|
Keith Henry, M.D., is Director of the HIV program at Regions Hospital,
St. Paul, Minnesota. Regions Hospital has an HIV research program and is
affiliated with the AIDS Clinical Trials Group (ACTG) at the University of
Minnesota. Many patients are indigent and initially uninsured, and many
have already had extensive antiretroviral treatment.
AIDS Treatment News: Explain your concern that the new treatment
Guidelines (published for comment by the U.S. Department of Health and
Human Services in July 1997) are leading doctors to rush into protease
inhibitor treatment for some patients who might be better off with less
aggressive approaches for now.
Dr. Henry: One of the take-home messages from the ICAAC meeting (September 28 -
October 1, in Toronto, Ontario) was that in the real world, up to half of
the people who have started on a protease-inhibitor based
antiretroviral combination have virologically failed that regimen. In my
practice I have also seen that treatment combinations which include a
protease inhibitor have failed to durably suppress HIV viral load in about
50% our patients. This is a scary figure; nationally, it would likely mean
that every day there are more people becoming resistant to protease
inhibitors than people being infected with HIV. Can we feel secure that the
impressive reduction in AIDS-related death and illness, which we have seen
and are still seeing today, can be maintained?
How can one prevent protease inhibitor resistance? And is there really
a need to rush and put everybody with a detectable HIV RNA level onto a
protease-inhibitor based regimen?
When working with patients I am often uncertain about what to do --
except in classic situations. For example, when a person is sick due to
HIV, and has a high viral level and low CD4 count, I treat them very
aggressively, and usually start with four antiretrovirals. On the other
end, if somebody feels great and has a very high CD4 count and very low RNA
level, I may choose to just follow them and not prescribe antiretroviral
drugs at this time.
But it is hard to know how to treat everybody in the middle. With only
50% success in getting durable viral suppression with a
protease-inhibitor regimen, I am reluctant to jump into something that
perhaps we do not know as much about as we think we do.
Physicians may be lulled into complacency with the new drugs, because
the initial viral load drop looks good. But after the viral load goes below
the test limit, they probably see people perhaps once every three months to
repeat the test; this is a typical recommendation. If suddenly the virus
rockets back, three months is a long time and there could be resistance to
several kinds of drugs, and then further treatment choices are limited.
The natural history of HIV suggests that a number of people have an
excellent prognosis for years, and that we can identify many of them with
clinical evaluation and blood tests. If we are trying to keep a person
healthy for 20 to 30 years, and almost certainly we will have better and
simpler drugs several years away, does one need to commit now to regimens
that have a 50% virological failure rate?
ATN: Is that failure rate only for your patients who are heavily
pretreated, who therefore are the hardest to treat?
Dr. Henry: Yes. But in the United States today, an estimated
three quarters of a million people already have HIV, and 40,000 a year
become newly infected, and an increasing percentage of them are pretreated.
For any given clinic population, a relatively small percent are
treatment-naive. And it will remain that way, if the push is always to
start treating almost everybody quickly. If we do not know the best way to
start patients and maintain them, even in the ideal patient population
where they have never seen drugs before, a high percent may at first
achieve viral suppression below the test limit, but then have the virus
return.
ATN: Why are we seeing differences between clinical trials and
clinical practice?
Dr. Henry: Major clinical trials are reporting that 80% to 90%
of treatment-naive volunteers have complete viral suppression that looks
like it may last for years. But very few doctors are claiming this high a
success rate in their clinics, even with the treatment-naive patients who
are easiest to fully suppress.
The reason may be that the resources are not there to provide education
and adherence strategies that would equal what happens in a clinical trial.
In the trial, you tend to select for adherent patients. And typically there
is one study nurse to 20 volunteers; it is the nurse's job to have them
come in and take their medicines. In a typical busy practice, the ratio of
nurses to patients may run from one nurse per 150, to one per 300 or even
500 patients. What can be accomplished with phone calls, checking up on
things, is quite different in this clinic setting.
And the reimbursement for adherence counseling is almost nonexistent. I
would be curious if any practitioner in any state is being paid
appropriately by coding that task somehow [on insurance forms]. Here in
Minnesota we do not get paid anything for it.
It is a big deal for any individual patient to be started on
antiretroviral treatment. Bells should ring, cannons should fire; it is a
huge step for that person. Yet it has become somewhat trivialized because
we do it so often, and with so many different people, and in the clinic we
want to spend more time talking about it, but we cannot. I wish I could
have a team of pharmacists and clinical educators work with the patient,
but that is not available.
