AIDS TREATMENT NEWS #215, January 20, 1995
On January 12, many newspapers carried page-one stories about a
major scientific advance in understanding AIDS. Some of the
reporting has been unclear and confusing to readers, however.
This article will outline what the discovery is and why it is
important to people with HIV or AIDS. We will also explain our
concern that this new understanding, while indeed important,
shows signs of being misunderstood in ways that could seriously
harm AIDS research.
The occasion of the recent press coverage was the publication
of two articles and a commentary in the January 12 issue of
Nature,(1,2,3) which is one of the most prestigious scientific
journals. The two articles are by two well-known research
groups which independently arrived at essentially the same
conclusions. AIDS researchers worked with mathematicians to
help them reach a deeper understanding of the data -- which
came from blood tests of patients with advanced HIV disease as
they started taking certain experimental antivirals in small
The discovery announced in the newspapers on January 12 was not
news to the scientific community, which has heard the central
idea at a number of public conferences during the last year. We
published a preview last November, in an extensive quote from
Douglas D. Richman, M.D., who spoke on November 12 to a
physicians' conference in San Francisco (see "Viral Load, Small
Trials, and Immune Recovery," AIDS TREATMENT NEWS #211,
November 18, 1994).
New Understanding of HIV
The new understanding of how HIV behaves in the body was made
possible by the development of better antiviral drugs, which
can now shut off HIV reproduction almost completely (although
so far only for a short time since resistant virus develops in
a few weeks) -- and also by the new blood tests which measure
plasma HIV RNA, a much more accurate indication than the old
p24 antigen tests of how much virus is in the blood. The three
antiviral drugs used were the Abbott protease inhibitor
(ABT-538), the Merck protease inhibitor (L-524, also called
L-735,524), and Boehringer Ingelheim's nevirapine (which is a
non-nucleoside reverse transcriptase inhibitor); all three of
these drugs are experimental and currently in clinical trials.
AZT, ddI, and other approved anti-HIV drugs do not stop the
virus well enough to make the new discovery obvious. [Note: the
HIV RNA blood tests used are today available to physicians; for
background, see "HIV RNA: New Blood Tests for Individualized
Therapy and Faster Trials," AIDS TREATMENT NEWS #204, date
August 5, 1994, and later issues.]
Both the research teams measured the HIV RNA frequently (at
least weekly) when patients started taking the new experimental
antivirals. They found that once the reproduction of new virus
was shut off, the viral level in the blood declined very
rapidly; about 30 percent of the virus is cleared from the
blood each day (and this percentage may be about the same
regardless of stage of the disease). Since blood levels are
fairly stable from day to day when the patient is not taking an
antiviral, about as much new virus must be produced every day
as is destroyed. The essence of the new discovery is that the
"turnover" of new virus in the body is far more rapid than
people had believed -- in just a few days, most of the virus in
the blood has been removed and replaced with fresh virus.
Also, these drugs work by stopping the infection of new cells
-- not by shutting off viral production from chronically
infected cells -- and still they often cause the amount of
virus in the blood to drop by 99 percent or more. This means
that almost all of the enormous amount of virus being produced
continuously during HIV infection is coming from newly infected
The researchers also measured the recovery of CD4 cells (T-
helper cells) while the reproduction of HIV was temporarily
shut off. They found that CD4 cells increased very rapidly
during this time. This seems to mean that they are also being
destroyed equally rapidly when the virus is not shut down.
(Although an alternative possibility, which we do not believe
is ruled out by the new Nature articles, is that active HIV
infection produces something that suppresses production of new
CD4 cells; if this is the case, CD4 cell turnover in HIV
disease may be less than now thought.)
[Note on terminology: Until now AIDS TREATMENT NEWS has
preferred the term "T-helper cell" to "CD4 cell," as being
better for communicating with the public. But now the more
accurate term "CD4 cell" (or "CD4+ cell," which is the same
thing) is coming into wider use, and will be our preferred
usage in the future.]
Through simple calculations based on this data, the researchers
estimated that about 100 million virions (individual virus
particles) are normally being produced and destroyed each day.
(Note that this applies only to patients with fairly advanced
HIV disease, who were studied in this research; they had a
median CD4 count of 102 in one of the papers, a mean of 180 in
the other. In earlier stages of asymptomatic HIV infection, the
number of virions would usually be much less.) Also, the
researchers estimated that about two billion CD4 cells are also
being created and destroyed each day in these patients. Each
infected cell can produce many copies of the virus; the
researchers suspect, therefore, that most infected CD4 cells
are destroyed by the immune system before they ever have a
chance to produce complete virus.
