|
On October 30 AIDS Treatment News interviewed David D. Ho.,
M.D., director of the Aaron Diamond AIDS Research Center in
New York City, and a professor at The Rockefeller University.
Aaron Diamond, the world's largest independent AIDS research
laboratory, "focuses on basic research efforts to increase
understanding of the structure and function of HIV and to
define the mechanism by which it destroys the immune system."
We asked Dr. Ho to tell our readers what he believes is most
important now in HIV treatment research.
Hit Hard Hit Early
AIDS Treatment News: Some see a backlash against the "hit
hard hit early" philosophy. Some physicians say that is right
for some people, but that others -- especially those who have
an excellent prognosis anyway for several years -- maybe we
should hold the fort with a regimen such as d4T plus ddI, and
reserve protease inhibitors, non-nucleoside RT inhibitors,
and other treatments where viral resistance can develop
rapidly, until we have better drugs and combinations
available. How do you answer?
Dr. Ho: Some people are looking at failures on combination
therapies which include a protease inhibitor, as an
indication that the concept of hit early hit hard is
incorrect. My view is the opposite, given what we have seen
over the past year. The failures on combination therapy have
generally occurred when patients are more advanced in the
disease process, and have taken prior antiretroviral therapy,
usually as monotherapy or dual therapy. And one must consider
the issue of compliance or adherence; poor adherence is akin
to hitting inadequately hard.
Those are the principal reasons for failure on combination
therapy. Our experience suggests that when treatment is
successful, patients maintain control of HIV replication on
combination therapy. Most of them were drug naive, or at
least without extensive antiretroviral drug experience. The
earlier you start, the greater the chance of sustaining the
antiretroviral effect. So the practical experience does
support the hit-hard-hit-early notion.
ATN: When you are the treating physician, who if anybody
would you suggest should not start treatment at all?
Dr. Ho: There are certain patients who are HIV infected, but
have either a normal or slightly depressed CD4 count, and a
viral load which is spontaneously quite low. Their prognosis
without treatment is quite good, based on those numbers, and
there is no reason to rush into any form of therapy. One
could comfortably monitor their viral load and CD4, and delay
making any decision on starting treatment.
ATN: A big concern is that many people are not ready to start
this complex regimen, and do not have the commitment they
will need to see it through for year after year after year.
The fear is that they will start something and not be ready
to maintain it.
Dr. Ho: This is obviously a very important decision. If the
commitment is not completely there, then I would agree. If a
patient cannot face a very complicated regimen for several
years at least, then starting therapy may in fact be bad.
Without the commitment, adherence will likely be poor,
leading to drug failure, and many drugs will be used up. That
would not be in the patient's best interest.
ATN: What would you say about treatments like d4T plus ddI
that are not recommended in the guidelines, but are picked to
be drugs where resistance tends to develop slowly -- as
intermediate between no treatment, and the full "hit hard hit
early"?
Dr. Ho: I think if the situation were not as desperate, if
somebody is pretty well off and wants to take that sort of
approach, of reserving stronger drugs for later use, I think
an argument could be made in support of that strategy.
However, with a drug combination such as the one you
mentioned, it is fairly clear that you cannot adequately
control the virus, there will be residual ongoing
replication, and some variant forms will eventually emerge.
So the antiviral effect is unlikely to be very durable. And
even though ddI may not be the most crucial drug to have in
your arsenal, you will still use it up. It takes away one
card that we may want to play in the future. Overall, I
cannot endorse such an approach.
Maintaining Blood Levels
ATN: At the recent ICAAC conference there were posters on the
correlation of the nadir of indinavir blood levels [the
trough or lowest concentration of indinavir (Crixivan®),
for example, in the blood between doses -- which usually occurs
just before the next dose], with how durable the effect was.
If the drug level became too low between doses, viral
resistance and rebound would likely develop. So there has
been an interest in the community -- maybe less so in industry -- in individualizing doses, by checking blood levels when a
person starts on a drug, to make sure it is being absorbed,
and that the nadir stays high enough to maintain an antiviral
effect. What do you think of this approach? Does it make even
theoretical sense? And if so, what about the practicalities
of getting it introduced?
