The trial proposal, the trial-specific information provided to volunteers, the consent form and substantial protocol amendments (if applicable) were reviewed by a recognized Research Ethics Committee and by the Medicines and Healthcare products Regulatory Authority.

The study was conducted in compliance with UK Clinical Trial Regulations and any amendments, which include compliance with the principles of good clinical practice, and the study abided by the principles of the Declaration of Helsinki.

All volunteers provided written informed consent to participate in the trial on the basis of appropriate information and with adequate time to consider the information and discuss the trial with the principal investigator or her delegate. Participants were asked to explain what the trial involved in their own words, to ensure the volunteer understood the intensity of the schedule and the issues associated with taking part in a trial of a candidate HIV vaccine. All volunteers were made aware that they were free to withdraw without obligation at any time and that such an action would not adversely affect any aspect of their medical care or legal rights.

A recombinant uncleaved clade C HIV-1 envelope gp140 protein (CN54gp140) produced by Polymun Scientific (Klosterneuburg, Austria) to GMP specification, which has previously been reported to be immunogenic in a number of preclinical and clinical studies (26, 27), was used for the X001 clinical trial. The vaccine Ag CN54gp140 was administered intramuscularly into the deltoid muscle of the upper arm at a dosage of 100 µg CN54gp140 formulated with 5 µg GLA-AF [Infectious Disease Research Institute, Seattle, WA, USA (28)] in a total volume of 0.4 ml. GLA has been shown to enhance antibody responses in mice, non-human primates, and humans without causing adverse reactions (29, 30). Good adjuvanticity was observed in a previous influenza clinical trial where they used 2.5 µg GLA and a clinical trial of a leishmaniasis vaccine using 2 or 5 µg GLA in an oil-in-water emulsion formulation (GLA-SE), which was safe and well tolerated (30, 31). Since much of the reactogenicity of GLA-SE is believed to be associated with its oil-in-water form, the aqueous formulation GLA-AF was assumed to be less reactogenic. Subsequent animal studies revealed that GLA-AF has broadly equivalent adjuvant activity to GLA-SE and based on these findings for the current trial a 5-µg dose of GLA was selected (32). Further, good tolerability of GLA-AF at a dose of 5 µg in combination with the CN54gp140 Ag has been demonstrated in previous clinical trials conducted by us (27, 33).

The trial was open labeled without placebo control, and we sought to enroll 12 healthy volunteers, male or female between the ages 18 and 45 who did not report high-risk behavior for HIV infection. They were randomly assigned to two groups of six each within a screening period of 42 days prior to the first vaccination. Trial participants were required to follow a rigorous visit schedule and depending on the study group underwent up to 17 blood draws (Tables S6 and S7 in Supplementary Material).

Study participants in each group received three injections (priming phase) followed by a booster IM at 6 months (Group A) or 12 months (Group B) of the study vaccine CN54gp140 formulated with GLA-AF. The study schema is detailed in Table 1. Blood samples and mucosal secretions (cervicovaginal, urethral, rectal, and seminal) were collected at designated time points (Tables S6 and S7 in Supplementary Material) for immunogenicity and vaccine safety analysis.

The primary endpoint was to evaluate the safety and tolerability of two vaccine regimens consisting of three initial injections of CN54gp140 (priming phase) at a dosage of 100 µg formulated with 5 µg GLA-AF and a final booster injection at 6 months (Group A) or 12 months (Group B). The secondary outcome was to assess B and T cell-mediated and antibody binding responses in the peripheral blood.

Collected data were identified only by a volunteer identification number. Safety and tolerability were addressed by examining the overall rates of solicited and unsolicited adverse events and serious adverse events that were potentially associated with vaccination and the number of volunteers who experienced these events. All clinical and routine laboratory work was included in the safety analysis (Tables S1 and S2 and Figures S1 and S2 in Supplementary Material).

Study staff performed HIV testing and any associated pre- or posttest counseling, according to national guidelines at time points defined in the Schedules of Procedures (Tables S6 and S7 in Supplementary Material).

