mRNA Therapeutic Vaccine for Hepatitis B Demonstrates Immunogenicity and Efficacy in the AAV-HBV Mouse Model

Abstract

Chronic infection with hepatitis B virus (HBV) develops in millions of patients per year, despite the availability of effective prophylactic vaccines. Patients who resolve acute HBV infection develop HBV-specific polyfunctional T cells accompanied by neutralizing antibodies, while in patients with chronic hepatitis B (CHB), immune cells are dysfunctional and impaired. We describe a lipid nanoparticle (LNP)-formulated mRNA vaccine, optimized for the expression of HBV core, polymerase, and surface (preS2-S) antigens with the aim of inducing an effective immune response in patients with CHB. Prime and prime/boost vaccination with LNP-formulated mRNA encoding for core, pol, and/or preS2-S dosing strategies were compared in naive C57BL/6 and BALB/c mice. Immune responses were assessed by IFN-γ ELISpot, intracellular cytokine staining (ICS), and ELISA for antibody production, whereas anti-viral efficacy was evaluated in the AAV-HBV mouse model. The mRNA vaccine induced strong antigen-specific polyfunctional T cell responses in these mouse models, accompanied by the emergence of anti-HBs and anti-HBe antibodies. After three immunizations, the antigen-specific immune stimulation resulted in up to 1.7 log₁₀ IU/mL reduction in systemic HBV surface antigen (HBsAg), accompanied by a transient drop in systemic HBeAg, and this was observed in 50% of the AAV-HBV-transduced mice in the absence of additional modalities such as adjuvants, HBsAg reducing agents, or checkpoint inhibitors. However, no treatment-related effect on viremia was observed in the liver. These results warrant further optimization and evaluation of this mRNA vaccine as a candidate in a multimodal therapeutic regimen for the treatment of chronic HBV infection.

Keywords: AAV-HBV mice; HBsAg reduction; chronic hepatitis B; lipid nanoparticles; mRNA vaccine; therapeutic vaccination.

Figure 1

Vaccine composition (a) The vaccine is composed of mRNA encoding for antigens of the HBV virion: core, polymerase, and preS2-S envelope protein; (b) LNPs containing mRNA encoding for a single HBV antigen; (c) the coformulated mRNA vaccine is a mixture of the three different LNPs, each containing mRNA encoding for a single HBV antigen; (d) the co-encapsulated formulation contains mRNA encoding for all antigens, encapsulated within one LNP. (One mRNA copy per LNP is for illustration only; created with BioRender.com, accessed on 19 January 2024.)

Figure 2

Evaluation of induced immune responses in naïve C57BL/6 mice after vaccination with MC3-formulated mRNA vaccine encoding for one HBV antigen. (a) Dosing scheme for prime and prime/boost strategy. Mice that were only primed received a vaccine on day 0 where responses are evaluated on day 14. In a prime/boost strategy, mice received a prime (day 0) and a boost (day 21) with an evaluation of the responses on day 28 (created with BioRender.com, accessed on 19 January 2024). (b) Induced immune responses, as measured by IFN-γ ELISpot (vertical axis), in a dose range (horizontal axis) finding (5, 2, 1, and 0.2 µg) after a prime dose of core-expressing mRNA (n = 6). (c) Induced immune responses, as measured by IFN-γ ELISpot (vertical axis), split per peptide stimulus (horizontal axis) in a comparison between prime (white) and prime/boost (grey) dosing schedules for the dosed formulated mRNA core (circles), pol (triangles), and preS2-S (open circles) at 5 µg (n = 6). (d) Induced immune responses, as measured by IFN-γ ELISpot (vertical axis), split per peptide stimulus (horizontal axis) in a comparison between different doses (10 µg white, 5 µg light grey, and 2 µg dark grey) of formulated mRNA core (circles), pol (triangles), and preS2-S (open circles) in a prime/boost dosing strategy (n = 6). Results are shown as mean ± standard deviation. Statistical analysis is performed by one-way ANOVA of the mean of log₁₀ transformed data points (* p ≤ 0.05, ** p ≤ 0.01, ****, p ≤ 0.0001).

Figure 3

Evaluation of dose finding study in naive C57BL/6 mice after vaccination with MC3-coformulated mRNAs. Naive C57BL/6 mice (n = 6) were dosed via prime/boost strategy using 15 µg (white bars), 6 µg (light grey bars), and 3 µg (dark grey bars) of coformulation encoding for the HBV antigens. (a) Induced immune responses, as measured by IFN-γ ELISpot (vertical axis), split per peptide stimulus (horizontal axis). (b) Induced concentrations of anti-HBs (vertical axis), as measured by CLIA, on day 14 and 28 (horizontal axis). % of polyfunctional CD8⁺ T cells (c) and CD4⁺ T cells (d) defined as % of IFN-γ and TNF-α double positive cells of parent (vertical axis), split per peptide stimulation (horizontal axis). Results are shown as mean ± standard deviation. Statistical comparison is performed by unpaired t-test or one-way ANOVA on log₁₀ transformed data for ELISpot and on logit transformed data for ICS (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, and **** for p ≤ 0.0001).

