Abstract
BACKGROUND: Determination of programmed death-ligand 1 (PD-L1) protein expression level in tumor cells and tumor-associated immune cells is critical for identifying patients eligible for immunotherapy. PD-L1 manual scoring algorithms can generally be divided into two categories: cell counting and visual estimation. Cell counting can be time-consuming and is not in sync with pathology practice, which classically uses a Gestalt approach based on pattern recognition and visual estimation. In this study, we introduce the Tumor Area Positivity (TAP) score, which is a novel, straightforward method for scoring tumor cells and immune cells together using visual estimation. METHODS: To demonstrate the reproducibility of TAP scoring among pathologists, between- and within-reader precision studies were performed both within (internal) and outside of (external) our organization. We also compared the TAP score to the Combined Positive Score (CPS), which is based on cell counting, for concordance and time efficacy. RESULTS: The average positive agreement, average negative agreement, and overall percent agreement between and within readers were all above 85% for both internal and combined external reader precision studies. TAP score had high concordance rate at 5% cutoff compared with CPS at cutoff 1: positive percent agreement, negative percent agreement, and overall percent agreement were all above 85%. CONCLUSIONS: Our study showed the TAP scoring method to be straightforward, significantly less time-consuming, and highly reproducible with a high concordance rate between TAP score and CPS.