Dr. Ho: It makes a lot of theoretical sense. When we first
tested protease inhibitors, we noticed that there are lots of
individual variations in drug concentrations. In those days
we were doing pharmacokinetics regularly. The same dose could
achieve concentrations that vary considerably between
patients, suggesting significant individual differences in
drug absorption and metabolism.
We found quite a bit of individual variation in how these
drugs are absorbed and metabolized. If those concentrations
could be measured easily and dosing could be tailored on an
individual basis, this would certainly make sense. But on a
practical level, even for researchers, it is not easy or
inexpensive to get these measurements made. In clinical
practice, it would be even more difficult.
ATN: Some researchers have told us that companies have not
been cooperative with them in developing these tests -- for
example, refusing to provide pure samples of the drug to
calibrate their tests. Have you had any such experience?
Dr. Ho: I do not know much about that. Many of the companies
do not even do these assays [for drug levels in blood]
themselves; they contract out to certain commercial
laboratories that measure these drug concentrations by
specialized chemical techniques. I find that on a practical
level, it is difficult to get these measures done.
ATN: This may be a bit blue sky, but one HIV drug now being
developed, T-20, will have to be given by injection. So they
are looking at the infusion pumps, the little programmable
ones that look like a beeper; I'm told they are working well
for diabetics, for giving insulin. It would seem that this
should be investigated for other antiretrovirals, since it
would get around problems of adherence, the number of pills,
food or no food, acidity in the stomach, and individual
absorption from the GI tract. Most importantly, continuous
infusion might provide better control of blood level and HIV
replication. What do you think of eventually looking to an
infusion pump for delivering some of the antiretroviral
combinations?
Dr. Ho: We had some experience using such pumps many years
ago, back in the late 1980s, to administer soluble CD4. We
know it is do-able, although it will not be easy. Also, in
testing the early generation of the Abbott protease
inhibitor, the compound referred to as 77003, a pump was
tried for a short period, because the oral pharmacokinetics
were not good.
During those periods, the sentiment was that parenteral
[injection] administration was so infeasible that we should
move to oral medications. Today we have mostly developed
orally available drugs; we are faced with the adherence
issue, and people are thinking of ways to improve adherence.
If we have a drug that needs to be administered not daily but
occasionally, I think such a strategy may be an important
adjunct. But I find it difficult to think that we are going
to be able to administer three or four different drugs by a
device such as that; on a practical level it would be quite
difficult.
Instead, to deal with the adherence or compliance issue, I
think it is more important for companies to develop drugs
that are easier to take -- for example, to move from three to
two to one time per day. Many pharmaceutical companies in the
HIV field have taken that on as an important objective,
because the easier the regimen, the higher the adherence, and
the higher the success. And of course some of the non-adherence problems are related to toxicity -- so companies need
to make related compounds that have fewer adverse effects. I
think that on a practical level, these approaches might work
better. Certainly I would not be opposed to trying the pump,
but I think that daily use of injectable drugs would be
tough.
ATN: I hear that in diabetes, people have no trouble wearing
the device permanently.
Dr. Ho: But that is just delivering one drug, insulin. With
our patients, it's unlikely that any patient will get just
one drug. And some drugs could not be administered in
intravenous or subcutaneous forms. And if the different drugs
need to go into different solutions, then you will be
constantly reloading the device with different drugs. If it
were a single drug, as in diabetes, it might work.
After Virological Failure
ATN: When there is virological failure with a triple
combination, what can people do next? We hear many anecdotal
reports of people whose treatment has "failed" virologically
with some of the protease inhibitor combinations, in that
viral load comes back up but usually not all the way to
baseline; however their CD4 count seems to stay high, and
they seem to be doing well clinically. What do we know about
what's going on? And how should these people continue their
treatment?
Dr. Ho: There are several scenarios. Take the patient who is
previously antiretroviral naive, who gets treated with a
triple combination, who has good suppression for a while,
then after some period of time, say a year, the viremia
reappears. In that case, you could have viremia that bounced
back to the prior baseline -- which means that the virus is
resistant to two or three of the drugs. The thing to do there
is to change to a new regimen. Since the patient is
previously drug naive, the doctor still has several other
options to work with.
