A panel‐based mutational signature of homologous recombination deficiency associates with response to PARP inhibition in metastatic castration‐resistant prostate cancer

Abstract Background The PARP inhibitor (PARPi) olaparib is approved for homologous recombination repair (HRR) gene‐altered metastatic castration‐resistant prostate cancer (mCRPC). However, there is significant heterogeneity in response to PARPi in patients with mCRPC. Better clinical biomarkers are needed to identify patients likely to benefit from PARPi. Methods Patients with prostate adenocarcinoma and panel sequencing at Dana‐Farber Cancer Institute were identified. Mutational signature analysis was performed using SigMA to characterize tumors as HRR deficient (HRD). The validity of SigMA to identify patients likely to benefit from olaparib was compared to the current FDA label (presence of a deleterious alteration in one of 14 HRR genes). Results 546 patients were identified, of which 34% were HRD. Among patients with HRR gene alterations, only patients with BRCA2 two‐copy loss (2CL) were more likely to be HRD compared to patients without HRR gene alterations (74% vs 31%; P = 9.1 × 10 ‐7 ). 28 patients with mCRPC received olaparib, of which 13 were HRD and 9 had BRCA2 2CL. SigMA improved upon the current FDA label for predicting PSA50 (sensitivity: 100% vs 90%; specificity: 83% vs 44%; PPV: 77% vs 47%; NPV: 100% vs 89%) and rPFS > 6 months (sensitivity: both 92%; specificity: 93% vs 53%; PPV: 92% vs 63%; NPV: 93% vs 89%). On multivariate analysis, incorporating prognostic clinical factors and HR gene alterations, SigMA‐predicted HRD independently associated with improved PSA‐PFS (HR = 0.086, p = 0.00082) and rPFS (HR = 0.078, p = 0.0070). Conclusions SigMA‐predicted HRD may better identify patients likely to benefit from olaparib as compared to the current FDA label. Larger studies are needed for further validation.