September 27, 2002
Use of protease inhibitors has been linked with improved survival and fewer AIDS complications among persons with HIV infection. The first protease inhibitor treatment regimens were complicated, however, with high pill burden, dietary restrictions, side effects and toxicities. There has been a considerable move towards the use of non-nucleoside inhibitor (NNRTI)-based regimens as a strategy to offset the limitations of protease inhibitors (PIs). The two most commonly prescribed NNRTIs are efavirenz (EFV, Sustiva) and nevirapine (NVP, Viramune); the advantages of switching include the lower pill counts, dosing frequency and side effect profiles.
Several large prospective studies have demonstrated the long-term potency of efavirenz-based regimens in direct comparison with several unboosted protease inhibitors; a largely unanswered question is whether these two NNRTIs are equivalent in potency. The optimal study to answer this question (a head-to-head, randomized, prospective study) is underway, but the results will be months away. Until that time, we must rely on less rigorous, retrospective studies (ones that look at how patients have done once started on medications, but not studies that randomly assign patients to receive treatment) to provide guidance.
In this retrospective study, 130 HIV-infected persons from France or Portugal who were on successful PI-based treatment and previously switched to either efavirenz (54 persons) or nevirapine (76 persons) were identified. After two years of follow up, there were only subtle differences between the two groups, with 80 percent of subjects sustaining virologic control. There was a greater proportion of treatment failures (due to adverse effects) in the efavirenz group, compared with the nevirapine group.
Overall, the study shows that a strategy of switching persons on successful PI regimens to either NNRTI should be safe. A well-designed previous study from DuPont (now Bristol-Myers Squibb) called 049 showed that a switch from PI to efavirenz resulted in improvements in virological control and adherence. Differences in the side effect profile of these two NNRTIs is well known -- these data show that treatment discontinuations because of unacceptable neuropyschiatric side effects of efavirenz can lead to treatment discontinuation for some persons. The study is limited in strength because of the relatively small size; because of the retrospective nature of the study, the possibility of bias in patient or treatment selection cannot be ruled out. Nevertheless, the long period of follow up provides useful information. The study does not address the important question of which NNRTI is better to start among treatment-naive persons; the ongoing 2NN study, which is prospective and randomized in nature, should help answer this point. In day-to-day clinical practice, these data are important in that they suggest comparable performance in this "stable-switch" situation. For persons with unacceptable PI-induced side effects, an NNRTI switch to either drug appears to be a reasonable and evidence-based approach to improve the burden of treatments.