There is another scenario, however -- closer to the one you
presented. Viremia reappears, but at a much lower level.
Therefore, that lower level does not seem to impact the CD4
count or the clinical picture. What do you do?
This is a difficult issue. Clinically and immunologically,
you are still benefiting the patient by continuing the
medications -- because the viremia is suppressed compared to
the baseline, and the CD4 count is still at a higher level.
The concern here is that there is now residual virus
replication; we are all aware of the possibility that
resistant viruses will eventually emerge, and the question
is, will that take a few more months, or will that take
another year or two?
We do not know what the right answer is. For those inclined
to be more aggressive, they may go ahead and add a drug, and
see if they could get it back down to undetectable levels -- to
arrest the evolution of viral sequences, and therefore block
the further development of resistant virus. Others do not
like to add just a single drug; they would rather change the
whole regimen. We do not have enough experience to know how
persistent such a situation could be.
Then there are patients who have taken several drugs before,
and begin a more powerful combination and suppress viremia
for a period of time, but then the viremia comes back at some
intermediate level. Here it is hard to make pronouncements or
general guidelines about what to do; you have to deal with it
case by case, depending on the specifics of the situation,
the prior drug experience, and what options are left to come
up with a combination of drugs that a patient has not seen
before.
ATN: What could be said about the odds or percentages -- either
for naive patients starting a hit-hard-hit-early regimen, or
people who have had much prior experience and have been
through most of the drugs besides protease inhibitors, and
then start their first protease-inhibitor-containing regimen?
Are there any situations where we have fairly good data to
let people know what their chances may be? And to what extent
are the problems due to lack of adherence, and therefore
under control to some extent by the patient?
Dr. Ho: We have some indications, but not much cohort or
trial data to guide us. In drug naive patients who start
these combinations reasonably early, and adhere to the
combinations, in our experience clearly over 80% to 90% have
durable suppression for over a year, and some have been on
treatment successfully now for over two years. But adherence
to these difficult regimens is a very real problem. And often
it is hard to predict in advance who will stick to the
regimen, and who will not. In addition to the true drug
failure, we are going to have somewhere around a third who
will be significantly non-adherent to the regimen -- even in
motivated patients such as our trial subjects.
But if someone is very confident that they can stick to a
twice-daily regimen for several years, and if that patient is
drug naive and reasonably early in disease progression, the
chances of prolonged suppression are quite good.
On the other extreme are patients who are in and out of the
hospital, who have taken several of the reverse transcriptase
inhibitors in the past, and then start on a typical triple
combination. I think the recent San Francisco General data
suggests that about half of those patients will fail. That
takes into consideration everything: the prior drug
experience, the more advanced disease status, and includes
any adherence problems.
ATN: Yet so far I have heard that most of those patients are
doing well, even the ones who have virologically failed.
Dr. Ho: That is a very important point -- what do you mean by
failure? That failure in the San Francisco study was defined
as any virus being detectable; it was a virological
definition. If you have ongoing, detectable virus, the
concern is that the result will be more resistant virus, more
viral load, and clinical failure. But having a little bit of
viremia but still having the virus under reasonable control
is clinically very beneficial.
The San Francisco study was portrayed in the lay press as
terrible news. It is certainly not great news, but not so bad
if you consider the clinical perspective.
And you need to realize that was a chart review study,
looking back at the early experience with these treatments,
in a hospital setting with advanced patients. That study
design was prone to yield a more pessimistic view of
combination therapy including a protease inhibitor.
ATN: I talked to Dr. Deeks, and we published a short article
on that report ["ICAAC: Newspaper Headlines Misleading on Protease Inhibitor Failure," AIDS Treatment News #281,
October 17, 1997], to give people a different perspective
than was misreported in the headlines.
Viral Load -- More Sensitive Tests
ATN: Now there is data suggesting that a viral load below 400
copies may not be good enough; for the most durable viral
control, you would really like to know that you are below a
much lower cutoff, for example below 20 copies. The practical
problem is that most patients cannot get the Ultrasensitive
test that goes down to 20. As a stopgap, would it make sense
to use the available test and look for a result of zero, no
virus detected at all, instead of a number which is above
zero but below the quantification limit of 400?
