• The Strategy of Antiretroviral Switch Studies -- A Review
  Presentation 1375
  William G. Powderly

Dr. Powderly undertook a review of the literature regarding switch studies prior to this conference. Switching is done for two reasons:

1. Simplicity

   • Improve adherence

   • Decrease pill burden and/or complexity

   • As a result to improve effectiveness

2. Toxicity

   • Reversal of side effects due to or perceived to be due to a component of the regimen

   • Avoidance of future toxicity

Several strategies have been employed:

For NVP there are two randomized trials and four cohort studies, for EFV, one randomized trial and five cohort studies, and for ABC, two randomized trials and two more detailed metabolic substudies of other trials. Dr. Powderly expressed concern that most of the studies to date have been based on observational data, rather than on metabolic studies. These would include insulin sensitivity, lipid levels in a fasting state, and body composition studies.

Abacavir-switch studies have shown a higher failure rate in patients switched to ABC from a PI, especially in patients with NNRTI use prior to starting HAART. Several studies at this conference pointed to underlying NRTI resistance as the reason for failure. Caution must be exercised in making this switch in experienced patients without genotyping.

Other studies have shown mixed results.

Dr. Powderly's conclusions:

1. Protease inhibitors appear to have a causal role in development of insulin resistance.

2. Protease inhibitors affect triglycerides in some patients.

3. Effect of protease inhibitors on cholesterol metabolism is unclear -- may contribute to decreased HDL in some patients.

4. Most studies on fat redistribution are poorly performed and/or inadequately controlled.

5. Current switch data do not support a major role of protease inhibitors in maintenance of fat loss or gain (although they may be necessary to cause it).

6. Stavudine may have a role in lipodystrophy.

The bottom line about substituting a protease inhibitor with a different class after reviewing current studies was:

1. Appears safe with NNRTI.

2. Caution with ABC especially with prior NRTI use before HAART.

3. Decreased insulin resistance.

4. Usually decreased triglycerides, although not necessarily back to a normal level.

5. Inconsistent effects on cholesterol and HDL.

6. No consistent effect of fat gain or loss.

7. No consistent effect on body habitus.

The only thing we can say so far about substituting d4T with another nucleoside is that it may decrease fat loss.