Circulating immune and plasma biomarkers of time to HIV rebound in HIV controllers treated with vesatolimod.

ABSTRACT

Background: Antiretroviral therapy (ART) for HIV-1 treatment has improved lifespan but requires lifelong adherence for people living with HIV (PLWH), highlighting the need for a cure. Evaluation of potential cure strategies requires analytic treatment interruption (ATI) with close monitoring of viral rebound. Predictive biomarkers for HIV-1 rebound and/or duration of control during ATI will facilitate these HIV cure trials while minimizing risks. Available evidence suggests that host immune, glycomic, lipid, and metabolic markers of inflammation may be associated with HIV-1 persistence in PLWH who are treated during chronic HIV-1 infection. Methods: We conducted post-hoc analysis of HIV controllers who could maintain low levels of plasma HIV-1 without ART in a phase 1b vesatolimod trial. Baseline and pre-ATI levels of immune, glycomic, lipidomic, and metabolomic markers were tested for association with ATI outcomes (time of HIV-1 rebound to 200 copies/mL and 1,000 copies/mL, duration of HIV-1 RNA ≤400 copies/mL and change in intact proviral HIV-1 DNA during ATI) using Spearman's correlation and Cox proportional hazards model. Results: Higher levels of CD69+CD8+ T-cells were consistently associated with shorter time to HIV-1 rebound at baseline and pre-ATI. With few exceptions, baseline fucosylated, non-galactosylated, non-sialylated, bisecting IgG N-glycans were associated with shorter time to HIV rebound and duration of control as with previous studies. Baseline plasma MPA and HPA binding glycans and non-galactosylated/non-sialylated glycans were associated with longer time to HIV rebound, while baseline multiply-galactosylated glycans and sialylated glycans, GNA-binding glycans, NPA-binding glycans, WGA-binding glycans, and bisecting GlcNAc glycans were associated with shorter time to HIV rebound and duration of control. Fourteen bioactive lipids had significant baseline associations with longer time to rebound and duration of control, and larger intact proviral HIV-1 DNA changes; additionally, three baseline bioactive lipids were associated with shorter time to first rebound and duration of control. Conclusion: Consistent with studies in HIV non-controllers, proinflammatory glycans, lipids, and metabolites were generally associated with shorter duration of HIV-1 control. Notable differences were observed between HIV controllers vs. non-controllers in some specific markers. For the first time, exploratory biomarkers of ATI viral outcomes in HIV-controllers were investigated but require further validation.