These findings are important to people with AIDS/HIV because:
* They strongly suggest that if HIV reproduction could be
largely stopped for a longer time, and kept at a very low
level indefinitely, then the immune system would have much
more ability to recover, even without immune-based treatment,
than most physicians and scientists had believed.
Note that this does NOT mean that the virus will be eradicated.
Huge reservoirs of latent virus will still likely remain, in
the DNA of cells in lymph nodes and elsewhere. While it is
latent, it is not doing damage. But some of the latent virus
may become activated later and re-establish the active
infection. It will probably still be necessary to take drugs to
keep the virus suppressed -- although if the immune system
recovers, it may be an important help in this process.
* It is now clear that researchers can test potential antiviral
drugs and combination treatments in people in small, rapid
trials -- which get results in weeks. This means that it is
feasible to test many new treatments quickly, greatly
speeding the development of better AIDS/HIV treatments.
Of course it will be necessary to continue testing the
treatments which do show antiviral activity, to see how long
they work before viral resistance develops. This should not be
hard to do, since patients will naturally want to continue
taking a treatment which is working.
Other steps are also necessary before FDA approval. [We analyze
the critical bottleneck in the whole process -- the need to
prove actual clinical benefit to patients -- in "1995 Outlook
-- Research Strategy," below.]
* They emphasize again the major problem of HIV developing
resistance to drugs. The very rapid turnover of HIV in the
blood means that in just two to four weeks, the drug-
susceptible viral population in a patient can be replaced by
a drug-resistant population, meaning that the drug will no
longer work, or at least not work as well.
The researchers emphasized the need to use combination
treatments to prevent drug resistance. Drug combinations help
to prevent resistant viruses from developing, because only a
small fraction of the virus will initially be resistant to any
candidate drug. If a second drug is added, only a small
fraction of that small fraction will be resistant to both. The
more drugs that are added to the combination, the less likely
it will be that any virion will be resistant to all of them.
But with billions of virions in the body, it may take a number
of drugs in combination to stop all of them.
Some researchers have suggested that combination treatments
should be used early in infection, when the viral load in the
body, and also the variation of the virus, is less. (It is
generally believed that HIV infection starts with only one
variant of HIV, but as the disease progresses, this one strain
evolves into a great many different "quasispecies" within the
individual patient. In each patient, this evolution is
Comment: What This Discovery Does Not Explain
The most important unanswered question about HIV disease may be
how it progresses from early, asymptomatic infection to
late-stage illness. For years the immune system controls the
virus to a degree, and blood levels stay relatively low. But
somehow the body gradually loses this ability (in most people,
but not in all), allowing the virus in the blood to increase by
as much as hundreds of times, leading to the development of
Many theories have been proposed to explain how HIV suppresses
the immune system -- including, for example, direct killing of
CD4 cells, and also indirect mechanisms such as molecular
mimicry, cytokine or possibly endocrine dysregulation,
autoimmunity, abnormal apoptosis, loss of CD8 cells, etc. If
researchers knew why the immune system gradually loses its
ability to control HIV, then it might be possible to treat HIV
infection by correcting this specific problem, and keep people
healthy indefinitely, perhaps without the need for antivirals.
We are concerned by indications that some researchers, as well
as medical reporters, may uncritically assume that the new
understanding of HIV also explains how the infection progresses
from the early, asymptomatic stage, to the later stages of
greatly increased HIV levels and clinical illness. Such a
premature conclusion could suppress interest and research in
other potential mechanisms of progression.
The new information recently published in Nature was based only
on studies of persons with advanced HIV infection. And each
patient was studied only for a brief snapshot of time; there
was no long-term data on how infection progresses over time.
The Nature papers do not claim to have shown how HIV infection
progresses. But they are often being implicitly interpreted
that way. Some of these interpretations do not stand close
* One assumption is that, due to the great number of CD4 cells
infected and destroyed, the immune system eventually becomes
exhausted and unable to keep up, and then the disease
progresses because the body cannot produce enough CD4 cells
to replace those that are lost.
