For the past several years, a few regimens have competed for the title of the "best" regimen with which we should initiate treatment for HIV. For years multiple HIV treatment guidelines have agreed with the well-supported approach of a dual-nucleoside backbone with either efavirenz (EFV, Sustiva, Stocrin) or a boosted protease inhibitor.

But specifically which dual-nucleoside/nucleotide reverse transcriptase inhibitor [NRTI] backbone should be selected has been the focus of numerous, important studies over the last few years. In the past year, the toxicity associated with a few years' use of stavudine (d4T, Zerit) has led to a decline in support for selecting this drug in initial regimens, given the data that supports the contention that we have more effective options.

Another battleground of the NRTIs remaining in our "preferred" lists has been the contrast between zidovudine (AZT, Retrovir) and tenofovir (TDF, Viread). Study 934, presented at ICAAC, allows a judgment of the outcomes when these 2 drugs are compared in a large, randomized study.

In this study, 517 patients were randomized to receive either tenofovir plus emtricitabine (FTC, Emtriva) or the coformulated tablet of zidovudine/lamivudine (AZT/3TC, Combivir). Both arms were taken in combination with efavirenz. The data presented here was a 24-week endpoint, stated as a planned interim analysis.

In this study, the median viral load was about 5 log, and the CD4+ cell count about 233-241. By week 24, 11% of the patients had discontinued the tenofovir arm, while 21% on the zidovudine arm had stopped. Withdrawals due to adverse events was the main difference observed in these 2 arms.

As a result, by intent-to-treat analyses, there was a superior outcome for the tenofovir arm, with 87% of the patients (compared to 78% on the zidovudine arm) achieving suppression to P = .01) and a 95% confidence interval that shows superiority of the tenofovir arm. By a 50-copy cutoff, the arms had a 73% vs. 65% response on the tenofovir and zidovudine arms, respectively.

There was also a superior CD4+ response to tenofovir, with an increase of 129 cells, vs. 111 cells on the zidovudine arm. Resistance outcomes were similar in these arms, and no patient developed either the K65R mutation or any zidovudine "TAMs" (short for thymidine analog-associated mutations).

The adverse event of anemia was further defined for patients who withdrew from the trial due to this outcome: A striking drop in the hematocrit from a mean of 40 at baseline to 22 was documented at the time of discontinuation -- changes that are clinically meaningful. There were no reported changes in serum creatinine on the tenofovir arm, and 2 reported increases in creatinine on the zidovudine arm.

This study will present powerful data to both clinicians who are considering initiation of therapy for their patients and for patients deciding upon their first regimen. While people who do tolerate zidovudine/lamivudine can do well in terms of virologic suppression and, at least short-term safety, there are clear differences in the likelihood of success and toxicity between the zidovudine/lamivudine option and the tenofovir plus emtricitabine option.

On every measurement presented, the arm containing tenofovir was superior for safety, as well as for immunologic outcomes. These data will likely become another benchmark in the history of studies that influenced the field of HIV medicine.