DNA priming and gp120 boosting induces HIV-specific antibodies in a randomized clinical trial

Abstract

BACKGROUNDRV144 is the only preventive HIV vaccine regimen demonstrating efficacy in humans. Attempting to build upon RV144 immune responses, we conducted a phase 1, multicenter, randomized, double-blind trial to assess the safety and immunogenicity of regimens substituting the DNA-HIV-PT123 (DNA) vaccine for ALVAC-HIV in different sequences or combinations with AIDSVAX B/E (protein).METHODSOne hundred and four HIV-uninfected participants were randomized to 4 treatment groups (T1, T2, T3, and T4) and received intramuscular injections at 0, 1, 3, and 6 months (M): T1 received protein at M0 and M1 and DNA at M3 and M6; T2 received DNA at M0 and M1 and protein at M3 and M6; T3 received DNA at M0, M1, M3, and M6 with protein coadministered at M3 and M6; and T4 received protein and DNA coadministered at each vaccination visit.RESULTSAll regimens were well tolerated. Antibodies binding to gp120 and V1V2 scaffold were observed in 95%-100% of participants in T3 and T4, two weeks after final vaccination at high magnitude. While IgG3 responses were highest in T3, a lower IgA/IgG ratio was observed in T4. Binding antibodies persisted at 12 months in 35%-100% of participants. Antibody-dependent cell-mediated cytotoxicity and tier 1 neutralizing-antibody responses had higher response rates for T3 and T4, respectively. CD4+ T cell responses were detectable in all treatment groups (32%-64%) without appreciable CD8+ T cell responses.CONCLUSIONThe DNA/protein combination regimens induced high-magnitude and long-lasting HIV V1V2-binding antibody responses, and early coadministration of the 2 vaccines led to a more rapid induction of these potentially protective responses.TRIAL REGISTRATIONClinicalTrials.gov NCT02207920.FUNDINGNational Institute of Allergy and Infectious Diseases (NIAID) grants UM1 AI068614, UM1 AI068635, UM1 AI068618, UM1 AI069511, UM1 AI069470, UM1 AI069534, P30 AI450008, UM1 AI069439, UM1 AI069481, and UM1 AI069496; the National Center for Advancing Translational Sciences, NIH (grant UL1TR001873); and the Bill & Melinda Gates Foundation (grant OPP52845).

Keywords: AIDS vaccine; AIDS/HIV; Vaccines.

Conflict of interest statement

Conflict of interest: The authors have declared that no conflict of interest exists.

Figures

Figure 1. HVTN 105 CONSORT statement flow diagram.

Figure 2. IgG binding-antibody responses in HVTN 105 participants over time, as measured by binding-antibody multiplex assay (BAMA) against aggregate vaccine-matched gp120 antigens, 1 gp140 antigen, and aggregate V1V2 antigens.

Shown are the positive-response rates and 95% CIs estimated using the score test method (top panels) and the geometric mean response magnitudes among all participants and 95% CI based on an assumption of log(IgG) following a normal distribution (bottom panels) by time point and treatment group (n = 25, 26, 26, 25 in T1–T4, respectively). The lines connect the response rates/geometric mean magnitudes between time points. Vaccine-matched gp120 antigens: A244.AE, MN.B, and ZM96.C. gp140 antigen: Con S gp140 CFI. V1V2 antigens: 1086.C V1V2, CaseA2_gp70_V1V2.B, CaseA2_V1/V2/169K.B, and A244.AE V1V2. Arrows indicate the second, third, and fourth vaccinations. D, DNA; A, AIDSVAX B/E.

Figure 3. IgG binding-antibody responses 2 weeks and 6 months after the fourth vaccination in HVTN 105, as measured by binding-antibody multiplex assay (BAMA) against 3 V1V2 antigens.

Shown are the positive-response rates (top panels) and the distribution of the response magnitudes (positive responders in red circles and nonresponders in blue triangles) and the box-and-whisker plots among the positive responders (the midline of the box-and-whisker plot denotes the median and the ends of the box-and-whisker plot denote the 25th and 75th percentiles) (bottom panels) by time point and treatment group (n = 25, 26, 26, 25 in T1-T4, respectively). V1V2 antigens: A244.AE V1V2, 1086.C V1V2, and CaseA2_gp70_V1V2.B. Bars on the top of plots indicate the significant differences between treatment groups within the same visit (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001) without multiple-comparisons adjustment. The comparisons between treatment groups were performed using Fisher’s exact test for response rates and Wilcoxon’s rank-sum test for magnitudes. Fractions above bars on top panels indicate numbers of positive responses over total numbers of responses (negative and positive) by time point and treatment group. D, DNA; A, AIDSVAX B/E.