Cervicovaginal secretions were collected from female volunteers using the INSTEAD™ Softcup, a commercially available, self-inserted menstrual cup made of polyethylene. There is a risk of dislodging an intrauterine device (IUD) when removing the Softcup and women who have had toxic shock syndrome (TSS) should not use internal sanitary protection products. For these reasons, women with IUDs or a history of TSS were excluded from the study.

Urethral secretions were collected from male volunteers using a 2-mm diameter polyvinyl alcohol (PVA) sponge. Sponges could be self-inserted following explanation or were inserted by the clinical trials team at the NIHR/Wellcome Trust Imperial Clinical Research Facility at Hammersmith Hospital.

Rectal secretions were collected from male and female volunteers using a pre-moisturized (with 50 µl PBS) sponge.

Semen samples were either donated while within the ICRF or brought into the facility following donation at home (within 3 h of ejaculation). Participants were instructed to donate semen following 24 h of sexual abstinence.

Mucosal secretion collection occurred at the time points indicated in the Schedules of Procedures (Tables S6 and S7 in Supplementary Material), except at visits when female volunteers were menstruating, in which case cervicovaginal secretions were not collected.

Prior to processing the mucosal samples, the extraction buffer (EB) was prepared by adding 1 ml of 100× protease inhibitor cocktail set (Calbiochem, 539131) and 10% NaN₃ (Sigma-Aldrich, S2002-25G, UK) to 100 ml sterile PBS.

Prior to analysis, Softcups containing cervicovaginal secretions in 50 ml tubes were removed from −80°C, allowed to thaw on ice overnight, and centrifuged at 400 × g for 15 min at 4°C. An equal volume of EB was added to the secretion samples, mixed and aliquoted.

The 2-mm diameter PVA sponge used for the collection of urethral samples, and 500 µl EB was added to the top chamber of a Costar Spin-X tube (Corning) and centrifuged for 15 min at 13,000 rpm and 4°C. The eluate was collected and aliquoted.

Tubes containing the rectal swabs were removed from −80°C, thawed on wet ice, and placed into a 50 ml tube containing a sterile falcon filter unit (70 µm). To each tube, 500 µl EB was added before centrifugation at 400 × g for 10 min at 4°C. To remove debris, the eluate was transferred into the top chamber of a Spin-X tube, centrifuged at 13,000 rpm for min at 4°C, and aliquoted.

Seminal secretions were transferred into one or two sterile, plastic, 2 ml microcentrifuge tubes before centrifugation at 1,200 × g, for 10 min, at 4°C. The supernatant (seminal plasma) was transferred into 2–3 prelabeled, sterile, plastic, 2 ml microcentrifuge tubes. All mucosal sample aliquots were frozen at −80°C until analysis via ELISA.

Peripheral blood mononuclear cells were isolated using density gradient separation from heparinized whole blood, used fresh or frozen (within 8 h of blood collection) in a mixture of fetal bovine serum (Sigma-Aldrich, St. Louis, MO, USA) and DMSO at a 90:10 ratio using a Kryo 560-16 rate controlled freezer (Planer, Sunbury-On-Thames, UK). PBMCs were stored in vapor phase liquid nitrogen at the IAVI Human Immunology Laboratory or St Mary’s Campus, Imperial College London.

