IN BATTLE AGAINST HEROIN, SCIENTISTS ENLIST HEROIN

Taking a bold approach to addiction, researchers at Columbia
University are offering free heroin to addicts and paying them to use
it under a federally authorized program aimed at finding a cure for
their drug habits.

Since September, 14 addicts have received regular doses of pure heroin
after being given naltrexone (pronounced nal-TREX-own), buprenorphine
(boo-pruh-NOR-fin) or methadone, which are medications found to be
effective in neutralizing heroin's high.

Because increasingly pure heroin is being sold on the streets, the
researchers want to learn whether stronger doses of these medications
are needed and can be prescribed without risk.

"A dose that was probably good enough 15 years ago may not be good
enough for heroin now," said Dr. Herbert D. Kleber, a professor of
psychiatry at Columbia who directs the medical college's division on
substance abuse and is medical director of the National Center on
Addiction and Substance Abuse at Columbia.

Heroin and cocaine have been tested on mice and other laboratory
animals. But Columbia's study is now the only one in the country
testing addicts with heroin, although researchers at Harvard have done
so and researchers at Johns Hopkins plan to.

"If you have a promising drug in animal trials," Dr. Kleber said, "you
need to do it in human trials to see if it adequately works, because
we're different from animals in physiology, metabolism and the ability
to report on subjective effects" of drugs like heroin.

Dr. Marian W. Fischman, a psychologist at Columbia who studies
medications for drug abusers, said the volunteers, who agree to stay
in the hospital for six or seven weeks, had rejected offers to treat
their heroin habit. "We would not give a drug of abuse to someone who
was seeking treatment," she said. "We refer anyone to treatment who
would even moderately consider it." The medications being tested
affect the brain different ways. Naltrexone is an antagonist, meaning
that it blocks the effect of heroin on the brain's receptors.
Methadone is an agonist, which stifles the craving for heroin by
binding to the brain receptor.

The third, buprenorphine, acts as an agonist at lower doses and as an
antagonist at higher doses.

Unlike methadone, naltrexone could be available with just a doctor's
prescription. But because a dose lasts only a day or so and leaves
some users feeling anxious and unhappy, many tend to quit. Fewer than
2,000 heroin addicts nationwide take naltrexone, compared with 115,000
methadone users. In one study, the researchers are experimenting with
a slow-release form of naltrexone that lasts 30 days.

The second study, involving buprenorphine and methadone, gives
participants who have received a dose of either drug the chance to
perform laboratory tasks for heroin or for money. If the drug works
properly, the addict will logically choose the cash. But to earn a
full dose of heroin or $20, the participant must work hard, by pushing
a button as many as 11,500 times. The study is taking place while
buprenorphine is undergoing approval by the Food and Drug
Administration.

"Buprenorphine may have certain advantages over methadone," Dr. Kleber
said. "It's longer acting, easier to withdraw from and has a ceiling
effect" on depressing breathing, he said, making it unlikely that a
user could overdose on it. The heroin addicts are recruited for the
studies through newspaper advertisements as well as word of mouth.
Strict standards for acceptance exclude anyone who is dependent on
other drugs, was convicted of a violent crime, is on probation or
parole or has medical or psychological problems making them unsuitable
participants. And because of the hospital time involved, the addicts
would not be able to hold a job during the test. The National
Institute on Drug Abuse, which is financing all six studies with
$708,000 in grants this year, has issued the program a certificate of
confidentiality so that the names of participants cannot be subpoenaed
by prosecutors or the police.

"We have to guarantee them confidentiality," said Dr. Kleber, who
declined to let addicts in the studies be interviewed or photographed.
According to the protocol summaries, nearly half of them are white,
with the rest primarily black or Hispanic. Roughly a third are women.
"We do everything possible to minimize the risks in terms of the
selection of subjects," Dr. Kleber said. They must be physically
healthy, from 21 to 45 years old, and dependent only on heroin, not on
other drugs.

