       GMHC Treatment Issues, Volume 9, Number 10 - October 1995
       ---------------------------------------------------------

       Roaches of Inner Space
       by Dave Gilden

       The Interscience Conference on Antimicrobial Agents and
Chemotherapy (ICAAC) is an annual affair that brings together
thousands of infectious disease experts to ponder the progress
modern medicine has made in their area of specialty.

       This year's conference, which was held last month in San
Francisco, counted the disappointments as much as the victories in
the battle against pathogens. The theme was new and re-emergent
disease.

       AIDS is, of course, the archetypal new disease. It received
plenty of attention, especially given the lack this year of an
International Conference on AIDS (now held only biannually). In
addition, people with AIDS are both the source and target of
drug-resistant microbes. This double threat arises because their
immune systems do not back up the therapeutic effect of drugs the
way competent immune systems do. Drug-resistant germs have a greater
opportunity to evolve in such an environment, which is precisely the
environment in which the need for potent drug therapy is greatest.

       In the following pages, we describe research presentations at
ICAAC related to HIV and AIDS. The virus has turned out to be the
most modern of bugs -- like the cockroach, HIV's high reproduction
rate gives it a supreme capacity for mutating to evade the toxic
chemicals attacking it. But, there has been at least incremental
progress made in formulating strategies to circumvent this issue.

       A More Prolific Virus...

       At ICAAC, David Ho, M.D., director of the Aaron Diamond AIDS
Research Center in New York, presented a continuation of his group's
observations on the rapid pace of the HIV infectious cycle. (See
these researchers' earlier observations in Nature, January 12, 1995,
pages 123-6.) These results are based on following the blood levels
of HIV and infected cells after patients begin receiving the
powerful protease inhibitor ritonavir.

       According to Dr. Ho's current estimates, at least ten billion
HIV particles, on average, are produced each day in people infected
with HIV. The lifespan of those free virions averages about eight
hours, and the cells they infect live for an average of 2.2 days
after infection. Dr. Ho calculated that the time between the
creation of a new virus particle and the point that particle starts
producing its own progeny within the cell it infects is about 1.2
days, making for about 300 replication cycles per year. These
calculations show the high frequency with which we can expect
mutations in the virus, which apparently undergoes the equivalent of
thousands of generations of human evolution in the course of a
single person's disease.

       Dr. Ho claims that any single mutational change in any of the
HIV genes probably occurs multiple times per day in a given human
being, and forms of a single HIV gene with multiple mutations
probably occur as well in the course of a day. Mutations alter the
amino acid sequence and shape of an HIV enzyme or protein, and such
alterations nullify the action of antiviral drugs or immune defenses
that bind to and neutralize specific amino acid chains.

       Steven Wolinsky, M.D., of Northwestern University, who gave
one of the lectures at the ICAAC opening session on AIDS, pointed
out that the ability to rapidly evolve may be the central reason why
HIV ultimately defeats the immune system. The immune response to HIV
depends on detecting specific portions of HIV's structural protein
(epitopes) either on the virus itself or on the surface of
HIV-infected cells.

       Cytotoxic lymphocytes (CTLs) kill cells recognized as
infected, and antibodies bind to their target regions on free virus
particles and neutralize the virus. This activity creates selective
pressure in favor of mutant HIV with proteins not targeted by
existing CTLs and antibodies. The immune system may at first be able
to engender new CTLs and antibodies to catch the escaping HIV, but
eventually HIV comes up with so many diverse strains that an
individual's immune system cannot take care of them all (the
"diversity threshold" is exceeded). (Much of this material was
described by Martin Nowak and Andrew McMichael in the June, 1995
Scientific American.)

       The HIV evolutionary system that works to evade the natural
hazards posed by the human immune system also functions efficiently
to escape the perils posed by antiviral drugs that bind to the
reverse transcriptase and protease enzymes. In the lecture preceding
Dr. Wolinsky, John Coffin of Tufts University in Boston examined the
rapid rate of appearance of resistance to our most powerful
antiviral drugs.

       The rapid appearance of resistance is testimony to the vast
diversity of virus variants always present in the body, according to
Dr. Coffin. He thinks mutant viruses with at least modest resistance
to a given drug exist even before a person starts taking that
medication. These mutants may not be detectable because they do not
multiply quite as fast as the dominant "wild type" strain, but they
rapidly takeover when HIV is exposed to a particular drug.

       Dr. Coffin cited the example of 3TC. A mildly resistant
mutant strain is indeed detectable before therapy, and then a more
highly resistant single mutation shows up a few weeks after 3TC
administration begins. "There is no way to hit the virus hard enough
to prevent mutations," he concluded.

       A final problem is that a few infected cells are chronically
infected and do not naturally succumb at the rate Dr. Ho predicted.
(These latently infected and/or long-lived, slow producers of HIV,
probably macrophages and quiescent CD4 cells, are often termed
"virus reservoirs.") Since present drugs only block HIV replication
and cannot eliminate infected cells, there persists a small amount
of virus production (0.1 to one percent of normal production even
for the most powerful protease inhibitors). Virus reservoirs
inescapably limit how hard any of the present medications can "hit"
HIV over the short term, and mutant escape versions of HIV continue
to have the opportunity to arise after drug therapy commences.

       Dr. Ho calculates that the solution to the problems HIV
variation poses is not necessarily powerful therapy, but more varied
therapy. "We need combinations of drugs that force the virus to
mutate at multiple points, preferably more than four, before the
numbers favor the host," he said.

       Previously, combination therapies have been promoted because
of the drugs' additive or synergistic effects in reducing HIV
replication. Dr. Ho is speaking about combination therapy from a
different perspective, inhibiting the evolution of unwanted
drug-resistant HIV strains by both reducing HIV replication and
requiring an untenable number of mutations.

       And a Proliferation of Resistance Factors

       More and more researchers are accepting Dr. Ho's contention
that the ultimate failure afflicting all current anti-HIV therapies
is strictly associated with genetic mutations conferring drug
resistance. And combination therapies seem a logical way to defeat
HIV's talent for these mutations. A satellite conference held after
ICAAC explored these two assumptions further.

       The first lecturer, Daniel Kuritzkes, M.D., of the University
of Colorado warned that all is not so straightforward. Citing two
government-sponsored studies, ACTG 116B/117 and ACTG 192, Dr.
Kuritzkes remarked that switching to ddI after AZT use apparently
confers benefits regardless of whether a patient's HIV is still
susceptible to AZT. More remarkably, the people with AZT-resistant
virus do worse after switching than those with susceptible HIV. This
is counter-intuitive: one would think that ddI would have the
biggest impact in people with AZT-resistant HIV. They are the ones
receiving the most minimal benefit from their present drug, and
there is no evidence from laboratory cultures that AZT-resistance
alone causes any decrease in susceptibility to ddI.

       Fears have been raised for the past several years that some
mutation in AZT-resistant HIV causes an increase in the virus'
virulence. A similar, though less serious phenomenon is present
during d4T and ddC therapy: CD4 count falls and disease progression
do not seem related to the emergence of resistant strains. This is
true, too, for saquinavir, Roche's protease inhibitor. According to
the ICAAC report on using high dose saquinavir (ICAAC presentation
LB-5, also see Treatment Issues, September, 1995, pages 6-7), HIV
plasma levels bottom out and begin to rebound four weeks after
starting the 7.2 gm/day dose, yet resistance mutations do not start
to appear before the twentieth week of therapy.

