Re-boost immunizations with the peptide-based therapeutic HIV vaccine, Vacc-4x, restores geometric mean viral load set-point during treatment interruption

Abstract

Background: Vacc-4x, a therapeutic HIV vaccine candidate has previously induced a significant reduction in viral load (VL) set-point compared to placebo upon interruption of combination anti-retroviral therapy (ART) (2007/1 study). This study, (2012/1), explored the potential to maintain Vacc-4x effect by re-boosting eligible 2007/1 study participants.

Methods: Participant inclusion required 2007/1 participants to have completed all Vacc-4x immunizations and interrupted ART for up to 26 weeks. At weeks (wk)0 and 2, participants received intradermal (i.d.) Vacc-4x booster immunizations (1.2mg) on ART with GM-CSF (60μg) i.d. as a local adjuvant. ART was interrupted for up to 16 weeks (wk12-wk28). Participants were then followed on ART until wk36. VL set-point, total proviral DNA (pvDNA) and immunogenicity assessed by IFN-γ ELISPOT, T-cell proliferation and delayed type hypersensitivity (DTH) reactions were compared to participants' values in the 2007/1 study where available.

Results: This open, multicenter, clinical study enrolled 33 participants from 9 clinical trial sites in the US and Europe. In the per-protocol (PP) population, the VL set-point geometric mean (GM) 18162 copies/mL was not significantly changed compared to the 2007/1 study (GM VL 22035 copies/mL), (p = 0.453, n = 18). For participants with available preART VL values, the VL set-point (GM 26279 copies/mL) remained significantly lower than the preART VL set-point (GM 74048 copies/mL, p = 0.021, n = 13). A statistically significant reduction in pvDNA (49%) from baseline to wk4 was observed (p = 0.03, n = 26). DTH responses (wk4) increased significantly from baseline (p = 0.006, n = 30) and compared to the 2007/1 study (p = 0.022, n = 29) whilst the proportion of participants with ELISPOT and T-cell proliferation responses was similar between the two studies.

Conclusions: Vacc-4x booster immunizations safely maintained the mean VL set-point at that established following primary Vacc-4x therapeutic immunization. The reduction in pvDNA during ART supports the potential for Vacc-4x immunization to reduce HIV reservoirs and thereby contribute to combination HIV cure strategies.

Trial registration: ClinicalTrials.gov NCT01712256.

Conflict of interest statement

We have the following interests: This study was funded in part by the sponsor Bionor Immuno AS. During the course of the study Jürgen K Rockstroh, Giuseppe Pantaleo and Barry Peters were members of Bionor Immuno’s Clinical Advisory Board. Mats Ökvist (MO) and Maja A. Sommerfelt (MAS) were employees of Bionor Immuno and had shares in the company. Currently, MO and MAS are no longer employed at Bionor Immuno and do not have shares in the company. Darren Jolliffe (DJ) is employed by S-Cubed Biometrics Ltd., Kim Krogsgaard (KK) is employed by KLIFO. Neither S-Cubed nor KLIFO provided funding for the study. The remaining co-authors have no competing interests. The following patents related to this study are owned by Bionor Immuno AS: HIV PEPTIDES, ANTIGENS, VACCINE COMPOSITIONS, IMMUNOASSAYS KIT AND A METHOD OF DETECTING ANTIBODIES INDUCED BY HIV WO00/52040 and METHOD FOR REDUCING AND/OR DELAYING PATHOLOGICAL EFFECTS OF HUMAN IMMUNODEFICIENCY VIRUS I (HIV) OR FOR REDUCING THE RISK OF DEVELOPING ACQUIRED IMMUNODEFICIENCY SYNDROME (AIDS) WO2016/005508. There are no further patents, products in development or marketed products to declare related to this study. This does not alter our adherence to all the PLOS ONE policies on sharing data and materials.

Figures

Fig 1. CONSORT diagram for the 2012/1 clinical study.

ITT: Intention to treat population. PP: per protocol. One participant was lost to follow-up after completing the week 28 visit, but was included in outcomes analysis in the ITT population. ART: antiretroviral therapy.

