts) from historically underrepresented and underserved populations receiving care at community practices. Herein, we present an interim analysis of ~500 unselected pts who underwent standard of care GGT at a rural community oncology practice. Methods: The UNITY (UNIversal germline Testing in the communitY) trial (NCT05416710) is a prospective, observational study of pts with newly or previously diagnosed cancer from July 1, 2022-August 1, 2024 (censor date). GGT was performed largely via an 80+ gene panel and insurance-billed. Patient demographic and clinical features were collected by clinicians. NCCN criteria for the pt’s primary cancer at the time of GGT determined if the pt was in-criteria (IC) or out-of-criteria (OOC). Differences among groups were determined by two-tailed Fisher’s exact, one-way ANOVA and Tukey’s HSD tests with significance set at p < 0.05. Results: 462 pts had complete data: 73% were female; most common cancers: breast (54%), colorectal (14.5%), lung (9%), head & neck (6%), prostate (4%); mean age at diagnosis and testing: 63.2 and 67.6; 21% stage IV/metastatic; 66% Non-Hispanic white, 30% Black/African-American; 17% > 1 cancer diagnosis; 77% family history of cancer; 48% met NCCN criteria; 61% commercial insurance; 27% annual income < $25,000. 47 pathogenic germline variants (PGV) were identified in 41 pts (8.9%). 12 pts (3%) carried a single PGV in a gene associated with autosomal recessive cancer risk (e.g. MUTYH ) and were excluded from further analyses, resulting in 6.3% (29/462) pts with a PGV. There was no significant difference in the rate of PGVs in IC vs. OOC pts (5.9% vs. 6.7%, p = 0.85). 16/29 (55%) pts with PGVs were OOC, with CHEK2, ATM, BRCA2 being the most frequently mutated genes. Additionally, the majority of PGVs were potentially clinically actionable (25/29, 86%) and most (14/25, 56%) were OOC. 190 (41%) pts had a variant of uncertain significance (VUS) in the absence of a PGV, with Black/African-American pts having significantly higher rates of VUS-only findings compared to non-Hispanic White pts (50% vs. 36%, p = 0.04). Conversely, PGV rate showed the opposite pattern (0.7% Black/African-American; 8.5% non-Hispanic White, p = 0.01). Conclusions: Universal GGT in this diverse cohort identified PGVs in nearly 1 in 15 pts, most of which were potentially clinically actionable, but > 50% would have been missed by NCCN criteria. As Black pts had significantly lower odds of carrying a PGV and higher odds of a VUS result, broader GGT testing criteria would help mitigate racial disparities by increasing the number of (diverse) individuals tested resulting in better representation of genetic variation. Clinical trial information: NCT05416710 .