Figure 4

Evaluation of induced immune responses in naive C57BL/6 mice after vaccination with coformulated versus co-encapsulated (MC3) mRNA encoding for HBV antigens. Naive C57BL/6 mice (n = 6) were dosed via prime/boost strategy using coformulated (open circles) or co-encapsulated (closed circles) mRNA encoding for the three HBV antigens using a total of 15 µg. (a) Induced immune responses, as measured by IFN-γ ELISpot (vertical axis), split by peptide stimulation (horizontal axis). Levels of induced anti-HBs (b) and anti-HBe (c) antibodies (vertical axis) in the serum measured via CLIA one day −1, 14, and 28 (horizontal axis). % of polyfunctional CD8⁺ T cells (d) and CD4⁺ T cells (e) defined as % of IFN-γ and TNF-α double positive cells of parent (vertical axis), split per peptide stimulation (horizontal axis). Results are shown as mean ± standard deviation. Statistical comparison is performed by unpaired t-test or one-way ANOVA on log₁₀ transformed data for ELISpot and antibodies, and on logit transformed data for ICS (* p ≤ 0.05).

Figure 5

Evaluation of induced immune responses in naive BALB/c mice after vaccination with ALC-0315-co-encapsulated mRNA encoding for HBV antigens. Naive BALB/c mice were dosed via a prime/boost strategy using co-encapsulated mRNA encoding for the three HBV antigens using 6 µg (black circles, n = 8) and were compared to saline control (grey circles; n = 4). (a) Induced immune responses, as measured by IFN-γ ELISpot (vertical axis), split by peptide stimulation (horizontal axis). (b) Induced concentrations of anti-HBs (vertical axis), measured by CLIA, on day −1, 14, 21, and 28 (horizontal axis). % of polyfunctional CD8⁺ T cells (c) and CD4⁺ T cells (d) defined as % of IFN-γ and TNF-α double positive cells of parent (vertical axis), split per peptide stimulation (horizontal axis). Results are shown as mean ± standard deviation. Statistical comparison is performed by unpaired t-test or one-way ANOVA on log₁₀ transformed data for ELISpot and antibodies, and on logit transformed data for ICS (** p ≤ 0.01, *** p ≤ 0.001, and **** for p ≤ 0.0001).

Figure 6

Evaluation of ALC-0315-co-encapsulated mRNA vaccine in AAV-HBV-transduced C57BL/6 mice. (a) Scheme of AAV-HBV-transduced mouse experiment where the mice were dosed 77 days after transduction (n = 8). Vaccination with co-encapsulated mRNA vaccine occurred three times with 21-day intervals where the first dose was set on day 0. Blood was drawn weekly (red tubes), and on day −14, 7, 28, 49, and 79 post dosing, leucocytes were isolated for IFN-γ ELISpot (plates). On all other weeks, HBsAg, HBeAg, anti-HBs, and anti-HBe were measured. Mice were vaccinated at 10 µg (black) compared to saline control (grey). (created with BioRender.com, accessed on 19 January 2024) (b) Induced immune responses (mean ± SD) over time (horizontal axis), as measured by IFN-γ ELISpot (vertical axis), split per peptide stimulation (core: circles; pol1: triangles; pol2: reverse open triangles; and preS2-S, squares). (c) Induced concentrations (vertical axis) of anti-HBs (triangles) or anti-HBe (circles) antibodies, measured via CLIA for vaccinated (black) and saline control (grey) shown for each individual mouse. Log₁₀ change from baseline (vertical axis) for concentrations of HBsAg (d) and HBeAg (e) measured via CLIA for mRNA-vaccinated mice (blue circles for the mice without effect on HBsAg levels and black for the mice with >1 log₁₀ reduction in HBsAg levels, for which each mouse is visualized with a different symbol that reoccurs in subsequent figures); and saline control (grey) shown for each individual mouse.

Figure 7

Evaluation of liver and spleen at the end point of the AAV-HBV-transduced mouse experiment. (a) Schematic overview of performed assays on liver and spleen at the end of the efficacy study in AAV-HBV-transduced mice. Flow cytometry was performed on purified splenocytes, ALT-levels on serum, and viral parameters were assessed on FFPE slides of liver. The spleen was used to check for polyfunctional T cells (created with BioRender.com, accessed on 24 February 2024). (b) ALT levels measured over time in the AAV-HBV experiment. (c) % of the liver area positive of HBsAg and the proportion of HBcAg-positive cells (vertical axis) for vaccinated (black and blue; corresponding to Figure 6d,e) and saline control group (grey) measured using immunohistochemistry on FFPE slides from the liver (dotted bars). % of polyfunctional CD8⁺ T cells (d) and CD4⁺ T cells (e) defined as % of IFN-γ and TNF-α double positive cells of parent (vertical axis) in spleen, split by stimulation peptide (horizontal axis). For (c–e): blue circles are used for the mice without any effect on HBsAg levels and black for the mice with >1 log₁₀ reduction in HBsAg levels, for which each mouse is visualized with a different symbol corresponding to the symbols in Figure 6d,e); saline control in grey. Results are shown as mean ± standard deviation. Statistical comparison is performed by unpaired t-test or one-way ANOVA on logit transformed data (* p ≤ 0.05, *** p ≤ 0.001, and **** for p ≤ 0.0001).