Stanford University Home

Stanford News Archive

Stanford Report, January 5, 2000

Test helps predict which HIV patients will respond to therapy


A newly available drug resistance test can help doctors make better choices in treating HIV infection in patients for whom combination therapy has failed, according to Stanford researchers.

More than half of HIV patients treated with the new drug "cocktails" fail to do well because of drug resistance, said Andrew Zolopa, MD, assistant professor of medicine and lead author of the latest study.

Doctors typically have relied on standard clinical factors, such as a patient's T-cell count, viral load and treatment history, in deciding what follow-up treatment to pursue with these drug-resistant patients, Zolopa said.

Zolopa and his colleagues found new genotype tests are more helpful in predicting a patient's response to therapy than a traditional clinical evaluation, he said. The gentotype tests reveal the specific mutations in the genetic sequence of a virus that enable it to resist drug treatment.

"The study indicates that these tests can provide the clinician with valuable information in crafting new combinations of medications to better control HIV infection in patients with partially resistant virus Zolopa said.

"We obviously need new and better drugs to treat HIV infection for many patients in our care now, but short of new drugs, these resistance tests should help clinicians use the drugs we have more wisely," said Thomas Merigan, MD, the Becker Professor of Medicine and one of the study's senior authors.

The study was published in the December 7, 1999, issue of the Annals of Internal Medicine.

The researchers evaluated 54 HIV-infected patients at the Stanford Positive Care Clinic, which Zolopa directs. The patients had all received a combination of saquinavir and ritonavir ­ two types of protease inhibitor drugs ­ between October 1996 and February 1998. In addition, all of the patients had been previously treated with nucleoside analogs, the earlier class of AIDS drugs that include AZT and ddI.

The researchers checked the patients' response to therapy three times over a six-month period, testing their viral loads, or the amount of virus circulating in the blood. About one-third of the patients had a good response, one-third had a partial response, and one-third had no response at all to the treatment, Zolopa said.

The researchers then compared these results with the results of genotyping tests done on viral samples stored at the start of the study. The genotyping test results were not known at the outset and were not used as a criteria for selecting patients, Zolopa said.

The researchers found a direct correlation between the patients' response to the new drug regimen and the genotyping test. They also found a link between the patients' response to treatment and the standard clinical evaluation, but the connection wasn't nearly as strong, Zolopa said.

With both pieces of information available ­ the genotype test and the clinical assessment ­ doctors could make the best treatment predictions, he said.

"We believe the results of this study support the use of resistance testing in the care of HIV-infected patients who are experiencing less-than-optimal response to the medications," he said.

Zolopa noted that interpreting the results of genotype tests can be difficult and requires special training. Stanford faculty are developing educational programs for doctors from around the country to learn how to better use these tests, he said.

The other investigators in the study are Robert Shafer, MD, assistant professor of medicine; Ann Warford, PhD, director of the Clinical Virology Laboratory at Stanford; Jose Gilberto Montoya, MD, assistant professor of medicine; Philip Hsu, BA, an undergraduate; David Katzenstein, MD, associate professor of medicine; and Bradley Efron, PhD, professor of statistics at the university.

Partial support for the study was provided by Roche Molecular Systems in Alameda, Calif. SR