Poster presentation at:
3rd IAS Conference on HIV Pathogenesis and Treatment
24th July 2005 - 27th July 2005
Rio de Janeiro, Brazil
Revell AD1, Larder BA1, Wang D1, Wegner S2, Harrigan R3, Montaner J3 and Lane C4.
1: The HIV Resistance Response Database Initiative (RDI), London, UK; 2: US Military HIV Research Program, Rockville, MD, USA; 3: The BC Centre for Excellence in HIV/AIDS Vancouver, BC, Canada; 4: National Institute of Allergy and Infectious Diseases (NIAID), Bethesda, MD, USA.
Objective: To compare 'local' Artificial Neural Network (ANN) models trained with data from a single clinic with 'global' models trained with data from over 200 clinics in terms of the accuracy with which they predict virologic response from genotype, viral load and drug regimen.
Methods 40 and 38 treatment change episodes (TCEs) from clinics A and B, were partitioned from the RDI database as test data sets. Ten local ANN models were trained with the remaining data from clinic A (337 TCEs) and ten global models trained with 3,168 TCEs from >200 clinics, including A. The models' predictions of viral load (VL) response were compared with the actual VL responses in the two test sets. The ANN models were also used to identify alternative regimens for treatment failures from clinic A
Results Test set A: Correlations between predicted and actual VL gave r@ values of 0.70 vs 0.78 for global and local models respectively and the mean absolute difference between predicted and actual VL was 0.55 vs 0.49 logs (ns). Test set B: Correlations between predicted and actual VL gave r@ values of 0.23 vs 0.07 and the mean differences between predicted and actual VL were 0.66 vs 0.97 logs (p<0.05) for the global and local models respectively.
Both sets of models identified effective alternative regimens for the 19 treatment failures. The mean reduction in VL predicted for the best alternative was 1.83 logs and 1.71 logs for the global and local committees respectively (ns).
Conclusions Global ANN models can perform as accurately as local models, trained with data from a single clinic, in predicting response for patients from that clinic. Global models appear superior to local models for other clinics, suggesting that they may be the most powerful way to exploit ANN as a generally available treatment decision-making tool.
