Vaccination Against Androgen Receptor Splice Variants to Immunologically Target Prostate Cancer

Abstract

Background/Objectives: Androgen receptor (AR) expression and signaling are critical for the progression of prostate cancer and have been the therapeutic focus of prostate cancer for over 50 years. While a variety of agents have been developed to target this axis, many of these fail due to the emergent expression of AR RNA splice variants, such as AR-V7, that can signal independently of ligand binding. Other therapies, such as vaccination against prostate-specific antigens, have achieved FDA approvals but have fallen short of being incorporated as standard-of-care therapies for advanced prostate cancer. This may be due to the elevated level of immunosuppression observed in prostate cancer, which remains largely refractory to immune checkpoint blockade. Methods: We developed a vaccine targeting AR-V7, a common isoform associated with treatment resistance, and demonstrated its ability to elicit AR-V7-specific immunity and enable anti-tumor responses against AR-V7+ cancers in subcutaneous tumor models. Results: Our studies also revealed that AR-V7 expression conferred an immune suppressive phenotype that was significant in a non-AR-dependent prostate cancer model. Notably, in this model, we found that vaccination in combination with enzalutamide, an AR antagonist, suppressed these aggressive immune suppressive cancers and resulted in enhanced survival in comparison to control vaccinated and enzalutamide-treated mice. While anti-PD-1 immune checkpoint inhibition (ICI) alone slowed tumor growth, the majority of vaccinated mice that received anti-PD-1 therapy showed complete tumor elimination. Conclusions: Collectively, these results validate the importance of AR signaling in prostate cancer immune suppression and suggest the potential of AR-V7-specific vaccines as therapeutic strategies against prostate cancer, offering significant protective and therapeutic anti-tumor responses, even in the presence of androgen signaling inhibitors.

Keywords: androgen receptor; castration resistance; immune checkpoint inhibition; immune suppression; prostate cancer; vaccines.

Figure 3

(a) Tumor growth curves for P3CA parental in B6 males with or without enzalutamide combination treatment (n = 5); (b) Western blot for AR in P3CA-Empty, P3CA-AR, and P3CA-AR-V7 cell lines; (c) Luciferase reporter assay on P3CA-Empty and P3CA-AR-V7 cell lines; (d) GO pathway analysis on differentially expressed genes up (black bars) and down (white bars) in P3CA-AR-V7 compared to P3CA-Empty; (e) Heatmap of the z-scores for the 30 most significantly differentially expressed genes between P3CA-Empty and P3CA-AR-V7 from bulk RNA sequencing (* p < 0.05).

Figure 3

(a) Tumor growth curves for P3CA parental in B6 males with or without enzalutamide combination treatment (n = 5); (b) Western blot for AR in P3CA-Empty, P3CA-AR, and P3CA-AR-V7 cell lines; (c) Luciferase reporter assay on P3CA-Empty and P3CA-AR-V7 cell lines; (d) GO pathway analysis on differentially expressed genes up (black bars) and down (white bars) in P3CA-AR-V7 compared to P3CA-Empty; (e) Heatmap of the z-scores for the 30 most significantly differentially expressed genes between P3CA-Empty and P3CA-AR-V7 from bulk RNA sequencing (* p < 0.05).

Figure 3

(a) Tumor growth curves for P3CA parental in B6 males with or without enzalutamide combination treatment (n = 5); (b) Western blot for AR in P3CA-Empty, P3CA-AR, and P3CA-AR-V7 cell lines; (c) Luciferase reporter assay on P3CA-Empty and P3CA-AR-V7 cell lines; (d) GO pathway analysis on differentially expressed genes up (black bars) and down (white bars) in P3CA-AR-V7 compared to P3CA-Empty; (e) Heatmap of the z-scores for the 30 most significantly differentially expressed genes between P3CA-Empty and P3CA-AR-V7 from bulk RNA sequencing (* p < 0.05).

