Accelerated Approval: Statement to FDA

[The U.S. Food and Drug Administration has requested public
comment for the September 12-13 meeting of its Antiviral
Advisory Committee, which will examine accelerated approval
-- the regulations now in effect which allow critically
needed drugs to be approved based on blood tests which show
virological or immunological improvement, with further
research to be done later. (The alternative to accelerated
approval is to require statistical proof that the new drug
improves survival or reduces serious illness before it is
approved, and such studies commonly involve thousands of
people and take several years to complete.) Drug-approval
policy requires a weighing of risks and benefits, and the FDA
has heard proposals that it retreat from its accelerated-
approval system to make sure that companies get better data
before drugs are approved.

[The meeting will be at the Holiday Inn Plaza Ballroom, 8777
Georgia Ave., Silver Spring, Maryland, starting at 8:30 a.m.
on September 12 and September 13; all parts of this meeting
are open to the public, and there might be times to make a
brief statement to the Committee even without prior
arrangement. Also, written statements can be submitted after
the meeting. Address them to Antiviral Advisory Committee,
c/o Lee Zwanziger, Ph.D., Advisors and Consultants Staff
(HFD-9), Parklawn Building, Room 8B45, 5600 Fishers Lane,
Rockville, MD 20857; or fax them to 301/443-0699. For more
information, call Dr. Zwanziger at 301/443-5455.]

The following is an outline of our talk to the Committee:

* While everybody agrees that better data is needed, what
AIDS Treatment News is hearing from people with HIV disease
is the strong, widespread feeling that access remains most
important to them, and must not be sacrificed. This is true
whether or not people are highly educated about treatments.
When conventional treatments are failing, no special
education is needed to know that other possibilities have
shown promise, and to want to try them.

* Parallel track and other pre-approval access programs
cannot replace accelerated approval, as companies are
increasingly refusing or unable to pay for them.

* The concerns around accelerated approval of ddC and other
AIDS drugs reflect the inadequacy of the drugs, not
fundamental flaws in the approval process.

* Within the research and medical communities, there is a
rapidly growing consensus that plasma RNA is a much better
indicator of antiviral effect than the measurements we have
had in the past. This new marker will almost certainly
improve future decisions about antivirals. Accelerated-
approval decisions will not be based on any one marker,
however, but will reflect a judgment involving all
information available. We should avoid premature decisions or
doctrines which would limit flexibility and prevent timely
and effective use of new technologies as professional
consensus develops.

For example, it may be tempting to impose the doctrine that
whenever accelerated approval is given, it must be in the
context of a development plan that will eventually use
clinical endpoints to prove or disprove the drug's
superiority to standard treatment. In some cases, this
approach may make sense; in other cases, different ways of
organizing post-marketing studies may be more effective in
obtaining the information physicians need to optimize use of
the drug. And the human costs of clinical-endpoint trials
(which tend to keep people on protocols that don't work for
them so that body counts can be obtained) are often
underestimated. These costs are highest for marginal drugs,
where the resulting information is of least value.

* A major problem in drug development is that most credible
treatment possibilities do not even begin clinical research,
because the cost barriers to eventual approval are too high.
This is especially tragic in view of the historical
experience that important medical advances are seldom
predictable -- meaning that the few drugs which do get big-
corporate backing are not the cream of the crop, but closer
to a random selection. Critical treatment advances may be
quietly lost, if they do not look good ahead of time and
therefore never get the chance to begin developing
credibility. Accelerated approval can reduce cost barriers by
deferring major expenses until after there is an income
stream, encouraging early testing by making the ultimate
development path more feasible.

* In addition, the movement toward individualized therapy
will make accelerated approval more important than it has
been in the past, as physicians and patients rely less on
averages and more on viral and other measurements for rapid
indications of which treatments are working for which people.