Depth-dependent PD-L1 sampling of head and neck squamous cell carcinoma (HNSCC) – implications for PD-1/PD-L1 checkpoint inhibitor immunotherapy

At present, PD-L1 scoring is the gold standard in evaluating a cancer patient's suitability for receiving PD-1/PD-L1 checkpoint inhibitor immunotherapy (CIT). Either tumor proportion score (TPS) or combined positive score (CPS) can be used to select head and neck squamous cell carcinoma (HNSCC) patients for immunotherapy. However, the current clinical workflow only uses one histological section for PD-L1 scoring. The considerable under-sampling of a patient's tissue may lead to an inaccurate PD-L1 score, potentially affecting treatment outcomes. In this study, we selected formalin-fixed, paraffin-embedded (FFPE) tissue blocks from five HNSCC patients. The thickness of each of these tissue blocks allowed us to determine the tumor proportion score (TPS) and combined positive score (CPS) at five depths with adjacent layers separated by 500 μm. We found that although there are variations of scores among different tissue sections, the scores are generally consistent across the tissue block from each patient. We also compared TPS and CPS values at different sampling frequencies with a single section and sections separated by 1000, 1500, and 2000 μm. We found that in all cases, the values all fall within one standard deviation of the average values of all five depths. Considering the different threshold values of TPS and CPS for treatment decisions, our results suggest that in cases where the patients' scores are close to that of the threshold values, scoring of two histological sections separated by 2000 μm should be considered.