"The extent of viremia, measured by HIV RNA, is the best available surrogate marker of HIV disease progression. Use of HIV RNA as a surrogate marker should help guide future therapeutic research and individual patient management."
--John Mellors and others. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272: 1167-1170. May 24, 1996.
Using Chiron Corporation's branched-chain DNA (bDNA) test, John Mellors and colleagues at the University of Pittsburgh conclude that viral load predicts the risk of HIV disease progression (time to AIDS and death) better than CD4 count. The study population consisted of 180 gay and bisexual men enrolled in the Pittsburgh subset of the Multicenter AIDS Cohort Study (MACS). The Pittsburgh researchers' new findings have important implications for the management of HIV disease. Based on these and other study results, viral load testing is expected to supercede CD4 count as the principal marker for guiding individual HIV treatment decisions and for evaluating the effectiveness of anti-HIV drugs in clinical studies.
Armed with accurate measurements of the amount of HIV in their blood plasma as measured by HIV viral load testing, physicians and patients can make more informed decisions about when to start anti-HIV therapy, when to stop using an ineffective treatment and when to add or switch to a new treatment. In addition, monitoring viral load over time allows patients to make treatment decisions much earlier, prior to a significant loss of CD4 cells and well before clinical decline occurs. CD4 cell loss is thought to be a relatively late result of increased HIV replication. Therefore, it appears more beneficial to make anti-HIV treatment decisions based on HIV viral load rather than on CD4 cell count alone, particularly when the CD4 count is greater than 500 cells/mm3 (see Research Notes).
When to Start Anti-HIV Therapy?
The article on viral load in Science adds a new dimension to the ongoing debate about the optimal time to start anti-HIV treatment. The Pittsburgh cohort data suggest that the appropriate time to initiate therapy is when HIV viral load exceeds 10,000 copies/mL, regardless of CD4 cell count. Many individuals in the study with CD4 counts greater than 500 cells/mm3 progressed as rapidly to AIDS and death as those with much lower counts when their viral load levels were greater than 10,190 copies/mL. In current clinical practice, a CD4 cell count of fewer than 500 CD4 cells/mm3 is commonly used as the trigger to start anti-HIV treatment. This recommendation needs to be reconsidered, given the researchers' finding that 50% of the men in the study with greater than 500 CD4 cells/mm3 (median CD4 count 781 cells/mm3) at study entry and a viral load greater than 10,190 copies/mL died within 6 years after entering the study. In comparison, only 5% of those in the same cohort with similar CD4 counts (median CD4 count 787 cells/mm3) at study entry and viral loads less than 10,190 copies/mL died within 6 years.
The implication of these findings is clear: the decision to begin anti-HIV therapy should not be based solely on CD4 cell counts. Individuals should consider starting anti-HIV therapy when their viral load is greater than 10,000 copies/mL, regardless of their CD4 cell count. These conclusions do not diminish the value of CD4 cell testing in the management of HIV disease, which continues to serve as a reliable marker for predicting the risk of opportunistic infections and for determining the appropriate timing of initiating preventive treatment for these infections. In addition, many clinicians believe that a CD4 count less than 350 cells/mm3 represents an indication for starting anti-HIV therapy, regardless of HIV viral load.
-Viral load level (HIV RNA copies/mL): less than 4,531 Median time to AIDS (years): greater than 10 Median survival time (years): greater than 10
-Viral load level (HIV RNA copies/mL): 4,531-13,020 Median time to AIDS (years): 7.7 Median survival time (years): 9.5
-Viral load level (HIV RNA copies/mL): 13,020-36,270 Median time to AIDS (years): 5.3 Median survival time (years): 7.4
-Viral load level (HIV RNA copies/mL): greater than 36,270 Median time to AIDS (years): 3.5 Median survival time (years): 5.1
Viral Load and Disease Progression
The University of Pittsburgh investigators followed study participants for up to 11 years. They determined 4 groups of increasing viral load levels at study entry, and correlated each one with progression to AIDS and survival. The following chart outlines the results of their findings.
In this cohort of men, baseline HIV viral load levels correlate directly with time to AIDS diagnosis and with survival time. Simply stated, the lower the viral load, the longer the time to AIDS diagnosis and the longer the survival time. Conversely, the higher the viral load, the shorter the time to AIDS and the shorter the survival time. The study results indicate that HIV RNA levels can predict disease progression as far as 10 years into the future. The investigators also noted that when study participants are divided into 2 levels based on their viral load at entry -- greater than 10,190 or less than 10,190 copies/mL -- a surprising trend emerges. The 10-year rate of survival was 70% for those with less than or equal to 10,190 copies/mL compared to a survival rate of only 20% for those with greater than 10,190 copies/mL, including those in this group who had greater than 500 CD4 cells/mm3 at entry!
New Treatment Recommendations
Researchers at the University of California at San Francisco-affiliated San Francisco General Hospital (SFGH) have formulated interim recommendations on how to interpret viral load test results in conjunction with CD4 cell counts (see Research Notes). A group of researchers and clinicians from the International AIDS Society-USA also has published recommendations on how to interpret viral load test results (Michael Saag, MD, and others. Nature Medicine 2: 625-629. May 1996).
