• Effect of Amprenavir (APV) Hypersusceptibility on the Response to APV/Ritonavir (RTV)-Based Therapy in ART-Experienced Adults Selected by Baseline Susceptibility (ESS40006): 24-Week Data (Oral 143)
  Authored by R. Schooley, R. Haubrich, M. Thompson, D. Margolis, S. Schneider, D. Richman, K. Pappa, L. Yau, S. Hessenthaler, J. Hernandez
  View the original abstract

Hypersusceptibility has been reported for different classes of drugs including for NRTIs, NNRTIs and now protease inhibitors. For non-virologists reading this, we say that a virus is hypersusceptible if the amount of drug it takes to inhibit it in a phenotypic assay is less than the amount of drug it takes to inhibit wild type virus.

Normally, the dosing of drugs has been calculated to reach levels high enough in blood to inhibit replication of wild type virus. If hypersusceptibility really does occur in patients, these hypersusceptible viruses would be "killed" easier than regular wild-type viruses. But the notion of hypersusceptibility is a result of an in vitro test, done in a plate, not in a human. Thus, the important question is if hypersusceptibility occurs in patients or if it is some sort of a test tube artifact and has no clinical significance. If it really does occur, it might be important how we sequence drugs and how we select drugs when we need to "rescue" patients from virologic failure.

This study was a substudy of a study that had already been presented. In this study, subjects who had failed a previous protease inhibitor-containing regimen received abacavir (ABC, Ziagen), EFV (EFV, Sustiva) (if they have never received an NNRTI) or tenofovir (TDF, Viread) (if they have received an NNRTI). Most of the people who were naive to EFV became undetectable and were excluded from the analysis.

The proportion of the NNRTI-experienced subjects who achieved undetectable virus was 64 percent (38/59, ITT obs.) and 50 percent (38/76, ITT M=F). A predictor for the success of the new regimen included baseline viral load, which is not a big surprise. In almost all the antiretroviral trials it predicts the probability of virologic response in treatment-experienced patients and also in a majority of trials in naive patients (especially in the pre-EFV era).

The other predictor for success was hypersensitivity to amprenavir (APV, Agenerase) (defined as having a score less than 0.66 in Virologic's assay) and this remained true even in the multivariate analysis, suggesting that this hypersensitivity phenomenon in protease inhibitors is clinically relevant. Several studies have proved previously that this is also true for hypersensitivity to NRTIs and NNRTIs.

Does all this affect clinical practice? No, not a lot. One cannot make a virus hypersusceptible. It just happens. However, clinicians could pay more attention to phenotypic results when they are requested. Normally the phenotypic test results report the sensitivity of the virus in three categories: sensitive, intermediate and resistant. If among the drugs that a virus is sensitive to there is one drug that the virus is hypersensitive or hypersusceptible to, that drug might be a good choice, because, at least based on this study, the probability of virologic success seems to be a little higher.

One last comment: This study used the old formulation of APV, which will be history as soon as the new formulation GW433908, also known as 908, the calcium phosphate ester pro-drug of APV, becomes available. The rumor is that the U.S. Food and Drug Administration (FDA) will approve 908 this year during the summer. The formulation has better pharmacokinetics and tolerability and will certainly increase the use of APV, a drug currently underutilized because of difficulties with the current formulation.