Abstract Objective: Spontaneous neo-antigen specific immune responses are rare in animals and humans. Our AB1 model, one of the few induced by a natural human carcinogen, has several. Given that current immunotherapies rely on unleashing responses to existing tumour neo-antigens, we evaluated the capacity of such responses to inform subsequent therapies, determining if these responses are: a) more sensitive than imaging to detect tumor recurrence after surgical resection b) able to predict responses to immune checkpoint blockade (ICPB) therapy c) able to provide evidence of the immunogenicity of chemotherapy. Methods: To examine this, we interrogated the immune response to tumour neo-antigens in BALB/c mice bearing the AB1-HA tumor model.AB1-HA whole exome sequencing and RNAseq data were analysed and the MHC-I binding affinity of mutated sequences were predicted and expression confirmed with RNA-seq data. A total of 180 neo-antigens were screened, in addition to the responses to the known neo-antigens UNC45a and UQCRC2, spontaneous immune responses to two novel neo-antigens, VMN2R67 and ELP2 were found in tumour bearing mice. a) To determine if responses to these neo-antigens detected post-resection tumor recurrence tumours were surgically resected then growth of IV luciferase transfected cells determined sequentially by an In Vivo Imaging System (IVIS) and imaged for peak bioluminescence in a Lumina II Imager (Figure), PET-CT (IV 15-FDG at 7-15MBq). For neo-antigen specific T cell detection, IFNg ELISPOT assays were performed. b) To determine if pre-treatment neo-antigen responses are able to predict responses to ICPB, we utilised a version of the dual-tumor model developed by Lesterhuis et al, enabling pre-treatment neo-antigen responses to ICPB blockade (anti-CTLA-4 + anti-GITR) compared in subsequent responders versus non-responders. c) To determine if a known immunogenic chemotherapy (gemcitabine) could boost the efficacy of vaccines designed based on spontaneous neo-antigen reactivity, we treated tumour bearing mice with gemcitabine ± a vaccine containing the neoantigen peptides Unc45a, Uq2, Vmn2r67, and Elp2 with Poly I:C. Results: a) A combination of 3 neo-antigen specific T cell responses detected metastases detection prior to IVIS or PET-CT (86±11 SFU/105 cells for the recurrence group compared to 17±5 SFU/105 cells in the non-recurrence group, p<0.0001).b) Pre-treatment dLN UNC45a response was higher in the responders (24 ± 14 SFU/105 vs 5 ± 2 SFU/105, p = 0.0004). Thus pretreatment responses to some neo-antigens could strongly predict response to ICPB therapy.c)Gemcitabine + vaccine delayed tumour growth compared to either treatment or adjuvant alone (p<0.05). Conclusions: Neo-antigen specific T cell responses increases sensitivity of recurrence/metastasis detection, predict ICPB therapy outcome and inform vaccine design for synergy with immunogenic chemotherapy. Citation Format: Bruce W. Robinson, Vanessa Fear, Catherine Forbes, Shaokang Ma, Jonathan Chee, Sam Neeve, Scott Fisher, Ian Dick, Anna Nowak, Nandini Makwana, Jessica Boulter, Jenette Creaney, Alec Redwood. Neo-antigen specific T cell responses can inform cancer therapy [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3292.
Read full abstract