October 27, 2008
This is Bonnie Goldman, Editorial Director of TheBody.com. I'm here today with Dr. Joel Gallant. Dr. Gallant has been treating people with HIV since the beginning of the pandemic. He's also a leading HIV researcher and a professor at Johns Hopkins University School of Medicine in Baltimore, Md.
Welcome, Dr. Gallant.
Thanks, Bonnie. Good to be here.
There are a number of clinical trials related to initial therapy that I think were really important at this meeting. Probably the most important one was the STARTMRK trial. This was a big, phase 3 trial comparing Isentress [raltegravir, MK-0518], which is the new integrase inhibitor from Merck, against Sustiva [efavirenz, Stocrin], both of those in combination with Truvada [tenofovir/ FTC] for people who are on their first regimen.
We'd seen earlier, smaller studies that suggested that Isentress was going to be as good as Sustiva and maybe better tolerated. STARTMRK was the big trial, with lots of patients enrolled. It showed that both drugs did great, with very high rates of suppression at one year. To use a statistical term, Isentress was what we call non-inferior to Sustiva.
Also, as you might expect, Isentress didn't cause the central nervous system side effects that we associate with Sustiva, so we didn't see the dizziness, the dreams, those kinds of things. In that sense, Isentress was as good as Sustiva, but better tolerated. There are also some little advantages to Isentress in terms of triglycerides, but both drugs are pretty good in terms of their effect on lipids. This trial, I think, brings integrase inhibitors, specifically Isentress, into the list of drugs that we could consider for first-line use.
Now, the one disadvantage of Isentress is that it's a twice-a-day drug. Of course, Sustiva is usually given in combination with Truvada in the form of a single pill, called Atripla, which you take once a day. That's an enormous advantage and people like that a lot. On the other hand, for some people, the tradeoff in terms of side effects may be worth taking a twice-a-day combination.
The researchers also presented some data looking at why Isentress might end up being a once-a-day drug. They haven't studied it, but there is at least a scientific rationale for why this drug may be able to be given once a day. They're going to proceed with studies comparing the once-a-day and twice-a-day formulations, so it may not be a twice-a-day drug forever. I think this study was particularly important because, up until this conference, the choices for initial therapy were two nukes [nucleoside reverse transcriptase inhibitors or NRTIs] plus a non-nuke [non-nucleoside reverse transcriptase inhibitor or NNRTI] like Sustiva, or two nukes plus a protease inhibitor [PI]. Now we have a third option.
Do you think it's true what they've been saying about Isentress, that it's one of the most powerful HIV medications we've ever seen?
I don't think we know that it's more powerful. It's certainly very effective. It's very well tolerated. Those are the advantages. I don't think we can say that it is more powerful. Potency is a different issue than what we call effectiveness, or efficacy. The combination of being effective and being relatively free of side effects is really kind of the claim to fame.
The other issue, of course, is that Isentress is part of an entirely new drug class, so nobody's resistant to it yet, if they haven't already taken it. For that reason, in treatment-experienced patients, it'd offer a lot because no matter how resistant you were to other drugs, you'd respond to this drug. Maybe that gave people the sense that it was powerful. It's a great drug, but it is not the most powerful. The STARTMRK trial moves it from a drug that was primarily for salvage into the first-line arena.
Now, this wasn't the only trial for treatment-naive patients. I stressed it first, because it was really the most novel of the studies, but we also heard longer-term data on a number of other trials, such as the ARTEMIS study looking at once-a-day Prezista [darunavir, TMC114] with Norvir [ritonavir] versus Kaletra [lopinavir/ritonavir] for first-line therapy. Again, now at two years, Prezista is actually looking better than Kaletra for first-line therapy, both in terms of its effectiveness and its tolerability, as well as its effect on lipids. As many of you know, the FDA [U.S. Food and Drug Administration] just approved Prezista for once-a-day use in treatment-naive patients at a dose of 800 mg with 100 mg of Norvir. The 400-mg tablet has just become available and is in pharmacies.
