The pharmaceutical industry has made significant contributions to HIV vaccine research, but it has been rare for a major company to take the leading role in propelling a candidate toward efficacy testing. The only example to date occurred around the turn of the millennium with Merck's adenovirus serotype 5 (Ad5) vector, which showed some evidence of promise in macaque models but was ultimately discontinued after being associated with an enhanced -- rather than reduced -- risk of HIV acquisition in the STEP and Phambili trials.
Given Merck's cautionary tale, it is all the more notable that Janssen Vaccines and Prevention B.V. is now rapidly advancing an experimental prime-boost HIV vaccine regimen, with plans announced at the IAS 2017 conference to launch a large-scale "proof of concept" efficacy trial in women in late 2017/early 2018.
The Dutch biotech company Crucell, which Janssen acquired several years ago, initially produced the main vaccine candidates being studied. An iterative process involving both macaque and human studies has been ongoing, informing decisions about which versions of the vaccine constructs and which HIV antigens will be selected for efficacy evaluations. While Janssen is taking the lead, the program is a large collaboration including the Beth Israel Deaconess Medical Center (BIDMC), the Bill & Melinda Gates Foundation, the National Institute of Allergy and Infectious Diseases (NIAID), the HIV Vaccine Trials Network (HVTN), the International AIDS Vaccine Initiative, the US Military HIV Research Program, the Ragon Institute and the Los Alamos National Laboratory.
At IAS 2017 Hannah Schuitemaker, Vice President and Head of Viral Vaccine Discovery and Translational Medicine at Janssen Vaccines and Prevention B.V., presented results from a key clinical trial named APPROACH (the company has chosen monikers from mountaineering terminology to designate their steps toward the hoped-for summit of a licensable HIV vaccine). The trial randomized 400 participants into eight groups of 50, with each receiving different prime-boost recipes drawn from four possible ingredients: adenovirus serotype 26 (Ad26) vectors, modified Vaccinia Ankara strain (MVA) vectors and high or low dose trimeric clade C gp140 protein in alum adjuvant.
The Ad26 and MVA vectors used in the trial encode mosaic HIV antigens, which represent an amalgam of elements from multiple different HIV variants generated with the aid of computer software by Better Korber at the Los Alamos National Laboratory. The aim is to induce immune responses capable of recognizing HIVs from all circulating strains.
The crux of the APPROACH results is that the best-performing regimen comprised two priming immunizations (at months 0 and 3) with trivalent Ad26 vectors followed by two booster immunizations (at months 6 and 12) with a combination of the same Ad26 vectors plus a high dose soluble trimeric clade C gp140 Env protein with alum adjuvant. Trivalent refers to a mix of three Ad26 vectors, two containing different mosaic versions of Gag and Pol proteins (designated Ad26.Mos1.Gag-Pol and Ad26.Mos2.Gag-Pol) and one containing a mosaic Env protein (designated Ad26.Mos1.Env).
Gratifyingly for the researchers, this regimen had also shown the highest protective efficacy in a macaque study conducted by Dan Barouch at BIDMC. In 12 vaccinated animals given a series of low-dose challenges with the pathogenic SIV/HIV hybrid virus SHIVsf162p3, per-exposure risk of acquisition was reduced by 94% and, after a total of six challenges, 66% of the macaques remained uninfected. For comparison, the next-best regimen in the study used MVA vectors in the boosting phase and reduced per-exposure risk by 87%, with 42% of the group of 12 animals still uninfected after the completion of six challenges.
Performance of the regimens in the APPROACH trial was judged based on measures of HIV-specific immune responses that correlated with protective efficacy in the macaque studies. These included the magnitude of binding antibodies to the HIV Env protein and Env-specific T cell responses, as well as antibody-dependent cellular phagocytosis (ADCP, an antibody function that can mediate killing of virus-infected cells). After the final boost in APPROACH, 80% or more of participants displayed the requisite immune responses at levels above the desired magnitude. These results greatly exceeded the targets set for making the go/no go decision about proceeding to efficacy testing.
Results are still pending from one other preparatory clinical trial, TRAVERSE, which is comparing the trivalent mix of Ad26 vectors with a tetravalent version that adds an additional mosaic Env protein (Ad26.Mos25.Env). If all goes according to plan, the tetravalent combination will be used in the planned efficacy trial, HPX2008/HVTN 705, which aims to enroll 2,600 sexually active cisgender women aged 18-35 at sites in South Africa, Zambia, Zimbabwe, Malawi and Mozambique. Participants will be randomized 1:1 to receive either the active vaccine regimen or placebo.
Schuitemaker's IAS 2017 presentation is available on the conference YouTube channel (it's first in the Translational Vaccinology conference session). An excellent overview of Janssen's HIV vaccine development work was provided earlier this year by Maria Grazia Pau on an AVAC webinar, which is also available on YouTube.
An issue that has gone largely unmentioned in relation to Janssen's program is the lingering uncertainty regarding exactly what caused the enhanced risk of HIV infection observed in recipients of Merck's Ad5 vaccine vector. As discussed at a special summit convened by NIAID in 2013, there are some concerns that the same problem could arise with Ad vectors of other serotypes.
For the most part, theories to explain the outcome of the Merck trials have centered on the induction of HIV-specific CD4 T cells, which could have been vulnerable to HIV infection in the absence of any antibody responses (some limited data from a small macaque study supports this possibility). The antibody component of the Janssen approach, in addition to the use of an Ad vector from a different serotype, is likely to obviate any fears that this type of problem could arise.
But another suggested scenario involves an Ad vector-induced increase in the number of Ad-specific CD4 T cells engaged in immune surveillance of natural adenovirus infections at mucosal sites (particularly penile sites), which might be a concern with the Ad26 vector due to the extensive cross-reactivity of Ad-specific CD4 T cells.
In other words, the Ad26 vector could potentially induce Ad26-specific CD4 T cells capable of recognizing adenoviruses of other serotypes that recipients had encountered (or encountered during the trial) -- natural adenovirus infections have been found to be relatively common and potentially persistent, so increasing the number of Ad-specific CD4 T cells at these locations might conceivably provide additional targets for HIV. A similar mechanism has been proposed to explain the well-documented association between HSV-2 infection and increased risk of HIV acquisition.
Because the initial efficacy study involving the Ad26 vector is in women, there does not appear to be any clear cause to worry about enhanced susceptibility to HIV infection, as it was only observed in male recipients in extended follow up of participants in the Phambili trial of Merck's Ad5 vector. But, with trials likely to include men in the future, it would be prudent to conduct a review of the recommendations that emerged from the 2013 NIAID summit on Ad vector platforms and the status of relevant studies, to ensure that the safety of participants is maximized. A logical venue for this discussion might be a future meeting of the AIDS Vaccine Research Subcommittee (AVRS).
Richard Jefferys is project director at Treatment Action Group.
[Note from TheBody.com: This article was originally published by Treatment Action Group on Aug. 25, 2017. We have cross-posted it with their permission.]