Endobronchial Ultrasound-Guided Transbronchial Needle Aspiration for PD-L1 Testing in Non-small Cell Lung Cancer

Background: Programmed death-ligand 1 (PD-L1) expression on cancer cells is a clinically important biomarker to select patients with non-small cell lung cancer (NSCLC) for treatment with programmed death-1/PD-L1 inhibitors. Clinical trials of immunotherapy in patients with NSCLC have required histologic evidence for PD-L1 testing; in clinical practice, cytologic samples commonly are acquired in patients with advanced disease. Research Question: This study aims to investigate whether endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) samples are adequate for PD-L1 testing in NSCLC. Study Design and Methods: This study investigates the sampling adequacy of EBUS-TBNA for PD-L1 testing when compared with other methods. Furthermore, the relationship between clinicopathologic characteristics and PD-L1 expression in the study population have been examined. Five hundred seventy-seven NSCLC specimens were analyzed from consecutive patients with NSCLC across six centers in the United Kingdom and one center in the United States between January 2015 and December 2016. Results: In the EBUS-TBNA group (189 specimens), the overall percentage of patients with successful PD-L1 testing was 94.7%. There was no significant difference in sampling adequacy with other methods of tissue acquisition. Older patients had higher failure rates of PD-L1 testing (OR, 1.06; P =.008). In multivariate analysis, advanced N-stage (P =.048) and presence of brain metastasis (P <.001) were associated with high PD-L1 expression. Interpretation: This large multicenter study shows that EBUS-TBNA provides samples adequate for PD-L1 testing and that advanced N stage and the presence of brain metastasis are associated with high PD-L1 expression.