Neoantigen cancer vaccine augments anti-CTLA-4 efficacy

Abstract

Immune checkpoint inhibitors (ICI) based on anti-CTLA-4 (αCTLA-4) and anti-PD1 (αPD1) are being tested in combination with different therapeutic approaches including other immunotherapies such as neoantigen cancer vaccines (NCV). Here we explored, in two cancer murine models, different therapeutic combinations of ICI with personalized DNA vaccines expressing neoantigens and delivered by electroporation (EP). Anti-cancer efficacy was evaluated using vaccines with or without CD4 epitopes. Therapeutic DNA vaccines showed synergistic effects in different therapeutic protocols including established large tumors. Flow cytometry (FC) was utilized to measure CD8, CD4, Treg, and switched B cells as well as neoantigen-specific immune responses, which were also measured by IFN-γ ELIspot. Immune responses were augmented in combination with αCTLA4 but not with αPD1 in the MC38 tumor-bearing mice, significantly impacting tumor growth. Similarly, neoantigen-specific T cell immune responses were enhanced in combined treatment with αCTLA-4 in the CT26 tumor model where large tumors regressed in all mice, while monotherapy with αCTLA-4 was less efficacious. In line with previous evidence, we observed an increased switched B cells in the spleen of mice treated with αCTLA-4 alone or in combination with NCV. These results support the use of NCV delivered by DNA-EP with αCTLA-4 and suggest a new combined therapy for clinical testing.

Fig. 1. Immune responses induced by M8 vaccine.

C57Bl/6 mice were vaccinated at day 0, 14 and 28, while neoantigen-specific immune responses were analyzed on day 35 on splenocytes by IFNγ ELISpot stimulated with individual peptides. ConA served as a positive control. Neoantigen-specific measurements are reported corresponding to the neoantigens listed in Table 1. Each symbol represents an individual sample (n = 3). A representative image of the experiment is on the right. Values are from one of two experiments.

Fig. 2. Therapeutic effect of M8 in combination with ICI.

C57Bl/6 mice were inoculated s.c. with MC38 cells and treated with M8 vaccine delivered by EP and ICI according to the experimental scheme. a Tumor volumes followed over time. b Final tumor volume measurements, single tumor volumes are depicted from one experiment out of two performed. Mann–Whitney tests were conducted *p < 0,05 **p < 0.01 ***p < 0.001. c At day 27, mice were sacrificed and IFNγ producing cells were evaluated by IFNγ ELISpot assay with splenocytes restimulated with neoantigens pool or ConA as a positive control. Significance was determined using Mann–Whitney test, *p < 0,05 **p < 0.01 ***p < 0.001. Six animals per group were utilized. Each symbol represents an individual sample with the error bars representing the s.e.m. Only the most relevant statistics were reported.

Fig. 3. Prophylactic C20-IM induces a significant reduction in tumor growth.

a Balb/c mice were vaccinated with the C20 vaccine at days 0, 21, and 42 and challenged with CT26 cells on day 62. One week after the last immunization (day 49), mice were bled retro-orbitally to monitor T cell immune response against CT26-neoepitopes by intracellular staining. Panel describes CD8 and CD4 neoantigen-specific effector and central memory T cell responses measured by FC using the gating strategy depicted in Supplementary Fig. 2. The stimulation pool included the 15 peptides listed in Table 2. b The panel depicts CT26 tumor growth overtime of one out of two experiments performed. Five animals per group were utilized. Each symbol represents an individual sample with the error bars representing the s.e.m. Significance was determined using Mann–Whitney test (*p < 0,05).

Fig. 4. Therapeutic effect of C20 in combination with ICI in two different schedules of vaccination.

a Balb/c mice were inoculated s.c. with CT26 cells and treated with C20 and ICI starting from day 2 as depicted in the experimental scheme. Tumor volume and survival curve. b, c Balb/c mice were inoculated s.c. with CT26 cells and treated with ICI and NCV according to the experimental scheme. b Tumor (50–100 mm³) bearing mice were randomized at day 6 and treated with αCTLA-4 and vaccinated with C20 the day after. The treatment was repeated weekly as described in the scheme. b Tumor volume measurements and survival curve. This experiment was conducted only once (C) CT26 tumor growth in CD4 or CD8 depleted mice treated as in panel (b). This experiment was repeated twice with similar results. Six animals per group were utilized with the error bars representing the s.e.m. Significance was determined using Mann–Whitney and Log-rank (Mantel–Cox) test **p < 0.01 ***p < 0.001.

Fig. 5. CD8⁺IFNγ⁺ response increased in αCTLA-4 + C20-IM treated mice.

Tumor-bearing mice were treated as described in the experimental scheme and neoantigen-specific immune response was analyzed by FC in the peripheral blood one week after the last treatment (on day 28). Production of IFNγ and TNFα by CD4⁺ and CD8⁺ T cells was measured by FC using the gating strategy depicted in Supplementary Fig. 2 upon restimulation with the pool of neoantigen peptides listed in Table 2. This experiment was conducted twice with similar results. Each symbol represents an individual sample (n = 6). Significance was determined using Mann–Whitney test, *p < 0,05 **p < 0.01.

Fig. 6. C20 and αCTLA-4 increased neoantigen-specific CD8 and switched B cells.

On day twenty, nine mice per group were sacrificed and tumors and splenocytes were analyzed by FC using the gating strategy depicted in Supplementary Fig. 2. a Frequency of E2f8 neoantigen-specific response was measured using the specific dextramer (Dx+) gated on CD45⁺CD3⁺ CD8⁺ in live cells. b Treg population (CD25⁺Foxp3⁺) were gated on CD45⁺CD3⁺CD4⁺ in live cells and expressed as ratio with DX⁺ cells (c). d Switched B cells were identified as CD19⁺IgG⁺ cells and gated on CD45⁺CD3⁺ in live cells. Nine animals per group were utilized and each symbol represents an individual sample. Significance was determined using Mann–Whitney test *p < 0.05 **p < 0.01 ***p < 0.001 ****p < 0.0001.