Prevalence and Distribution of Unexpected Actionable Germline Pathogenic Variants Identified on Broad-Based Multigene Panel Testing Among Patients With Cancer.

Abstract

In patients with a variety of malignancies undergoing multigene panel testing (MGPT), we examined the frequency of a pathogenic/likely pathogenic variant (PV) that would not have been predicted on the basis of the patient's personal and family history of cancer. This is a retrospective review of patients with cancer ascertained from a single academic cancer center who underwent broad-based MGPT of ≥20 cancer predisposition genes not selected on the basis of personal or family cancer history from 2015 to 2021. Low-penetrance variants and recessive inheritance genes were excluded. Deidentified pedigrees were analyzed to determine clinical suspicion of PV. MGPT was performed on 10,975 patients with cancer: 1,134 (10.3%) were found to have ≥1 PV in a moderate or highly penetrant cancer susceptibility gene. Three hundred seven (2.8%) of the PVs were not predicted on the basis of patient's personal cancer history alone, and 192 (1.7%) remained unsuspected after patient's cancer diagnosis and review of family cancer histories were considered. Unexpected PVs accounted for 16.9% of the 1,134 patients with a moderate- or high-penetrance PV. Most frequent unexpected variants were <i>MITF</i> (n = 18), <i>PMS2</i> (n = 18), <i>ATM</i> (n = 17), <i>BRIP1</i> (n = 17), <i>HOXB13</i> (n = 14), <i>SDHA</i> (n = 12), <i>CHEK2</i> (n = 11), <i>BRCA2</i> (n = 7), <i>MSH6</i> (n = 7), <i>SDHC</i> (n = 7), <i>PALB2</i> (n = 6), and <i>TP53</i> (n = 6). Low-penetrance or recessive variants were found in 519 (4.7%) patients. Variants of uncertain significance were found in 3,775 (34.4%). In patients with cancer, MGPT identified a rate of 1.7% PV in unexpected actionable cancer predisposition genes. Findings were more often unexpected (2.8%) when considering only the patient cancer history. These findings may justify consideration of broader MGPT panels in patients with cancer, given implications for subsequent surveillance, cascade testing, and treatment options dependent on specific findings.