Viral Load: New "Strategy" Trial in 15 U.S. Cities

A new kind of trial which focuses on treatment strategy instead of testing individual drugs is now recruiting 1,100 volunteers with CD4 (T-helper) counts of 300 or less, in the following cities: Albuquerque, Atlanta, Baltimore, Camden NJ, Chicago, Denver, Detroit (2 sites), New Orleans, New York City, Newark, Philadelphia, Portland OR, Richmond VA, San Francisco, and Washington DC.

The main purpose of this trial is to see if testing for plasma HIV RNA ("viral load") can help doctors make better decisions, leading to provable benefits to patients. To test this, half of the volunteers will be randomly assigned to use viral load in making treatment decisions, and the other half will be assigned not to use viral load. The tests will be free for those assigned to use them.

This trial will not tell patients or physicians which treatments to use; that will be decided in the normal course of medical practice. But viral load results will be available to half the patients and physicians to help in making these decisions, and not available to the other half. The trial is expected to run for two to three years.

Anyone is eligible to enter this trial if they are at least 13 years old, have HIV, have a CD4 count of no more than 300, and are in reasonably good health and willing to take antiretrovirals if needed. They must be receiving treatment from a physician who is participating in the CPCRA program. This trial imposes no restriction on what drugs the patient may use.

This trial is designed and run by the Community Programs for Clinical Research on AIDS (CPCRA), a program of the U.S. National Institute of Allergy and Infectious Diseases (NIAID); the principal investigator is Melanie Thompson, M.D. of the AIDS Research Consortium of Atlanta. Chiron Corporation will donate at least 10,000 bDNA tests for viral load; and SmithKline Beecham, which does medical laboratory testing, will donate the transportation of the samples, a substantial expense since every sample must be shipped in dry ice by overnight express. (The test which will be used is the same one physicians can order, with a cutoff of 10,000 copies of RNA -- not the newer research-only test with a cutoff of about 500 copies.)

A fraction of patients assigned to NOT use HIV RNA will have samples drawn anyway for an HIV RNA test, but the results will not be reported to them or their physician until the end of the trial. The purpose of these samples is to help in scientific analysis of the results.

Patients who are already using viral load or planning to do so should not enter this study, which is for patients and physicians who have not already decided whether or not to use this test.

This trial is named CPCRA 036; its official title is A Randomized Study of the Clinical Effects of Initiating or Changing Antiretroviral Therapy Based on Plasma HIV RNA Quantitation Compared with Initiating or Changing Therapy Based on Current Clinical Practice Alone. For more information about volunteering, call the AIDS Clinical Trials Information Service, 800/TRIALS-A, for the contact telephone number of the CPCRA site in one of the 15 cities listed above.

Comment

This study will develop important information about how physicians are deciding when to switch antiretroviral therapies -- whether those physicians are using viral load tests or not. Also, if it succeeds in proving that viral load tests can have value, it will help to change everyone's standard of care to make these tests more available, not only for those with money or good insurance, but for all.

We do have reservations about being randomized to not use viral load testing to guide therapy. We believe that everyone with HIV should get at least one baseline viral load test -- partly to have it available for comparison with future values, and especially to help decide on how aggressive to be with therapy. If the viral load is high (certainly anything over 100,000 copies is high, and many would aggressively treat values lower than that), the test should be repeated to make sure it was not a temporary high value (which can result from a minor infection, a flu shot or other shot, or from other causes), or due to laboratory error. If the repeat test is also high, we believe that one should use whatever drug combinations, other treatments, or lifestyle changes may be necessary to lower the viral load and keep it down.

Many experts do not agree with our view. Some suspect that the CD4 count "mirrors" viral load sufficiently closely that viral load does not add enough new information to be significant. Also, while everyone agrees that people with high viral load do worse on the average than those with low viral load, some experts argue it has not yet been definitively proved that lowering the viral load by drug treatment will improve a patient's prognosis. It is even possible that viral load testing could indirectly result in harm, by prompting many changes of therapy, exposing patients to more different drug toxicities than they would otherwise face.

Also, relying on viral load tests before we know exactly what they mean may lead researchers to discard drugs too soon. For example, in some trials of protease inhibitors, viral load greatly declined but then became high again, while CD4 count rises were much longer lasting; no one knows what this means, but it is possible that the drug is providing some benefit even when the viral load test is not showing it. Also, it is not known that viral load in the plasma accurately shows viral activity in all body compartments, for example in the central nervous system; it would be unfortunate to discard a drug that provided a benefit which was not readily seen through this test.

The CPCRA viral load study will produce a unique database to answer questions about how well viral load predicts clinical outcome, how it changes with changes in therapy, and how physicians make decisions about antiretroviral treatment. This information will be reviewed monthly as the study proceeds, and will be made available throughout the study to the participating physicians, and through them to the volunteers -- and perhaps even by a World Wide Web page available to everyone. In addition, this study is saving plasma samples from some volunteers for later studies of the effects of drug resistance.

Patients and physicians who are already committed to use viral load should not join this study. But those who would not otherwise use viral load can volunteer without reducing the care they receive. And of course if they change their mind later, they can withdraw from the study at any time.