Dr. Ho: If the threshold of detection is 400, then there is
no point in reporting a specific number below that 400; that
would confuse people, and is not reliable.
You can find some patients who consistently have a result
between 20 and the lower limit of the more available viral
load tests [400 or 500]. Some people on the modern
combination therapies are in that category.
How we interpret this situation, and whether we should do
anything about it, is a gray area now. It is unlikely that a
viral load of, say, 150 copies per ml will be associated with
rapid progression and clinical disease. So clinically, you
probably do not need to do very much to modify whatever the
patient is on. However, this test result tells us that there
is continuing viral replication; the virus will continue to
evolve, and we have to be careful about a breakthrough later
on. Since there is not a lot of clinical experience to guide
us, some people are saying we should just follow these people
carefully; others, if there is flexibility, would add a drug
to see if the virus could be dropped below 20 copies per ml.
But these assays [those which accurately measure viral load
when it is below 400 copies] are only available today to
research groups, or to patients who pursue it directly with
the companies.
Viral Resistance Testing
ATN: On viral resistance testing, what role do you see for
these tests -- both now, and in the future, for selecting which
drugs to start somebody on, or which drugs to change to?
Dr. Ho: I think these tests probably could be very helpful,
if they could be made simpler, with faster turnaround, and at
some reasonable cost, say equivalent to a viral load test, or
only slightly above that. We do not have anything today that
meets these criteria; however, a number of people and
companies are working toward those goals.
These assays will most likely be useful when a patient goes
on combination and is failing it, and you need to need to
know what to do next. While the patient is still taking the
failing regimen, so that the viruses are breaking through
despite the presence of the drugs, you could take the virus
and measure their resistance profile, and get a reasonable
reflection of what is going on. Then if the viruses are, say,
resistant to two of the drugs the patient is taking, and
sensitive to the third, you would use that drug in
combination with two or three other new drugs. In that
setting, resistance testing will be fairly useful.
One has to be careful about using any of these tests on a
patient who has gone off the drugs already. Say the patient
has taken AZT plus 3TC in the past, and has resistant virus
to those two drugs, but the drugs are no longer being used.
The resistant virus would revert back to a minority
population, say only accounting for 5% of the viruses. The
resistance assays used today are not sensitive enough to pick
that up; you would miss it, and get a false sense of
security, and may use drugs that would be doomed to fail.
So while the patient is on the medication, I think the tests
are useful. But once the patient has gone off the drugs, and
the resistant virus is the minor population, the tests we
have now would not be reliable. There are ways to deal with
this; surely the companies will come up with such strategies
in the future, but they are not there yet.
Therefore, at the start of therapy, I think that taking a
history [of previous antiretroviral drug use, where some of
those drugs are likely to have been discontinued] may be at
least as useful as measuring the virus -- unless you are
dealing with a patient who was infected with a drug-resistant
virus in the beginning.
ATN: Looking ahead maybe three to five years in the future,
do you think most of the usefulness will be from the
phenotypic tests [which see how well the virus can grow in
the presence of drugs] or the genotypic tests [which look for
mutations associated with resistance to particular drugs]?
Dr. Ho: I think that ultimately we would like to have a
phenotypic assay -- the same approach used successfully for
testing antibiotic resistance very quickly in the laboratory,
to strep and staph and other pathogens. The genotypes are
quite difficult to interpret; because of mutational
interactions, their clinical consequences are often very
difficult to predict. We want a phenotypic assay in the long
run.
Targeting Resistant Protease
Dr. Ho: Here is one project I would like to see
pharmaceutical companies take on. They know how to make
protease inhibitors now. They could take a very resistant
virus -- for example, a virus highly resistant to indinavir --
and crystallize that protease, and start the whole effort
over, and develop a second generation protease inhibitor that
will attack the drug-resistant variants. Companies have been
through the process [of developing protease inhibitors], they
know the strategy, and these strategies have been quite
successful over the last few years. Why not repeat it, and
come up with compounds that would be active against the
viruses that are resisting the first generation protease
inhibitors? I think that would add a lot to the arsenal for
our patients.
ATN: Isn't this the approach that Abbott used to develop its
second generation?