Because of the new Guidelines, a doctor who senses that patients could
not be adherent to a complex regimen, and does not give protease
inhibitors, may feel deficient. I believe this is a problem with the
current recommendations. Might it be better, for many patients, to wait
several years and use no therapy now, or use a treatment that would
basically hold down the fort without lots of resistance -- until we can do
a better job with adherence, and have simpler drugs? Or should we go with
the Guidelines and go for broke? I do not think there has been enough
debate about this. The results from ICAAC shook up people who have only
been reading the headlines of the publications in the various journal
articles, saying that the results with the aggressive treatments are good.
Even experienced physicians sometimes feel bad when they are not in
compliance with the recommendations. What do doctors with less experience
feel? I have patients who say they want a certain treatment, but it is
clearly not the right thing at that time. Because of my experience I can
tell them that, and they believe I am looking out for their best interest.
But physicians with less experience may just start protease inhibitors
because it is the thing to do. That is not ideal for developing a long-term
strategy.
ATN: Ever since the Guidelines came out, many people have
thought that they were too aggressive for some patients.
Dr. Henry: I also think they are not aggressive enough for many
others. The line is too thin if you may have just one key drug in a
combination. So I often go straight from two drugs to four, double protease
inhibitors and double nucleosides. I feel more comfortable with the margin
of error. And often the twice-daily dosing which the four-drug regimen can
allow is an important advantage. For many patients the four-drug
combination might be a better, more durable therapy than three drugs with
only one protease inhibitor.
Non-Protease Regimens
ATN: For those who do not need such intensive therapy, what are
some of the treatments you use to "hold down the fort" for those
who are not very advanced?
Dr. Henry: The d4T plus ddI regimen is one I am using more and
more -- with the ddI once a day, at night, and d4T twice a day. Some people
also add hydroxyurea. The virologic response seems good. With d4T it has
been very difficult to know when or if someone is resistant. People have
criticized this approach, saying there is more cross resistance than we
know about; but I have not seen any data suggesting that when this
treatment approach is used, people have quickly failed the next regimen.
Did we really need to use protease inhibitors in many of these
patients, my own included? Or were we in too much of a rush to do it,
without thinking of some of the long-term implications?
ATN: How reversible is the immune damage caused by HIV?
Dr. Henry: One of the strong arguments for starting maximum
treatment immediately is that the immune damage is irreversible. I am not
sure about that. Most of the clinical data on the high-potency regimens
suggest that even people with fairly advanced disease have improvements in
their immune function to the extent that they are quite protected from most
infections -- with no major holes which would allow particular infections
break through. Even in people with advanced disease -- the worst case -- it
appears that there is considerable protection. And for those with CD4
counts over 150, it seems that immune damage is quite reversible. Some of
the work we have done with Ashley Haase and others, in looking at levels of
T-cells in tissues vs. the blood(1), found people whose lymph nodes were shot, with germinal
centers you could hardly detect, yet who recovered with therapy. So it is
an open debate about reversibility. We have to learn much more about this.
Most people agree there are better, simpler drugs on the horizon. There
needs to be more critical discussion of what is the best time to use a
complex, highly active regimen.
ATN: Are your more advanced patients still doing well, despite
the virologic failure? Are the gains in reduced death, hospitalization, and
opportunistic infections still there?
Dr. Henry: Absolutely. I am taking care of about 300 patients,
and we have projected for 1997 that we may have about eight patients who
will have died this year. But that is down from 40 deaths per year two to
three years ago, about an 80% reduction. For people who came into 1997 and
did not have a near-terminal condition, and were on antiretroviral therapy,
I have only had one new AIDS opportunistic infection that we have diagnosed
in 1997 to date. This amazing improvement leaves me puzzled about how much
of the prophylaxis to continue.
The people I am now seeing in the hospital are mostly new patients.
They have not been treated, they come in with pneumocystis just like the
old days.
I consider it excellent advice that it is rarely an emergency to get
people started on antiretroviral therapy. Some leading physicians almost
never start patients on treatment until it is their fourth or fifth visit.
That is probably wise; you get to know the person, and hopefully all of
their questions have been answered. Things were urgent when we were under
the gun of all those deaths and infections. But now that these consequences
have slowed down, we should start to be more thoughtful about when to apply
different strategies.
