AIDS Treatment News Issue #228, August 4, 1995
Marcus Conant, M.D., heads the Conant Medical Group, one of
the largest HIV practices in San Francisco.
AIDS Treatment News: How has your approach to treating HIV
infection changed since a year or two ago?
Dr. Conant: Today when we find that someone is HIV positive,
we are using three indicators, instead of one or two, to try
to measure where they are in the course of the disease.
Historically, the first measurement we had of disease
progression was symptoms. It was obvious that people who had
fever, fatigue, weight loss, night sweats, who started
developing skin conditions or other diseases, were further
along in the course of the disease, and therapy might be
instituted then. If someone was doing very well and suddenly
developed a yeast infection, even if they didn't have any
symptoms, even if the CD4 (T-helper) count had not dropped
greatly, that would tell us that their immune system was not
working as well as it had before, because they could not
suppress the yeast any more.
Then it became clear that the CD4 count -- or better, the CD4
percentage -- was an important indicator. So we measured CD4
about every four months. The most sophisticated patients knew
that it wasn't a drop in the number that was important, it
was a drop in the percentage.
The classical indicators that had been used in medicine, the
history, the physical findings, and the laboratory studies
(particularly the CD4 count or percentage, in the case of HIV
infection) were what we used until last year to decide who
should go on therapy. So if a patient had a CD4 count of 800
and was totally asymptomatic, we did not initiate therapy. If
someone had a CD4 count of 550, but had a lot of fatigue, and
had had zoster a year ago, and then had developed hairy
leukoplakia, many of us would have gone ahead and put that
patient on combination therapy -- saying that the CD4 counts
looked good, but because of the increase in symptoms and
signs, we had clinical evidence of HIV disease progression,
and we now needed to be more aggressive.
Viral Load Testing
In the last year we have added viral load to the formula. So
now we are looking at symptoms, measuring the CD4 count, and
also looking at viral load. While the information has become
more complex, it has actually become easier to use, because
the viral load can give us more information than all of the
others put together.
If we saw a patient who was totally asymptomatic, had no
physical finding, no cutaneous or other infections suggesting
disease progression, and had a very high CD4 count, say 800,
and had a viral load measured by branched DNA (bDNA) or
quantitative PCR at less than 10,000 copies, we would not
treat that patient, but would repeat the branched DNA test
every four to six months. If at any point one of the markers
started telling us this patient was progressing -- if the
branched DNA, for instance, suddenly went up to 40,000, or
the CD4 count dropped precipitously, or the patient started
getting lots of symptoms -- then we would likely decide to
[Note: The branched DNA test -- abbreviated bDNA -- is one
technology for measuring the viral load, the number of copies
of HIV RNA per milliliter of blood plasma. Other technologies
for measuring the viral load are quantitative PCR, and NASBA.
Incidentally, each HIV virus particle has two copies of its
RNA, so the actual number of copies of the virus is half the
number of copies of HIV RNA.]
ATN: If the viral load suddenly goes up, would you repeat the
branched DNA or other viral load test?
Dr. Conant: In my experience, the bDNA test is very
reproducible, far more reliable than the CD4. So if I get a
bDNA that is elevated, I usually do not repeat it a month or
two later to see if it is accurate; instead, I respond to it.
But if the patient was totally asymptomatic and had a stable
CD4 percent, and the bDNA count went sky high, say from
10,000 to over 100,000, then I would retest to make sure a
mistake had not been made.
Initial Antiviral Treatment
ATN: What antiviral therapies are you using first?
Dr. Conant: If there is evidence of disease progression, we
are first treating many patients with very high dose d4T. The
dose I usually start with is 20 mg. twice a day; then I have
the patient increase it by 20 mg. each week, until they get
neuropathy, or until we get to 80 mg. twice a day. That is
twice the FDA-recommended dose.
Why are we going that high? The answer is that the
recommended dose of 40 mg twice a day was picked because that
was the dose at which 50 percent of the patients in the early
trials developed neuropathy. But the studies also showed that
you got a four-fold increase in antiviral effect, every time
you doubled the dose; from 40 to 80 total daily dose, you get
a four-fold increase in suppression. So if patients can
tolerate the dose, I go all the way up to 80 mg per day. And
in people with high CD4 counts, we can get to 80 in almost
all the patients. We are not seeing much neuropathy, if the
patient is asymptomatic with a high CD4 count when you start
aggressive treatment with d4T.
Once I get to the maximum dose of d4T, then about a month
later I repeat the bDNA test. With CD4 counts, we used to
wait four months or so before repeating it. But with viral
load, there is clear evidence that you can see measurable
decreases within two weeks of initiating effective
antiretroviral therapy. By a month you ought to be able to
get a good measure.
Say we had a patient whose viral load had gone from less than
10,000 to 40,000 copies, and we put him on 80 mg of d4T, and
a month later we measured it, and it is less than 10,000
again. We would continue that patient on the therapy, seeing
him or her at least once every four months, reassessing where
we are with their antiviral measurement -- and reassessing
their history and physical signs, and CD4 count.
