that has been gathered subsequent to the filing of
the NDA.
Federal law requires that FDA issue a notice of
action within 180 days of filing the NDA. There are
three forms of action: approval, approvable, or non-
approvable. Approvable letters must indicate all the
deficiencies that FDA has identified that can, upon
rectification, lead to approval. If such deficiencies
require the submission of additional data, however
slight, then FDA has another 180 days to review the
application. If the deficiencies are administrative
(e.g. debate over the precise wording in labeling),
then the FDA must act within 90 days of a resub-
mission. Lack of agreement on labeling is the major
reason for issuance of an approvable letter rather
than an approval letter. Although most FDA rev-
iewing divisions will only negotiate the proposed
label, word for word, after the issuance of an
approvable letter, some companies have been able
to go directly to an approval letter as the first action
by a combination of good communication with the
FDA and the submission of a realistic package
insert. The labeling negotiation itself will often be
done by fax and counter-fax, possibly culminating
in a face-to-face meeting at FDA premises.
NDA approval is sometimes contingent on the
sponsor making various commitments. Most
recently, the Company is being asked to conduct
post-marketing surveillance studies for safety
issues that may be more or less well defined.
Post-NDA safety report frequency will also be
agreed prior to approval. Occasionally, there may
be a toxicology study that FDA regards as out-
standing but not crucial to drug launch. There
may also be stated requirements for additional
indications that have been refused by the Company
at the initial NDA approval.
31.7 Sources of guidance
Both CDER and CBER have published a large
number of guidance documents that are now also
available at the http://www.fda.gov web site. Some of
these are simply ICH documents in English. How-
ever, FDA has gone far beyond this, in supplying a
large amount of valuable information. Guidances
are not binding (on either sponsor or FDA), but it
would be fair to say that clear reasons would have
to be enunciated by FDAwhen requiring the guide-
line to be exceeded, and by the sponsor when
suggesting a variance from them.
One of thedifficulties in dealingwith FDA is that
reviewing divisions interpret these guidances dif-
ferently. These differences can be profound. The
term ‘adequate and well-controlled studies’ is used
to describe the requirement for complying with the
need to demonstrate drug efficacy. Most reviewing
divisions in CDER still tend to interpret this to
mean two independent, large-scale phase III clin-
ical trials despite the clarification in FDAMA. Yet,
CBER will approve drugs with a single phase III
study and some consistent phase II data. Similarly,
although the ICH guideline states that drugs used
for intermittent, acute therapy do not need to have
lifespan carcinogenicity tests, there can still be
different interpretations within FDA regarding
the definition of ‘intermittent’. Anesthetic drugs
are usually exempt from these long and resource-
intense animal studies; but should this apply to
acute treatments for disease, labeled for a maxi-
mum of three doses per week, and with relatively
short half-times of elimination, or not?
Another example was the Computer Assisted
NDA (CANDA). The Cardiorenal division within
CDER embraced this technology rapidly and
developed its own guidelines as to the technical
parameters for this innovation. When the rest of
CDER caught up (several years later), it was clear
that consistency with an established and successful
CANDA format was not on the agenda.
A further example relates to the pre-IND meet-
ing. Some FDA divisions do not like them, and if
reviewers attend, they have a tendency to provide
less valuable information than they would for an
EOP2 meeting or a pre-NDA meeting, the so-
called ‘entitled meetings’. But, within the industry,
there are a number of companies who have similar
attitudes about the value of pre-IND meetings. The
notable difference is that the FDA has to go to the
pre-IND meeting if scheduled. The companies who
see little value merely do not schedule them.
The bottom-line is that guidances are merely
that – guidances. Individual reviewers at FDA are
unlikely always to agree with what are essentially
consensus documents.404 CH31 UNITED STATES REGULATIONS