       One must remember all the factors that contribute to drug
failure. The first of these is that since the currently approved HIV
therapies are not really very good at stopping viral replication,
they do not halt the progressive immunologic decline leading to
AIDS. With less and less immune system backup, any drug will perform
less and less well (this circumstance was described for CMV
therapies in Treatment Issues July/August, 1995, page 1). Another
factor is cells' increased ability to avoid metabolizing nucleoside
analogs like AZT into their active, or phosphorylated, form (this
"cellular resistance" was described by Michael Dudley in ICAAC
presentation S110). Finally, the present drugs do not reach all the
cell types harboring HIV. All these elements abet the rise of drug
resistance by allowing further HIV replication and more mutations.
The final outcome of the genetic rearrangement that evolves, as the
virus attempts to optimize itself in the presence of drugs, is
uncertain.

       Although it is obviously important to look at the whole
environment in which HIV and drugs interact, an elegant report by
Charles Boucher, M.D., at the satellite resistance conference,
showed how well merely following the HIV genetic response to
chemical therapy can explain the observed results.

       Dr. Boucher and his associates looked at a group of trial
participants in Amsterdam who first received 3TC monotherapy, to
which AZT was later added. The unique, rapidly appearing mutant that
creates 3TC resistance is supposed to block the effects of the AZT
resistance mutations, but Dr. Boucher says that this is only true
for the initial alteration of the reverse transcriptase that causes
low-level AZT resistance. The succeeding mutations, which result in
high-level resistance (and possibly greater viral virulence), are
little affected.

       In the Amsterdam 3TC-resistant patients, high-level AZT
resistance occurred later than usual, in the second year of AZT/3TC
therapy, via several of the usual mutations plus a novel one.

       The ability of high resistance to AZT to coexist with 3TC
resistance suggests that it will not help so much to add 3TC to the
regimens of people with lengthy experience on AZT (in whom the
mature AZT-resistant genetic pattern is frequent). And indeed, the
long-term observed antiviral effect in clinical trials of adding 3TC
to prior AZT monotherapy was comparable to adding ddC. (See
Treatment Issues, February, 1995, pages 3-4.)

       Dr. Boucher nevertheless held out the hope that all these
mutations will render reverse transcriptase a less functional
enzyme, and the HIV producing it will replicate less well. The
extent to which this holds true, as well as the physical benefit
patients receive, will have to be determined in clinical trials. A
previous attempt to create "replication- deficient" HIV -- through
the so-called convergent combination therapy -- was not very
successful (see Treatment Issues, January, 1995, page 6).

       Beyond the Valley of the Drugs

       A combination therapy that includes AZT, ddC and Abbott
Laboratories' protease inhibitor ritonavir is performing
significantly better than any past regimen, with HIV levels
continuing to decrease in a handful of patients, so far, out beyond
at least five months (see page 6). An ongoing trial testing AZT plus
3TC and indinavir (the Merck protease inhibitor) may show that this
combination performs as well as the ritonavir-containing one.

       Abbott investigators are already talking about "draining the
reservoirs" of HIV. It is quite a leap to think that this type of
combination can completely clear the body of HIV, though. One
immediate problem is that protease inhibitors penetrate the brain
poorly if at all. Most likely, people with HIV will be required to
take an effective drug therapy absolutely every day for the rest of
their lives, regardless of side effects. If therapy is missed for
just a few days, HIV replication will start to return, with the
heightened possibility of a virus evolving that is both virulent and
drug-resistant. This scenario is similar to the one that exists for
tuberculosis, only worse.

       We still have a hard time controlling TB, and drug-resistant
tuberculosis is becoming a major problem. For ultimate HIV control,
then, maybe we will have to hit the virus not just "hard," but from
different directions. Hydroxyurea, as reported in our last edition,
and other compounds that make the cellular milieu less compatible to
HIV, offer a variety of new approaches. Compounds that break HIV's
"zinc fingers" are also a possibility here (see page 7).

       Immune-based therapies, which work by ameliorating the immune
response, could be the final answer since they would work on the
heart of the problem -- the lack of immune backup to chemotherapy.
Unfortunately, the initial attempts have not worked well. IL-2 (see
Treatment Issues, February, 1995, pages 7-11), for example, failed
to lower HIV levels over twelve months of treatment although, as in
the past, it did double CD4 counts (ICAAC presentation LB-8). No
clinical benefit from IL-2 has yet been demonstrated, either.

       Another approach has been the so-called therapeutic vaccine,
in which synthesized HIV proteins or killed HIV stripped of its
outer coat has been injected into people with HIV in the hope of
coaxing a broader immune defense that reacts to more critical HIV
epitopes. Like IL-2, these vaccines have had no confirmed benefit in
terms of HIV levels or patients' health.

       There were two early reports at ICAAC on the most
sophisticated therapeutic vaccine yet, Viagene's gene therapy
product which induces cells in the body to mimic HIV infection by
producing HIV proteins. The presenters claimed improvements in
immune response, but detected no significant change in HIV levels or
CD4 counts (presentations I75 and I76).

       An alternative way of modulating the immune response to HIV
would be to reduce aspects of that response that are considered
counterproductive. Use of thalidomide to inhibit the inflammatory
cytokine and HIV-stimulator tumor necrosis factor (TNF) has been
widely discussed (see Treatment Issues, May 1995, pages 3-6). At
ICAAC, one report described using a fusion protein including part of
cells' TNF receptor to "mop up" excess TNF. A pilot study of this
molecule again observed no effect on HIV levels or CD4 count
(presentation I78).

       Nevertheless, immune-based therapy seems like an important
avenue to pursue at this point. "It's clear that antiretroviral
therapy has been limited by a number of factors that are quite
straightforward It might make sense to refocus our thinking on the
virus-host interaction," contended Robert Schooley, M.D., of the
University of Colorado, during a review of immune-based therapies
(presentation S51).

       We need to identify the responses that keep HIV in check
early in disease and in long-term nonprogressors. A successful
immune-based therapy would then reinforce or recreate such responses
in the human body. As drug therapy keeps HIV levels low, a powerful
and properly cultivated immune response could seek out and eliminate
the last vestiges of infection wherever they may be.

       ******************
       ACTG 175 and Delta
       by Gabriel Torres, M.D.

       First-line treatment may need to be re-evaluated according to
the findings of two studies, ACTG 175, and Delta. ACTG 175 was a
National Institute of Allergy and Infectious Diseases (NIAID)
sponsored study, presented at ICAAC (abstract LB-01). The data from
Delta, a European/Australian co-operative trial, were reported in
September at the Fifth European Conference on the Clinical Aspects
and Treatment of HIV Infection in Copenhagen, Denmark, September
26-29, 1995.

       ACTG 175 compared monotherapy with AZT or ddI to combination
therapy with AZT/ddC and AZT/ddI, as well as benefits of immediate
vs. delayed combination therapy. The study population included both
antiretroviral-naive and -experienced people with HIV and CD4 cell
counts between 200 and 500. The trial involved a higher proportion
of women, African-Americans, Hispanics and injecting drug users than
in previous ACTG studies. The study's primary endpoints were the
times to a 50 percent decline in CD4 cell count or a clinical
condition, namely an AIDS-defining condition or death.

       Participants who reached either the CD4 or clinical endpoint
were switched in a blinded manner to receive combination therapy
with AZT/ddI or AZT/ddC if they had previously received AZT or ddI
monotherapy or to receive the alternate combination if previously on
combination therapy. This allowed for a comparison of immediate
versus delayed combination therapy.

       At the conclusion of the study, 69 percent of the endpoints
occurred in the antiretroviral experienced group, and 31 percent in
the naive group. Over two-thirds of the endpoints were declines in
CD4 cell counts. The rate of side effects in this study was 19
percent and did not vary significantly among the treatment arms
(although in the naive group there were more adverse events in the
AZT monotherapy arm). More patients discontinued AZT/ddC than other
regimens due to adverse effects, mostly peripheral neuropathy. The
study found no significant difference between immediate versus
delayed combination therapy. Other findings are summarized in the
table on page 2.

