Programmed Death-Ligand 1 Expression in Lung Cancer and its Association with Clinicopathological Characteristics, Tumor Microenvironments, Driver Mutations, and Clinical Outcomes

• Main Authors: Ching-Yao Yang, 楊景堯
• Other Authors: Yih-Leong Chang
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/dv2xp2

博士 === 國立臺灣大學 === 病理學研究所 === 106 === Lung cancer is the leading cause of cancer-related death worldwide despite a lot of advances in cancer treatment have been achieved in recent decades. Currently, the treatment strategy for advanced lung cancer has been moved forward from conventional platinum based chemotherapy, molecular target therapy, to the era of immunotherapy. Immune checkpoint inhibitors are now the most promising immunotherapy which can unleash the immunosuppression cancer cell posed on host immune system. Among the various immune checkpoints, programmed-death 1 (PD-1) and its ligand, programmed death-ligand 1 (PD-L1) are the most predominating ones which are found in lung cancer and many other types of neoplasms. Expression of PD-L1 on cancer cells can deliver an inhibitory signal through PD-1 on T cells to down-regulate the activity of T cells and prompt cancer escape from host immune systems. Therefore, the expression pattern of PD-L1 on cancer cells becomes an important issue, either for recognition of tumor-host immune balance, or for potential prediction of anti-PD-1/PD-L1 immunotherapy. We conducted serial studies to examine tumor PD-L1 expression in two stage I cohorts of non-small cell lung cancer (NSCLC). The first one is an adenocarcinoma cohort consisting of 163 patients. Using ≥ 5% membranous positive staining cells as cutoff value, we found PD-L1 expression was positive in 65 of 163 patients (39.9%). Higher grade of differentiation and vascular invasion were associated with PD-L1 expression. The multi-variate analysis for survival showed PD-L1 expression was an independent good prognostic factor of disease free survival (DFS), while higher grade of differentiation and abnormal carcinoembryonic antigen (CEA) were poor prognostic factors. The only favorable and independent prognostic factor for overall survival was earlier stage (IA vs IB). In the subsequent study, we conducted a deeper and thorough evaluation of PD-L1 expression in another cohort of stage I squamous cell carcinoma (SqCC) (n=105). Because tumor microenvironments are quite crucial for immune-oncology, we analyzed the composition of tumor infiltrating lymphocytes (TIL) in addition to tumor cell PD-L1 expression and clinicopathological features. Using the same cutoff value of 5%, we found that PD-L1 expression was positive 59 of 105 patients (56.2%), which was more prevalent than the ADC cohort (SqCC vs ADC, 56.2% vs 39.9%, p<0.001). The exploration of TIL composition showed that PD-L1 expression positively correlated with tumor epithelial CD8+ T cell and tumor stromal CD4+ T cell infiltration, and negatively correlated with PD-L1+ immune cells infiltrations in tumor stroma. For the individual TIL, tumor epithelial CD8+ T cell and tumor stromal CD4+ T cell infiltrations were associated with a better DFS and OS; while tumor stromal regulatory T cells infiltrations were associated with a poor DFS and OS. By multivariate analysis, PD-L1 and tumor stromal CD4+ T cell infiltrations were independent prognostic factors of OS but only had a trend toward better DFS. Our results may suggest that in early stage NSCLC, PD-L1 expression reflects a phenomenon of adaptive resistance that an effective host immunity may still exist and is powerful enough to force cancer cell develop an escape mechanism, especially by expressing PD-L1. However, the existence of PD-L1 expression in early stage lung cancer may not point to a status of successful immune escape. We hope our serial studies may help the delineation of cancer immunology and tumor microenvironment of NSCLC.