Cell-Mediated Immune Predictors of Vaccine Effect on Viral Load and CD4 Count in a Phase 2 Therapeutic HIV-1 Vaccine Clinical Trial

BACKGROUND: In a placebo-controlled trial of the peptide-based therapeutic HIV-1 p24(Gag) vaccine candidate Vacc-4x, participants on combination antiretroviral therapy (cART) received six immunizations over 18 weeks, followed by analytical treatment interruption (ATI) between weeks 28 and 52. Cell-mediated immune responses were investigated as predictors of Vacc-4x effect (VE) on viral load (VL) and CD4 count during ATI. METHODS: All analyses of week 28 responses and fold-changes relative to baseline considered per-protocol participants (Vacc-4x:placebo = 72:32) resuming cART after week 40. Linear regression models with interaction tests were used. VE was estimated as the Vacc-4x–placebo difference in log(10)-transformed VL (VE(VL)) or CD4 count (VE(CD4)). FINDINGS: A lower fold-change of CD4+ T-cell proliferation was associated with VE(CD4) at week 48 (p = 0.036, multiplicity adjusted q = 0.036) and week 52 (p = 0.040, q = 0.080). A higher fold-change of IFN-γ in proliferation supernatants was associated with VE(VL) at week 44 (p = 0.047, q = 0.07). A higher fold-change of TNF-α was associated with VE(VL) at week 44 (p = 0.045, q = 0.070), week 48 (p = 0.028, q = 0.070), and week 52 (p = 0.037, q = 0.074). A higher fold-change of IL-6 was associated with VE(VL) at week 48 (p = 0.017, q = 0.036). TNF-α levels (> median) were associated with VE(CD4) at week 48 (p = 0.009, q = 0.009). INTERPRETATION: These exploratory analyses highlight the potential value of investigating biomarkers in T-cell proliferation supernatants for VE in clinical studies.