NIAID News Release

New Finding May Help Doctors Prevent HIV Drug Resistance

November 22, 2001

This article is part of The Body PRO's archive. Because it contains information that may no longer be accurate, this article should only be considered a historical document.

When an individual starts taking antiretroviral drugs for HIV infection, it is important to know as soon as possible if the drugs are adequately suppressing the virus. If they are not, some patients can develop drug-resistant strains of HIV in a matter of weeks, reducing the chance that a new drug regimen will be successful. Conventional wisdom, however, says doctors must wait four to eight weeks to accurately assess how well the drugs work.

But new findings suggest doctors can get the answer in one week, according to researchers at the National Institute of Allergy and Infectious Diseases (NIAID) and the National Cancer Institute (NCI). Their study shows the rate at which the virus disappears from the blood during the first week of therapy accurately predicts a drug regimen's long-term effectiveness in an individual. A report describing this work appears in the current issue of The Lancet.

"We now have a quick, simple and direct way of telling if the drugs are working," says Michael A. Polis, M.D., M.P.H., an NIAID infectious disease specialist and lead author on the Lancet paper. "In addition to heading off HIV drug resistance, this approach might also avoid exposing patients to the toxicities of antiretroviral drugs that aren't effective for them."

Currently, about 17 different antiretroviral drugs are available to fight HIV. They are prescribed in a variety of combinations because the fast-mutating virus quickly develops resistance to one drug alone. Many of the drugs work in a similar manner, however, so resistance to one drug can make the virus at least partly resistant to others. How well any drug will work in an individual is difficult to predict.

The standard approach to gauge a drug regimen's effectiveness in an individual is to measure the amount of virus in the blood after four to eight weeks of therapy. "Before our work, no one supposed an accurate prediction of drug efficacy could be made any earlier," says Dimiter S. Dimitrov, Ph.D., Sci.D, a computational biologist at NCI and senior author of the Lancet paper.

To monitor patients for drug resistance, doctors also analyze the virus' genetic material, RNA, to see if individuals harbor resistant viral strains. Trying to grow virus samples from patients in the presence of drugs is another way to check for resistance. "These tests, while useful, don't provide information as quickly or definitively as our measurement," says Dr. Polis.

Dr. Dimitrov and Polis' team mined the data of three previous clinical trials that tested a variety of antiretrovirals in HIV-infected adults and children. Once therapy began, the viral loads of those patients were measured daily. The researchers found that patients whose viral loads had dropped 50-fold or more at day six almost always responded well to the drug regimen long term. In contrast, patients whose viral loads dropped only five-fold or less at day six almost always responded poorly, that is, the virus concentration rebounded three months later. This single measurement accurately predicted therapy outcome more than 95 percent of the time.

"We are also trying to identify additional parameters that may help more accurately evaluate an individual's response to therapy," says Dr. Dimitrov.

These findings may be particularly important in light of recent reports that the years-long sharp decline in the number of Americans developing AIDS and dying from the disease appears to be leveling off, says Robert Yarchoan, M.D., study author and NCI researcher. "This news is evidence that the current crop of drugs are reaching the limits of their effectiveness," Dr. Yarchoan says. "One way to overcome that problem would be to develop new drugs, but another way would be to use the drugs we have more effectively. Our findings can help doctors to do that."

The scientists caution that a good result at day six is no guarantee a person will get better. "It is much easier to say when a therapy isn't working than to guarantee it will be ultimately successful," Dr. Dimitrov says. Dr. Polis adds, "Larger studies in more diverse groups of people need to be done to see if this approach will be useful in a real-life, clinical setting."


  1. MA Polis et al. Reduction in plasma HIV-1 RNA by one week following initiation of antiretroviral treatment correlates with longer-term efficacy. The Lancet 358(9295):1760-65 (2001).

This article was provided by U.S. National Institute of Allergy and Infectious Diseases.