February 11, 2009
While at CROI 2009, I had the chance to talk with Mitchell Warren, the executive director of the AIDS Vaccine Advocacy Coalition (AVAC). We discussed what's new in the field of HIV vaccine development.
Can you tell us where we are in 2009 in terms of HIV vaccine research?
Where we are now is actually quite an interesting moment in the field. I can't help sitting at CROI 2009 and not think about CROI 2008. At CROI 2008 was the official presentation of the data from the STEP study, which was a big vaccine trial of a candidate vaccine for Merck.1,2 It was a trial that had a result that was disappointing and very surprising, I think, to most people.
In addition, there were two plenary presentations at CROI '08 that basically asked, "Has the field lost its way? Did these results show us that we actually know what we're doing?"3,4 After CROI, there was a big vaccine summit at the NIH [U.S. National Institutes of Health] and lots of hand-wringing in the field. [Click here to view a webcast of the Summit on HIV Vaccine Research and Development.]
Could you remind us about the results of the STEP study?
The STEP study was a 3,000-person study in the United States, Latin America and Australia, testing a candidate vaccine called adeno 5 [adenovirus type 5] made by Merck; it's sometimes called MRK Ad5.2 Adeno 5 is a basic cold virus. It had specific inserts of HIV -- pieces of HIV that don't themselves cause AIDS, can't cause AIDS, but that are used to prime the immune system in case it does encounter infection.
In one of the interim analyses by the Data Safety and Monitoring Board, they found first that it was futile to continue -- that based on the interim analysis, the trial couldn't possibly prove efficacy. [Click here to read an update to the U.S. National Institute of Allergy and Infectious Diseases (NIAID)'s official statement on the halting of the trials.] Shortly thereafter there was an additional data analysis, and it showed a very troubling finding: It seemed that the vaccine increased susceptibility to HIV amongst a specific subset of individuals.
That subset has been sliced and diced. It's looking very much like people who had preexisting exposure to adeno 5 and who were uncircumcised men -- in that subgroup, they seemed to have this increased risk of HIV infection. People were surprised that it didn't work at all, because this was seen from early animal studies and earlier phases of human trials as one of the more promising vaccine candidates. The fact that it didn't work was a setback. The fact that it potentially caused increased risk to some people is obviously an even bigger setback.
That data are continually being analyzed, and that is really interesting, because people talked about it being a failed trial. The trial didn't fail. We said this a year ago -- we say this today in '09 -- that trial didn't fail. Obviously, the product failed. The Merck Ad5 product didn't work. But the trial is continuing to give us answers.
Interestingly, coming out of CROI last year, people said, "We should go back to basic science. We should stop human trials, because this is a waste of time." That was the tone of the conversation, but a lot of us were advocating even then. You can't just abandon human clinical trials, because there are data that can only come from clinical human trials.
There was a big debate over the last year, so what was really wonderful at CROI this year is that there were data presented from the samples of participants in the STEP trial that are beginning to help answer some really specific and interesting questions about the immune response and about HIV.5
There was one presentation that was perhaps the first time that someone from the Microsoft Corporation presented at a CROI: A man named David Heckerman was presenting data that he's been working on with Bruce Walker and others, looking at what are good epitopes and potentially bad epitopes.6
Could you explain what an epitope is?
There's a lot of talk about epitope mapping in HIV and in the immune response. Basically, you begin to get that different epitopes exist; some may actually be good in terms of helping the immune response, and some may be bad. We don't know which are which. So being able to map that out is really important. While the data are just a mini-step in understanding, what's important is: That data couldn't come from an animal study. That data could not come from the test tube. That data can only come from human trials. I don't want to oversell the importance of STEP2, but I do want to make it clear that that's one little bit of the finding.
One of the other pieces is: A lot of specimens are collected in a vaccine trial, and those specimens really begin to answer a lot of other questions too. I've not seen data presented yet, but there's a lot of work looking at HSV [herpes simplex virus] in the participants in the STEP trial. What was the interaction? What was the effect? Again, things that you could only find in a clinical trial.
There's a lot of debate about basic science, clinical science. We need to understand that we can answer some basic science questions through clinical research and clinical investigation. That was a real big take-home message here.
