The Odyssey of Therapeutic Vaccines for HIV
by Richard Jefferys
In the earliest days after the discovery of HIV in the mid-1980s, uncertainty reigned regarding how the immune system responded to the virus. Initially, it was thought that the time between HIV infection and the development of severe immuno- deficiency and disease represented a period of viral inactivity or latency. In this context, it seemed logical to propose that perhaps vaccination could be used to bolster the immune response to HIV and thus delay or even prevent the development of illness.
But the first efforts toward this goal quickly mired therapeutic vaccine research in controversy, casting an initial pall across the field that was compounded by the failure of any candidate to show significant efficacy. Additionally, the scientific rationale for the approach evolved as more was learned about the pathogenesis of HIV infection and the types of immune responses that may be effective—and ineffective—at controlling the virus. After a period in which enthusiasm regarding the prospects for therapeutic vaccines waned, the recent resurgence in interest in research aiming to cure HIV infection has offered new reasons to pursue their development.
The controversy that attended the earliest research into therapeutic vaccination began in the mid-1980s when a French scientist named Daniel Zagury obtained a vaccinia virus from the National Institutes of Health that had been modified to include several components from HIV, and proceeded to test it in both HIV-infected and -uninfected individuals in Paris and Zaire without appropriate regulatory approval (the vaccinia construct had been created only for the purpose of conducting studies in animals). Several of the HIV-infected participants died, and this fact was omitted from published reports about the experiments (which instead attempted to suggest the vaccine was efficacious).
Following quickly on the heels of this debacle were two more woeful contretemps relating to therapeutic HIV vaccine candidates. A company named MicroGeneSys created a vaccine containing the HIV gp160 protein, and Robert Redfield, a scientist with the Walter Reed Army Institute of Research, conducted trials in people with HIV. At the International AIDS Conference in Amsterdam in 1992, Redfield claimed the preliminary results were encouraging but quickly came under fire for overstating the findings. The situation was aggravated by a successful attempt to secure a $20 million congressional appropriation specifically to conduct an efficacy trial of the vaccine, bypassing normal research review mechanisms (this money was ultimately redirected after Redfield’s initial analysis was shown to be unreliable).
The International AIDS Conference in Berlin in 1993 was the site of the third blow to the credibility of therapeutic HIV vaccine research. A great deal of enthusiasm had attended Jonas Salk’s venture into the field in the late 1980s, when he described the development of a vaccine comprising a whole-killed HIV isolate that was intended to be tested as a preventive and therapeutic vaccine. Due to regulatory concerns about the safety of killed vaccines in HIV-negative individuals, Salk focused on therapeutic studies. Results were hotly anticipated and due to be presented in Berlin, but they were not debuted at the conference itself, but rather at a news conference; this decision fostered distrust and anger among attendees before the data were even described. The unimpressive outcomes of the trials, which Salk and the Immune Response Corporation (the company set up to produce the vaccine) tried to spin positively, served as the final insult.
As these disasters piled up, scientific advances were also undermining the original rationale for the approach. The notion that HIV was latent during the asymptomatic phase of the infection was overturned by data showing that the virus was constantly replicating, and that this replication was accompanied by the ongoing proliferation and death of CD4 T cells. Improvements in tools for evaluating immune responses revealed that there is a massive specific response to HIV that, in most individuals, is unable to control viral replication, leading to a situation where the immune system essentially flails away at the virus throughout the course of infection. Furthermore, CD4 T cells responding to HIV (HIV-specific CD4 T cells) were shown to be preferentially infected, contributing to their poor functionality and inability to deliver appropriate help to the other vital components of an antiviral immune response: CD8 T cells, whose primary task is to recognize and kill virus-infected cells, and B cells, which generate antibodies that—when effective—glom onto free floating viral particles and prevent them infecting new cells. These findings seriously called into question the idea that adding more HIV antigens into the mix via therapeutic vaccination— when the virus itself was failing to induce protective immunity—would be beneficial. While research did not entirely come to a halt, it was not viewed as a priority, and hopes for a successful therapeutic vaccine faded.
A Second Try
The burgeoning success of triple combinations of antiretroviral drugs (ART) in the mid-1990s might have been expected to further erode interest in therapeutic vaccines, but it ultimately led to a mild revival in interest, for two main reasons. Firstly, the drugs were clearly imperfect in terms of safety and side effects, leading to interest in approaches that might allow intermittent or delayed use of ART. Secondly, the profound suppression of HIV replication mediated by ART facilitated reconstitution of the immune system, and some scientists speculated that this may offer an opportunity to use vaccines to induce new HIV-specific immune responses that could develop (or “mature” in vaccine parlance) without interference from HIV because the drugs were keeping the virus at bay.
These ideas prompted a slew of new trials combining a variety of vaccine candidates with ART. These candidates included Salk’s whole-killed vaccine (now called Remune), attenuated viruses used as vectors to deliver HIV antigens (such as the canarypox-based ALVAC and cowpox-based MVA), and “naked DNA” constructs that deliver the genetic code for making vaccine antigens into cells. Data were generated showing that HIV-specific CD4 and CD8 T-cell responses could be induced in individuals with suppressed viral loads, and in some cases laboratory tests suggested that the functionality of these T-cell responses was markedly superior to those present prior to vaccination. But the harder question to answer was whether these apparent immunologic effects of therapeutic vaccines could be translated into a measurable health benefit.
