Safety and broad immunogenicity of HIVconsvX conserved mosaic candidate T-cell vaccines vectored by ChAdOx1 and MVA in HIV-CORE 006: a double-blind, randomised, placebo-controlled phase 1 trial in healthy adults living without HIV-1 in eastern and southern Africa

Abstract

Background: Even within the context of antiretroviral treatment and prevention, an HIV-1 vaccine remains the best strategy for ending the HIV/AIDS epidemic. A vaccine is particularly needed in sub-Saharan Africa, where HIV-1 greatly affects people's lives and economy. Here, we aimed to assess the safety and immunogenicity of candidate T-cell vaccines in African populations.

Methods: HIV-CORE 006 was a double-blind, randomised, placebo-controlled phase 1 trial conducted across four clinical research centres in Uganda, Kenya, and Zambia. Eligible participants were not pregnant, were living without HIV-1 or HIV-2, had a low likelihood of acquiring HIV-1, were aged 18-50 years, fully comprehended the purpose and details of this study as outlined in the participant information sheet, and passed an assessment of understanding before providing written informed consent. Participants were randomly assigned (9:2) to receive either a vaccine regimen or a placebo. The vaccine was administered as ChAdOx1.tHIVconsv1 (C1) followed by MVA.tHIVconsv3 (M3) and MVA.tHIVconsv4 (M4) in regimen C1-M3M4. The first primary outcome was the vaccines' safety assessment, assessed in all participants who received at least one vaccine or placebo dose. The second primary outcome evaluated the C1-M3M4 regimen's induction of HIVconsvX-specific T-cell responses by assessing the proportion of vaccine recipients who responded to the vaccination, assessed in all participants who received all doses of vaccine or placebo as per protocol. This study is registered with ClinicalTrials.gov, NCT04553016, and the Pan-African Clinical Trials Registry PACTR202006495409011, and is now closed.

Findings: Between July 15, 2021, and Nov 2, 2022, 89 healthy adults living without HIV-1 were randomly assigned, with 88 receiving either the vaccine (n=72) or placebo (n=16). Of these 88 participants, 57 (65%) were male and 31 (35%) were female. The C1, M3, and M4 vaccine components were well tolerated and induced HIVconsvX-specific responses in 70 (99%) of the 71 participants who completed all vaccine doses. Vaccine-elicited T cells peaked at a median of 2310 (IQR 1080-4480) IFN-γ spot-forming units per 10⁶ peripheral blood mononuclear cells and recognised a median of eight (five to ten) of ten peptide pools spanning the HIVconsvX immunogen. The total frequencies of elicited T cells decreased 4·6 times over a 40-week follow-up period compared with the peak responses. Upon antigenic re-exposure, T cells proliferated, exhibited multiple effector functions, and inhibited HIV-1 representatives from clades A, B, C, and D.

Interpretation: Results from key sub-Saharan African populations supported the safety of the vaccine regimen previously shown in the first-in-human trial in the UK. The induction of T cells and their characteristics encourage vaccine integration into HIV-1 cure strategies, which could inform HIV-1 prevention efforts.

Funding: The European and Developing Countries Clinical Trials Partnership.

Figure 1

Vaccine design and trial schema Alignments of curated global HIV-1 protein sequences available from the Los Alamos National Laboratory HIV Sequence Database as of September, 2013, were used to compute two mutually complementary mosaic sequences of the complete Gag and Pol proteins and to identify highly conserved regions within them. These regions were arranged into six distint proteins collectively referred to as HIVconsvX and the individual genes were inserted into simian adenovirus or poxvirus vectors to generate vaccines C1, C62 (not tested in HIV-CORE 006 due to delayed manufacturing), M3, and M4. In the HIV-CORE 006 trial, people living without HIV-1 in sub-Saharan Africa received either the vaccine regimen (C1-M3M4; n=72) or placebo (n=16) and were followed up for 40 weeks following the last vaccination. C1=ChAdOx1.tHIVconsv1. C1/P=C1 or placebo. C62=ChAdOx1.HIVconsv62. M3=MVA.tHIVconsv3. M4=MVA.tHIVconsv4. M3M4/P=M3M4 or placebo.