An extensive ELISA platform was developed and employed to measure both total and Ag-specific antibody responses in human plasma/serum samples and mucosal secretions. The protocols of the various Ig ELISAs are detailed in the Supplementary Material (Table S4 in Supplementary Material). In brief, Ag-specific gp140-binding antibodies were measured using a standardized ELISA platform. In plasma/serum samples Ag-specific IgA, IgM, IgG (total), and IgG subclasses were measured, while both total and Ag-specific IgA/G antibody responses were measured in the mucosal compartment (cervicovaginal, urethral, seminal, and rectal). A total of 96-well high binding MaxiSorp plates (Nunc) were coated with 50 μl/well recombinant CN54gp140 (Polymun Scientific) at 2.5 µg/ml in PBS for overnight at 4°C. As reference material, we used standard Igs (Table S4 in Supplementary Material), which were captured with a combination of antihuman kappa and lambda light chain specific mouse antibodies. These capture antibodies were coated onto the plates overnight at 4°C, and coated plates were washed four times with PBS-T before blocking with the appropriate assay buffer (Table S4 in Supplementary Material). Following further washing, diluted sera or mucosal samples were added to the Ag-coated wells and titrations of Ig standards were added to the kappa/lambda capture antibody coated wells at 50 μl/well and incubated for 1 h at 37°C. Plates were washed four times prior to the addition of secondary antibody and incubated for 1 h at 37°C. Plates were washed four times and developed with 50 μl/well of KPL SureBlue TMB substrate (Insight Biotechnology, UK). The detection of Ag-specific IgA/G in mucosal samples and the detection of Ag-specific IgA/M/G2 required a streptavidin-(poly)-HRP amplification step prior to TMB development (Table S4 in Supplementary Material). The reaction was stopped after 5 min by adding 50 μl/well of 1 M H₂SO₄, and the absorbance read at 450 nm on a KC4 spectrophotometer.

Cellular immunogenicity was assessed by IFN-γ T cell ELISpot using frozen PBMC. Prior to use, ELISpot plates, pre-coated with anti-IFN-γ antibody (Mabtech), were washed with PBS and blocked with RPMI supplemented with 10% heat-inactivated fetal calf serum (R10) for a minimum of 2 h at 37°C. PBMCs were plated out at 2 × 10⁵ cells/well in 50 µl. For CN54gp140-specific responses, two pools of 15mer peptides with 11 amino acid overlap (P1 and P2) matched to the HIV-1 sequence vaccine insert were plated in quadruplicate wells. Peptides were initially dissolved and pooled in DMSO before dilution in R10 with final assay DMSO concentration of 0.45% v/v. Eight negative no-peptide control wells were cultured in R10 supplemented with 0.45% DMSO. Positive controls in duplicate wells were cultured with 10 µg/ml phytohemagglutinin (Sigma-Aldrich, Dorset, UK) and a pool of CMVpp65-peptides. Each individual HIV and CMV peptide was at a final assay concentration of 1.5 µg/ml. PBMCs were thawed and rested overnight in RPMI 20% HIFCS prior to ELISpot assay setup. On the day of assay setup, viable PBMCs were counted using a Beckman Coulter Vi-cell XR counter and plated at 2 × 10⁵ cells/well in 50 µl and incubated for 16–24 h at 37°C, in 5% CO₂. Spots were visualized using biotinylated anti-IFN-γ (Mabtech) combined with ABC streptavidin-HRP (Vector labs), and the color was developed using AEC substrate (Sigma). The reaction was stopped after 4 min by rinsing with tap water.

Plates were air dried overnight and the spots counted using an AID ELISpot Reader and version 5.0 software (AID GmbH). After quality control of the data, the mean number of spot forming units (SFU) in no-peptide wells was subtracted from test wells, and the results were expressed as the median net SFU/10⁶ PBMC. Responses at least 38 SFU/10⁶ PBMCs above background and at least 4× background were scored as positive. These response cutoffs were derived from assessing the 97.5% percentile of the peptide pool responses in PBMC from 95 HIV seronegative individuals.

