Late boosting of the RV144 regimen with AIDSVAX B/E and ALVAC-HIV in HIV-uninfected Thai volunteers: a double-blind, randomised controlled trial

Abstract

Background: The RV144 phase 3 vaccine trial in Thailand demonstrated that ALVAC-HIV (vCP1521) and AIDSVAX B/E administration over 6 months resulted in a 31% efficacy in preventing HIV acquisition. In this trial, we assessed the immunological effect of an additional vaccine boost to the RV144 regimen at varying intervals between the priming vaccine series and the boost.

Methods: RV306 is a double-blind, placebo-controlled, randomised clinical trial done at three clinical sites in Thailand. Eligible volunteers were HIV-uninfected individuals aged 20-40 years who were at low risk for HIV infection and in good health. A randomisation schedule was centrally generated with fixed sized strata for Research Institute for Health Sciences Chiang Mai and combined Bangkok clinics. Participants were randomly assigned to one of five groups and then further randomly assigned to either vaccine or placebo. All participants received the primary RV144 vaccine series at months 0, 1, 3, and 6. Group 1 received no additional boost, group 2 received additional AIDSVAX B/E and ALVAC-HIV (vCP1521) or placebo at month 12, group 3 received AIDSVAX B/E alone or placebo at month 12, group 4a received AIDSVAX B/E and ALVAC-HIV or placebo at month 15, and group 4b received AIDSVAX B/E and ALVAC-HIV or placebo at month 18. Primary outcomes were safety and tolerability of these vaccination regimens and cellular and humoral immune responses compared between the RV144 series alone and regimens with late boosts at different timepoints. Safety and tolerability outcomes were assessed by evaluating local and systemic reactogenicity and adverse events in all participants. This trial is registered at ClinicalTrials.gov (NCT01931358); clinical follow-up is now complete.

Findings: Between Oct 28, 2013, and April 29, 2014, 367 participants were enrolled, of whom 27 were assigned active vaccination in group 1, 102 in group 2, 101 in group 3, 52 in group 4a, 51 in group 4b, and 34 combined placebo across all the groups. No vaccine-related serious adverse events were recorded. Occurrence and severity of local and systemic reactogenicity were similar across active groups. Groups with late boosts (groups 2, 3, 4a, and 4b) had increased peak plasma IgG-binding antibody levels against gp70 V1V2 relative to group 1 vaccine recipients with no late boost (gp70 V1V2 92TH023 adjusted p<0·02 for each; gp70 V1V2 CaseA2 adjusted p<0·0001 for each). Boosting at month 12 (groups 2 and 3) did not increase gp120 responses compared with the peak responses after the RV144 priming regimen at month 6; however, boosting at month 15 (group 4a) improved responses to gp120 A244gD- D11 (p=0·0003), and boosting at month 18 (group 4b) improved responses to both gp120 A244gD- D11 (p<0·0001) and gp120 MNgD- D11 (p=0·0016). Plasma IgG responses were significantly lower among vaccine recipients boosted at month 12 (pooled groups 2 + 3) than at month 15 (group 4a; adjusted p<0·0001 for each, except for gp70 V1V2 CaseA2, p=0·0142) and at month 18 (group 4b; all adjusted p<0·001). Boosting at month 18 versus month 15 resulted in a significantly higher plasma IgG response to gp120 antigens (all adjusted p<0·01) but not gp70 V1V2 antigens. CD4 functionality and polyfunctionality scores after stimulation with HIV-1 Env peptides (92TH023) increased with delayed boosting. Groups with late boosts had increased functionality and polyfunctionality scores relative to vaccine recipients with no late boost (all adjusted p<0·05, except for the polyfunctionality score in group 1 vs group 4b, p<0·01).

Interpretation: Taken together, these results suggest that additional boosting of the RV144 regimen with longer intervals between the primary vaccination series and late boost improved immune responses and might improve the efficacy of preventing HIV acquisition.

Funding: US National Institute of Allergy and Infectious Diseases and US Department of the Army.

Conflict of interest statement

DECLARATION OF INTERESTS

J. T., and C. D. are employees of Sanofi Pasteur. C.D. is a shareholder at Sanofi Pasteur. S.P. was an employee of Sanofi Pasteur during the conduct of this study and analysis. F. S. is an employee of Global Solutions for Infectious Diseases. P.D. and Y.Z. are employees of Emmes, and receive other funding support for statistical analyses from the Henry M. Jackson Foundation. J.K. reports other from US Department of Defense, during the conduct of the study, non-financial support from GSK, and personal fees from Takeda, outside the submitted work. N.L.M. reports grants from US Army and NIAID/NIH, during the conduct of the study. M.R. reports grants from the Henry M. Jackson Foundation, during the conduct of the study. All other authors report no potential conflicts.

Copyright © 2020 Elsevier Ltd. All rights reserved.

Figures

Figure 1:. RV306 study design

Each RV306 participant received ALVAC-HIV and AIDSVAX® B/E either alone or in combination (abbreviated Combo), at the indicated time points. Participants were randomized to 1 of 5 groups and further randomized within each group to receive either vaccine product or placebo injections at the ratio indicated for each group displayed on the right. Participants were followed for 24 months in total.

Figure 2:. Trial profile

Participant screening and enrollment by study group.

Figure 3:. Plasma IgG HIV-1 Env and V1V2 binding antibody levels increase with late boosting

IgG binding antibody responses to HIV-1 gp120 and scaffolded gp70 V1V2 antigens. Reciprocal titers against gp120 A244gD- D11 (upper left), gp120 MNgD- D11 (upper right), gp70 V1V2 (92TH023) (lower left), and gp70 V1V2 (case A2) (lower right) are shown. Vaccination timepoints and the last visit are shown on the x axis. All peak immunogenicity measurements were performed two weeks post vaccination. Each panel graphically displays geometric mean titers, color-coded by group as per the legend. Error bars depict 95% confidence intervals. Volunteers completing all vaccinations are depicted. The responses against HIV-1 gp70 V1V2 in Group I were significantly lower than each of the late boosting groups (adjusted p

Figure 4:. Plasma IgA HIV-1 Env and V1V2 binding antibody levels do not increase with late boosting

IgA binding antibody responses against HIV-1 gp120 and scaffolded variable regions 1 and 2 (V1V2) antigens. Reciprocal titers against gp120 A244gD- D11 (upper left), gp120 MNgD- D11 (upper right), gp70 V1V2 (92TH023) (lower left), and gp70 V1V2 (case A2) (lower right) are shown. Vaccination timepoints and the last visit are shown on the x axis. All peak immunogenicity measurements were performed two weeks post vaccination. Each panel graphically depicts geometric mean titers, color coded by group as per the legend. Error bars depict 95% confidence intervals. Volunteers completing all vaccinations are depicted, at vaccination time points as indicated above the x-axis.

Figure 5:. TZM-bl neutralizing antibody levels against subtype B MN.3, subtype AE TH023.6, and subtype C MW965.26 increase with late boosting

ID50 against MN.3 (upper left), TH023.6 (upper right), MW965.26 (lower left), and SF162.LS (lower right) are shown. Vaccination timepoints and the last visit are shown on the x axis. All peak immunogenicity measurements were performed two weeks post vaccination. Each panel graphically depicts ID50, color-coded by group as per the legend. Error bars depict 95% confidence intervals. Volunteers completing all vaccinations are depicted. The neutralizing antibody levels against TH023.6 and MW965.26 in Group I were significantly lower than each of the late boosting groups (adjusted p