PD-L1 immunohistochemistry in oesophageal, gastric and gastro-oesophageal junction adenocarcinoma: comparison of combined positive score across 22C3, and SP263 assays

Introduction and objectives: Programmed death-ligand-1 (PD-L1) is a potential target for immune checkpoint inhibitors in various cancers. The current study aimed to evaluate the concordance of combined positive score (CPS) across 22C3, and SP263 assays when interpreting PD-L1 immunohistochemistry on oesophageal, gastric and gastro-oesophageal junction (GOJ) adenocarcinoma.
Methodology: Twenty-two biopsies and eight resection specimens of oesophageal, gastric and GOJ adenocarcinoma were evaluated for PD-L1 SP263 expression at the Department of Pathology, Royal Surrey Hospital, over a period of four years (January 2021 to December 2024). These specimens were re-evaluated for PD-L1 22C3 assay at Poundbury Cancer Institute by an independent pathologist who was blinded to the PD-L1 SP263 assay results. The two results were compared for concordance.
Results: There was a very strong positive correlation between the SP263 and 22C3 scores of the two assays (Pearson correlation co-efficiency; r=0.89). Discordance was observed in categorising tumours at cut offs of CPS ≥5 and CPS ≥10, with the discordance being more at a cut off of CPS ≥5.
Discussion and conclusion: There are three major commercially available immunohistochemistry assays to quantify PD-L1 expression in cancer cells; SP263, 22C3 and 28-8. A CPS ≥1 is currently used to classify a tumour as PD-L1 positive. According to NICE guidelines, oesophageal or GOJ adenocarcinomas with a PD-L1 CPS ≥5 are eligible for treatment with nivolumab, while a CPS ≥10 qualifies for pembrolizumab therapy. This study demonstrated analytical concordance between the SP263 and 22C3 assays when CPS scoring was applied, consistent with previously published data.