Dr. Ho: I am not sure if that is the strategy they used. ABT-
378 has greater potency against the wild-type virus, and good
pharmacokinetics. It tries to overcome the resistant virus
through greater potency and higher drug concentration. To my
knowledge, it was not designed against a resistant virus.
I think this process might work better if we start from
scratch, with a drug-resistant viral protease.
Is Eradication Possible?
ATN: What do you see, in the next several years, on
eradication possibilities?
Dr. Ho: We are learning quite a bit. Eradication is still a
goal; it is not something we or anyone else has achieved so
far. But we are learning a lot about different viral
compartments, and how quickly they could be eliminated once
strong therapies are applied.
The available therapies have helped us eliminate over 99% of
the virus in infected individuals. But we are learning about
additional viral compartments. In our patients who have been
treated very early and very hard, and have received this
treatment two years now, we are still finding a residual pool
of virus resting in certain CD4 cells, in a very quiescent
way. This poses a new obstacle to deal with.
Some people would say it means eradication is not possible.
But those of us who are working on the science look at it
differently, and say this is information we need to achieve
eradication. Now with this information, how do we move on?
How do we flush that residual pool out, how do we protect the
individual from that virus reactivating and spreading the
infection?
Much fairly exciting basic research is underway, by several
groups. If we go back a few years to 1994 or 1995, we would
not have predicted that we would be dealing with such a small
pool of virus after a year and a half or two years. But on
the other hand, the cells that contain this virus are
quiescent, so they will not turn themselves over very fast.
So we need to think about continuing antiviral therapy, but
also using certain immunological strategies to activate those
cells so they could be flushed out. It poses a new obstacle;
but it is a scientific challenge which has certain
theoretical solutions. The research groups engaged in these
issues are taking that on. It is quite exciting from a
scientific perspective.
Can One Be Re-Infected with a Second Strain of HIV
ATN: Is there much re-infection with a different strain of
HIV, after one strain has become established?
Dr. Ho: We can show in monkeys that re-infection occurs; we
do know from the various recombinant viruses that we find in
people throughout the world, that dual infection or
superinfection must occur on occasion. But on a practical
level, within our patients, there have been precious few
documented cases of superinfection. Some groups have looked
for evidence that this has occurred, and have not found it.
Treatments to Watch
ATN: Over the next three to five years, what are some of the
particular drugs and treatment strategies and approaches that
you think we should be keeping an eye on?
Dr. Ho: Some of my answers will be the standard ones you hear
from others. We will have several more reverse transcriptase
inhibitors, both nucleoside and non-nucleoside. We should
have a few more protease inhibitors. But I also think that
the companies that are making these drugs will come up with
equivalent ones that are less toxic and easier to take. If we
could move some of the protease inhibitors from three times a
day to one time a day, it would be a major advance. Some of
the companies are working toward these goals.
Also, a couple major pharmaceutical companies are working on
integrase inhibitors, a potentially very important class of
drugs.
The recent breakthroughs in chemokine receptors, as cofactors
necessary for viral entry, have led to much screening to find
drugs that might inhibit one or several of these co-
receptors. At the screening stage, the effort has been quite
successful; there are many "hits" [compounds that work in the
laboratory screening tests], and they are being pursued. We
probably will not hear much about them for another year or
two, when they get to toxicity testing, and some to clinical
studies. But there is much activity already.
And there are other new kinds of drugs being developed, such
as T-20 that attacks the envelope fusion process, and other
drugs that attack the viral protein.
Key Issues
ATN: What do you see as the most important issues to leave
with our readers?
Dr. Ho: The issue of eradication, the goal, what we are
learning about the obstacles, and the strategies to deal with
those. The failures: it is important that patients understand
what are the principal causes of failure; many of them
involve treating advanced-stage patients who have prior
antiretroviral therapy. And the issue of adherence.
When we hear statistics, we have to understand under what
setting were the numbers generated -- and not just walk away
and say that study says it's 90% successful, and this study
says it's 50%, so somebody is wrong. One has to interpret all
of these results in the context of the patient population
that is being studied.
Also, the word "failure" is being thrown around without a
precise definition. Is it virologic failure, is it clinical
failure? We need to keep this difference in mind.
|
