You are leaving Medscape Education
Cancel Continue
Log in to save activities Your saved activities will show here so that you can easily access them whenever you're ready. Log in here CME & Education Log in to keep track of your credits.


ART Advances for Drug-Naive Patients: An Expert Interview With Calvin J. Cohen, MD


This feature requires the newest version of Flash. You can download it here.

Editor's Note:
Researchers and front-line clinicians have made great strides in the almost 10 years since highly active antiretroviral therapy (HAART) first changed the face of HIV clinical care. Better-tolerated, simple regimens are now the rule for most antiretroviral-naive patients. But even with the increased range of therapeutic options, clinicians and patients must still weigh a number of critical considerations prior to treatment initiation. At the 3rd IAS Conference on HIV Pathogenesis and Treatment, July 24-27, 2005, in Rio de Janeiro, Brazil, Scott Williams, Editor and Program Director, Medscape HIV/AIDS, interviewed Calvin J. Cohen, MD, Research Director at both Community Research Initiative of New England and Harvard Vanguard Medical Associates in Boston, Massachusetts, about optimizing antiretroviral treatment for drug-naive patients.

Medscape: At this year's IAS meeting there were new longer-term data that compared tenofovir (Viread) + FTC (emtricitabine, Emtriva) vs AZT + 3TC (zidovudine + lamivudine, Combivir), both combined with efavirenz (Sustiva) in antiretroviral-naive patients.[1] What do the new data tell us about differences in efficacy and tolerability between these 2 combination antiretroviral medications in this patient population?

Dr. Cohen: What they tell us is that the combination of tenofovir and FTC was superior on several aspects of care when compared with Combivir, both in efficacy as well as in safety over the past 48 weeks. Virologically, both regimens do well, but there's a rough 10% advantage of the overall performance of tenofovir/FTC vs Combivir. It's tempting to say Truvada, although in that study tenofovir and FTC were actually given as separate tablets, not as Truvada [the coformulated tenofovir + emtricitabine tablets]. Nevertheless, the results would be similar in either case.

They differed in terms of safety and adverse events. We've known for several months that there's a difference in the rate of anemia. At this study timepoint, 6% of patients dropped out of the Combivir arm due to various grades of anemia judged serious by the clinician, while no one on the tenofovir/FTC arm had anemia leading to discontinuation. Therefore, when we look at the adverse event table, we see superiority in terms of the ability for people to tolerate and stay on tenofovir and FTC. In terms of lipids, while both arms do okay we saw superiority to the tenofovir/FTC arm, with less of an increase in total cholesterol on TDF, and a greater increase of cholesterol in the AZT arm. This is something that's not necessarily been appreciated in too many prior studies, but this study does document a difference favoring the tenofovir arm.

Finally, at this meeting we saw differences in DEXA [dual-energy x-ray absorptiometry] scanning of body fat. Clearly for many patients, the issue of lipoatrophy is one that's critical to decision making. What we saw here is not a comparison of baseline to week 48, since unfortunately patients did not have a baseline DEXA scan. Nevertheless, if we assume that randomization works and people were similar at baseline, what we see at week 48 is already a meaningful difference in the 2 arms favoring tenofovir/FTC, with a greater amount of body fat vs a somewhat smaller amount in the AZT arm. These data are in the context of many other trials that show that over time AZT can be associated with fat loss. This study is simply just one more piece of evidence to support that tenofovir appears to have much less lipoatrophy than does AZT as demonstrated here and certainly compared with d4T [stavudine] in prior studies. When we put it all together, when we look at rates of discontinuation from toxicity, rates of safety as measured by lipids, and overall ability of somebody to go on and stay on a regimen safely, what we see is that the tenofovir/FTC arm outperformed Combivir.

Medscape: Tenofovir and ddI (didanosine) originally was a drug pair that people were interested in because they are both once-daily agents and this would obviously be a combination that would potentially facilitate ease of dosing. However, a lot of research showed that many patients, when these 2 drugs were combined, had precipitous CD4 declines. What are the cautions that providers should be aware of if they do combine these 2 drugs in naive patients, when potentially other options might be less preferable due to the presence of resistance mutations?

