Looking into the Future of HIV Therapy
An interview with David A. Cooper, MD, DSc
Long before Europeans knew for sure that Australia existed, they intuited a hefty land mass lurking somewhere near the bottom of the globe. What else, they asked, could balance those many northern lands and keep the Earth from rolling over? The physics of these Western sages was decidedly pre-Newtonian, but their intuition was right on the money. Although the first European explorers to wet their hulls in the Indian Ocean managed to sail right past Australia, in the end it proved too big to miss.
To the boreocentric HIV research community, Australia may sometimes seem under the equator and out of mind. But its contributions to HIV research, like its mass, have proved too big to ignore. And the balance it adds to sometimes top-heavy northern ideas can yield a sustaining equilibrium. Many of those contributions-and much of that balance-come from David Cooper.
While working at the Dana-Farber Cancer Institute in 1983, he was part of the team that described the CD4+ receptor on helper T cells. Returning to Sydney, he hatched a plan that pinned the quicksilver seroconversion syndrome to the clinical map. Cooper was among the first to challenge primary infection with pharmacotherapy and continues to refine that strategy today. With his studies of acyclovir, he helped forge the link between antigenic stimulation and HIV disease progression. His early championing of combination therapy has been vindicated by massive clinical endpoint studies. And, lest the balance slip too dangerously toward antiretroviral solutions, he continues to assert the need for immune-based therapies.
Cooper is Professor of Medicine at the University of New South Wales, Director of the National Centre in HIV Epidemiology and Clinical Research, and Director of the HIV Medicine Unit at St. Vincent's Hospital in Sydney. The current President of the International AIDS Society, he also sits on the editorial boards of several international journals of HIV medicine and immunology, including this publication.
The Journal interviewed Cooper (twice, because of a balky tape recorder) during the November 16-19 conference of the Australasian Society for HIV Medicine.
Detecting and defusing primary HIV infectionJournal: How did you come to describe primary infection?
Cooper: From 1981 to 1983 I was working on human cellular immunology in Stuart Schlossman's lab at the Dana-Farber Cancer Institute in Boston. Some people may recall that that was the lab in which CD4+ was first described.1 At that time the epidemic was just being described in the United States, and of course I followed the news with interest.
As these events were unfolding, I knew that I was coming back to my previous appointment at St. Vincent's Hospital, which is situated in an inner-city area of Sydney. Within one kilometer of the hospital one way is the gay area of the city, and within one kilometer the other is the drugs and prostitution area. It wasn't too hard to predict that, if there was going to be a problem in Australia, it would be likely to show its head at St. Vincent's Hospital fairly promptly. I wasn't wrong.
When I got back we started a special immunology clinic. There weren't any cases of AIDS for quite a few months in 1983, but there were obviously things going on. Lots of people had lymphadenopathy, fatigue, skin rashes, and so on. So we persuaded the New South Wales Health Department to set up a small prospective study amongst gay men in Sydney to have a look at their clinical and immune status, study their sexual behavior, and save sera-the kinds of things that were interesting people at that time. In fact the New South Wales Health Department was very generous in funding that study in the absence of a single case of AIDS. We gathered together a group of inner-city general practitioners who had large gay practices and were really very expert in the treatment of sexually transmitted diseases amongst gay men. In those days confidentiality was a very big issue. In some states in Australia homosexuality was still outlawed, although things were changing. So gay men tended to go to these private practices that specialized in STDs, and at that time the rate of STDs amongst gay men was quite extraordinary.
These colleagues turned out to be very astute clinicians, as STD docs often are. They started observing cases of acute febrile illnesses-and one particular recurring story: It was very popular for Australian gay men to vacation in the United States. The typical vacation was a week in California, either in San Francisco or Los Angeles, where much time was spent in the bath houses, followed by a week on the beach somewhere in Hawaii on the way home. When these men came back home they complained to the GP that, instead of spending the week in Hawaii on the beach recovering from the previous week in the bath houses, they spent it in the hotel room absolutely wiped out with fever, profound fatigue, and a rash. So that was our first clue. From there, it wasn't terribly difficult once we got the serological test, because we had saved the sera from the prospective study and were able to demonstrate seroconversion associated with these illnesses. It was Dr. Jay Levy from UCSF [University of California, San Francisco] who actually gave us the assay-the immunofluorescence assay, not an ELISA assay.