The other big issue that's challenging the vaccine field is that, in addition to Merck developing their Ad5, there was a number of other vaccine candidates that were being developed based on adeno. So you have this response in the STEP trial, and then you have a number of other candidates that were variations on the theme. In fact, just before the STEP trial was stopped, there were plans to start an even larger trial of another vaccine that was based on Ad5, but in a combination with a DNA vaccine. That trial was put on hold. It's now been redesigned and made much smaller. It's being changed dramatically because of the STEP trial. That trial will start sometime this year presumably.
So there's certainly been a delay in this, because it would be unethical and just bad science to not take the best information you can from the current trials to inform future trials. One of the things about CROI this year is that there's not a lot of new information on HIV vaccine candidates in clinical trials, instead there's a lot more around the basic science.
But the STEP trial wasn't the only vaccine trial ongoing at that time, was it?
That's a great question, because we focus a lot on STEP. There were, at the time -- and there still are -- almost 30 other clinical trials going on. Now, most of those, actually 29 of them, are small phase 1 and phase 2 trials. But I think what we mustn't forget, sitting here in 2009, is that the world's largest HIV vaccine trial ever is just finishing up. It's in Thailand. It's called the Thai Prime-Boost trial, and it's testing two vaccines in combination, priming and boosting: one from Sanofi Pasteur, a canarypox vaccine, and the other made by VaxGen, called AIDSVAX.
Now, people may remember, the only candidate before the STEP trial to have ever been tested in an efficacy trial was the VaxGen product, AIDSVAX.7 [Click here to read an article outlining the halting of the first phase III HIV vaccine trial.] It didn't work in either of two efficacy trials. But this trial was designed with the concept that you're priming and boosting, and that perhaps you might get benefit from the VaxGen product when it's combined with the Sanofi product. We don't know if it's going to work, but over 16,000 Thais enrolled in the trial and there has been an incredible retention of participants throughout the trial. That result is probably going to come out sometime around October of this year.
Were there interim results?
There were not. All we know is that it's had a number of interim analyses by the independent monitoring board. It's not been stopped for any safety concerns.
In prevention research, we've had a hard couple of years. In addition to the STEP trial, there have been a number of microbicide candidates for which trials were stopped early because Data Safety and Monitoring Boards have seen trends toward harm.8 So due to the fact that the Thai trial hasn't been stopped, we know that there are no safety issues right now. But we have no sense of whether it's going to work.
Conventionally, we create vaccines to prevent a person from getting infected. In addition to safety, every vaccine trial always looks to see if the study vaccine prevents infection. But the Merck trial and the Thai trial are looking also at the impact of their vaccines on viral load, on the assumption that their vaccines may help control the virus in someone who is already infected. Delayed disease, or perhaps delayed time until a patient has to go on treatment -- there may be benefit in that regard.
We don't know if that's going to work in the trial. And even if it does, we don't know how that will necessarily translate into clinical benefit.
The people in Thailand who are on the trial, they're not asked to leave the trial if they test positive? I imagine they're being tested along the way.
They're testing along the way. Different trials do it different ways, but normally it's every month, every three months or every six months. Everyone starts the trial HIV negative. These are preventive vaccine trials, which is important to note because there actually are some therapeutic vaccine trials, but right now all the vaccines we're talking about are preventive vaccines.
So, people coming into the trial are HIV negative. They are given not only vaccine or placebo, but they're also given state-of-the-art counseling, access to condoms, all of the prevention package that we have. They're then monitored very closely for their risk behaviors -- they're counseled to always use condoms and safer sex practices.
Tests are made. People who are HIV positive continue in the trial and are monitored very closely, because these trials are looking at this issue of the impact on viral load. The researchers want to establish what the set point was and then see if perhaps the vaccine actually controlled viremia more than placebo. Again, we don't know how that would translate over long-term.
In fact, the now smaller, follow-up trial to STEP of the DNA/Ad5 prime-boost vaccine that is being developed by the NIH's vaccine research center is likely not to have prevention as an endpoint. It's specifically looking at controlling viral load.
Again, I think we need to just take a step back. There are two parts to the immune system: the cellular immune response and the humoral immune response. Most vaccines focus on antibody responses, or the humoral immune response. Right now, the vaccines we're talking about in HIV research primarily look at cellular immune responses. They may not prevent infection, but they may modulate the disease progression. That's going to be a very interesting thing, even if we get a positive result out of a trial.
