Triple-Drug Therapy Still Worth Testing, Despite Laboratory Error

The treatment approach called "convergent combination
therapy" is the combined use of multiple drugs (usually three
or more) which all act against the same target in a virus.
This differs from the usual approach to combining drugs for
treating infection, which is to target different stages in
the life cycle of the bacterium, virus, or other infecting
organism. The theory of convergent combination treatment is
that viruses, which are much simpler than bacteria, may only
have a limited range of possible changes in a key enzyme,
before they become unable to reproduce. Therefore, it might
be possible to stop the virus by a combination of drugs, each
of which requires the virus to have one or more different
mutations in the same enzyme. Either the virus has all the
mutations and is not viable, or it does not have them all and
is susceptible to at least one of the drugs.

Convergent combination therapy first received widespread
publicity in February 1993, after the publication of a major
article in Nature (February 18) describing laboratory tests
of the concept. (For background, see AIDS TREATMENT NEWS
#170, March 5, 1993.) More recently a laboratory error has
been found in the work reported in the February paper. The
error also received widespread publicity, with a page-one
article in The New York Times on July 22.

We believe that both the original work and the error have
often been overinterpreted by the public, which
underestimates the degree to which laboratory results are
only hints or guides to treatment development, and not strong
evidence of what will happen in people. Convergent
combination therapy never promised more than a limited
advance; yet even after the discovery of the highly-
publicized error, it is still well worth testing.

The Nature paper reported two different experiments, which
people have sometimes confused:

* The Harvard team artificially constructed an HIV provirus
(the virus in its DNA form) containing mutations giving
resistance to AZT, and to ddI, and also to nevirapine.
(Nevirapine, also called BI-RG-587, is an anti-HIV drug being
developed by Boehringer Ingelheim Pharmaceuticals, Inc.; it
inhibits reverse transcriptase, an enzyme which is required
for HIV to reproduce, but which is not found in uninfected
humans. AZT and ddI also inhibit reverse transcriptase.
Resistance can develop to each of these drugs, through
mutations in the reverse transcriptase, but different
mutations are required for each drug.) In separate tests, the
researchers also combined AZT, ddI, and pyridinone (also
called L-661, a drug being developed by Merck & Co., Inc.).

The Nature paper reported that one such engineered virus,
labeled "mutant 4," was dead, and hypothesized that the
accumulation of mutations in the reverse transcriptase had
weakened the virus to the extent that it was no longer
functional. The authors suggested that since one combination
of mutations led to a dead virus, it meant that one
mutational pathway to the development of multidrug resistance
to AZT, ddI, and nevirapine might be blocked.

* In a different experiment, the Harvard team showed that by
using a combination of AZT, ddI, and nevirapine (or
pyridinone), at concentrations achievable in patients' blood,
viral spread in cell cultures was effectively prevented.

Perhaps because of the way the two different experiments were
reported in the same paper, many people have misinterpreted
the goal of the treatment as using the drug combination to
force HIV to mutate into a dead virus. In fact, the
construction of mutated viruses in the laboratory only
suggested that not every combination of resistance mutations
to the different drugs automatically gave rise to multidrug
resistance. From that, the researchers theorized that
multidrug resistance might be delayed or made more difficult
by use of multiple drugs against the same viral target
(reverse transcriptase in this case).

The Error

At the Noordwijk Drug Resistance Workshop in the Netherlands,
and at the IX International Conference on AIDS in Berlin,
both in June of 1993, researchers from other laboratories
showed that they could not reproduce the results of a key
experiment performed by the Harvard group. The discrepancy
between the original report, and the independent findings of
two teams, one led by Dr. Brendan Larder at Wellcome Research
Laboratories in the UK, and the other led by Dr. Emilio Emini
at Merck Research Laboratories in the U.S., centers on the
artificially constructed HIV mutants.

In an effort to confirm the original report by the Harvard
team, Drs. Larder and Emini constructed HIV mutants with the
intended four mutations (at positions 74, 103, 215, and 219
in the genetic code for the reverse-transcriptase enzyme).
The result was a virus that could reproduce. Immediately
following these reports, the Harvard team went back to their
laboratory and discovered that one of their artificially
constructed viruses, "mutant 4," had at least one unintended
mutation in addition to the intended ones. Additional
experiments should have been performed to confirm that all
HIV mutants were built as specified and did not have any
unwanted mutations. All other data in the original Nature
paper have been reviewed and were found to be correct.

According to the Nature paper, HIV did not develop resistance
to the triple-drug combination after passing the virus 10
generations in the presence of the drugs in cell cultures.
But, using recent data obtained from longer-term cell culture
experiments, the Harvard team reported at the Berlin
conference that they found multidrug resistance after 20 to
40 generations under different conditions. These new findings
are in agreement with those from Dr. Larder's group. They
suggest that HIV which is resistant to AZT, ddI, and
nevirapine may emerge in treated patients, especially those
who started this triple-drug regimen with viruses already
resistant to AZT and ddI.

Clinical Trial Continues

A large clinical trial sponsored by the National Institute of
Allergy and Infectious Diseases' AIDS Clinical Trials Group
(ACTG), designed to evaluate the effectiveness of convergent
combination therapy with 400 patients, is currently underway.
Despite the announcement of the experimental error, the
clinical trial will proceed as originally planned. All study
participants will be contacted and will have an opportunity
to discuss the information with their investigators. In
addition, a "Letter to Patients" has been sent out to all
participants in the study.

These recent findings have demonstrated to the research
community that the HIV reverse-transcriptase enzyme has more
adaptability and ability to evade antiviral treatments than
previously believed. And caution must always be used in
trying to apply laboratory findings to clinical treatment.
Nevertheless, the current trial should determine whether
triple-drug combination is a more effective anti-HIV strategy
than the more standard combination of only AZT and ddI. In
addition, the study should provide a better understanding of
the overall biological activity of nevirapine in combination
with AZT and ddI, and show whether the three-drug combination
will delay the development of resistance in patients.

About the author: Henry E. Chang is an AIDS treatment
activist and the director of research and development at
Shared Medical Research Foundation in Los Angeles,
California. He can be reached at 818/345-2172.