Experimental Viral Tests for HIV Disease Progression

A book recently published in France -- Viral Quantitation in HIV
Infection, edited by Jean-Marie Andrieu -- provides a technical but
understandable background on the state of the art of measuring HIV
levels in humans (and also in animal and laboratory studies). The book
presents the proceedings of the International Workshop on Viral
Quantitation in HIV Infection, June 13-24, in Paris; it is entirely in
English, the language of the conference, although more of the research
presented was from France than from any other country.
For several reasons it is important for AIDS activists, medical
professionals, and others involved with the epidemic -- not only
scientific specialists -- to understand what is happening in the
development of better tests for HIV disease status and progression:
* Better "surrogate markers" of disease status are critically
important for designing faster, more feasible, and less expensive trials
of new drugs. Early clinical trials to test drug efficacy were set up
to run for years, because they looked for statistical differences in
death or disease progression. More recently, T-helper counts have
become generally accepted as an indicator of disease progression -- but
only for one class of drugs, nucleoside analogs such as AZT, ddI, and
ddC. And T-helper counts are highly variable and have other drawbacks.
Today it appears that a better surrogate marker for drug trials --
proportion of T-helper cells which are infected by HIV (see description
below) -- might be feasible. But experience has shown that improvements
in AIDS drug testing and in treatment do not happen automatically.
Activists must understand them and push for them.
* The same tests which could allow faster drug trials could also be
important for improving the treatment of individual patients. For
example, the test mentioned above -- proportion of HIV-infected T-helper
cells -- may be a good prognostic indicator, showing who is likely to
progress to serious illness well before T-helper cells start to decline.
These patients may want to start more aggressive treatment early, while
those with a better prognosis might choose a more conservative approach.
* Another use for such tests would be to monitor treatments,
telling which ones are working for which patients. HIV disease is very
different in different people, so no one treatment will fit everybody.
Treatments could be selected based on their efficacy for a particular
individual, then discarded if they later stop working. (T-helper counts
are already used for this purpose, but this measurement is a crude one.)
* Better tests for disease progression could have public- health
importance in the campaigns beginning now to get people into early
treatment for HIV infection. Better targeting of early treatment could
reduce the cost and increase the effectiveness of these programs, making
it easier to get them funded. Better targeting could also make early
treatment more attractive to patients, since they would receive AZT or
other drugs only when laboratory tests showed that they clearly needed
them.
* Better viral testing could make it easier to screen the various
"alternative" treatments -- those without pharmaceutical-company or
government sponsorship -- which have come into popular use without
official approval. The nutritional, herbal, and other treatments which
have become widely used have usually been fairly safe and relatively
inexpensive -- some have been called the "why not" drugs -- and usually
they have some scientific rationale. The problem is that usually there
is little or no hard evidence that they actually work as AIDS/HIV
treatments in people. The great expense of clinical trials has made it
very difficult to finance careful trials of substances which are
unpatentable, or are otherwise commercially unattractive. (And this
same expense all but guarantees that when a pharmaceutical company does
develop a drug, the eventual price will exclude most of the world's
population.)
Better ways to tell how well a drug is working would make trials
more precise, more credible, and less expensive. And even if trials of
non-commercial products still could not be financed, better information
about drug efficacy would give physicians a better practical sense of
what is working and what is not.
Dozens of "alternative" treatments are now in use. If even a few
were clearly found to be useful, they could be immensely important.
* The new book on viral quantitation was published in Europe, with
no U. S. distributor, so it may not be widely seen in this country.
Some of the leading U. S. experts on viral measurement attended the
conference to present their own results, and they will be familiar with
the state of the art in this field. But it is also important that
front-line physicians, activists, and organizers involved with AIDS
policy understand this field, and the new testing options which may
become available. Because of the ongoing lack of U. S. political
leadership on AIDS, there is no one in a position to survey the
technological advances and develop national policy to respond to the new
opportunities and make sure that important leads are effectively
followed up. Activists, by default, have had to take on this job and do
what they can.
Viral Quantitation in HIV Infection is divided into three sections.