To most people -- on-the-street policy makers, community members -- the idea of a vaccine is, "Oh, I got the vaccine. I'm safe." We may be talking about very different kinds of vaccines. They'll have great benefit to people, but we're going to be really challenged to figure out how to explain it to people. These are not the conventional ways of vaccines -- and that's not just unique to HIV. There are a lot of other vaccines in development looking at T cell-based responses. But it definitely challenges us and our communities. Certainly at AVAC, we focus on how we engage communities in the search for a vaccine. It's a re-education for a lot of us in terms of what a vaccine could do and how we describe it.
Due to the results of the STEP trial and the publicity surrounding it, has it become harder to recruit people for vaccine trials?
We don't know yet. In fact, the DNA/Ad5 trial is going to take place only in the United States, only among men who have sex with men [MSM]. They will need to be circumcised and not have been exposed to adeno 5. Again, the STEP trial taught us that that particular vaccine did increase risk for people who were not circumcised and had preexisting Ad5 exposure. So to do the next trial ethically, we need to ensure it is what we call equipoised -- that it adequately balances the risks and benefits to the individuals. You can't really enroll an uncircumcised, Ad5-positive person, because we know from the science that they may be at risk.
We don't know that this same vaccine would do that. It's a different vaccine, but because it's based on the same adeno 5, there's a reason to be worried. So the trial is designed in a very specific way. I think it is going to be a complicated trial to explain, and I think it could be complicated to enroll patients.
But I will say this: There was another trial in addition to STEP called Phambili, which tested the same Merck product in South Africa.9 It had started much later than STEP, and it was stopped very early because of the STEP result. Both the STEP trial and the Phambili trial unblinded everybody right upon knowledge that this trial potentially was causing harm. People were told whether they got the placebo or the vaccine. I was actually in South Africa last year where they were unblinding participants. The most amazing thing we heard -- and we heard this in the U.S. as well -- at the STEP trial sites was that individuals were so grateful that they were in the placebo arm. And when asked why, it wasn't because they were worried about the increased risk. They were pleased they were in the placebo arm because they were asking the trial staff: "When's the next trial?"
I think we need to understand: Why do people participate in clinical trials? Certainly, in treatment trials, a lot of it is about the fact that you not only want to help the world to develop better therapies, but you're also looking for self-benefit in therapy. In vaccines, certainly in these trials, and sitting in 2008, 2009, we're not thinking right now that these vaccines are going to be shown to work and then we're going to license them. We don't know that they have great benefit to the individual yet. But people participate in prevention vaccine trials largely because they want to be part of something that helps lead to the end of the epidemic. What's remarkable is -- even in the midst of negative press of scientists last year at CROI saying, "We lost our way," and "The sky is falling" -- the actual participant said, "I want to be in the next trial." So I think it's going to be a hard trial to explain, potentially a hard trial to recruit for, but I'm relatively hopeful.
There's a whole debate in the field, in the science, that we can't afford another trial because if it doesn't go well, people are going to start losing faith. I actually disagree with that. I believe that communities will support research and individuals will participate in trials as long as the research community is honestly talking about what we're doing. As long as we explain that there are risks; as long as we explain what happened in STEP; as long as we engage in a conversation, I'm convinced that communities will participate in the research, because they've got the most to benefit from it.
So for me, it's a question of how we honestly communicate with communities, not that we should not put products in trials because we don't know what the number of trials is going to be. We have a long road to go in vaccines. And I think people are willing to stay on that road as long as we're talking to them throughout the entire journey, so to speak.
Just to put things in perspective: How many other viruses have vaccines, and how long did it take to create those vaccines?
It's interesting. We in AIDS vaccines, I think we sometimes suffer from magical thinking. If you go back to history, there's only one vaccine that was developed in less than 25 years, between when you identified the cause of the disease, the pathogen, and the vaccine that prevented it. That was for hepatitis B.10 Everything else took longer than 25 years. In fact, sometimes we're talking hundreds of years. With respect to malaria and TB [tuberculosis], the causes of which we discovered long before we discovered HIV, we still don't have vaccines for malaria and there's a need to develop better vaccines for TB. So this idea that we're going to develop a vaccine overnight is unrealistic.
But one of the reasons we got in that place is that hepatitis B vaccines were developed in the '80s, right around the time of HIV. The early vaccine candidates, in AIDS vaccine trials, were based entirely on the concept of the hepatitis B vaccine: taking pieces of the outer envelope of the virus. That was the early thinking around what was called the gp120 vaccine, which actually is kind of where we are with the VaxGen product that's still in this big Thai trial.
