Poster presentation at:
3rd IAS Conference on HIV Pathogenesis and Treatment
24th July 2005 - 27th July 2005
Rio de Janeiro, Brazil
Larder BA1, Wang D1, Revell AD1, Harrigan R2, Montaner J2, Wegner S3, and Lane C4.
1: The HIV Resistance Response Database Initiative (RDI), London , UK ; 2: The BC Centre for Excellence in HIV/AIDS Vancouver , BC , Canada ; 3: US Military HIV Research Program, Rockville , MD , USA ; 4: National Institute of Allergy and Infectious Diseases (NIAID), Bethesda , MD , USA .
Objectives: A shortcoming of genotypic resistance tests is their limited sensitivity for detection of minority populations of resistant virus. Artificial Neural Networks (ANN) can successfully predict virological response to combination therapy from genotype but their accuracy may be limited by this shortcoming. This study was designed to address whether inclusion of treatment history data, as a 'surrogate' for minority mutant populations, can increase the accuracy of ANN in predicting response to combination therapy.
Methods Two 'committees' of 10 ANN were trained using 2,559 TCEs. The 'basic' committee was trained using input variables of genotype, baseline viral load (VL), drugs in the new combination regimen, time to follow-up and the output variable of VL response. The 'treatment history' models were trained with these variables plus four additional treatment history input variables: AZT, 3TC, any NNRTI and any PI. The two committees were tested using 51 TCEs from 23 patients, partitioned at random as an independent test set . During testing the ANN models were given the input variables of the test TCEs and required to predict the VL response. The models' performance was assessed in terms of the 'committee average' (the average VL prediction of all 10 models for each test TCE).
Results Correlations between predicted and actual VL change for the basic and treatment history committees gave r@ values of 0.30 and 0.45 respectively (p<0.01). The mean absolute difference between predicted and actual VL response was 0.88 and 0.78 respectively (p=0.05).
Conclusions The addition of treatment history data significantly improved the accuracy of ANN confirming that historical exposure to antiretroviral drugs can influence response to a new regimen. This suggests that treatment history may act as a surrogate for minority mutant populations and can enable ANN to overcome this shortcoming of current genotyping.
New predictive models accurately predict HIV treatment response despite missing data
Predicting HIV therapy response despite missing data - new models published
2021 update to HIV-TRePS: a highly flexible and accurate system for the prediction of treatment response from incomplete baseline information in different healthcare settings
New models that predict treatment outcome without the need for a baseline CD4 count or genotype added to HIV-TRePS
The most accurate models to date for predicting response to HIV therapy are released for clinical use
RDI expands the range of models available through HIV-TRePS to include their most accurate models to date
AI models predict absolute virological response to HIV therapy over time