Therapeutic AIDS vaccines

Are they feasible and is their development a separate endeavor from preventive vaccines?

By Simon Noble, Ph.D.*

Developing an AIDS vaccine that will prevent HIV from establishing infection in a healthy immune system is a daunting enough challenge. But some researchers are working on what is almost certainly an even more formidable undertaking—developing therapeutic vaccines that are intended to boost the immune response to HIV in people who are already infected. Given that the target of HIV, the immune system itself, is already compromised in these individuals and that established infection means high virus loads and diversity, as well as virus-infected cells, most AIDS vaccine researchers are highly skeptical that it will be feasible to manipulate the immune response in infected individuals to significantly improve their health status. Even so, many are convinced that there are still compelling reasons for trying to do so.

It is still far from clear in HIV infection precisely what are the “correlates of protection”; that is, the relative importance of the immune parameters (neutralizing antibody, CD4+ and CD8+ T cells, plus many others) that need to be stimulated to give a robust immune response against HIV. Many within AIDS vaccine research think that, beyond any potential benefit to infected individuals, investigations into therapeutic vaccination modalities are likely to give indicators as to what is required for a preventive vaccine response. “A therapeutic vaccine as an adjunct to HAART to counter re-emerging virus is a perfectly reasonable idea, but the task of a therapeutic vaccine is more difficult than a prophylactic one—the amount and diversity of virus it will have to counter initially is greater in a therapeutic setting,” says John Moore of Cornell University, “you might as well try and make a prophylactic vaccine because it would seem to be easier.” The requirement of HAART is an important qualifier, and he says it is difficult to “even contemplate using [a therapeutic vaccine] in the absence of virus suppression by HAART because you’re just making the [virus] escape even more inevitable.”

Progress to date

There has been a long history of therapeutic vaccine strategies that have proven ineffective at improving clinical outcome, although a number of studies have shown moderate positive effects on some HIV-specific immune responses. A whole inactivated HIV immunogen called Remune, made from whole HIV particles depleted of gp120, has been tested in combination with incomplete Freund’s adjuvant in a number of clinical trials involving HIV-infected patients receiving HAART. Cumulatively, results have indicated that this agent can induce some improvement in several HIV-specific immune responses, including CD4+ cell responses, but there has been no suggestion that these can affect patients’ disease progression. Similarly, recombinant gp160 and gp120 have been tested in a total of six Phase II efficacy trials in asymptomatic patients with early-stage HIV infection, again with no effect on clinical outcome.

More recently, some of the more promising immunogens that are being tested in preventive vaccine trials are now going into human trials in therapeutic settings. The Adult AIDS Clinical Trials Group (AACTG) focuses on therapeutic approaches generally, and now has a number of therapeutic vaccine trials planned or underway. One trial that begins next month will test Merck’s Ad5-gag adenovirus construct in 120 patients whose virus is well-contained by HAART in a randomized, placebo-controlled trial. These patients will be immunized, their HAART suspended and then they will be monitored to see how well virus replication can be controlled.

But it is still far from certain, given the biological unknowns, that an effective therapeutic AIDS vaccine can be developed. As Emilio Emini of IAVI says, “the nature of [HIV] infection is a balance between the virus and the immune system and the ongoing interaction between the two during the course of persistent infection. From a prophylactic vaccine perspective, the objective is to alter that balance clearly in favor of the immune response during the early stages of infection. Studies performed in monkeys, as well as anecdotal human observations, suggest that once that balance is established it’s very difficult to alter the equilibrium, even under a situation in which virus replication is being suppressed with antiviral chemotherapy. Therefore, the unknown with regard to therapeutic vaccination is whether or not the damage that’s caused to the immune system during the initial phase of HIV infection is essentially irreversible and unlikely to be enhanced by anti-HIV immunization.”


The initial goals of therapeutic vaccines will be to try to increase the durability of HAART regimens (that is, increase the time to virological failure when drug-resistant virus emerges) and to allow patients “drug holidays,” the off-therapy periods that mean patients can relax from the strict adherence that can compromise quality of life. If therapeutic vaccines prove particularly effective these goals will become more ambitious, with the removal of all ARV therapy being the ultimate, but presently very distant, goal.

