FDA Home Page | CDRH Home Page | Search | CDRH A-Z Index | Contact CDRH

Dissemination of Information on Unapproved/New Uses for Marketed Drugs, Biologics, and Devices

[Federal Register: November 20, 1998 (Volume 63, Number 224)]
[Rules and Regulations]
[Page 64555-64588]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr20no98-15]

[[Page 64555]]

_______________________________________________________________________

Part IV

Department of Health and Human Services

_______________________________________________________________________

Food and Drug Administration

_______________________________________________________________________

21 CFR Parts 16 and 99

Dissemination of Information on Unapproved/New Uses for Marketed Drugs,
Biologics, and Devices; Final Rule

[[Page 64556]]

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Food and Drug Administration

21 CFR Parts 16 and 99

[Docket No. 98N-0222]
RIN 0910-AB23


Dissemination of Information on Unapproved/New Uses for Marketed
Drugs, Biologics, and Devices

AGENCY: Food and Drug Administration, HHS.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: The Food and Drug Administration (FDA) is issuing final
regulations pertaining to the dissemination of information on
unapproved uses (also referred to as ``new uses'' and ``off-label
uses'') for marketed drugs, including biologics, and devices. The final
rule describes the new use information that a manufacturer may
disseminate and describes the content of and establishes procedures for
a manufacturer's submission to FDA before it may begin disseminating
information on the new use. The final rule also describes how
manufacturers seeking to disseminate information on a new use must
agree to submit a supplemental application for that use within a
specified period of time, unless a supplemental application already has
been submitted or FDA has exempted the manufacturer from the
requirement to submit a supplement. The final rule provides for
requests to extend the time period for submitting a supplemental
application for a new use and describes how a manufacturer can seek an
exemption from the requirement to submit a supplemental application for
the new use. Additionally, the final rule discusses FDA actions in
response to manufacturers' submissions, corrective actions that FDA may
take or require, and recordkeeping and reporting requirements. The
final rule implements sections 551 through 557 of the Federal Food,
Drug, and Cosmetic Act (the act) (21 U.S.C. 360aaa through 360aaa-6) as
amended by section 401 of the Food and Drug Administration
Modernization Act of 1997 (FDAMA).

DATES: The final rule is effective November 20, 1998. Written comments
on the information collection requirements should be submitted by
January 19, 1999.
ADDRESSES: Submit written comments on the information collection
requirements to the Dockets Management Branch (HFA-305), Food and Drug
Administration, 5630 Fishers Lane, rm. 1061, Rockville, MD 20852.

FOR FURTHER INFORMATION CONTACT:
    Regarding biological products and devices regulated by the Center
for Biologics Evaluation and Research: Toni M. Stifano, Center for
Biologics Evaluation and Research (HFM-602), Food and Drug
Administration, 1401 Rockville Pike, Rockville, MD 20852, 301-827-3028;
    Regarding human drug products: Laurie B. Burke, Center for Drug
Evaluation and Research (HFD-40), Food and Drug Administration, 5600
Fishers Lane, Rockville, MD 20857, 301-827-2828;
    Regarding medical devices: Byron L. Tart, Center for Devices and
Radiological Health (HFZ-302), Food and Drug Administration, 2098
Gaither Rd., Rockville, MD 20850, 301-594-4639.

SUPPLEMENTARY INFORMATION:

I. Introduction

    In the Federal Register of June 8, 1998 (63 FR 31143), FDA
published a proposed rule that would add to title 21 of the Code of
Federal Regulations (CFR) a new part 99 entitled, ``Dissemination of
Information on Unapproved/New Uses for Marketed Drugs, Biologics, and
Devices.''
    The proposed rule was intended to implement section 401 of FDAMA.
In brief, section 401 of FDAMA amended the act to permit drug,
biologic, and device manufacturers to disseminate certain written
information concerning the safety, effectiveness, or benefits of a use
that is not described in the product's approved labeling to health care
practitioners, pharmacy benefit managers, health insurance issuers,
group health plans, and Federal and State Government agencies, provided
that the manufacturer complies with certain statutory requirements. For
example, the information that is to be disseminated must be about a
drug or device that is being legally marketed; it must be in the form
of an unabridged reprint or copy of a peer-reviewed journal article or
reference publication; and it must not be derived from another
manufacturer's clinical research, unless that other manufacturer has
given its permission for the dissemination. The information must be
accompanied by certain information, including a prominently displayed
statement that the information discusses a use or uses that have not
been approved or cleared by FDA. Additionally, 60 days prior to the
dissemination, the manufacturer must submit to FDA a copy of the
information to be disseminated and any other clinical trial information
that the manufacturer has relating to the safety or effectiveness of
the new use, any reports of clinical experience that pertain to the
safety of the new use, and a summary of such information.
    A detailed description of section 401 of FDAMA appeared in the
preamble to the proposed rule (see 63 FR 31143 at 31144 and 31145).

II. Highlights of the Final Rule

    Although the statute is very detailed, and the final rule closely
tracks its provisions, there are some places where the regulation fills
in the details of the statutory requirements. For example, the final
rule defines terms that were not defined in the legislation (e.g.,
``supplemental application'' and ``clinical investigation,'' and it
explains concepts that required additional explanation (e.g., what is
meant by the term ``unabridged''). The final rule also sets forth the
more detailed procedures for how to submit the required information to
FDA before disseminating any new use information (e.g., where the
information should be submitted and how many copies are required).
Finally, the final rule defines what is meant by the basic criteria
that the statute sets forth for granting an exemption from the
requirement to submit a supplement application on the basis that it
would be unethical or economically prohibitive to conduct the studies
needed to submit a supplemental application.
    The final rule has been revised in response to comments received on
the proposal. For example, Sec. 99.3 was revised to add a definition
for pharmacy benefit manger, which is not included in the statute. The
definition of ``clinical investigation'' in Sec. 99.3 also was revised.
Section 99.101 was revised to reflect FDA's position that most journal
articles and reference texts (as those terms are defined in the
regulation) would be considered to be scientifically sound and to
describe specific instances (e.g., letters to the editor, Phase 1
trials in healthy individuals) when that would not be the case.
    Section 99.103 revised the mandatory statement that the
disseminated information has not been approved or cleared by FDA. That
section also was revised to ensure that the financial disclosures
required under this part would be consistent with FDA's final rule on
financial disclosures by clinical investigators.
    Sections 99.201(a)(4)(i)(B) and (a)(4)(ii)(B), 99.203(b), and
99.401(b) were revised to clarify that for purposes of computing time
periods that begin on the date of initial dissemination, FDA will look
to the date that dissemination can begin. This clarification was

[[Page 64557]]

necessary because FDA will not know when a manufacturer actually begins
to disseminate materials.
    Sections 99.203 and 99.303 were revised to clarify that there are
two different ways that FDA can extend the time period for completing
the studies needed to submit a supplemental application for a new use:
One before any studies have begun and one after the studies have begun.
FDA also revised the standard for granting an exemption from the
requirement to submit a supplemental application on the basis that it
would be economically prohibitive. The focus is now on the revenue from
the new use rather than the revenue from the product.
    In Sec. 99.301, FDA clarified when it would require a manufacturer
to keep records identifying the individual recipients of new use
information as opposed to just the categories of such recipients.
Finally, the final rule was revised to ensure that a decision on a new
use submission would be made within 60 days.

III. Responses to Comments on the Proposed Rule

    FDA received over 50 written comments on the proposed rule. In
addition, on July 8, 1998, FDA held a public meeting on the proposed
rule. Thirteen speakers commented on the proposal. In general, the
comments expressed a diverse range of opinions, both favoring and
opposing the proposed rule, and were submitted by health professionals,
medical organizations, consumer groups, patient groups, a medical
journal, members of Congress, trade associations, and manufacturers.

A. General Comments

    Several comments addressed the concept of disseminating information
on unapproved or new uses rather than the proposed rule itself. Other
comments sought further restrictions on the dissemination of
information on unapproved or new uses, while still other comments
sought to expand the rule to cover more products.
    1. A number of comments expressed concern that the proposed rule
could result in harm to patients. One comment expressed concern over
the self-policing aspects of the rule. Another comment cited several
examples where drugs were administered for unapproved uses and proved
to be harmful. The comment stated that dissemination of information on
unapproved uses for approved drugs would further encourage the use of
``untested'' drugs and discourage clinical trials that would show
whether the drugs are safe and effective for their intended uses. The
comment asked FDA to ``revise or abandon these regulations so as to
continue to protect consumers from untested and potentially dangerous
drugs.'' One comment argued that the new rule was not ``warranted''
because the disseminated information may be inappropriate and would
pose a significant risk to public health. The comment further argued
that current practices in this area are the best way to handle
information on unapproved uses. Finally, a number of comments expressed
concern that FDA does not have sufficient resources to implement the
regulation in a manner that can adequately protect the public health.
Such comments urged FDA to direct adequate resources to implementation.
    Section 401(c) of FDAMA required FDA to issue regulations to
implement sections 551 through 557 of the act by November 21, 1998. The
final rule, which closely tracks the statutory language, represents
FDA's effort to comply with that requirement. FDA is committed to
implementing this new statutory authority consistent with its
obligation to protect the public health.
    2. Several comments claimed that dissemination of information on
unapproved or new uses of drugs for which pediatric labeling is not
available would be contrary to section 505A of the act (21 U.S.C. 355A)
as it pertains to pediatric studies of drugs because it would impede
the development of pediatric data. Several comments said that
dissemination of information on unapproved uses for pediatric therapy
should be limited to drugs that have ``sufficient labeling in the ages
of the children addressed by the information disseminated.'' Another
comment noted that dissemination of information for an unapproved use
of a drug in children when the drug's approved use has not been tested
for safety in pediatric patients may pose even more risk than
unapproved uses generally. Others said that for drugs without labeling
for pediatric populations or specific age populations, drug
manufacturers should not be able to disseminate unapproved use
information about pediatric populations or about specific age
populations not specified in the label, unless such information is
specifically requested by the physician.
    FDA declines to amend the rule as suggested by the comments. It is
FDA's hope that the statutory scheme set forth in section 401 of FDAMA
and implemented by this part will actually stimulate research and the
development of data on new uses, including pediatric uses. Moreover,
nothing in section 401 of FDAMA or its legislative history suggests
that Congress intended to exclude pediatric uses from section 551 of
the act or to further limit how information on such uses can be
disseminated. Finally, the act does not require that the disseminated
information be specifically requested by a physician in order to be
disseminated.
    Although FDA is not amending the codified language in any way, it
does recognize that the potential dangers of unapproved uses in
children may be greater than for adults because few drugs have been
tested in children. The agency will take this into account in making a
determination as to whether a proposed dissemination of information on
a new use poses a significant risk to public health such that the
dissemination under this part should not be permitted.
    3. One comment would revise the rule to exclude drugs that may be
covered by orphan drug exclusivity. The comment explained that a
manufacturer may obtain orphan drug exclusivity for a particular use of
a drug, but that other manufacturers could be marketing the same drug
for non-orphan indications. The comment stated that such other
manufacturers could disseminate information on the orphan indication,
thereby undermining the value of orphan drug exclusivity.
    There is no indication in section 401 of FDAMA or its legislative
history that Congress intended the dissemination of information on
unapproved uses of drugs and devices to undermine patent protection or
exclusivity granted to a product under the Orphan Drug Act, the Waxman-
Hatch Amendments, or the pediatric exclusivity provisions in section
111 of FDAMA. Therefore, an indication that is not included in a
particular sponsor's approved product labeling because the indication
is protected by patent or exclusivity is not eligible for dissemination
under part 99.
    4. Several comments urged FDA to broaden the proposal to include
over-the-counter (OTC) drug products being marketed under an OTC
monograph.
    Section 401 of FDAMA requires that in the case of a drug, there be
in effect for the drug an application filed under section 505(b) or (j)
of the act. OTC drugs being marketed under an OTC monograph do not have
an application filed under section 505(b) or (j) of the act in effect.
Therefore, FDA declines to revise the rule as suggested in these
comments.
    5. One comment stated that companies sometimes assist physicians
and patients in obtaining reimbursement from Medicare, Medicaid, and
private insurers by furnishing copies of journal articles and reference
publications on unapproved