Figure 4. IgG3 and IgG4 binding-antibody responses in HVTN 105 participants over time, as measured by binding-antibody multiplex assay (BAMA) against 2 V1V2 antigens.

Shown are the positive-response rates and 95% CI for IgG3 (A) and IgG4 (B) estimated using the score test method (top panels) and the geometric mean response magnitudes among all participants and 95% CI based on an assumption of log(IgG) following a normal distribution (bottom panels) by time point and treatment group (n = 25, 26, 26, 25 in T1–T4, respectively). The lines connect the response rates and geometric mean magnitudes between time points. Subtype AE V1V2: A244.AE V1V2; subtype C V1V2: 1086.C V1V2. Arrows indicate the second, third, and fourth vaccinations. D, DNA; A, AIDSVAX B/E.

Figure 5. IgA binding-antibody responses in HVTN 105 participants over time and IgA/IgG ratio 2 weeks after the fourth vaccination, as measured by binding-antibody multiplex assay (BAMA) against consensus A gp140 and A244.AE antigens.

(A) Shown are the positive-response rates and 95% CI estimated using the score test method (top panels) and the geometric mean response magnitudes among all participants and 95% CI based on an assumption of log(IgA) following a normal distribution (bottom panel) by time point and treatment group (n = 25, 26, 26, 25 in T1–T4, respectively). Arrows indicate the second, third, and fourth vaccinations. (B) The distribution and box-and-whisker plot of the IgA/IgG ratio in T3 and T4 (the midline of the box-and-whisker plot denotes the median and the ends of the box-and-whisker plot denote the 25th and 75th percentiles). Shown at the top of plots are the percentages of positive responders to IgA and IgG, respectively, in T3 and T4; and the P value is the testing difference in IgA/IgG ratio between T3 and T4 from Wilcoxon’s rank-sum test. D, DNA; A, AIDSVAX B/E.

Figure 6. Antibody-dependent cell-mediated cytotoxicity (ADCC) at 2 weeks and 6 months after the fourth vaccination in HVTN 105.

Shown are the response rates (top panels) and the distribution of AUC of granzyme B (GzB) activity (positive responders to peak granzyme B activity in red circles and nonresponders in blue triangles) and the box-and-whisker plots among the positive responders to peak granzyme B activity (the midline of the box-and-whisker plot denotes the median and the ends of the box-and-whisker plot denote the 25th and 75th percentiles) (bottom panels) by time point and treatment group (n = 25, 26, 26, 25 in T1–T4, respectively). Subtype C gp120: ZM96.C; subtype AE gp120: A244.AE; and subtype B gp120: MN.B. Bars and asterisks presented on the top of plots indicate the significant differences between treatment groups within the same visit (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001) without multiple-comparisons adjustment. The comparisons between treatment groups were performed using Fisher’s exact test for response rates and Wilcoxon’s rank-sum test for magnitudes. Fractions above bars on the top panels indicate numbers of positive responses over the total numbers of responses (positive and negative) by time point and treatment group. D, DNA; A, AIDSVAX B/E.

Figure 7. Neutralizing-antibody responses in HVTN 105.

(A) Tier 1 Env-pseudotyped viruses (BaL.26.B, MN.3.B, MW965.26.C, SF162.LS.B, and TH023.6.AE) were tested in the TZM-bl neutralization assay 2 weeks after the fourth vaccination (peak time point). Bar plots show positive-response rates by treatment group on top panels. The bottom panels show the distribution of response titer (positive responses in filled red circles and negative responses in open blue triangles) and the box-and-whisker plots of response titer among positive responders (the midline of the box-and-whisker plot denotes the median and the ends of the box-and-whisker plot denote the 25th and 75th percentiles) by treatment group (n = 25, 26, 26, 25 in T1-T4, respectively). (B) Neutralizing-antibody magnitude breadth (AUC-MB) curves for TZM-bl based on all 5 isolates. (C) Neutralizing-antibody responses to MN.3.B and TH023.6.AE at 6 months after the fourth vaccination. Bars and asterisks presented on the top of plots in A and C indicate the significant differences between treatment groups within the same visit (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001) without multiple-comparisons adjustment. The comparisons between treatment groups were performed using Fisher’s exact test for response rates and Wilcoxon’s rank-sum test for magnitudes. The comparisons of AUC-MB at peak time point in B between treatment groups from Wilcoxon’s rank-sum test show that AUC-MB in T4 is significantly higher than T1–T3, with P < 0.001, P < 0.001, and P < 0.003, respectively, and AUC-MB in T1 is significantly lower than T2–T3, with P < 0.001. Fractions above bars in A and C indicate the numbers of positive responders over the total number of responses (positive and negative). D, DNA; A, AIDSVAX B/E.