       Delta compared AZT to AZT/ddI and AZT/ddC in AZT-naive (Delta
I) or -experienced (Delta II) individuals with CD4 cell counts
between 50 and 350 (see table). The average CD4 cell count for
participants in Delta I was around 200, much lower than in the naive
population of ACTG 175. Delta I was also larger than the naive
subset in ACTG 175, and more convincing because of the greater
number of clinical endpoints counted. Delta II was comparable in
size to the experienced cohort in ACTG 175, but participants in this
study again had much lower CD4 cell counts than the AZT-experienced
cohort in ACTG 175 (189 versus 338). More patients stopped AZT/ddI
than AZT/ddC due to side effects, mostly nausea and vomiting.
Delta's findings are also summarized on page 2.

       The recommendations potentially resulting from these trials
include: 1) for antiretroviral naive patients, initiate therapy with
either AZT/ddC or AZT/ddI or (according to ACTG 175) ddI
monotherapy. This latter finding contradicts the results of a
previous ACTG trial (116A) which found AZT monotherapy superior to
ddI monotherapy in naive patients. It is consistent with the results
of ACTG 152 which showed superior survival for ddI or AZT/ddI over
AZT in children. 2) Again, based on ACTG 175, for AZT experienced
patients: switch to or add ddI rather than staying on AZT
monotherapy. Adding ddC to AZT in the experienced group did not
provide additional benefits. This is consistent with ACTG 155, a
trial that found no clinical benefit (in terms of progression of
disease or survival) to AZT/ddC therapy in people with previous long
AZT experience.

       Final guidelines utilizing the results from both studies are
expected to be forthcoming from a state-of-the-art panel which will
review the data in the near future. This must await the results of
the NuCombo trial. This study is being conducted by the Community
Programs for Clinical Research on AIDS (CPCRA) in the U.S. and
compares AZT to either AZT/ddC or AZT/ddI in advanced patients with
CD4 counts below 200. Results from this trial should be available by
the end of the year.

       Summary of Data from ACTG 175 and Delta Trials

       These trials compared AZT to AZT plus ddI or ddC and also (in
ACTG 175) to ddI alone. Trial participants either had prior
experience with AZT or were AZT-naive.

       The following conclusions from the trial are based on the
effect of each treatment on progression to AIDS or death. ACTG 175
AZT-naive: AZT/ddC was superior to AZT monotherapy. AZT-experienced:
AZT/ddI was superior to AZT monotherapy.

       (Overall: AZT/ddI and ddI monotherapy were superior to AZT
monotherapy.)

       Delta AZT-naive: either AZT/ddI or AZT/ddC was superior to
AZT monotherapy.

       AZT-experienced: all drug regimens were equal. (Overall:
AZT/ddI and AZT/ddC were superior to AZT monotherapy.)

       For more details and the implications of these trial's
findings, see pages 3 and 4.

       [ACTG Table omitted from electronic addition]

       *****************************
       Making Sense of 175 and Delta
       by Theo Smart

       Some activists and researchers have heralded the findings of
ACTG 175 and Delta as "the most important trial results
ever." Much time already has been spent pouring over the
studies' statistics, comparing populations, and drop-out
rates in a valiant attempt to understand what these studies
mean. What it all boils down to is that AZT monotherapy is
not the best treatment available, whether as initial therapy
or in the drug-experienced. This finding does not impress
some. "Treating clinicians have known this for years," said
Stacy Kreiswirth, a nurse practitioner from New York
attending ICAAC, "show me something that will help my
patients."

       It is hard indeed to get excited about ddI and ddC, drugs
that have been around for so long and that clearly do so
little, as evidenced by Delta II, in which more than a
quarter of the participants had died by the end of the two-
year study, regardless of the treatment received. These
studies' findings do little to expand people's treatment
options.

       Some of these studies' findings are counter-intuitive. Why
didn't AZT/ddI fare better in delaying disease progression or
death than ddI alone? In the experienced population this may
have been due to AZT resistance, but in the naive population,
one would expect to have seen at least a trend in favor of
the combinations AZT/ddI and AZT/ddC. Both did have a more
pronounced effect on viral load than ddI in these patients,
as reported by Daniel Kuritzkes, M.D. (ICAAC abstract LB-02),
but in the experienced patients AZT/ddC performed
substantially worse clinically than ddI. ACTG 175 appears to
call into question the clinical utility of viral load tests.

       Delta II, which is still continuing, found no difference
between combination and monotherapy in antiretroviral experienced
people, and so contradicts the findings of ACTG 175. The
participants in Delta II were at a much later stage of disease,
though, so comparisons are difficult. There also could have been
more resistance to AZT, which could decrease the benefit received
from treatment with ddI or ddC (see page 14).

       The findings of these studies may become quickly irrelevant
because of new and more potent drugs being added to the
therapeutic arsenal. Where will d4T or 3TC fit into the
picture? Certainly, AZT/3TC is a much more potent
combination, although we have little clinical information.
Glaxo-Welcome has requested approval for AZT/3TC as first-
line treatment. Meanwhile, test tube data suggests that virus
resistant to 3TC is less susceptible to ddI and ddC. How
should this effect treatment decisions? Also, clinical data
on the protease inhibitors will soon become available.
Treatment options are increasing, but when to employ which
regimen is affected by such issues as cross-resistance,
synergy and toxicity. Without the clinical data, it is
impossible to know which variable is more important in making
treatment decisions.

       Finally, neither study provides information about the optimal
time to start therapy or when to switch regimens, and there
are no studies currently underway that will answer these
questions soon. ACTG 175 and Delta did yield some precious,
though somewhat contradictory data on clinical benefit. How
to apply the lessons learned beyond the particular
populations and regimens under study remains a mystery.

       *****************************************
       More Clinical Data on Protease Inhibitors
       by Theo Smart

       The four companies whose protease inhibitors are furthest
along in clinical development all presented clinical
information on their respective drugs at ICAAC. Overall, a
number of studies suggested that higher doses of these drugs,
and combination with nucleoside analogs appear to produce
more potent and durable antiviral effects than seen in
earlier protease inhibitor studies.

       Ritonavir: The combination of Abbott Laboratories' ritonavir,
AZT and ddC rapidly achieves a CD4 count jump of more than
110 cells, and a greater than 99 percent reduction in viral
load which is sustained for the duration of a twenty week
study. Still more strikingly, there is an increasing
proportion of participants whose HIV level becomes so low
that it is below or at the limit of detection by the PCR
assay used to detect viral load. Daniel Norbeck, M.D., of
Abbott presented this very preliminary data from a French
study, conducted by Jacques Liebowitz, M.D., in twenty-nine
volunteers (not all of whom have completed twenty weeks of
treatment). After two months, five of 27 participants had
undetectable HIV levels. Seven out of twenty had undetectable
HIV after five months, while four more experienced 99.9
percent of the maximum detectable reduction in viral load.

       Dr. Norbeck suggests that "the reservoirs of virus are
emptying," in the treated population. By reservoir, he means
that the longer-lived infected cells, such as macrophages and
monocytes, which continue to produce virus for weeks (unlike
an actively infected CD4 cell, which dies in two days). Many
researchers believe that since present drugs do not kill
infected cells, their use can only eliminate all of the virus
from the body if they can keep suppressing HIV replication,
even in the face of mutations that confer partial drug
resistance, for a period longer than these cells' lifetimes.
It is debatable whether Dr. Norbeck can claim such success on
the basis of one small study.

       Abbott appears even more excited about the promise of
combination therapy with ritonavir and Hoffmann-La Roche's
protease inhibitor saquinavir (abstract LB-7). Although
cross-resistance between protease inhibitors is a concern,
the initial resistance mutations arising with ritonavir and
saquinavir therapy are distinct, and the hope is that two-
protease combination therapy could suppress viral replication
enough to delay the emergence of resistance indefinitely.
Saquinavir's antiviral effect, however, is limited by the low
blood levels of saquinavir achieved with the current dose and
formulation. People are not getting enough saquinavir,
because the little that is absorbed is rapidly broken down in
the liver.