[[Page 64558]]

uses to the insurer or government agency when reimbursement is denied
on the ground that use of the product is experimental. The comment
concluded that this practice appeared to be legal prior to the passage
of section 401 of FDAMA and asked FDA to clarify that it did not become
illegal as a result of FDAMA.
    Prior to passage of FDAMA, the practice described in this comment
was not permissible unless the unapproved use information was provided
in response to an unsolicited request for such information. FDA's
policy, which allows manufacturers to provide unapproved use
information in response to an unsolicited request, was not affected by
FDAMA (see section 557(a) of the act). Accordingly, manufacturers who
wish to furnish unapproved use information as described in the comment
may do so if it is in response to an unsolicited request. Otherwise,
they must comply with the requirements set forth in section 401 of
FDAMA and this part.
    6. One comment asserted that the proposal should recognize the
specific legal authorization for manufacturers to provide off-label
information to health care practitioners in response to an unsolicited
request.
    Section 401 of FDAMA added a new section 557(a) of the act, which
provides that nothing in section 551 of the act shall be construed as
prohibiting a manufacturer from disseminating information in response
to an unsolicited request from a health care practitioner. Although FDA
does not construe section 557(a) of the act as a specific legal
authorization for manufacturers to provide off-label use information to
health care practitioners in response to an unsolicited request,
Sec. 99.1(b) of the final rule recognizes this statutory provision.
    7. One comment stated that FDA should exempt manufacturers from the
``pre-approval and reporting requirements'' when the primary focus of a
publication is on the approved uses of the product.
    Section 401 of FDAMA and this part do not cover publications
regarding approved uses. FDA intends to permit manufacturers to
disseminate certain information that focuses primarily on approved uses
and that report the results of studies that have been relied on by FDA
in its approval or clearance of a drug or device without meeting all of
the requirements set forth in this part. (Cf. Guidance to Industry on
Dissemination of Reprints of Certain Published Original Data (61 FR
52800, October 8, 1996). The agency was enjoined from applying this
guidance document in Washington Legal Foundation v. Friedman, CA No.
1:94CV1306 (D.D.C. July 30, 1998) (hereinafter referred to as WLF v.
Friedman). FDA sought clarification on the scope of the order through a
motion to amend the judgment in that case.) FDA plans to issue guidance
on this issue at some time in the future pending clarification by the
court.
    8. One comment suggested that FDA exempt manufacturers from the
requirements set forth in this part if the new use that is the subject
of the information being disseminated has been accepted as standard
medical practice (i.e., indications listed in the United States
Pharmacopoeia Drug Information for the Health Care Professional (USP
DI) or American Hospital Formulary Service, etc.).
    FDA declines to create an exemption from the entire rule as
suggested by the comment. Regardless of whether the unapproved use is
listed in the USP DI or American Hospital Formulary Service, the
statutory requirements in sections 551 through 557 of the act apply to
a manufacturer who intends to disseminate information on the unapproved
use for an approved product to health care practitioners, pharmacy
benefit managers, health insurance issuers, group health plans, or
Federal or State governmental agencies. Evidence that the unapproved
use represents standard medical care may, however, enable the
manufacturer to seek an exemption from the requirement to submit a
supplemental application for the unapproved use if the manufacturer can
demonstrate that it would be unethical to conduct the studies necessary
for a supplemental application for the new use. A discussion of the
``unethical'' exemption appears later in section III of this document.
    9. Some comments stated that the proposal properly reflects the
intent of Congress and achieves the important goals of assuring the
public health and encouraging the dissemination of information. Others
argued that the proposal is contrary to congressional intent,
paternalistic and cumbersome, and would restrict, rather than
facilitate, access to information about new uses.
    Although FDA drafted the proposed rule to reflect congressional
intent, the agency has revised the rule in response to specific
comments. These revisions are meant to ensure that the final rule more
accurately reflects congressional intent.

B. Comments on Specific Provisions

1. Subpart A--General Information
    a. Scope (Sec. 99.1). Proposed Sec. 99.1 described the scope of
part 99, explaining that the part applies to the dissemination of
information on human drugs, including biologics, and devices where the
information to be disseminated pertains to the safety, effectiveness,
or benefit of a use that is not included in the approved labeling for
an approved drug or device or in the statement of intended use for a
cleared device and the information is to be disseminated to a health
care practitioner, pharmacy benefit manager, health insurance issuer,
group health plan, or Federal or State Government agency.
    10. Several comments urged FDA to add pharmacists to the list of
recipients of information under this part.
    Section 401 of FDAMA specifically lists who can receive the new use
information under this provision and proposed Sec. 99.1 tracked that
statutory provision. Therefore, FDA declines to amend the regulation as
requested. However, to the extent that pharmacists fall within the
definitions of ``health care practitioner,'' ``pharmacy benefit
manager,'' health insurance issuer,'' or ``group health plan'' (see
Sec. 99.3) they will be included as recipients of this information.
    b. Definitions (Sec. 99.3). Proposed Sec. 99.3 defined various
terms, such as ``clinical investigation'' (proposed Sec. 99.3(b)),
``health care practitioner'' (proposed Sec. 99.3(d)), ``new use''
(proposed Sec. 99.3(g)), ``scientific or medical journal'' (proposed
Sec. 99.3(i)), and supplemental application (proposed Sec. 99.3(j)).
    11. One comment urged FDA to include a definition for ``pharmacy
benefit manager'' and to include pharmacists in that definition.
    Although the statute defines the other recipients of information
under this provision (i.e., health care practitioner, health insurance
issuer, and group health plan), it does not define pharmacy benefit
manager. FDA has revised the rule to define a ``pharmacy benefit
manager'' (PBM) as ``a person or entity that has, as its principal
focus, the implementation of one or more device and/or prescription
drug benefit programs.'' PBM's, which generally include pharmacists,
typically provide claims processing services for devices and/or
prescription drugs; negotiate device and/or prescription drug prices;
negotiate volume purchase agreements with medical device and/or
pharmaceutical manufacturers, develop formularies, and institute
formulary

[[Page 64559]]

compliance programs (e.g., mandatory generic substitution programs).
The new definition is in Sec. 99.3(h) and the agency has redesignated
the remaining definitions accordingly.
    12. Proposed Sec. 99.3(b) defined a ``clinical investigation'' as
an ``investigation in humans that is prospectively planned to test a
specific clinical hypothesis.'' Several comments argued that FDA should
delete the proposed definition of ``clinical investigation.'' They
argued that restricting clinical investigations to those that are
prospectively planned is not part of the statute, that it would
preclude the use of retrospective studies, modeling studies, open label
studies, metanalysis, reference articles, and consensus standards,
which these comments assert may be useful, and that Congress never
intended for the definition to be limited in this manner. One comment
argued that the prospective planning criteria should not have to meet
the criteria for investigational new drug applications (IND's).
    FDA believes that many of these comments misconstrued what the
agency meant by the phrase ``prospectively planned.'' FDA does not
consider modeling studies, which are not actual studies, but rather
extrapolations of information or data that are used to predict how a
study might come out, to be clinical investigations. Moreover, FDA does
not consider consensus standards and reference articles to contain
adequate detail about ``clinical investigations'' as defined by this
rule. However, it was the agency's intent that the definition could
include historically controlled studies, retrospective analyses, open
label studies, and metanalyses if they are testing a specific clinical
hypothesis. To avoid any confusion, FDA is eliminating the phrase
``prospectively planned'' from the definition of ``clinical
investigation.'' In the final rule, FDA has defined a clinical
investigation to mean ``an investigation in humans that tests a
specific clinical hypothesis.''
    13. Several comments urged FDA to revise the definition of ``health
care practitioner'' in Sec. 99.3(d) to include pharmacists.
    Section 556(1) of the act (21 U.S.C. 360aaa-5(1)) defines the term
``health care practitioner'' to mean a physician, or other individual
who is a provider of health care, who is licensed under the law of a
State to prescribe drugs or devices.'' FDA's proposed regulation
tracked this statutory definition. FDA declines to revise the
definition. To the extent that pharmacists fall within this definition,
they will be eligible to receive information disseminated under this
part.
    14. Proposed Sec. 99.3(g) defined ``new use'' to mean a use that is
not included in the approved labeling of an approved drug or device, or
a use that is not included in the statement of intended use for a
cleared device. The preamble to the proposed rule explained that a new
use is one that would require approval or clearance of a supplemental
application in order for it to be included in the product labeling.
    The preamble to the proposed rule explained that ``new uses,''
include, but are not limited to: A completely different indication;
modification of an existing indication to include a new dose, a new
dosing schedule, a new route of administration, a different duration of
usage, a new age group (e.g., unique safety or effectiveness in the
elderly), another patient subgroup not explicitly identified in the
current labeling, a different stage of the disease, a different
intended outcome (e.g., long-term survival benefit, improved quality of
life, disease amelioration), effectiveness for a sign or symptom of the
disease not in the current labeling; and comparative claims to other
agents for treatment of the same condition (see 63 FR 31143 at 31145).
    A number of comments supported FDA's definition of new use.
However, others disagreed with the specific examples set forth in the
preamble as too broad. Most of the latter comments objected to the
inclusion of patient subgroups and comparative claims for approved
indications. They argued that their inclusion in the definition is
inconsistent with the agency's prescription drug advertising
regulations, which permit companies to promote patient subgroups and
comparative claims if certain conditions are met. Several comments
disagreed with the inclusion of a new age group--specifically
children--in the definition of new use. One comment argued that
children should not be considered a ``use,'' but a ``user.'' One
comment stated that the definition should focus only on information
that differs from the current labeling; it should not include
information that is consistent with, but more detailed than what is
described in the approved labeling. Finally, one comment disagreed with
the agency's characterization of a different intended outcome as an
off-label use.
    FDA agrees with the comments discussed previously, which note that
FDA's prescription drug advertising regulations permit companies to
make comparative claims about two approved uses, without getting the
claims on the approved label if the companies have on file, substantial
evidence or substantial clinical experience to support such claims.
(See Sec. 202.1(e) (21 CFR 202.1(e)).) FDA did not intend to change the
provision found in its prescription drug advertising regulations. In
addition, FDA agrees that as long as the comparison is between two
approved claims, there technically is not a new ``use'' involved.
Therefore, FDA is deleting comparative claims about approved uses from
its interpretation of ``new use.'' Manufacturers who want to make such
claims for a drug, must submit a labeling supplement or must meet the
requirements set forth in FDA's drug advertising regulations. (See
Sec. 202.1(e).) Manufacturers who want to make such claims for a
medical device must meet the requirements set forth in
Secs. 807.81(a)(3)(ii) or 814.39 (21 CFR 807.81(a)(3)(ii) or 814.39).
    With respect to claims of efficacy in a new patient subgroup,
including a new age group, claims that are more detailed than the
approved labeling, and claims that relate to different intended
outcomes (as well as with respect to some of the other types of new use
claims listed in the preamble to the proposed rule), FDA's prescription
drug advertising regulations may permit companies to make such claims
about prescription drugs in certain circumstances, without submitting a
supplement, provided they have on file the required evidence to support
the claim. (See Sec. 202.1(e).) However, FDA does consider such claims,
including claims regarding children, to be new uses in some cases. In
cases where such claims constitute new uses, manufacturers also can use
the procedures set forth in this part to disseminate journal articles
and reference publications about those claims. For medical devices,
manufacturers can use the procedures set forth in this part to
disseminate journal articles and reference publications about these
types of claims. Otherwise, they must comply with the requirements set
forth in Secs. 807.81(a)(3)(ii) or 814.39.
    15. Proposed Sec. 99.3(i) (now redesignated as Sec. 99.3(j))
defined ``scientific or medical journal,'' in part, as a journal that
is indexed in Index Medicus. It excluded scientific and medical
publications that are in the form of special supplements that have been
funded in whole or in part by one or more manufacturers. One comment
agreed that special supplements are not appropriate for dissemination
under this part. One comment, however, stated that the definition was
too narrow by requiring that the publication be listed

