Panoramic p250 scanner

Slides were digitally scanned using the Panoramic P250 scanner, manufactured by 3D HISTECH in Hungary. The analysis was conducted using the 3D Histech Quant Center software (Version 2.1, Budapest, Hungary). The digital tissue scanner and imaging system were used to capture scan files and images of the immunohistochemical sections. The Servicebio Image analysis system automatically measured the tissue measurement area. Expression score = ∑(pi × i) = (percentage of weak intensity area ×1) + (percentage of moderate intensity area × 2) + (percentage of strong intensity area × 3), where pi represents the percentage of the pixel area of positive signals and i represents the positive rating [18 (link)]. The expression score is a value ranging from 0 to 300, where a higher score indicates a stronger comprehensive positive intensity [19 (link),20 (link)].

Slides belonging to the two cohorts were digitized in the respective pathology centers using a Panoramic P250 scanner (3DHistech) at different resolution levels. For each WSI, tumor tissue was manually annotated in QuPath [36 (link)] (v.0.2.3) by a trained observer (MA) under expert supervision (VB) (see supplementary material, Figure S1). An automated Python‐based pipeline (https://github.com/MiriamAng/TilGenPro) was implemented to tessellate the identified tumor areas into nonoverlapping tiles of 512 × 512 pixel edge length, perform quality filtering, and stain‐normalization (see supplementary material, Figure S2). Further details are provided in Supplementary materials and methods.