So those were the keys: realizing that there was no AIDS around but that there was something happening, meticulous clinical observations by astute STD docs, being fortunate enough to be given the serological test by Jay Levy. And we published our findings in the Lancet in 1985.2
Journal: Now primary infection is a target for clinical study. What percentage of people do you think can be identified during primary infection and so take advantage of any benefit that might come from such early therapy?
Cooper: In a retrospective study that compared cases and controls for seroconversion illness,3 we showed that at least 50 percent and probably up to 90 percent of people who seroconvert have a seroconversion illness. Because the illness is often nonspecific, it's often missed. Clinical investigators usually see only people at the more severe end of the spectrum, when they're hospitalized with meningitis, encephalitis, or very severe fever and prostration. But we believe that most people have an illness. In Sydney, for example, amongst GPs who care for many gay men, the most common cause of a mononucleosis-like illness is HIV seroconversion. In our setting there is a very high index of suspicion, and we provide these primary care docs with excellent laboratory backup to diagnose people quickly.
It's interesting to speculate, based on what we know now, that the very rapid early spread of the epidemic in some cities with large gay populations may have been due to sexual contact with men who were in the seroconversion phase or very early postseroconversion and may have had extremely high viral loads and therefore may have been extremely infectious. That-and of course the number of sexual contacts-may explain the extraordinarily rapid spread of HIV amongst gay communities in some cities that now have high case loads: Amsterdam, Sydney, Los Angeles, San Francisco, and New York.
Journal: What is the most that can be gained from an effective antiretroviral combination during primary infection?
Cooper: With Luc Perrin and his colleagues in Geneva,4we studied zidovudine versus placebo in primary infection. Although the benefits in that study were very modest, as has been the case in all studies of zidovudine monotherapy, we showed very clearly that seroconversion studies can be carried out well in a proper randomized trial. People can be identified at that stage of infection and can be recruited into clinical trials.
What we have to do now is try to improve on the clinical results, just as we have been making progress with combination therapies in other stages of HIV disease. Given that we can identify these populations and randomize them in trials, we should look at fairly aggressive triple and quadruple combinations to see if we can reduce viral load when viral diversity is not great and perhaps permit the immune system to kick in and abort the viral infection. This is the concept of ablative therapy proposed by Marty Hirsch and Luc Perrin at the Lisbon combination therapy conference. [See the September 1995 issue of the Journal.]
Journal: So you're saying there's a possibility that therapy for primary infection could cure a person.
Cooper: I think the probability of that is low, even though some American investigators are calling it a home-run strategy. But, given what we now understand about pathogenesis, the concept must be tested. And I don't believe it will harm any patients. We've already set up protocols and begun enrolling patients with our European and American collaborators. This multicenter study, sponsored by Merck, will look at the triple drug combination of the protease inhibitor indinavir plus AZT and 3TC versus AZT plus 3TC versus indinavir alone in 30 to 40 people with primary infection. [At the Aaron Diamond AIDS Research Center, AZT/3TC is also being combined with Abbott's ritonavir in a primary infection trial.]
Combination therapy: Take a big first biteJournal: For the people who can't take advantage of whatever benefits such therapy might provide because their infection is detected later, what's going to be the best antiretroviral strategy?
Cooper: I think there's no doubt that it's going to be combination therapy. It's going to be three or four drugs, and I think we are going to be able to identify combinations that will give very worthwhile benefits compared with the very limited benefits of monotherapy or even, for that matter, double nucleoside therapy as in Delta and ACTG 175. The questions that remain are when to intervene and how much benefit we can derive from the drugs that we have now. Having about a dozen drugs certainly gives us a lot of tools to explore those questions.
For the vast majority of asymptomatic HIV-infected persons, I think that it's still very unclear when we should intervene. There may be people whose immune systems are doing a very good antiretroviral job-keeping viral loads low and placing them in the long-term survivor category. Clearly, they may not need immediate treatment. One may need just to watch them carefully, and viral load monitoring is one way to do that. But I think the key will be to try to identify the moment to intervene before the repertoire of the immune system is irreversibly damaged. And to do that, in addition to viral load assays, we're going to need better immunologic tools. Some of the data that Martyn French presented at this meeting on the CMI multitest, which is the only in vivo test of cell-mediated immunity that we have, could play a role. [See the article on the Australian meeting in this issue on page 6.] Those kinds of tools, in addition to CD4+ cell counts, might be useful ways of detecting deteri oration of the immune system before it's too late.
