Laboratory analyses of blood and other medical measurements, which help health practitioners make diagnoses and detect toxic effects of medication, can also help people with HIV track their health. Robert Coombs, MD, PhD, is an associate professor in the Departments of Laboratory Medicine, Division of Virology, and the University of Washington School of Medicine in Seattle. He has done extensive research on measurements of HIV viral levels and is internationally recognized as an expert in this area. He analyzes the viral-load measurements of POZ founder Sean O. Strub.

Sean is like many PWAs who are so far responding well to the relatively new treatment approach called highly active antiretroviral therapy (HAART), using combinations of drugs that drastically reduce viral load. In Sean’s case, that has most recently meant indinavir (Crixivan), d4T (Zerit) and delavirdine (Rescriptor). The result? He’s had an undetectable viral load for six months now, down from a high of 3.3 million viral copies two years ago. The combination of this with his increase in CD4 count from one cell two years ago to 102 currently is most encouraging. These laboratory results clearly demonstrate the efficacy and short-term durability of his antiretroviral therapy. Research has made it clear that such results are generally associated with exactly the sorts of clinical benefits that Sean has experienced, including an increased sense of well-being and improved energy. Such results are also generally associated with decreased risk of disease progression. However, despite recent encouraging clinical-trial data, we don’t yet know whether these favorable short-term laboratory responses to therapy will ultimately translate into long-term survival.

For now, there are important questions that remain about the value of viral loads in making therapy decisions. First, how accurate is the number that appears on the lab printout?  With the current rush of enthusiasm for using this test as the basis for making treatment decisions, it may surprise many to learn that it is considerably less reliable than we would like. How the blood specimen is handled, whether samples of blood drawn at different times are measured individually at the time each sample is taken or are later measured all at once, the inherent imperfections of the biochemical technique used for the count and the day-to-day variation in the level of virus in the bloodstream can all influence the final result.

With the current tests, there is an expected variability of at least five-fold. This means that a person’s viral load taken from a single blood specimen could vary by up to five-fold above or below the original value if a second blood sample is taken around the same time. For example, the “true” viral load of someone who tests at 2,000 might range anywhere from 400 to 10,000; a load of 5,000 could range from 1,000 to 25,000; 20,000 could be 4,000 o 100,000; and 200,000 could range from 40,000 to 1,000,000.

And that undetectable viral load? If the test measures down to 200 copies (the standard test now being used), with anything below that labeled as “undetectable,” then we have to presume that the level could actually be 400. Thus, a second test could find a range of from 80 to 2,000. Since the low end of these ranges could be associated with a substantially lower risk of disease progression than the high end, fine-tuning your assessment by running a second test is obviously important. The simple rule on his one? Never use viral load as part of the basis for deciding to change your antiretroviral therapy without at least two measurements.

The other important question about the value of viral load is whether a person like Sean who had a high viral load that was substantially suppressed by drugs down to an undetectable level has the same outlook as a person who naturally has an undetectable or even moderately low level. The answer is “probably not,” because immunologically these latter people are totally different. This is a fundamental flaw in the current widespread presumption that everyone with a low viral load has the same prognosis.

Changes in the viral can only explain a portion of the clinical response to therapy. Preliminary data from ongoing clinical trials lead us to believe that other factors probably account for half or more of the outcome. Included are both the characteristics of the individual’s immune system function and of the viral population that is present. In other words, how powerful are someone’s HIV-specific immune responses and what is the virus’ ability to kill the particular cells that are infected in that person? We know that these may vary greatly from person to person. Unfortunately, outside of research labs, there are no currently available tests to answer these questions, and even if answered, there are no changes in therapy that we could make to affect these things.

In simple terms, we know that what defines one’s quantity of virus is the interaction of the virus with the host-the living, breathing human body where the virus lives. However, the factors that are responsible for containing the virus are poorly understood. There are several immune markers that should be studied to determine which correlate with survival and, thus, which might have value for making better therapy decisions. It will be particularly important to study activation markers since we know that an immune system that’s activated is proving the best possible environment for viral replication. A better understanding of these markers could substantially improve our assessment of disease progression risk, as could a better understanding of the virus.

We also need data from clinical trials to determine what weight should be assigned to various viral characteristics-the type of virus, the presence of mutations that could contribute to drug resistance, the results seen with pretesting for sensitivity to various HIV drugs, and others-when making therapy decisions. Without such additional information, we may continue to have people who inappropriately change what may well be successful therapy regimens because of the failure to understand the limitations of the viral-load test.

Until the researchers can provide this information, how does Sean or anyone else make the best decisions about therapy? For now, the best suggestion is to use the combined results of at least two viral load measurements (both done within two to three weeks) and two recent CD4 counts, and cautiously assess all those results in the context of the overall clinical picture. So far, the changes in lab results that have followed Sean’s HAART, combined with his improved well-being, seem very encouraging. Hopefully, the durability of his response to therapy will continue to translate into improved overall health, a slow but steady increase in CD4 count and continued suppression of his HIV to very low levels.