At CROI 2017, Michel Nussenzweig from Rockefeller University presented evidence that early administration of a short course of two broadly neutralizing antibodies (bNAbs)—3BNC117 and 10-1074—led to prolonged immunological control of SHIV viral load in macaques. The study was subsequently published in Nature. At the end of last month, results from the first human trials of the same dual bNAb regimen were announced in papers in Nature and Nature Medicine.
The report in Nature by Pilar Mendoza and colleagues covers a study involving 15 people with HIV who were on suppressive ART at the time of enrollment. The screening process for the study included tests to assess whether HIV resistance to either bNAb could be detected in viruses grown from the latent reservoir, and individuals with samples displaying evidence of resistance were excluded (approximately half the individuals screened could not enroll due to evidence of resistance to one of the antibodies).
Participants received three infusions of the two bNAbs, spaced three weeks apart. An analytical ART interruption (ATI) was initiated two days after the first infusion in order to evaluate the potential of the bNAbs to maintain HIV viral load suppression.
Efficacy data are presented from 11 individuals (four participants were excluded from the analysis due to exhibiting viral loads above 20 copies/ml prior to the first bNAb administration). The median time to viral load rebound—defined as a confirmed viral load over 200 copies/ml—was 21 weeks, an improvement over the 6-10 weeks documented in a prior trial that gave only 3BNC117.
Nine of the 11 participants were able to remain off ART for at least 15 weeks, and two out of these nine did not meet the viral load rebound criteria for the entire 30 weeks of follow up. The bNAb combination appeared safe and well tolerated, with two cases of mild fatigue representing the most serious adverse events.
Additional detailed analyses revealed that the two participants who rebounded earliest—five and seven weeks into the ATI, respectively—had low-level HIV variants that were resistant to one of the two bNAbs (these variants had been missed by the less sensitive screening test). Therefore they had essentially received bNAb monotherapy during the ATI, explaining the rapidity of their viral load rebounds.
The two bNAbs were found to have somewhat different half-lives, with 10-1074 persisting longer in the body than 3BNC117. As a consequence, six of the seven participants experiencing later viral load rebounds had developed resistance to 10-1074 (after 3BNC117 levels declined, there was a period during which 10-1074 effectively became a monotherapy).
Measures of the HIV reservoir did not reveal any evidence of reductions caused by the dual bNAbs. Quantitative virus outgrowth assays (QVOA) were performed at study entry and 12 weeks into the ATI for a subset of participants, and no significant changes were observed.
In a presentation at the NIAID Strategies for an HIV Cure meeting yesterday (now available in the NIH videocast archive, starting at the 5:32:22 mark), Marina Caskey from Rockefeller University provided an update on the two individuals who maintained viral suppression the longest during the ATI. One remains off ART after 52 weeks while the other experienced viral load rebound at week 50, restarting ART shortly afterward. Although the study had been intended for people with chronic HIV infection, it turned out the former individual started ART fairly early, about 4-5 months after HIV infection, raising the question of whether post-treatment control might have occurred without an additional intervention. Caskey noted that their viral load at the time of ART initiation was high, at around 800,000 copies/ml.
Preliminary analyses of HIV-specific T cells in these two participants suggest that receipt of the bNAbs may have been associated with improvements in responses to Gag and Pol antigens, echoing an observation made in the macaque study, but Caskey took pains to emphasize that this work is still at a very early stage.
The second paper in Nature Medicine presents results from a cohort of seven individuals who were not on ART when they received the dual bNAbs. Significant viral load declines were achieved, but only one participant with a very low viral load at baseline (730 copies/ml) suppressed to undetectable levels. Furthermore, three of the seven participants were found to have evidence of resistance to the bNAbs that had been missed by the assays used at screening.
Taken together, the results show that combination bNAbs are safe and can have significant and prolonged activity against HIV. The issue of pre-existing resistance presents a challenge that may require the use of additional long-acting antiretroviral drugs or bNAbs if therapeutic regimens are to be designed for clinical use. The authors point out that modified, longer-acting versions of these and other bNAbs are now under investigation, suggesting that the development of combination approaches that require infrequent administration—perhaps every 3-6 months—is a realistic goal.
In the context of HIV cure research, the potential of bNAbs to enhance virus-specific immunity is of keen interest, and will undergo further exploration in several trials. There is also the possibility that bNAbs could mediate anti-reservoir activity via antibody-dependent cellular cytotoxicity (ADCC), with the caveat that this could also run up against the problem of HIV resistance—the bNAbs would have to be capable of recognizing and binding to the Env proteins of diverse viruses in the reservoir in order to flag the virus-infected cells for destruction by ADCC, and as yet it is unclear how many bNAbs would need to be combined for optimal coverage of the HIV variants present in most people.