Journal: Once people reach that stage where they need to start antiretroviral therapy, are they going to need different combinations over time? Is the one triple or quadruple combination they start with eventually going to lose its punch and require them to move on to something else?
Cooper: If you look at the principles of other infectious diseases or even malignancies, the first bite's the best. I would predict that we're going to find the same thing in HIV. So I think it's critical, as Joep Lange and David Ho have said, "-in other words, to use a really potent combination that's tolerable and has a durable effect.
But we also have to look beyond that first bite. Patients, I find, don't have any goal orientations in their HIV treatment programs, and that makes compliance extremely difficult. Let's say that we had antiretroviral strategies that started with four drugs for a fixed period-for example, four to six months to maximize viral load reduction-and then slipped back to a very tolerable two-drug maintenance of pills that need to be taken only once or at most twice a day. I think you would find overall that the compliance would be better and that you may be able to maintain the low viral loads you've established with an aggressive four-drug combination. If patients saw this as their goal-perhaps feeling a bit worse during that first period but getting over it-I think we might achieve a lot more.
Journal: You've been pretty outspoken in criticizing sequential monotherapy. But if someone today is looking at the ACTG 175 results, they might conclude that ddI monotherapy held its own against nucleoside combinations. Or they might look at some protease inhibitor studies and figure that taking ritonavir or indinavir alone might work pretty well, cause fewer side effects than a combination, and cost less too. How would you answer arguments like that?
Cooper: I stand by my view that sequential monotherapy is a very impoverished strategy. I think that the benefits of ddI in ACTG 175 are probably related to what we know about ddI-that a switch to ddI after AZT use or failure gives a worthwhile modest benefit. But I think we can do better. So I would say that if you're going to change from AZT, at least change to two or three new drugs simultaneously. That won't be as good as if you started with a combination, but it will be much more worthwhile than just having monotherapy 3TC or monotherapy saquinavir or monotherapy indinavir or whatever.
I really feel that these expanded access programs are not serving anyone any useful purpose-except for the manufacturers themselves. The community is putting pressure on the pharmaceutical industry to have these drugs before they are approved, and they tend to be used by patients with very advanced disease who have no other option. They clamor for the drugs, but the benefit in general is negligible compared with the benefits they might get at that stage from aggressive OI prophylaxis. Expanded access becomes a marketing exercise for the pharmaceutical industry under the guise of allowing them to evaluate the safety of their drug in huge numbers of people, particularly in those with advanced disease, which is the most worrisome area for safety. It's a very perverse way of making drugs available, and it's getting worse. I think what will happen is a form of therapeutic anarchy, where everyone will use something different, getting hold of whatever they can through these expanded access programs. As a result, I'm afraid we won't find the best ways to use combinations.
Hopes for restoring the immune repertoire
Journal: Some people are now starting therapy with a combination regimen but at a lower CD4+ count-200 or fewer CD4+ cells. If they respond with an increase in CD4+ cells, are those new CD4+ cells going to be functional?
Cooper: Whatever we do in terms of early diagnosis and treatment of HIV infection, there will always be people who, for whatever reason, present relatively late. I think we must find a treatment strategy that is valid for such people, and there are some interesting data that are emerging on that. For example, in the phase I/II study of ritonavir in Europe and Australia,5 we looked at the recovery of immunity in fairly advanced patients receiving ritonavir monotherapy.6 And there are very impressive returns of immunity, both of CD4+ and CD8 cells, and in functional assays that return is quite impressive. I think the data that were presented by Judy Falloon of Cliff Lane's NIAID group at ICAAC,7 showing the possibility of actually rescuing advanced patients with a combination of the protease inhibitor indinavir and interleukin 2 [IL-2], are extremely intriguing and need to be strongly pursued.
Even if the immune repertoire is greatly damaged at that late stage, we do know that, in patients who have had total nodal irradiation for Hodgkin's disease or patients who have had transplants and are heavily immunosuppressed, their repertoire is depleted but nevertheless they can carry on with a pretty reasonable quality of life. Analogously, in HIV we may not be able to fully restore the repertoire, but we may be able to keep such patients going for very worthwhile periods of time.
