The A5345 prospective treatment interruption (TI) study in the setting of modern ART previously demonstrated a modest delay in HIV rebound in those who had initiated ART during early infection. From A5345, we sought to identify pre-TI predictors of time to HIV rebound that could be crucial for designing and evaluating interventions for HIV remission.
A5345 participants initiated ART during chronic (N=33) or early (N=12) HIV infection with ?2 years of suppressive ART, and restarted ART if two viral loads ?1,000 copies/mL after TI. Viral reservoir markers quantified pre-TI included: unspliced cell-associated RNA (CA-RNA), total HIV DNA (total-DNA), intact proviral DNA (IPD) by the IPDA, integrase single-copy assay (iSCA), and infectious units/million rCD4 cells (IUPM) by viral outgrowth assay. We assessed associations between reservoir measures with time to HIV rebound and cellular and humoral immune responses.
Early-treated participants had smaller and less transcriptionally active HIV reservoirs as reflected by significantly lower levels of CA-RNA (9-fold), total-DNA (3-fold), IPD (4-fold) and IUPM (1.4-fold) compared to chronic-treated participants. IPD correlated with total-DNA and IUPM (Spearman r=0.45, P=0.004 for both). As part of the primary objective, we found that lower reservoir activity was modestly associated with a delay in HIV rebound (CA-RNA: r=-0.26, P=0.08). The strongest predictor of time to HIV rebound was IPD in chronic-treated participants (r=-0.37, P=0.04) and iSCA in early-treated participants (r=-0.68, P=0.02). Amongst all participants, a higher percentage of HIV gag-specific CD8+ T cell cytotoxic response (CD107a) was associated with a lower IPD (r=-0.37, P=0.02). In early-treated participants, a higher percentage of CD8+ T cells expressing ?2 effector cytokines was also associated with lower IPD (r=-0.66, P=0.05). Levels of HIV antibodies were positively correlated with both reservoir size (total-DNA, IPD, and IUPM) and activity (CA-RNA).
This TI trial found that smaller size and lower activity of the HIV reservoir predicted only a modest delay in HIV rebound for most participants, with the degree of association differing by early versus chronic ART initiation. However, the restricted reservoir size and activity in early-treated participants support previous reports that early ART may enrich for post-treatment control. We also identified key HIV-specific CD8 immune responses associated with smaller intact HIV reservoir size.