Figure 2

Study population and the CONSORT diagram (A) Age distribution of the study individuals. (B) CONSORT diagram. C1=ChAdOx1.tHIVconsv1. CFHRZ=the Center for Family Health Research Zambia. KAVI-ICR=the KAVI-Institute for Clinical Research. KWTRP=the KEMRI-Wellcome Trust Research Programme. M3M4=MVA.tHIVconsv3 and MVA.tHIVconsv4. MUL=the Medical Research Council/Uganda Virus Research Institute and London School of Hygiene and Tropical Medicine Uganda Research Unit.

Figure 3

Strong and broad T-cell responses induced by the HIVconsvX vaccines in individuals without HIV-1 (A) The HIVconsvX vaccine-specific T-cell responses were enumerated in an IFN-γ ELISPOT assay using HIVconsvX peptide pools one to ten on day 0, day 35 (peak), and day 308. The p values reported were calculated using the Wilcoxon matched-pairs signed rank test. (B) A summary of the total peak magnitudes and the number of peptide pools recognised out of 10 at the peak of T-cell responses on day 35 is provided. Medians with individual specific T-cell frequencies on day 35 in female and male vaccine recipients (Mann–Whitney test; C) and for individual HIVconsvX pools (D) are presented. (E) Functionality of HIVconsvX-specific CD8⁺ and CD4⁺ T cells. Five strong responders from each CRC were selected for the polychromatic flow cytometry analysis following stimulation of day 35 peripheral blood mononuclear cells with personalised peptide pools. Average frequencies are depicted as central pie charts illustrating the number of functions of individual cells depicted in shades of grey with functions colour-coded around the periphery (left). Total frequencies for each function are displayed as median (range) boxes indicating individual values (right). For gating, refer to the appendix (p 12). (F) T-cell memory structure. The memory structure of vaccine-elicited T cells was analysed by flow cytometry for their subpopulations defined as T_EM (effector memory; CD45RA^NegCCR7^LoCD27^Neg), T_TM (transitional memory; CD45RA^NegCCR7^LoCD27^Pos), T_CM (central memory; CD45RA^NegCCR7^HiCD27^Pos), T_TEMRA (terminal effector memory re-expressing CD45RA^PosCCR7^HiCD27^Pos), T_TD (terminally differentiated; CD45RA^PosCCR7^LoCD27^Neg), CD45RA^PosCCR7^HiCD27^Pos), and T_EFF (effector; CD45RA^PosCCR7^LoCD27^Pos). See appendix for gating (p 12). Graphs show averaged percentages over five strong responders from each CRC. CFHRZ=the Center for Family Health Research Zambia. CRC=clinical research centre. KAVI-ICR=the KAVI-Institute for Clinical Research. KWTRP=the KEMRI-Wellcome Trust Research Programme. MUL=the Medical Research Council/Uganda Virus Research Institute and London School of Hygiene and Tropical Medicine Uganda Research Unit. P=peptide pool. PMBC=peripheral blood mononuclear cell. SFU=spot-forming unit.