Dr. Cohen: At this meeting we saw yet more evidence that in patients initiating therapy for the first time, a regimen containing tenofovir, ddI, and efavirenz underperforms other available regimens. One study presented here was called TEDDI, simply tenofovir, ddI, efavirenz.[2] The TEDDI study is quite similar to other studies we've seen in the past year showing that overall the performance of this regimen is not as good as other regimens, such as tenofovir/FTC/efavirenz and other regimens. There are many reasons this might be, but of increasing concern is a possibility of some intracellular problems when these 2 drugs are combined.

It's hard to totally put this story together -- for example, in treatment-experienced patients in BMS 045,[3] in which patients were randomized to either boosted atazanavir or lopinavir. In that study, roughly half of the patients were on tenofovir and ddI. As you mentioned, when resistance tests are analyzed from patients who are switching therapy, often tenofovir and ddI may look to have retained activity and may be selected by clinicians as was done in that study. What's interesting is that, on average, patients did well on that combination in terms of CD4+ cell responses and safety and tolerability. That contrasts with what we see in these studies of antiretroviral-naive patients.

So what does this mean? First, when somebody is starting therapy on a nonnucleoside, all 3 drugs probably have to be fully potent and not be vulnerable to resistance and cross-resistance. We are starting to appreciate that tenofovir and ddI may have some overlapping impact from their resistance profiles, ie, the 74V and 65R mutation pathways. If that's the explanation, then that may explain why efavirenz with 2 nucleosides that might have overlapping resistance pathways may yield to resistance sooner than regimens that have 3 drugs with 3 different resistance pathways.

Of course, that doesn't explain the CD4+ cell count differences, and it's hard to make sense of all the data. In this study that BMS did, for instance, the CD4 response has been fine, while in other studies there's a concern for CD4 response. The current theory, however, has to do with levels of these drugs. In studies that have put people on ddI 250 mg in combination with tenofovir, by and large the CD4 response has been similar to that of other combinations.

I think the unanswered questions are several, including understanding what is the mechanism that is causing this, as well as when (or if) do we ever recommend tenofovir and ddI, given that there are concerns about overlapping resistance as well as intracellular toxicity. Nevertheless, I think for naive patients the answer's clear: Those 2 drugs in combination do not have a recommended role at this time. For treatment-experienced patients, it's fair to say that we're always struggling to do the best we can. If tenofovir and ddI look the best in terms of the most activity, it may be that with proper monitoring we can figure out in whom these 2 drugs are tolerated, and in whom these 2 drugs have too much toxicity.

Medscape: Atazanavir and tenofovir are obviously 2 of the newer antiretrovirals that we have in our arsenal, and both have very favorable tolerability profiles compared with other agents in their classes, which make them also particularly attractive for antiretroviral-naive patients. What do clinicians need to consider in terms of pharmacokinetic issues, dosing adjustments, and related concerns when coadministering these 2 drugs?

Dr. Cohen: I agree that tenofovir and atazanavir do have attractive features, particularly with regard to safety parameters, including lipids. So how do we give these 2 drugs together if that seems like the best choice? We've had consistent answers for years about what to do, although it remains a bit uncertain why we have to do it. The influence of tenofovir is to lower atazanavir exposure, and it still remains unclear how this happens. Regardless, if we want to use tenofovir with atazanavir, the correct dose of atazanavir is 300 mg once daily with ritonavir 100 mg once daily, given together, and tenofovir also given 300 mg once daily. Tenofovir does not have to be given with the atazanavir but the ritonavir/atazanavir must be coadministered, and tenofovir given similarly once daily. That approach has led to some very successful responses in patients and accounts for the way in which tenofovir lowers atazanavir exposure by further boosting the atazanavir levels to a comfortable and protective level.

How do we know that's the right dose? In the BMS 045 trial, what we saw is that when patients in that study received 300 mg/100 mg atazanavir/ritonavir with tenofovir, the design of that study had that regimen compared with lopinavir/ritonavir (Kaletra) with tenofovir, and in each case with a third nucleoside. There were comparable results over the 96 weeks, and that gives us confidence that that ratio of atazanavir/ritonavir 300 mg/100 mg with tenofovir 300 mg provides ample antiretroviral activity compared with a good standard-of-care regimen.

