New computer models predict patient response to HIV drug therapy
The HIV Resistance Response Database Initiative developed the new models for use in settings where genotyping is not affordable. An HIV genotype is a test used in wealthier countries to determine the genetic code of the virus to select drugs that will be the most effective.
The new models estimate the probability that any combination of HIV drugs will reduce the amount of virus in patients whose current therapy is not working. Using data from tens of thousands of patients worldwide, the new models were approximately 80 percent accurate, far more accurate than the 57 percent accuracy achieved using genotyping.
"This study and these models are proof of principle that this could be a very helpful approach for selecting effective therapy in highly resource-constrained settings, such as southern Africa," Robin Wood, a co-author on the paper, said. "As more of our patients fail first and even second line therapy, it is critical to optimize the selection from our limited range of drugs to achieve maximum suppression of the virus and this system could be very useful."
The system requires a measure of the amount of virus in the patient's bloodstream, also known as the viral load. While the viral load test is not widely used in resource-limited settings, the potential cost savings offered by the system could cover the costs of viral load testing many times over.
The new models are now freely available online to be used by healthcare professionals as part of the RDI's HIV Treatment Response Protection System.
"Currently, most HIV patients in resource-limited settings are treated according to WHO public health guidelines that offer very limited treatment options," Hugo Tempelman, a co-author on the paper, said. "The HIV-TRePS system, incorporating these models, enables doctors to tailor the HIV treatment based on the cost and predicted effectiveness of the treatment. What is wonderful is that HIV-TRePS provides us with high predictive value at no cost."