ICAAC 2005; Washington, D.C.; December 16-19, 2005

Key Links:

By Topic:


See Also
A dual-NRTI backbone for multiple patients >

UNBP3028 1/17

The Body PRO Covers: The 45th Interscience Conference on Antimicrobial Agents and Chemotherapy

Two NRTIs or Three? Makes No Difference When Taken With Efavirenz, ACTG 5095 Finds

December 16, 2005

When prescribing the potent antiretroviral efavirenz (EFV, Sustiva, Stocrin) as part of a first-line treatment regimen, there is no discernible difference between completing the regimen with two nucleoside reverse transcriptase inhibitors (NRTIs) or three NRTIs, according to 144-week results from the landmark ACTG 5095 study. Treatment-naive patients responded excellently, regardless of the total number of medications in their regimen. However, both regimens appeared inexplicably less successful in African-American patients.

ACTG 5095, which enrolled 1,147 patients in the United States, was initially designed to compare three antiretroviral regimens for the initial treatment of HIV infection: 1) zidovudine/lamivudine/abacavir (AZT/3TC/ABC, Trizivir); 2) zidovudine/lamivudine (AZT/3TC, Combivir) + efavirenz; and 3) zidovudine/lamivudine/abacavir + efavirenz. This was a randomized, double-blinded, placebo-controlled study, so neither the patient nor the study staff knew which drugs each patient was receiving. Patients were required to have a plasma viral load above 400 copies/mL at treatment initiation, but could have had any CD4+ cell count. Patients were followed for virologic failure (defined as two consecutive viral loads above 200 copies/mL).

In 2003, Roy Gulick, Director of the Cornell HIV Clinical Trials Unit, and his fellow ACTG 5095 researchers sent shock waves through the world of HIV care when they revealed that the zidovudine/lamivudine/abacavir-only arm of their study yielded poorer virologic outcomes than the other two arms. The findings, which were eventually published in the New England Journal of Medicine,1 were so convincing that the zidovudine/lamivudine/abacavir-only arm was discontinued entirely, and zidovudine/lamivudine/abacavir fell out of favor as a preferred option for initial therapy of HIV. Meanwhile, the two ACTG 5095 arms that contained efavirenz continued on: the "three-drug arm" (zidovudine/lamivudine + efavirenz) and the "four-drug arm" (zidovudine/lamivudine/abacavir + efavirenz). At that point, of course, all patients remaining in the study knew they were taking efavirenz. However, they were still blinded as to whether they were receiving abacavir.

There were 765 patients randomized to the three-drug and four-drug arms. Among this group, 59% were people of color (35% black, 21% Hispanic), 19% were women and 11% had a history of injection drug use. The mean baseline viral load was 4.86 log10 copies/mL and the median baseline CD4+ cell count was 240 cells/mm3. Patients were observed for a median of 144 weeks on study. Ninety-nine percent of the patients actually started their assigned treatment; at the end of follow-up, 61% were still on their original regimen, 11% had discontinued study treatment and were on a study-prescribed second-line regimen, 14% had discontinued study treatment while remaining in follow-up, and 14% had discontinued study treatment for other reasons, including loss to follow-up or death.

The 144-week results showed that, in short, there was no difference in the virologic response between the three-drug arm and the four-drug arm. Overall, 25% of the patients reached protocol-defined virologic failure -- 25% in the four-drug arm and 26% in the three-drug arm, a difference that was not statistically significant. There was no difference even when looking at patients with a baseline HIV RNA above 100,000 copies/mL. Likewise, the mean CD4+ cell count change was similar (up by 291 cells/mm3) at week 144. There was no difference in the rate of Grade 3 or 4 adverse events. Interestingly, 10% of the patients in the four-drug arm (the one with abacavir) and 7% of the patients in the three-drug arm (the one without abacavir) were diagnosed with a hypersensitivity reaction, but this was not a statistically significant difference.