Figure 3

Strong and broad T-cell responses induced by the HIVconsvX vaccines in individuals without HIV-1 (A) The HIVconsvX vaccine-specific T-cell responses were enumerated in an IFN-γ ELISPOT assay using HIVconsvX peptide pools one to ten on day 0, day 35 (peak), and day 308. The p values reported were calculated using the Wilcoxon matched-pairs signed rank test. (B) A summary of the total peak magnitudes and the number of peptide pools recognised out of 10 at the peak of T-cell responses on day 35 is provided. Medians with individual specific T-cell frequencies on day 35 in female and male vaccine recipients (Mann–Whitney test; C) and for individual HIVconsvX pools (D) are presented. (E) Functionality of HIVconsvX-specific CD8⁺ and CD4⁺ T cells. Five strong responders from each CRC were selected for the polychromatic flow cytometry analysis following stimulation of day 35 peripheral blood mononuclear cells with personalised peptide pools. Average frequencies are depicted as central pie charts illustrating the number of functions of individual cells depicted in shades of grey with functions colour-coded around the periphery (left). Total frequencies for each function are displayed as median (range) boxes indicating individual values (right). For gating, refer to the appendix (p 12). (F) T-cell memory structure. The memory structure of vaccine-elicited T cells was analysed by flow cytometry for their subpopulations defined as T_EM (effector memory; CD45RA^NegCCR7^LoCD27^Neg), T_TM (transitional memory; CD45RA^NegCCR7^LoCD27^Pos), T_CM (central memory; CD45RA^NegCCR7^HiCD27^Pos), T_TEMRA (terminal effector memory re-expressing CD45RA^PosCCR7^HiCD27^Pos), T_TD (terminally differentiated; CD45RA^PosCCR7^LoCD27^Neg), CD45RA^PosCCR7^HiCD27^Pos), and T_EFF (effector; CD45RA^PosCCR7^LoCD27^Pos). See appendix for gating (p 12). Graphs show averaged percentages over five strong responders from each CRC. CFHRZ=the Center for Family Health Research Zambia. CRC=clinical research centre. KAVI-ICR=the KAVI-Institute for Clinical Research. KWTRP=the KEMRI-Wellcome Trust Research Programme. MUL=the Medical Research Council/Uganda Virus Research Institute and London School of Hygiene and Tropical Medicine Uganda Research Unit. P=peptide pool. PMBC=peripheral blood mononuclear cell. SFU=spot-forming unit.

Figure 4

Cross-clade inhibition of HIV-1 growth by vaccine-elicited CD8 T cells All data concerning the inhibition of eight HIV-1 IMCs growth inhibition by vaccine-elicited CD8⁺ effectors in autologous CD4⁺ cells are provided in the appendix (p 14). (A) The graphs show the median vaccine-elicited inhibition by CD8⁺ T cells over eight HIV-1 IMCs for each CRC. The overall inhibition across all CRCs and all viruses was assessed and compared at pre-vaccination (day 0), peak response (day 35), and study conclusion (day 308) using the Friedman test with Dunn’s multiple comparison-adjusted values. p values are indicated above the graphs. (B) A comparison of four CRCs regarding the overall inhibition of eight IMCs on day 35 is presented, showing the medians for each IMC inhibition. The Kruskal–Wallis multiple comparison test with Dunn’s correction found no statistically separable results except for a significant difference between MUL and CFHRZ. (C) The same data from day 35 are presented as medians for each IMC and CRC. Subsequently, vaccine recipients were categorised into three age groups (D) and by gender (E) to compare overall IMC inhibition (left) and the number of inhibited IMCs by ≥0·1 log₁₀ (right). Medians are shown alongside individual values with significant p values shown above, obtained through the Mann–Whitney test and the Kruskal–Wallis test with multiple comparison corrections, respectively. (F) The graph illustrates the correlation between peak ELISPOT frequencies and the growth inhibition caused by effector CD8⁺ cells elicited by the C1-M3M4 vaccine regimen applying simple linear regression. CFHRZ=the Center for Family Health Research Zambia. CRC=clinical research centre. IMC=infectious molecular clones. KAVI-ICR=the KAVI-Institute for Clinical Research. KWTRP=the KEMRI-Wellcome Trust Research Programme. MUL=the Medical Research Council/Uganda Virus Research Institute and London School of Hygiene and Tropical Medicine Uganda Research Unit. PBMC=peripheral blood mononuclear cell. SFU=spot-forming unit.