H00002263 m01 clone 1g3

FGFR2 protein level was examined using immunohistochemistry and quantified in digitalized slides as described previously [20 (link)]. Immunohistochemical staining (IHC) was conducted on 5‐μm paraffin tumour sections using a mouse monoclonal anti‐FGFR2 antibody (H00002263‐M01; clone 1G3, Abnova, Taipei City, Taiwan). All slides were digitalized (Pannoramic 1000 Scanner, 3DHistech, Sysmex, Kobe, Japan) for quantification in 0–300 H‐score scale. Cases from 1st tercile of H‐score were regarded as FGFR2low and cases from 2nd and 3rd terciles were classified as FGFR2high.

Immunohistochemical staining (IHC) for FGFR2 in all tumors was conducted using a mouse monoclonal anti-FGFR2 antibody (H00002263-M01; clone 1G3, Abnova, Heidelberg, Germany) (Figure 2). To confirm specificity of the staining, additional IHC with a mouse anti-FGFR2 antibody (Sc-6930, Santa Cruz, CA, USA) was performed in randomly selected samples. Following manufacturer’s recommendations, tissue samples of gastric adenocarcinoma and lymph node were used as positive and negative controls for IHC, respectively. Immunohistochemical procedures were carried out on 5-µm paraffin sections, as reported previously [9 (link),15 (link)]. All slides were digitalized using Pannoramic 1000 Scanner (3DHistech, Sysmex, Kobe, Japan). FGFR2 levels were quantified according to the semiquantitative H-score approach by two independent pathologists (MB, HR). The data were presented in 0–300 scale resulting from multiplication of percentage of positive cells by intensity of staining: 0—no staining, 1–3—increased intensity of both cytoplasmic and membrane staining (subgroups by H-score: 0–75 for negative/weak; 76–150 for moderate; 151–225 for strong; 226–300 for very strong expression) (Figure 2). Cases from 1st tercile of H-score were regarded as FGFR2low and cases from 2nd and 3rd terciles were classified as FGFR2high.