Medscape: What about selection of the third drug, in terms of either a protease inhibitor (PI) or a nonnucleoside reverse transcriptase inhibitor (NNRTI) for initial HAART in ART-naive patients? What's your own mental checklist when weighing which drug is most appropriate for an individual patient who presents at your clinic?

Dr. Cohen: One of the more vigorous debates we still have in our field is, do I start with a nonnucleoside, do I start with a boosted PI, and how do we pick? Over the years we certainly have had numerous studies that support and suggest over and over that whichever we pick it will work well, so it's not a question of which one is good and which one is bad, it's a question of which good approach makes the most sense.

So how do we pick? One easy question: Is there baseline resistance? Unfortunately, perhaps 7% to 8% of patients who have baseline resistance have baseline NNRTI resistance. That's been seen in populations certainly across the United States as well as across Europe, where these drugs have been introduced. Increasingly in places like Africa, where single-dose nevirapine has been used in mother-to-child transmission studies, the spread of NNRTI resistance is upon us. Therefore, in those circumstances there isn't a debate; the boosted PI is the preferred approach. In patients who don't have resistance, how do we choose? We still await the results of a randomized, comparative study of efavirenz vs lopinavir, or in another study, vs boosted atazanavir, and such studies are under way. The atazanavir comparison study will compare efavirenz vs boosted atazanavir, each in combination with some of the best nucleoside backbones we have, either Truvada or Epzicom (abacavir + lamivudine).

Until we have that data, how do clinicians choose? It's fair to say that in some patients efavirenz isn't a good choice. If, for example, a woman is considering pregnancy, we know that efavirenz is associated with birth defects if the drug is administered in the first trimester, so we should avoid using efavirenz. A boosted PI would be preferred since atazanavir received a schedule B classification from the FDA for safety in pregnancy. Also, efavirenz certainly does have some CNS toxicities in the first week, and some clinicians have ways in which we choose which patients either could or could not tolerate that medication.

I think with that checklist we also have one other factor, which is that if we use boosted atazanavir there are probably 2 issues to remember. One is that atazanavir, in about 5% to 10% of patients, might have some jaundice or scleral icterus. Some patients are willing to find out if they're in that 5% to 10%; for others, if anybody saw that I had yellow eyes that would mark me in some way that I don't want to be marked. Therefore, for some patients, that issue is one that leads them in another direction.

Second, atazanavir must be given with food, and it requires acid for absorption. If somebody requires acid-lowering medications, we've learned recently that H2 blockers are fine to be coadministered, and the atazanavir levels are acceptable even in the presence of an H2 blocker. But if somebody with bad reflux requires the use of a proton-pump inhibitor such as omeprazole, atazanavir is inadequately absorbed, and those patients should not be on atazanavir and should probably be on efavirenz or some other third agent.

As we put that checklist together, we come to a series of patients who could be on either drug, and after that point we wind up with the reason we do randomized trials: There are a number of patients in whom there's no reason not to use either with confidence, and at that point we usually just present these options to our patients and ask them to help us choose. For some, pill burden is a major consideration. Truvada + efavirenz is 2 pills once daily, and it might be 1 pill once daily if research allows. Atazanavir/ritonavir is 3 different pills, the dual NRTI Truvada (or Epzicom) is another pill. For some the difference is too small to matter, but there actually are patients for whom the choice of 2 pills vs 4 is enough for them to say "I'd rather have 2 pills," and for those patients we have choices. That's at least how I approach my patients.

Medscape: Let's talk a little bit about special populations of ART-naive patients. First let's discuss naive patients, who may present with more advanced disease, perhaps those with a very high baseline viral load, low CD4+ cell count, opportunistic infection (OI), or perhaps a combination of these characteristics. Is there perhaps a 4-drug induction regimen that's warranted, possibly followed by a standard 3-drug maintenance regimen after an extended period of viral suppression? Are you of the mindset that a boosted PI or efavirenz is preferable as a third drug? What are some guidelines for this type of naive patient?

