VENUS, a Novel Selection Approach to Improve the Accuracy of Neoantigens' Prediction

Abstract

Neoantigens are tumor-specific antigens able to induce T-cell responses, generated by mutations in protein-coding regions of expressed genes. Previous studies demonstrated that only a limited subset of mutations generates neoantigens in microsatellite stable tumors. We developed a method, called VENUS (Vaccine-Encoded Neoantigens Unrestricted Selection), to prioritize mutated peptides with high potential to be neoantigens. Our method assigns to each mutation a weighted score that combines the mutation allelic frequency, the abundance of the transcript coding for the mutation, and the likelihood to bind the patient's class-I major histocompatibility complex alleles. By ranking mutated peptides encoded by mutations detected in nine cancer patients, VENUS was able to select in the top 60 ranked peptides, the 95% of neoantigens experimentally validated including both CD8 and CD4 T cell specificities. VENUS was evaluated in a murine model in the context of vaccination with an adeno vector encoding the top ranked mutations prioritized in the MC38 cell line. Efficacy studies demonstrated anti tumoral activity of the vaccine when used in combination with checkpoint inhibitors. The results obtained highlight the importance of a combined scoring system taking into account multiple features of each tumor mutation to improve the accuracy of neoantigen prediction.

Keywords: MC38; VENUS; cancer vaccine; neoantigen; prediction.

Figure 1

Schematic description of the vaccine-encoded neoantigens unrestricted selection (VENUS) RSUM prioritization method: Schematic description of the ranking procedure applied with VENUS RSUM score. The mutations-encoded neo-peptides are ranked independently three times using three different parameters. For each neo-peptide, the individual ranks are summed and corrected according to weighting factors (k; WF; details in methods) that penalize neo-peptides with a predicted IC50 > 1000 nM (k) and mutations that fall in regions with low read coverage or within not expressed genes according to the next generation sequencing (NGS) mRNA transcriptome data (WF).

Figure 2

VENUS ranking of experimentally validated neoantigens: Ranking of 20 experimentally validated neoantigens according to VENUS RSUM score. Better ranks correspond to lower values on the y-axis. Grey crosses represent all neo-peptides generated by somatic mutations in each patient. Experimentally validated CD8+ and CD4+ reactivities are depicted in red and blue, respectively. Horizontal lines indicate thresholds for the top 20 (black) and top 60 (green) neoantigens.

Figure 3

Comparison of VENUS performance against alternative methods that include less parameters. Light blue bars represent the percentage of validated neoantigens included in the top 60 (A) or top 20 (B) selection performed using the indicated alternative single or combination of parameters. For comparison, the percentage of validated neoantigens included within the top 60 (A) or 20 (B) neoantigens by VENUS. Funnel filtering is performed by retaining only the neo-peptides predicted as major histocompatibility complex (MHC)-I binders (9 mer; predicted IC50 ≤ 500 nM) and encoded by an expressed gene (transcripts for million (TPM) ≥ 0.50).

Figure 4

Vaccination with VENUS-identified neoantigens encoded in a GAd vector is effective in the established MC38 tumor model in combination with anti-PD1. (A) Schematic of the approach used to identify MC38 tumor specific mutations and generation of the vaccine. (B) In vivo immunogenicity of GAd-MC38-62. T-cell responses were measured by IFN-γ ELISpot on splenocytes of naive mice 3 weeks post immunization with 5 × 10⁸ vp of GAd-MC38-62. Responses against the eight immunogenic neoantigens (nAgs) are shown. Neoantigens IDs inducing CD8+ or CD4+ T-cell responses are indicated in red and blue, respectively. Data are representative of two independent experiments. (C) Mice were inoculated s.c. with MC38 cells. One week later, animals were randomized according to tumor volume and treated with anti-PD1 alone or in combination with GAd-MC38-62. Vaccine was administered at day 0 (i.m.) following randomization, whereas anti-PD1 was given twice per week until day 17 (i.p.). Tumor growth over time is shown for individual mice. Black curves indicate responder mice showing a complete response post treatment. (D) Frequency of tumor free (black) and tumor bearing mice (white) upon treatment with GAd-MC38-62 and anti-PD1 depleted for CD8+ T cells or undepleted.