Related Experiment Video
Updated: Jun 22, 2026
19:57
An Affordable HIV-1 Drug Resistance Monitoring Method for Resource Limited Settings
Published on: March 30, 2014
Predictive value of HIV-1 genotypic resistance test interpretation algorithms
Soo-Yon Rhee1, W Jeffrey Fessel, Tommy F Liu
1Division of Infectious Diseases, Dept. of Medicine, Stanford University, Stanford, CA 94305, USA. syrhee@stanford.edu
The Journal of Infectious Diseases
|June 26, 2009
Summary
Interpreting HIV-1 genotypic drug resistance is complex. This study found that regimen-specific genotypic susceptibility scores (rGSSs) from four algorithms strongly predicted virologic response in HIV-1 patients.
Area of Science:
- Virology
- Infectious Diseases
- Clinical Pharmacology
Background:
- Interpreting human immunodeficiency virus type 1 (HIV-1) genotypic drug-resistance test results presents challenges for clinicians.
- Existing drug-resistance interpretation algorithms need validation with current clinical data.
Purpose of the Study:
- To evaluate the predictive value of four HIV-1 drug-resistance interpretation algorithms.
- To assess the ability of genotypic susceptibility scores (GSSs) to predict virologic response (VR).
Main Methods:
- Examined 734 treatment-change episodes (TCEs) for predicting virologic response (VR).
- Calculated drug-specific GSSs, weighted GSSs using antiretroviral (ARV) potency factors, and regimen-specific GSSs (rGSSs).
- Used multivariate logistic regression and 10-fold cross-validation to estimate predictive value.
Main Results:
- 65% of TCEs achieved VR.
- rGSSs from the four algorithms were the strongest predictors of VR (adjusted odds ratios 1.6–2.2, P < .001).
- Area under the receiver operating characteristic curve improved from 0.76 to 0.80 with weighted rGSSs.
Conclusions:
- Both unweighted and weighted rGSSs from four genotypic resistance algorithms independently predicted VR.
- Optimizing ARV weighting in GSS calculations may enhance VR prediction accuracy.
