Researchers led by a UC San Francisco scientist have developed a model for predicting the emergence of antiviral drug resistance and for identifying the key factors that generate drug resistance.
The model, reported in the June 1 issue of Nature Medicine and lauded in an accompanying review by a Yale School of Management professor of management sciences and medicine, looked specifically at the likelihood of drug resistance occurring with increased use of a drug to treat genital herpes. The findings predicted that while increased use would lead to elevated levels of drug resistant strains, the number of cases of the disease would significantly decrease.
However, the greater significance of the study, said the principal investigator, Sally Blower, PhD, an associate professor of microbiology, immunology and medicine at UCSF, is the potential use of the new methodology for predicting drug resistance in many infectious diseases prone to antibiotic or antiviral resistance, including HIV.
"The methodology we've developed and applied to genital herpes can be used as a health policy tool to predict the likelihood of future levels of antibiotic or antiviral resistance for numerous infectious diseases, including HIV," said Blower.
A tool to predict future drug resistance is sorely needed, according to Edward H. Kaplan, PhD, of Yale in his review of the study. "This new model," he said, "provides a way to challenge the inertia often seen in public health decisions in which the overwhelming cost and timing required by empirical studies renders such analyses unfeasible, leading by default to decisions not to promote drug treatment when there is a risk of drug resistance developing."
The UCSF mathematical model, which traced the dynamics of drug-sensitive and drug-resistant herpes simplex virus in the general population, was designed to predict how much drug resistance could be expected to emerge over the next 50 years if treatment rates increased, and to identify the key factors that would determine the emergence of drug resistance.
Using two forms of statistical analyses, known as uncertainty and sensitivity analyses, the researchers examined a wide range of numerical values considered plausible for every parameter in the model, ranging from the best, most likely, to worst case values. These numbers were approved by a panel of infectious disease experts.
"The results of the uncertainty and sensitivity analyses demonstrated," said Blower, "that increasing treatment rates would be extremely beneficial if there is a low probability that acquired drug-resistance will develop and/or if the drug-resistant strains that emerge are of low transmissibility. Currently many infectious disease experts believe that these conditions appear plausible but the precise values of these two parameters must be determined by future empirical studies."
Genital herpes is only treated in 10 percent of cases in the United States and often goes untreated altogether in Third World countries. The apparent reason, at least in the United States, is that most people who contract the condition don't find it bothersome enough to warrant the effort.
However, "the overwhelming finding of the study was that treating herpes epidemics would always be beneficial, in terms of the number of cases prevented in relation to the number of drug resistant cases that developed," said Blower.
The study found that, even in the worst case, at least five drug-sensitive infections could be prevented on average for each case of drug-resistant genetic herpes.
And since genital herpes is a risk factor for developing HIV, increasing treatment rates of genital herpes could potentially also have an important indirect and rapid effect on decreasing the incidence of HIV, she said. The study also found, however, that increasing treatment over a 50-year period correlated with a significant increase in drug resistant strains of the virus, which can be spread through sexual transmission or evolve naturally during treatment. In fact, after 50 years of increased treatment with the drug acyclovir, drug resistant strains accounted for 29 percent of infectious cases. Importantly, the model also identified the most influential factors in the emergence of strains of genital herpes resistant to drugs, which could lead to preventive measures, said Blower. They included the average number of infectious episodes per year, the transmission probability of drug-resistant strains and the average number of new sex partners.
The next step, however, she said, must be biological studies to determine the transmissibility of drug resistant strains and the likelihood of acquiring permanent drug resistance during episodic treatment.
As far as evaluating drug resistance in other infectious diseases, says Blower, "we suggest that the novel health policy analyses that we have presented and applied to genital herpes should now be applied to other infectious diseases."
The co-authors of the study were Travis Porco, formally a postdoctoral fellow in Blower's research group at UCSF and now a research scientist in the AIDS office at the San Francisco Department of Public Health, and Graham Darby, PhD, the international director for viral diseases research at Glaxo Wellcome Research & Development, which markets the drug acyclovir, used to treat genital herpes.
Journal
Nature Medicine