Functional Homologous Recombination Assay on FFPE Specimens of Advanced High-Grade Serous Ovarian Cancer Predicts Clinical Outcomes

Abstract

Purpose: Deficiency in homologous recombination (HR) repair of DNA damage is characteristic of many high-grade serous ovarian cancers (HGSC). It is imperative to identify patients with homologous recombination-deficient (HRD) tumors as they are most likely to benefit from platinum-based chemotherapy and PARP inhibitors (PARPi). Existing methods measure historical, not necessarily current HRD and/or require high tumor cell content, which is not achievable for many patients. We set out to develop a clinically feasible assay for identifying functionally HRD tumors that can predict clinical outcomes.

Experimental design: We quantified RAD51, a key HR protein, in immunostained formalin-fixed, paraffin-embedded (FFPE) tumor samples obtained from chemotherapy-naïve and neoadjuvant chemotherapy (NACT)-treated HGSC patients. We defined cutoffs for functional HRD separately for these sample types, classified the patients accordingly as HRD or HR-proficient, and analyzed correlations with clinical outcomes. From the same specimens, genomics-based HRD estimates (HR gene mutations, genomic signatures, and genomic scars) were also determined, and compared with functional HR (fHR) status.

Results: fHR status significantly predicted several clinical outcomes, including progression-free survival (PFS) and overall survival (OS), when determined from chemo-naïve (PFS, P < 0.0001; OS, P < 0.0001) as well as NACT-treated (PFS, P < 0.0001; OS, P = 0.0033) tumor specimens. The fHR test also identified as HRD those PARPi-at-recurrence-treated patients with longer OS (P = 0.0188).

Conclusions: We developed an fHR assay performed on routine FFPE specimens, obtained from either chemo-naïve or NACT-treated HGSC patients, that can significantly predict real-world platinum-based chemotherapy and PARPi response. See related commentary by Garg and Oza, p. 2957.

Figure 1.

RAD51-based assay to determine fHR capacity from chemo-naïve and NACT-treated clinical HGSC specimens. A, Diagram showing the sample collection. Chemo-naïve samples were obtained from PDS or DL. NACT-treated specimens were obtained from IDS. B, Workflow of the fHR assay. Example images of geminin (green) and RAD51 (red) double stained fHRD and fHRP samples with ImageJ analysis illustration. Number of RAD51 and geminin double positive nuclei divided by the number of geminin-positive nuclei provides the fHR score. C and D, Distribution of fHR scores in chemo-naïve samples (C), as well as in the IDS (NACT-treated) samples (D), shown separately for discovery and validation cohorts. Dashed line indicates the proposed fHRD versus fHRP cutoffs. Colored squares depict HRD estimates from genomics-based assays, with blue shades corresponding to HRD and red shades to HRP. “Non-matched treatment stage” refers to cases where the genomics-based estimate of the patient was obtained from a different surgery sample (PDS/DL vs. IDS) than the fHR score. Deleterious mutations in HR genes identified from WGS/WES data are indicated for each patient. For the IDS validation cohort, only BRCA1/2 mutational testing results from the clinic were available. Asterisks indicate patients who received bevacizumab as part of their subsequent maintenance treatment. E, Comparison of fHR scores from chemo-naïve and IDS (NACT-treated) samples, obtained from the same patient (n = 13 patients). Abbreviations: ND, no data. (A, Created with BioRender.com.)

Figure 2.

Low fHR scores correlate with better primary therapy response and with longer PFI. A, For patients with chemo-naïve samples, the median of fHR scores is lower in CR and PR groups compared with SD/PD (Mann–Whitney test, two-tailed). Only patients with fHRP tumors had SD/PD after primary therapy. B, The median of fHR scores was lower in CR and PR groups compared with SD/PD also for patients with fHR score from IDS/NACT-treated sample (Mann–Whitney test, two-tailed). C, Proportions of CR, PR, and SD/PD are significantly different between fHRP and fHRD groups (Fisher exact test, two-tailed). D and E, PFI and OS significantly correlate with lower fHR scores (linear regression). F, Multivariate hazard ratio analysis for PFI with fHR status and prognostic clinical parameters. The fHRD status and success of cytoreduction significantly correlated with longer PFI (Cox proportional hazards regression). Blue = fHRD, red = fHRP.

Figure 3.

Kaplan–Meier survival analysis. fHRD status significantly predicts longer PFS and OS in chemo-naïve (A and C) and IDS (B and D) cohorts (Log-rank, Mantel–Cox test). Patients treated with PARPi in the first-line setting or with unknown PARPi treatment status were excluded from the PFS analysis (A and B). In addition to first-line and unknown PARPi treatment statuses, patients treated with PARPi at recurrence were excluded from OS analyses (C and D).

Figure 4.

fHRD status indicates longer survival in patients treated with PARPi maintenance therapy at recurrence. A, In Kaplan–Meier survival analysis, time-to-progression from PARPi treatment start did not significantly differ between fHRD and fHRP patients. B, OS of patients in the fHRD group was significantly longer than in the fHRP group. (Log-rank, Mantel–Cox test).

Figure 5.

Comparison between different HRD estimates. A, Hazard ratio analysis with the different HRD estimates available and PFI. Only fHR and ovaHRDScar tests produced significant results (Cox proportional hazards regression). B, Sensitivity and specificity of the different HRD tests. ovaHRDScar has the best sensitivity in detecting clinically meaningful HRD (>1y PFI), and fHR test has the best specificity in detecting clinically meaningful HR proficiency (<1y PFI).