Figure 3

(a) Tumor growth curves for P3CA parental in B6 males with or without enzalutamide combination treatment (n = 5); (b) Western blot for AR in P3CA-Empty, P3CA-AR, and P3CA-AR-V7 cell lines; (c) Luciferase reporter assay on P3CA-Empty and P3CA-AR-V7 cell lines; (d) GO pathway analysis on differentially expressed genes up (black bars) and down (white bars) in P3CA-AR-V7 compared to P3CA-Empty; (e) Heatmap of the z-scores for the 30 most significantly differentially expressed genes between P3CA-Empty and P3CA-AR-V7 from bulk RNA sequencing (* p < 0.05).

Figure 1

(a) Androgen receptor splice variant exon structures for AR, AR-V7 and AR-V12; (b) IFNγ ELISpot counts on splenocytes from B6 mice two weeks post-IM vaccination (n = 10); (c) IFNγ ELISpot counts on splenocytes from DO mice two weeks post-IM vaccination (n = 10, two-way ANOVA, with Bonferroni multiple comparisons); (d,e) Intracellular IFNγ+ staining of CD8+ T cells (d) or CD4+ T cells (e) from DO mice two weeks post-IM vaccination (n = 10, two-way ANOVA, with Bonferroni multiple comparisons) (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 1

(a) Androgen receptor splice variant exon structures for AR, AR-V7 and AR-V12; (b) IFNγ ELISpot counts on splenocytes from B6 mice two weeks post-IM vaccination (n = 10); (c) IFNγ ELISpot counts on splenocytes from DO mice two weeks post-IM vaccination (n = 10, two-way ANOVA, with Bonferroni multiple comparisons); (d,e) Intracellular IFNγ+ staining of CD8+ T cells (d) or CD4+ T cells (e) from DO mice two weeks post-IM vaccination (n = 10, two-way ANOVA, with Bonferroni multiple comparisons) (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 2

(a) Luciferase reporter assay on HEK293T cells shows that AR- and AR-V7-overexpressing lentiviral vectors result in strong signaling in ligand-dependent or ligand-independent manner, respectively; (b) Tumor growth curve of CT-26 tumors expressing AR, AR-V7 or AR-V12 in the flank of BALB/c male mice (n = 5); (c) CT-26-AR-V7 tumor growth curves in mice that were vaccinated IM two weeks prior to tumor implantation with Ad-AR, Ad-AR-V7 or Ad-Control (n = 10, two-way ANOVA, with Bonferroni multiple comparisons); (d) Kaplan–Meier plot of BALB/c male mice vaccinated before CT26-AR-V7 tumor implantation (n = 10, Gehan–Breslow–Wilcoxon test); (e) IFNγ ELISpot counts on splenocytes from tumor-rejecting mice 100 days post-tumor-implantation (n = 1, n = 6, n = 7 in Ad-Control, Ad-AR, and Ad-AR-V7, respectively); (f) CT26-AR-V7 tumor growth curves in mice that were vaccinated IM one day post-tumor-implantation with Ad-AR, Ad-AR-V7 or Ad-Control (two-way ANOVA, with Bonferroni multiple comparisons); (g) Kaplan–Meier plot of BALB/c male mice vaccinated after CT26-AR-V7 tumor implantation (n = 10, Gehan–Breslow–Wilcoxon test) (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 2

(a) Luciferase reporter assay on HEK293T cells shows that AR- and AR-V7-overexpressing lentiviral vectors result in strong signaling in ligand-dependent or ligand-independent manner, respectively; (b) Tumor growth curve of CT-26 tumors expressing AR, AR-V7 or AR-V12 in the flank of BALB/c male mice (n = 5); (c) CT-26-AR-V7 tumor growth curves in mice that were vaccinated IM two weeks prior to tumor implantation with Ad-AR, Ad-AR-V7 or Ad-Control (n = 10, two-way ANOVA, with Bonferroni multiple comparisons); (d) Kaplan–Meier plot of BALB/c male mice vaccinated before CT26-AR-V7 tumor implantation (n = 10, Gehan–Breslow–Wilcoxon test); (e) IFNγ ELISpot counts on splenocytes from tumor-rejecting mice 100 days post-tumor-implantation (n = 1, n = 6, n = 7 in Ad-Control, Ad-AR, and Ad-AR-V7, respectively); (f) CT26-AR-V7 tumor growth curves in mice that were vaccinated IM one day post-tumor-implantation with Ad-AR, Ad-AR-V7 or Ad-Control (two-way ANOVA, with Bonferroni multiple comparisons); (g) Kaplan–Meier plot of BALB/c male mice vaccinated after CT26-AR-V7 tumor implantation (n = 10, Gehan–Breslow–Wilcoxon test) (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 4