A lot of the early enthusiasm for AIDS vaccines was, "Well, we did hep B in 16 years. We can do HIV." What we're finding, and what HIV vaccine research has taught us, is that this virus is unbelievably complicated in ways that we continually learn about.
We now know more about this virus than almost any other virus in the world, after 26 or 27 years of research, but a lot of that knowledge is telling us how hard it is. The virus mutates very rapidly. There are lots of different circulating subtypes. Even at CROI this year, we're seeing more and more information about the acute infection phase, and learning that -- whether we're talking about a vaccine, a microbicide or a pre-exposure prophylaxis [PrEP] pill -- we have a very limited window of when we can intervene to stop infection.11 It's literally within hours or days. It's not like you have a lot of time once you get exposed to the virus. That's a huge challenge for vaccine development.
The good news is that we also have learned why a vaccine is possible. One can ask the question -- after all these years, after products that didn't work, after lots of money spent -- "Why are you still doing this? What a waste of time." Well, that's a silly approach, I think. There's a lot of good scientific rationale of why we think a vaccine is possible. We're seeing more information presented this week at CROI about people who are called elite controllers or long-term nonprogressors.12-15 There's something going on in some people's immune system naturally that is fighting off the virus. We don't know how to translate that into a vaccine, but we know there's something going on in the immune system that works for some people some of the time. We need to replicate that.
Similarly, we have increasing data from animal studies that we are finding vaccines that can prevent SIV [simian immunodeficiency virus], the simian version of HIV, with complete protection.16 This is true in PrEP, in oral pills. It was presented this week.17 [Click here to read or listen to TheBodyPRO.com's interview with an investigator in this study of oral PrEP.] It's true in vaccine animal studies. How do you replicate that into human trials? We don't know yet. But these are the clues that tell us that a vaccine is possible. We know it's possible; that's the scientific process.
So people in the vaccine world are still excited?
They are -- and I must say, they're excited about different things. It used to be getting really excited about starting a 16,000-person clinical trial. Now the excitement is around what's happening in that cohort of long-term nonprogressors, what's happening with the epitope mapping. There are little clues.
I think that for me, the big issue is making sure all of us -- scientists, laypeople, practitioners in the field -- that we understand the research process. You often have these slides. You have a vaccine concept. You do a phase 1 trial, a phase 2 trial, a phase 3 trial, and you license the product. That's the ideal world. It doesn't exist. The scientific process is thousands of little experiments to get us a vaccine candidate, and then a phase 1 trial. And then, maybe, after a successful one, a phase 2 trial. And almost inevitably, after phase 2, back to phase 1 -- making it better, learning what you did, figuring out, "Oh, we need to do this differently," maybe getting back into a phase 2 trial.
The STEP trial was actually called a test-of-concept trial, or a phase 2b trial; it wasn't even a phase 3 licensure trial. We learn from that, and you see this iterative process. That's not a bad thing. That's how science happens. We just haven't often explained it that way -- in large part, I think, because we've been so desperate for success. We so desperately need a vaccine; we so desperately need a microbicide; we so desperately need PrEP to work that we sometimes forget that research is about a lot of different experiments that take us in lots of different directions, with a bit of a feedback loop to have better experiments. We'll get there.
For me, one of the exciting things this week was to see a microbicide trial that actually showed at least a trend toward potential protection, though it wasn't statistically significant. This well-conducted trial was studying the PRO 2000 gel. Luckily, another trial is already in the field testing that same candidate. What does one do with a 30% protective microbicide? How do we think about its use? All really important questions, but it did give us a clue. It gave us a positive clue.
Other trials have given us negative clues. In fact, interestingly, PRO 2000 was in a study with another microbicide candidate [BufferGel] that didn't work.18 It had a total flat result. That's good. We obviously want things to work, but knowing what doesn't work is important. We know something was happening with PRO 2000; it wasn't a huge effect, but it was modest and in the right direction. We have to build on that.
So I look at that result, and I look at some of the data coming out of the STEP trial presented here that have me feeling like we better understand the research process. We understand that it's going to be iterative, that it's going to be little, little pieces of the puzzle, but we have more of them now than we did before.
Thank you very much.
This transcript has been lightly edited for clarity.