Ultimately we will also need to be more cost effective. If a patient is
looking at possibly 50 years of treatment, and each year costs $10,000,
that would be half a million dollars. These resources are valuable. Of
course we do not want anybody to get sick when it could be prevented, but
the clinical community has an obligation to use resources as carefully as
possible. Today I am unsure about what is wise and what is not.
Protease Inhibitor Resistance
Dr. Henry: Some people believe that once you are resistant to
one protease inhibitor, you will be resistant to all of them, that there is
a common pathway for resistance. If that is true, the decision to use a
protease inhibitor is a momentous one; and you have to do all you can to
use it at the right time, with the right other drugs. I am not sure that we
now know how to do that for many people.
On the other hand, some people say you can use one protease inhibitor,
and if you get a viral breakthrough, you can switch to a different protease
inhibitor. But there is not much data, except that if saquinavir is the
initial drug, following that up with indinavir generally does not work very
well. Also, there is a lot of cross resistance between ritonavir and
indinavir. And for most people it does not look like nelfinavir can rescue
people who are highly resistant to other protease inhibitors.
One of the key issues now is whether there is any difference starting
with nelfinavir instead of indinavir. There is very little data on that
issue. We are very happy to have both drugs available. Both now use three
times a day dosing, which is somewhat inconvenient to maintain
indefinitely. Both have fairly good data on durable response in ideal
situations. Indinavir has impressive clinical data from ACTG 320, and the
one and two year data that Roy Gulick presented from their pivotal clinical
trial.
But if indinavir fails to keep the virus controlled, how well can you
suppress that virus with other treatments? The data from ICAAC overall was
worrisome, indicating that this is very difficult to do.
A presentation from St. Vincent's, with a small number of patients, did
not show much success in suppressing virus after treatment with nelfinavir
had failed. [Abstract LB-5]
In Toronto we presented our data from patients who had been on
Agouron-sponsored studies utilizing nelfinavir. We reported, particularly
for patients who did not have advanced disease, that even though their
virus had broken through to a high level, usually we were able to suppress
virus initially with a ritonavir-saquinavir based regimen, and at least
some patients look like they are going to have a durable response.
Research Needed
ATN: What are some of the studies needed now?
Dr. Henry: There are appallingly few studies on how to treat
people who have developed protease inhibitor resistance. After much hard
work the AIDS Clinical Trials Group has a new study, called ACTG 359, for
volunteers who have had indinavir failure; this drug is the market leader,
and yet this is the first such study looking in an organized manner at
people for whom this drug has failed to control the virus.
The data from the Toronto conference (ICAAC) on protease inhibitor
failures was largely anecdotal; it is painfully clear that this area has
been overlooked for several years now. There has not been enough support
from either government or the companies to develop protocols for people
whose protease inhibitors have failed, so there is little knowledge on what
to recommend for them.
The protease inhibitors are good drugs. But I am not sure that we can
tell the average clinician and patient that now is the time to go for
broke. We need more studies to look at how to optimize adherence, and we
need plans in place for when the recommended therapies fail.
Clinical research in the next several years needs to focus on how we
can do better. We need simpler and less expensive medications -- once or
twice a day dosing, and a much better safety margin for nonadherence, or
for cases of resistance to perhaps one of the drugs. Today that margin is
often thin; sometimes you really have just one key drug that you are
relying on, a single protease inhibitor or perhaps a non- nucleoside. If
something goes wrong with adherence, or if the pharmacokinetics are not
ideal, the regimen will likely fail due to viral resistance.
Also, it is hard to know what our absolute target should be for HIV
RNA. We have been working with a lab running the Roche Ultrasensitive assay
for six months on our patients who are below the limit of the standard
Chiron assay. In our clinic population, about 55% of patients at any given
time are below the limit of the Chiron test (500 copies). Roughly 25% of
our patients are below 50 copies, and maybe 15% are below 20 copies.
Yet our patients are doing very well clinically. So the question is,
for all those people who are not at that absolute lower limit, do I need to
pour it on? Or should I try basically not to cause damage (with the drugs)?
We need studies to address this critical group, people who are doing well
clinically, but maybe they are not below a viral cutoff of 20 copies. We
would be able to enroll many patients in such studies. Now I do not know
the best way to treat them. There are so many unanswered questions. I wish
more good studies were available.
References
1. Haase AT, Henry K, Zupancic M and others,
"Quantitative Image Analysis of HIV-1 Infection in Lymphoid
Tissue," Science, November 8, 1996, pages 985-989.
| 
