• Switching Protease Inhibitors to Nevirapine (NEV), Efavirenz (EFV) or Abacavir (ABA): A Randomized, Multicenter, Open-Label, Simplification Trial (WeOrB1262)
  Authored by E. Martinez, D. Podzamczer, E. Ribera, P. Domingo, H. Knobel, D. Dalmau, M. Riera, J.M. Gatell
  View the original abstract

This was an update by Dr. Martinez reporting the 18-month results of this trial. It was previously presented at the 9th Conference on Retroviruses and Opportunistic Infections (CROI) Late Breaker session in Seattle, Washington, in February 2002. To date this has been one of the best switch studies presented.

Subjects were enrolled between December 1999 and January 2001. Eligible subjects must have been on at least one PI plus two NRTIs with plasma HIV-1 RNA less than 200 copies/ml. Subjects were to continue their two NRTIs and switch their PI according to random assignment to one of three arms, nevirapine (NEV) (n = 155), efavirenz (EFV) (n = 156) or abacavir (ABA) (n = 149). Final enrollment in the three arms were as follows: NEV = 117, EFV = 107 and ABA = 117.

Subjects were followed every three months for a total duration of 24 months. The majority of patients had been on an indinavir-containing regimen prior to switch. The primary endpoint was the proportion of subjects whose viral load remains less than 200 copies/ml. Baseline characteristics did not differ among the three arms. The median age was 40, 75 percent were male, one third were classified as having AIDS and the median CD4+ count was 500 cells/mm³.

There was a lot of enthusiasm after Dr. Martinez's presentation of the 12-month data in CROI, in which he reported no significant difference among the three arms virologically. Intent-to-treat (ITT) analysis: NEV 78 percent, EFV 74 percent, ABA 77 percent. On-treatment (OTT) analysis: NEV 94 percent, EFV 94 percent, and ABA 87 percent. In addition, at 12 months, Dr. Martinez reported less adverse effects (AE) and an improvement in the total cholesterol among patients randomized to the ABA arm at 12 months.

However, we are beginning to see a virologic difference among the three arms at 18 months. Although there was no significant difference in the ITT analysis, there was a significant difference in the OTT analysis between the NNRTIs and ABA. There is a higher probability of maintaining virologic control in the NEV and EFV arms as compared with ABA. The number of subjects with viral loads greater than 200 copies/ml at 18 months are as follows: NEV = 8, EFV = 7 and ABA = 23.

However a point to consider is the previous antiretroviral history of the subjects. Twenty-one out of twenty-three ABA failures had received sub-optimal therapy previously compared with 5/8 in the NEV arm and 6/7 in the EFV arm. In fact, the outcomes of the three arms were similar when one considers switching from first-line PI-based regimens only. This was clearly evident when one reviewed the genotype data which revealed 12/23 of the patients had nucleoside analog mutations compared with only four in the NEV arm and one in the EFV arm. There was no significant difference between the arms regarding CD4+ counts. The adverse experiences (AE) could be predicted. The predominant AEs were: rash (n = 13) in the NEV arm; neuro-psychiatric (n = 23) in the EFV arm; and gastrointestinal (n = 5) and hypersensitivity (n = 4) in the ABA arm. There was no significant difference among arms in regards to triglycerides; however, there was a significant decrease in total cholesterol in the ABA arm. The clinical significance of this is unknown given the vast majority of patients had normal triglycerides and total cholesterol on entry and throughout the study. Although there were trends for improvement in both fat accumulation and lipoatrophy, there was no significant improvement in any of the arms.

This is an important study. The 24-month data will be important to see the durability of the switch virologically. In fact, I hope the team considers extending the study. So far it appears that the three arms are similar in the proportion of subjects who have remained virologically suppressed. Although Dr. Martinez reports a significant difference with more virologic failure among the ABA arm, it is critical to remember that this is among subjects who previously had received suboptimal therapy -- presumably nucleoside mono or dual therapy. For patients with first time PI-failures, the data is encouraging that all three arms have remained virologically suppressed. It is not surprising that no differences were noted metabolically. The vast majority of these patients had normal lipids and the lipodystrophy was based upon self-report. One will need a population with known dyslipidemias, insulin resistance and body fat alterations in order to determine if switching off a PI will be beneficial.