R-HEV: BETTER TEST FOR NEW AIDS TREATMENTS?

A new blood test developed in France may allow testing of proposed AIDS treatments in clinical trials much more rapidly than is now being done -- and to give individuals early information about how well a particular treatment is working for them. AIDS TREATMENT NEWS interviewed the test's principal developers, Jacques Leibowitch, M. D. and Dominique Mathez, M. D., clinical immunologists at the Raymond-Poincare Hospital/University Rene-Descartes, Paris-Ouest. Leibowitch stressed that these results must be considered very preliminary as they have not yet been confirmed by peer review. (The work has been submitted for publication, but not yet formally published; a short, early poster was presented at the June 1989 AIDS conference in Montreal.)

A retrospective study using frozen blood samples examined how the test performed in predicting disease progression. Researchers tested 131 samples from 57 HIV+ patients followed over five years; none of the patients received any anti-HIV treatment during this time. Because the test can be run on frozen blood, it is possible to study its prognostic value quickly, without waiting for years to see how AIDS/HIV progresses in the patients tested, as would be necessary if fresh blood were required. The new test, called R-HEV, performed much better than the P24 antigen test in predicting who would do badly or do well in the future.

In a separate study, the R-HEV test clearly showed the antiviral effect of AZT -- an effect which was partial and temporary -- and also the effect of alpha interferon. (Some of the 57 patients whose blood was tested in the earlier study were later treated with these drugs, and frozen blood samples collected after treatment began were tested.) These results suggest that R-HEV may allow researchers to determine very quickly whether and to what extent a new drug is working. Also, much might be learned quickly and inexpensively about the efficacy of existing drugs by studying stored blood samples of patients who have already been treated, without needing to wait for new clinical trials to be funded, organized, and conducted.


WHAT IS THE R-HEV TEST?

R-HEV, an abbreviation for "radiation-resistant HIV expression ex vivo," is based on a technique which has been used in viral research for over 20 years (Henle and others, 1967), but has not previously been applied to HIV. It basically consists of viral cultures on cells which have been treated with radiation. The goal is to measure the HIV expression levels of the patient's virus, on the theory that radiation will suppress cells carrying HIV but not expressing it in the patient, and with the assumption that viral expression in the person is required for HIV disease to progress.

Ordinary viral cultures are unreliable as a measure of disease progression. Part of the reason is that latent virus, which is not causing any immediate problem, can be stimulated to become active by the culturing process itself, causing a positive result which does not reflect disease progression or poor prognosis for the patient.

In the R-HEV test, the radiation treatment causes the cells to die shortly after the time that culturing begins. If the virus was latent at the time the blood was drawn from the patient, the cells containing the latent virus die without being able to infect cells in the culture medium. But if the virus was active in the patient, some of the infected cells will just have time to transmit the infection to cells in the culture medium. Eight separate wells are cultured for each test, and the result reported by the R-HEV test is the percentage of wells which do grow virus.

Hundreds of sites in the U. S. already have the equipment needed to do the R-HEV test. The radiation can be provided by a machine already used in blood banks to irradiate blood before transfusion to persons with immune deficiencies. HIV cultures are somewhat expensive today -- just the materials for each R- HEV test cost $40 -- but the technology is already available to reduce the cost enough to allow regular use of the test in medical practice, not only in research.


THE AIDS PROGRESSION STUDY

Of the 57 HIV+ patients whose frozen blood and five-year followup results were available, 22 already had AIDS or ARC, or T-helper cells under 400, at the time of their first visit. The other 35 had few or no symptoms, and T-helper cells over 550, at their first visit.

Of the latter 35 (the apparently healthy patients), 13 remained well and had T-helper counts which remained above 500. These 13 were called slow progressors. The other 22 had rapid clinical deterioration or rapid fall in T-helper cells (the fast progressors). How well did the R-HEV test do in distinguishing who was already ill, and more importantly, in predicting who would become ill in the future?

Of the 131 specimens available from the 57 HIV+ subjects, 91 tested positive on R-HEV; the other 40 were negative. Of these 40, all but one came from 15 patients with few or no symptoms and high T-helper cells. Among these 15 were ten who remained R-HEV negative for an average of 32 months; all ten of them remained well, and their T-helper counts remained over 500.

On the other hand, 43 of the 44 patients with serious illness or bad prognosis (the 22 with AIDS or ARC, and the 22 fast progressors) tested R-HEV positive. (And the one who tested negative did test R-HEV positive on alveolar lavage cells.)

