Abstract 3578: Optimization of shared neoantigen vaccine design to increase vaccine potency: From bench to bedside and back

Abstract Vaccines targeting neoantigens identified from common tumor driver mutations are of increasing interest as evidence of clinical benefit builds, and opportunities to combine such vaccines with immune modulators are growing. Our individualized neoantigen vaccine (GRANITE) has shown clinical benefit and strong, consistent CD8 T cell induction in patients. We have developed an analogous off-the-shelf product (SLATE) targeting shared neoantigens that offers manufacturing simplicity and faster administration to patients with shared driver mutations. A Phase 1/2 trial of a heterologous prime/boost vaccine regimen using a chimpanzee adenovirus (ChAd) prime and self amplifying mRNA (SAM) boosts (SLATE, NCT03953235) was initiated to assess safety, tolerability, and immunogenicity in patients with advanced cancers. SLATE version 1 encodes 20 unique neoantigens to various shared driver mutations (KRAS, TP53, etc.). Patients were selected if their tumors harbored one of the 20 neoantigens encoded by the vaccine cassette and an HLA Class I allele that presents that neoantigen. Administration of ChAd prime and repeated administration of 30, 100, or 300µg SAM doses were safe and well tolerated in all subjects dosed (n=26), with no evidence of increasing reactogenicity with sequential dosing. Early efficacy signals (molecular responses; one unconfirmed RECIST response) were observed in NSCLC subjects all treated with and progressed on prior anti-PD(L)1. Analysis of T cell responses pre and post immunizations by ex vivo IFNγ ELISpot did not show robust responses to KRAS neoantigens across all patients. However, objective CD8 T cell responses to KRAS antigens post vaccination were detectable after in vitro stimulation, suggesting the induction of low-level KRAS specific T cell responses in vivo. In contrast, HLA-matched responses to TP53 neoantigens encoded by the vaccine were consistently detected via ex vivo ELISpot in these same patients. These data suggests that an immunodominant T cell response to the TP53 mutations may have outcompeted the response to the less immunogenic KRAS mutations restricted and presented by the same HLA in vivo. Differential surface peptide-HLA (pHLA) density may explain these discordant findings, and subsequent targeted mass spectrometry analyses revealed detection of TP53 pHLA complexes at a higher frequency compared to KRAS mutations in single HLA-allele cell lines. Redesigned vaccine cassettes excluding the TP53 epitopes and repeating KRAS epitopes demonstrated increased immune responses (ex vivo IFNγ ELISpot) compared to cassette version 1 in HLA transgenic mice, further supporting the tumor neoantigen immunodominance hierarchy observed in humans dosed with SLATE version 1. A re-designed product (SLATE v2) focusing exclusively on KRAS mutations (G12C, G12D, G12V and Q61H) is currently being assessed in phase 2 in patients with advanced KRAS-driven tumors. Citation Format: Amy R. Rappaport, Christine D. Palmer, Annie Shen, Claudia X. Dominguez, Meghan G. Hart, Lauren D. Kraemer, Sonia Koulavouth, Martina Marrali, Jason R. Jaroslavsky, Charmaine N. Nganje, Ciaran D. Scallan, Sue-Jean Hong, Leonid Gitlin, Monica Lane, Daniel V. Catenacci, Chrisann Kyi, David P. Carbone, Hossein Borghaei, Raphael Rousseau, Andrew Ferguson, Karin Jooss. Optimization of shared neoantigen vaccine design to increase vaccine potency: From bench to bedside and back [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3578.