Low Adenovirus Vaccine Doses Administered to Skin Using Microneedle Patches Induce Better Functional Antibody Immunogenicity as Compared to Systemic Injection

Abstract

Adenovirus-based vaccines are demonstrating promising clinical potential for multiple infectious diseases, including COVID-19. However, the immunogenicity of the vector itself decreases its effectiveness as a boosting vaccine due to the induction of strong anti-vector neutralizing immunity. Here we determined how dissolvable microneedle patches (DMN) for skin immunization can overcome this issue, using a clinically-relevant adenovirus-based Plasmodium falciparum malaria vaccine, AdHu5-PfRH5, in mice. Incorporation of vaccine into patches significantly enhanced its thermostability compared to the liquid form. Conventional high dose repeated immunization by the intramuscular (IM) route induced low antigen-specific IgG titres and high anti-vector immunity. A low priming dose of vaccine, by the IM route, but more so using DMN patches, induced the most efficacious immune responses, assessed by parasite growth inhibitory activity (GIA) assays. Administration of low dose AdHu5-PfRH5 using patches to the skin, boosted by high dose IM, induced the highest antigen-specific serum IgG response after boosting, the greatest skewing of the antibody response towards the antigen and away from the vector, and the highest efficacy. This study therefore demonstrates that repeated use of the same adenovirus vaccine can be highly immunogenic towards the transgene if a low dose is used to prime the response. It also provides a method of stabilizing adenovirus vaccine, in easy-to-administer dissolvable microneedle patches, permitting storage and distribution out of cold chain.

Keywords: adenovirus; anti-vector immunity; antibody; dose; microneedle; skin; stability; virus vector vaccine.

Figure 1

Incorporation of adenovirus vaccine into dissolvable microneedle (DMN) patches significantly stabilized the vaccine for up to 6 months. AdHu5–PfRH5 containing patches were stored in a stability chamber at 25 °C or 40 °C, with the surrounding chamber maintained at 60% or 75% relative humidity (RH), for the indicated time. (A) The initial vaccine dose was 1 × 10⁷ ifu per 1 cm² patch. The amount of remaining viable adenovirus was measured at 1 week and then 1, 2, 3, and 6 months after incubation of samples (n = 6 per group per timepoint, except at 6 months when n = 4 per group). The % remaining in the sample was determined compared to time zero. No physical changes were observed in DMN stored at accelerated stability conditions. Points represent median, error bars represent inter-quartile range. ** p < 0.01, *** p < 0.001, compared to time zero, Kruskal–Wallis, one-way ANOVA, with Dunn’s multiple comparison test. (B) 25 °C, 60% RH, or (C) 40 °C, 75% RH, for 6 months.

Figure 2

Effect of vaccine dose and route on the induction of IgG antibodies to PfRH5 antigen and to the adenovirus vector. Female Balb/c mice (n = 5 per group) were immunized on day 0 with AdHu5–PfRH5 by the IM route or using DMN patches. The following doses were administered by IM (from left to right); 1 × 10⁸ ifu, 1 × 10⁷ ifu, 1 × 10⁵ ifu, or 1 × 10⁴ ifu to each mouse. Alternatively, 5 × 10⁷ ifu, 5 × 10⁵ ifu, or 5 × 10⁴ ifu were administered in two patches to each mouse; one patch per ear. Serum IgG responses were assessed 8 weeks after the priming immunization (light grey bars). All mice were boosted on day 54 with 1 × 10⁸ ifu AdHu5–PfRH5 by the IM route and post-boost IgG responses were assessed 2 weeks after this boost (dark grey bars). (A) IgG responses to the encoded antigen PfRH5. * p < 0.05, ** p < 0.01, *** p < 0.001 by one-way ANOVA, Kruskal–Wallis test with Dunn’s multiple comparison. Light grey lines and stars indicate significant differences across post-prime samples, dark grey lines indicate significant differences across post-boost samples. (B) IgG responses to the AdHu5 vector. * p < 0.05, ** p < 0.01, *** p < 0.001 compared to control, IM 1 × 10⁸ ifu. (C) The ratio of anti-PfRH5 IgG titre to anti-vector IgG in each animal. * p < 0.05, ** p < 0.01, *** p < 0.001 compared to control, IM 1 × 10⁸ ifu. $ p < 0.05, $$ p < 0.01 of post-prime compared to post-boost within the same group. (D) Neutralizing antibody responses to AdHu5 vector 2 weeks after boosting. “IM lo” and “DMN lo” refers to priming mice with a low dose of 1–5 × 10⁴ ifu by IM injection or DMN patch, whereas “IM hi” and “DMN hi” refer to a high dose of 1–5 × 10⁷ ifu by IM or DMN patch-based administration in the prime * p < 0.05, compared to control, IM 1 × 10⁸ ifu. Bars represent median, error bars represent interquartile range, dots represent individual samples.