This certainly offers another option for people who haven't yet used PIs. We also heard more data from the CASTLE study looking at Reyataz [atazanavir] and Norvir compared to Kaletra. Again, it looks like there are advantages to Reyataz over Kaletra, specifically in terms of GI [gastrointestinal] tolerability and effect on lipids.
My own feeling is that both the ARTEMIS study and the CASTLE study are telling us that if you have never been on a PI before, and you don't have any PI resistance, then you shouldn't have to take more than 100 mg of Norvir. The problem with Kaletra is that -- as good as it is and as durable as it's been over the years -- the only way you can take Kaletra is to be on 200 mg of Norvir. When you're taking 200 mg of Norvir, you're going to have more diarrhea, more nausea, and more effect on your cholesterol and triglycerides.
I think the way we're moving in terms of protease inhibitors is to take the lowest amount of Norvir that you can get away with. We're seeing with Prezista, with Reyataz and also with Lexiva [fosamprenavir, Telzir] that 100 mg of Norvir is just fine.
The problem with Kaletra is that it contains 200 mg of Norvir. You can't take it out, so you're taking more than you need.
Exactly. However, that's also the advantage of Kaletra, of course. You don't need an extra pill. You don't need an extra co-pay. You don't need refrigeration. The Kaletra includes Norvir, but the disadvantage is you're stuck with that dose of 200 mg.
A question about integrating Isentress into care: So far, everyone's pretty happy with Atripla. It's good enough, and in cases where it's not, there's Reyataz. How many options do we need for first-line that are really powerful?
That's a good question. Atripla has still never really been beat. You could argue that Isentress came close to beating Atripla, not in terms of effectiveness, but just in terms of tolerability. But Atripla, or at least Sustiva-based regimens, especially those that include Viread [tenofovir] or Truvada, really remains the gold standard. We haven't seen anything be better.
I think Atripla is still going to be the first-line regimen for most people, but keep in mind there are people who can't take this drug. For example, people who have a lot of the psychiatric and neurological side effects, such as difficulty concentrating, mood changes and insomnia. Most of those things go away in a few days, but there are people in whom the side effects go on for weeks and weeks, so they can't stay on it. There are women who want to become pregnant or who are preventing pregnancy who may not be good candidates for that.
There certainly are people who end up on something else, such as boosted Reyataz as you mentioned, and I think now boosted Prezista is going to enter that realm too. Isentress may compete against the idea of always using a protease inhibitor as a backup when we can't use Atripla. Some people, I suspect, will simply hear about the side effects of Atripla and say, "I don't even want to try it. I don't mind taking pills twice a day. I'd rather go with something that is unlikely to cause side effects." Now that becomes a possibility.
There's also the question: What about people who can't take nucleosides? Atripla contains Truvada and Truvada contains Viread. Viread is a drug that can rarely cause kidney damage, but for people who already have kidney problems, this may be a concern. We are now more and more concerned about abacavir [Ziagen], which is an ingredient in Epzicom [abacavir/3TC, Kivexa], in terms of whether it's less effective at reducing viral loads and whether it could cause heart problems. There may be some people -- for example, people who have a lot of risk factors for heart disease and kidney disease -- who may be better off on no nukes at all. Now, we have the option to use a protease inhibitor and Isentress, or a protease inhibitor and Sustiva, without using nucleosides at all. I don't think that will be a common approach, but there may be people for whom it's appropriate.
Has that kind of regimen been studied?
