Comparison of PARPi efficacy according to homologous recombination deficiency biomarkers in patients with ovarian cancer: a systematic review and meta-analysis

Background: Mutations in the BRCA1/2 (BRCA) genes are associated with response to poly(ADPribose) polymerase (PARP) inhibitors (PARPi). In addition, there are different homologous recombination and myChoice (R) score] that identify patients who can benefit from PARPi. Inconsistencies in biomarkers used in PARPi clinical trials make it challenging to identify clinically relevant predictive biomarkers. This study aims to compare clinically available HRD biomarkers in terms of benefits from PARPi.Methods: We performed database search for phase II or III randomized clinical trials comparing PARPi versus chemotherapy, and meta-analysis using generic inverse variance and a Random Effects model. Patients were classified according to their HRD status: (I) BRCAm (patients with BRCA mutation of germline or somatic origin); (II) non-BRCA HRD [patients BRCA wild-type (wt) with another HRD biomarker- gLOH or myChoice (R)]; and (III) homologous recombination proficiency (HRP) (BRCAwt and without HRD biomarkers). From those that were BRCAwt, we compared myChoice (R)+ with gLOH-high.Results: Five studies (3,225 patients) analyzing PARPi in first line setting were included. Patients with BRCAmut had progression-free survival (PFS) with hazard ratio (HR) 0.33 [95% confidence interval (CI): 0.30-0.43]; patients with non-BRCA HRD had a PFS HR 0.49 (95% CI: 0.37-0.65), and patients with HRP had a PFS HR 0.78 (95% CI: 0.58-1.03). Eight studies (5,529 patients) with PARPi including first line and recurrence settings were included. BRCAmut had PFS HR 0.37 (95% CI: 0.30-0.48), BRCAwt & HRD 0.45 (95% CI: 0.37-0.55) and HRP 0.70 (95% CI: 0.57-0.85). Patients with BRCAwt & myChoice (R) >= 42 had PFS HR 0.43 (95% CI: 0.34-0.56), similar to patients with BRCAwt & gLOH-high with PFS HR 0.42 (95% CI:Conclusions: Patients with HRD derived significantly more benefit from PARPi when compared to patients with HRP. The benefit of PARPi in patients with HRP tumors was limited. Careful costeffectiveness analysis, and alternative therapies or clinical trial enrollment should strongly be considered for patients with HRP tumors. Among patients with BRCAwt, a similar benefit was found in patients with gLOH-high and those myChoice (R)+. The clinical development of further HRD biomarkers (e.g., Sig3) may help identify more patients who benefit from PARPi.