Benefit vs. Risk: How FDA Approves New Drugs
By Dixie Farley
Under current law, all new drugs need proof that they are effective, as well as safe,
before they can be approved for marketing. But it's important to realize that no drug is
absolutely safe. There is always some risk of an adverse reaction. It's when the benefits
outweigh the risks that the Food and Drug Administration considers a drug safe enough to
approve.
In fact, it was little more than 30 years ago that U.S. drug law first embraced the
idea of risk vs. benefit that is now the key to new drug approval. Providing evidence of
safety before marketing was first required by the Federal Food, Drug, and Cosmetic (FDC)
Act in 1938, but not until the Kefauver-Harris Drug Amendments of 1962 did firms also have
to show a drug's effectiveness before marketing.
Before any drug gets on the market today, the FDA decides--as quickly as a thorough
evaluation allows--whether the studies submitted by the drug's sponsor (usually the
manufacturer) show it to be safe and effective for its intended use. Here's what goes into
those decisions.
"Take AZT, for example," says Robert Temple, M.D., director of the Office of
Drug Evaluation I, in the FDA's Center for Drug Evaluation and Research. (AZT stands for
azidothymidine, the former generic name of the drug now known generally as zidovudine and
marketed as Retrovir to treat AIDS.) "It has significant toxicity. If you weren't
quite sure it had a benefit, it would be hard to describe it as 'safe.' But we know, from
well-controlled studies, that it has a benefit. In the first large clinical study with the
drug, there were 19 deaths in patients taking a placebo [an inactive substance], but only
one death in those on AZT."
Zidovudine was approved in March 1987 in a record 107 days. But no corners were cut.
Indeed, the FDA expended an estimated eight staff years at a cost of $600,000 on
zidovudine's evaluation. That the review was so rapid was due largely to the fact that the
FDA was involved with the drug every step along the way from the start of clinical studies
in AIDS patients.
The FDA has in a number of ways taken steps to make urgently needed drugs available
sooner. These are drugs for treating serious or life-threatening diseases that have no
good treatment.
Under the accelerated approval rule, the agency can rely as a basis for drug approval
on a reasonable "surrogate" endpoint--that is, an effect of a drug on a marker
of the disease, rather than an actual effect on survival or illness. (An example of a
marker would be CD4 cell counts, used to measure the strength of the immune system.
Usually such a surrogate can be assessed much sooner than such an endpoint as survival.)
In accelerated approval, the FDA approves the drug on condition that the sponsor study the
actual clinical benefit of the drug.
According to FDA Commissioner David A. Kessler, M.D., "We cannot wait for all the
evidence when people are suffering and dying from a devastating disease. But, we must
ensure that all the evidence we need eventually does get collected."
Promising Experimental Drugs
Today's policies also allow broader use of some investigational drugs even before they are
approved for marketing.
These new policies include the Treatment IND (IND stands for investigational new drug
application) and the parallel track mechanism.
Both allow promising drugs, not yet approved for marketing, to be used in
"expanded access" protocols--relatively unrestricted studies in which the intent
is not only to learn more about the drug, especially about its safety, but also to provide
treatment for people with no real alternative. But these expanded access protocols also
require researchers to formally investigate the drug in well-controlled studies and to
supply some evidence that the drug is likely to be helpful.
"This expanded access does not represent just 'giving the drug out,'" Temple
says. "The sponsor has the obligation to develop the drug properly, so we will know
whether it really is useful."
The FDA participates actively in the drug development process, seeking to provide clear
standards and expectations. Sponsors are encouraged to meet with the FDA, Temple says, at
an "end of phase 2 conference" before carrying out the large-scale controlled
clinical trials. (For information about the various phases of drug study, see "Testing Drugs in People.") At this conference, the FDA gives
advice about the design of the sponsor's study plan to ensure that the trials will be
acceptable.
As Temple puts it: "We try to find and eliminate flaws in the individual studies
and overall development plan that we know will give us trouble later on in the NDA review.
We don't want people to carry out a large study that has no chance of being considered
adequate and well-controlled."
The FDA also provides advice, he says, in the form of guidelines on how to study
particular classes of drugs and on how to submit and analyze data in a marketing
application.
In addition, to ensure that institutional review boards meet the FDA's rules for the
protection of the rights and welfare of research subjects, the agency routinely inspects
the boards every five years. "We may go more often, if there are problems," says
Frances O. Kelsey, Ph.D., M.D., director of the FDA's division of scientific
investigations.