To characterize T follicular helper cell responses, study samples at 2 weeks after the third priming injection and after the respective booster injection along with matched baseline samples were interrogated for phenotype and cytokine secretion using cross-qualified polychromatic flow cytometry. Cryopreserved PBMCs were thawed and rested overnight and resuspended at 1 × 10⁶ in 200 µl RPMI medium supplemented with 10% FBS and co-incubated with HIV-peptide pools, 1 μg/ml staphylococcal enterotoxin B (Sigma-Aldrich, Poole, Dorset, UK) or mock stimuli, BD Golgistop (Becton Dickinson, San Jose, CA, USA) and Brefeldin A (Sigma-Aldrich, Poole Dorset, UK) for 6 h at 37°C. Cells were stained with 50 µl LIVE/DEAD fixable aqua viability dye L34957 (Invitrogen, Paisley, UK), anti-CD4-V605 (Biolegend, UK), anti-CD45RO-PerCP-Cy5.5, anti-CXCR5-A488, anti-CXCR3-PeCF594, anti-PD-1-BV421, and anti-CCR7-PE-Cy7 (Becton, Dickinson, Oxford, UK) and stained intracellularly with anti-CD154-PE and anti-IFN-γ-APC (Becton, Dickinson, Oxford, UK) (see Table S5 in Supplementary Material). A minimum of 30,000 viable CD3+ CD4+ lymphocyte events were acquired on a BD LSR Fortessa, which had been set up according to the Perfetto protocol (34, 35). Data were analyzed using FlowJo (Treestar, Ashland, OR, USA), PESTLE, and SPICE (courtesy of Mario Roederer, Vaccine Research Center, National Institutes of Allergy and Infectious Diseases) software. An intracellular cytokine secretion responder was considered positive if the background-subtracted values were twice those of the mock stimulus for each CD4-cytokine combination.

To assess the magnitude of mBC responses, we adapted a previously published protocol (36). In brief, frozen PBMC from baseline (D0), 28 days post-priming phase (D84) and 28 days post-booster injection (D196 or D364) were resuspended at 1 × 10⁶ PBMC/ml in stimulation media composed of R10, 5 ng/ml IL-2 (Roche, UK) and 0.5 µg/ml R848 (Vac-r848, Invivogen, UK), cultured for 4 days and washed with R10 prior to plating.

Sterile 96-well ELISpot plates (MSIPS4510 from Millipore) were pre-wet with 15 μl/well 70% EtOH for 1 min. After washing the plates five times with 200 μl/well sterile water, 100 μl/well of the CN54gp140 Ag (5 µg/ml) and 100 µl of the capture antibody MT91/145 (Mabtech, at 15 µg/ml) were added to the ELISpot plate and incubated overnight at 4°C. To remove excess Ag, plates were washed five times with sterile 200 μl/well PBS before blocking with 200 μl/well R10 for 1–2 h, at 4°C. Following the removal of R10, cells were added at 10,000/well for total IgG and 125,000/well for Ag-specific response in R10. Each condition was tested in triplicate, and plates were incubated for 6 h at 37°C, 5% CO₂. For spot development, plates were first washed six times with 200 μl/well PBS and then 100 μl/well of detection antibody MT78/145 (Mabtech, 1:500) was added at 1 µg/ml in PBS/0.5% FCS and incubated overnight at 4°C. The next day plates were washed five times with 200 μl/well PBS and 100 μl/well streptavidin-HRP (Mabtech, 1:1,000), diluted in PBS/0.5% FCS, was added and incubated for 1 h at room temperature. Fifteen minutes before developing the plate, the AEC substrate (BD Biosciences) was prepared according to the manufacturer’s instructions. After washing the plates five times with 200 μl/well PBS, 100 μl/well of the AEC substrate solution was added and incubated for a maximum of 5–10 min before stopping the spot development with tap water. SFU were counted using an automated AID ELISpot reader (Autoimmun Diagnostika GmbH, Strassberg, Germany).

A customized multivariate multiplex assay was developed using a panel of gp140 Ags (Clade C: CN54, 1086, Clade A: UG37, Clade D: UG21–NIH AIDS Reagents) coupled to magnetic fluorescent multiplex beads (Bio-Rad, AU) as described previously (37). Biotinylated dimeric Fc-gamma-receptors (FcγRIIa-H131, FcγRIIIa-V158) were produced as previously described (16).