[[Page 64560]]

in Index Medicus and by excluding special supplements.
    The definition in FDA's rule, which excludes journals not indexed
in Index Medicus and scientific and medical publications that are in
the form of special supplements that have been funded in whole or in
part by one or more manufacturers, tracks the statutory definition.
(See section 556(5) of the act.) Accordingly, no changes to the final
rule have been made.
    16. Proposed Sec. 99.3(j) (now redesignated as Sec. 99.3(k))
defined ``supplemental application'' as a supplement to support a new
use to an approved new drug application (NDA) for human drugs or a
supplement to an approved license application for biologics. Several
comments argued that the definition of a supplemental application for a
drug should be expanded to include the possibility that a ``new use''
could require a new NDA rather than just a supplemental NDA. One
comment claimed that there are certain review divisions in the Center
for Drug Evaluation and Research (CDER) that require NDA's for all new
uses.
    There may be times when a manufacturer would be required to submit
an NDA rather than a supplemental NDA to support a new use. In these
instances, the unapproved use would not be covered by this part.
However, it would not be appropriate to exclude new uses from this part
merely because a review division assigns a new NDA number to the
supplement for administrative convenience. In the latter instance, the
difference would be in name only. Therefore, although FDA is declining
to revise the regulation as suggested by the comments, FDA will treat
applications that have been assigned a new NDA number for
administrative convenience as a supplemental NDA for purposes of this
part.
    17. One comment recommended expanding the definition of
supplemental application to cover OTC drugs that are subject to a
monograph.
    As set forth previously, OTC drugs that are subject to a monograph
are not covered by this provision. Therefore, FDA declines to expand
the definition as requested.
    18. For devices, proposed Sec. 99.3(j) (now redesignated as
Sec. 99.3(k)) defined ``supplemental application'' as a new 510(k)
submission, if the device that is cleared for marketing is the subject
of a 510(k) submission, or a supplement to an approved premarket
approval application (PMA), if the device that is marketed is the
subject of an approved PMA. One comment recommended expanding the
definition of supplemental application for devices to include a 510(k)
to a 510(k) exempt device.
    FDA agrees that the statutory provision covers 510(k) exempt
devices and so has amended the definition of supplemental application
accordingly.
    19. Several comments disagreed with FDA's definition of
supplemental application for devices because it did not include a PMA
for a new use for a device on the market under section 510(k) of the
act (21 U.S.C. 360(k)).
    Because there are no supplemental applications for 510(k) devices,
FDA could have interpreted the statute to exclude all 510(k) devices
from the scope of the rule. FDA drew a distinction between those that
require a new 510(k) and those that require a PMA because the agency
determined that this was similar to the distinction between a
supplemental NDA and an NDA (i.e., a supplemental NDA and a 510(k) are
filed on products about which the agency has some accumulated knowledge
and experience such that it is not required to start its review from
scratch; an NDA and a PMA are filed for products about which the agency
has no such accumulated knowledge or experience upon which to base a
decision).
    FDA disagrees with the comment that an original PMA submission
should be included in the definition of ``supplemental application''
for a device that entered the marketplace through the 510(k) process.
The 510(k) process and the PMA process are designed to provide
different ways to market regulated products, are supported by a
different extent and kind of data, and are predicated on different
concepts of how to assure consumer protection.
    A product entering the market via the 510(k) process does so
because the agency agrees with the sponsor that the new device is
substantially equivalent to a device commercially distributed before
May 28, 1976, or to a newer predicate device for the same intended use.
For a 510(k) product, the consumer protection objective of the act is
met in part by the accumulated experience with the predicate devices
and the review and establishment of the device category in the
appropriate class and a modicum of device specific information.
Information on manufacturing and premarket assurance of conformance to
good manufacturing practices (GMP's) are not addressed. The agency does
not, in the case of a 510(k), make an individual product determination
of safety or effectiveness.
    The act requires a PMA for a device for which there is a new
intended use with no predicate, or which raises new issues of safety
and effectiveness. Evidence required under a PMA is substantial and the
sponsor must show, through the use of well-controlled clinical trials
or, at the discretion of the agency, other valid scientific evidence,
that there is a reasonable assurance the product is safe and effective
for its intended use. As part of its review of a PMA, FDA reviews and
audits clinical trial information and the GMP's employed by the
manufacturer.
    Allowing an original PMA submission to be regarded in this context
as a supplement for a device already marketed under a 510(k) would
undermine the statutory and regulatory requirements established to
ensure the safety and effectiveness of products subject to PMA's. It
would be analogous to applying the dissemination provision to new
devices that were never legally marketed. For a PMA product, a new
intended use supplement is intended to provide the agency with
additional data supporting a new use for an approved device. It relies,
in large part, on information previously reviewed regarding product
materials, biocompatibility, design, performance, and basic safety
data. For a 510(k) product, a PMA would not be providing additional
information; it would be providing all of the information.
    To illustrate, a product not currently marketed, but that was
marketed as a general use tool without any known labeling or identified
product specific intended use in the 1960's preamendment period may be
re-introduced through a 510(k) for that same (implied) intended general
tool use (e.g., it ablates or thermally destroys tissue). The product
will be regarded as an unclassified preamendment product. If a
manufacturer wished to market it for a specific intended purpose where
that new purpose creates a new use with attendant questions of safety
and effectiveness of the new use, it must do so through a PMA. In a
recent instance, a company sought to market its unclassified
preamendment product, an interuterine probe for a cryosurgery machine
(using freezing to thermally destroy tissue), for ablation of the
uterine endometrium with ultrasound control of the location and extent
of tissue being frozen to control excessive menstrual bleeding. By
moving to a tissue and anatomic specific intended use and indication,
as well as by incorporation of a new (external) control procedure, the
manufacturer has created a new intended use. The product's underlying
safety and manufacture have never been evaluated. Even the presumption
that ultrasound

[[Page 64561]]

measurement of the extent of tissue being frozen accurately predicts
the extent of tissue necrosis and allows proper positioning of the
probe remains unevaluated. Nevertheless, the comments would argue that
this product could be the subject of an article or text disseminated
under section 401 of FDAMA.
    In passing section 401 of FDAMA, Congress intended to provide
health care practitioners important scientific information about
unapproved uses of approved products. The risks to the public of
disseminating information in a case such as that described previously
are closer to the risks from instances where there has never been an
approved product than those for a new use of a previously approved
product. FDA believes that these risks are far greater than those
authorized by section 401 of FDAMA.
2. Subpart B--Information To Be Disseminated
    a. Information that may be disseminated (Sec. 99.101). Proposed
Sec. 99.101 discussed the types of information concerning the safety,
effectiveness, or benefit of a new use that a manufacturer may
disseminate. For example, the proposal required (among other things)
that the written information to be disseminated concern a drug or
device that has been approved, licensed, or cleared for marketing by
FDA and be in the form of an unabridged reprint or copy of a peer-
reviewed scientific or medical journal article or an unabridged
reference publication that pertains to a clinical investigation
involving the drug or device and that is considered scientifically
sound by experts who are qualified to evaluate the product's safety or
effectiveness. Proposed Sec. 99.101 also described criteria for
determining whether the information to be disseminated is false or
misleading, whether a clinical investigation is ``scientifically
sound,'' and whether a reprint or copy of an article or reference
publication is ``unabridged.''
    20. One comment urged FDA to include a 60-day window in advance of
a drug's Prescription Drug User Fee Act date during which time a
manufacturer could submit proposed material for review. In other words,
the comment urged FDA to accept dissemination materials for review
before a drug has been approved.
    FDA declines to adopt this approach. The statute does not direct
FDA to accept submissions on products that have not yet been approved
or cleared. If FDA accepts submissions on products that have not yet
been approved or cleared, it may be wasting resources reviewing
submissions on products that never get approved or cleared.
    21. One comment urged FDA to make clear that this part does not
permit the verbal dissemination of unapproved use information. Another
comment suggested that companies that disseminate information on a new
use should be permitted to discuss the clinical investigation that is
the subject of the disseminated materials with the recipient.
    FDA agrees with the first comment that neither this part nor
section 401 of FDAMA, would permit the verbal dissemination of
information about unapproved uses. Section 551(a) of the act and
Sec. 99.101 refer clearly and specifically to ``written'' information.
Therefore, a manufacturer (or its representatives or agents) is not
permitted to discuss with a recipient the clinical investigation that
is the subject of the written materials disseminated under this part.
    22. Several comments asked whether Internet or electronic
dissemination would be permitted under this part.
    Although, as set forth previously, FDA agrees that the provision
was not meant to cover verbal dissemination, it could cover electronic
dissemination. However, a manufacturer seeking to disseminate
information electronically would have to ensure that all of the
requirements under this part could be met for electronic dissemination.
For example, the manufacturer would have to ensure that the recipients
of the information are appropriately limited and that all of the
required information and disclosures can be attached in accordance with
this part. FDA may, in the future, issue guidance on this subject.
    23. One comment noted the importance of requiring manufacturers to
disseminate unabridged journal articles so that information from a
clinical study is not pulled out of context or released without all
relevant data.
    FDA agrees with this comment. Both the statute and the regulation
require that a journal article or reference publication disseminated
under this part be unabridged.
    24. Several comments objected to the requirement that a reprint or
copy of an article be published prior to submission for FDA for review.
These comments argued that manufacturers should be allowed to send FDA
final manuscripts. Another comment opposed allowing submissions to
include manuscripts or preprints of articles that have been accepted
for publication. This comment stated that it could take months for
these manuscripts to be published and that they might be submitted
before the peer-review process is complete.
    FDA understands manufacturers' desire to disseminate new use
information as quickly as possible. However, section 552 of the act (21
U.S.C. 360aaa-1) requires that the peer-reviewed journal articles
disseminated under this part be published. If FDA were to accept
manuscripts before publication, it could not be sure that what gets
published, and then disseminated, is exactly what it was given to
review. The agency might not even be sure that the peer-review process
has been completed. FDA does not have the resources to verify this
information or to conduct duplicative reviews. Therefore, FDA is not
revising the rule to permit submission of unpublished manuscripts.
    25. Several comments took issue with the statement in the proposal
that information can be false or misleading if it includes only
favorable publications. These comments argued that dissemination should
not be prohibited if the only information that has been published is
favorable and the research is scientifically rigorous. These comments
noted that FDA should make clear that a single favorable publication
can be disseminated if it is objective, balanced, and discusses
appropriate safety information. One comment noted that a more
appropriate manner in which to state the issue would be to cite the
exclusion of an unfavorable publication as the example.
    FDA agrees that new use information is not necessarily without
balance or misleading just because there is no unfavorable information
disseminated with it and FDA did not intend to suggest the contrary.
FDA agrees that it would be inappropriate to find a favorable article
misleading just because it is disseminated without an unfavorable
publication when no unfavorable publication exists. What FDA will be
looking for is whether the manufacturer has failed to include
unfavorable information that exists and that is necessary to provide
balance. FDA has revised the rule to clarify this point.
    26. One comment said that proposed Sec. 99.101(a)(4) was unclear on
what ``other information concerning risks and adverse effects that are
or may be associated with the new use'' a company would have to include
to ensure that the disseminated information is not false or misleading.
    The other information refers to the additional information that FDA
can require under Sec. 99.103(a)(4). FDA has revised the rule to
clarify this point.