Not well. Not in treatment-naive patients, but there certainly are ongoing studies looking especially at protease inhibitors plus Isentress. I think we'll see more of those and we'll have more information. It also brings up the point that you could now go from a first-line regimen to a second-line regimen with no overlap in classes. Meaning that, let's say you started Atripla and failed Atripla for some reason. That's not common, but it could happen. You could then go to a protease inhibitor and an integrase inhibitor where there'd be absolutely no concern about cross-resistance because you're using two entirely new classes of drugs in your second-line. You could just as easily flip the order where you started with the protease inhibitor and integrase inhibitor, and then went to something like Atripla. We're really opening up a lot of options for how we can start, how we can sequence from one class to another.
Who's going to teach the physicians of America about this new treatment paradigm? It sounds like a sea change, meaning what was true last year is not true this year. It's hard to keep up and physicians don't necessarily adapt easily to new things. What is a patient to do?
Fortunately, the things that were true last year are still true in the sense that Atripla is still a great way to start and Truvada plus a boosted PI is still a great way to start. It's just that now we have this other option. Of course, I've always said that people with HIV should be managed either by an expert or by their primary doctor in regular consultation with an expert. This shouldn't be prescribed by somebody who's dabbling with a few patients with HIV.
It's another reason, I think, why guidelines are so important. We recently had the IAS-USA [International AIDS Society-USA] guidelines updated. They're pretty current, although they do not reflect this Isentress data. Then, we have the DHHS [U.S. Department of Health and Human Services] guidelines that get updated on a regular basis. Guidelines are another good way for doctors to keep up. I would point out that, while a lot of people think the guidelines are just a table of what to start and when to start, they are in fact a wonderful detailed document. If you knew the material in those guidelines, you'd really be up-to-date, I think, on HIV. It's important not just to know the bottom line, but to also study that document and make sure you're keeping up in that way. Of course, there are lots of CME [continuing medical education] programs, courses and updates, and things like that for people who can't come to these meetings.
I wasn't going to bring it up, but since you mentioned the guidelines I'll as the question: Seeing how you are on the DHHS guidelines committee, do you have any idea when they'll be updated to reflect all these changes?
The guidelines get updated when they're updated. There's no regular schedule. It depends on how soon the group can come to agreement. There are subcommittees that look at different aspects. So I couldn't predict, except to say that all the data from ICAAC/IDSA 2008 will be considered by the panel.
Because there are so many things that happened since January 2008 when the DHHS guidelines were last updated. It's amazing.
Yes, well, you'll just have to interview me again when the time comes.
OK. Is there anything else that stood out for you at the conference?
I think one of the most important studies from a treatment standpoint was the study from the NA-ACCORD [North American AIDS Cohort Collaboration on Research and Design]. NA-ACCORD is this big conglomeration of 22 different clinical cohorts in Canada and the United States. It represents a huge number of patients. What the researchers decided to look at was the question of when to start therapy.
They specifically looked at people who started treatment within a year and a half of their first CD4 count between 350 and 500. They compared those people with those who did not start within a year and a half of meeting that point. What they found was that those who started early had a 70 percent better survival rate than those who waited. They even corrected for things like hepatitis C and injection drug use [IDU], and still found the same thing. These same investigators are going to go further with this, hopefully at the next meeting, which is CROI [Conference on Retroviruses and Opportunistic Infections] in February. Then, they're going to look at those who started above 500 to see if they see the same effect.
There are problems with observational studies like this. On the other hand, there's talk of the START study, which is going to randomize people to early versus late therapy. If that study happens at all in the current budget climate, it's going to be a very expensive study. It could take years to get the results. My concern is that by the time we get the results of that study, nobody's going to care because we'll already have moved on and will be treating much earlier based on big observational studies like NA-ACCORD.
My own feeling is that these observational studies, when done well, are extremely important for guiding these questions that are very hard to study in the traditional way where you randomize people to two different arms. I think the study presented here has moved us further toward the idea of treating earlier at higher CD4 counts.
But in reality, what's the average T-cell count of somebody who walks into a clinic today?