If we see viral escape -- if the viral load starts going up
again -- I would add ddI to the d4t, and I would continue
that, monitoring every four months.
AZT and 3TC
If the patient's CD4 count still drops -- and if it drops low
enough so that they could qualify for the 3TC expanded-access
program -- we often put those patients on AZT plus 3TC. But
we wait to start AZT until we could start 3TC simultaneously.
Because it did not seem to make sense to put them on AZT
alone for a year or two, let the virus get resistant to that
drug, and then add the drug that helps to prevent the
emergence of resistance.
If the patient could not qualify for the 3TC access program,
I would probably wait, and continue to play with d4T and ddI
-- and maybe even add saquinavir -- and hold the AZT until I
could combine it with 3TC.
ATN: How is this different from your approach a year or two
Dr. Conant: The change in view you are hearing from me is
that we have traditionally used AZT as the first drug of
choice. That was the first drug we had, and we continued to
use it first. And yet, nothing says that we cannot first use
drugs that may not be as effective as AZT, and use up their
efficacy, and then go to stronger drugs later on.
If tomorrow the FDA approved 3TC, then I would probably begin
with a combination of AZT plus 3TC plus ddC, much earlier in
the course of the disease.
But in San Francisco you have an aging epidemic; most
patients have experienced these drugs in some dose, for some
period of time, over the last five years. So my
recommendation is that the patient and physician sit down,
find the drugs that were tolerated the best, and try to
combine two or perhaps three of those. For example, if the
patient could take AZT and had no problem with ddC, combine
the two drugs, and see what happens to the viral load. If the
viral load goes down, monitor them every four months; and if
the viral load starts going up again, shift to ddI and AZT.
Play with the drugs to get the viral load down as low as
With saquinavir now available on expanded access, if I were
the patient, even though that drug in the dose being used is
not as effective as the new protease inhibitors coming along,
I would add that to my cocktail, using the bDNA about a month
after I changed to see how effective the treatment was. If
the viral load goes down, great; then I would wait another
four months and check it again.
ATN: The Consensus Symposium on Combined Antiviral Therapy is
being held in Lisbon July 25-27. My concern is that there is
not much consensus, and we might end up with a lowest common
denominator kind of recommendation.
Dr. Conant: More and more my colleagues are using similar
strategies. They are starting with combinations earlier, and
they are using bDNA to tailor what the combination is.
ATN: I have heard of a few people who had very low CD4 counts
who were treated with AZT plus 3TC, and in a few cases their
total lymphocyte count went way down, but the CD4 and CD8
counts stayed about the same or decreased only a little. Have
you noticed any problem like that, perhaps where the
hematological toxicity of AZT was getting amplified?
Dr. Conant: We have not seen that. In our experience, 3TC has
been the best tolerated of the antiretrovirals. Many patients
feel better after they start taking it. There can be problems
with peripheral neuropathy -- especially in patients who have
had peripheral neuropathy in the past. And there are
occasional reports of minor dizziness, hair loss, or joint
CD4 Percent Vs. CD4 Count
ATN: You mentioned that the CD4 percent is actually a better
measurement that CD4 count, even though the latter is the one
generally used. With CD4 percent, what danger points do you
watch for? And could you explain why the percent is the
Dr. Conant: When the CD4 percent goes below 20 I start
We need to keep stressing to patients that if your CD4 count
drops a little -- from 300, for example, down to 230 -- then
look at changes in your CD4 percent. If the percentage was 21
percent, and the count dropped to 230 but you still have 21
percent, there really has not been a drop.
This is because of the way the CD4 count is calculated. You
start with the white blood count and see what percent of
those cells are lymphocytes. Then you see what percent of
those are CD4 cells. For example, if your white count is 4000
per milliliter, and your percentage of lymphocytes is 50
percent, then that means that 2000 of the cells are
lymphocytes. And if the percent of lymphocytes that are CD4
cells is 25 percent, that means that 500 of them were CD4
cells, so your CD4 count would be 500.
Why is the CD4 percent more important than the CD4 count?
Because your white blood count can vary from hour to hour,
just because of time of day, or exercise, or a small
infection. If the white blood count changes for these minor
reasons, it may look like the CD4 count has gone down or up.
It is well known that the percentage is most important; and
yet patients, unfortunately, are still stuck on the number.
We doctors did that; we continued to talk in terms of numbers
instead of percentages, although as early as 1988, at the
International Conference on AIDS in Stockholm, it was clear
that percentage was really the most important figure.
ATN: Where do you see AIDS treatment heading in the near
Dr. Conant: We are very excited about both the Abbott and the
Agouron protease inhibitors. We are seeing drops in viral
load of two to three logs (100 to 1000 fold), and preliminary
data suggests that those reductions are sustained for
extended periods of time, without the emergence of
resistance. I would predict, based on the data we are seeing
today, that we may well see protease inhibitors, instead of
nucleoside analogs, as the first line of therapy.