New Nation-Wide Trial Tests Timing of Switch from AZT

A new clinical trial -- the first conducted jointly by the Federally-
sponsored AIDS Clinical Trials Group and military institutions -- will
study whether an advanced blood test can show the best time to
switch from AZT alone to other treatments. This trial is seeking 300
volunteers, 150 from civilian and 150 from military sites, and will
last for up to four years. This is the first time an AIDS clinical trial
has used a virological test to change the treatment regimen at the
first sign of viral resistance to the original treatment.

All volunteers will begin this trial taking only AZT as the active drug.
They will also take placebos for ddI, and for nevirapine, an
experimental drug proposed for the "convergent combination"
approach to triple-drug therapy.

As the trial proceeds, their blood will be tested for the presence of a
particular mutation of HIV, at "codon 215" (position 215 in the viral
gene which codes for reverse transcriptase, the enzyme which is
targeted by AZT, and also by ddI and nevirapine). At least five
different mutations are known to cause AZT resistance; however, the
one at position 215 is the most serious.

When the mutation is detected (which can be at any time in the
course of the trial, or never), the patient will be secretly assigned to
one of three treatment groups: continuation of AZT, switch to AZT
plus ddI, or switch to the triple combination of AZT plus ddI plus
nevirapine. Neither the patient nor the research nurses, physicians,
or study administrators will know if or when the switch occurs.
Instead, one or more of the placebos will be changed to active drug,
without their knowledge.

This will make the pill-taking somewhat complex. All patients,
throughout the trial, will take AZT every eight hours, ddI placebo (or
real ddI) every 12 hours, and nevirapine placebo (or real nevirapine)
once per day. And when they take the nevirapine, they will always
take one pill from each of two different bottles. The reason for this is
the protocol calls for starting nevirapine at half dose for the first
eight weeks it is administered; the half dose is given by changing
only one of the bottles from placebo to real drug. Eight weeks later,
the other bottle also will be switched from placebo to drug. Once the
mutation has been detected and randomization has occurred, the
assigned treatment will be maintained throughout the trial, unless
side effects or other problems force it to be discontinued.

Note that this trial is designed to look for changes in blood
measurements, not the development of clinical disease. It's main
objective is "to validate that (the '215' mutation) precedes the
increase in viral burden ... and decline in CD4 count which been
observed in association with clinical failure" on AZT, and to see if
adding the other drugs will prevent the increase in viral burden and
decline in T-helper count. (The importance of the '215' mutation has
been suggested by a small retrospective study of stored specimens.
Seventeen patients with the mutation had a mean 50 percent
decrease in T-helper count, during treatment of approximately three
years; the other 21 patients had a mean increase of 11 percent. See
Kozal MJ, Shafer RW, Winters MA, Katzenstein DA, and Merigan TC,
Mutation in HIV Reverse Transcriptase and Decline in CD4
Lymphocyte Numbers in Long Term Zidovudine Recipients, Journal of
Infectious Diseases 1993; pages 526-532.)

The protocol chair for this study is Thomas C. Merigan, M.D., Stanford
University Center for AIDS Research. The co-chair is Douglas L.
Mayers, M.D., Walter Reed Army Institute of Research and Naval
Medical Research Institute.

Switch to Non-Blinded Therapies

What about those who do poorly on the study, who may be randomly
assigned to continue AZT after the resistance mutation has
developed? The protocol calls for taking the volunteer off the study
medication if (1) dose-limiting toxicity has developed, (2) an AIDS-
defining condition develops (using the 1977 definition, as amended
for this study), or (3) the patient's T-helper count has declined by
more than 50 percent, as shown by two consecutive measurements
at least 72 hours apart. If any of these occur, the patient's physician
can begin any other treatment; the patient may continue to be
enrolled in the trial for followup. However, the study will not pay for
AZT or ddI once the patient is off the study medication; whether
nevirapine (which is not commercially available) can be provided is
being negotiated as this article goes to press. Also, the blind will not
be broken until the study is finished and the data analyzed, so the
volunteer who is taken off study medication will not be told whether
it was AZT alone or something else that did not work for them.