       Abbott may have a solution: ritonavir shuts off the metabolic
pathway in the liver that breaks down saquinavir as well as
ritonavir. In rats, co-administration of the two drugs led to
an eighteen-fold increase in peak levels of saquinavir in the
blood (from a single dosing) and extended the drug's half-
life fifteen-fold. (The cumulative increase in drug levels
from multiple dosing could be much higher.) Dr. Norbeck
concluded that ritonavir and saquinavir may have a "two-
dimensional synergy."

       Follow-up studies must be done immediately to see whether the
interaction occurs in people. Although no dose-limiting
toxicities have been encountered yet with saquinavir, no one
knows about higher blood levels, and an effect of this
magnitude may be extremely dangerous. As GMHC's David Barr
remarked, "people may want to trade in their livers after
using this combination," if this interaction is indeed seen
in people.

       It may not be. The antifungal medication ketoconazole also
slows down the same liver pathway. According to Noel Roberts
of Roche Laboratories in Great Britain, co-administration of
ketoconazole and saquinavir also dramatically boosted
saquinavir levels in rodents, but the increase was much less
impressive in humans. Still, just a three- or four-fold rise
in blood levels could be helpful, particularly in those
people receiving low doses of saquinavir in the current phase
III trials and expanded access program.

       [As Treatment Issues went to press, Miklos Salgo, M.D.,
Roche's director of clinical research notified us that Abbott
and Roche have reached an agreement to start drug interaction
studies in both dogs and people before the end of the year.]

       Indinavir sulfate (Crixivan): Higher doses of this Merck
protease inhibitor have a much more profound and sustained effect in
naive patients than lower doses, reported John Mellors of the
University of Pittsburgh (abstract I172). The findings come from a
study comparing indinavir to AZT, discussed previously in Treatment
Issues (January, 1995, pages 1-2). In this study, the dose of
indinavir was increased to 600 mg four times a day after Merck found
that HIV quickly becomes resistant to lower doses. Increasing the
dose had little effect in the patients who had initiated therapy at
lower doses, but researchers suspected that the CD4 and viral load
responses would be better in patients who begin therapy at a higher
dose.

       This proved to be the case in the sixteen trial participants
who had been randomized to AZT and were crossed over (at week
24) to the protease inhibitor at the 600 mg four times a day
dose. These patients began indinavir therapy with median CD4
cell counts of 110. After six months of treatment, the
majority of these patients had sustained reductions in viral
load, although there was a trend back toward baseline. At
week eight, half of the patients had more than a 1.98 log
(almost 99 percent) suppression in viral load. At week 24,
half maintained at least a 90 percent reduction in virus, and
a third still had 99 percent less virus than when they
started therapy. Median CD4 counts remained more than 100
cells above baseline.

       The findings were similar in a 24 week study comparing
indinavir sulfate (600 mg four times a day), AZT, and
indinavir and AZT combined (abstract LB-6). This study
enrolled 73 volunteers with CD4 cell counts below 500 (median
221) and viral load levels greater than 20,000 copies per ml,
(median of approximately 80,000 copies per ml).

       At the point of maximum antiviral effect, AZT monotherapy
produced around a 75 percent reduction in viral load (0.6
log), while indinavir monotherapy and combination therapy
reduced viral load by medians of 99.5 and 99.75 percent,
respectively (2.3 and 2.6 logs). At week 24, viral load was only
slightly (0.2 log) below baseline for AZT monotherapy. The viral
load was still significantly below baseline for indinavir
monotherapy and combination therapy, 1.5 log and 2.5 log,
respectively. Although the difference between the combination arm
and indinavir monotherapy arm was not statistically significant,
there was a clear trend toward a rebound in viral load on
monotherapy compared to a more sustained suppression of HIV on
combination therapy. CD4 cell analysis was not complete for this
study, yet there already is a clear benefit in favor of the arms
containing indinavir.

       In a meeting with activists several weeks before ICAAC, Dr.
Jeffrey Chodakewitz of Merck reported data from a dose-
ranging indinavir monotherapy study in more than 70 people.
For the 800 mg thrice daily dose (the dose now used in
Merck's trials), median viral load reductions were slightly
greater than 99 percent at maximum, with a slight trend back
to baseline by week twenty. CD4 counts increased by an
average of more than 100 cells.

       Activists asked Merck whether the company plans to run any
studies of indinavir in perinatal transmission. Merck has
expressed an interest in such studies, but the increase in
bilirubin levels that indinavir induces are a more serious
problem in infants than adults. Bilirubin can pass through
the immature blood/brain barrier and cause severe disability.

       AG1343 (VIRACEPT): Graham Moyle, M.D., reported on the
British low-dose study, (abstract LB-3) previously reported
in Treatment Issues (June, 1995, page 4). Six out of 22
patients had a greater than 90 percent reduction in viral
load at day 28. This response was sustained for the duration
of treatment (three to four months), and participants gained
an average of 100 CD4 cells. Diarrhea was the most common
side effect, occurring in eight out of ten of the people on
the highest dose tested.

       Marty Markowitz, M.D., of the Aaron Diamond AIDS Research
Center in New York, announced the results of a dose-ranging
study (that went higher than the British study) in people
with CD4 cell counts greater than or equal to 200, and viral
load levels greater than or equal to 20,000 copies per ml
(abstract LB-4). The study evaluated three doses -- 500 mg,
600 mg and 750 mg -- all twice daily in 30 subjects. At day
28, CD4 counts had increased by more than 100 cells; the
greatest response was seen in the patients on the highest
dose. Several HIV-related conditions such as lymphadenopathy,
hairy leukoplakia and candida resolved or improved in a
number of patients.

       The average maximum decrease in viral load ranged from 90 to
94 percent, and in some patients viral load dipped below the
level of detection. There was no difference in anti-HIV
effect by dose, although the viral load on the low dose did
appear to begin to rebound toward baseline levels. However,
Dr. Markowitz noted that individuals with higher drug levels
in their plasma had more profound responses -- but higher
doses did not always achieve higher blood levels of drug.

       Overall, 21 participants had a greater than one log reduction
needed to continue on VIRACEPT after the initial 28 day trial
period. This criterion for study continuation caused
consternation with AIDS activists on the West Coast, who
noted that some participants who were dropped had desired to
continue on therapy, largely because of improvements in CD4
cell counts. Dr. Markowitz countered that in patients who
were not seeing at least a one log reduction, long-term
therapy might induce cross resistance to other protease
inhibitors, so treatment discontinuation was in their best
interest. The activists felt that this decision should rest
with the individual and their primary care provider.

       Another area of concern to activists is VIRACEPT's shaky
position as fourth protease inhibitor in a race for approval,
where pivotal studies for the three competing drugs have already
begun. Hoffmann-La Roche has already filed for approval of
Saquinavir, and Merck and Abbott both plan to file for approval in
the first quarter of the coming year. Agouron's pivotal clinical
endpoint study in people with less than 50 CD4 cells is not slated
to begin until the end of the year. The competing drugs almost
certainly will be approved before this study is completed. It is
unlikely that people with less than 50 CD4 cells will be compliant
to the protocol when comparable drugs become available. Agouron's
best chance of salvaging its phase III program would be to
dramatically increase the size of this trial. This would decrease
the time it would take to reach a conclusion, since increasing the
number of participants will increase the number of endpoints that
can be counted. Otherwise, the study may not find statistically
significant evidence of the clinical benefit of the drug, and the
company may have to start all over again, in a world where the other
protease inhibitors are standard of care.