Fig 2. Study schedule for CT-BI Vacc-4x 2007/1 and CT-BI Vacc-4x 2012/1 clinical studies.

Thick arrows indicate Vacc-4x immunizations. Analyses carried out during the 2012/1 clinical study are shown. Scr.: Screening; ELISPOT: Enzyme-linked Immunosorbent Assay; T-cell prolif.: T-cell proliferation; pvDNA: proviral DNA; DTH: Delayed type hypersensitivity; CD4 and CD8 T-cell counts; VL: viral load.

Fig 3. Viral load set-points in the per protocol population for participants with all three time points (n = 13).

The box plots show the minimum, first quartile (Q1), median, third quartile (Q3) and maximum values for each set-point. Diamonds correspond to GM values. The maximum VL value for preART VL is 750 000 copies/mL. PreART GM VL set-point: 74048 copies/mL; 2007/1: 22475 copies/mL; 2012/1: 26279 copies/mL. The p-values are derived from a paired t-test on logged set-point values.

Fig 4. Geometric mean of proviral DNA (pvDNA) levels per million peripheral blood mononuclear cells (PBMC) in the per protocol population.

One participant lacked pvDNA data from week 1. p-value was derived from a non-parametric Wilcoxon signed-rank test.

Fig 5. The proportion of per protocol participants showing positive ELISPOT assay responses during the 2012/1 and the 2007/1 studies.

For participants with values at week (wk) 0 in the 2012/1 and wk1 in the 2007/1 study (n = 23), the paired difference in proportions of participants with positive assay response was not statistically significant (-0.87; CI: -0.202, 0.028; p = 0.5). The two-sided p-value was derived from exact version of McNemar’s test. The 95% CI may not be valid as analysis is based on a small sample of data. The sample size was too small for a comparison of ELISPOT assay responses between wk28 in the 2012/1 study and wk52 in the 2007/1 study in participants that had data for both time points (n = 11). *Participants contributing data to the p-value.

Fig 6. The proportion of per protocol participants showing ‘Vacc-4x ELISPOT Responder’ profile during the 2012/1 study and at week (wk) 52 for the 2007/1 study.

The difference in the proportion of participants with ‘Vacc-4x ELISPOT Responder’ profile at wk0 in the 2012/1 and wk1 in the 2007/1 study (n = 11) was not statistically significant (-0.429, CI:-0.637, 0.092, p = 0.375). Two-sided p-value was derived from exact version of McNemar’s test. The 95% CI may not be valid as analysis is based on a small sample of data. *Participants contributing data to the p-value.

Fig 7. The proportion of per protocol participants showing positive T-cell proliferation assay responses during the 2012/1 study at week (wk)1 and wk52 in the 2007/1 study.

For participants with values at wk0 in the 2012/1 and wk1 in the 2007/1 study, the difference in the proportion of participants with positive assay responses for the CD4 T-cell population was statistically significant (-0.462; CI: -0.733, -0.191, p = 0.031 n = 13), but not for the proportion of participants with CD8 T-cell proliferation responses (-0.200, CI: -0.420, 0.002, p = 0.250, n = 15). The two-sided p-value was derived from exact version of McNemar’s test. The 95% CI may not be valid as analysis is based on a small sample of data. The sample size was too small for a comparison of T-cell proliferation assay responses between wk28 in the 2012/1 study and wk52 in the 2007/1 study in participants that had data for both time points (CD4 n = 7; CD8 n = 7). *Participants contributing data to the p-value.

Fig 8. The proportion of per protocol participants showing ‘Vacc-4x T-cell proliferation Responder’ profiles in CD4 and CD8 T-cell subpopulations for the 2012/1 study and wk52 of the 2007/1 study.

The sample size was too small for a comparison of ‘Vacc-4x T-cell proliferation Responders’ between wk28 in the 2012/1 study and wk52 in the 2007/1 study in participants that had data for both time points (CD4 n = 6; CD8 n = 7).

Fig 9. Median CD4/CD8 ratio over time for the per protocol population.

ART was stopped at week 12 and resumed at week 28. Bars indicate interquartile range.