Development of a Novel IHC Assay for PD-L1 Detection in Non-Small Cell Lung Cancer

Background/Objectives: Programmed cell death-ligand 1 (PD-L1) is one of the key biomarkers for immune checkpoint inhibitors. We are developing a novel PD-L1 CAL10 immunohistochemistry (IHC) assay (Leica Biosystems) on BOND-III staining system and have analyzed its initial performance by comparing it to the PD-L1 SP263 assay (Ventana) assay in a feasibility study. The study objective was to determine the concordance of the Leica Biosystems PD-L1 CAL10 assay with the comparator SP263 assay at the tumor proportion score (TPS) cutoff of ≥50% in non-small cell lung cancer (NSCLC) tissue samples. Additionally, the concordance between the two assays at the TPS cutoff of ≥1% was also evaluated. For informational purposes, we also evaluated the concordance between manual slide reads vs. digital reads (whole slide images generated using the Aperio GT 450) for the CAL10 assay. Methods: Two pathologists read and scored the glass slides. The CAL10 PD-L1 assay concordance with the PD-L1 SP263 assay was evaluated by assessing the agreement rates between the two assays. Results: The lower bound of the 95% confidence interval (CI) of the overall percent agreement (OPA) at ≥50% cutoff was 86.2%, while for ≥1% TPS cutoff, it was 94.0%, which met the predefined target of a minimum OPA lower bound of the 95% CI value of 85%. Conclusions: The Leica Biosystems CAL10 PD-L1 assay has demonstrated comparable performance to the SP263 assay. Additionally, the PD-L1 CAL10 stained glass slides and the corresponding whole side images generated by GT 450 showed comparable concordance rate.