AI systems in clinical practice

Decision support systems
HIV decision support system for genotype interpretation

developed by clinical domains keywords
The decision support technology used in the implementation of RetroGram™ is AREZZO, based on PROforma and developed by InferMed Ltd., London. The medical content was developed by Virology Networks, HIV Pharmacology. RetroGram is jointly owned by Virology Networks and Hoffmann-La Roche. HIV, HIV-Genotype interpretation, drug regimens Decision support systems, antiretroviral drugs, Arezzo, PROforma
location commissioned status
RetroGram™ is used by more than 250 clinicians in European, Australian and US centers in resistance management of over 2,000 HIV-1 infected patients. RetroGram version 1.0 was launched in 1999. In clinical practice. Under continuing development and update - web-enabled v1.6 released in August 2002).

System features

  • RetroGram relates complex genomic information on HIV-1 resistance to the clinical suitability of anti-retroviral drugs (the main task supported by Arezzo technology).
  • RetroGram generates a suitability ranking of all FDA approved drugs based on their expected clinical efficacy for an individual patient.
  • A unique feature of RetroGram is that it includes the experience and opinions of experts in the algorithm to weigh the effect of specific genotypic changes.
  • RetroGram is available both as a stand-alone Windows-based application and in a Web-enabled version.
  • RetroGram's clinical utility was highlighted in a large multi-centre study in Spain. Investigators concluded in their study that "HIV-1 genotyping interpreted by a software package improves the virological outcome when it is added to the clinical information as a basis for decisions on changing anti-retroviral therapy".
  • RetroGram is currently being used in more than ten international comparison studies of all available genotype interpretation systems.

C Tural et al. Clinical utility of HIV-1 genotyping and expert advice: the Havana Trial. AIDS. 2002; 16:209-218.

[PubMed]   []   [Reuters report on the Havana trial]

" OBJECTIVE: To determine whether HIV-1 genotyping and expert advice add additional short-term virologic benefit in guiding antiretroviral changes in HIV+ drug-experienced patients. DESIGN: A two factorial (genotyping and expert advice), randomized, open label, multi-center trial. The patients were stratified according to the number of treatment failures. PATIENTS AND METHODS: HIV-1 infected patients on stable antiretroviral therapy who presented virological failure were included into the study. Genotypic testing was performed by using TrueGene HIV Genotyping kit and the results were interpreted by a software package (RetroGram, version 1.0). An expert advisory committee suggested the new therapeutic approach based on clinical information alone or on clinical information plus HIV-1 genotyping results. Plasma HIV-1 RNA load, CD4+ cell count and adverse events were recorded at baseline and every 12 weeks. RESULTS: A total of 326 patients were included. The baseline CD4+ cell count and plasma HIV-1 RNA were 387 (+/- 224) x 10(6) cells/l and 4 (+/- 1) log(10) respectively. The proportion of patients with plasma HIV-1 RNA < 400 copies/ml at 24 weeks differed between genotyping and no genotyping arms (48.5 and 36.2%, P < 0.05). Factors associated with a higher probability of plasma HIV-1 RNA < 400 copies/ml were HIV-1 genotyping [odds ratio (OR), 1.7; 95% confidence interval (CI), 1.1-2.8; P = 0.016] and the expert advice in patients failing to a second-line antiretroviral therapy (OR, 3.2; 95% CI, 1.2-8.3; P = 0.016). CONCLUSIONS: HIV-1 genotyping interpreted by a software package improves the virological outcome when it is added to the clinical information as a basis for decisions on changing antiretroviral therapy. The expert advice also showed virologic benefit in the second failure group. "

Torti C, Quiros-Roldan E, Keulen W; GenPherex Study Group of the MaSTeR Cohort. Comparison between rules-based human immunodeficiency virus type 1 genotype interpretations and real or virtual phenotype: concordance analysis and correlation with clinical outcome in heavily treated patients. J Infect Dis. 2003 Jul 15;188(2):194-201.

[PubMed]   []

" We compared 2 rules-based genotype interpretation systems and real or virtual phenotype through a retrospective analysis of a prospective trial. Genotypes were determined with VircoGEN II (VIRCO) and were interpreted with either RetroGram 1.4 or TRUGENE HIV-1 (guidelines 3.0) or original virtual phenotype (Virtual Phenotype; VIRCO), as available in the year 2000. Among 188 human immunodeficiency virus (HIV) type 1 isolates, overall concordance (kappa agreement) was observed for the 2 rules-based systems, whereas striking discordances were noted between them and real and virtual phenotype interpretations for stavudine, didanosine, zalcitabine, abacavir, and amprenavir (kappa <0.4). Clinical evaluation of a subset of 173 patients showed that both rules-based sensitivity scores were independently associated with HIV RNA loads <400 copies/mL at week 16 of during-treatment analysis (TRUGENE: odds ratio [OR], 2.90; 95% confidence interval [CI], 1.52-5.52; P=.001; RetroGram: OR, 2.34; 95% CI, 1.21-4.55; P=.012), whereas, in contrast to real or virtual phenotype, interpretations according to biological cut-offs were not (OR, 1.91; 95% CI, 0.77-4.76; P=.162). "

De Luca A, Cingolani A, Di Giambenedetto S et al. Variable prediction of antiretroviral treatment outcome by different systems for interpreting genotypic human immunodeficiency virus type 1 drug resistance. J Infect Dis. 2003 Jun 15;187(12):1934-43.

[PubMed]   []

" To determine the variability of genotypic human immunodeficiency virus (HIV) type 1 drug-resistance interpretation by available expert systems and its clinical implications, 261 subjects for whom a potent antiretroviral regimen was failing who were starting salvage therapy were evaluated. The association of the genotypic susceptibility score (GSS) of the salvage regimen, according to 11 interpretation systems, with HIV RNA outcomes for 6 months was examined. GSS was highly variable, as determined by the different interpretation systems, and showed independent correlation with changes from baseline HIV RNA levels at 6 months with 5 systems--Stanford hivdb, GuideLines 3.0, Retrogram 1.4, HIVresistanceWeb, and Sao Paulo University. Most GSSs predicted virologic response in regimens containing stavudine, lamivudine, efavirenz, or indinavir. Selected systems predicted response in regimens containing didanosine, abacavir, or nelfinavir, and no system predicted outcome of boosted protease inhibitors. GSSs predicted changes in HIV RNA levels better in adherent patients than in nonadherent individuals. Interpretation may be improved, and knowledge should be used uniformly throughout different expert systems. "

contact links


 bullet  RetroGram web site  bullet  Demonstration of RetroGram [OC]  bullet  Demonstration of RetroGram []
Entry on archive: April 10 2002
Last main update: August 21 2002
Search this site


Privacy policy User agreement Copyright Feedback

Last modified:
© Copyright OpenClinical 2002-2011