But this theory ignores the fact that HIV disease progression
occurs also at earlier stages, when the level of HIV
infection and destruction of T-cells is a small fraction of
what the body will be able to keep up with later. For
example, there is data showing that in persons with a CD4
count around 200, the level of virus in the blood may be ten
to a hundred times more than the level when the CD4 count is
500.(4) How could this exhaustion theory explain the
progression from a CD4 count of 500 to 450 or 400, if at 500
the body still has at least ten to a hundred times the
capacity needed to replace the cells which are lost at that
* Another assumption -- seen repeatedly in the January 12 New
York Times article -- is that HIV infection progresses
because the virus has a "slight statistical edge" over the
immune system, leading to its eventual victory. But this
theory is unlikely to hold up to an analysis of the dynamics
of stability vs. instability. The levels of both the opposing
sides in the battle -- the virus, and the CD4 cells -- can
change greatly in a time frame of days or weeks. It seems
unlikely that these two opposing forces, which each can
change in days, will just happen to stay approximately
balanced for the much longer time period of years that it
takes for HIV disease to progress.
To visualize the problem, imagine an acrobat on a high wire who
momentarily loses balance, and fights in an attempt to regain
it. Usually only seconds will pass until either balance is
regained, or the acrobat falls off. We do not expect to see the
struggle go on for ten years.
In other words, some mechanism other than a slight statistical
edge must be controlling the timing of HIV disease progression.
No one knows what sets this clock; it should be a central goal
of AIDS research to find out. The statistical-edge theory, by
giving false confidence that the answer is already known, could
distract research from this key goal.
Note that in most viral and other infections, the dynamics of
the interaction between the replicating organism and the immune
system does not lead even temporarily to a homeostasis, to a
balance or steady state. Instead, the immune system either
eradicates or almost eradicates the disease-causing organism,
or the infection progresses until the person dies. (Nor does
the limited supply of cells to infect explain the limited
growth of HIV, except perhaps at end-stage illness; at earlier
stages, the supply of CD4 cells would clearly allow more viral
growth, as shown by the fact that viral levels will
substantially increase later, when there are fewer such cells
available.) No one knows why HIV reaches a certain blood level
and then stops (until the level changes over months or years).
When we discover the mechanism by which this control becomes
established and maintained, but then gradually lost, we will
very likely have a new way of controlling HIV disease.
* One of the recent Nature articles(1) noted, "The difference
in lifespan between virus-producing cells and latently
infected cells (PBMCs) suggests that virus expression per se
is directly involved in CD4+ cell destruction. The data do
not suggest an 'innocent bystander' mechanism of cell killing
whereby uninfected or latently infected cells are indirectly
targeted for destruction by adsorption of viral proteins or
by autoimmune reactivities."
We are concerned that readers may take away from this statement
something different from what it actually says. While the data
cited may not support indirect mechanism for the decline of CD4
cells, they do not rule out such mechanisms, either. Again, our
concern is that important research could be prematurely
downgraded or closed off, due to the rush of enthusiasm which
has greeted the new findings.
(And if it turns out that active HIV infection somehow
suppresses development of new CD4 cells, or suppresses their
release into the blood -- and CD4 counts increase rapidly when
the infection is stopped, not because they are usually being
created and destroyed that fast, but because the suppressive
effect is removed -- then the whole issue of the turnover and
lifespan of infected CD4 cells will have to be rethought.)
Research Strategy Summary
The most important impact of the new understanding of HIV
disease, published in two articles and a commentary in the
January 12 Nature, will be on the strategy of research and
development of new AIDS treatments. We see the following
* The new work validates the use of small, rapid screening
trials, in a few patients, to learn which potential drugs
have antiviral activity in people.
* It also shows that if the virus can be suppressed completely
enough, the immune system has much more ability to recover
than many had suspected.
* It also confirms the belief that overcoming drug resistance
is likely to be the biggest challenge to making anti-HIV
* On the minus side, possible misinterpretation of the new
findings (in the current atmosphere of great attention and
enthusiasm for them) may impede important work of learning
how HIV infection progresses -- work which could lead to a
completely different class of treatments which do not attack
the virus directly, but preserve the immune system's ability
to control it.
1. Wel X, Ghosh SK, Taylor ME and others. Viral dynamics
in human immunodeficiency virus type 1 infection. Nature.
January 12, 1995; volume 273, pages 117-122.
2. Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM,
and Markowitz M. Rapid turnover of plasma virions and CD4
lymphocytes in HIV-1 infection. Nature. January 12, 1995;
volume 273, pages 123-126.
3. Wain-Hobson S. Virological mayhem. Nature. January
12, 1995; volume 273, page 102.
4. Phillips AN, Sabin CA, Elford J, and others. Viral
burden in HIV infection. Nature. January 13, 1994; volume 367,