[[Page 64562]]

    27. Proposed Sec. 99.101(a)(5) required that the disseminated
information not be derived from clinical research conducted by another
manufacturer unless the manufacturer disseminating the information has
the permission of such other manufacturer to make the dissemination.
    One comment noted that the rule should clarify that contracts or
agreements between sponsors may specify how the data are to be used by
the sponsoring companies. In other words, cosponsoring companies should
be responsible for maintaining their own agreements without FDA input.
Several other comments opined that once a peer-reviewed article is
published, it is in the public domain and a sponsor should be able to
pursue use of the data published by the original sponsor (i.e., without
first obtaining permission) as long as proper credit is given. One
comment asked FDA to clarify the rule to show that research conducted
by an independent academic or similar organization can be disseminated
if the information meets the standards for dissemination and is legally
available for such use.
    Section 551(b)(3) of the act prohibits the dissemination of
information derived from research conducted by another manufacturer
without that other manufacturer's permission. The fact that an article
has been published does not eliminate the need to get permission from
the researching company. If it did, this requirement in the statute
would be meaningless because all information disseminated under this
part must be published. Therefore, FDA declines to revise the rule to
permit the dissemination of all published articles reporting on
research conducted by another manufacturer without that manufacturer's
permission. However, FDA agrees that cosponsoring companies can make
agreements without FDA's input and that research conducted by
independent parties does not, by the terms of the statute, require that
party's permission.
    28. One comment noted that reference publications will include many
unapproved use discussions that reflect research conducted by other
manufacturers and that proposed Sec. 99.101(a)(5) would appear to make
the disseminating company get permission from every one of those
manufacturers.
    As set forth in the proposal, FDA expects that manufacturers that
disseminate reference publications under this part will flag the
section of the text that describes the clinical investigation of a
specific unapproved use (otherwise, they would have to commit to study
all of the unapproved uses discussed in the reference publication).
Therefore, FDA would expect that a manufacturer would be required only
to seek the permission of another manufacturer if that other
manufacturer conducted the study for that specific discussion of an
unapproved use.
    29. Proposed Sec. 99.101(b)(1) provided that the determination of
whether a clinical investigation is considered to be ``scientifically
sound'' will rest on whether the design, conduct, data, and analysis of
the investigation described or discussed in a reprint or copy of an
article or in a reference publication reasonably support the
conclusions reached by the authors. It further provided that a clinical
investigation described or discussed in an article or reference
publication must include a description of the study design and conduct,
data presentation and analysis, summary of results, and conclusions
pertaining to the new use. The proposal also stated that a clinical
investigation presented in a format that does not represent a
reasonably comprehensive presentation of the study design, conduct,
data, analyses, and conclusions (e.g., letters to the editor, review
abstracts, abstracts of a publication) would not qualify for
dissemination under this provision.
    The preamble to the proposal provided that in order to provide a
basis for determining whether the conclusions are reasonably supported
and the findings represent evidence of safety and effectiveness of the
new use, the article or reference publication should provide, where
applicable, evidence that the investigation: (1) Was prospectively
planned; (2) enrolled an appropriately defined and diagnosed patient
population for the specific clinical condition of interest; (3)
accounted for all patients enrolled, including all patients who
discontinued therapy prematurely; (4) utilized clinically meaningful
endpoints or utilized surrogate endpoints that are reasonably likely to
predict safety and effectiveness; (5) used a well described treatment
regimen with a clear description of dose, schedule, duration, and route
of administration; (6) used an appropriate control group or made
reference to an appropriate historical control; (7) collected and
reported adequate information on adverse experiences, and the need for
dose reductions and treatment interruptions due to toxicity; and (8)
was analyzed in a scientifically appropriate manner. (See 63 FR 31143
at 31146 and 31147.)
    Some comments supported FDA's interpretation and applauded the
agency's efforts to ensure that journal articles and reference
publications are scientifically sound. These comments noted that FDA's
interpretation reflected what is required by most peer-reviewed
journals.
    In contrast, a number of comments objected to FDA's approach. Some
of these comments objected to FDA making any determination that an
article or reference publication is scientifically sound. They stated
that it was not Congress' intent to have FDA ``do its own peer
review.'' Others criticized the criteria set forth in the proposed
codified language and/or the eight criteria in the preamble to the
proposal. They argued that FDA would be requiring more detail than is
ever found in articles or reference publications and/or that FDA's
standard is akin to that for a supplemental application. One comment
said that FDA should require only enough detail to determine if the
article or publication is scientifically sound. One comment urged FDA
to adopt a broader definition of scientifically sound by removing the
specific requirements, i.e., prospectively planned, and recognizing the
value of scientifically sound studies as long as any limitations (e.g.,
epidemiological data) are fully disclosed. One comment said that FDA
should require the journal article to include the ``typical level of
detail'' and, if it does not, then the company should be able to attach
it to the article. Several comments opposed the specific exclusion of
abstracts. Finally, a number of comments specifically criticized the
requirement that the clinical investigation be prospectively planned.
    FDA has a role to play with respect to whether an article or
reference publication is scientifically sound. The statute includes a
requirement that the disseminated article or reference publication
pertain to a clinical investigation that would be considered to be
scientifically sound by experts qualified by scientific training or
experience to evaluate the safety or effectiveness of the drug or
device involved. FDA believes that this provision indicates that
Congress meant for FDA to look at whether experts would find that the
article or publication is about an investigation that experts would
consider to be scientifically sound. However, FDA also believes that
its role in determining whether an article or publication is
scientifically sound is limited. This approach is consistent with the
proposed rule and FDA fully expected that most journal articles about a
clinical investigation from reputable peer-reviewed journals would meet
the definition of scientifically sound set

[[Page 64563]]

forth in its proposal. Nevertheless, to ensure that the provision will
be implemented consistent with congressional intent, FDA is revising
Sec. 99.101(b)(1) to provide that FDA will find that all journal
articles and reference publications (as those terms are defined in
Sec. 99.3) are scientifically sound except: (1) Letters to the editor;
(2) abstracts of a publication; (3) those regarding Phase 1 trials in
healthy people; (4) flagged reference publications that contain little
or no substantive discussion of the relevant clinical investigation;
and (5) those regarding observations in four or fewer people that do
not reflect any systematic attempt to collect data, unless the
manufacturer demonstrates to FDA that such reports could help guide a
physician in his/her medical practice.
    Section 552(a)(2) of the act prohibits the dissemination of
information that is false or misleading. That provision prohibits the
dissemination of journal articles and reference publications that
contain conclusions that are not supported by the study results. FDA
has revised Sec. 99.101(a)(4) accordingly.
    30. One comment asked what FDA would do if an article discussed
multiple unapproved uses, but the manufacturer wanted to focus on just
one unapproved use.
    FDA expects that there may be articles that discuss multiple
unapproved uses and that such articles may be disseminated only if the
requirements are met for each of those uses. There also may be
instances when an article discusses multiple unapproved use(s), but
there is one (or more) predominant unapproved use(s) discussed in the
article. Under certain circumstances, it may make sense for the
manufacturer to have to meet the requirements set forth in this part
only for the predominant use(s). However, FDA will have to make this
determination on a case-by-case basis.
    31. One comment argued that dissemination of reference publications
is not consistent with the purpose of section 401 of FDAMA because, by
their very nature, reference publications are considerably out of date
at the time of their publication. The comment further opined that
because the authors do not report the methods used to assess the
current scientific literature, reference publications should be
considered the authors' opinion and thus, not scientifically sound.
    FDA agrees that many reference publications may not be up to date.
However, Congress did include reference publications within the scope
of section 401 of FDAMA. There is no basis to presume that all
reference publications are not scientifically sound.
    32. Several comments opposed the requirement that disseminated
information in the form of a reference publication ``pertain to a
clinical investigation regarding the drug or device.'' Instead, they
argued, the reference publication should ``include information about''
such a study. Some comments interpreted this to mean that the study
should meet all of the criteria to establish scientific soundness, but
the information about such a study should not be required. One comment
said that the language means that the information needs to be based on
a scientifically sound clinical investigation, it need not be about or
describe such clinical investigation.
    Both the act and this part provide that reference publications must
``include information about a clinical investigation.'' However, this
does not mean that the information about that clinical investigation
should be any less complete than the information included in a journal
article. It means only that the text may have a lot of additional
information that is not about the clinical investigation. The idea
behind the dissemination provision is that physicians and other
recipients be in a position to make treatment decisions based on
published reports of clinical trials. If the information that is
disseminated gives them little or no information about the actual
trial, then it would be difficult to argue that they have a reasonable
basis upon which to make such treatment decisions.
    33. A number of comments argued that the proposal has written
reference publications out of the statute by requiring the same level
of detail as would appear in journal articles. One comment said that
FDA should accept the dissemination of peer-reviewed reference
publications. Some comments argued that the proposal would make text
book dissemination more difficult than it was prior to passage of FDAMA
and that FDA should adopt a final rule that is consistent with its
existing reference text guidance or it should leave that guidance in
place. One comment argued that the statute makes it clear that FDA must
allow the dissemination of reference publications that meet the
requirements of the statute and that the agency's decision to issue a
guidance document on this issue is not an option.
    As set forth previously, FDA does not believe that Congress meant
that reference publications disseminated under this part could have
less detail about clinical investigations than journal articles. In
addition, reference publications are not subject to classic peer-
review. Therefore, FDA rejects the comment that FDA accept all peer-
reviewed reference publications. As discussed in the preamble to the
proposal, however, FDA recognizes that it will be difficult for many
reference publications to meet the statutory criteria. Moreover, as set
forth in many of the comments, the new statutory scheme in most
respects makes it more difficult to disseminate reference publications
than was possible before FDAMA. Thus, FDA plans to permit companies to
distribute unabridged reference publications (as defined in the statute
and Sec. 99.3(i)) without meeting all of the requirements set forth in
this part if the company does not focus on or point to a specific
unapproved use in the publication and it includes a disclaimer that the
publication includes information about unapproved uses. (Cf. Guidance
for Industry Funded Dissemination of Reference Texts (61 FR 52900,
October 8, 1996). The agency was enjoined from applying this guidance
document in WLF v. Friedman. FDA sought clarification on the scope of
the order in that case through a motion to amend the judgment.) FDA
plans to issue guidance on this issue at some time in the future
following clarification by the court. Of course, manufacturers that
want to focus or point to a specific unapproved use will have the
option of doing so by meeting the requirements set forth in this part.
    34. One comment argued that Congress intended for manufacturers to
be able to disseminate reference publication chapters.
    Section 552(a)(1) of the act clearly requires that the reference
publication be unabridged. A chapter from a textbook does not meet this
requirement.
    35. Proposed Sec. 99.101(b)(2) provided that journal articles and
reference publications disseminated under part 99 cannot be
disseminated with any information that is promotional in nature. One
comment strongly agreed with the concept of prohibiting promotional
material to be distributed with scientific information on a new use.
One comment opposed the concept, stating that there is no policy or
legal rationale for prohibiting companies from distributing information
on approved uses with these reprints. A number of comments requested
clarification of this statement. These comments were concerned that it
could preclude a sponsor from delivering a promotional piece on a
labeled use during the same office visit or detail. These comments
suggested that FDA

