Meta-analysis supports triple therapy for HIV infection
Analysis by British researchers of 54 randomised control trials involving more than 20,000 patients strongly supports the use of triple therapy as the initial treatment for HIV-infected patients.
The University of Birmingham collaborators searched six electronic databases and citation lists and consulted pharmaceutical companies for trials of HIV-seropositive patients at least 12 years old. The research reports included trials lasting 12 weeks to nearly five years. Dr Rachel Jordan and associates report their findings in the British Medical Journal for March 30.
Triple therapy with two reverse transcriptase inhibitors (RTIs) and a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor reduced the risk of progression or death to 0.6 times that of patients receiving therapy with two RTIs. Dropout rates were similar between triple and double therapy if no protease inhibitor was involved. However, trial arms that included a protease inhibitor had significantly higher withdrawals.
Dr Jordan’s team notes that “no fully published evidence on the effectiveness of quadruple or higher combination” has been presented as of February 2001.
Dr Charles Carpenter, of Brown Medical School in Providence, Rhode Island, comments in an associated editorial that patients who adhere to triple therapy do so well that it is difficult to show a significant benefit to a four-drug regimen.
“However, a sound scientific rationale exists for using an initial four drug regimen that includes two protease inhibitors,” he adds. Dr Carpenter recommends including a low dose of ritonavir, in addition to a second protease inhibitor, to provide a booster effect while lowering the risk of viral resistance.
- Jordan R, Gold L, Cummins C et al. Systematic review and meta-analysis of evidence for increasing numbers of drugs in antiretroviral combination therapy, BMJ 2002; 324:757-760
- Carpenter C. Editorial; Initial antiretroviral regimens: In general three drugs are better than two are better than one, BMJ 2002;324:747-748