|
|
|
|
Articles
back
How to Test Drugs on Kids? FDA Starts
the Debate
Little Information Is Available on Medications'
Effects on Children
In March 1958, researchers at the University of Southern California in Los
Angeles began a study to determine the effects of the antibiotic
chloramphenicol on premature infants. By February 1959, over 60% of the babies
who had been given the drug were dead. In total, 52 of the 126 children
enrolled in the study eventually died. But the researchers never halted the
study, despite knowing that chloramphenicol could kill.
"We discussed stopping the study early, and the decision was made ... that
unless you have convincing evidence, nobody's going to believe you," one of the
researchers said later. The study's goal was to discredit the use of that
antibiotic in infants -- a practice that may have resulted in thousands of more
deaths, the researchers said in their defense.
Now, more than 30 years later, U.S. health officials once again want convincing
evidence about drugs' effects on children. To make it happen, the FDA has
passed a rule requiring drug makers to conduct clinical trials aimed at
determining the effectiveness and effects of commonly used medications on
children, beginning this December.
There is a surprising lack of information about the appropriate use of
prescription drugs in children, the FDA says. The agency estimates that more
than half of the drugs approved every year that are likely to be used in
children are neither adequately tested nor labeled for children, leaving
doctors to determine the appropriate doses. Information on drug safety and
effectiveness is especially sparse for children under age 2.
But in an effort to avoid past mistakes, FDA officials this time are consulting
experts on how to establish the guidelines for these types of studies. Those
consultations began this week with a number of meetings between FDA officials,
pediatric specialists, and experts in other medical specialties such as
oncology. The issue at the heart of the discussions: To what level of risk or
discomfort can a child be ethically exposed to in a clinical trial?
Dianne Murphy, MD, associate director of pediatrics at the FDA, says the agency
believes that these trials should be conducted using only people who may
benefit. But whether the studies ethically can include a placebo group largely
remains a point of contention, she says.
A placebo generally is a sugar pill or other facsimile of the real drug being
studied used for comparison purposes. It has no treatment benefit. While some
scientists and the FDA believe that side effects and clinical benefits of a
drug cannot be determined without the use of a group taking only a placebo,
others argue that it is unethical to leave children untreated -- even when the
child suffers from a nonfatal disease, such as a skin rash.
"An investigator's chief concern ought to be the health and well-being of her
patients," Charles Weijer, MD, cautioned FDA officials at the meeting, which
included several representatives from Europe who are working on developing
international standards. The standards set by the FDA also will set a precedent
for what happens abroad, cautioned Weijer, a bioethics and an assistant
professor of medicine at Dalhousie University in Nova Scotia.
Still, there are several areas in which researchers agree, says Robert Temple,
MD, director of medical policy at the FDA's Center for Drug Evaluation and
Research. Overall, they agree that placebos may be used when there is no
alternative therapy. They also agree that neither children nor adults should be
denied treatment if they are suffering from a life-threatening disease.
As for nonfatal diseases, using alternative study designs could help bridge the
opinion gap among researchers, Temple says. Among those is a simple method that
would test the standard treatment vs. the standard treatment plus the new drug
-- in effect making the group getting the standard treatment act as the placebo
group.
But while this type of study could help test certain drugs in fields such as
oncology, or cancer, it is unlikely to help researchers looking at disorders
for which there is no standard treatment regimen, such as depression, Temple
says. In those areas, the most effective studies would test the drug in
question against a placebo, while allowing patients to withdraw if they exhibit
serious symptoms.
The debate over placebos is not the only issue the FDA faces in its quest to
involve children in clinical trials, Murphy says. For example, the FDA must
determine how to design the informed consent form, a form designed to ensure
that study participants understand the risks.
Ironically, when members of Congress grilled the University of Southern
California researchers in 1972 about the chloramphenicol experiment, they at
first criticized the researchers for withholding needed drugs by including a
placebo group, Murphy says. It was not until someone observed that most of the
deaths were among babies in the group receiving treatment that the researchers
were criticized for administering the antibiotic.
All of this leaves the FDA with a lot of ground to cover in a short time. FDA
officials themselves have requested 332 studies involving children, of which
they estimate about 282 will be conducted. Thanks to financial incentives
passed by Congress in 1997, such as granting six additional months of marketing
exclusivity rights for drug makers who conduct voluntary studies of their
products on children, the number of studies initiated by someone other than the
FDA has exploded over the past few years.
But for the FDA, that is good news, Murphy says. While some may believe that
the USC experiment was unfounded, the principle that guided those researchers
40 years ago still applies today, she says. "Unless you have convincing
evidence, nobody's going to believe you."
|
|
back
|
|
|
