With a unanimous 12-0 vote, members of the U.S. Food and Drug
Administration (FDA) Psychopharmacologic Drugs Advisory Committee (PDAC) told
the agency to reverse its recently implemented requirement calling on drug
companies to submit both short-term (eight to 12 week) and longer-term (more
than six months) efficacy data as part of an initial application for approval
to market a psychiatric medication.
The PDAC members expressed concern that the requirement would slow drug
development and timely approval of new medications for the treatment of mental
illness. In addition, they expressed concern over how longer-term data should
be collected and by whom.
"We'll likely follow your advice," said Thomas Laughren, M.D.,
acting director of the FDA's Division of Psychiatry Products. Laughren
officially spoke for the agency at a day-long public hearing late last
Legally, the FDA is not bound by the votes of its advisory panels.
Historically, it has nearly always followed its committees' leads, although
there have been notable exceptions recently in areas other than psychiatric
The unanimous "no" vote was in response to the first of a list
of 12 questions Laughren had submitted to the PDAC for its consideration. Of
the 12 questions (some multipart), formal votes were requested for four, with
the remainder of the questions intended only to generate discussion. The
PDAC's answer to the first question, "Is it a reasonable expectation
that a sponsor would have accumulated data for both acute and longer-term
efficacy trials at the time of filing of an application for a drug treatment
for major depressive disorder?" seemed to be a foregone conclusion
nearly from the outset of the hearing. Indeed, of those speaking throughout
the hearing, Laughren seemed to be the only one who supported the notion.
Yet, several members of the PDAC expressed concern over how the public
might view the panel's "no" vote. Some members said they feared
that the panel's vote would be considered a rebuke to the FDA for attempting
to require "a higher standard of evidence" for approval of
In an unusual move, the PDAC unanimously voted to approve the release of a
public statement saying: "The advisory committee recognizes the need for
evidence to inform clinical practice regarding long-term treatment efficacy,
without potentially slowing progress of new drug development. We encourage
collaborative efforts by industry, the National Institutes of Health, and the
FDA to further research on long-term treatment."
At the end of the day, the remaining questions that the FDA had presented
to the PDAC were not directly addressed, and no further votes were taken.
An unusual coalition involving 10 pharmaceutical companies (all of which
are direct competitors in the psychiatric drug market) provided nearly three
hours of joint testimony on how the requirement for long-term data at the time
of initial filing would effect the drug-development process. The overwhelming
message was that the requirement would significantly slow down drug
development, perhaps obstructing it altogether. Bringing psychiatric
medications to the market would take longer and be more costly, the companies
"Both acute and long-term efficacy data are needed for psychiatric
disorders that are recurrent or chronic; everyone agrees with that,"
said Frederick Goodwin, M.D., a professor of psychiatry at George Washington
University, who introduced the companies' joint testimony. "But [an
indication to treat an acute episode and an indication for maintenance
treatment] are very different indications and should not be made
Yet, FDA's Laughren argued throughout the day for requiring"
longer-term data," saying at the outset that "most
psychiatric illnesses are chronic, and all psychiatric illnesses for which
psychiatric drugs have approved indications in the U.S. are
Even so, he emphasized, all of the drugs currently approved for psychiatric
illnesses gained their approvals with short-term efficacy data (generally
eight to 12 weeks). In contrast, he noted, "most treatment guidelines
for chronic psychiatric illness" recommend between four and six months
of medication treatment.
"Clinicians," he said, "have generally not had sufficient
evidence base to support what is the standard of practice."
Prior to six months ago, new medications for psychiatric indications were
routinely approved on the basis of two randomized, placebo-controlled clinical
trials, lasting eight to 12 weeks, that showed a drug was efficacious for the
indication sought and safe to use.
Recognizing the need for longer-term data, the FDA increasingly over the
last several years has added postmarketing research commitments (commonly
known as Phase IV clinical trials) to drug-approval letters. Phase IV
commitments have often included requests to complete longer-term studies
generally ranging from six to 12 months and focused on maintenance of efficacy
and/or relapse prevention.
