Epidemic Potential of Drug Resistant HIV-1
January 18, 2003

Widespread antiretroviral therapy could have deleterious or beneficial effects on the course of the HIV-1 epidemic. Indeed, treatment of cases of infectious diseases has been the mainstay of epidemic control for tuberculosis, gonorrhea, and syphilis but has not been promoted for HIV-1 because of fear that drug-resistant HIV-1 has epidemic potential. For example, the increasing prevalence of HIV-1-infected persons, transmission of drug-resistant virus, and disinhibition of risk behavior are some effects of treatment that could have a deleterious effect on the epidemic or even lead to a second epidemic of HIV-1. Opposing these deleterious effects, therapy decreases viral load and infectiousness, which could decrease the number of new HIV-1 cases.

To explicitly consider this complicated situation, Dr. Grant collaborated with Dr. Sally Blower (University of California, Los Angeles) to develop a mathematical model of drug-resistant HIV-1 epidemiology. This model assumes that people can be infected with either drug-resistant or drug-sensitive HIV-1. Some of the infected persons undertake treatment that decreases viral load and infectiousness for some time but can lead to resistant viral infection. The model was simulated using parameter estimates derived from data available in San Francisco, including rates of infection, treatment, treatment failure, and mortality. Uncertainty in parameter estimates was evaluated using latin hypercube sampling of probability distributions that span the range of possible parameter values. This model indicated that increasing the proportion of persons who are treated to 90% from the current value of 50% would decrease the incidence of new infections and lower the mortality rate if risk behavior did not change. However, a 15% increase in risky behavior could completely offset the benefits of widespread therapy. This analysis also identified three factors that are critical in predicting the epidemic outcome of widespread therapy: treatment rates, the infectiousness of persons with treated infections, and the amount of risky behavior in treated persons. Studies to measure these effects were initiated based on the findings of the mathematical model.

Transmission of HIV-1 resistant to antiviral drugs is a complication of antiviral therapy with profound epidemiologic consequences. Although transmission of viruses resistant to a single drug has been reported, multidrug-resistance in HIV-1 had not been demonstrated until 1998. In January 1998, Dr. Grant’s laboratory identified a drug-naive subject who was infected with HIV-1 that had multiple genetic markers of decreased sensitivity to several commonly used antiviral drugs, including AZT, 3TC, and protease inhibitors. Decreased sensitivity to these drugs was confirmed through a collaboration with scientists at ViroLogic, who have developed a novel test-vector-based assay of drug-resistance phenotype. To evaluate time trends and the epidemic potential of drug-resistant HIV-1, primary resistance from 1996 to 2001 was analyzed in 243 subjects from San Francisco. Genotypic primary resistance increased from 16.7% to 27.6% for any drug class, increased steadily from 0% to 17% for nnRTI’s (P=0.003), and increased steadily from 0 to 10.3% for PI?s. Genotypic primary resistance to nRTI’s initially decreased from 25% to 6% in 1998-99 then increased to 20% by 2000-01 (P=0.03). Primary resistance to 2 drug classes increased from 1.5% to 13.5% (P=0.01). There was only one case of 3-class MDR primary resistance, detected in 2000. Mean phenotypic susceptibility decreased for nnRTI’s and the prevalence of greater than 10 fold resistance to PI’s and nnRTI’s increased over time, confirming trends in genotypic resistance. These data indicate that primary resistance is prevalent and rising in San Francisco, especially for nnRTI and dual-class resistant HIV-1. Primary 3 class MDR is still rare. Diminished transmission fitness of HIV-1 resistant to certain drug classes corresponds with diminished fitness for replication, which spares CD4+ cells in drug naive persons.

 Tell a friend
 Printer friendly version