Data to be shared and accessible over the Internet
SEVILLE, SPAIN; 4th July 2002
A radical new initiative in HIV management was launched today at the XI International HIV Drug Resistance Workshop. The International HIV Resistance Response Database Initiative, or RDI, will combine data from thousands of HIV patients around the world into a freely accessible database that could enhance the ability of physicians to choose the most effective drugs for their patients.
"The relationship between genetic changes in HIV and response to drug therapy is highly complex and predicting one from the other requires vast amounts of data", commented RDI founding member Dr Brendan Larder (Chief Scientific Officer, Visible Genetics Inc). "The RDI is the first global attempt to share the clinical data necessary to develop a predictive tool that could be the key to managing HIV drug resistance in the future."
The RDI is made up of an expanding group of individuals, companies and institutions who are pooling a number of key pieces of data from many thousands of patients worldwide, in particular: their 'genotype' (the genetic fingerprint of their HIV), the drugs used in their treatment and their response to those drugs (changes in the amount of virus in the blood). In the future, physicians will be able to run a genotype test and then search this database for information from patients with HIV that is genetically similar to their own patient. They will then be able to retrieve data on the response of these patients to different drugs, enabling them to predict with increased confidence which drugs will work best for their patient.
One of the main reasons why HIV treatment fails is that mutations in the genetic code of HIV cause resistance to HIV drugs. Tests are already available to read these genetic changes ('genotypic resistance tests'). The RDI database is designed to enhance the predictive power of such tests by relating genetic changes in HIV directly to virological response to drug therapy, using the largest collection of clinical data of its kind in the world.
There are already several organizations and individuals involved in the RDI including Visible Genetics, the US Military HIV Research Group, The US AIDS Clinical Trials Group (ACTG), BC Center for Excellence in HIV/AIDS, Vancouver, Canada and the University of Siena, Italy. The group is actively seeking other organizations with appropriate data to participate in this unique global initiative.
"A number of us had been attempting to develop databases independently over the past few years", commented Dr Larder. "No single organization has sufficient data to do the job, however, and the RDI represents a pooling of our resources to make this happen. We all agreed that to maximize the benefit from the database there should be free access for people to interrogate it."
Results presented at the conference from a study conducted by Visible Genetics provided proof of principle that the RDI approach can work. Bioinformatics specialists employed sophisticated proprietary analytical techniques, including neural networks, to model the relationship between genotype and virological response using data from just 400 patients - the first to be added to the database. This resulted in predictions of viral load changes in response to each of the available drugs. Further studies are required to validate the approach. The database now contains information from 3,500 patients and the intention is to collect many times this amount of data making the prediction of response highly reliable.
In a separate study presented today (DiRienzo et al), biostatisticians from Harvard University investigated the amount of data required to make reliable predictions of changes in viral load after 24 weeks of treatment. Using historical data on the relationship between genotype and response to the protease inhibitor amprenavir as the basis for their assumptions, they calculated that data from 2,000 to 4,000 patients would ultimately be required for each drug combination (or combinations with similar resistance profiles) in order to predict moderate effects on viral load.
All the data collected by the RDI will be from studies that have obtained full, informed patient consent, giving permission for the data to be used in a confidential manner in studies and initiatives such as this. In order to comply with the data protection and privacy laws of each country involved, any information relating to the identity of the patient is removed before the data is shared with the RDI, so patient confidentiality is assured.
Date published: 4th July 2002