May/June 2000, Vol. 17, No. 5 & 6 Software To Help In Fight Against HIV/AIDS "Virtual" HIV drug resistance testing now available VirtualPhenotype, a new quantitative approach to HIV resistance testing, is now available in the United States via Laboratory Corporation of America Holdings (Burlington, N.C.; www.labcorp.com). Developed by Virco (Mechelen, Belgium; www.vircolab.com), a multinational biotechnology company, the new system uses the company's proprietary interpretation software to interrogate one of the largest databases of resistance information and predict which drugs the patient's HIV will respond to and which it will be able to resist, based on the genetic code of their virus. HIV/AIDS mutates very readily as it replicates, and any mutants that are resistant to drug therapy can swiftly become the dominant strain. This is one of the main causes of HIV treatment failure. Resistance information can be used to help select the optimal combination of drugs for the individual patient. Currently, two main approaches are used to assess HIV drug resistance in a patient. One method examines the ability of the virus to replicate in the presence of each of the available drugs and is a direct measure of resistance — the phenotype. The other approach is to 'read' the genetic code of the patient's virus — the genotype. This reveals which 'letters' of the code have mutated. However, given the existence of over 100 mutations known to cause HIV drug resistance, the interpretation of genotypic data can often be extremely complex. The Virco test first analyzes the patient's genotype, then searches the company's database of over 65,000 HIV samples for previous genotypes with the same patterns of mutations. It then retrieves the phenotypes that correspond to these samples, averages the information for each drug and predicts to which drugs the current sample will be sensitive or resistant. This is called the VirtualPhenotype. An analysis of over 10,000 patient samples showed that the VirtualPhenotype is highly reliable, with discordance between the VirtualPhenotype and the real phenotype in less than 1 percent of cases. Additional data also shows that the VirtualPhenotype can be more reliable than the current rule-based interpretation approaches that simply use lists of mutations for each drug to predict resistance or sensitivity. Table of Contents Web Site © Copyright 2000 by Lionheart Publishing, Inc. All rights reserved. Lionheart Publishing, Inc. 2555 Cumberland Parkway, Suite 299, Atlanta, GA 30339 USA Phone: 770-431-0867 | Fax: 770-432-6969 E-mail: lpi@lionhrtpub.com Web: www.lionheartpub.com