Journal: Mario Roederer at Stanford 8 showed that there's a predominance of memory CD8+ cells as infection progresses-and a shortage of naive cells that can learn to respond to new antigens. Do you see that as a concern?
Cooper: Yes. I think his data are good. What's encouraging, however, is that we've seen a return of both memory and naive cells, particularly CD4+ cells, in our ritonavir studies.5,6 They have an apparently different rate of return in that the memory cells come back fairly quickly, while the naive cells seem to have a delay of a few weeks in returning. I think the markers for naive and memory cells have had immunologists somewhat flummoxed for many years because there can be switches back and forward between naive and memory phenotypes. But nevertheless I think it does give us some encouragement that we may be able to have some restoration of those functions.
Immune-based therapies: dreadnoughts and dingosJournal: Where do you see immune-based therapies fitting into the strategy for people with HIV infection? What approaches do you think are worth pursuing?
Cooper: This is a viral disease of the immune system. The approach in the long term will be antiretroviral therapy plus immunotherapy. Antiretroviral therapy seems to be demonstrating a better effect at the moment mainly because more agents have been better investigated. A lot more work needs to be done on immune-based therapies. To date, the work of Cliff Lane and his colleagues with IL-2 is extremely important and intriguing. Now we have to accumulate more data showing that the CD4+ cell increases seen in these patients are viable increases and not just expansions of the few remaining cells that don't work. I think that the data are sufficiently strong that we ought to be exploring phase III trials looking at clinical endpoints for this kind of therapy.
Although I believe that additional clinical endpoint trials are probably not necessary for the further development of nucleoside analogs and probably won't be necessary for protease inhibitors once the large clinical endpoint studies for those drugs are finished and show a benefit, I do share the belief that we will need clinical endpoint studies to prove the principle of immunotherapy. And I think that IL-2 is really one of the strong candidates for clinical endpoint studies. If you look at the mini-meta-analysis that Joe Kovacs presented at the ICAAC meeting,9 comparing the three randomized studies of intravenous IL-2 versus standard-of-care controls, it's very impressive that there are more clinical endpoints in the control groups compared with the interleukin-2-treated groups. That result must be interpreted cautiously, however, because these studies were not powered to compare clinical endpoints in the two groups. Nevertheless, the endpoints in the patients who were receiving IL-2 are not occurring at high CD4+ cell counts. They're occurring at very low CD4+ cell counts in people who have failed to respond to interleukin therapy. So I think it's extremely important that we push on and show definitely whether this therapy is going to be part of our arsenal against HIV.
Journal: Are there other immunotherapeutic strategies that look intriguing to you? And which just look like dogs?
Cooper: We can do the dogs first, if you like, or the dingos as we would call them in Australia. I think that therapeutic vaccination, although initially attractive, is probably going to turn out to be a waste of time. The pathogenesis research of the last year has pretty clearly put that strategy to bed. It seems extremely unlikely that a tiny amount of antigen would ever do anything when enormous amounts of antigen are being presented to the immune system every day. So I think that we need to make that one disappear very quickly. There is a danger that this approach may be exploited in developing countries, where it may appear attractive because it costs less than antiretroviral therapy.
The studies in which CD8 cells are cloned, expanded, and reinfused to try to achieve antiviral effects are important because they will help us understand pathogenesis as well as give particular patient groups who are keen on therapy a possible benefit. But I can't see such an approach being translated into routine clinical practice.
On the positive side, I think that some of the immune activation blockers-things like corticosteroids, cyclosporine analogs, and some of the cytokine inhibitors such as thalidomide and monoclonal antibodies to tumor necrosis factor-are really very intriguing, and the preliminary data look encouraging. It's important to explore those approaches fairly aggressively.
Third-world strategies that can work right nowJournal: If you were directing policy for UN AIDS [formerly WHO's Global Programme on AIDS], and all the countries in the world agreed that they would let you set research priorities for developing countries, what priorities would you rank highest?
Cooper: If I were to consider only those interventions that are based on biomedical models, there are several things that would be relatively simple and relatively cheap and that deserve a lot of attention. The first is the prevention of tuberculosis in HIV-infected populations because that may in fact improve survival. Another would be reducing the general incidence of sexually transmitted disease. I think the study published in the Lancet this year,10,11 showing a reduced incidence of HIV when STDs were controlled, is extremely important because that's a relatively inexpensive and simple intervention.
