Abstract

Purpose
Ovarian cancer presents significant treatment challenges due to its aggressive nature and poor response to immune checkpoint inhibitors (ICIs). The lack of standardized programmed cell death-ligand 1 (PD-L1) assays and cut-off values complicates clinical decision-making. We evaluated the concordance among commonly used PD-L1 assays and assessed changes in the expression of PD-L1 following chemotherapy.
Materials and Methods
Tissue samples from 29 patients with ovarian cancer were analyzed using five validated PD-L1 immunohistochemistry assays: Dako 22C3, Ventana SP263, Ventana SP142, Dako 28-8, and Ventana 22C3. PD-L1 positivity was assessed using a combined positive score (CPS), immune cell (IC), or tumor proportion score (TPS) at 1%, 5%, and 10% cut-offs. Concordance rates, including overall percent agreement (OPA) and Cohen’s kappa coefficient, were analyzed. In addition, changes in the expression of PD-L1 pre- and post-chemotherapy were evaluated.
Results
Positivity rates ranged from 15.8% (SP142) to 29.8% (Dako 22C3 and SP263) at the 1% CPS cut-off. SP142 consistently exhibited the lowest concordance, whereas Dako 22C3 displayed high agreement with SP263, 28-8, and Ventana 22C3. Chemotherapy increased PD-L1 positivity, with 28% of patients converting from negative to positive.
Conclusion
The expression of PD-L1 in ovarian cancer varies across assays and scoring methods, emphasizing the need for standardized testing protocols. Increased PD-L1 expression post-chemotherapy underscores the importance of assessing its status at appropriate times to guide ICI therapy. Larger studies are required to validate these findings and refine clinical applications.

• Keywords: Ovarian neoplasms, Programmed cell death ligand 1 (PD-L1), Immune checkpoint inhibitors, Immunohistochemistry