Doctors’ Roundtable
edited by Enid Vázquez
The following transcript
is edited down from a teleconference hosted by HIVandHepatitis.com,
as part of their HIV Treatment Live! monthly teleconferences.
The topic is “Report from the 7th Conference on Retroviruses
& Opportunistic Infections,” which was held during the
first week of February. For a complete transcript, contact
ronbaker@pacbell.net
or write P.O. Box 14288, San Francisco, CA 94114. Request
to be put on their mailing list for notices of future teleconferences.
The clinicians and researchers speaking in this edited version
are Michael S. Saag, University of Alabama at Birmingham;
David Cooper, National Centre in HIV Epidemiology and Clinical
Research, University of New South Wales in Australia; and
Stephen Becker, private practice in San Francisco.
Saag (moderator): Let’s
jump right into antiretroviral therapy, especially on issues
of when to start treatment. Steve, there was a poster by John
Bartlett about a “meta-analysis.” Can you tell us
about that?
Becker: Sure, although
I think it’s important to say that the question of when
to start therapy was not answered at this conference. It has
been a pressing question and I think one that has in fact
become more pressing as we’ve come to appreciate the
complications of therapy. The poster was an analysis of approximately
10 randomized trials in treatment-naïve [no prior therapy]
patients. These were trials performed in the United States
and Europe and they included a number of drugs, including
protease inhibitors (PIs), the non-nucleoside inhibitors [NNRTIs],
and combinations that included three nucleoside [NRTI] drugs.
Dr. Bartlett looked at the predictors of a successful virologic
response—in other words, driving the HIV viral load to
less than 50 copies and sustaining that for a 48-week period.
Among the factors that he looked at was the baseline or entry
CD4 T-cell count, the HIV viral load count, the type of drug
that was used, and the number of pills taken in the regimen
under study. The factor that emerged as the most predictive
factor for treatment success was the number of pills. Study
regimens that had the fewest number of pills were associated
with the greatest success in terms of HIV viral load reduction
[HIV in the blood].
Saag: David, could
you summarize some of the data that was discussed about complications
of treatment, especially metabolic complications?
Cooper: There was a
lot of data presented at the meeting. In fact some people
who were around the poster area on Sunday actually nicknamed
the poster area “Side Effect Sunday”—that there
were just so many posters there. I think the issues relate
to how common are the metabolic disorders and lipodystrophy.
I think that most cohorts are reporting on average about 50%
of patients have some form of lipodystrophy. However, I think
it’s also pretty clear from these cohorts that only about
10% are severely affected with them—and that’s probably
a little bit encouraging. People of course are very disturbed
about the body shape changes, the central fat accumulation
and the fat loss in the arms and legs and face. But the big
question is whether the other features that occur in these
metabolic disturbances, such as the development of early diabetes
which is related to insulin resistance and high cholesterol
and high triglycerides are going to be problematic for you
in the long term. There’s no simple answer to that, but
there are some initiatives now to look at that quite rapidly
by taking very large numbers of people—about 30,000—all
over the globe and looking at them for the development of
any heart disease or stroke. So that’s an important development.
Becker: David, could
you comment on the difference between some of the body composition
changes and normal aging changes? There was some discussion
about the need to explore this as well.
Cooper: One of the
difficulties is that the fat wasting is relatively easy to
recognize and that’s probably not an aging phenomenon.
But the fat accumulation is. We tend to get fatter—particularly
we become apple or pear-shaped—as we get older depending
on our fat distribution. It’s a little bit hard to distinguish
the fat accumulation from the drugs from just this natural
tendency. There are studies going on to look at that in the
U.S.
I think there was some encouraging
data about lipodystrophy syndrome and its occurrence rates
from the HOPS (HIV Outpatient Study) data. They showed that
the more risk factors you had for lipodystrophy, besides being
on the drugs, the more likely you were to have it. And that
really is an important issue—that we ought to be watching
these other risk factors for heart disease and fat changes
and diabetes very carefully.
Saag: What can we do
about it if a patient has, for example, hyperlipidemia [high
blood fats]? Switch the components of the regimen or intervene
with specific anti-lipid agents?
Cooper: There was quite
a bit of data to give us some indication on that. There were
at least a dozen what are called “switch studies”
where people were switching off protease inhibitors to non-nucleosides.
Basically, sadly, they were pretty disappointing. The body
shape changes, particularly the fat wasting, didn’t really
improve. But the lipid levels—the cholesterol and triglycerides—did
go down substantially. So that was a strategy. There was also
one important abstract at the late breaker session which gives
us some indication about the use of “statin” drugs.
We’ve always been worried about these statin drugs because
they can be quite toxic, particularly sometimes in the presence
of other drugs—and this theoretically would include protease
inhibitors. And there was some data from the ACTG, the AIDS
Clinical Trials Group in the U.S., from their pharmacology
groups to show that some of the statins at least were safe
to use. Pravastatin (Pravachol) was probably the best, atorvastatin
(Lipitor) was second, and it was rather hard to use simvastatin
(Zocor)—you got very high levels with that.