       **************************
       Protease Drug Interactions

       The following drug interactions have been reported for the
       protease inhibitors:

       Ritonavir (Abbott): Cannot be taken together safely with
codeine, Demerol, Darvon, Feldene, Nizoral, Monostat-4,
Tranxene, Valium, Versed (anesthesia), Halcion, Zanax,
Tambocor, Rythmol, Vascor, Hismanol, Seldane, Tagamet, Paxil,
Clozaril. Also: the dosage of clarithromycin should be
reduced when taken concurrently with ritonavir.

       Indinavir (Merck): elevates blood levels of rifabutin, so the
dose of rifabutin must be halved when administered with indinavir.

       **********************************************
       Update on Protease Inhibitor Compassionate Use

       Indinavir: Merck is increasing the size of its indinavir
compassionate use protocol to enroll 850 additional people
from 23 foreign countries. There has been much activist
criticism of the course of the U.S. program, though. Merck
had promised to get drug into 1,400 PWAs by the end of the
year, and the company is taking its time doing it, doling out
drug to 125 more lottery winners each week. Many people who
were chosen in the lottery assumed that they would receive
drug much sooner. Let us hope that they survive the wait!

       Saquinavir: Hoffmann-La Roche is increasing the size of the
saquinavir access program, as reported in last month's Treatment
Issues. The lottery will be held on November 3, 1995, and will
supply drug to 2,000 more people, for a total of 4,280 individuals
in the U.S. The program has enrolled 1,700 in the rest of the world.
Participants who did not win in the first drawing will be rolled
over into the second lottery.

       VIRACEPT: In a meeting with AIDS activists during ICAAC,
Agouron announced its intention to open a compassionate use
protocol subsequent to enrollment and interim analysis of its
pivotal phase III studies.
-- TS

       ******************************************
       Zinc Fingers -- The Next Antiviral Target?
       by Theo Smart

       The next target for antiviral drugs may be HIV's zinc
fingers, according to data presented by researchers from the
National Cancer Institute (ICAAC abstracts I134, I135, and
I136). Zinc fingers are chains of amino acids found in
cellular proteins which bind to DNA or messenger RNA, and
play important roles in a cell's life cycle. They are called
zinc fingers because they capture a zinc ion, which
contributes to the array's binding to RNA or DNA. Zinc
fingers found in nature can have one of a number of
structures. In HIV's zinc finger, four amino acids are
responsible for capturing the zinc ion. The sequence is
unique, found elsewhere only in other lentiviruses, the
family of retroviruses that includes HIV-1 and 2, cancer
causing HTLV, feline immunodeficiency virus, and murine
leukemia virus, to name a few.

       There are two zinc fingers in HIV's nucleocapsid (NC), a core
viral protein. The zinc fingers are involved in binding and
packaging viral RNA into new virions budding from an infected cell.
Experiments in which the zinc fingers have been deleted have shown
that new budding virions do not incorporate RNA, which instead
spills out of the infected cell. The NC protein and zinc fingers
also play some role during the process of reverse transcription. The
exact role of the protein is unclear, but it may anchor the RNA
molecule while the reverse transcriptase enzyme builds HIV DNA from
the RNA template. HIV that lacks zinc fingers is unable to infect
new cells. The zinc fingers are therefore essential for two phases
of the viral life cycle. Since both zinc fingers structures are
identical, the same compound could inhibit both (two targets for the
price of one).

       Researchers at the NCI have altered the structure of these
zinc fingers to see whether they could mutate and still work. They
cannot; changes in even one of these amino acids in the array make
the zinc fingers dysfunctional. In another study in which the entire
zinc finger array was replaced with the types of zinc fingers
commonly found in cellular proteins, RNA could be successfully
incorporated into budding virions, but the resulting virus was
non-infectious. HIV may not be able to mutate and escape the effects
of drugs targeting its zinc fingers.

       The team at the NCI has developed a laboratory test to
quickly screen for compounds that bind to HIV's zinc fingers.
In their first mass screening of substances in their chemical
library, they quickly found about 60 different active
substances that eject the zinc from the zinc finger, altering
its function. Many of these compounds were active against HIV
at doses that were not toxic in cell cultures. One of the
first, but least potent, of these compounds was disulfiram
(Antabuse), a drug that has been on the market for years as a
treatment for alcoholism. Unfortunately, this compound didn't
work when used in monkeys with SIV.

       Since HIV's zinc fingers are identical to the zinc fingers of
retroviruses which infect rodents, cats, and monkeys, animal models
may be useful in the preclinical evaluation of these drugs. Some
lead compounds have been evaluated in mice infected with murine
leukemia virus. In one study, untreated mice survived an average of
35 days, while mice treated with one of these compounds lived ten to
fifteen days longer (which is significant for laboratory mice).

       Preclinical data on two of the leading compounds were
presented at the ICAAC meeting in San Francisco. One class of
compounds is being developed by Parke-Davis Pharmaceuticals, and
another by Octamer Inc., a six-year old biotech company in Mill
Valley, California focused on AIDS and cancer research.

       The Parke-Davis compound may be further along. It is
currently being studied for activity, toxicity and
bioavailability in monkeys with SIV. Although shown to be
very safe in mice, the doses required to sustain adequate
blood levels were very high (250 mg per kg of body weight,
twice a day). If the doses are similar in people (and
absorption and metabolism may be much different in humans),
this could add up to more than thirty grams a day. Such high
doses could tax Parke-Davis' production facilities and may
have toxic effects in the body.

       It is not yet clear what dose will be necessary with the
Octamer drug, but if dosage is a problem, at least the
compound's structure is simple enough that it should be cheap
to produce, according to Octamer's president, Ron Brown,
Ph.D. The current lead compound, 4-Iodo-3-nitrobenzamide, is
a prodrug that becomes active in the cell, with very little
cellular toxicity. Octamer is currently studying its
bioavailability and efficacy in mice infected with
retroviruses. If the drug works well in the mouse model, the
company plans to approach the FDA to go directly into humans
studies.

       Other active compounds could be awaiting discovery in other
pharmaceutical companies' drug libraries. To ferret these
out, the NCI has opened a CRADA (co-operative research and
development agreement) offering to screen any company's drugs
for anti-zinc finger activity and to possibly help with the
development of such compounds.


       ******************************************************
       Opportunistic Infection Highlights from the 35th ICAAC
       by Gabriel Torres, M.D.

       Several important studies evaluating treatment and prevention
strategies for opportunistic infections were presented at
ICAAC. Especially noteworthy were a variety of presentations
on cytomegalovirus and mycobacterium avium complex, both of
which are undergoing rapid advances in treatment and
prophylaxis. The effect on HIV levels of opportunistic
infections also received considerable attention.

       Epidemiology

       A report from the Centers for Disease Control and Prevention
(CDC) examined HIV-related mortality data from 1987 to 92
(ICAAC abstract I22). The authors found a significant drop in
death related to pneumocystis pneumonia (PCP) (from 32.5
percent to 13.8 percent of annual HIV-associated deaths),
cryptococcal meningitis (from 7.7 percent to 5.0 percent) and
candidiasis (from 2.3 percent to 1.7 percent). The cause of
the decline in death rates are probably attributable to
prophylaxis with anti-PCP drugs and anti-fungal agents such
as fluconazole.

       In contrast, death due to other conditions increased.
Complications with a rising death rate included mycobacterium
avium complex (MAC) (from 6.7 percent to 12.2 percent of
annual HIV-related deaths), cytomegalovirus (CMV) (from 5.2
percent to 9.9 percent), bacterial sepsis (from 9.0 percent
to 11.5 percent) and non-Hodgkin's lymphoma (from 3.9 percent
to 5.7 percent). There were smaller increases in the
incidence of progressive multifocal leukoencephalopathy
(PML), bacterial pneumonia, cryptosporidiosis, or pneumonias
due to unspecified organisms.

       In 1992, unspecified pneumonia (18.6 percent of annual HIV-
related deaths) and Kaposi's sarcoma (KS) ranked with PCP as
the leading fatal opportunistic conditions.

