Post-exposure prophylaxis with highly active antiretroviral therapy could not protect macaques from infection with SIV/HIV chimera

Abstract

In HIV-seropositive patients, the potent efficacy of the combination of nucleoside analogues and protease inhibitors leading to the suppression of viral load below the detection threshold, suggested the use of highly active antiretroviral therapy (HAART) for post-exposure prophylaxis of HIV transmission. To date, only one study [1] has reported that zidovudine (ZDV) alone may protect from occupational post-exposure infection with an efficacy estimated at 81%. However, a retrospective case–control study is not the optimal design for assessing the efficacy of such strategies, thus limiting the significance of this observation. As for vaccine trials, the prophylactic treatment of HIV infection needs to be assessed in prospective placebo-controlled studies, which are difficult, if not impossible, to set up in humans. Pre-clinical approaches in non-human primate models of AIDS enable this experimental evaluation to be carried out, and offer the possibility to determine precisely the conditions of its efficacy [2]. Currently, the infection of macaques with pathogenic strains of SIV, or related chimera expressing the envelope of HIV-1 (SHIV), represents the most relevant model of human HIV infection and AIDS [3,4]. The objective of this study was to evaluate, in a macaque model, the efficacy of the combination of ZDV, lamivudine (3TC) and indinavir, administrated early after experimental intravenous inoculation of the pathogenic SHIV89.6P. This chimera expresses the envelope gene of HIV-1 89.6 primary isolate, uses both CCR-5 and CXCR-4 co-receptors like many other primary isolates of HIV-1, and induces AIDS in macaques [4]. In vitro, the replication of this virus is inhibited by ZDV, 3TC or indinavir with IC90 ranging between 50 and 100 nM. Three groups of cynomolgus macaques (Macaca fascicularis) were inoculated intravenously with 50 MID50 of a cell-free stock of SHIV 89.6P grown on monkey cells (kindly provided by Dr Anne Marie Aubertin, Strasbourg, France). This dose corresponds to the infectious titre that may be present in 25 μl (estimated needle volume) of plasma obtained from patients with high viral load, as is the case during primary infection or late stages of the disease. Three control animals were treated with placebo and four animals were treated, as recommended in humans [5,6], by the combination of ZDV (4.5 mg/kg), 3TC (2.5 mg/kg) and indinavir (20 mg/kg) 4 h after experimental intravenous inoculation of SHIV89.6P. The treatment was administered twice a day by the oral route through a nasogastric catheter and was continued until day 28 post-inoculation. A third group of four macaques received the same combination of drugs but treatment was initiated 72 h after virus inoculation and continued until day 28 post-inoculation. All the animals became infected, demonstrating that initiating this treatment as early as 4 h post-inoculation does not protect from intravenous inoculation of cell-free virus. However, during primary infection, plasma viral RNA load was reduced by at least a hundredfold in both groups of macaques receiving HAART (Fig. 1a). In the control group, the numbers and percentages of CD4 circulating T lymphocytes dramatically decreased during the second week post-inoculation (Fig. 1b) and remained very low thereafter until the end of the study, 12 months later (mean ± SD 183 + 94 cells/μl). A dramatic decrease in CD4 T cell percentages was also observed by day 14 post-inoculation in peripheral lymph nodes (data not shown). By contrast, during the 4 weeks of treatment, no significant changes in circulating and peripheral lymph node CD4 T lymphocyte percentages were noticed in both treated groups, although a mild decrease in CD4 T cell counts in the blood was observed between days 14 and 28 post-inoculation (Fig. 1b). One week after treatment arrest, a significant and transient decrease of this cell population was noticed, coinciding with a transient rebound of cell-associated viraemia or plasma viral load. One year after infection, i.e. 11 months after treatment arrest, CD4 and CD8 circulating T cell numbers remained elevated in previously treated animals (619 ± 221 CD4 cells/μl and 752 ± 122 CD4 cells/μl for monkeys treated with HAART initiated at 4 or 72 h, respectively).Fig. 1.: Numbers of copies of viral RNA, measured using a branched DNA assay (Bayer's diagnostics, Amsterdam, the Netherlands), in the plasma of macaques treated with the placebo (a1), or with highly active antiretroviral therapy (HAART) initiated at 4 h post-inoculation (a2) or 72 h post-inoculation (a3). Numbers of CD4 circulating T cells, measured as previously described [2], in the plasma of macaques treated with placebo (b1), or with HAART initiated at 4 h post-inoculation (b2) or 72 h post-inoculation (b3).This is the first demonstration that post-exposure prophylaxis of HIV transmission with a therapeutic design recommended in humans could not protect macaques from experimental challenge with a pathogenic lentivirus closely related to HIV-1. However, a reduction of viral load during primary infection in treated animals preserved CD4 T lymphocyte subsets from dramatic decrease. Our results suggest that HAART should be initiated as early as possible, within minutes, or at the latest, within hours after exposure, and stresses the need for the discovery of more efficient prophylactic strategies, combining, for instance, potent antiretroviral drugs and immunotherapeutic approaches. Acknowledgements The authors gratefully acknowledge the efficient collaboration of Drs R. Bidault and D. Lapierre from Glaxo Wellcome, Drs M.-C. Gervais and P. Duprat from Merck Sharp and Dohme-Chibret. They also thank B. Boson, B. Delache, D. Renault, P. Pochard, J.C. Wilk and R. Rioux for excellent technical assistance. Roger Le Grand Bruno Vaslin Jérome Larghero Olivier Neidez Hugues Thiebot Pierre Sellier Pascal Clayette Nathalie Dereuddre-Bosquet Dominique Dormont