Remune underwent testing in a large, randomized, placebo-controlled phase III trial that evaluated whether vaccination reduced morbidity and mortality in people with HIV, the vast majority of whom were on ART. No significant differences in the incidence of opportunistic infections or deaths were seen, but the interpretation of the results was complicated by the fact that the standard of care for ART evolved from dual- to triple therapy while the trial was ongoing, and that there were—happily—very few endpoints in both the Remune and placebo arms. Because the effectiveness of ART made it essentially impossible to demonstrate an additional benefit from therapeutic vaccination, alternative study designs became more common. There were two main approaches. The first was to immunize individuals on ART and then evaluate the effects on CD4 T-cell levels and viral load during an ART interruption (in hopes of allowing extended breaks from ART). The second was to administer therapeutic vaccines to individuals with early infection prior to ART initiation (in hopes of being able to show a delay in reaching CD4 thresholds indicating a need for ART). Data from these types of trials occasionally hinted that receipt of therapeutic vaccines was associated with better preservation of CD4 T-cell counts and slightly lower viral loads during ART interruptions, although at least one trial of ALVAC showed the opposite. A still-unpublished South African trial of a DNA vaccine suggested that it might have slightly delayed CD4 T-cell declines and the associated indication for ART.
New Dawn Fades
Once again, however, scientific advances served to undermine the rationale behind these studies. Specifically, the idea that ART could be safely interrupted as long as CD4 T-cell counts were maintained was shown to be erroneous by the sobering results of the Strategic Management of Antiretroviral Therapy (SMART) trial. SMART had the specific goal of assessing whether intermittent, CD4-guided ART could be as effective as continuous ART, but the trial had to be stopped early because individuals in the intermittent arm experienced a doubling in risk of illness and death. Analyses demonstrated that these events were associated with inflammation resulting from unsuppressed viral load, prompting additional investigations into the link between inflammatory markers, uncontrolled HIV replication, and health outcomes. This type of research has now been conducted in multiple cohorts in diverse global settings, and it has reinforced the conclusions from SMART: inflammatory markers are linked to viral load and show significant associations with morbidity and mortality; measures of cumulative exposure to viral load prior to ART initiation have also been shown to be associated with risk of morbidity and mortality after starting ART.
The window of opportunity for therapeutic HIV vaccines therefore narrowed once more, as it was clear that slight diminutions in viral load would be insufficient to offer benefit. Some therapeutic vaccine developers have unfortunately been slow to acknowledge this shift in the research landscape; for example, Bionor Pharma conducted a trial attempting to show that their candidate Vacc-4x could delay the need to restart ART after a six-month interruption, but the SMART results had already shown that this type of trial design was risky and outdated. The company has since conducted an analysis (not planned in the original trial design) looking at viral loads among study participants, claiming that vaccination was associated with a difference off therapy of around 1 log (22,300 vs. 61,900 copies). But it is known that a viral load of 22,300 copies likely poses long-term health risks and is not low enough to retard disease progression; furthermore, prior studies strongly suggest that the duration of such an effect is likely to be transient.
Third Time’s a Charm?
Although it will present a problem for the commercial development plans of some companies, it is clear that the bar for therapeutic vaccines has been raised. The key question has become: is it possible for a therapeutic vaccine to generate HIV-specific immune responses capable of completely containing viral replication when ART is interrupted? This may seem like a dauntingly high hurdle given results to date, but it dovetails with emerging research that has recently resurrected therapeutic HIV vaccines for the third time. This research is in pursuit of the ultimate goal: a cure for HIV infection.
Presentations at the 2012 Conference on Retroviruses and Opportunistic Infections (CROI) conspired to highlight this new rationale for therapeutic vaccines. A major focus of cure research is identifying and eliminating the reservoirs of HIV-infected cells that persist in the body despite ART (latently infected cells). For several years, scientists have been evaluating compounds that can awaken dormant HIV, but it has been unclear if this strategy will be sufficient to ensure that infected cells are killed. At CROI, Liang Shan from Robert Siliciano’s laboratory at Johns Hopkins presented compelling evidence that simply rousing HIV is not sufficient; CD8 T cells are needed to deliver the coup de grace and kill the infected cells. Shan showed that in most people with chronic HIV infection, HIV-specific CD8 T cells were not functional enough to accomplish the task, but required stimulation with HIV antigens prior to being mixed with infected CD4 T cells— essentially a laboratory dish equivalent of therapeutic vaccination. The study was published in the journal Immunity on March 8, 2012 and the authors are unequivocal about the implications, writing: “Our study strongly suggests that boosting CTL [CD8 T cell] responses through vaccination prior to virus reactivation may be essential for eradication of HIV-1 infection.”
There is another complementary reason for studying therapeutic HIV vaccines in the context of cure research. Studies have shown that a portion of the latently infected CD4 T cells that persist in the face of ART are specific for HIV antigens, suggesting that stimulation with a therapeutic vaccine might also reactivate the virus in these cells. A study of therapeutic vaccines in children with HIV has offered some support for this idea, as it uncovered evidence of a transient decline in the numbers of latently infected CD4 T cells during immunizations. An ongoing trial in adults—named Eramune 02—intends to explore this possibility in greater detail.
The Road Ahead
Despite the history of controversy and uncertainty, the ascendency of cure research has provided a strong and scientifically sound rationale for further studies of therapeutic HIV vaccines. The goals are now far clearer: to achieve containment of HIV replication and prevention of disease in the absence of ongoing treatment (now described as a “functional cure”), or complete elimination of the virus (a “sterilizing cure”). The first evaluations of therapeutic vaccines in this new context are getting underway, but significant questions remain to be answered, particularly in terms of delineating the ideal immune responses that should be induced and evaluating whether they can be effective and sustained. Researchers also need to explore which other antilatency strategies should be combined with therapeutic vaccines, and whether different vaccine candidates should themselves be combined to achieve the best results. There might even be a role for therapeutic vaccines in the context of gene-therapy approaches, as a means to boost numbers of gene-modified, HIV-specific CD4 T cells. While there is clearly some road ahead, there is at least a sense, finally, that therapeutic HIV vaccines are headed in the right direction. •