EGFR L858R mutation as a possible target for individual-independent immunotherapy in Chinese population.

8538 Background: Neoantigens arise from tumor-specific mutations and potentially provoke immune responses. General vaccines targeting these peptides could be beneficial for patients suffering from common cancers, like lung cancer. Therefore, a retrospective analysis was performed on 799 non-small cell lung cancer (NSCLC) tissue samples previously profiled using our 1021-gene panel. Each sample was collected from a unique patient, from whom peripheral blood or normal tissue was also obtained as control. Methods: Sequencing data were generated and pre-analyzed according to our in-house procedures. HLA typing was done using OptiType v1.0 (required sequences were captured by 1021-gene panel) and neoantigens were predicted by netMHCpan v4.0 based on typed HLA alleles and curated non-frameshift somatic mutations with frequency > 5%, which were called in pre-analysis. A neoantigen is considered mutant-specific if IC50 mut is < 500 nM and IC50 wild is > 500 nM, and especially, it is considered a strong-binder if IC50 mut is < 50 nM. Results: HLA typing returned 141 unique alleles, with the top 3 by carrier frequency being A*1101 (39%), C*0102 (33%) and A*2402 (28%). A further investigation into HLA alleles, mutations and neoantigens revealed two mutations on EGFR as candidates for off-the-shelf vaccine development: (1) L858R mutation (19%, 151 out of 799) and (2) E746_A750del mutation (13%, 106 out of 799). Among the four neoantigens derived from EGFR L858R mutation is HVKITDFGR, which can be recognized by A*3303 (IC50 mut = 22.93 nM and IC50 wild = 12,733.96 nM) and the combination is shared by 3% of the patients (23 out of 799), despite that A*3303 is not a very frequent allele in this population (16%, 127 out of 799). Two neoantigens were derived from EGFR E746_A750del mutation, including IPVAIKTSPK, which is mainly recognized by A*1101 (IC50 mut = 158.16 nM and IC50 wild = 31,132.66 nM). This combination is shared by 5% of the patients (41 out of 799). Conclusions: (1) EGFR L858R mutation and HLA-A*3303 could be a good target for individual-independent vaccine development. (2) HLA-A*1101 is the most frequent allele in this population. However, HLA-A*1101 and E746_A750del mutation is not so ideal for off-the-shelf vaccine development.