       Valaciclovir for CMV Prophylaxis

       The final results of ACTG 204, an international study on
preventing CMV infection sponsored by both the National
Institute of Allergy and Infectious Diseases and the Glaxo-
Wellcome company, were presented by Judith Feinberg, M.D.,
now at the University of Cincinnati (abstract I214).

       This double-blind randomized trial compared the acyclovir
prodrug valaciclovir (Valtrex) -- which achieves acyclovir
blood levels thought sufficient to inhibit CMV -- to either
high dose acyclovir (800 mg four times per day) or low dose
acyclovir (400 mg twice daily) in HIV-positive individuals
with CD4 counts below 100 and evidence of prior CMV exposure
(positive CMV antibody). The valaciclovir dose was two grams
four times per day, the equivalent of about eight grams of
acyclovir four times per day.

       ACTG 204 enrolled 1,227 participants who were followed for
the development of CMV end organ disease (e.g. retinitis,
gastrointestinal disease). Two-thirds of the patients had CD4
counts below 50 on entry, most were gay males, and 80 percent
were concurrently receiving antiretroviral therapy.

       After an average of 33 weeks of treatment and 57 weeks of
follow-up, those receiving valaciclovir had developed fewer
confirmed CMV endpoints (11.7 percent) as compared to those
receiving either high or low dose acyclovir (17.5 percent).
This represents a 33 percent reduction of CMV disease, most
of which (79 percent) was CMV retinitis. The time to
confirmed CMV disease was significantly longer for those
receiving Valtrex.

       There was, however, a strong trend toward earlier death in
those receiving valaciclovir. The final mortality difference
was not statistically significant, but it did lead to early
study closure after an interim analysis. Causes of death were
similar in all three groups. More patients on valaciclovir
had to discontinue study medication due to toxicity. The most
common side effect associated with valaciclovir was
gastrointestinal distress, which occurred in 14.5 percent of
the participants.

       Dr. Feinberg concluded that Valtrex reduced CMV disease rates
by one-third. This reduction is similar to what has been
reported in the Syntex Laboratories study of oral
ganciclovir. (See Treatment Issues, July/August, 1995, page
4-5.) On the other hand, the Syntex study found a trend
favoring improved survival on oral ganciclovir. The cost was
a greater frequency of adverse events from oral ganciclovir
than were observed with valaciclovir in ACTG 204.

       Oral Ganciclovir for Preventing CMV

       Dr. Carol Brosgart from Berkeley, CA reported the results of
CPCRA Study 023, which compared oral ganciclovir (3 grams a
day) to placebo for primary prevention of CMV end organ
disease in AIDS patients with CD4 counts below 100 and prior
CMV exposure (abstract LB-10). This study was amended in
July, 1994 when the results of the Syntex study became
available, allowing patients in both arms the option of open-
label ganciclovir. A total of 994 patients were randomized in
a two to one ratio to oral ganciclovir or placebo and
followed for CMV disease endpoints.

       Patients did not have mandatory ophthalmologic evaluations on
entry or at routine intervals to screen for CMV retinitis but
rather were identified at the time of the onset of symptoms
such as deterioration of vision. The median follow-up was
fifteen months. Statisticians performed an intent-to-treat
analysis, which separated trial participants as if they had
stayed on their original medication -- oral ganciclovir or
placebo -- for the entire time.

       This analysis found that the rate of CMV disease was similar
in the oral ganciclovir and placebo groups, as were the death
rates. Patients receiving oral ganciclovir experienced more
adverse effects, including neutropenia (25 percent vs.
sixteen percent). Dr. Brosgart concluded that the results did
not support the use of oral ganciclovir for primary
prevention of CMV disease.

       The results of this study contradict those of the Syntex
study, which found an extraordinarily high rate of CMV
retinitis in its placebo arm, 39 percent, which was reduced
by half in the oral ganciclovir arm. The Syntex study relied
on scheduled retinal exams with fundoscopy for detection of
early CMV retinitis. Dr. Brosgart claims that some sixty
percent of the participants in the CPCRA trial also had
regular fundoscopic exams on their own. There nevertheless
remains considerable controversy as to whether the looser
requirements of the CPCRA trial meant that much early CMV
retinitis was missed both at study entry and during the
follow-up period.

       Syntex (now Roche Bioscience) is also criticizing the use of
open-label oral ganciclovir by members of the CPCRA placebo
arm, which could have obscured the observed benefits of
ganciclovir. How significant this contamination of the
placebo group with oral ganciclovir is in dispute. In the
CPCRA trial, exposure to oral ganciclovir averaged 9.3 months
in the group originally randomized to the drug and 2.1 months
in the initial placebo group. Roche Bioscience, meanwhile,
has filed an application with the Food and Drug
Administration to market oral ganciclovir for prevention of
CMV.

       While the arguments fly, clinicians must decide whether oral
ganciclovir prophylaxis is worthwhile in persons with HIV and
very low CD4 counts (below 50). Everyone agrees that oral
ganciclovir prophylaxis should be more precisely targeted at
those most at risk of CMV, if it is used at all. New assays
to assess CMV activity, such as PCR, quantitative cultures
and pp65 antigenemia, may help select the appropriate
population who could benefit the most from preemptive CMV
therapy.

       Resistant CMV

       In a preliminary analysis (abstract 135), researchers from
the Syntex oral ganciclovir prophylaxis trial found that the
incidence of ganciclovir-resistant CMV was very low (one
percent). One patient with resistant virus failed to respond
to intravenous ganciclovir treatment after developing active
CMV retinitis. Several posters, meanwhile, reported cases of
CMV infections in which viral isolates were resistant to
several well-known CMV drugs after treatment with only one of
the drugs.

       In one study (abstract H115) from the University of
California San Diego, CMV resistant to foscarnet was isolated
after only two months of foscarnet therapy in two patients,
whose CMV had previously been found resistant to ganciclovir.
In a follow-up study of nine patients with ganciclovir-
resistant strains, reduced sensitivity to foscarnet was
observed although the patients had not yet been treated with
foscarnet. A genetic analysis of these individuals' CMV found
mutations in the virus's gene for DNA polymerase, an enzyme
necessary for virus replication. These mutations, which cause
lowered sensitivity to foscarnet in laboratory cultures,
probably indicate that treatment with foscarnet will rapidly
fail.

       In another report (abstract H114), CMV strains isolated from
two patients after prolonged ganciclovir therapy were
resistant to both ganciclovir and cidofovir (HPMPC). The
second patient's strain also had greatly reduced sensitivity
to foscarnet. A genetic analysis of the first patient's CMV
found a mutation in the UL97 gene as well as in DNA
polymerase gene. The UL97 gene creates the CMV enzyme
(thymidine kinase) which converts ganciclovir into its
active, phosphorylated metabolite. The UL97 mutation renders
the thymidine kinase inactive and is well known for causing
resistance to ganciclovir but not the other drugs.

       On the bright side, herpes simplex virus with reduced
thymidine kinase activity was found resistant to acyclovir
but more sensitive to cidofovir (abstract H10). But
laboratory strains of CMV resistant to both ganciclovir and
cidofovir could be bred by investigators from Gilead
Pharmaceuticals, the makers of cidofovir (abstract H117). The
Gilead researchers could achieve this cross-resistance by
adding increasing concentrations of either ganciclovir or
cidofovir to a CMV lab culture. It was not necessary for the
CMV to have been exposed to both of the drugs at any point.

       Cidofovir (HPMPC) for CMV

       Cidofovir previously has shown promise in the treatment of
refractory CMV disease. At ICAAC, a late breaker presentation
(LB-9) described 60 patients with relapsed CMV retinitis who
were randomized to receive different regimens of cidofovir (5
mg/kg of body weight once weekly for two weeks, then 5 mg/kg
once every other week, or 3 mg/kg once every other week). To
minimize kidney toxicity, intravenous saline and probenecid
were co-administered.

