"The good news is that transmission of drug-resistant HIV will not become a major public health problem," said Dr. Sally Blower, lead author and UCLA professor of biomathematics and AIDS Institute member. "The bad news is that the prevalence of drug-resistant HIV is already high and will continue to substantially increase."
Antiretroviral drugs currently offer the best means for controlling the progression and symptoms of HIV disease. But combination drug therapy, or the triple-drug "cocktail," demands a complicated dosage regimen that is difficult to maintain and often provokes severe side effects.
According to the authors, physicians treating people with HIV may unwittingly contribute to the drug-resistant epidemic if they don't recognize the risks associated with incorrect use of the antiretroviral medications.
"These drugs are as dangerous as chemotherapy," warned Dr. James Kahn, UCSF associate professor of medicine and last author of the study. "General practitioners should not be using them. You really need a skilled HIV specialist to prescribe the medications and closely monitor the patient's adherence and response to treatment."
Blower's team used a mathematical model to understand the evolution of drug-resistant HIV strains in the San Francisco gay community from 1996 to 2001, and to predict the epidemic's growth from 2001 to 2005.
Their theoretical model included such variables as the number of infected drug-sensitive cases, the treatment rate, increases in risky sexual behavior and the rate at which drug-resistant strains of HIV emerge during treatment. Blower's team modeled the evolution of 1,000 different strains of drug-resistant HIV.
Blower's team estimated that only 3 percent of cases in San Francisco were drug-resistant in 1997. However, by 2005, they predict that 42 percent of all HIV cases will be drug-resistant.
Using their mathematical model, the research team determined that the rise in the number of drug-resistant cases was mainly due to the conversion of drug-sensitive cases to drug-resistant cases during antiretroviral therapy. Sexual transmission of drug-resistant virus did not -- and will not -- play a major role in fueling the epidemic of drug resistance. Blower's team estimated that in 2000, only 8 percent of the new HIV drug-resistant infections were due to transmission of resistant strains.
"In the future, the vast majority of new HIV infections will still be drug-sensitive," Blower said. "We predict that even in 2005, only 16 percent of new infections will be drug-resistant." The team also determined that the transmission of drug-resistant strains has not increased, and will not increase, the overall number of new HIV infections.
Blower and her co-authors point out that physicians and policymakers can take steps to minimize the prevalence and the transmission of drug-resistant HIV. Based upon their findings, they recommend four epidemic-control strategies:
1. Delay drug treatment as long as possible in order to maximize the medical benefit and reduce side effects.
2. Create clinical centers of excellence for HIV/AIDS treatment to most effectively limit the rate of acquired drug resistance.
3. Develop therapies more effective for treating patients with drug-resistant viral strains.
4. Reduce the amount of time a drug-resistant patient is on ineffective therapy.
Despite the predicted high prevalence of drug resistance, the authors emphasize that people shouldn't consider their findings an argument against antiretroviral drug treatment in San Francisco or in developing countries.
"We have shown that the surging number of drug-sensitive HIV cases -- which are due to increases in high-risk sex -- pose a substantially greater public health problem than the transmission of drug-resistant virus," Blower said.
Based on their findings in San Francisco, the researchers strongly advocate the expanded use of antiretroviral drugs in developing countries. However, they caution that these therapies must be administered carefully and coupled with effective efforts to reduce the risk of infection.
"Antiretroviral treatment will do the most good when the patient is ready to follow it. But the optimal timing is a real unknown," Kahn said. "We need scientifically proven guidelines to help HIV specialists work with their patients in making this complicated decision."
The UCLA/UCSF study is discussed in an accompanying News & Views in the same issue of Nature Medicine entitled, "Will the drugs still work? Transmission of resistant HIV," by Andrew Philips of the Royal Free and University College Medical School, London.
The National Institute of Allergy and Infectious Diseases, a branch of the National Institutes of Health, and the University of California AIDS Research Program funded the study. Nick Aschenbach and Hayley Gershengorn, research assistants in Blower's lab, are co-authors of the study.