Anti human pd l1 antibody clone 22c3

Multiplexed IF was performed on a total of 45 FFPE samples (29 pre-BCG tumor and 16 post-BCG specimens) with the Opal system and images were acquired using the Vectra 3.0 Automated Quantitative Pathology Imaging System (Perkin Elmer) with 4′,6-diamidino-2-phenylindole (DAPI) as the nuclear marker as previously described (26 (link)). The antibodies used are anti-Human CD4 (clone EPR6855; Abcam), anti-Human CD8 (clone C8/144B; DAKO), anti-Human FOXP3 (clone 236A/E7; Abcam), and anti-Human PD-1 (clone NAT105; Abcam). From these 45 FFPE samples, another single-plexed staining of PD-L1, anti-human PD-L1 antibody (clone 22C3; DAKO), was separately performed on consecutive slide of 12 samples (six pairs of matched pre- and post-BCG tumor from non-responders). Full tissue sections were used for all samples. For specimens smaller than 1cmx0.5cm, whole tissue was imaged and quantified; whereas for specimens larger than 1cmx0.5cm, at least 10 areas of 2 mm x 3 mm with high infiltration of immune cells were imaged and quantified. Quantification was done by ImageJ and the mean was calculated as the cell density (number/mm²) for each patient sample, while area with PD-L1 positive staining was quantified and the mean was reported as PD-L1+area/mm². The median value of density for each immune subsets was used as the cutoff point to dichotomize the patients into two groups (low versus high).

PD‐L1 staining was performed using the monoclonal mouse antihuman PD‐L1 antibody (clone 22C3, Cat No. M3653; Dako). A minimum of 100 viable tumor cells must be present in the specimen slide for the PD‐L1 expression to be calculated with complete or partial membrane staining. PD‐L1 assay results were interpreted according to the scoring guidelines as previously described.¹⁸