Resistance profile of etravirine: combined analysis of baseline genotypic and phenotypic data from the randomized, controlled Phase III clinical studies.

OBJECTIVE

To refine the genotypic and phenotypic correlates of response to the nonnucleoside reverse transcriptase inhibitor etravirine.

DESIGN

Initial analyses identified 13 etravirine resistance-associated mutations (RAMs) and clinical cutoffs (CCOs) for etravirine. A multivariate analysis was performed to refine the initial etravirine RAM list and improve the predictive value of genotypic resistance testing with regard to virologic response and relationship to phenotypic data.

METHODS

Week 24 data were pooled from the phase III studies with TMC125 to Demonstrate Undetectable viral load in patients Experienced with ARV Therapy (DUET). The effect of baseline resistance to etravirine on virologic response (<50 HIV-1 RNA copies/ml) was studied in patients not using de-novo enfuvirtide and excluding discontinuations for reasons other than virologic failure (n = 406). Clinical cutoffs for etravirine were established by analysis of covariance models and sliding fold change in 50% effective concentration (EC50) windows (Antivirogram; Virco BVBA, Mechelen, Belgium). Etravirine RAMs were identified as those associated with decreased virologic response/increased etravirine fold change in EC50. Relative weight factors were assigned to the etravirine RAMs using random forest and linear modeling techniques.

RESULTS

Baseline etravirine fold change in EC50 predicted virologic response at week 24, with lower and preliminary upper clinical cutoffs of 3.0 and 13.0, respectively. A fold change in EC50 value above which etravirine provided little or no additional efficacy benefit could not be established. Seventeen etravirine RAMs were identified and attributed a relative weight factor accounting for the differential impact on etravirine fold change in EC50. Virologic response was a function of etravirine-weighted genotypic score.

CONCLUSION

The weighted genotypic scoring algorithm optimizes resistance interpretations for etravirine and guides treatment decisions regarding its use in treatment-experienced patients.