       The median time to progression was 114 days for the 5 mg/kg
group and 49 days in the 3 mg/kg group, a difference that was
not deemed statistically significant. Side effects of
cidofovir included kidney dysfunction in 7 percent, and
probenecid reactions (fever, nausea and rashes) in 48 percent
of trial participants. Survival was similar in both groups
(39 percent vs. 36 percent). Cidofovir is now available
through a "Treatment IND" for patients with CMV retinitis who
have relapsed or cannot tolerate ganciclovir and foscarnet.

       Call 800/GILEAD5 to obtain the drug.

       Mycobacterium avium complex (MAC)

       A large French study called Curavium, comparing two oral drug
regimens for the treatment of disseminated MAC, was presented
at a late breaker session (abstract LB-19). The study
compared clarithromycin (two grams for two months, then one
gram thereafter) in combination with either clofazimine 200
mg daily for two months then 100 mg daily thereafter (two-
drug combination) or with rifabutin 450 mg daily and
ethambutol 1200 mg daily (three-drug combination).

       One hundred twenty-three patients with CD4 counts of greater
than 100 and positive blood cultures for MAC were enrolled
and randomized to one of the two regimens. They were followed
with serial blood cultures.

       The three-drug combination proved superior to the two-drug
combination with improved survival after 24 weeks.
Breakthrough MAC bacteremia was more common with two drugs
(eighteen cases) than with three drugs (six cases), and
emergence of clarithromycin resistance was less common using
the three drug regimen. Side effects included
gastrointestinal distress from clarithromycin and uveitis
from rifabutin.

       In a Canadian study (abstract LB-20), the three-drug
combination of clarithromycin, rifabutin, and ethambutol was
superior to a four-drug combination of rifampin, ethambutol,
ciprofloxacin and clofazimine.

       The rifabutin dose was halved (from 600 mg daily to 300 mg
daily) due to the development of uveitis at the high dose.
Rates of MAC clearance from the blood were 69 percent and 24
percent for the three- and four-drug combinations,
respectively. A significant survival difference (258 days
versus 153 days) also was seen in this study. Both studies
indicate that clarithromycin and rifabutin are the most
active drugs in the treatment of MAC and can be used
together.

       Opportunistic Infections and HIV Viremia

       A study from Henry Ford Hospital in Detroit examined HIV RNA
plasma levels in patients suffering from acute opportunistic
infections (abstract I236). Increases in HIV plasma levels
were found (median 36,000 copies/ml) in eight patients with
the advent of an opportunistic infection. With recovery from
disease, there was a decline in the HIV plasma level (median
change was -30,000 copies/ml). In contrast there were no
changes in p24 antigen levels or CD4 counts with either the
advent or recovery from opportunistic infections. The authors
concluded that opportunistic infections increase HIV
production and may exacerbate immune decline. These OIs also
may confuse the interpretation of disease progression and
drug failure based on changes in HIV load (see Treatment
Issues, September, 1995, pages 12-14.)

       A similar study showing increases in HIV levels following
herpes simplex reactivation in ten HIV-positive patients was
presented as a poster (abstract I235). The median increase in
HIV viral load was 55 percent, and the extent of viremia
seemed to be related to the size and number of herpes
lesions. It remains to be determined whether acyclovir
suppression of HSV can influence HIV viremia.

       Immunizations

       Very disturbing reports of the effects of influenza and
pneumococcal immunizations on HIV viral load received
considerable attention at the conference.

       One study (abstract LB-11) compared influenza vaccine to
placebo in 47 HIV-positive patients. This double-blind
randomized test found significant increases in plasma HIV
levels in the vaccinated group four to six weeks after the
injection. Three months after the vaccine, the CD4 percentage
had dropped by 1.8 percent in the vaccinated group as
compared to a rise of 0.2 percent in the placebo group. The
authors concluded that annual influenza vaccination for HIV-
positive persons may be harmful, although the study was
unable to demonstrate any clinically significant adverse
outcome to the vaccine.

       In another study (abstract I237), twelve asymptomatic HIV-
positive volunteers with an average CD4 count of 374 who
received the pneumococcal vaccine (Pneumovax) had their
plasma HIV levels assessments by quantitative PCR before and
after vaccination. Immunization with Pneumovax was followed
by a rapid and sometimes profound increase in viral burden
(mean rise of 83-fold). This rise correlated with the
volunteers' antibody production in response to the vaccine
and with the extent of their CD4 cell activation. HIV levels
in four patients remained increased twelve to sixteen weeks
after the vaccine.

       Although these studies are small, and neither showed adverse
consequences to the patients' health status, they indicate need for
further placebo-controlled studies of the vaccines and due caution
when administering them to patients with high viral loads.

       **************************
       Cryptosporidiosis at ICAAC

       Even the most rigorous of water purification systems could
not keep the intestinal parasite cryptosporidia out of the
tap water according to one poster presented by Susan
Goldstein of the Centers for Disease Control and Prevention
(CDC) (abstract K63). This study investigated the cause of a
1994 outbreak of 78 cases of cryptosporidiosis in Clark
County, Nevada (Las Vegas). People with AIDS made up 81
percent of these cases. The CDC found that the only risk
factor for the infection was drinking tap water (the subgroup of
cases with CD4 cell counts less than 100 were twelve times more
likely to have drunk tap water than matched controls). This outbreak
occurred despite Clark County's state-of-the- art water filtration
system, which produced a water quality better than what the national
standard mandates, and despite the fact that no cryptosporidia could
be detected either in the source or in the finished water. These
findings led the CDC to suggest that immune-compromised people drink
either boiled or bottled water.

       On the treatment front, one study evaluated the use of total
parenteral nutrition (TPN) in eight people with cryptosporidiosis
and weight loss (abstract I42). Although the four participants who
survived for more than five months did eventually gain weight, it
was not clear whether this was due to TPN or these patients' better
prognosis. The authors concluded that more study was needed to
determine which patients were more likely to benefit from the
expensive procedure (more than $800 a day).

       A number of studies meanwhile evaluated new anti-
cryptosporidia treatments in the laboratory. A group of researchers
from Emory University found dramatic activity with a number of
herbicides, at doses that may be safe enough to administer to humans
(abstract E42). Researchers from the University of Texas and DuPont
Merck reported that compounds inhibiting an enzyme unique to
cryptosporidia block an essential phase of the parasite's life cycle
(abstract B56). Although the compounds utilized in this study are
not absorbable in the gut, this finding may lead to new drugs
eventually.

       More effective therapy for cryptosporidiosis might require
combination strategies. One cell culture study at the University of
Arizona found synergistic activity with two available drugs,
paromomycin (Humatin) and clarithromycin, (abstract E43). This
combination could easily be evaluated in people. Currently, the
National Institutes of Health is conducting test tube studies of
combinations of NTZ (see Treatment Issues, September, 1995, page 14)
and other compounds.	- TS

       ********************
       Candidiasis at ICAAC
       by David Barr and Theo Smart

       Several presentations on oral candidiasis treatment occurred
at ICAAC. Unfortunately, there were no human studies on
treatments for vaginal candidiasis.

       One study (ICAAC abstract I221) compared fourteen day
treatments of an oral suspension of fluconazole (100 mg
daily) to liquid nystatin suspension (five ml swished or
gargled four times daily). A total of 166 patients were
enrolled, 83 in each arm. Of the evaluable patients, 59 out
of 68 patients on fluconazole were clinically cured compared
to only 33 of 68 cured on nystatin. Those on nystatin also
had significantly higher rates of relapse. Overall,
fluconazole was found to be clinically superior to nystatin.