The researchers are mindful of reducing the harm that heroin users
cause themselves. "They take less drugs with us than they would on the
street, and they're being taken better care of," said Dr. Kleber, who
with Dr. Marc Galanter was co-editor of "The American Psychiatric
Press Textbook of Substance Abuse Treatment" (1994), widely used by
health professionals. First, the volunteers go through several days of
detoxification, in which their heroin is stopped and they are put on
sufficient buprenorphine to suppress the pangs of withdrawal. Then
they are given naloxone, a medication used by emergency room doctors
to treat heroin overdoses, to make them ready for naltrexone. Once
this is administered, they go back to heroin again to see whether it
is neutralized by the naltrexone and for how long. Because street
heroin varies widely in purity, the National Institute on Drug Abuse
supplies the studies with pharmaceutically pure heroin, with the
approval of the Drug Enforcement Administration.

"It is given under very safe conditions," Dr. Kleber said. "We
carefully monitor the effects of drugs on the body. And they have to
keep responding on a computer to tell us how they feel."

The testing takes place in a small room, where the addict is wired to
a battery of machines measuring vital functions like heart rate and
blood pressure. During the individual sessions, which last about two
hours, the addicts answer a succession of questions from the computer.
Are they yawning or sweating? Do they feel depressed, sedated or
energetic? How much would they pay for the heroin they feel, on a
scale from nothing to $20? On the street, a bag of heroin supplying a
single high costs $10.

"Addicts don't like to be able to take drugs in controlled
circumstances like this," Dr. Kleber said. "It's not the way they
prefer to take drugs. They want to use drugs when they want, as much
as they want." The addict is also asked to perform tasks, like
duplicating patterns on a computer screen. "People who are intoxicated
have an impairment of higher functions," said Dr. Adam M. Bisaga, a
psychiatrist who helps supervise the tests. The researchers watch from
behind a one-way mirror. Every 20 minutes, a researcher steps in and
takes a snapshot of a volunteer's pupil to measure the heroin's
effect. (Heroin constricts it; cocaine dilates it.) During their
hospital stays, the addicts must attend lectures about heroin's
effects on their bodies and learn about ways to shake the habit.
"There's potential value for them at a later date if they decide they
do want treatment," Dr. Kleber said. And, he added, "they're free to
leave at any time." The addicts agree to participate in the studies
for $25 a day, but if they complete the study, the amount doubles to
$50 a day, for a maximum of about $2,700 which they must pick in
weekly intervals of $300 at the hospital after the study is over.

When the participants pick up their checks, study workers try to
counsel them to seek treatment. They are also paid for several weeks
of detoxification and preparation before the studies officially begin.
"This is for staying in the hospital for six weeks, which is something
that most people don't want to do," Dr. Kleber said. "There's
isolation, boredom, inactivity."

A third study financed under the Federal grant focuses on heroin users
who do want treatment. They are given naltrexone, but no heroin, as
outpatients, with a spouse or friend as a monitor.

"It's a very sophisticated approach, teaching somebody to become part
of the therapy team," Dr. Fischman said.

Here, the volunteers are paid only $25 a week, plus $3 for
transportation, if they keep taking the naltrexone and produce clean
urine in a drug test. The money is given in vouchers for food, clothes
for a job interview and what Dr. Kleber called other "pro-social" needs.

About 810,000 Americans regularly use heroin, according to the White
House's Office of National Drug Control Policy, and 3.6 million are
regular cocaine users, but a proven cure for their addiction has yet
to be found. And even when an addict withdraws from heroin, a cocaine
habit often remains. Dr. Edward V. Nunes, the investigator for the
cocaine studies, estimated that at least 25 percent of people who take
methadone to block their addiction to heroin continue abusing cocaine.
"Part of the trouble with cocaine is convincing the patient that they
need help." Dr. Nunes said. "The majority of them really dance around
it."

To that end, the three other Columbia studies are looking at cocaine
and how its use is affected by desipramine, an anti-depressant;
respiridone, an anti-psychotic, and memantine, a drug marketed in
Europe for dementia and Alzheimer's disease. No cocaine is dispensed
in these studies because the volunteers have accepted the offer for
treatment. Earlier studies have suggested that such medications may be
useful in blocking or diluting a cocaine high.

"No one has been successful in developing a drug for cocaine," Dr.
Kleber said, "so we're looking at treatment for sub-groups" of addicts
whose underlying emotional and mental problems may have led them to
cocaine abuse. Dr. Kleber said that finding a cure for drug addicts
would have spillover benefits of reducing crime and cutting welfare
and health-care costs. "The community should support treatment," he
said, "not only because it is good for the addict but also because
it's good for the rest of us."