Researchers in London announced October 6 the launch of a free-of-charge, online computer modeling system to help providers choose the antiretroviral (ARV) combinations that will most likely work for their patients.

For people starting their first ARV regimen, the choices are usually pretty clear. Given that all of the combinations recommended by the Department of Health and Human Services HIV treatment guidelines panel have proved effective at controlling HIV, the choice usually comes down to which combination will be easiest to take and have the least troubling side effects.

For people who are on their third, forth or fifth regimens, however—or those who are infected with drug-resistant HIV—treatment decisions can be challenging. Currently, providers must sort through the often complex results of genotype tests, which identify the drug-resistant mutations a person’s HIV carries, along with a person’s treatment history to determine what combinations are most likely to work. In the most complex cases, providers often consult with resistance experts at university academic centers.

Now, a group of researchers belonging to a non-profit group called the Response Database Initiative has launched a new online computer modeling program to help providers make more accurate predictions about the best regimens for their patients. Called the HIV Treatment Response Prediction System (HIV-TRePS), the system uses computer simulations based on the medical records of over 70,000 people with HIV around the world to determine which treatments will be best for a specific individual.

To use the system, a provider enters a person’s genotypic test results, viral load, CD4 count and treatment history. Within seconds, the system returns to the provider a list of potential combinations, sorted by their potency and tolerability.

“This is a very exciting development—the system literally puts the experience of treating thousands of different patients at the doctor’s fingertips,” commented Julio Montaner, MD, from the BC Centre for Excellence in HIV & AIDS in Vancouver. “This has the potential to improve outcomes for people living with HIV and AIDS around the world, particularly where resources and expertise are scarce.”

HIV-TRePS is only about 78 percent accurate in its predictions, and its developers caution that it should not be substituted for expert medical guidance. It has, nevertheless, proven more effective than other available methods, which have about 54 percent accuracy, for predicting which regimens will be most likely to work.

“We are really excited about the launch of this system, which is a milestone for us, our research partners around the world and also for the use of bioinformatics in medicine,“ said Brendan Larder, PhD, scientific chair of RDI. ”We believe this approach can make a significant difference in a variety of settings and diseases."