Coupled microspheres were premixed in multiplex assay buffer (PBS + 0.1% BSA + 0.05% Tween), creating a working mixture of 12.5 microspheres/bead type/μl. Using a black, clear bottom 96-well plate 40 µl of the working microsphere mixture (1,000 beads of each type/well) was added to 40 µl of 100× diluted serum (diluted in PBS). The plate was covered and incubated overnight at 4°C on a plate shaker. The plate was washed three times with 200 µl of assay wash (PBS-1×, 0.1% BSA, 0.5% Triton-100X) using a Bio-Plex Pro plate wash station (Bio-Rad, AU). Ag-specific antibody binding to dimeric FcγRs was detected by adding FcγRs followed by streptavidin PE, at 1.0 µg/ml with 50 μl/well. After 2 h incubation at room temperature on a shaker, the plate was washed three times with 200 µl of assay wash, and microspheres were resuspended in 100 µl of sheath fluid.

A Bio-Plex reader (Bio-Plex MAGPIX, Bio-Plex Manager 5.0, Bio-Rad) was used to detect the microspheres, and binding of PE was measured to calculate a median fluorescence intensity (MFI). Background signal, defined as the average MFI observed for each microsphere set when incubated with the PE-conjugated detection reagent in the absence of clinical antibody sample, was subtracted from the MFI for each sample.

Sera were tested for neutralizing activity against HIV-1 at Monogram Biosciences (San Francisco, CA, USA) as described previously (38, 39). A panel of viruses (92BR020, 92TH021_043, 93IN905_040, 94UG103_023, MGRM-C-026_024, SF162, TH023) covering clades A, B, C, E, and including the neutralization sensitive laboratory strain NL43 and the primary isolate JRCSF, was used. Neutralizing activity is displayed as the percent inhibition of viral replication (luciferase activity) at each antibody dilution compared with the specificity control (aMLV). Titers were calculated as the reciprocal of the plasma dilution conferring 50% inhibition (IC50).

Statistical analysis was carried out using Prism 7.0a (GraphPad, CA, USA) or the R software (R3.3.2) for statistical programming (40).

Fifteen study volunteers were assessed for eligibility with 14 deemed eligible according to the pre-defined inclusion and exclusion criteria. Two participants in study Group A withdrew due to personal reasons and were not included in the final immunogenicity analysis (Figure 1).

The majority of participants were male, and the mean age was 27.2 years (Table S1 in Supplementary Material). Overall, the vaccine was safe and well tolerated. Table S2 in Supplementary Material shows the proportion of volunteers with Grade 2 or higher reactogenicity events through day 7 post any vaccination. Most systemic reactogenicity events (fever, malaise, myalgia, headache, nausea, and vomiting) were graded as Grade 1 or Grade 2. The overall frequency of any Grade 2 or higher systemic reactogenicity following any vaccination was 42.9% (95% CI: 17.7–71.1; Table S2 in Supplementary Material). One Group B volunteer, X00143007, reported Grade 3 malaise on Day 2 after the third vaccination. All other reactions were Grade 1 or Grade 2 among volunteers in both study groups. All local reactogenicity (pain, tenderness, erythema/skin discoloration, and swelling/hardening or thickening) were graded as present or Grade 1.

The humoral response to two homologous prime-boost HIV-vaccine regimens (Group A, n = 6; Group B, n = 5) was initially assessed by Ag-specific ELISA before and after sequential intramuscular IMs with the clade C ENV protein CN54gp140 (Figure 2). A low responder in Group B (Figure S3 in Supplementary Material) with significantly reduced levels of Ag-specific antibody levels was excluded as an outlier from the immunogenicity analysis in Figure 2.

The initial priming phase (first three IMs, Table 1) was common to both groups. Ag-specific IgG, IgA, and IgM antibodies were detected as early as 14 days after the first priming IM in both groups. IgG responses were significantly higher than IgA and IgM responses (p < 0.0001, Group A and B). Peak Ag-specific IgG, IgA, and IgM levels were reached after second IM, while the third IM had little additional impact on the magnitude of the response (Figures 2A,B). The decay in Ag-specific levels (IgG and IgA) between days 84 and 168 was similar for both groups (Group A/B: 2.4-fold/3.4-fold for IgG and 3-fold/3.7-fold for IgA). Although IgM levels appeared lower at day 168 in group B relative to group A, the differences were not statistically significant (p > 0.9999), where levels for both groups were at the lower limits of detection. In the boosting phase the fourth IM at 6 months (Group A) and 12 months (Group B) had a similar impact on the magnitude of boosted responses with no statistical differences between the respective endpoints of Ag-specific IgG, IgA, and IgM (Group A versus Group B: IgG: p > 0.99, IgA: p > 0.99, IgM: p = 0.56).

