Impact of neoadjuvant chemotherapy on homologous recombination deficiency test results in patients with advanced high-grade serous ovarian cancer.

Homologous recombination deficiency predicts response to platinum-based chemotherapy and poly(adenosine diphosphate-ribose) polymerase inhibitors in advanced high-grade serous ovarian carcinoma. However, the optimal timing of homologous recombination deficiency testing remains unclear for patients receiving neoadjuvant chemotherapy, as it may affect test informativity and results. We evaluated the concordance of genomic and functional homologous recombination deficiency testing before and after neoadjuvant chemotherapy in patients with high-grade serous ovarian carcinoma. Matched tumor samples collected before and after neoadjuvant chemotherapy from patients with high-grade serous ovarian carcinoma treated at the European Institute of Oncology (Milan, Italy, July 2018-December 2021) were analyzed. Genomic homologous recombination deficiency assessment included the Genomic Instability Score and tumor BRCA1/2 mutation testing using SOPHiA DDM Homologous Recombination Deficiency Solution. Functional homologous recombination deficiency assessment was performed through RAD51 foci formation immunofluorescence. Cohen's kappa coefficients (κ) assessed genomic and functional homologous recombination deficiency testing concordance for matched pre- versus post-neoadjuvant chemotherapy results, and functional versus genomic homologous recombination deficiency testing concordance at all time points. Samples collected before and after neoadjuvant chemotherapy from 23 patients with high-grade serous ovarian carcinoma were analyzed. Homologous recombination deficiency informativity was higher before neoadjuvant chemotherapy (87%, 20/23) than in samples collected afterward (65%, 15/23), whereas Genomic Instability Score informativity was 91% (21/23) and 78% (18/23), respectively. Concordance of the Genomic Instability Score was moderate (κ = 0.52), while homologous recombination deficiency concordance was substantial (κ = 0.67) in matched samples collected before and after neoadjuvant chemotherapy. Two of 15 matched informative samples (13%) lost Genomic Instability Score positivity after chemotherapy, but their homologous recombination deficiency test remained positive due to BRCA mutations. Functional homologous recombination deficiency assessment showed poor concordance between time points and with homologous recombination deficiency testing at each time point. Genomic homologous recombination deficiency tests were concordant in matched tumor samples collected before and after neoadjuvant chemotherapy for high-grade serous ovarian carcinoma, but tissue collection before chemotherapy should be prioritized due to higher informativity. Loss of informativity may result in missed opportunities for poly(adenosine diphosphate-ribose) polymerase inhibitor therapy.