[[Page 64564]]

clarify that so long as the promotional material concerns an approved
use and is kept physically distinct from the unapproved use
information, FDA would not consider the two to be distributed together.
    FDA did not intend to prohibit a sponsor from delivering
promotional pieces on an approved or cleared use during an office visit
or detail in which it has delivered information on an unapproved use.
Any unapproved use information, however, must be kept physically
distinct from the promotional materials, and the sponsor may not
verbally promote the unapproved use or include materials about the
unapproved use, beyond those permitted or required under this part.
    b. Mandatory statements and information (Sec. 99.103). Proposed
Sec. 99.103 described the information that must accompany the journal
article or reference publication. For example, it required a
prominently displayed statement disclosing (among other things) that
the information being disseminated is about a use that has not been
approved or cleared by FDA and is being disseminated under section 551
et seq. of the act and, if applicable, a statement that there are
products or treatments that have been approved or cleared for the use
that is the subject of the dissemination. It also required the official
labeling and a bibliography of other articles to accompany the
disseminated information. In addition, the proposal described what is
meant by a ``prominently displayed'' statement by setting forth
criteria that are consistent with the agency's regulations on
prescription drug advertising (Sec. 202.1(e)(7)(viii)) and labeling (21
CFR 201.10(g)(2)). Proposed Sec. 99.103 required the statement that the
use has not been approved and the additional information required by
FDA to be attached to the front of the disseminated materials and that
all other mandatory information be attached to the disseminated
information.
    36. Although some comments supported FDA's position on mandatory
statements, there were others that thought the proposal was unduly
restrictive. For example, although some comments supported the
requirement for a uniform statement disclosing that the new use has not
been approved by FDA, there were a number of comments that thought
manufacturers should be allowed to use alternative language to convey
this message. One comment specifically objected to the phrase ``and is
being disseminated under section 551 of the Federal Food, Drug, and
Cosmetic Act.'' This comment said that the phrase was unnecessary and
could be confusing.
    FDA continues to believe that it is important to have a uniform
disclosure stating that the new use has not been approved by FDA.
Different statements can be confusing and recipients of the information
may believe that they have different meanings. FDA agrees, however,
that the phrase: ``and is being disseminated under section 551 et seq.
of the Federal Food, Drug, and Cosmetic Act'' is unnecessary and has
therefore dropped it from the final rule.
    37. One comment stated that clarification is needed regarding
articles that discuss more than one use because, as written,
Sec. 99.103(a)(1)(i) uses singular and plural forms in a way that is
confusing.
    FDA agrees that clarification was needed and has revised the final
rule accordingly.
    38. Proposed Sec. 99.103(a)(1)(iii) required a statement disclosing
any authors who have a significant financial interest in the
manufacturer. One comment noted that, although the disclosure is
appropriate, the final rule should make clear that such disclosure be
in line with the level required by the rule on financial disclosure and
should apply only to the financial interests at the time the study was
conducted and not the author's current interest.
    In the preamble to the proposed rule, FDA stated that an author
would have a significant financial interest in a manufacturer when
there is a relationship that may give rise to actual or perceived
conflicts of interest and that when there is a question as to whether a
relationship is significant, it should be disclosed (see 63 FR 31143 at
31147). Manufacturers may consult the final rule on financial
disclosure by clinical investigators (codified at 21 CFR part 54) to
learn the types of financial interests of greatest concern to the
agency. However, because the purposes and terminology of this final
rule and the final rule on financial disclosure by clinical
investigators are different, manufacturers should consult the
provisions of this final rule for the requirements that apply to
disclosures regarding authors. FDA agrees that the financial disclosure
should not necessarily apply to the author's current financial
interest. FDA believes, however, that it should apply to the author's
financial interests during the time the study was conducted up through
1 year after the time the journal article or reference publication was
written and published. FDA has revised the final rule to reflect this
time limitation. FDA's revision is consistent with part 54.
    39. One comment urged FDA to require that the statement that there
are products or treatments that have been approved or cleared for the
use that is the subject of the dissemination list the names of other
drugs that have been approved by FDA. Another comment asked whether
such statement should address adjuvant or supporting therapies.
    FDA's regulation tracks the statute, which does not require a
manufacturer to identify the specific products that have been approved
or cleared for the new use or the adjuvant or supporting therapy for
the new use. (See section 551(b)(6)(A)(v) of the act.) Although FDA can
see the benefit of having those specific product names listed, it would
be difficult to develop a complete and accurate list. Moreover, the
information could be misleading if the manufacturer merely provided a
list of names. FDA also does not believe that the statement should
address adjuvant or supporting therapies. The idea behind the
disclosure is to let health care practitioners and other recipients
know that approved/cleared alternatives exist. Therefore, FDA is
retaining the requirement that the manufacturer only disclose that such
approved/cleared products exist.
    40. Proposed Sec. 99.103(a)(2) provided that the manufacturer must
attach the official labeling of the product to the unapproved use
information. In the preamble to the proposed rule (63 FR 31143 at
31147), FDA noted that devices, unlike drugs, do not always include a
package insert in the same form and manner as drugs. Therefore, the
agency would expect device manufacturers to provide the same
information that is generally found in package inserts, namely: (1) The
name of the device, including its trade or proprietary name; (2) the
manufacturer's name, address, and telephone number; (3) a statement of
intended use, including a general description of the diseases or
conditions that the device is intended to diagnose, treat, cure, or
mitigate; (4) a description of the patient population for which the
device is intended; (5) a description of indications that have been
approved or cleared by FDA; (6) a description of any limitations or
conditions that have been placed on the sale, distribution, or use of
the device; and (7) all warnings, contraindications, side effects, and
precautions associated with the use of the device.
    One comment suggested that a device's official labeling be
interpreted as: (1) The package insert for the device; (2) the
accompanying documents that a manufacturer distributes with its legally

[[Page 64565]]

marketed device to comply with the requirements of 21 CFR 801 or 809.10
for in vitro diagnostic products; or (3) the new labeling vehicle
created by a manufacturer that contains the listed items from the
preamble.
    FDA agrees that this interpretation of official labeling for
devices is appropriate provided the third option is used only when the
first two options are not available or not feasible and provided the
third option includes only the information listed in the preamble
(i.e., no promotional statements or representations are included).
    41. Proposed Sec. 99.103(a)(3) required the manufacturer to attach
a bibliography of other articles (that concern reports of clinical
investigations both supporting and not supporting the new use). One
comment noted that a bibliography is not required every time--only when
one is not present in the disseminated information. Another comment
stated that the bibliography requirement is vague regarding what needs
to be included and under what circumstances a bibliography included in
the publication is sufficient.
    FDA's proposal provided that the manufacturer need not include a
separate bibliography if the disseminated information already includes
a bibliography that meets the requirements set forth in
Sec. 99.103(a)(3). The bibliography requirement would be met by a list
of all other published articles from scientific reference publications
or scientific or medical journals that discuss clinical investigations
and are specific to the new use discussed in the disseminated
information. The bibliography must include articles about clinical
investigations that both support and do not support the new use and it
must identify which articles relate to the new use. A bibliography
already included with the disseminated information would meet this
requirement only if it includes all other such published articles. The
manufacturer would still have to include its search strategy to show
that it took reasonable steps to ensure that the bibliography includes
all relevant published articles as described in Sec. 99.103(a)(3).
    42. Proposed Sec. 99.103(a)(4) required a manufacturer to include
any additional information required by FDA, including objective and
scientifically sound information pertaining to the safety or
effectiveness of the new use that FDA determines is necessary to
provide objectivity and balance, including information that the
manufacturer has submitted to FDA or, where appropriate, a summary of
such information, and any other information that can be made publicly
available; and an objective statement prepared by FDA, based on data or
other scientifically sound information, bearing on the safety or
effectiveness of the new use of the product.
    Several comments noted that this provision should specify that FDA
must provide the manufacturer notice and an opportunity to meet before
requiring such information.
    FDA agrees that a manufacturer must be provided notice and an
opportunity to meet before being required to include this additional
information. Redesignated Sec. 99.301(a)(2) provides this opportunity
and FDA has revised the final rule at Sec. 99.103(a)(4) to include a
reference to Sec. 99.301(a)(2).
    43. Several comments opposed the requirement that the statement
that the use has not been approved and the additional information
required by FDA be attached to the front of the disseminated materials
and that all other mandatory information be attached to the
disseminated information. One comment suggested that the FDA-required
information be attached to the back, and that FDA permit the use of a
sticker on the front of the disseminated material stating that the FDA-
required information is attached to the back.
    FDA believes that it is important to permanently affix the
statement indicating that the disseminated information is about an
unapproved use to the front of the materials. The recipients of such
materials should know, in advance, that they are reading information
about an unapproved use. However, FDA agrees that it could be
appropriate to attach the additional information required by FDA to the
back of the materials, provided there is a sticker or notation on the
front referring the recipient to that information. The agency has
amended Sec. 99.103(a)(4) accordingly.
    FDA also believes it is important to attach the remaining
information to the disseminated materials. Congress included this
mandatory information because it determined that it was important for
the recipient to receive it. If such information is not attached, it
can easily be separated from the disseminated material and never seen
by the recipient. This is the information that helps to ensure that the
disseminated materials are objective, balanced, and not misleading.
    44. Although some comments stated that the criteria in proposed
Sec. 99.103(c) for determining whether the mandatory information is
prominently displayed are appropriate, others opposed the factors that
FDA will consider in determining whether the mandatory information is
prominently displayed. The latter comments argued that manufacturers
should retain some flexibility and discretion in this area.
    FDA's approach is flexible. Section 99.103(c) sets forth the
factors that FDA will consider and provides that the required
statements shall be outlined, boxed, highlighted, or otherwise
graphically designed and presented in a matter that achieves emphasis
or notice and is distinct from the other information being disseminated
(emphasis added). Such an approach is not as proscriptive as the
comments imply. FDA has retained this approach in the final rule.
    45. One comment suggested that FDA permit manufacturers to post
information, such as balancing articles required by FDA, on the
Internet so long as the Internet address is prominently displayed on
the information that was disseminated. The comment said that this would
reduce paperwork burdens and provide a continuous source of current
information.
    FDA does not think that it would be appropriate for manufacturers
to use the Internet to balance a published reprint disseminated in hard
copy format or to provide recipients of unapproved use information with
only part of the information required by the statute and regulations.
The idea behind the provision was that physicians would receive, at one
time, a balanced package. Such balance would not be achieved if a
manufacturer could hand a physician an article and then advise the
physician that he/she has to take steps on his/her own to retrieve the
balancing information.
    46. Several comments urged FDA to require manufacturers to provide
patient labeling for drugs that are the subject of the disseminated
information. The comments noted that such labeling should identify the
drug by name, notify consumers that the drug has been promoted for an
unapproved use, and indicate FDA-approved uses for the drug. They
further argued that the patient labeling must include information about
the potential risks of the drug and meet the quality and content
standards of FDA's 1995 proposed Medication Guide rule. This comment
said that FDA-approved patient labeling must be in commercial
distribution at the level of the pharmacy before dissemination under
this part can begin. One comment stated that the labeling should state
that these products are not tested in certain populations and should
say ``use at your own risk.''

