Given this new approach, both physicians and HIV-infected persons would greatly benefit from a laboratory test capable of predicting the future progression of disease. Such a test would enable the physician to better counsel the patient and optimize the timing of treatment, and it might also be of considerable value in monitoring the patient's response to treatment in the absence of clinical symptoms. As a research tool, such a test could significantly accelerate the evaluation of new anti-HIV drugs and prospectively allow the identification of populations of rapid progressors, slow progressors, and nonprogressors for important comparative studies.

Unfortunately, the early search for predictive surrogate markers of subsequent HIV disease has proved singularly disappointing. Serum tests for immune activation markers such as ß₂-µglobulin and neopterin have had limited value, as have measurements of HIV p24 antigenemia. Enumeration of CD4 T-lymphocytes in peripheral blood remains the most commonly accepted way to monitor HIV disease progression, but these counts are often not depressed early in the course of infection and may fluctuate significantly for reasons unrelated to HIV infection [4-6].

A series of four recent studies [7-10] from three different groups now gives cause for renewed optimism about finding a predictive test. Each of these three groups have assembled data to support the notion that the increased expression of HIV messenger RNA (mRNA) in peripheral blood mononuclear cells forecasts future progression to AIDS. Still unsettled is the question of whether the measurement of multiply spliced or unspliced viral mRNA species is more informative. The results of a larger, longitudinal investigation involving 150 asymptomatic HIV-infected men from the New York City Prospective AIDS study are reported by Saksela and colleagues in this issue [10]. This study reconfirms and significantly extends the earlier pilot work. Detection of HIV mRNA in peripheral blood mononuclear cells (multiply spliced) was found to strongly predict future progression to AIDS. Although generally correlating with the immunologic status of the person being tested, the results of this test had a predictive power independent of CD4 cell counts. Indeed, the subsequent development of disease could even be predicted in persons with normal CD4 T-cell counts. Perhaps most striking was this test's ability to identify persons who would fail to progress to disease during the ensuing 6 to 7 years of follow-up. Again, CD4 cell counts were of no value in distinguishing this group. From a scientific viewpoint, the ability to prospectively identify slow progressors or nonprogressors is exciting, because these persons may yield important insights into HIV pathogenesis and the nature of protective immunity.

Measurement of cell-associated RNA species using reverse transcription and the polymerase chain reaction is not now routinely done in hospitals or commercial clinical laboratories. Thus, one significant question is whether this technology can be successfully transferred to the clinical laboratory in a way that will make it available to the practicing physician. Certainly, the need to handle cells and process labile RNA molecules complicates the task. Notably, alternative tests capable of measuring viral RNA within circulating virions in plasma are becoming more widely available [11]. Whether such measurements of viral RNA load in plasma will have the same predictive power as peripheral blood mononuclear cell HIV RNA measurements is currently under intense study, and early results are promising [12, 13]. It should be noted that these tests examine different parts of the retroviral life cycle and are subject to influence by different factors. Accordingly, they may not be fully interchangeable. Plasma viral load probably reflects a dynamic balance between the daily production of nearly 1 billion new virions [14, 15], countered in part by the destruction and physical trapping of these virions in the dendritic cell network of the lymph node or other lymphatic tissues [16, 17]. Levels of HIV mRNA in peripheral blood mononuclear cells may instead principally measure the trafficking of infected cells out of the lymph node coupled with new infections occurring in the periphery. In terms of such targets for infection, recent studies [14] suggest that as many as 2 billion CD4 cells are produced daily to replace the losses induced by HIV. This finding underscores both the marked cytopathicity of HIV and the remarkable resiliency of the immune system. Like viral load, the number of peripheral blood mononuclear cells transcribing viral RNA at any point in time is the product of an array of diverse processes. Nevertheless, for the HIV-infected person, the clinician providing care, and the scientist attempting to unravel the mysteries of the HIV virus, the identification of an assay potentially capable of predicting future progression or nonprogression to AIDS is very good news indeed.

@copy; 1995 American College of Physicians