1987P Discordant genomic correlates of PD-L1 expression in lung adenocarcinoma among multiple cohorts using dissimilar PD-L1 testing techniques

Abstract

Programmed death-ligand 1 (PD-L1) is a crucial biomarker predicting efficacious treatment of immune checkpoint inhibitors (ICIs). Understanding the genetic association of PD-L1 expression provides insights into the tumor-immune interaction and further exploration of immunotherapeutic biomarkers. However, the diverse and dissimilar quantifying assays might lead to inconsonant and confusing results, not conducive to the following research. Previous studies identified distinct staining utility among multiple PD-L1 antibodies, however, whether the genomic correlate of stained PD-L1 is also impacted by different testing assays is largely unknown. We evaluated 873 Chinese LUAD patients with lung adenocarcinoma in whom FDA-approved PD-L1 testing (22C3, Dako, Glostrup, Denmark; SP263, Ventana, Tucson, AZ) and targeted next-generation sequencing (3D Medicines Inc.) was performed on the same tissue. Data from Memorial Sloan Kettering Cancer Center (MSKCC), The Cancer Genome Atlas (TCGA) and the OAK trial were analyzed for further validation. In the 3DMed database (n=873), PD-L1 expression in tumor cells was positively correlated with tumor mutational burden. (p<0.001). PD-L1 positivity was more common in metastatic lymph nodes compared to primary lung tumors and other metastatic samples (p =0.009). Deleterious mutations in KRAS, TP53, ALK and MET significantly associated with PD-L1 high expression (each p<0.05) and EGFR and ERBB2 mutations associated with PD-L1 negativity (each p<0.05). Comparing the results in the 3DMed, MSKCC, TCGA, and OAK cohorts using different methods to quantifying PD-L1 expression, most of the positive results were validated in at least two cohorts, except ERBB2. Specific genomic features are associated with differential expression of PD-L1. The differed results related to STK11, ALK, and ERBB2 alterations might be on account of the distinction between PD-L1 testing techniques, laboratory-developed E1L3N method and FDA-approved SP263/22C3 assays, and the racial differences of LUAD between western and Chinese population. A large comparative study into the regulation of PD-L1 expression by these genes across races is warranted.