[[Page 64566]]

    FDA recognizes the importance of providing consumers access to
information about the products they use. Since 1968, FDA has
occasionally required and often encouraged manufacturers to produce
patient labeling for certain prescription drugs. However, the comments'
request for additional patient labeling on drugs that are the subject
of information disseminated under part 99 is outside the scope of
section 401 of FDAMA.
    47. Several comments argued that the lack of availability of
pediatric studies on a particular use should be clearly and prominently
stated in the information being disseminated to health professionals.
These comments also urged FDA to require an additional statement for
drugs that have not undergone pediatric testing: ``Safety and
effectiveness in pediatric populations have not been established for
this product for the use that has been approved by FDA or for the use
suggested by this information.''
    The suggestion that for drugs and devices that have not undergone
pediatric testing, the disseminated information should include a
statement to that effect is beyond the scope of this rule. However, for
unapproved pediatric uses that are the subject of the information being
disseminated, there will be a statement that the use has not been
approved or cleared by FDA.
    c. Recipients of information (Sec. 99.105). Proposed Sec. 99.105
identified who may receive information disseminated under this part.
Specifically, a health care practitioner, pharmacy benefit manager,
health insurance issuer, group health plan, or Federal or State
Government agency could receive information disseminated under part 99.
    48. Several comments urged FDA to add pharmacists to the list of
recipients of information under this part.
    As previously discussed, section 401 of FDAMA specifically lists
who can receive the unapproved use information under this provision. To
the extent that pharmacists are included in the definitions of ``health
care practitioner,'' ``pharmacy benefit manager,'' ``health insurance
issuer,'' or ``group health plan'' (see Sec. 99.3), they will be
included as recipients of this information.
3. Subpart C--Manufacturer's Submissions, Requests, and Applications
    a. Manufacturer's submission to the agency (Sec. 99.201). Proposed
Sec. 99.201 described the contents of a manufacturer's submission to
FDA. This submission would be made 60 days before disseminating
information on an unapproved or new use and would include items such as
a copy of all of the information to be disseminated, all other clinical
trial information that the manufacturer has relating to the safety or
effectiveness of the new use, any reports of clinical experience
pertinent to the safety of the new use, and, if a supplement for the
new use has not been submitted, a certification that the manufacturer
will submit a supplement or an application for an exemption from the
requirement to submit a supplement. The proposal also discussed what
types of information must be submitted when the certification provides
that the studies have been completed or that studies will be conducted
as well as the contents of the certification. Proposed Sec. 99.201 also
provided that the 60-day period begins to run when FDA receives a
complete submission.
    49. One comment agreed that manufacturers should have to submit any
clinical trial information that they have relating to the safety and
effectiveness of the new use. However, another comment argued that the
requirement for any clinical trial information is far more exhaustive
than that required by the statute.
    Section 551(b)(4)(B) of the act requires manufacturers to submit
``any clinical trial information the manufacturer has relating to the
safety or effectiveness of the new use, any reports of clinical
experience pertinent to the safety of the new use, and a summary of
such information.'' Proposed Sec. 99.201(a)(2) tracked this requirement
and described what it included. In the final rule, FDA is making clear
that, for effectiveness information, the requirements are limited to
information on clinical investigations of the new use; safety
information is broader and must include all relevant new data from
human experience.
    50. One comment urged FDA to require manufacturers to report only
those adverse experiences that they have received directly because
companies do not have access to the details of cases submitted to other
manufacturers and thus, are unable to evaluate the reports. That same
comment stated that FDA should permit adverse experience reports to be
submitted in summary or tabular form rather than as individual case
reports. Several other comments requested the ability to reference
files that FDA already has about adverse experiences. Finally, one
comment noted that the search requirements for adverse reports should
be more clearly delineated.
    Under the statute and these regulations, manufacturers would have
to submit only those adverse experience reports that they have. This
would include reports originally made to other manufacturers. If the
reports were originally submitted to other manufacturers and the
disseminating manufacturer does not know whether to attribute the
adverse experience to the new use, it should submit the information to
FDA. Manufacturers can submit adverse experience reports in summary or
tabular form if FDA already has the individual case reports. With
respect to search requirements for postmarket adverse event reports,
FDA does not think that it is necessary to be any more specific.
Manufacturers gather this information on a regular basis.
    51. One comment said that the literature search requirements in
Sec. 99.201(a)(3) should be more clearly delineated. Several comments
stated that the requirement for the submission of a search strategy is
not required by statute and should be eliminated because it is
unnecessary and burdensome and could delay the process.
    FDA believes that it is necessary to include the search strategy.
This is how FDA will be able to determine whether the bibliography
meets the statutory criteria. FDA has revised Sec. 99.201(a)(3),
however, to clarify the bibliography search strategy requirements.
    52. FDA, on its own initiative, revised Sec. 99.201(a)(4)(i)(B) and
(a)(4)(ii)(B) to clarify that, for purposes of computing time periods
that begin on the date of initial dissemination, FDA will look to the
date on which dissemination can begin. This clarification was necessary
because FDA will not know when a manufacturer actually begins to
disseminate materials. The same revision was made to Secs. 99.203(b)
and 99.401(b).
    53. Proposed Sec. 99.201(a)(4)(ii) required a manufacturer that has
planned studies that will be needed for a supplement to submit the
proposed protocols and schedule for conducting such studies. The
protocols must comply with FDA's IND or investigational device
exemption (IDE) regulations. One comment asked FDA to clarify whether a
manufacturer who has planned studies and wishes to disseminate
information must submit a complete IND or IDE in addition to the
information required in a submission under this rule. One comment
stated that if the protocols are to be treated as IND's, IDE's, or
amendments thereto, the manufacturer should be able to commence the
studies within 30 days unless the agency places the study on clinical
hold. The same comment said that if the agency does not place a

[[Page 64567]]

clinical hold on the protocol within 30 days, the agency should not be
able to determine that the protocols are inadequate on day 60 and if
the protocol is put on clinical hold within 30 days, it should not be
dispositive of the decision. The comment further stated that if the
agency decides that the protocols are adequate, it should be bound by
this decision and the final rule should reflect this. Finally, several
comments urged FDA to permit manufacturers to cross reference IND's and
IDE's rather than resubmitting such information.
    FDA intends that the protocols for planned studies under this
provision be submitted in compliance with the IND or IDE regulations.
However, a manufacturer will not be required to submit these materials
twice. If a protocol has already been submitted to an IND or IDE, the
IND or IDE can be cross referenced in the dissemination submission.
    Moreover, FDA does not intend to change, in any way, the IND or IDE
regulations, including the timeframes. If an IND or IDE is submitted
and a clinical hold is not issued within 30 days, the manufacturer can
commence the study or studies. However, the fact that FDA does not
issue a clinical hold within 30 days, does not prevent FDA from
determining, within 60 days, that a protocol is inadequate. FDA can
issue a clinical hold at any time after the 30-day period if the
requirements for issuing a clinical hold are met. If the protocol is
put on clinical hold within 30 days, it may not be dispositive of the
issue because the sponsor may remedy the reason for the clinical hold
within the 60-day period. However, if the reason for issuing the
clinical hold is not resolved, it will be dispositive of the issue.
Finally, FDA is declining to revise the rule to provide that if the
agency finds that the protocols are adequate, it will be bound by this
decision. FDAMA addressed the issue of agreements regarding the
parameters of the design and size of clinical trials. (See, e.g.,
section 505(b)(4)(C) or section 520(g)(7)(A) through (g)(7)(C) of the
act (21 U.S.C. 360j(g)(7)(A) through (g)(7)(C)).) FDA will abide by
these statutory directives.
    54. Proposed Sec. 99.201(a)(4)(ii) required a manufacturer that has
planned studies that will be needed for the submission of a
supplemental application for the new use to certify that it will
exercise due diligence to complete such studies and submit a supplement
within 36 months of dissemination. FDA has revised this section to
reflect the possibility that FDA may determine, before the
certification is submitted, that the studies needed to submit a
supplemental application cannot be completed and submitted within 36
months. This change is further reflected in Sec. 99.203.
    55. One comment requested that the 36-month timeframe for
submitting a supplement not override the time limits created under
separate regulatory or statutory authority. This comment was concerned
that if FDA finalizes its proposed 1997 regulation on pediatric
research and it includes compliance dates for completing the pediatric
studies that are less than 36 months, the 36-month period in this part
not override that shorter timeframe.
    As FDA has stated elsewhere in this document, nothing in this
regulation is meant to change or supersede other regulatory
requirements.
    56. One comment asked FDA to clarify the submission requirements
and FDA action requirements with respect to nonsignificant risk
devices.
    Protocols submitted for studies for devices considered to be
nonsignificant will be reviewed by FDA only to ensure that the protocol
for the study is consistent with the new use information to be
disseminated. Manufacturers must present the protocol for the
nonsignificant risk device study to an institutional review board (IRB)
for approval before starting the study. (See 21 CFR 812.1(b)(1).)
However, all reporting requirements under this part will apply to
nonsignificant risk device studies.
    57. One comment requested that the agency provide the sponsor an
opportunity to meet with FDA promptly to review what changes can be
made to the protocol to ensure that it meets requisite standards.
    Sections 505(b)(4)(B) and 520(g)(7)(A) and (g)(7)(C) of the act
provide sponsors with an opportunity to meet regarding their proposed
protocols. Therefore, no changes to this rule are necessary.
    58. One comment recommended that all statements submitted under
this part be certified by an officer from the manufacturer's executive
committee. Another comment recommended that the language in the
certification should include ``to the best of my knowledge'' to reduce
the risk that a certifying official could be penalized for an
inadvertent mistake not within his/her knowledge.
    The final rule requires that the manufacturer's attorney, agent, or
other authorized official sign the submission. Although an officer from
the manufacturer's executive committee may be an authorized official,
FDA does not think it is necessary for the submission to be signed by
such an officer. FDA also does not agree that it would be appropriate
to include the words ``to the best of my knowledge'' in the
certification. The attorney, agent, or other authorized official who
signs the submission and certification on behalf of the manufacturer,
and ultimately the manufacturer itself, is responsible for what is
submitted to the agency under this part.
    59. Proposed Sec. 99.201(c) described the component in each FDA
center that will receive a submission under this part. Several comments
noted that it would be appropriate for the review divisions in the
centers to also receive copies of the information submitted under this
part.
    In the final rule, FDA is retaining the requirement that the
submissions go to a single office within each center. Those offices
will forward the information to the appropriate review divisions within
the agency. The regulation need not spell out all of FDA's internal
procedures for processing these submissions.
    60. One comment stated that FDA needs to clarify the required
physical organization of the documents submitted under this part.
    FDA does not think it is appropriate to include that kind of detail
in this regulation. Nevertheless, FDA expects that materials in a
submission will be organized and labeled in accordance with the
submission requirements described in this part. If FDA subsequently
determines that manufacturers need more guidance in this area, it will
issue a guidance document.
    61. A number of comments objected to proposed Sec. 99.201(d), which
provided that the 60-day (post submission) period shall begin to run
when FDA receives a complete submission and that a submission shall be
considered complete if FDA determines that it is sufficiently complete
to permit a substantive review. These comments argued that FDA would
use this provision to extend the 60-day time period. The concern was
that FDA would, on day 59, advise a manufacturer that their submission
was not complete and therefore the 60-day time period had not begun.
The comments said that Congress meant for FDA to give a final answer
within the 60-day time period.
    As further described below, FDA is committing to give manufacturers
a final decision within 60 days. FDA has revised Sec. 99.201(d) to
provide that the 60-day period shall begin when FDA receives a
manufacturer's submission, including, where applicable, a