(a) Tumor growth curves for SCID and B6 albino male mice implanted subcutaneously with P3CA-Empty or P3CA-AR-V7 (n = 5); (b) Kaplan–Meier plot of SCID male mice with P3CA-Empty or P3CA-AR-V7 (n = 5, Log-rank test); (c) Kaplan–Meier plot of B6 albino male mice with P3CA-Empty or P3CA-AR-V7 (n = 5, Log-rank test); (d) Tumor growth curves for P3CA-Empty and P3CA-AR-V7 in B6 albino males with or without CD8 depletion (n = 5); (e) MHC-I expression on P3CA cell lines by flow cytometry (dotted—no secondary control, dashed—P3CA-Empty, solid—P3CA-AR-V7); (f) IHC quantification from P3CA-Empty and P3CA-AR-V7 tumors from B6 albino males, collected 7 days post-implantation (n = 5, unpaired t test with Welch correction and Holm-Šídák correction for multiple comparisons) (* p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 4

(a) Tumor growth curves for SCID and B6 albino male mice implanted subcutaneously with P3CA-Empty or P3CA-AR-V7 (n = 5); (b) Kaplan–Meier plot of SCID male mice with P3CA-Empty or P3CA-AR-V7 (n = 5, Log-rank test); (c) Kaplan–Meier plot of B6 albino male mice with P3CA-Empty or P3CA-AR-V7 (n = 5, Log-rank test); (d) Tumor growth curves for P3CA-Empty and P3CA-AR-V7 in B6 albino males with or without CD8 depletion (n = 5); (e) MHC-I expression on P3CA cell lines by flow cytometry (dotted—no secondary control, dashed—P3CA-Empty, solid—P3CA-AR-V7); (f) IHC quantification from P3CA-Empty and P3CA-AR-V7 tumors from B6 albino males, collected 7 days post-implantation (n = 5, unpaired t test with Welch correction and Holm-Šídák correction for multiple comparisons) (* p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 5

(a) Tumor growth curves of C57BL/6 male mice on enzalutamide-diet-vaccinated two weeks prior to P3CA-AR-V7 tumor implantation (n = 10, two-way ANOVA, with Bonferroni multiple comparisons); (b) Kaplan–Meier plot of C57BL/6 male mice on enzalutamide-diet-vaccinated two weeks prior to P3CA-AR-V7 tumor implantation (n = 10, Gehan–Breslow–Wilcoxon test); (c) Tumor growth curves of B6 albino male mice vaccinated two weeks prior to P3CA-AR-V7 tumor implantation, with or without anti-PD-1 antibody therapy (n = 5, two-way ANOVA, with Bonferroni multiple comparisons); (d) Kaplan–Meier plot of B6 albino male mice vaccinated two weeks prior to P3CA-AR-V7 tumor implantation, with or without anti-PD-1 antibody therapy (n = 5, Gehan–Breslow–Wilcoxon test) (* p < 0.05, ** p < 0.01, **** p < 0.0001).

Figure 5

(a) Tumor growth curves of C57BL/6 male mice on enzalutamide-diet-vaccinated two weeks prior to P3CA-AR-V7 tumor implantation (n = 10, two-way ANOVA, with Bonferroni multiple comparisons); (b) Kaplan–Meier plot of C57BL/6 male mice on enzalutamide-diet-vaccinated two weeks prior to P3CA-AR-V7 tumor implantation (n = 10, Gehan–Breslow–Wilcoxon test); (c) Tumor growth curves of B6 albino male mice vaccinated two weeks prior to P3CA-AR-V7 tumor implantation, with or without anti-PD-1 antibody therapy (n = 5, two-way ANOVA, with Bonferroni multiple comparisons); (d) Kaplan–Meier plot of B6 albino male mice vaccinated two weeks prior to P3CA-AR-V7 tumor implantation, with or without anti-PD-1 antibody therapy (n = 5, Gehan–Breslow–Wilcoxon test) (* p < 0.05, ** p < 0.01, **** p < 0.0001).