Poster
Artificial Intelligence–Powered and Manual Assessment of PD-L1 Are Comparable in Predicting Response to Neoadjuvant Atezolizumab in Patients With Resectable Non-Squamous, Non-Small Cell Lung Cancer
AACR 2022
Study Background
• Immunohistochemistry (IHC) evaluation of programmed death-ligand 1 (PD-L1) expression is a reliable predictor of the efficacy of anti–PD-L1/programmed cell death protein 1 (PD-1) cancer immunotherapy in patients with resected and metastatic non-small cell lung cancer (NSCLC)
• Exploratory analyses in patients with metastatic NSCLC suggest a potentially limited enrichment for the efficacy of anti–PD-L1/PD-1 therapy in patients with high PD-L1 expression and squamous histology.2 Data in early NSCLC are limited to date
• Preoperative treatment with atezolizumab (anti–PD-L1 antibody) in patients with untreated early-stage resectable NSCLC resulted in a 20% major pathologic response (MPR) rate in the LCMC3 Phase II study
• Computer vision can be used on pathology slides to identify biological entities, including cell types, tissue types, and biomarker expression, and to compute slide-level scores or digitally quantify pathology features
• Using manual and digital PD-L1 scoring methods, we assessed PD-L1 expression in tumor cells as a potential predictor of pathologic response to neoadjuvant atezolizumab treatment in patients with non-squamous and squamous NSCLC tumors
• Exploratory analyses in patients with metastatic NSCLC suggest a potentially limited enrichment for the efficacy of anti–PD-L1/PD-1 therapy in patients with high PD-L1 expression and squamous histology.2 Data in early NSCLC are limited to date
• Preoperative treatment with atezolizumab (anti–PD-L1 antibody) in patients with untreated early-stage resectable NSCLC resulted in a 20% major pathologic response (MPR) rate in the LCMC3 Phase II study
• Computer vision can be used on pathology slides to identify biological entities, including cell types, tissue types, and biomarker expression, and to compute slide-level scores or digitally quantify pathology features
• Using manual and digital PD-L1 scoring methods, we assessed PD-L1 expression in tumor cells as a potential predictor of pathologic response to neoadjuvant atezolizumab treatment in patients with non-squamous and squamous NSCLC tumors
Authors
Abel et al.