The perinatal transmission study of combination AZT/3TC that is being done by UN AIDS to see if short-course regimens will work in developing countries, particularly regimens that could be given during labor and delivery and particularly in breast-feeding populations, are very important studies. Clearly the ACTG 076 model, although proof of principle, cannot be translated into a developing country because it requires therapy for several months before delivery as well as intravenous therapy during labor, which is obviously impractical in the developing world.
I think that research on female-controlled methods of prevention is an extremely high priority in developing countries. In most developing countries women lack empowerment, so it would be an enormous benefit if those women could prevent HIV transmission with something they control themselves. I'm thinking of vaginal microbicides, cheap female condoms that are less imperfect than those now on the market, perhaps postexposure regimes with non-nucleoside-type drugs-I think there are a lot of possibilities that could be explored. From what I understand, those are close to the current priorities of UN AIDS.
Acyclovir and survival: moving beyond the controversyJournal: You were a key investigator in two controlled trials that showed a survival advantage for people taking acyclovir. But other studies-controlled and observational-have come up with conflicting results see Table 1. With more powerful antiretroviral therapies emerging, do you think acyclovir is going to have a role beyond routine antiherpes indications?
Cooper: We did two randomized studies which clearly showed a survival benefit. Both studies have been validly criticized for minor flaws, the first one12 because the acyclovir group was slightly healthier than the placebo group. A randomized study is a randomized study, but randomization doesn't always give you perfection. In the second study, the UK and Australian study,13 the problem was that CMV [cytomegaloviral] disease was the primary endpoint and survival was secondary. That study was stopped because of a lack of benefit on CMV disease, and then the survival benefit became apparent.
Having said that, I would argue that it's hard to fluke it twice. I think there's something more to the study results than chance. The MACS [Multicenter AIDS Cohort Study] analysis in the United States14supports that; another cohort study in the United States15 doesn't support it. But their methodologies were somewhat different, and that may explain the differences in their results. The acyclovir study in early disease sponsored by Burroughs Wellcome16 was unable to show clinical endpoints because the patients were early in their disease process. And the ACTG 063 investigators said themselves that their trial17 was full of difficulties from the beginning. It was an unpopular study because other trials were looking at more interesting double nucleosides, it was slow to recruit and had a high dropout rate, and the follow-up was difficult.
So where do we go? I think that ACTG 204 may be helpful because it looked at low-dose acyclovir, high-dose acyclovir, and high-dose valaciclovir. [See the November 1995 issue of the Journal, page 27.] The ACTG 204 virologists did assays to quantitate the various herpesviruses that may be cofactors, ranging from herpes simplex, Epstein-Barr virus, and varicella zoster, to HHV-6, -7, and -8. So we may be able to dissect out which viruses are affected by acyclovir and develop a future strategy around this.
Can we go on and do another placebo-controlled trial of acyclovir? I doubt it, because many patients in this population will use acyclovir anyway for herpes prophylaxis. I think the only place we can really go is to have a look at some of the new drugs that are very intriguing-like the PMEA group. These seem to have a very broad spectrum of antiviral activity-including anti-HIV, antiherpes, anti-hepatitis B-and see if we can use those kinds of drugs in an advanced-disease population to look at survival. They're not the most potent agents against HIV that we have but nevertheless have good activity and appear to be safe. I think we'll see those studies evolving over the next year or two.
Journal: In the meantime, should clinicians be talking to their patients about using acyclovir routinely now?
Cooper: In Australia practitioners do by and large use acyclovir. They feel comfortable with the lower doses [0.8 to 1.6 g daily]-even though the original studies evaluated 3.2 g a day-simply because the MACS study showed the survival benefit with the standard antiherpes doses. Clinicians in other countries are, I believe, more skeptical about the positive results but will nevertheless use acyclovir for reasonable herpes indications in people with HIV disease. It's unfortunately another "watch this space" situation. But I think emerging data will help solve this continuing controversy in HIV disease.
See interview for circumstances in each study that may have confounded the results
Mark Mascolini is a freelance writer specializing in HIV infection.
This article was provided by International Association of Physicians in AIDS Care. It is a part of the publication Journal of the International Association of Physicians in AIDS Care.