[[Page 64568]]

certification statement or an application for an exemption.
    62. A number of comments were made regarding the appropriateness of
public disclosure of information submitted under this part. Some
comments argued that both the fact of the submission and all
information in the submission is confidential and should not be
released. Other comments argued that all of the previous information
should be public because the public, including the patient community,
wants to be involved and has a right to know about a submission, the
data in such submission, FDA action on the submission, what studies are
being conducted, and the status of those studies. Several comments
argued that upon receiving a submission, FDA should publish in the
Federal Register, the citation for the article and the bibliography,
and solicit additional published information that might be appropriate
for dissemination. One comment argued that the public should have an
opportunity to comment prior to FDA's granting approval for
dissemination of information and that FDA should hold an advisory
committee meeting and let the public participate in its decision on
whether an exemption from the requirement to submit a supplement should
be granted.
    FDA declines to amend the rule to require a notice and comment
process before permitting dissemination to proceed or before granting
an exemption. However, the Freedom of Information Act (FOIA) and FDA's
regulations will dictate what information submitted under this
provision can be disclosed. Because the agency was required to issue
this regulation within such a short period of time, it has been unable
to fully examine all issues of disclosability. However, the agency will
continue to examine these issues separately.
    b. Request to extend the time for completing planned studies
(Sec. 99.203).   Section 554(c)(3) of the act (21 U.S.C. 360aaa-3)
describes two types of extensions of time regarding planned studies.
Section 554(c)(3)(A) of the act provides that the 36 month period for
completing planned studies and submitting a supplemental application
may be extended by the Secretary of Health and Human Services (the
Secretary) if the Secretary determines that the studies needed to
submit such application cannot be completed and submitted within 36
months. This type of extension would be granted before such studies are
begun. Section 554(c)(3)(B) of the act provides that the period for
completing planned studies and submitting a supplemental application
may be extended by the Secretary if the manufacturer submits a written
request for the extension and the Secretary determines that the
manufacturer has acted with due diligence to conduct the studies in a
timely manner. The latter extension cannot exceed 24 months. Proposed
Sec.  99.203 set forth the procedures that a manufacturer must follow
to request an extension of time for submitting a supplemental
application and the content of a request for an extension. The
provision covered only the extension in section 554(c)(3)(B) of the
act.
    63. The comments to this provision indicated that there was some
confusion regarding the two different statutory procedures. Several
comments asked FDA to more clearly set out the two procedures
contemplated by the statute.
    Although the statute specifically refers to a manufacturer request
in connection only with the procedure described in section 554(c)(3)(B)
of the act and FDA agrees that the agency can, under section
554(c)(3)(A), on its own initiative determine before the studies have
begun that more than 36 months is needed, FDA believes that
manufacturers will come to FDA and ask FDA to make a determination
under section 554(c)(3)(A) of the act. Therefore, FDA has revised
Sec. 99.203 to establish procedures for the two different types of
extensions. The first extension, set forth in Sec. 99.203(a), relates
to a request for an extension by the manufacturer at or before the time
it submits its dissemination package to FDA because the 36-month period
is not enough time to complete a study or studies of the new use and
submit a supplemental application. Revised Sec. 99.203(b) sets forth
the procedures that a manufacturer must follow to request an extension
of time for submitting a supplemental application after a study has
begun and the content of a request for an extension.
    c. Application for exemption from the requirement to file a
supplemental application (Sec. 99.205). Proposed Sec. 99.205 set forth
what a manufacturer must submit when seeking an exemption from the
requirement to file a supplemental application for a new use for
purposes of disseminating information on that new use. It required the
manufacturer to include an explanation as to why an exemption is sought
and to include materials demonstrating that it would be economically
prohibitive or unethical to conduct the studies needed to submit a
supplemental application for the new use.
    64. A number of comments supported the standards that FDA proposed
to determine whether it would be economically prohibitive or unethical
to conduct the studies needed to submit a supplemental application.
Some noted that FDA's standards are consistent with congressional
intent that exemptions be limited in scope and infrequent or rare. One
comment argued that pediatric exemptions should be extremely rare. One
comment stated that exemptions should never be granted.
    FDA agrees that Congress intended that exemptions from the
requirement to file a supplemental application for a new use be granted
in limited circumstances (see H. Conf. Rept. No. 399, 105th Cong., 1st
sess. at 100 (1997); 143 Congressional Record S9,837 (daily ed. Sept.
24, 1997) (Statement of the Managers)). There is nothing in the statute
or legislative history that gives FDA authority to apply a different
standard in the case of pediatric exemptions. Moreover, the act
provides for exemptions, so FDA does not agree that such exemptions
should never be granted. In light of the comments received to the
standards set forth in its proposal (discussed in more detail below),
FDA is adopting a different standard for the economically prohibitive
exemption. Although, FDA is not changing the standard for the unethical
exemption, it has, as discussed in the following paragraphs, clarified
how it will apply that exemption.
Economically Prohibitive Exemption
    Under proposed Sec. 99.205(b)(1), a manufacturer seeking an
exemption from the requirement to file a supplemental application on
the basis that it would be economically prohibitive to conduct the
needed studies would have to: (1) Explain why existing data, including
data from the scientifically sound study described in the information
to be disseminated, are not adequate to support approval of the new
use; and (2) show, at a minimum, that the estimated cost of the
necessary studies would exceed the estimated total revenue from the
product minus the cost of goods sold and marketing and administrative
expenses attributable to the product and that there are not less
expensive ways to obtain the needed information.
    Proposed Sec. 99.205(b)(1) set forth the type of evidence that the
manufacturer would have to include to meet the requirements for an
economically prohibitive exemption. These included:
    (1) A description of the current and projected U.S. patient
population for the product and an estimate of the current and projected
economic benefit to the

[[Page 64569]]

manufacturer from the use of the drug or device in this population. The
estimate would assume that the total potential market for the drug or
device is equal to the prevalence of all of the diseases or conditions
that the drug or device will be used to treat and involve the following
considerations: (a) The estimated market share for the drug or device
during any exclusive market period, a summary of the exclusive market
period for the product, and an explanation of the basis for the
estimate; (b) a projection of and justification for the price at which
the drug or device will be sold; and (c) comparisons with sales of
similarly situated drugs or devices, where available.
    (2) A description of the additional studies that the manufacturer
believes are necessary to support the submission of a supplemental
application for the new use and an estimate of the projected costs for
such studies; and
    (3) An attestation by a responsible individual of the manufacturer
verifying that the estimates included with the submission are accurate
and were prepared in accordance with generally accepted accounting
procedures. The data underlying and supporting the estimates shall be
made available to FDA upon request.
    65. As set forth previously, some of the comments agreed with FDA's
construction of ``economically prohibitive'' These comments argued that
such exemptions should be granted rarely and that the criteria for such
an exemption should be rigorous. One comment argued that the cost for
the studies should substantially exceed revenues to qualify for the
exemption. Several comments opposed such an equation.
    FDA agrees that exemptions should be granted only in limited
circumstances. As set forth below, however, FDA was convinced by the
comments that the standard set forth in its proposal was inappropriate
and has revised the standard.
    66. A number of comments objected to how the agency proposed to
determine what is economically prohibitive. First, they objected to the
agency's use of the term ``rare'' in describing when such exemptions
would be granted. One comment opined that Congress meant for the
exemption to arise in a ``fair number of circumstances.'' Second, they
objected to the absence of the criteria listed in the statute and
report language from the standard set forth in the codified regulation.
Third, they claimed that the proposed rule's standard for determining
what is economically prohibitive is too high.
    One comment argued that the exemption should be granted if it does
not make economic sense to pursue a supplement. Others argued that it
should be based on the revenue from the new use, not all uses of the
product. Some argued that the standard should be whether the cost of
the studies would exceed the revenues from the new use; others argued
that it should be whether the cost of the studies exceeds the new use
revenues that resulted from approval of the supplement (i.e., the
increase in revenues from the new use that result from submission of
the supplement). Several comments argued that FDA should automatically
grant an exemption if the new use is for a rare disease or condition
because for such use there is no reasonable expectation that the cost
of developing and making available a drug for such disease will be
recovered from sales in the United States of such drug. Several
comments argued that the economically prohibitive exemption should
automatically be granted if: (1) There is no market exclusivity for the
product (from patent, orphan drug status, or Waxman-Hatch); or (2) the
patient population likely to be served by the new indication will not
exceed an established number (e.g., 1,000). One comment opined that
interpreting ``prohibitive'' to mean anything other than the point at
which an economically rational company will not pursue research ignores
the needs of patients with rare disorders.
    FDA agrees that Congress did not use the term ``rare'' in the
legislative history. Nevertheless, Congress did state that exemptions
to the requirement to submit a supplement would be appropriate only in
``limited circumstances,'' which in FDA's view implies fewer than in a
``fair number of circumstances.'' Moreover, Congress strongly
emphasized the critical importance of getting information about new
uses onto the label. Although FDA did not include the criteria listed
in the statute and the legislative history in the standard for
economically prohibitive, they were included as types of evidence that
would be required to support the exemption.
    FDA's proposed criterion did not focus solely on sales from the new
use because the agency believed that there might be many circumstances
where the cost of the study requirements would exceed the sales from
just the new use. The agency explained that in some of these
situations, even if it were not economically ``wise'' to conduct the
studies, the cost would not rise to the level of being ``prohibitive.''
This view was judged consistent with the legislative history, which
foresaw the granting of economic exemptions only in limited
circumstances. The agency noted, however, that defining a practical
``economically prohibitive'' exemption was particularly troublesome,
because it would be so difficult for the agency to assess cost and
income projections. In view of these difficulties, FDA acknowledged
that it was not certain that the proposed approach was optimal and
sought comment on other possible ways to define economically
prohibitive.
    Unfortunately, the agency has received widely conflicting public
comment on this issue and remains uncertain about the elements of a
standard that would be most appropriate and effective in achieving the
statutory goals. An approach that would grant automatic exemptions if:
(1) The new use were for a rare disease or condition; (2) there was no
market exclusivity for the product (from patent, orphan drug status, or
Waxman-Hatch); or (3) the patient population likely to be served by the
new indication would not exceed an established number (e.g., 1,000)
would be inappropriate. Neither the statute nor the legislative history
provide for automatic exemptions in these circumstances. Rather, they
direct FDA to take both market exclusivity and population size into
account. The legislative history made clear that the size of the
patient population would not necessarily justify an exemption. In fact,
the legislative history stated that an exemption based on the size of
the patient population was intended to be the exception rather than the
rule in cases of populations suffering from orphan or rare diseases or
conditions. The legislative history made clear that FDA should consider
the importance of getting products for these diseases or conditions
approved. It noted that for many years, Congress has sought to
encourage research into orphan diseases and support the approval of
innovative drugs for their treatment. Congress, therefore, has directed
FDA to recognize the vital importance of encouraging applications for
new products intended to treat rare diseases and to examine very
carefully whether an exemption from filing a supplemental application
might hinder such research (see H. Conf. Rept. No. 399, 105th Cong.,
1st sess. at 100 (1997); H. Rept. No. 310, 105th Cong. 1st. sess. at 62
(1997)).
    Because the agency remains uncertain about the elements of a
standard that would be most appropriate and effective, FDA plans to
continue its search for a policy that would satisfy the congressional
expectation of approving exemptions in only limited

