RDI’s system for predicting how HIV and AIDS patients

Two multinational clinical studies indicate that the RDI’s system for predicting how HIV and AIDS patients will respond to different drugs could be a useful tool with potential clinical and economic benefits. The studies, published in the January issue of AIDS Patient Care and STDs, involved highly experienced physicians in the USA, Canada and Italy who used the system to help them select the optimum combination of HIV drugs for patients whose therapy was failing.

The HIV Treatment Response Prediction System (HIV-TRePS) harnesses the power of complex computer models that have been trained with data from thousands of patients around the world. In these studies, physicians entered their patient’s data and their selection of the next combination of HIV drugs, via the Internet. A prototype version of HIV-TRePS predicted how the patient would respond to hundreds of alternative combinations of HIV drugs. Within seconds, the physician received a report listing the drug combinations that the models predicted were most likely to work. Having reviewed the report, the physicians recorded their final treatment decision and completed an online evaluation.

The results demonstrated that use of the system was associated with a change of treatment decision in one-third of cases to combinations with fewer drugs overall, which were predicted to result in better virological responses. Evaluations indicated that the physicians found the system to be easy and useful. Based on these findings, use of the system could potentially improve patient outcomes and reduce the overall number – and therefore cost – of drugs used. An improved version of HIV-TRePS is now available free of charge over the Internet (via the RDI web site) as an experimental tool.