The first, on quantifying HIV by using viral cultures, consists of ten
research papers. The second, on animal research, has three papers. The
third, on using PCR (polymerase chain reaction) to measure HIV, has
eight papers. Instead of trying to summarize all this research, we will
describe one report which may be very important.
Measuring the Proportion of T-Helper Cells Which Are Infectious
A paper by JM Andrieu, G Manolikakis, and W Lu, of the Laboratory
of Tumor Immunology at the University of Paris, is titled "Clinical
Application of Viral Quantitation by Measuring the Frequency of
Infectious Virus-Harboring Cells in Blood Samples of HIV-1 Seropositive
Individuals." The basic idea, that the proportion of T-helper cells
which are infectious in viral cultures can be a useful measure of
disease status, is not new; considerable research has already been
published. But Andrieu and colleagues claim to have improved and
simplified the procedure; they are proposing a practical, potentially
standard method by which different clinical laboratories could monitor
patients with HIV infection and obtain comparable data.
When they tested their method with 38 selected asymptomatic
patients, some monitored for up to two years, they found that the
proportion of T-helper cells which were infective almost invariably
distinguished those who remained stable from those who were later found
to progress -- even though there was little initial difference between
these groups in T-helper counts. (Other results from patients with
symptomatic illness or AIDS were also reported, including data showing
limited response to AZT therapy in the group overall -- but also showing
that some patients had a very good, and continuing, response to the
AZT.)
Comment
It should not be surprising that the proportion of T-helper cells
which are infected would be a better measure of disease status than the
T-helper count. The count depends on many factors, such as time of day,
exercise, other illnesses, and probably many unknown factors as well.
The proportion of cells infected by HIV is a much more specific measure
of the progression of HIV disease.
The paper does leave some questions unanswered, so other scientists
will need to get details from the authors. For example, the 38
asymptomatic patients were "selected," but the paper does not tell how
they were selected. And ten of these patients progressed to AIDS and
three to ARC in a mean follow-up of 18 months, despite the fact that
these 13 patients started with a mean T-helper count over 500 (The other
25 patients remained stable -- although seven of them did cross the cut-
off line of more than one in ten thousand T-helper cells being
infectious, indicating that they were at risk for progressing in the
future.) Perhaps the fact that all 38 patients had to be ones from whom
the virus could be cultured caused this group to have a worse prognosis
than would usually be expected; however, this issue is not explained in
the paper. [Other researchers have been able to culture HIV from blood
cells of all patients with ARC or AIDS, and from a large majority
patients who are asymptomatic and have high T-helper counts. See, for
example, "Quantitation of Cellular Viral Load: Correlation with CD4
Cell Count, by A. Venet and others, also published in Viral Quantitation
in HIV Infection.]
Despite such questions, this very important study deserves to be
followed up, not rejected out of hand. And since all the viral
culturing done was from frozen blood cells, confirmatory studies by
other groups might be done quickly, from samples already in freezers,
without the need to start a new trial and then wait two years to see who
progresses.
Another bad reason to dismiss this research would be from
misinterpretation of the modern realization that the classical theory of
AIDS (that HIV infects and kills T-helper cells and thereby causes the
immune deficiency) is incorrect -- and that the real pathogenesis of
AIDS, the way the disease develops, is still unknown. Other cells --
such as macrophages, or cells in lymph nodes, or in other organs -- may
well be more important than T- helper cells as sites of HIV infection.
But even if the infection in T-helper cells is only an indicator of
infection elsewhere, if this indicator works as a measurement to show
who will progress to more serious disease, and whether particular drugs
are effective, then that is what we need for now. We must use the tools
we have and improve them later, instead of postponing practical science
in order to wait for better theories.
How to Buy the Book
Viral Quantitation in HIV Infection is published by John Libbey
Eurotext Limited, in Montrouge, France. The price is about $50,
although it can change with international exchange rates; the best way
to pay is by credit card. The book can be ordered from a distributor in
England, Saber and Saber Limited, telephone 44-0279-417-134; or directly
from the publisher in Montrouge, 33-1-46-57-10-09 (fax). It may be
possible to request airmail delivery.