Abstract
New reliable biomarkers are needed to predict the response to immune checkpoint inhibitors against programmed death-1 (PD-1) and its ligand (PD-L1), because PD-L1 expression on tumor cells has limited power for selecting patients who may benefit from such therapy. Here we investigated the significance of PD-L1 and PD-L2 gene copy number gains using fluorescence in situ hybridization as well as PD-L1 and PD-L2 expression in 654 patients with resected non-small-cell lung cancer. The prevalence of PD-L1 amplification and polysomy was 3.1% and 13.2%, respectively. The PD-L1 gene copy number status was in agreement with both the PD-L2 and Janus kinase 2 gene copy number statuses. PD-L1 and PD-L2 expression was observed in 30.7% and 13.1%, respectively. Both PD-L1 copy number gains and expression were associated with smoking-related tumors. Tumor cells with PD-L1 genomic gains exhibited significantly higher levels of PD-L1 expression than those without, but PD-L2 copy number gains were not related to PD-L2 augmentation. PD-L1 gene amplification and polysomy were independently associated with PD-L1 expression, with high immune infiltrates and EGFR expression in a multivariate logistic regression model. Comparative analysis between primary tumors and synchronous regional lymph node metastases revealed that the PD-L1 gene copy number alterations were highly consistent and reproducible compared with the PD-L1 expression. Both PD-L1 amplification and level of protein expression were predictors of poor survival using Cox univariate analyses. Therefore, we conclude that an increase in PD-L1 gene copy number can be a feasible alternative biomarker for predicting response to anti-PD-1/PD-L1 therapy.