Comprehensive assessment of homologous recombination deficiency via simultaneous methylation and mutation analysis in epithelial ovarian cancer: implications for PARP inhibitors efficacy

BACKGROUND: The advent of poly (ADP-ribose) polymerase inhibitors (PARPi) over the past decade has significantly altered the management of epithelial ovarian cancer (EOC). We proposed that the etiology of homologous recombination deficiency (HRD) might underlie the variable responses to PARPi observed across patient populations. METHODS: As part of the phase 2 study of the Chinese HRD Harmonization Project, we developed a genomic methylation sequencing (GM-seq) pipeline facilitated by the TET enzyme for the simultaneous identification of methylated modifications and genetic variations in EOC tumor samples, and compared with established DNA sequencing-based HRD assays. RESULTS: Somatic mutation and HRD scores were confounded by low tumor purity in our cohort of 98 locally advanced/advanced EOC patients. In samples with tumor purity ≥ 30% (n = 45), the GM-seq pipeline showed high consistency with DNA sequencing-based HRD assay, identifying genetic variations in homologous recombination repair (HRR) genes and HRD score with 92.6% (25/27) and 97.1% (33/34) consistency respectively, in addition to conducting methylation profiling. Moreover, different underlying mechanisms of HRD were associated with varying degrees of PARPi efficacy, with BRCA1/2 LOH group having the best efficacy (median PFS, undefined), followed by BRCA1 methylation group (median PFS, 23.4 months), and those with unknown etiology of HRD having the worst efficacy (median PFS, 8.8 months, p < 0.001). CONCLUSION: Our findings underscore the importance of considering HRD etiology when evaluating PARPi efficacy in EOC patients. The GM-seq pipeline, represents a significant advancement in HRD detection, enabling more accurate predictions of PARPi response.