The FDA routinely inspects animal laboratories every two years, or more often, Kelsey
says, "if a review division has a question about a specific animal study."
Reviewing NDAs
The documentation required in an NDA is supposed to tell the drug's whole story,
including: what happened during the clinical tests; how the drug is constituted--its
components and composition; results of the animal studies; how the drug behaves in the
body; and how it's manufactured, processed and packaged, especially the quality controls.
The FDA also requires samples of the drug and its labels.
Full reports of a drug's studies must be submitted so that the FDA can evaluate the
data. The controlled clinical trials are especially important because they provide the
only basis, under law, for demonstrating effectiveness. They answer the question,
"Does this drug work for the proposed use?" The whole data bank is used to look
for adverse effects. From analyses of the data, FDA reviewers assess the benefit-to-risk
relationship.
The human studies also generate information that will be in the drug's professional
labeling, the guidance approved by the FDA on how to use the drug. This is the package
insert that accompanies a drug in all shipments to physicians and pharmacies.
Whenever an NDA is submitted to the FDA, the agency lists it in a computer database,
and the division of scientific investigations learns of that NDA by routinely checking the
database.
According to Alan Lisook, M.D., who works in the division with Kelsey, "After
determining the important studies supporting approval, we send assignments to the field to
make on-site inspections of the investigators who did the work, to verify that it was
valid." The division may also participate in the inspections.
Since more and more foreign studies are being accepted as primary evidence for drug
approval, the agency has been doing a larger number of foreign inspections, Lisook says,
"the same as we do here. We compare the data submitted with those data available on
site." The sponsor makes sure the FDA has access to the research, he says,
"because we have the bottom-line authority that if we can't inspect it, we don't
believe it."
If the FDA's evaluation of studies reveals major deficiencies, substantially more work
by the sponsor may be needed, ranging from further analyses to the conduct of new
studies--in either case thereby extending the evaluation time and delaying approval.
"It's particularly important," Temple says, "that sponsors use the
opportunities FDA offers during the IND to discuss the critical studies and overall plans,
so that they know what we expect with respect to study design, conduct and analysis. This
can greatly reduce the chance that the application will 'recycle.' "
The FDA has undertaken various ways to reduce drug review time, which during the past
several years has averaged (median) about two years, down from about two-and-one-half
years.
For example, funds provided by the Prescription Drug User Fee Act of 1992 allow the
agency to hire several hundred additional reviewers and support staff and expedite its
move to accepting computerized NDAs.
Writing to Congress in September 1992, Commissioner Kessler listed the FDA's goals for
using the additional resources, including the following five-year goals for review and
approval time frames for new prescription drugs:
Within six months of an NDA's submission date, review and act on complete applications
for "priority" drugs (those appearing to represent an advance over available
therapy). Within 12 months, review and act on NDAs for "standard" drugs (those
appearing to have therapeutic qualities similar to those of an already marketed drug).
(In both classifications, major amendments received within three months of the action
due date extend a time frame three months.)
Within six months, review and act on supplements not requiring clinical data review.
Within 12 months, review and act on supplements requiring clinical data review. Within six
months, review and act on complete applications resubmitted after receipt of a "not
approvable" letter (which describes deficiencies that preclude approval unless
corrected).
Priorities
The order in which applications are looked at is determined with the aid of a
classification system. The idea is to give priority to drugs with the greatest potential
benefit. For example, all AIDS drugs receive the highest priority, and all drugs that
offer a significant medical advance over existing therapies for any disease are considered
"priority drugs."
Which of the FDA's review staffs gets an NDA depends on the drug. For example, cancer
treatments go to the division of oncology and pulmonary drug products, and contraceptive
drugs go to the division of metabolism and endocrine drug products. Generic drugs go to a
different office, the Office of Generic Drugs.
The FDA frequently seeks advice from its 17 standing advisory committees on drugs. This
is especially true when an approval decision is a "close call."
To be sure approval decisions reflect the most recent safety data, the FDA requires
safety updates four months after the NDA is submitted, again after it sends the firm an
"approvable letter," and at other times if necessary. Updates must report new
adverse reactions and important changes in the frequency or severity of effects that are
known.