Dr. Cohen: For the past few years we've had some data to support the idea that if there's any vulnerability to the 2 NRTIs plus efavirenz approach, it might be seen in patients at the extremes -- low CD4+ cell counts or very high viral loads. Indeed, in the Gilead 903 study,[4] those few patients who did have a viral rebound and resistance to tenofovir were those patients who started with very low CD4+ cell counts, often < 50 cells/mcL, and who had very high viral loads, often > 200,000 copies/mL.

Is that a problem that's shared by boosted PIs? We don't have randomized trials to answer the question, but what we do see is fairly consistent evidence that in studies that have used a boosted PI in ART-naive patients, even in those with such extreme numbers, if we don't get complete viral suppression, we also do not see resistance to the boosted PI even in those circumstances. That difference has been one that's swayed some clinicians to prefer a boosted PI when resistance to the regimen is a worry. If they tolerate the boosted PI, they can remain on it. But they also could simplify the regimen at least by pill burden, and substitute a nonnucleoside after suppression; that's one version of induction-maintenance therapy.

Do we need 4 drugs to get suppression in patients, and then go to 3 drugs for those who present at the extremes of lab numbers? We lack any data at this point that 4 are ever better than a good, potent 3-drug regimen. In fact, most studies that have been done show the opposite. When we use a fourth drug, we don't improve the chances of suppression. If anything, we add the fourth drug's toxicity and don't improve the outcomes at all.

So at this point, 2 nucleosides plus efavirenz or a boosted PI seem ample for almost all patients, but it doesn't seem that we ever need a fourth drug to help those regimens. They're fully potent, and if somebody takes them, those 3 drugs work well to stop this virus from replicating.

Medscape: What about pregnant women or women who might become pregnant, in whom efavirenz is obviously contraindicated? What are the preferred options for the third drug in these ART-naive patients, and what NRTIs should also possibly be avoided?

Dr. Cohen: As you pointed out, the recent data on efavirenz reinforced the initial animal data and have led to review of our options.[5] If a woman is not currently on therapy, if the CD4+ cell count is > 250 cells/mcL, we also have a caution to not initiate nevirapine, and that has to do with a black-box warning from the FDA suggesting increased rates of toxicity in women based on certain baseline CD4+ cell counts.[6] If a woman has lower CD4+ cell counts, nevirapine remains an option. For some clinicians 250 cells/mcL is just a line in the sand, and if it's 249 cells/mcL, is that different from 251 cells/mcL? We all have to make some judgment about what CD4+ cell count leads us to be comfortable regarding the use of nevirapine.

In addition to nevirapine, we have the PIs. Fortunately, a few of them do have the confidence of the FDA to give them schedule B status: That includes atazanavir, saquinavir, and also the use of ritonavir to boost these drugs. There's certainly a great deal of experience, at least with nelfinavir, as well. At this Rio meeting, we saw data for the first time on the use of atazanavir boosted in a small number of women in Springfield, Massachusetts, led by Dr. Anne Morris in our research group at CRI New England.[7] What we report is a good safety track record for boosted atazanavir, and these clinical data are some of the first to increase the amount of understanding regarding our options in treating pregnant HIV-infected women.

Medscape: Since you have your finger on the pulse of the latest developments in drug research, what advances do you see in initial therapy in the near to medium term?

Dr. Cohen: There's a great deal of confidence in the current safety, track record, efficacy, durability of the current options. The approaches we have now are extraordinarily more successful than those of even a few years ago on many metrics. I think that research is always looking to see how we can do better, but I think the bar is going to be set high to not just do as well as those options, but to do better. We've established a great current profile of medications that gives us confidence that we can stop this virus from replicating in a durable way. That doesn't stop research, however. At this meeting we saw evidence of people who are working with induction-maintenance strategies and going to monotherapy with boosted PIs. As you mentioned, new drug classes are coming -- CCR5 inhibitors. Even further are other drug classes, such as newer nucleosides and newer PIs. What their role will be is exciting to consider, but again the bar's going to be set high given how well we're doing with current treatment.

  • Print