This reliance on HAART means conducting therapeutic vaccine trials in developing countries would be extremely difficult right now. “I think this is a case where you really do have to do the first trials in settings where ARV drug treatment is freely available, so that you don’t induce volunteers to participate by the offer of free treatment—you have the full range of drug treatment available for people who participate in the trials just the same as if they hadn’t participated,” says Andrew McMichael of John Radcliffe Hospital, Oxford. “I think it would be very hard to set this up in many developing countries at this stage, but if [therapeutic vaccination] looked promising then it would be very important to do it in developing countries.”

Need for new interventions

Particularly in the last six months, a great deal of attention has centered on antiretroviral (ARV) treatment and sweeping pledges have come from many quarters to provide funds to improve global access to these powerful drugs. Many preventive AIDS vaccine trial sponsors have committed to ensure that ARVs will be made available to trial participants who become infected with HIV (see article on p. 1). This progress has raised hopes that some of the grave inequities in ARV access might finally be addressed. But McMichael still thinks that, if an effective therapeutic vaccine can be developed, “therapeutic vaccination may be one of the few hopes for effective therapy in developing countries, where it’s unlikely for a long time that lifelong treatment with ARVs is going to be available for everybody.”

Those fortunate enough to have access to these powerful ARV regimens can effectively manage disease progression, possibly for up to twenty years, but lifelong adherence is not easy and virus suppression comes at a price. Some patients experience severe adverse effects of therapy, including lipodystrophy, myocardial infarction and mitochondrial toxicity resulting in neuropathy and myopathy. And even when patients don’t suffer these toxicities, drug-resistant viruses often emerge which means they have to switch to alternative regimens. Eventually they may run out of options.

Alexandra Trkola of University Hospital Zurich works with patients who are controlling their virus with HAART and is familiar with the associated problems. She is investigating therapeutic approaches that require patients to come off their ARV regimens periodically. “Patients, sadly enough, are very keen in getting some off-time from taking drug therapy, some take every chance they can for a ‘drug holiday,’” she says. “Some patients don’t want to go back on therapy even though they are urged to because of the rise in viral load.”

Feasibility of therapeutic vaccines

There is some scientific evidence that the immune system of HIV infected individuals can be manipulated to improve their immune response to HIV, although it’s widely agreed that this evidence is only suggestive. A number of trials have looked at structured treatment interruption (STI) where patients who are receiving HAART to suppress their virus have their treatment periodically withdrawn under strict, often weekly, clinical monitoring. The rationale is that the HAART-induced suppression of virus allows the immune system to recuperate and then, when HAART is stopped, the virus load rebounds (the virus begins to replicate again) and it acts as an “auto-vaccination”—it is hoped that the cells of the immune system again encounter the virus and can respond to mount an improved immune response. Early studies enrolled patients who had initiated HAART during the primary acute phase of their HIV infection, and some patients could effectively control their virus load for prolonged periods after the suspension of the HAART regimen.

However, follow up studies in patients who had begun HAART during the chronic phase of infection did not derive any benefit from STI, and viral load set-points after the STI protocol were comparable to the initial set-points before HAART was first initiated. This suggests that preservation of the immune system (and also possibly suppression of viral diversity) by early intervention with ARV chemotherapy is crucial. The vast majority of patients do not begin their ARV therapy in the acute phase of infection, meaning that STI is unlikely to be a practical therapeutic approach to boost HIV-specific responses.

Since the ARV regimen in STI is intermittent and consequently sub-optimal, another concern raised is that this approach could lead more readily to the emergence of drug-resistant virus. But in a wider context, “what STI showed is that the immune system can help you even in HIV disease. If you can do that then it’s functionally possible to design a vaccine that could induce a more robust response and therefore be able to suppress viral replication below a certain level,” says Roger Pomerantz of Thomas Jefferson University.

One of the most promising strategies in preventive vaccine modalities in recent years has been the prime-boost approach. The prime vaccine, often a DNA plasmid encoding HIV epitopes, is given to trial participants and then, some weeks later, a boost vaccine is given, most usually an attenuated viral vector expressing HIV proteins. The scientific mechanism behind this approach is still not fully clear, but most immunologists think that the prime activates a broad repertoire of T-cell clones that the boost then selectively amplifies to give what is hoped will be a robust and appropriate immune response against HIV antigens. McMichael, a pioneer of this approach in preventive AIDS vaccines, is planning to use his preventive prime-boost vaccine in a therapeutic vaccine trial setting, and suggests that “it may be easier to boost the immune response in someone who is already well primed by natural infection than in somebody in whom you’re trying to use the vaccine to prime the immune response.” But he is keen to stress that he thinks both preventive and therapeutic vaccine approaches should be pursued.