Saag: One of the things
we’re also noticing is that people are failing therapy,
or therapy is failing them, let’s say, virologically
[looking at viral load in the blood]. And some of the approaches
to dealing with that included some new regimens, adding two
PIs together.
There was a late breaker which
showed that if you had a backbone of drugs that had either
amprenavir (Agenerase) as a single PI or amprenavir with a
number of other PIs including either nelfinavir (Viracept)
or indinavir (Crixivan) together, that the dual PI combination
actually had a better outcome. The concerning part about this
is, while every regimen did include an NNRTI, as you might
predict, the NNRTI naïve patients had a much better shot
at getting the viral load under control. And probably the
best that you could see in the NNRTI-experienced patients
was about a 20% success rate at getting below 200 copies [vs.
45% for the naïve group].
There are other approaches
though that may be coming into play, and one that’s been
talked about a lot is withdrawing therapy from patients who
have had virologic failure for a period of time. Steven Deeks
is here and he presented some data on that.
Deeks: This whole concept,
this so-called “treatment interruption strategy”
(STI) or “drug holiday,” is based on the theory,
the premise, the observation, that if you take away your medications
and you no longer have drugs around that maintain the resistant
mutations of the virus, that those resistant mutations will
go away over time. We did a study that we reported today where
we took 18 people who had been on therapy for a long time
and had a detectable viral load for like 2 1/2 years now—so
long-term virologic failure—people who were otherwise
doing okay who stopped their drugs, and we observed that over
a matter of anywhere from 6-12 weeks that their resistance
disappeared completely. And we really had trouble detecting
it during that period of time.
Now this was associated, as
one might guess, with a significant drop in CD4 T-cell counts—T-cell
counts dropped by about 100 cells. These people coming into
the study had 200, 250 cells to begin with—so people
in general lost about half of their T-cells as a consequence
of stopping their medications. And this is because the medications
they stopped were, to a certain degree, still working. So
we looked at this issue, when you stop therapy, viral load
went up by about 10-fold, T-cells dropped, and resistance
sort of went away. Now, this raises questions as to whether
this is a good thing—no resistance—or a bad thing—loss
of T-cells—and there’s just no way really to address
that with our study. If this is a strategy that will be pursued
outside of a clinical trial, people need to be aware that
there is a lot of risk.
Saag: I think some
of the key points you made, Steve, just to kind of summarize,
is one, that when the drugs were stopped it took a little
while, 5 weeks or so, before the reversion to wild type [the
original HIV] from resistance occurred, but when that did
happen—which seemed to be pretty abrupt—that that’s
when the viral loads tended to shoot up and the CD4 counts
drop. And your point, I think, was that that may have been
an issue of fitness of that resistant virus, and when you
switch back to wild type, that was a more aggressive form
of the virus.
Deeks: Yes. In this
whole issue of virologic failure, one of the dominant themes
was this whole relationship between the fitness of a virus
and resistance. And I think our combined data, including the
“stop therapy” study, is that despite years of virologic
failure, despite having high levels of resistance to the drug
that people are on, the virus does not replicate that well.
And we believe that this is because the virus is not that
fit. And in our studies, this is associated with continued
persistent CD4, and presumably clinical, benefit. And I think
that that story was confirmed by multiple different observations
at this particular conference.
Saag: I think another
one that you presented a couple of days ago also plays to
that in that there seems to be some residual clinical benefit
even if the viral load is now detectable, but certainly significantly
below baseline. Could you briefly describe what that study
was about?
Deeks: Just looking
at 450 patients who we’ve been following for years now
at San Francisco General Hospital, we looked carefully and
tried to determine why some people’s T-cells remained
elevated when the virus comes up. And we pretty much, I think,
observed that again, in the state of long-term virologic failure,
the virus can go to very high levels but rarely goes all the
way back to baseline. And that for reasons that are unclear,
just having a small reduction in the viral load is associated
with a significant increase, and preservation of that increase,
of the CD4 T-cells. And I think that has implications. We
all know this, that there’s a significant delay between
virologic failure and what happens to the immune system.
Saag: Exactly. And
I think it’s consistent with some meta-analyses that
have been done in the past, looking at clinical trials that
had clinical endpoints from the old days. And that, if anything,
if you could achieve a 0.5 log decrease from baseline and
sustain that, that was associated with clinical benefit. And
if you lost that, then you were more likely to see progression
to a clinical endpoint [such as disease or death]. And while
your study didn’t exactly define a precise cut—it
seemed like it was more of a continuum—I think it’s
safe to say if you’re 0.5 log below baseline and don’t
have a whole lot of options to switch to, taking it all together,
you might be better off to ride that horse a little while
longer.
Deeks: I think, Michael,
that’s the key thing. If you run out of options, if you
no longer have the ability to get that virus down to undetectable,
or if you only have one class drug left and you don’t
want to use it up now (like the non-nukes), then I would take
these observations and say, “Yes, let’s keep the
virus as far below baseline as we can and ride it for as long
as we can.” I think that’s a reasonable strategy.
|
