• Long-Term Efficacy and Tolerance of Switching the Protease Inhibitor for Non-Nucleoside Reverse Transcriptase Inhibitors: A 52-Week, Multicenter, Prospective Study (Poster 673)
  Authored by J. L. Casado, J. Arrizabalaga, A. Antela, J. A. Iribarren, A. Moreno, F. Dronda, M. J. Perez-Elias, and S. Moreno for the BEGIN Study
  View the original abstract

Many HIV-infected individuals have benefited from years of treatment with combination antiretroviral therapy. The majority of patients started on treatment in previous years received protease inhibitors (PI) as part of their treatment. Protease inhibitor treatment is effective and has been unequivocally associated with decreases in the death- and complication-rates for those with advanced disease. Unfortunately, protease inhibitor-based therapy can have down sides, for example, complex dosing, high pill counts, gastrointestinal side effects. These adverse aspects to therapy can compromise long-term adherence for some patients. For these patients, treatment simplification (e.g., minimization of pill counts, and avoidance of side effects) may offer improvements in quality of life. One strategy for treatment simplification is a switch from protease inhibitor to other classes of antiretroviral medications. Treatment switches to non-nukes (NNRTIs) can offer decreases in pill burden and avoidance of some symptoms.

J.L. Casado presented this Spanish study. One hundred HIV-infected persons on stable dual NRTI/protease inhibitor therapy were non-randomly switched off their protease inhibitor to either efavirenz (EFV) or nevirapine (NVP). Nevirapine was administered with the usual dose escalation. The median CD4 count of these patients was 489, time on PI was 19 months, and had a viral load <50 for 12 months. Nearly all of these patients had intolerance or toxicity from their PI. Thirty-six subjects received EFV, 64 subjects received NVP. After 12 months of NNRTI therapy, 65% of EFV subjects and 62% of NVP subjects had viral loads less than 50 copies (intent to treat analysis; missing = failure). Twenty-seven patients developed symptomatic adverse events, though few required treatment discontinuation. Virologic failure occurred in 17% of NVP patients, 8% of EFV subjects. Lipid levels improved somewhat, with cholesterol decreasing from 241 to 197 (NVP) and 211 (EFV). Triglycerides also decreased from 287 to 217 (NVP) and 144 (EFV). Self reported adherence (defined as >95%) improved modestly, from 91% to 96% (NVP) and 98% (EFV).

The study's conclusions are limited by its non-randomized nature and the lack of documentation reasons for study dropout. How did patients wind up on particular treatment arms? It would be of interest to know how many of the virologic failures had low-level viremia and how many had high-level viremia.

This study offers only a small additional piece to the experience base of simplification from PIs to NNRTIs. Unfortunately, the relatively high rate of treatment failure and lack of full patient accounting limits our ability to assess the degree of relevance to North American treatment. A study that more completely addresses this question is the large DuPont DMP-049 study. This study has been reviewed elsewhere in The Body's conference coverage (see Dr. Cohen's review of abstract 20). In this study, patients with undetectable viral loads on protease inhibitor therapy were switched to EFV, with successful maintenance of virologic suppression and improvements in adherence. It is important not to unnecessarily fix a regimen that is not broken -- switches may place a patient at risks of new side effects or hidden drug resistance. However, when a switch is carefully performed with the back up of good clinical trial data, simplification of antiretroviral therapy can be an important tool in patient management.