Further subclass analysis was performed to dissect IgG-subclass responses. Ag-specific IgG1/2/3 were detected as early as 14 days after the first IM on D0 (Figures 2C,D). Ag-specific IgG1/2 antibodies dominated the antibody profile in both groups and were detectable across all visits 14 days post the first IM. Interestingly, while the third IM had no apparent impact on Ag-specific IgG1/2 levels, it induced Ag-specific IgG4 levels in both groups, and these continued to rise between days 84 and 168 (Figures 2C,D) and were significantly boosted by the respective fourth IM (Figure S4 in Supplementary Material). In contrast, concentrations of Ag-specific IgG3 rapidly declined after the third IM to below the detection limit of the assay but were re-boosted to similar levels following the fourth IM (D168, Group A and D336, Group B). Intriguingly Ag-specific IgG2 levels appeared to be better sustained over days 84–168 in both groups than IgG1, which decayed 4.3-fold (Group A)/6.9-fold (Group B) and was of similar magnitude to IgG2 by day 168 in Group A (Figure 2C; IgG1/IgG2 ratio = 1.4), dropping below the levels of Ag-specific IgG2 by day 336 in group B (Figure 2D; IgG1/IgG2 ratio = 0.2). There was no apparent difference in Ag-specific IgG-subclass response to early (6 months, Group A) and late (12 months, Group B) boosting IMs (fourth IM); however, these re-boosted levels of Ag-specific IgG in both groups to levels observed after the second IM.

Antigen-specific mucosal antibody levels were determined using optimized sampling protocols and a high sensitivity ELISA assay. Reference values for total IgG and IgA for the different mucosal secretions are given in Figure S5 in Supplementary Material. Low levels of Ag-specific IgG antibodies could be detected in seminal and rectal secretions (Figure 3) and, for the four female participants, in cervicovaginal secretions (Figure S6 in Supplementary Material).

Mucosal antibody levels were significantly lower than Ag-specific IgG in serum samples (Figures 3A–D), but significantly correlated (p < 0.0001) with Ag-specific IgG in serum samples from both groups (Figure 3E). Response curves in mucosal secretions and the systemic compartments were identical, with a relatively constant differential between mucosal and systemic antibody levels between the respective compartments in both study groups (Figures 3A–D). Comparable trends were observed for both Groups A and B (Figures 3A–D) with no statistical difference in peak Ag-specific mucosal antibody levels 2 weeks post the fourth IM. Ag-specific IgG was below the limit of detection in urethral samples and reflects the low level of detected total IgG (Figure S5 in Supplementary Material) by this method. None of the mucosal samples exhibited detectable levels of Ag-specific IgA (data not shown), reflective of the low serum IgA responses (Figure 2).

Vaccine-induced IgG responses toward linear epitopes in the HIV Env were mapped using a customized peptide microarray consisting of overlapping linear 15mer Env peptides 2 weeks after third and fourth IM in relation to baseline (Figure 4).

Both groups, irrespective of the interval between booster injections, showed similar IgG Env-peptide recognition patterns (Figure 4A) both before and after the fourth IM. No difference in epitope recognition between the two groups was observed. A dominant recognition of the V3 tip region (peak number 2) by all vaccines can be observed after the three priming IMs. The longer boosting interval (Group B) was associated with an apparently higher median response to the V3 tip when compared to Group A. This difference, however, was not significant due to the high inter-individual variation in the IgG response toward the V3 tip (Figure S7 in Supplementary Material). IgG responses toward other regions within gp120 (peaks number 1 and 3, Figure 4B) were of lower magnitude and observed with varying frequency (33–83% of all vaccines) at both time points and in both groups. Recognition of gp41 only seems to occur in Group A, albeit already after three IMs and therefore seems to be due to interindividual differences in the IgG response rather than attributable to the boosting interval. There were no significant differences between the two groups in the total sum of fluorescence intensity values for all peptide variants included in the array (Figure 4C).

