Neoantigen-based vaccination as a practical measure for head and neck cancer treatment

• Published in: Journal of clinical oncology Vol. 38; no. 15_suppl; p. e18528
• Main Authors: Zheng, Xiao-bin, Chen, Chuan-ben, Chen, Yu, Hao, Shiguang, Jun, Liu, Xiong, Jia-ni, Lin, Jin, Xu, Yaping, Guan, Yan-Fang, Li, Yi, Huang, Yingying, Yi, Yuting, Xia, Xuefeng
• Format: Journal Article
• Language: English
• Published: 20.05.2020
• ISSN: 0732-183X; 1527-7755
• DOI: 10.1200/JCO.2020.38.15_suppl.e18528

Abstract only e18528 Background: The view that neoantigens serve as potential vaccine targets has arisen in the last decade. Clinical and computational efforts have been done to increase the practicality of its application in real world. With these advances, we conducted a retrospective study on a Chinese population to explore the clinical feasibility of neoantigen-based vaccines for head and neck cancer treatment. Methods: Tumor and normal samples were profiled using a 1021-gene panel. Sequencing data were pre-analyzed according to our in-house standard procedures. Class I HLA typing was completed using OptiType v1.0. Curated somatic mutations in coding regions (SNVs and non-frameshift Indels with an allele frequency ≥ 5%) were collected and altered peptides produced by these mutations were analyzed using NetMHCpan v4.0. Peptides with an IC 50 < 500 nM were considered potential binders, and especially, those with an IC 50 < 50 nM were considered strong binders. An altered peptide was considered a neoantigen if IC 50 altere d is < IC 50 wildtype . Results: We analyzed a total of 243 patients and detected 114 unique HLA alleles. By carrier percentage, the top three alleles are C*01:02 (44%), B*46:01 (36%), and A*11:01 (33%). In total, 743 mutations were deemed eligible for neoantigen prediction and 223 unique neoantigens were found. Of these neoantigens, 67 (carried by 21% of patients) were strong binders, among which 26 (carried by 9% patients) exhibited a great fold change (≥ 5 folds) of binding affinity. Moreover, the neoantigens in these patients are unique, as only two neoantigens were shared. A search for shared neoantigens revealed a combination of mutation PIK3CA p.E542K and HLA A*11:01, which was detected in 0.54% of all patients. Additionally, 43.6% (106/243) of patients were diagnosed with nasopharyngeal carcinoma, among whom 42% (44/106) possessed predicted neoantigens, including 15 patients with strong-binder neoantigens. Conclusions: (1) Neoantigen-based vaccination is a practical measure to treat patients with head and neck cancer, as indicated by the percentage of patients harboring strong-binder neoantigens. (2) Off-the-shelf neoantigen vaccines may not be practical, given the result that the most common combination of a neoantigen-producing mutation and the corresponding HLA was only found in 0.54% of all patients.