Eligibility, How to Enroll

To be eligible for the study, volunteers must be asymptomatic, with
T-helper count between 300 and 600. They must have had AZT for at
least one month, uninterrupted, before being screened for the study,
but must not have had more than two years of AZT. They must be at
least 13 years old (at least 18 for the military sites). They must not
have a history of pancreatitis, and must not currently be alcohol or
drug abusers. Women are encouraged to enroll in this trial, but they
will be excluded if they are pregnant, or intend to breast feed during
the study. During the screening, they will be tested for the '215'
mutation, and be excluded if it has already developed. There are a
number of other medical and laboratory entry criteria.

[One possible advantage of being screened for the study is finding
out if one has the '215' mutation already; about half of those with T-
helper count between 300 and 600 who have been tested do have it,
and therefore they are not eligible to enroll in the study. They might
want to start treatment with something other than AZT, for example,
with ddI; however, there is no professional consensus yet about the
clinical meaning of this mutation, which is why the study is being
run. Unfortunately for some volunteers, the T-helper count is tested
first; both samples are drawn on the same day, to avoid an extra
visit, but if the T-helper test shows that the volunteer is ineligible,
the virology test for the '215' mutation is not run. As far as we know,
this test is not commercially available; it could become available,
however, well before the trial ends.]

The civilian trial sites are located in: Baltimore, Cleveland, Galveston,
Los Angeles, Minneapolis, New York City, Philadelphia, Stanford (Palo
Alto), Rochester, and Washington D.C. For the contact number to call
for information at a site near you, call the AIDS Clinical Trials
Information Service, 800/TRIALS-A. Ask for information about
study ACTG 244.

The military study slots are open to those who have an ID card which
shows that they are eligible to receive care at a Department of
Defense medical treatment facility. They can be a veteran, on active
duty, or a dependent (age 18 or older). The DoD will fly them to a
participating center, which will usually be either Wilfert Hall USAF
Medical Center, San Antonio, Texas, or Walter Reed Army Medical
Center, Washington, D.C., or National Naval Medical Center, Bethesda,
Maryland. (Those not on active duty can choose whether to enroll at
a military or a civilian site.) To enroll at a military site, call Douglas L.
Mayers, M.D., at Walter Reed Army Institute of Research, 301/217-
9410, or Kenneth Wagner, D.O., at National Naval Medical Center, HIV
Research Unit, 301/897-3290; they will refer you to a protocol nurse.

Comment

This study is expected to take four to six years to affect the standard
of care -- two to four years of treatment time (depending on how
many of the volunteers develop the '215' mutation), and additional
time for recruiting, getting all centers started, data analysis, and
publication, dissemination, and use of the results. And then the
results will apply directly only to AZT (and probably indirectly to
other drugs in the same class, such as ddI -- although a different test
would be used, as different mutations cause HIV to become resistant
to ddI).

Our concern is not with this trial itself, but with the broader
questions of research planning. Does a major study now to fine-tune
the use of AZT several years down the road reflect a successful
strategy for dealing with the epidemic? In treating bacterial
infection, it has long been common to test for resistance, in order to
avoid using antibiotics which will not work for a particular patient.
HIV is different from bacteria, but a case could be made for using the
'215' test now, at physicians' discretion, to switch from AZT to ddI or
some other drug as soon as AZT resistance is indicated. Meanwhile,
use clinical trials to test truly important leads, such as topotecan or
conocurvone, which have been neglected because of the
happenstance of politics.

The fundamental problem is that for over ten years, a research effort
which could not possibly save most people's lives has been
acceptable to the general consensus -- almost casually accepted as
scientifically inevitable. Little serious effort has been made to think
through the consequences and change direction. We wonder if this
basic root of the failure of AIDS research can be changed today.