M3653

One representative section of each surgical tumor resection or biopsy specimen was stained with antibodies against PD-L1 (M3653, Dako). The immunoreactivity of PD-L1 was evaluated according to the percentage of membranous (complete or incomplete) positively stained tumor cells (Tumor proportion score (TPS)) and the percentage of positively stained tumor associated immune cells (TAIs), i.e. all immune cell subtypes (categorization into 0%, 1%, 3%, 5%, 10%, 15%, 20%, 30%, etc. in both). Staining intensity was not considered. Specimens in which PD-L1 staining was observed in ≥1% of tumor cells or immune cells were considered PD-L1 positive.
Additionally, the combined positive score (CPS) and the interface pattern were evaluated. CPS was calculated by dividing the number of PD-L1 positive tumor cells, lymphocytes and histiocytes by the total number of vital tumor cells and then multiplying the result by 100. The interface pattern was described by Muro K et al. as a band of PD-L1-positive cells (mainly mononuclear inflammatory cells) at the interface between confluent areas of neoplastic cells and adjacent stroma [6 (link)]. We evaluated the presence of this pattern dichotomously (yes/no). All analyses were performed independently by two experienced pathologists (A.B., AI.S) and in case of differing results a consensus was reached together.

The primary antibodies used in this study include PD-L1 (22C3) (Dako, M3653, 1:50). The immunohistochemistry staining of PD-L1 (22C3) was performed on a Dako Autostainer Link 48 platform with the Dako K8002 detection kit, and the signals were amplified with a mouse linker (contained in the kit) and the signals enhanced with a DAB enhancer (Dako, S1961); the staining procedure was set similar to the FDA-approved PD-L1 (22C3) pharmDx staining procedure. All immunohistochemistry (IHC) sections were evaluated by two pathologists independently. PD-L1 ≥1% on tumor cells were defined as positive.