Abstract 86P
Background
Ephrin type-A receptors (EPHA) are members of the EPH family of receptor tyrosine kinases (RTKs) with nine EPHA receptors (EPHA1-8, EPHA10). Previous studies demonstrated that EPHA5 and EPHA7 mutations could predict the clinical benefit of immune checkpoint inhibitors (ICIs). We herein investigated the associations of EPHA3 mutations with tumor mutation burden (TMB), PD-L1 expression, microsatellite instability-high (MSI-H), tumor-infiltrating lymphocytes (TILs), and clinical response to ICIs. This will be very helpful for identifying patients who may benefit from immunotherapy.
Methods
The ICIs cohort from the MSKCC was used for exploring the associations between EPHA3 mutations and ICIs efficacy. The relationships between EPHA3 mutations and TMB, PD-L1 expression, and MSI-H were investigated in our Chinese cohort. The TCGA cohort was used for analyzing the link between EPHA3 mutations and TILs.
Results
First, we analyzed the associations between EPHA3 mutations and overall survival (OS) after ICIs therapy. For all enrolled patients, those who harbored EPHA3 mutations had a relatively longer OS compared to those without mutations (40 vs 18 months, P = 0.0102). However, EPHA3 mutations displayed divergent predictive roles in different cancer types. The median OS was significantly longer in the Mut group compared to the WT group for NSCLC (36 vs 11 months, P = 0.0167). There were no associations between EPHA3 mutations and ICIs efficacy for patients with melanoma, glioma, bladder, colorectal, esophagogastric, breast, and head and neck cancers. Second, the results from our cohort showed that the median TMB was higher in the Mut group for NSCLC (14.6 vs 6.6 muts/Mb, P < 0.0001). And we did not find the associations of EPHA3 mutations with PD-L1 expression and MSI-H. Third, the analysis in the TCGA cohort revealed that NSCLC patients with EPHA3 mutations were infiltrated with increased activated natural killer cells (P = 0.0131).
Conclusions
Our data indicated that EPHA3 mutations were associated with prolonged OS in NSCLC, rather than other cancer types, suggesting that it might act as a predictive biomarker for ICIs therapy in NSCLC. EPHA3 mutations were also correlated with higher TMB and increased activated natural killer cells in NSCLC.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.