Of the 22 apparently-healthy patients who were fast progressors, 17 were R-HEV positive at their first visit -- an average of one year before clinical symptoms or falling T-helper cells were seen; the others became R-HEV positive eventually (in one case, only when alveolar cells were tested.) By contrast, P24 antigen was positive in only eight of the 22 patients at their first visit. All together, 33 percent of the 44 patients who were seriously ill at some point in the five years were P24 antigen negative, vs only one of those patients who was R-HEV negative (and that patient was R-HEV positive when alveolar cells were tested).

These results suggest that R-HEV is much better than P24 antigen in correlating with disease state, and in providing early warning of poor prognosis -- long before T-helper cells decline. If future work confirms these results, then R-HEV will be the best test available for HIV-disease status, and an excellent HIV-specific "surrogate marker" to use in clinical trials to find out quickly whether or not a drug is working.


R-HEV AND ANTIVIRALS

How good is the R-HEV test for showing how well an antiviral is working? The very preliminary results now available appear promising.

Eighteen of the 57 patients mentioned above were later treated with AZT. While we do not have the exact figures, R-HEV scores were greatly reduced in almost all of them within six months of AZT treatment. However, between six and 12 months after the patients started AZT, their R-HEV scores were back to an average close to their original value.

Three patient was treated with alpha interferon; one improved greatly, going into a long clinical remission. Ordinary viral cultures decreased initially, but then produced as much virus as they had before, even though the patient was in remission and did not have P24 antigen -- illustrating that R-HEV correlated with this patient's condition, whereas a conventional viral culture was not. (The other two patients treated with interferon did not show as complete or as prolonged a remission.)

DISCUSSION

The R-HEV test is still very preliminary; until the early results outlined above are confirmed by other researchers, we cannot be sure that the test will prove useful. But if future work confirms the results obtained so far, then this test will have great importance in speeding the clinical trials of new AIDS drugs, by providing an objective, specific measurement of disease status which can be followed to see how well a treatment is working. A few weeks of testing in a small number of patients may provide a better measure of a new drug's efficacy than current trials -- which require hundreds of patients and commonly take two years or more, because they do not have a good measure of efficacy and have to wait for deaths or opportunistic infections to accumulate instead.

A detailed report on R-HEV will be presented at a scientific meeting in Bethesda, Maryland, later this month.

On September 11-12, a crucial meeting in Washington, DC will attempt to reach professional consensus on "surrogate markers" -- meaning results other than deaths or opportunistic infections which can be used as endpoints of clinical trials to determine whether treatments work. At this time, there is much scientific debate about whether existing blood tests -- especially P24 antigen and T-helper counts -- are meaningful enough to use as rapid proof of efficacy of a drug, without needing to wait for statistically significant numbers of "clinical events" -- deaths and major opportunistic infections -- which require the large, slow trials now causing bottlenecks and great delays in AIDS drug testing. While some leading scientists -- such as Samuel Broder, M. D., the principal developer of both AZT and DDI -- believe that new drugs could be tested using existing measurements such as P24 and T-cell subsets, it appears that most of the scientists involved doubt that those tests are good enough, and that clinical trials must therefore continue to recruit hundreds of subjects and take years to prove efficacy, as they have in the past.

AIDS TREATMENT NEWS has argued that statistical indices of patients' overall health should be used in clinical trials to determine whether drugs are working. But few physicians and scientists have been willing to accept this approach. They argue that measures of clinical condition are often "subjective" and are not HIV specific, and that AIDS does not follow a steady course, with patients' health often improving or deteriorating unpredictably. Because of these concerns, there is no chance that the approach we have proposed will emerge as a consensus from the October meeting.

We also fear that there may be no consensus on existing markers of HIV progression either -- that many scientists will not accept P24 antigen, T-helper counts, beta-2 microglobulin, etc., as good enough for proving efficacy in drug trials. In that case we would be left with the current "body count" clinical trials as the only accepted designs -- trials which cannot possibly test drugs fast enough to prevent tens of thousands, if not hundreds of thousands, of AIDS deaths.

This is why the R-HEV test, if its early promise is confirmed, would be so important. It is the kind of test which the scientific community might accept, and which could greatly speed AIDS treatment research. It is urgent that this approach to drug efficacy testing get prompt, objective consideration.


REFERENCES

Henle, W. and others. Herpes-type virus and chromosome marker in normal leukocytes after growth with irradiated Burkitt cells. SCIENCE, vol. 157, pages 1064-1065, September 1, 1967.

Leibowitch, J; Mathez, D; Cesari, D; Belilovsky, C; Gorin, I; Deleuze, J; and Paul, D. Morbidite systemique et expression
retrovirale (infection productive) chez le porteur HIV. V International Conference on AIDS, Montreal June 4-9, 1989, poster number W. C. P. 72.