Several noteworthy findings of ACTG 5095 revolved around predictors of virologic failure, about which Dr. Gulick presented extensive information at this conference. Gender, age, baseline CD4+ cell count and baseline viral load were not related to virologic failure in this study.

However, Gulick et al did find that patients who were hepatitis C antibody positive were more likely to have virologic failure (hazard ratio 1.6, P = .04). Even more intriguing was the relationship of race/ethnicity to virologic failure: The researchers found that non-Hispanic blacks were more likely to fail treatment than patients of other races/ethnicities. For example, the risk of virologic failure was 1.7 times higher for non-Hispanic blacks than non-Hispanic whites (P = .003). It is important to note that, even though black patients did not have the same degree of virologic suppression as other patients, they still had excellent results: 80% had a viral load below 50 copies/mL at week 144.

The researchers were at a loss to fully explain the racial/ethnic disparity in treatment outcomes. They said it could not be explained by differences in baseline CD4+ cell count or baseline drug resistance. Black patients did have a shorter time to the occurrence of grade 3 or 4 events, but the proportion of patients reporting excellent adherence was similar in all racial/ethnic groups.

Nonetheless, Gulick et al conducted a post-hoc analysis in which they looked at self-reported adherence at week 12 and the occurrence of subsequent virologic failure. Non-adherent black patients, defined as missing at least one dose in the past four days, were more likely to fail than adherent black, white or Hispanic patients, as expected. What was not expected was that non-adherent black patients were also more likely to fail than non-adherent white and Hispanic patients. The reason for this disparity is a mystery, although it is possible that black patients were non-adherent in different ways than other racial/ethnic groups or that similar rates of non-adherence somehow had different consequences in black patients.

Even with these unusual results for black patients, this study strongly supports the use of two NRTIs + efavirenz as a first-line antiretroviral regimen. The addition of a fourth drug, abacavir, did not improve effectiveness: Although the four-drug regimen was just as safe and tolerable as the three-drug regimen, there was no subgroup for which the four-drug regimen was shown to be superior. Of course, the reasons for the lower rate of virologic suppression in black patients have not been explained fully and will certainly be an area for further research. But it is important to note that this study in no way suggests that two NRTIs + efavirenz should not be used for black patients. The bottom line is that the rates of virologic suppression on this regimen were excellent in all racial and ethnic groups.


  1. Gulick RM, Ribaudo HJ, Shikuma CM, et al, for the AIDS Clinical Trials Group Study A5095 Team. Triple-nucleoside regimens versus efavirenz-containing regimens for the initial treatment of HIV-1 infection. N Engl J Med. April 29, 2004;350(18):1850-1861.

Abstract: ACTG 5095: zidovudine/lamivudine/abacavir vs. zidovudine/lamivudine + efavirenz vs. zidovudine/lamivudine/abacavir + efavirenz for initial HIV therapy (Slide H-416a)
Authored by: R Gulick, H Ribaudo, C Shikuma, C Lalama, B Schackman, W Meyer, K Squires, E Acosta, K Klingman, D Kuritzkes

Affiliations: Weill Cornell Med. Coll., New York, NY, Harvard Sch. of Publ. Health, Boston, MA, Univ. of Hawaii, Honolulu, HI, Quest Diagnostics, Baltimore, MD, Univ. of Southern California, Los Angeles, CA, Univ. of Alabama, Birmingham, AL, Div. of AIDS, NIH, Bethesda, MD, Harvard Med. Sch., Boston, MA

It is a part of the publication The 45th Interscience Conference on Antimicrobial Agents and Chemotherapy.

Please note: Knowledge about HIV changes rapidly. Note the date of this summary's publication, and before treating patients or employing any therapies described in these materials, verify all information independently. If you are a patient, please consult a doctor or other medical professional before acting on any of the information presented in this summary. For a complete listing of our most recent conference coverage, click here.

The content on this page is free of advertiser influence and was produced by our editorial team. See our content and advertising policies.