In the United States, the last I heard, it was about 180. So, for this average person who we're talking about, this discussion is completely irrelevant because they needed to start treatment years before they even got diagnosed. I'm hoping that will change if states and doctors start to follow the CDC [U.S. Centers for Disease Control and Prevention] recommendations saying that we should be testing everybody. That hasn't happened yet. We don't know how we're going to pay for care or how we're going to find enough doctors for all these people if they do get tested. But for somebody out there who's not in that situation, who was diagnosed early, I think these questions are very relevant.
Has your mind been changed by the NA-ACCORD trial? If a patient of yours with a CD4 count of 500 were to ask, "Should I start treatment?" Would you now respond yes?
I was already advising patients to consider early treatment. This study just gives me that much more supporting evidence. My approach has been to quote the guidelines and the guidelines currently say that you should start treatment if your CD4 count is below 350 and -- this is the part of the guidelines that a lot of doctors miss -- the footnotes say if it's above 350, you should consider treatment if you have hepatitis B, if you have HIV renal/kidney disease, if you have a partner who's negative, if your viral load is high or if your CD4 count is falling rapidly. All those are considerations that might make you want to start earlier.
Now, we have this study that suggests that there may be benefits for everybody to start earlier. A lot of that is because of what we used to call non-HIV-related morbidity and mortality (e.g., cancers, heart disease and liver disease). When these things happened before in smaller studies, we would just say, "Oh well, that has nothing to do with HIV." Now, we're beginning to realize they probably do. It's not good for you to have a virus in your blood.
My own approach toward patients who have a high CD4 count is to tell them that, while this is not standard of care yet, there's certainly a strong scientific rationale for treating almost everybody, especially if the person is likely to be adherent to therapy and is motivated. I wouldn't push them, but I would certainly suggest it. If they asked me, "What would you do?" I would choose treatment.
Is there any understanding of what the virus does to you that's so bad? Has somebody been studying the exact mechanism that makes it so harmful to humans?
There are a lot of different hypotheses and some of them have pretty good evidence. We know that untreated HIV increases inflammation in the body and inflammation can result in heart attacks, stroke and the like. We know that HIV can increase the blood's ability to clot, which could also have something to do with why people may develop more heart attacks when they're not treated. We're concerned about its effect on the brain and whether we will see long-term problems with not so much dementia, because we're not seeing a lot of dementia, but with just subtle differences in cognitive abilities with untreated HIV. We're worried about the premature aging that may be caused by HIV -- its effect on bones and on metabolics. We know that untreated HIV can affect lipid levels. So there are all sorts of mechanisms by which HIV may cause these sorts of less AIDS-specific problems.
This kind of study demonstrates that CD4 count and viral load are not the only things we should look at. We should also be aware that something else is going on that we're just not measuring.
Certainly, we've always known that the CD4 count has been great at predicting whether you're going to get PCP [Pneumocystis pneumonia], MAC [Mycobacterium avium complex] or CMV [cytomegalovirus], but we've moved on from that. We can prevent those things. Now, we're realizing that the CD4 count only measures those issues that are truly phenomena of immunosuppression; it's not going to tell you what's going on in terms of inflammation, immune activation, ability to clot and all those things.
Viral load may be a little bit more predictive. We know that, for example, the higher your viral load, the more inflammation and immune activation, so viral load may be telling us a little more. If you're not on treatment, it's probably better to have a low viral load than a high viral load, but it doesn't mean that there aren't things going on even when your viral load is relatively low that might be prevented by treatment.
Are there any ongoing studies that we should look forward to in the next six months that are going to again change things?
I'm going to pull a Sarah Palin here. I'm going to answer a question that you didn't ask.
One thing that we need to realize -- and we saw this here at this conference -- is that there aren't a lot of new drugs coming down the pipeline for people who have resistance. We heard a little bit about the maturation inhibitor bevirimat [PA-457]. Late 2007/early 2008, we had this burst of new drugs that really revolutionized treatment for HIV. Now, we're going to have this long dry spell with not much coming. Frankly, drug companies are very nervous right now about wanting to develop new drugs when everybody's got undetectable viral loads, and they're not even sure how they would develop or sell these drugs.
