Non-nucleoside reverse transcriptase inhibitors (NNRTIs) remain popular and important drugs in both initial regimens and for salvage. Nevirapine (NVP, Viramune) and efavirenz (EFV, Sustiva) each have advantages and disadvantages, but the direct head-to-head comparison study 2NN was eagerly awaited when it was presented last winter at the Retrovirus meeting.

The study essentially showed equivalent efficacy with counterbalancing differences in toxicity. Still, the study did not put all issues to rest. It was stopped at 48 weeks, leaving the possibility that subtle differences in drug potency might have been detectable if the study had been larger or longer. While a scientifically valid question, I think it is not clinically relevant, since the difference in side effects and personal preference are likely to be more important.

This study, however, tried to address the potency of the regimens through a different approach. David Ho and Alan Perelson showed through modeling that the early rate of drop in viral load (phase I viral delay) would be directly related to the potency of the regimen. The 2NN investigators reported a study comparing the phase I viral decay between the regimens. As you may recall, 2NN compared nevirapine twice daily with nevirapine once daily, efavirenz once daily and the combination of nevirapine and efavirenz. There were no major differences between the once-daily and twice-daily nevirapine arms so they were combined for this analysis.

Viral decay was calculated by frequently measuring the viral load during the first two weeks of treatment on day 0, 3, 7 and 14. This is a crude way to calculate decay compared to daily or every-other-day measurements, but the advantage of this study is the large number of patients.

Viral decay was faster in the first week than the second, falling almost 1.5 log in the first week. Viral decay was faster among those with higher baseline viral loads and in those with lower CD4 counts. There was no difference between the treatment regimens in viral decay rate.

Perhaps most interestingly, there was no relationship between the rate of viral decay and the probability of viral rebound. This was the only finding that was not, in my mind, fairly predictable.

What can we conclude from this intensive substudy? First, all four regimens were very potent, as predicted from the overall results. Second, there were no detectable differences in the decay rate, which confirms the overall results of the study, showing no detectable difference in efficacy. Third, although viral decay rate is useful to look at initial potency, and probably to make sure there is no hidden viral resistance to the regimen, it does not appear to be, in this study, a good predictor of long-term success.

As I said after the initial 2NN results, there is still room for marketers to battle here, but for doctors and patients, both nevirapine and efavirenz are extremely useful medications, as well as generally safe, and offer good options. The art is to understand the differences in side effects and their timing, and to try to find the best initial drug for each patient. Then, if one doesn't work, often the other drug will. An ominous issue on the horizon, however, is the increasing rate of transmitted resistance to NNRTIs.