       A Canadian/European study (abstract I-219) compared oral
fluconazole to an oral itraconazole solution for treatment of
oral candidiasis (thrush). The oral itraconazole solution has
improved bioavailability, can be administered without meals
and has a topical effect. The study included 244 patients
with thrush who received the itraconazole solution (100 mg
twice daily for seven days or 100 mg daily for fourteen
days), or oral fluconazole (100 mg daily for fourteen days).
Patients had weekly assessments of symptoms and oral cavity
cultures.

       The itraconazole solution seems to be a reasonable
alternative to oral fluconazole in the management of thrush.
Results showed that all three regimens were equivalent as
were the rates of relapse (20 to 22 percent). Side effects
were also similar.

       In a similar study, (abstract I220), oral itraconazole
solution appeared better than oral fluconazole. One hundred
ninety patients were randomized to receive itraconazole (200
mg daily for seven days or 200 mg daily for fourteen days) or
fluconazole (100 mg daily for fourteen days). The fourteen-
day itraconazole regimen produced the best results (97
percent success rate with the lowest rate of relapse). The
seven day itraconazole arm was comparable to the fluconazole
(response rates of 83 and 87 percent, respectively). Note,
though, that a total of seven itraconazole patients stopped
therapy because of adverse events compared to only one
fluconazole patient.

       The advantages of the itraconazole solution include that it
can be given along with rifabutin, rifampin and drugs that
suppress gastric acid (Tagamet, Pepcid) and that it has an
additional topical effect. How useful it will be in patients
with fluconazole-resistant candidiasis needs to be
determined. Janssen Pharmaceuticals, which makes
itraconazole, had no comment on whether it plans to market
the new oral formulation any time soon.

       Drug Resistance

       Since the organism that causes Candida albicans is a natural
part of the intestinal and vaginal flora, and thrush is
merely its overgrowth, drugs that keep candidiasis in check
may never completely eradicate it. The continued presence of
the organism during treatment makes the selection of
resistance more likely. Therefore, many clinicians do not
prescribe antifungal prophylaxis, for fear of creating yeast
strains resistant to treatment. Instead, they opt to treat
thrush as it occurs.

       It is unclear whether this approach is any better. Based on a
study of 153 patients, researchers from Dijon, France had
previously reported that prophylaxis with fluconazole can
cause resistance. At ICAAC, they reported that they saw just
as much resistance in 142 people who have not received
fluconazole prophylaxis (abstract C102). Some of those with
less susceptible yeast had been treated for thrush with
fluconazole or ketoconazole. A Spanish study of 130 episodes
of oral candidiasis in 95 people also found that fluconazole
resistance was not significantly more common in people
recently treated with the drug (abstract I103).

       A NIAID sponsored study, ACTG 816, in 202 people (56 percent
of whom became fluconazole resistant) appeared to disagree
with the Spanish researchers findings (abstract I101). This
trial found that reduced susceptibility to fluconazole was
highly correlated with both more than three episodes of
candidiasis in the previous year and previous treatment to
fluconazole.

       Doctors at Cochin Hospital in Paris evaluated the factors
that led to failure on fluconazole in fourteen patients
(abstract E62). While resistance was the most important
factor, concurrent use of other medications such as rifabutin
and rifampin, which lower the levels of fluconazole in the
blood was common in 43 percent of these patients. Other
cofactors, such as tobacco use and poor oral cavity condition
were also common. Treatment with doses of fluconazole at 400
mg a day was successful in 66 percent of these cases. Those
who failed these doses also did not respond to higher doses.
Such patients may have few treatment options, as a number of
studies reported that previous therapy with ketoconazole or
fluconazole could cause decreased susceptibility to the whole
azole class of antifungal drugs (abstracts E63, I102, and
I105).

       ****************
       Washington Watch
       by Derek Link

       New NCI Director Confronts Troubled Agency

       In his new role as Director of the National Cancer Institute
(NCI), 44-year-old Richard Klausner oversees the largest
federal biomedical research institute, with a budget of $2.4
billion. Dr. Klausner also is a leader in AIDS research. The
NCI received $213 million for AIDS research this year,
ranking it second among the 24 federal agencies that receive
AIDS research funds.

       Dr. Klausner has spent most of his career at the NIH. Before
his NCI appointment, Dr. Klausner served as the Chief of the
Cell Biology Branch at the National Institute of Child Health
and Development. He is one of the most published researchers
in the fields of molecular biology and immunology. His work
includes the biochemistry of a gene involved with human
kidney cancer, the function of T-cell receptors, and iron
metabolism.

       Dr. Klausner takes control of a troubled agency. Several
recent reports have criticized the NCI for lack-luster
science and mismanagement. The most prominent report, "A
Review of the Intramural Program of the NCI," popularly
called the "Bishop Calabresi Report" after Michael Bishop and
Paul Calabresi, the co-chairs of the panel that wrote the 43
page document, minces no words. The report describes the
"evolution of an elaborate bureaucracy" at the NCI, which
resembles a "fragmented feudal structure." The report
documents "cronyism" in the NCI's operations, calls its
organizational structure "unnecessary and wasteful", and
cites a "lack of financial accountability." The report makes
the scientific impact of these organizational and
administrative problems clear. It says many NCI programs
"cannot even be considered the best in their particular field
of inquiry."

       The Bishop Calabresi report uncovered some very disturbing
problems specific to the NCI's AIDS research effort. The
report notes that the NCI's in-house program alone spent $105
million on AIDS research in fiscal year 1994. The report said
this high level of AIDS funding was due, in part, to "liberal
definitions of what research might be related to AIDS." The
report noted that the NCI's AIDS programs are not integrated
with the overall national AIDS research effort, and
recommends greater coordination of NCI AIDS research by the
Office of AIDS Research. The report says a "significant
reduction" in NCI's AIDS research funds may be in order, and
that these funds should be used to increase the national AIDS
research effort overall. The report says NCI's AIDS research
funds might best be put "under control of a different
institute."

       Dr. Klausner is aware of the report's implications for AIDS
research at the NCI. In press reports, Dr. Klausner said the
report uncovered "something that is a real problem." He
acknowledges that it is "stretching it" to define much of the
NCI's so-called AIDS research as AIDS research. The NCI
component that receives $213 million represents over 16
percent of the entire federal AIDS research budget, a sum
considerably larger than any other single segment of the AIDS
research effort. Placing such a huge investment of federal
AIDS research funds in one site expects too much of a single
research community and shortchanges the rest of the AIDS
research enterprise.

       In a meeting with AIDS advocates in September, Dr. Klausner
said he has ordered a thorough review of all AIDS
expenditures at NCI, with the goal of re-programming and re-
classifying the money. He said he has already made $6.5
million of NCI AIDS research funds available for legitimate
AIDS research expenses and expects millions more to become
available. In another move greeted by AIDS advocates, Dr.
Klausner appointed Philip Pizzo, a noted pediatric AIDS
researcher, to lead the NCI's clinical sciences division.

       Dr. Klausner hopes to create a more collegial atmosphere with
his fellow institute directors. He told AIDS advocates that
the "Treaty of Bethesda" has been nullified. The Treaty of
Bethesda refers to an informal agreement reached in the mid
1980s between the National Institute of Allergy and
Infectious Diseases (NIAID) and the NCI. At the time a bitter
power struggle had erupted between the two institutes over
control of the AIDS research budget. NIAID director Anthony
Fauci prevailed, gaining the largest share of AIDS research
funds and making NIAID the "lead" institute for AIDS
research. Since that time, relations between the NCI and
NIAID have been strained, with collaboration and cooperation
difficult. Dr. Klausner hopes to integrate the NCI's AIDS
programs more closely with those of NIAID and other
institutes.

       Copyright (c) 1995 - GMHC Treatment Issues.  Distributed by
       AEGIS, your online gateway to a world of people, knowledge,
       and resources.  714.248.2836 * 8N1/Full Duplex * v.34+