Figure 3

Antibody responses induced using a homologous dose and route in the prime and the boost. Female Balb/c mice (n = 5 per group) were immunized on day 0 with AdHu5–PfRh5 by the IM route or using DMN patches. The conventional priming and boosting by the IM route, using 1 × 10⁸ ifu acting as a control. Alternatively, mice were administered a low dose of 5 × 10⁴ ifu by IM injection or DMN patch (“IM lo” or “DMN lo”) or a high dose of 5 × 10⁷ ifu by DMN patch (“DMN hi”). Serum IgG responses were assessed 8 weeks after the priming immunization. All mice were boosted on day 54 with a low or high dose of vaccine by the indicated route, and post-boost IgG responses were assessed 2 weeks after this boost. (A) IgG responses to the encoded antigen PfRH5. (B) IgG responses to the AdHu5 vector. (C) The ratio of the IgG titre to PfRH5 to the vector in each animal. * p < 0.05, ** p < 0.01, *** p < 0.001 by one-way ANOVA, Kruskal–Wallis test, with Dunn’s multiple comparison test compared to IM 10^8 in all graphs. Bars represent median, error bars represent interquartile range, circles represent individual samples.

Figure 4

The dose and route of immunization impacts on functional antibody activity against malaria parasites. Female Balb/c mice (n = 10 per group) were immunized on day 0 with AdHu5–PfRH5 by the IM route or using DMN patches. The conventional priming and boosting by the IM route, used 1 × 10⁸ ifu and acted as a control regime. Alternatively, mice were administered a low dose of 5 × 10⁴ ifu by IM injection or DMN patch (“IM lo” or “DMN lo”) or a high dose of 5 × 10⁷ ifu by DMN patch (“DMN hi”). All mice were boosted on day 54 with (A) the control dose of 1 × 10⁸ ifu by the IM route, or (B) with a low or high dose of vaccine by the indicated route. The level of growth inhibitory antibodies to P. falciparum 3D7 clone blood-stage parasites was assessed in purified IgG from serum taken 4 weeks post-boost and pooled into 2 to 6 samples, depending on volumes of available serum. * p < 0.05, by one-way ANOVA. Bars represent median, error bars represent interquartile range, circles represent individual samples.

Figure 5

IgG subtypes induced by repeated immunization with low or high dose of vaccine by IM or skin route. Female Balb/c mice (n = 7 per group) were immunized on day 0 with AdHu5–PfRH5 by the IM route or using DMN patches. Mice were administered the standard 1 × 10⁸ ifu by IM injection twice (“IM 10^8/IM 10^8”). Alternatively, mice were administered a low dose of 5 × 10⁴ ifu by IM injection or DMN patch (“IM lo” or “DMN lo”) or a high dose of 5 × 10⁷ ifu by DMN patch (“DMN hi”). All mice were boosted on day 54 with a low or high dose or 1 × 10⁸ ifu of vaccine by the indicated route. (A) The endpoint titre of RH5-specific IgG1 (dark grey bars) and IgG2a (light grey bars) in serum taken at day 84 (4 weeks post-boost). (B) The ratio of IgG1 to IgG2a in each animal. Bars represent median, lines represent interquartile range, triangles and circles represent individual samples. ** p < 0.01; *** p < 0.0005, **** p < 0.0001 by Kruskal–Wallis one-way ANOVA comparing all groups to IM 10^8/IM 10^8.

Figure 6

Effect of vaccine dose and route on the kinetics of IgG antibodies to PfRH5 antigen and to the adenovirus vector. Female Balb/c mice were immunized on day 0 with AdHu5–PfRH5 by the IM route or using DMN patches. Mice were administered the standard 1 × 10⁸ ifu by IM injection or with a DMN patch (“IM hi” or “DMN hi”). Alternatively, mice were administered a low dose of 5 × 10⁴ ifu by IM injection or DMN patch (“IM lo” or “DMN lo”). IgG responses to (A) the encoded antigen PfRH5 and (B) to the AdHu5 vector over time (n = 10 on day 17, n = 7 on days 35, 56, n = 4 on d119). Symbols represent median, bars represent interquartile range. For IgG responses to PfRH5, no significant differences in titres were observed over time for IM-lo or DMN-hi; for IM-hi; * p < 0.05 for day 17 compared to day 35 and day 119; * p < 0.05 for day 56 compared to day 119; for DMN-lo, * p < 0.05; day 17 compared to day 56; * p < 0.05 day 35 compared to day 119, ** p < 0.01; day 17 compared to day 119, by one-way ANOVA, Kruskal–Wallis test with Dunn’s multiple comparison. For IgG responses to AdHu5, in IM-hi; * p < 0.05 for day 17 compared to day 56 and day 119; in DMN-hi ** p < 0.01 for day 17 compared to day 56 and d119; in IM-lo; *** p < 0.001 compared to all other timepoints; no significant differences in titre were observed over time for DMN-lo, by one-way ANOVA, Kruskal–Wallis test with Dunn’s multiple comparison.