In these recommendations, the critically important numbers are fewer than 5,000 copies/mL and greater than 10,000 copies/mL. An HIV viral load test result of 5,000 copies/mL or less suggests a low level of viral replication, and probably no immediate need to start therapy, unless the CD4 count is less than 350 cells/mm3 (NOTE: some researchers argue that any level of HIV activity above the level of detection of the test used ought to prompt treatment!) A test result of 10,000- 50,000 copies/mL or greater suggests significant viral replication, and the SFGH recommendation is to consider therapy, regardless of CD4 cell count. The higher the viral load, the higher the risk for clinical decline and the more pressing the need to begin (or change) treatment. A test result of over 100,000 copies/mL may predict a rapid deterioration in clinical status. At this high level of HIV RNA concentration, the recommendation is to immediately start (or change) anti-HIV treatment.
FDA Approves Roche Viral Load Test for Prognosis
The U.S. Food and Drug Administration (FDA) approved the Amplicor HIV-1 Monitor Test (the viral load test from Roche Molecular Systems Inc.) for HIV disease prognosis on June 3, 1996, seven months after Roche submitted an application for approval. This test is called the reverse transcriptase polymerase chain reaction (RT-PCR) test or simply "PCR." FDA is expected to approve Quantiplex (the viral load test from Chiron Corporation) in the near future. The Chiron test is commonly called the branched-chain DNA test or simply "bDNA." Researchers have found that both the Roche PCR and the Chiron bDNA tests give comparable results in measuring HIV RNA levels in the blood plasma. Both tests cost $150-$200 per test, a price set by the laboratories where the tests are processed. The cost is expected to decline as laboratories face increased competition from each other. Roche has announced plans to offer 2 free HIV RNA baseline tests to all HIV positive patients in the U.S. over a 60-day period starting June 17, 1996. Call 888-TEST-PCR for more information.
Some researchers prefer the Chiron bDNA test because it is simpler to conduct and provides a direct quantification of the HIV RNA in plasma. Others prefer the Roche PCR test because it is more sensitive and capable of measuring HIV RNA levels as low as 400 copies/mL. Both Roche and Chiron have developed more sensitive second generation HIV RNA tests, which are not yet available except in research settings. The second generation bDNA test will measure HIV RNA levels as low as 300 copies/mL; the new Roche test will measure HIV RNA levels as low as 20 copies/mL.
Whichever test is chosen -- the Roche PCR or the Chiron bDNA -- it is important to continue using that same test to determine future HIV RNA values. At present, it is also advisable to use only the Roche or the Chiron test kits rather than "generic" viral load test kits from a laboratory. For now, only the Roche and Chiron tests can be expected to give consistent, reliable and comparable results. A third test that is also accurate and reliable -- the nucleic-acid sequence-based amplification (NASBA) test from Organon Teknika -- is more commonly used in Europe than in the U.S.
In using viral load testing to determine the level of HIV concentration, it is necessary to obtain a baseline value for each patient against which all future values can be compared. This is achieved by averaging the results from 2 viral load tests taken 2-4 weeks apart It is also important to base treatment decisions on sustained changes in viral load, not on a single measurement. Certain factors, such as vaccination for influenza and herpes simplex outbreaks transiently increase HIV levels in the bloodstream; other types of immunizations and other acute illnesses also may transiently increase HIV replication. Therefore, patients should avoid taking a viral load test for about 4 weeks following immunizations and after resolution of acute illnesses. When monitoring viral load results over time, it is important to note that only increases or decreases in viral load values of 3-fold or greater are considered significant enough to warrant a change in treatment regimen. For example, if an individual not on anti-HIV therapy with a baseline viral load test result of 4,000 copies/mL takes the test 3 months later and has a test result of 8,500 copies/mL, the increase is not considered significant because it is less than 3-fold. However, if the new test result is 12,500 copies/mL or higher, the change is significant (greater than 3-fold), and the physician and patient should discuss starting treatment.
Viral Load Results for Evaluating the Success of Therapy
In the coming months, more and more physicians will use the Roche and Chiron viral load tests to monitor the effects of anti-HIV therapy in their patients. Researchers have employed these test for several years in AIDS drug research to help evaluate the effectiveness of AIDS drugs. For example, FDA granted accelerated approval to the protease inhibitor drugs saquinavir, ritonavir and indinavir based in large measure on their ability to significantly decrease viral load as measured by HIV RNA in HIV positive individuals. Within a short time, viral load testing is expected to become an FDA-approved method for demonstrating how well a particular anti-HIV drug or drug combination works, without having to wait for clinical outcomes. In the future, anti-HIV drugs may be routinely evaluated based on whether or not they produce a sustained, significant decrease in viral load. Use of viral load testing also may dramatically shorten the time needed to test drug effectiveness. This could save millions of dollars in research costs by dramatically reducing the need for long clinical studies.
The increasing availability of the 3 new protease inhibitor drugs and of viral load testing has ushered in a new era in the treatment of HIV disease. With increased access to this powerful new technology and to a new class of potent anti-HIV drugs, we are closer to achieving the goal of making HIV disease a chronic manageable illness.