Jeff Lifson of the AIDS Vaccine Program, SAIC Frederick, Inc., at the US National Cancer Institute also thinks there may be some reason to be optimistic. “The most important thing is that there are some indications that, at least under some circumstances, these viruses can be controlled by some individuals. The question is how are they doing that and is there a generalizable mechanism, and can we induce such a mechanism in a broader segment of the population?”

And he cautions against dismissing the potential of therapeutic vaccines too soon. “The [HIV] field can have a relatively short attention span, relative to the timeframes required to actually figure any of this stuff out, so an idea like therapeutic vaccination will go from heresy to dogma to common wisdom to rejected paradigm before anyone has really had chance to systematically explore the idea at the level of actual data.” But Lifson goes on to qualify any optimism. “My intuitive feeling, based on the cumulative available data, even if [a therapeutic vaccine] does work it is probably only going to work in a meaningful way for a very small subset of patients, presumably those who have been treated relatively early in the course of disease and have limited cumulative immunological damage.”

In studies that support this more cautious stance, patients whose virus loads were suppressed with HAART were given a standard vaccine, like diphtheria toxin, and their subsequent immune responses measured. The strength of the immune response to the standard vaccine correlated with the time interval between their CD4+ cell count nadir and the initiation of HAART. In other words, HAART must be initiated soon after CD4+ cell counts reach their lowest levels in order to preserve the immune system so that it can subsequently mount a good immune response. Again, like the STI studies, this suggests that there is a progressive negative effect on the immune system over time.

Lifson further points out that “emerging data on superinfections in individuals who seemed to be controlling their original virus suggests that immune control may not be as robust as we’d hoped. So certainly there may be limits on what might be achievable with therapeutic vaccination.”

Experimental approach

Trkola, together with Huldrych Günthard, is conducting a passive immunization trial at the University Hospital Zurich. Analogous passive immunization strategies have been tried before, most notably using HIV-specific immunoglobulin (HIVIG) obtained from chronically HIV-infected asymptomatic donors. In those studies, HIV-infected patients were infused with HIVIG and monitored, but again no clinical benefits were seen.

Trkola’s study is different in that it will use well-defined human monoclonal antibodies that have unusually high and broad neutralizing activity against HIV. Patients on HAART are given a cocktail of three neutralizing monoclonal antibodies (called 2F5, 2G12 and 4E10), HAART is then suspended and they are infused weekly with the cocktail, after which their virus load and CD4+ cell count are closely monitored. The goal is to look for improvement of clinical parameters in the individual patients, but it is also a proof of principle trial to try to determine the importance of humoral immunity in the management of disease progression. “We are trying to mimic a therapeutic vaccine that elicits neutralizing antibodies. Such a vaccine isn’t available so we have to use passive immunization,” she says. Although passive immunization is unlikely to ever be a feasible long-term treatment option (it’s far too costly to produce the antibodies), the outcome will suggest whether or not antibodies are a vital component of the immune response in infected individuals. Trkola is also keen to see if antibody-dependent cellular cytotoxicity is an important mechanism.

This typifies the approach that those engaged in therapeutic vaccine research are adopting; the primary goal of any therapeutic clinical trial must always be the amelioration of an individual’s disease, but if designed appropriately, these experimental approaches offer a secondary advantage: lessons might be learned regarding which immune parameters are important in an effective immune response against HIV. This experimental attitude is endorsed by Pomerantz: “I look on it like chicken soup—it may not help but if it couldn’t hurt, why not? It’s important to think how therapeutic vaccines could hurt people, because it might be worth setting up these [therapeutic vaccination] studies, as long as you’re not going to hurt anyone.” Harriet Robinson of Emory University agrees, “I think the most promising current preventives should be tested in the therapeutic arena and find out how they do and build on that for long-term control.”

“There is a school of thought that you could use HAART-suppressed individuals as a ‘vaccine test-bed,’ as a way of testing [preventive] vaccine efficacy in a small number of patients in a short period of time. You have HAART-suppressed individuals, you vaccinate them and then take them off HAART and you can get some idea of vaccine efficacy. It’s an idea that’s worth considering,” says Moore. He emphasizes an important qualifier; an improvement in an individual’s HIV-specific immune response would be an important result, both for the patient and for vaccine development, but a negative result would not rule out that a vaccine candidate might have better efficacy in a prophylactic setting where it would be directed at a pristine immune system.

