I enjoyed the article "Premarket Approval or Premarket
Notification: An Update" by Jonathan S. Kahan (October 1984).
and I found it to be accurate and well written. Since I was formerly
a medical officer for FDA, I would offer some suggestions to
assist companies in preparing and presenting the clinical aspects
of a premarket approval (PMA) application.
Probably most important is learning to write a protocol. This
document is critical and can make or break a study. The protocol
should clearly state the motivation for the study with a reasonable
literature review and should discuss the objectives (i.e., purpose,
methods of measurement, length of time, and number of patients
required). The study design should include how patients will
be recruited, which patients will be accepted, and which patients
will be specifically excluded. FDA requirements on informed consent,
institutional review boards, etc., should always be conscientiously followed.
Choosing the proper clinical trial is also important; for this,
comparability is the key. The ideal clinical trial would be large,
multicentered, double-blind, and randomized; but such studies
are expensive and difficult to achieve. Therefore, you may need
to consider an alternative.
One option is to use nonrandomized concurrent controls, in which
two groups of patients, not selected on a random basis, are studied
with different treatments or tests. A disadvantage is that since
patients are not randomly selected, there is always the question
of selection bias. For example, were the sickest patients given
conventional care and the healthier patients given the new therapy?
Randomized control trials are those that are recommended.
Here, there is a comparative study in which the treatment and
control groups are assigned randomly to eliminate selection
bias. The three main advantages of such a study are the elimination
of bias in patient selection, the tendency for balance of the
baseline or prognostic factors (i.e., the severity of the illness),
and the determination of a valid test of significance.
Another option is to use an uncontrolled study. Although it is
rare that these studies are satisfactory, they may be useful
in collecting safety data when large numbers of patients are
required to determine a safety issue.
Since PMAs are reviewed by both FDA and an outside panel of experts,
knowing how to prepare for the presentation can prove equally
beneficial. Effective presentations require skill, experience,
and effort; clear conclusions should be achieved from a thorough
analysis of the subject.
1. First and foremost, know the submission requirements and
submit all documents.
2. Inquire about providing additional copies of your report so
that a reviewer does not have to be waiting for a copy.
3. Know your audience. The interests, backgrounds, and expertise
of those participating in the review should affect your presentation.
4. Explain why this new technology is important -- e.g., does
it save lives, time, or money? Does it improve the quality of
life?
5. Define your goals. Be clear and specific about your conclusions.
The sponsor should be absolutely clear and certain about what
is to be achieved, i.e., the labeling goals.
6. Know your visual aids and be prepared to continue your presentation
in the event of a mechanical failure.
7. Be mindful of the allotted time. Begin by stating your objective
and conclude by summarizing the achievement you expect.