Phenytoin (Dilantin) as Antiretroviral Treatment -- Negative Report?

like to publish "negative results" -- indications that
something does not work. As a result, other researchers do
not become aware of information they should have.

The following report suggests that the prescription drug
phenytoin (Dilantin) is not promising as a potential HIV
treatment -- despite early findings in laboratory tests. No
one knows for sure, however, since no trial has been done.
The potential value of discovering a well-known drug which
might be useful in HIV disease suggests that the scientific
book should not yet be entirely closed. However, given
current knowledge, there seems to be no rationale for people
trying Dilantin as an HIV treatment.]

This spring AIDS TREATMENT NEWS heard reports that phenytoin
(sold by Parke-Davis under the trade name Dilantin), an anti-
seizure medicine which has been used for many years to treat
epilepsy, had shown anti-HIV activity in test tube studies,
but that Parke-Davis had shown no interest in developing it
as an AIDS treatment. While these reports have proven
essentially true, it is not at all clear from existing data
how much -- if any -- value phenytoin might have as a
treatment for HIV infection.

In a study published in 1990 in Virology(1) Hans-Anton Lehr,
M.D., Ph.D. and colleagues (building on work begun in 1986
and published in two lesser-known German journals while Lehr
was an intern in Hamburg) compared lymphocytes from
phenytoin-treated non-epileptic patients with comparable
cells from healthy, untreated subjects. They found that the
drug affects the fluidity of the plasma membrane of CD4 cells
in a way that should reduce the ability of viruses such as
HIV to bind to them. Additionally, when lab cell lines were
incubated with phenytoin at concentrations equivalent to
those reached in anticonvulsive therapy, the drug almost
completely blocked HIV from infecting cells.

In an interview with AIDS TREATMENT NEWS, Dr. Lehr also
described apparent improvement in one AIDS patient he had
treated with phenytoin during roughly the same time period as
his lab experiments (the patient was treated for
approximately two months, then left the area, stopped the
treatment and subsequently died).

Because phenytoin is a "very well-known and well-
characterized drug with a very well-known spectrum of side
effects," Lehr says he was anxious to see it tested in
clinical trials on HIV/AIDS patients, but little was done.
Researchers they approached, he asserts, were more interested
in other drugs such as the then-emerging nucleoside analogs.
"We approached Parke-Davis several times," he continues, "but
the response we got was that they were basically not
interested" -- possibly fearing that association with AIDS
would hurt their ability to market the drug for approved
uses.

What little data exists beyond Dr. Lehr's work is less
encouraging. A small uncontrolled German study, involving 18
symptomatic HIV patients (and whose dosing and methodology
Lehr criticizes), showed little apparent effect from
phenytoin therapy.(2)

Further test tube work done by Dr. Miles Cloyd and associates
at the University of Texas Medical Branch in Galveston, also
published in Virology,(3) yielded mixed results. In
laboratory cell lines phenytoin both blocked new infection of
cells and reduced HIV expression (as measured by p24 antigen
levels) in chronically infected cells. Used in combination
with low-dose AZT, the drug was more effective than either
used alone.

But Cloyd's team was unable to get similar results using
normal lymphocytes taken from HIV-negative donors (instead of
cell lines grown in the laboratory). Here phenytoin showed
little or no effect. Further unpublished research, Dr. Cloyd
says, produced the same result, with or without AZT. "I did a
lot more lymphocyte studies, he told AIDS TREATMENT NEWS,
"and saw usually no effect at all and occasionally a very
minor, marginal effect." Feeling that the normal lymphocyte
studies represented a closer approximation of real-world
conditions than the cell-lines, the disappointing results
caused Cloyd to lose interest in phenytoin.

Lehr argues that the extensive processing required to
separate out and culture the lymphocytes used in such
research can "induce an enormous artifact in these cells,"
possibly altering their functioning. He doubts that any test
tube study can give us much more useful information, and
still would like to see a well-designed clinical trial that
might give a clear answer.

Parke-Davis spokesperson Sandy Horner says that the company
has never done any HIV-related research with phenytoin, but
insists the company has not had an aversion to such work. "If
we thought there were something there we would definitely be
willing to look into it," she says.

Comment

Unfortunately, this situation seems to tell us more about the
scattershot process of examining potential AIDS treatments
than it does about whether or not phenytoin has any value
against HIV. A definitive study has not been done, and may
never occur. That there has been no coherent followup with a
drug long in use, which has a well-established safety and
toxicity profile, and which showed at least a glimmer of
anti-HIV activity, is disconcerting.

It might be interesting to undertake a systematic review of
HIV patients who have been treated with phenytoin for
epilepsy or related disorders. If such a review showed any
indication of benefit, a clinical study might then be
undertaken.

In any case, the case of phenytoin seems to underscore the
need for an effective system for quickly moving drugs that
show promise in the lab into small, efficient, well-run
trials to examine their clinical usefulness.

References

1. Lehr HA, and others. Decreased binding of HIV-1 and
vasoactive intestinal peptide following plasma membrane
fluidization of CD4+ cells by phenytoin. VIROLOGY. 1990;
volume 179, pages 609-617.

2. Kern W, and others. Treatment of symptomatic HIV infection
with oral diphenylhydantoin. AIDS-FORSCHUNG (AIFO). 1988;
volume 6, page 334.

3. Cloyd MW, and others. Inhibition of human immunodeficiency
virus (HIV-1) infection by diphenylhydantoin (Dilantin)
implicates role of cellular calcium in virus life cycle.
VIROLOGY. 1989; volume 173, pages 581-590.