Even though vaccine-elicited antibodies did not neutralize virus in this study (Table S3 in Supplementary Material), we sought to further investigate the ability of Ag-specific antibodies to crosslink Fc receptors. Recently, the role of non-neutralizing HIV antibodies in vaccine-mediated protection has gained prominence as they can induce important Fc effector-mediated functions. We have previously developed an assay using recombinant dimeric Fc receptors that provide an indirect readout for influenza-specific ADCC and phagocytosis (15, 16). Here, we investigated any differences in the ability of vaccine-induced serum antibodies to interact with dimeric Fc receptors FcγRIIIa and/or FcγRIIa after binding to various HIV-1 clade env proteins (Figures S8–S10 in Supplementary Material). For C.CN54gp140, A.UG37gp140, or D.UG21 proteins, both FcγR-profiles (Figures 5A–D) and Ag-specific IgG levels were compared (Figures 5E,F).

In both groups, clade C and D binding vaccine-induced antibodies were detected by the dimeric FcγRIIIa and FcγRIIa, while vaccine-induced antibodies specific to clade A did not crosslink dimeric FcγRIIIa and FcγRIIa (Figures 5A–D).

As expected, in an Ag-specific ELISA the largest proportion of vaccine-induced antibodies bound the clade C CN54gp140 immunogen (Figures 5E,F). Fourteen days after the final IM (Figures 5E,F, indicated by 4°), the concentration of vaccine-induced antibodies binding to clade A and/or clade D was about 3- or 4.5-fold lower, respectively, in Group A and 2- or 5-fold lower, respectively, in Group B (Figures 5E,F). Overall, the fingerprint of Fc-receptor binding, associated with different HIV-1 clades, was similar in both pattern and magnitude for the two study groups (Figure 5; Figure S7 in Supplementary Material) and FcR engagement following binding to five HIV-clade proteins correlated significantly with levels of CN54gp140-specific IgG1, 2 and 3 in both groups (Figure 6). Similarly to a previous study of HIV-controllers (41), the strongest positive correlation with nearly identical spearman coefficients between Group A and B was observed between FcγRIIIa/FcγRIIa dimerization and CN54gp140-specific IgG1 (Group A/B: ***p < 0.001, ***p < 0.001). The increases after the initial IM and subsequent declines observed in the Ab response profiles (Figures 5E,F) for the two groups are generally matched as features in the FcR activity profiles (Figures 5A–D). Interestingly, however, FcγR activity was not merely reflective of Ab titer since, the dimeric FcγRIIIa showed similar binding activity to opsonized clade C and clade D (Figures 5A,B), despite a profound reduction in antibody detecting clade D HIV Env (Figures 5E,F). Significantly, this suggests the antibodies that did cross-react against clade D Env, though low in titer, had high Fc-receptor binding ability.

An mBC ELISPOT assay was used to compare the dynamics and magnitude of Ag-specific mBC responses, generated by the two homologous prime-boost HIV-vaccine regimens (Group A, n = 6; Group B, n = 5) at 28 days after the third and fourth IM, respectively (Figure 7).

The frequency (mean ± SD) of Ag-specific mBC responses was similar for both vaccine regimens at 28 days after the third priming IM (Group A: 0.37 ± 0.25%; Group B: 0.21 ± 0.19%) and 28 days after the fourth IM (Group A: 0.47 ± 0.29%; Group B: 0.57 ± 0.46%). In each group, the highest responders exhibited an Ag-specific mBC frequency of about 1% total IgG-positive mBC. In contrast to the mBC response, no positive Ag-specific T cell responses were detected, via IFNγ-ELISpot or intracellular staining for CD154 and IFNγ, 14 days after the third and fourth IM in either group (Figure S11 in Supplementary Material).