Validation of LOH-Germline Inference Calculator of Variant Origin on Real-World Data of Prevalence of Germline and Somatic <i>BRCA1/2</i> Mutations in High-Grade Serous Ovarian Cancer in Serbia.

Abstract

PURPOSE: High-grade serous ovarian cancer (HGSOC) is the most aggressive type of ovarian cancer, with a high mortality rate. Testing for alterations in BRCA1/2 genes from tumor tissue is currently recommended as a companion diagnostic procedure for selection of patients eligible for therapy with poly (ADP-ribose) polymerase inhibitors. However, tumor testing alone does not allow for identification of patients who may have a hereditary form of the disease, whereas unselected reflex testing for germline variants is not feasible in resource-limited settings. MATERIALS AND METHODS: We report real-world data from the consecutive unselected HGSOC patient population (n = 893) referred for BRCA1/2 mutation testing between 2016 and 2023. First, we evaluated the prevalence and distribution of germline and somatic BRCA1/2 mutations in the Serbian population. Second, we performed independent validation of the probabilistic model, LOH-Germline Inference Calculator (LOHGIC), for inference of germline origin of these alterations based on tumor purity, depth of sequencing, and allele frequency. RESULTS: The Serbian cohort shows approximately 21% overall prevalence of BRCA1/2 pathogenic mutations in HGSOC, of which 30%-50% are germline, none detected in women over 60 years. Concordance of the LOHGIC model for prediction of germline variants with test results from a subset of 121 patients with matched tumor and blood demonstrated a sensitivity of 0.44 (95% CI, 0.14 to 0.79) and a specificity of 0.67 (95% CI, 0.41 to 0.87). CONCLUSION: The moderate prevalence of germline variants in the Serbian population indicates the need for selection of patients for confirmatory testing from blood. However, probabilistic models alone are not sufficiently accurate and should be complemented with information on patient history and age of onset to rule out patients with likely somatic variant origin for a cost-effective testing program.