Ex vivo HIV entry into blood CD4+ T cells does not predict heterosexual HIV acquisition in women

BACKGROUND: A blood-based assay that could quantify HIV susceptibility would be very valuable for HIV prevention research. Previously, we developed and validated an ex vivo, flow-based, HIV entry assay to assess genital HIV susceptibility in endocervical CD4+ T cells. METHODS: Here we assessed whether this tool could be used to predict HIV risk using blood-derived CD4+ T cells in a rigorously-blinded, nested case-control study using blood samples collected from high-risk, HIV-uninfected South African women enrolled in the CAPRISA 004 clinical trial. Cases, subsequently acquiring HIV were sampled prior to HIV infection and compared with controls, who remained HIV-uninfected. The primary endpoint was ex vivo entry of a CCR5-tropic HIV founder virus into blood CD4+ T cells. Secondary endpoints included HIV entry into CD4+ central (T(CM)) and effector (T(EM)) memory T cells, and into CD4+ T cell subsets expressing CCR5, CD69, CCR6, α4β1 or α4β7. RESULTS: Compared to bulk CD4+ T cells (4.9% virus entry), CD4+ T cells expressing CCR5, CCR6 or α4β1 and T(EM) were highly susceptible (15.5%, 8.8%, 8.2% and 10.8% entry, respectively, all p<0.0001), while T(CM), CD69+ or α4β7+ CD4+ cells were moderately susceptible (6.4%, 6.0% and 5.8% respectively, p ≤ 0.003). While the proportion of the aforementioned highly susceptible cells correlated with overall virus entry into CD4+ T cells within an individual (r = 0.68, 0.47, 0.67, and 0.60 respectively, p<0.0001), blood virus entry did not predict subsequent mucosal HIV acquisition after controlling for sexual behaviour and condom use (OR 0.92, 95% CI 0.77–1.11, p = 0.40). CONCLUSIONS: Although virus entry identified several previously known highly susceptible cellular HIV targets, blood HIV entry did not predict subsequent heterosexual HIV acquisition. Assessment of mucosal HIV susceptibility may require sampling at the site of HIV exposure.