After the FDA primary reviewers finish their evaluation, additional review is given by
supervisory personnel. "In general," says the agency's Leah Ripper, "office
directors take final action on new molecular entities, switches from prescription to OTC
status, and other important actions, such as a major new use of a drug. Other approval
decisions are made at the division level." Ripper is special assistant to the
director of the Office of Drug Evaluation II.
Final Actions
In the final analysis, the FDA's decision whether to approve a new drug for marketing
boils down to two questions:
- Do the results of well-controlled studies provide substantial evidence of effectiveness?
- Do the results show the product is safe under the conditions of use in the proposed
labeling? Safe, in this context,
- means that the benefits of the drug appear to outweigh its risks.
When the review is complete, the FDA writes to the applicant to say the drug is either
approved for marketing, is "approvable," provided minor changes are made, or is
not approvable because of major problems. In the last case, the applicant can then amend
or withdraw the NDA or ask for a hearing. Once its NDA is approved, a drug is on the
market as soon as the firm gets its production and distribution systems going.
As reflected by the innovations of accelerated approval, the Treatment IND, and the
parallel track mechanism, the need for effective treatments for serious illnesses has been
so great that it has called for changing the drug approval process.
"The riskiest thing we can do when it comes to life-threatening diseases,"
says Commissioner Kessler, "is to be unwilling to take risks. But when we take risks,
we have to follow through."
Thus, while change is inevitable and often desirable, there are some constants at FDA.
Safety and effectiveness, risk vs. benefit, remain the pivotal issues in the FDA drug
review.
A Special System for OTC Drugs
The FDA has always applied the same standards to non prescription drugs as it does to
prescription ones whenever proposed over-the-counter (OTC) products meet the criteria for
"new drugs." In 1966, the FDA contracted for a review of the effectiveness of
all new drugs approved solely on the basis of their safety since passage of the 1938
Federal Food, Drug, and Cosmetic Act. Special attention soon became focused on OTC drugs:
Of the 512 OTC drug products evaluated, 75 percent lacked substantial evidence of
effectiveness.
That was when the FDA decided it was time to tackle a broader review of OTC drugs--no
small job, considering that more than 300,000 products were on the market. Those products,
however, involved only about 700 active ingredients. It didn't take long for FDA planners
to decide on a strategy: Classify the drugs by treatment category (antacids, laxatives,
and so on) and evaluate the ingredients. So, rather than review thousands of, say,
individual antacid products. The FDA evaluated the far fewer active ingredients found in
them--for example, aluminum hydroxide and magnesium carbonate.
That review, under the FDA's Office of OTC Drug Evaluation, is actually a three-phase
process of producing a final regulation (called a monograph) to establish standards for
each product-treatment category.
The first phase, conducted from 1972 to 1981, was a review by panels of outside
advisors who determined whether ingredients could be generally recognized as safe and
effective for self-use. The FDA published the reports in the Federal Register.
The second phase--still continuing--is FDA's review of the panel's findings on the
ingredients. In these reviews, the FDA takes into account public comments and any new
data. The conclusions are published as a proposed rule (or tentative final monograph).
After considering any new information and objections, the FDA publishes the final
regulation, or monograph.
An OTC drug product doesn't need specific approval before marketing so long as it meets
its category's standards.
Sometimes an approved prescription drug is deemed safe enough for self-use and is
switched to OTC status.
A number of ingredients were taken off the market as a result of the advisory panel's
OTC drug review. Among them were:
- camphorated oil, a liniment often accidentally ingested with frequently toxic results
- hexachlorophene, once common in deodorant soaps, but now available only by prescription
for special antimicrobial purposes because it may damage the central nervous system
- tribromsalan, removed from drugs and cosmetics because it was found to make skin extra
sensitive to light
- zirconium, still safe in most forms of antiperspirants, but removed from aerosols
because of concern it could cause lung nodules.
For lack of proof of effectiveness, the FDA banned some 200 ingredients in November
1990, including products used to treat problems ranging from acne and dandruff to diarrhea
and pain. In May 1993, the agency banned several hundred more, including products for such
problems as pain, digestive upsets, menstrual symptoms, and skin rashes.
Source: Food and Drug Administration (FDA)
"Benefit Vs. Risk: How FDA Approves New Drugs" originally appeared
in the December 1987-January 1988 FDA Consumer and was substantially revised for the FDA
Consumer Special Report on New Drug Development in the United States (January 1995).
|