On new drug applications (NDAs) for medications to treat mental illnesses,
the FDA has often asked drug companies to complete "randomized
withdrawal studies" in which patients receive initial treatment on an
open-label basis for a minimum period. Those patients who meet response
criteria and remain stable are then randomly assigned to either continue on
medication or switch to placebo. The endpoint in this type of trial is usually
either time to relapse or some measure of relapse rate.
However, beyond requirements for pediatric studies (imposed under the
Pediatric Research Equity Act of 2003), Laughren reminded committee members,"
I'm not aware of any regulatory way to enforce Phase IV
Simply put, the FDA can ask manufacturers to complete Phase IV commitments,
but has no regulatory authority to require their completion. As a result, many
requested studies have not been completed in a timely manner, and some have
not been completed at all.
Laughren said he would estimate that "70 percent to 80 percent of
requests for Phase IV studies are completed by manufacturers; however, it
often takes three to five years for the FDA to get the data." Industry
representatives at the hearing generally agreed with those statements.
Last spring, however, the FDA began asking drug manufacturers to submit
evidence of both short-term and long-term efficacy for drugs intended to treat
mental illness at the time the company submits its NDA, rather than on the
tail end of the approval process.
"[The FDA] has been thinking about this issue for a long time,"
Laughren said. The agency, he continued, has over the last six months required
data from a randomized withdrawal trial in which patients who have responded
to the medication have been stable for at least six months prior to being
randomized to continue medication or switched to placebo. After patients are
randomized, they must be followed for at least six months, making the minimum
duration of the clinical trial one year.
"This proposed policy has been met with considerable resistance and
questions," Laughren told PDAC members, "and for this reason, we
thought it would be useful to bring this general issue to the committee for
In essence, Laughren was asking the PDAC to endorse the proposed policy,
giving the agency's efforts to collect long-term data more credence.
The PDAC's discussion of the proposed up-front requirement for long-term
data was so heavily weighted against endorsing the requirement that PDAC
chair, Wayne Goodman, M.D., a professor of psychiatry at the University of
Florida College of Medicine, challenged committee members to discuss any
possible benefit to voting yes, "if only to play devil's
It appeared that no one discounted the need for long-term data on the
effectiveness and safety of psychiatric medications; however, several PDAC
members questioned Laughren on the intention behind bringing the question
before the advisory committee. PDAC consumer representative Jean Bronstein,
R.N., M.S., asked, "Is the intent here simply to give the FDA stronger
teeth to require long-term data?"
Laughren responded, "The issue is [that] we know that clinicians are
going to prescribe these medications long term, if they are working for a
patient. Right now, that prescribing is done without the benefit of any
He noted that other countries already require drug manufacturers to submit
long-term data, most notably the European Union
The majority of the discussion then appeared to center around two key
questions: How should long-term data be collected? And who should be
responsible for collecting it?
Nearly all of those who spoke at the meeting expressed concern that a
blanket requirement for long-term data would prove problematic. The primary
reason cited was the vast differences in treatment response both with
different medications in the same disorder and between different disorders.
Many expressed the need to tailor long-term research to specific disorders,
using different methods or different types of studies in different
With respect to who should collect long-term data, PDAC members noted that
it is already actively collected in large "real-world"
effectiveness trials funded by the National Institute of Mental Health (NIMH),
such as the recently reported data on antipsychotic medications from the CATIE
trial (Psychiatric News, October 21). NIMH has conducted large
effectiveness trials of medications for depression, bipolar disorder,
schizophrenia, and Alzheimer's disease.
"That's what NIMH is for though, it fills that role," said PDAC
That prompted a quick, but light response from Matthew Rudorfer, M.D.,
acting chief of the Adult Treatment and Preventive Interventions Research
Branch at NIMH. "They wouldn't let me bring the checkbook today,"
said Rudorfer, who is also a voting member of the PDAC.
More seriously, he continued, "Some forms of research are better at
framing different questions. The regulatory setting is good for acute efficacy
data, but maybe it is not the right venue for getting at the longer-term
Information on the public advisory committee hearing is posted at<www.fda.gov/ohrms/dockets/ac/cder05.html#Psychopharmacologic>.▪