Newsroom

A certain combination of AIDS drugs is superior to others when it comes to the initial treatment of HIV patients, NewYork-Presbyterian Hospital/Weill Cornell Medical Center researchers report in this week's (April 29) New England Journal of Medicine.

Roy Gulick, M.D., M.P.H., and a multicenter team of researchers specifically looked for a treatment that was relatively easy to take and that did not include protease inhibitors, highly effective AIDS-fighting drugs that can sometimes have undesirable side effects.

Today, we have 20 drugs that are approved for the treatment of HIV infection,and even though there have been great treatment successes, it's still not completelyclear what the optimal first drug regimen is, said Dr. Gulick, PrincipalInvestigator of the NIH Multicenter Clinical Trial and Director of the CornellHIV Clinical Trials Unit at Weill Cornell Medical College in New York City.

Many of the regimens that have been so successful in preventing AIDS-relateddeaths have been very complicated and caused side effects, so the whole fieldhas really tried to move toward regimens that are effective at controlling HIV,but, at the same time, are simpler and less toxic. We expect that patients willneed to take these types of medications for many years, and that's where ourstudy comes in, he said.

In the federally funded study, Dr. Gulick and colleagues randomly assigned 1,147 HIV-infected patients to one of three different treatment regimens: AZT/lamivudine/abacavir (marketed as Trizivir); AZT/lamivudine (marketed as Combivir) plus efavirenz (marketed as Sustiva); or a 4-drug regimen of AZT/lamivudine/abacavir plus efavirenz.

It was quite a large study, said Dr. Gulick, who is also AssociateProfessor of Medicine at Weill Cornell Medical College and Associate AttendingPhysician at NewYork-Presbyterian/Weill Cornell. African-Americans andLatinos were well-represented, so we had an excellent mix of patients trulyrepresentative of those who are living with HIV infection in the United Statestoday. Nineteen percent were women, and 60% of patients were non-white.

Efavirenz is a type of drug known as a non-nucleoside reverse transcriptase inhibitor, a class that also includes the drug nevirapine (marketed as Viramune). All other drugs in the study were in the same class of drugs as AZT (also known as zidovudine), which is a nucleoside analogue. Some of the drug combinations require as few as two pills a day, because two or three drugs have been combined in a single pill.

AZT/lamivudine/abacavir and AZT/lamivudine require two pills a day, while efavirenz is taken once daily. All patients took the same number of pills daily seven, including placebos so neither they nor the study staff would know what treatment regimen they were taking.

The study planned to follow patients for 96 weeks (almost two years). Treatment failure was defined as two blood tests in a row that showed there were more than 200 copies of HIV RNA (genetic material) per milliliter of blood. At about 32 weeks, the researchers discovered that patients taking the simpler triple nucleoside regimen were more likely to have treatment failure than those taking an efavirenz-containing regimen. Overall, 21% of those taking three nucleoside analogues had treatment failure at that point compared with only 11% of those taking an efavirenz-containing drug regimen.

In February 2003, the National Institute of Allergy and Infectious Diseases (NIAID) Data and Safety Monitoring Board, an independent group that monitors the study results, recommended that this arm of the study be halted. Patients taking the AZT/lamivudine/abacavir combination were switched to other regimens.

The Board recommended that we stop that arm of the study and go aheadand present the results to the treating community, said Dr. Gulick. TheNational Institutes of Health issued a Dear Doctor letter to alertphysicians and other health-care providers to the findings.

Prior to this study there were other studies that supported the potencyof the triple nucleoside combination. The simplicity of the drug combinationled a lot of people to consider it among the first-line treatments for HIV, whichwas endorsed by treatment guidelines, he said. Our results have leddirectly to a change in the current treatment guidelines.

The other two arms of the study are ongoing, and it's not clear if one of the efavirenz-containing regimens is superior to the other. Although the study did not include protease inhibitors, if treatment failure occurs with one of the treatment regimens in the study, it would be logical to switch a patient to this class of drugs, Dr. Gulick said.

Our study was designed to avoid protease-inhibitor containing regimensinitially because although they are quite potent drugs, they can be quite complexto take, and most involve taking a number of pills several times a day, hesaid. They have also been associated with liver toxicity and a fat redistributionsyndrome that is very troubling to patients, so we looked for simple, yet potentregimens that avoid these toxicities.

You could start with an efavirenz-based regimen, and if that failed, atsome point, you could move to a protease-inhibitor-based regimen, he said. Wecurrently have another study going on where we compare those two approaches head-to-head.

Such head-to-head comparisons really help to advance the field, said Dr. Gulick.

This new study is a positive step toward identifying what is the bestinitial regimen to start when you want to treat HIV infection, he said. Theother thing that this study does is raise caution about regimens that appearto be simpler. While simplicity and a lack of toxicity are important, potencyhas to be the first quality by which we judge a regimen.

Weill Cornell's Dr. Bruce R. Schackman was a co-author of the study, which also included investigators from Harvard School of Public Health, the University of Hawaii, University of Southern California Medical Center, and other centers.

The work was funded by the National Institute of Allergy and Infectious Diseases (NIAID), which is part of the National Institutes of Health.