What I would say, rather than answer your question about what to look forward to, is what not to expect. That is a steady supply of new drugs to bail us out when the old ones fail. The reason I bring that up is to emphasize how important it is to use these new drugs wisely, to make sure that you're getting expert care, and to take a combination of multiple active drugs if you have drug resistance. I'm really worried that a year or two from now, we'll be sitting and having a conversation about how to deal with integrase inhibitor resistance, specifically Isentress resistance. That's going to be a big mess, because we're not going to have much to help us out.
So adherence really counts now?
Oh, adherence really counts. Yes. Adherence always counted, but you know the more effective a drug is, the more important it is to be adherent. Back in the days when we gave people AZT [Retrovir, zidovudine] monotherapy, being non-adherent to a drug that only worked a little bit probably wasn't that big of a deal, but now that these drugs are highly effective and resistance can occur, adherence becomes really critical.
One last question: If you take your meds on time, every time and you don't have resistance, is it still possible to somehow acquire resistance?
Let's use Atripla as our example. There are certainly people who are infected with Sustiva-resistant virus. Most of the time, a baseline resistance test will pick that up, but it can miss that. It's possible. For example, let's say you've been infected for five years and you finally got diagnosed, saw the doctor, got a resistance test and it didn't even show any resistance. There is the possibility that there was a little bit of Sustiva resistance, but it just wasn't at a high enough level to get picked up by the test. Then you take Atripla and you fail through no fault of your own. You're taking your meds as prescribed, but what happened is you selected out that low-level resistant variant and it became the predominate strain. So that can happen.
If you don't have baseline resistance, you're put on a good regimen and you take it -- I would even add that with today's drugs, it's just taking them and not missing multiple doses that counts, because the timing is no longer as important as it used to --
You mean timing within two or three hours.
Yes. I'm not suggesting this, but if you were to miss a day's worth or even two days' worth of Atripla, for example, you probably would be fine because the drug lasts so long in the blood.
They stay in your system.
Don't make a habit of that. I'm just saying that, unlike the old days with Crixivan [indinavir], when you had to be right on the clock, every eight hours, these new drugs aren't really like that anymore.
They forgive a little bit.
They're quite forgiving. What they're not forgiving of is interrupting therapy. The worst thing you can do with Atripla, for example, is go away for a vacation for two weeks and forget to take your pills. That's where you get these slowly declining drug levels and we see a lot of resistance develop with that. As long as people are taking it every day, I don't so much care when they take it, and the same is true for most of these protease inhibitors -- if you get your drugs in every day with food, if that's required, that's really what counts.
Anyway, back to your question though. I don't see resistance occurring for no reason. Everybody I've seen who fails therapy is either failing because they're not taking their meds or they're failing because they have pre-existing resistance. Either it was resistance they were infected with or resistance that occurred while on a previous regimen. People who are starting therapy from scratch with a normal resistance test, those people aren't failing.
Bob Siliciano from Hopkins [Johns Hopkins University School of Medicine] presented some data at this conference showing that when you're on Atripla -- he used Atripla as an example, but I think it's true for any of the good regimens -- and your viral load is undetectable, it's truly undetectable. There's really no evidence that the virus is replicating at all. He showed that by adding additional drugs and not getting any further suppression. If you were able to measure your viral load using an assay that got down to one copy, you might see a viral load of, say, five or 10, but that's not really replicating viruses, just virus that's being released from reservoir cells. If there's no replication, then the virus isn't mutating and it's not developing resistance mutations, which means there's really no limit to how long one of these regimens could last. While you may not stay on the same regimen forever, because new drugs come out and things change, you wouldn't necessarily need to change because the regimen was failing.
Wow. Well thank you so much.
This transcript has been lightly edited for clarity.