The most sophisticated therapeutic vaccine clinical trials clearly illustrate that this experimental philosophy is at the fore. Nina Bhardwaj of New York University is taking dendritic cells from volunteers, incubating these cells with a mixture of HIV-derived peptides and then infusing them back into the volunteers. These dendritic cells are antigen presenting cells; their role is to “show” virus antigens in a particular context to lymphocytes, which subsequently go on to initiate a full immune response. Bhardwaj is conducting a trial that will compare the immune response induced in non-infected volunteers to that in infected patients who began their HAART early in infection. The primary goal is to augment HIV-specific immune responses (and ultimately improve the clinical outcome) in the infected patients, but she is also “teasing out the immune components that are important for inducing immunity—the longer term goal is to figure out how to activate these cells in situ, without having to pull them out of the body.” She is also conducting a two-arm study under the AACTG which is comparing the effects of therapeutic vaccination with canarypox AIDS vaccine vectors with or without a dendritic cell component.

Investigators are also looking at other novel strategies to try to augment any therapeutic vaccine-induced immune responses, and further innovations will probably be important. Cytokines are at the forefront now, but others like co-stimulatory signals may be beneficial, and again these approaches may well give indicators as to what’s required for a preventive vaccine response. Better understanding of the basic immunology governing these signals may increase their utility in the future.

Separate endeavor?

Beyond the challenges that come with an established infection and a compromised immune system, is there going to be any qualitative scientific distinction between an effective preventive and therapeutic vaccine? Lifson says that “according to immunological dogma, in a therapeutic vaccine setting antibodies may play a helpful role but cellular responses are probably going to be more important. And although we don’t know how to achieve it with any of the current prophylactic vaccine immunogens, the data [in monkey models] for passive immunization with monoclonal antibodies suggests that you can achieve sterilizing protection with antibody.” But he adds the caveat that “there may be scientific differences [between therapeutic and preventive vaccines] but still there are no compelling data that tell you what you ideally would want to control the virus in one setting or another.”

In the absence of hard evidence, should the two approaches really be seen as distinct? “There shouldn’t be two fields, we should be trying to make an effective prophylactic vaccine and if it works, therapy may be an additional use of it,” says Moore.

Emini agrees: “Given the present state of knowledge, it seems likely that no distinction will exist between vaccine-elicited immune responses that may be effective in either a prophylactic or a therapeutic setting. Of course, the two settings are not identical, but there is a high probability that studies performed in one of these settings will substantially inform the other.” He also thinks that, given our current understanding, there is “nothing qualitatively different regarding the development of the immunogen itself in terms of what you would do prophylactically or therapeutically.”

But some researchers give more weight to the suggestive evidence available so far, and think there might be important distinctions to be drawn between preventive and therapeutic vaccine approaches. McMichael emphasizes that a therapeutic vaccine will “need to stimulate T-cell immunity rather than antibody immunity, which is probably different from what you need to do for prophylactic immunity,” and points out that cell-mediated immunity “is a major player in controlling infection for 10 years in people who are untreated,” referring to the isolated cases of long-term non-progressors who are infected but show little sign of disease.

Robinson thinks that inducing CD4+ helper T cells will be the crux in distinguishing the two. “The difference is that the preventive [vaccine] is going to be able to use the normal mechanisms that a host has for providing CD4 help, but a therapeutic [vaccine] is going to have to use novel mechanisms for providing the CD4 help,” referring to the innovative strategies like cytokines that are still in early assessment.

A whole new paradigm

Most researchers are unconvinced that a highly effective therapeutic vaccine that will significantly improve HIV-infected patients’ disease progression can be developed. However, that doesn’t mean they think the effort will be fruitless. Given our lack of understanding of the basic biology of HIV disease, a therapeutic vaccine may perhaps turn out to be feasible. And it’s still not clear that the search for a preventive and a therapeutic vaccine are separate endeavors; lessons learned from one setting will most likely benefit the other.

Still, most researchers agree that developing a therapeutic AIDS vaccine is a much more difficult undertaking than a preventive one. As Pomerantz points out, “there is no such a thing as a therapeutic vaccine in any infectious disease, we’ve never made one. So we’re not only designing a new vaccine, we’re designing a whole new paradigm. It’s a tall order.”

*Simon Noble, Ph.D., is editor of the IAVI Report