Figure 8. CD4+ T cell responses elicited in HVTN 105, as measured by intracellular cytokine staining (ICS), and reported as the percentage of cells producing IFN-γ and/or IL-2 in each treatment group.

(A) CD4+ T cell responses to any HIV Env peptide pools (Any Env), all vaccine-matched: ZM96 gp140-Env1, ZM96 gp140-Env2, and 92TH023-Env. (B) CD4+ T cell responses to HIV Gag peptide pool (Any Gag): ZM96 Gag. Bar plots on top panels show the positive-response rates by time point and treatment group (n = 25, 26, 25, 25 in T1–T4, respectively). The bottom panels show the distribution of response magnitudes (positive responses in filled red circles and negative responses in open blue triangles) and the box-and-whisker plots of magnitudes among the positive responders (the midline of the box-and-whisker plot denotes the median and the ends of the box-and-whisker plot denote the 25th and 75th percentiles). Fractions above bars on top panels indicate the numbers of positive responders over the total numbers of responses (positive and negative). Bars and asterisks on top of plots indicate the significant differences between treatment groups (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001) without multiple-comparisons adjustment. The comparisons between treatment groups were performed using Fisher’s exact test for response rates and Wilcoxon’s rank-sum test for magnitudes. D, DNA; A, AIDSVAX B/E.

Figure 9. COMPASS CD4+ T cell polyfunctionality (PF) scores and mean probability of response heatmaps.

The distribution and the box-and-whisker plot of PF scores for all vaccine-matched HIV Env peptide pools: ZM96 gp140-Env1, ZM96 gp140-Env2, and 92TH023-Env (A) and for ZM96 Gag peptide pool (C) by time point and treatment group (n = 25, 26, 25, 25 in T1–T4, respectively). The midline of the box-and-whisker plot denotes the median and the ends of the box-and-whisker plot denote the 25th and 75th percentiles. Bars and asterisks on top of box-and-whisker plots indicate the significant differences between treatment groups using Wilcoxon’s rank-sum test (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001) without multiple-comparisons adjustment. Heatmaps for CD4+ T cell response to any Env (B) and ZM96 Gag (D) show the mean posterior probabilities of antigen-specific responses from COMPASS. Columns correspond to the different subsets of cytokines being considered and rows correspond to mean across the individual participants in each treatment group at each time point. Each cell shows the probability that the corresponding antigen-specific subset (column) is being expressed in the corresponding treatment group in average (row), and is color coded ranging from white (zero) to dark purple (one).

Figure 10. Radar plots.

Maximum response rates of each assay readout at 2 weeks (A) and 6 months (B) after the fourth vaccination. Antigens included in each assay readout are as follows: For IgG binding-antibody responses to Env, antigens include A244.AE, MN.B, ZM96.C, Con S gp140 CFI, as well as Con 6 gp120. For IgG binding-antibody responses to V1V2, antigens include 1086.C V1V2, CaseA2_gp70_V1V2.B, CaseA2_V1/V2/169K.B, A244.AE V1V2, as well as ZM96.C V1V2. For IgA binding-antibody responses, antigens include consensus A gp140 and A244.AE. For neutralizing-antibody (nAb) responses, antigens include BaL.26.B, MN.3.B, MW965.26.C, SF162.LS.B, and TH023.6.AE. For antibody-dependent cell-mediated cytotoxicity (ADCC) responses, antigens include ZM96.C, A244.AE, and MN.B. For CD4+ T cell intracellular cytokine staining, Env antigens are ZM96 or 92TH023 peptide pools (Any Env), and the Gag antigen is ZM96 peptide pool (Any Gag). D, DNA; A, AIDSVAX B/E.