Jitka Hausnerova, Petra Ovesna, Lucie Ehrlichova, Michal Felsinger, Monika Nalezinska, Kvetoslava Matulova, Lubos Minar, Leos Kren, Ondrej Slaby, Vit Weinberger, Marketa Bednarikova
PD-L1 expression assessed by CPS, particularly at a cutoff of ≥ 10, may have prognostic relevance in ovarian cancer and could help identify patients more likely to benefit from platinum-based chemotherapy. Prospective validation is warranted.
BACKGROUND: Programmed death-ligand 1 (PD-L1) has emerged as a potential biomarker for prognosis and treatment response in various solid tumors. However, its clinical relevance in ovarian carcinoma (OC) remains unclear, with published studies reporting inconsistent associations with platinum sensitivity and survival. These discrepancies are likely related to the biological heterogeneity of OC and methodological variability in PD-L1 evaluation. In the absence of a standardized, OC-specific assessment methodology, the Combined Positive Score (CPS) with cutoffs of ≥ 1 and ≥ 10, applied in other tumor types, represents a pragmatic approach.
METHODS: We retrospectively analyzed a real-world cohort of 146 patients with ovarian carcinoma treated with primary platinum-based chemotherapy at University Hospital Brno between 2010 and 2020. PD-L1 expression was assessed using the 22C3 antibody and CPS methodology on archival, treatment-naive tumor samples obtained from primary tumors or metastatic lesions. Associations between PD-L1 status, platinum treatment-free interval (TFIp), and disease-specific survival (DSS) were evaluated, including subgroup analyses by histological subtypes.
RESULTS: Among the 144 evaluable cases, 65 patients (45%) demonstrated positive PD-L1 expression (CPS ≥ 1), while 35 patients (24%) exhibited strong positive expression (CPS ≥ 10). In the high-grade serous carcinoma (HGSC) subgroup, CPS ≥ 1 and CPS ≥ 10 were detected in 58/121 (48%) and 30/121 (25%) cases, respectively. PD-L1 positivity did not differ significantly between HGSC and non-HGSC subtypes or between primary and metastatic lesions. PD-L1 positivity defined as CPS ≥ 10 was significantly associated with platinum sensitivity (TFIp ≥ 6 months; p = 0.035 overall; p = 0.028 in HGSC) and longer DSS (p = 0.029). No associations were observed when PD-L1 positivity was defined using a CPS ≥ 1 cutoff.
CONCLUSIONS: PD-L1 expression assessed by CPS, particularly at a cutoff of ≥ 10, may have prognostic relevance in ovarian cancer and could help identify patients more likely to benefit from platinum-based chemotherapy. Prospective validation is warranted.