SMAC Mimetic Plus Triple-Combination Bispecific HIVxCD3 Retargeting Molecules in SHIVCCH505-Infected, Antiretroviral Therapy-Suppressed Rhesus Macaques

The shock-and-kill human immunodeficiency virus type 1 (HIV-1) cure strategy involves latency reversal followed by immune-mediated clearance of infected cells. We have previously shown that activation of the noncanonical NF-kappa B pathway using an inhibitor of apoptosis (IAP), AZD5582, reverses HIV/simian immunodeficiency virus (SIV) latency. Here, we combined AZD5582 with bispecific HIVxCD3 DART molecules to determine the impact of this approach on persistence. Rhesus macaques (RMs) (n = 13) were infected with simian/human immunodeficiency virus SHIV.C.CH505.375H.dCT, and triple antiretroviral therapy (ART) was initiated after 16 weeks. After 42 weeks of ART, 8 RMs received a cocktail of 3 HIVxCD3 DART molecules having human A32, 7B2, or PGT145 anti-HIV-1 envelope (Env) specificities paired with a human anti-CD3 specificity that is rhesus cross-reactive. The remaining 5 ART-suppressed RMs served as controls. For 10 weeks, a DART molecule cocktail was administered weekly (each molecule at 1 mg/kg of body weight), followed 2 days later by AZD5582 (0.1 mg/kg). DART molecule serum concentrations were well above those considered adequate for redirected killing activity against Env-expressing target cells but began to decline after 3 to 6 weekly doses, coincident with the development of antidrug antibodies (ADAs) against each of the DART molecules. The combination of AZD5582 and the DART molecule cocktail did not increase on-ART viremia or cell-associated SHIV RNA in CD4(+) T cells and did not reduce the viral reservoir size in animals on ART. The lack of latency reversal in the model used in this study may be related to low pre-ART viral loads (median, <10(5) copies/ml) and low preintervention reservoir sizes (median, <10(2) SHIV DNA copies/million blood CD4(+) T cells). Future studies to assess the efficacy of Env-targeting DART molecules or other clearance agents to reduce viral reservoirs after latency reversal may be more suited to models that better minimize immunogenicity and have a greater viral burden. IMPORTANCE The most significant barrier to an HIV-1 cure is the existence of the latently infected viral reservoir that gives rise to rebound viremia upon cessation of ART. Here, we tested a novel combination approach of latency reversal with AZD5582 and clearance with bispecific HIVxCD3 DART molecules in SHIV.C.CH505-infected, ART-suppressed rhesus macaques. We demonstrate that the DART molecules were not capable of clearing infected cells in vivo, attributed to the lack of quantifiable latency reversal in this model with low levels of persistent SHIV DNA prior to intervention as well as DART molecule immunogenicity.