[[Page 64570]]

circumstances, without foreclosing the dissemination of useful
information by firms that could not otherwise conduct the needed
studies. In the meantime, FDA will implement the statute by basing its
evaluation of each exemption on a case-by-case determination of whether
the cost of the study for the new use reasonably exceeds the total
expected revenue from the new use minus the cost of goods sold and
marketing and administrative expenses attributable to the new use of
the product. This standard may not always meet a strict profitability
criterion because it considers all new use revenues, rather than just
the new use revenues that would result from approval of the supplement.
Nevertheless, it is consistent with most of the comments submitted by
the affected industry on this issue, it is consistent with the
statutory directive, and it attempts to strike a fair balance between
assuring the widest possible information dissemination while granting
economic exemptions only in ``limited circumstances.''
    The final rule sets forth the statutory standard and the
information that FDA would need to make this case-by-case
determination. This will include information about: (1) The cost of the
study for the new use; (2) the expected patient population for the new
use; (3) the expected total revenue for the new use minus the cost of
goods sold and marketing and administrative expenses attributable to
the new use of the product; (4) the amount of exclusivity for the drug
or new use; and (5) other information that the manufacturer believes
demonstrates that conducting the studies on the new use would be
economically prohibitive.
    As this revised criterion may significantly expand the number of
exemption applications beyond that anticipated by the Congress, the
agency is determined to review its experience with these requests as
they are submitted and, if necessary, to contract with outside economic
experts to help develop an approach that most appropriate and effective
and workable for the agency.
    67. A number of comments objected to the requirement to submit
detailed financial data. These comments argued that manufacturers
should be not required to submit highly sensitive and proprietary
information. Others felt that FDA is not qualified to review and
evaluate this data.
    Congress directed FDA to grant an economic exemption only upon
making a determination that conducting the studies and submitting a
supplement would be economically prohibitive. FDA cannot make this
determination without examining the relevant company data. Therefore,
the final rule retains these requirements.
    68. Several comments regarding FDA's approach to economic
exemptions recommended that FDA require a manufacturer to submit a
certified public accountant's (CPA's) opinion on the economic
feasibility of filing a supplemental NDA. FDA could contest the claim
by providing a CPA's statement to the contrary.
    FDA declines to adopt this approach because it removes the agency
from the statutorily-specified role of determining whether it would be
economically prohibitive to conduct the studies.
    69. One comment recommended that manufacturers be given the
flexibility to present whatever information they determine is relevant
to the ``economically prohibitive'' factor, that the manufacturer be
able to use its own assumptions, and that each situation be evaluated
on a case-by-case basis.
    As set forth previously, FDA is adopting a case-by-case
determination and has specified the information that is essential for
this determination. Nevertheless, manufacturers are free to provide
whatever additional information they think is relevant to the
determination. This could include information that would explain why a
study is so expensive to conduct. For example, one factor might be the
difficulty of enrolling patients in a clinical investigation if the new
use has become the standard of care.
    70. Proposed Sec. 99.205(b)(1)(ii)(A) stated that the estimated
economic benefit for a drug or device shall assume that the total
potential market is equal to the prevalence of the disease(s) or
condition(s) that such product will be used to treat. Several comments
argued that this assumption should be deleted because the potential
market for the drug or device may be less than the prevalence of the
disease in question if other therapies are likely to be used in some
portion of the total patient population.
    FDA agrees that this assumption should be deleted and has done so
in the final rule.
    71. One comment argued that the manufacturer should not be required
to provide a ``justification'' of the price at which the drug will be
sold. According to this manufacturer, only a projection is relevant.
    FDA has to be able to determine whether the manufacturer's proposed
price is reasonable. It may be that ``justification'' for the price is
not appropriate. Therefore, in Sec. 99.205(b)(ii)(C) of the final rule,
FDA will seek an explanation of the price at which the drug or device
will be sold.
    72. One comment opined that permitting an exemption because of cost
is an ethical decision because it is placing a monetary value on
people's lives and safety.
    FDA does not agree that an economically prohibitive exemption is
placing a monetary value on people's lives and safety. The standard in
FDA's regulation is intended to best effectuate the goals of the
statute.
    73. Proposed Sec. 99.205(b)(1)(ii)(C) required a manufacturer to
provide an attestation by a responsible individual of the manufacturer
verifying that the estimates included with a submission are accurate
and were prepared in accordance with generally accepted accounting
procedures. In addition, the data underlying and supporting the
estimates would have to be made available to FDA upon request. In the
preamble to the proposed rule, FDA noted that it had considered
requiring a report of an independent CPA with respect to the estimates
and FDA solicited comment on whether such a report should be required
in lieu of, or as an alternative to, the attestation that would be
required by the proposal.
    Some comments supported the submission of the CPA report discussed
previously, others felt that such a report should not be required.
Still other comments stated that the CPA report should be submitted in
lieu of the underlying data or that the CPA should make the
determination of economic feasibility instead of FDA.
    As stated previously, FDA refuses to adopt a procedure by which it
surrenders decision making to a CPA. However, FDA is not convinced that
it is necessary to require a report of an independent CPA with respect
to the estimates. Under Sec. 99.205(b)(1)(iii), therefore, FDA will
accept either an attestation by a responsible individual of the
manufacturer or by a CPA verifying that the estimates included with a
submission are accurate and were prepared in accordance with generally
accepted accounting procedures.
Unethical Exemption
    Proposed Sec. 99.205(b)(2) required a manufacturer seeking an
exemption on the basis that it would be unethical to conduct the
studies needed to submit a supplement, to: (1) Explain why existing
data, including data from the scientifically sound study described in
the information to be disseminated, are not adequate to support
approval of the new use; and (2) show that, notwithstanding the
insufficiency of existing data to support the submission of a
supplemental application for the

[[Page 64571]]

new use, the data are persuasive to the extent that withholding the
drug or device in the course of conducting a controlled study would
pose an unreasonable risk of harm to human subjects.
    The proposed codified language provided that an unreasonable risk
of harm would ordinarily arise only in situations in which the new use
of the drug or device appears to affect mortality or irreversible
morbidity. Evidence suggesting that the drug or device is the standard
of care for the new use can add weight to an argument that conduct of a
needed study or studies would be unethical.
    To support its argument that the conduct of a needed study or
studies would be unethical, the proposal provided that a manufacturer
would need to address the possibility of conducting studies in
different populations or of modified design (e.g., adding the new
therapy to existing treatments or using an alternative dose if
monotherapy studies could not be conducted).
    The proposal further provided that in assessing the appropriateness
of conducting studies to support the new use, the manufacturer may
provide evidence that the new use represents standard medical treatment
or therapy. Evidence that the new use represents standard medical
therapy can be one element of an argument that studies cannot ethically
be conducted, but the persuasiveness of available data is equally
important. Evidence that the new use represents standard medical
therapy might be obtained from a number of different sources. The
preamble to the proposal set forth the following possible
considerations: (1) Whether the new use meets the requirements of
section 1861(t)(2)(B) of the Social Security Act, which defines
``medically accepted indications'' with respect to the use of a drug;
(2)  Whether a medical specialty society that is represented in or
recognized by the Council of Medical Specialty Societies (or is a
subspecialty of such society) or is recognized by the American
Osteopathic Association has found that the new use is consistent with
sound medical practice; (3)  Whether the new use is described in a
recommendation or medical practice guideline of a Federal health
agency, including the National Institutes of Health, the Agency for
Health Care Policy and Research, and the Centers for Disease Control
and Prevention of the Department of Health and Human Services; and (4)
Whether the new use is described in a current compendia such as the
United States Pharmacopoeia Drug Information for the Health Care
Professional, the American Medical Association Drug Evaluations, or the
American Hospital Formulary Service (see 63 FR 31143 at 31150).
    74. A number of comments objected to FDA's proposed criteria for
the unethical exemption--particularly the emphasis on the requirement
that it ordinarily would arise only in situations in which the new use
appears to affect mortality or irreversible morbidity. Some comments
believed that the criteria set forth in the legislative history (that
are discussed in the preamble) should be in the codified language.
Finally, a number of comments argued that if the new use is the
standard of medical care, FDA must automatically grant an exemption.
    The act clearly does not require FDA to automatically grant an
exemption if a new use is the standard of medical care. The act says
that FDA must consider (among other considerations that the Secretary
finds appropriate) whether the new use is the standard of medical care,
and that is what FDA proposed to do. Moreover, an automatic exemption
would not be reasonable from a scientific standpoint because there are
many instances in which the results of a controlled clinical trial have
demonstrated that a drug or device is unsafe or ineffective for a new
use for which it is considered to be the standard of care.
    The standard set forth in Sec. 99.205 is consistent with how FDA
determines what studies are unethical in other contexts (i.e., when a
manufacturer argues that it would be unethical to conduct a study).
Moreover, the standard is consistent with the legislative history,
which provides that such exemptions should be granted in limited
circumstances. Therefore, FDA is retaining the proposed basic standard
for the unethical exemption in the final rule (i.e., the data are
persuasive to the extent that withholding the drug or device in the
course of conducting a controlled study would pose an unreasonable risk
of harm to human subjects). FDA continues to believe an effect on
irreversible morbidity or mortality is what ordinarily would be
required to show an unreasonable risk of harm. Nevertheless, there
could be other circumstances in which the agency would find that it
would be unethical to do the study, i.e., because there would be an
unreasonable risk of harm even though the new use does not affect
irreversible morbidity or mortality. In making a determination that it
would be unethical to conduct a study, the agency must consider whether
informed consent and proper IRB review would address the concerns
raised by questions about whether it is appropriate to conduct a study.
    FDA rejects the suggestion that the factors set forth in the
legislative history that FDA may consider in deciding whether to grant
an exemption be included as requirements in the codified language. FDA
has included the statutory factors in the codified language. The
legislative history provides that FDA may consider those factors among
other factors, and thus, consideration of these factors is neither
mandatory nor is it exclusive.
    75. One comment argued that the standard needs to take into account
the difficulty of enrolling patients in a study in which some subjects
will receive a placebo when a patient can go to a doctor and receive a
prescription for the drug. The comment further noted that physicians
refuse to participate in placebo controlled studies of therapies they
already believe to be effective.
    FDA agrees that it can be difficult to enroll patients in placebo
controlled trials and that this could be a relevant consideration.
Moreover, not all controlled studies are placebo controlled. Companies
may be able to conduct studies of a different design, depending on the
situation. For example, a company may be able to compare the new use to
another therapy that is known to work or may be able to rely on
historical controls. In some cases, the new use could be added to
existing therapy and compared with placebo added to existing therapy.
If these alternate study designs mean that the study or studies will
take longer, FDA can consider whether to extend the time to conduct the
studies and submit a supplemental application.
    76. One comment suggested that FDA should grant an exemption if the
new use is listed in the USP DI or Hospital Formulary. Another comment
suggested that an unethical exemption should be granted if the
unapproved use: (1) Is accepted in a monograph of the USP; (2) is
approved by another ``first world'' country; or (3) is approved by a
state FDA. Finally, one comment suggested that FDA should automatically
grant an unethical exemption if the new use: (1) Represents the
standard of care, as represented by inclusion in specified compendia or
practice guidelines, or (2) involves a combination of products or more
than one sponsor and should grant other exemptions on a case-by-case
basis.
     FDA does not agree that any of these individual factors is enough
to show that studying a new use would be unethical. Moreover, there is
nothing in the statute or legislative history to

[[Page 64572]]

suggest that any of the single factors should be sufficient to meet the
unethical exemption. FDA will, however, consider these factors in
making its determination of when it would be unethical to conduct a
study.
    77. One comment noted that, although it supported the list of
sources to be used to provide evidence that a new use represents
standard medical therapy, after 1998, the American Medical
Association's (AMA's) Drug Evaluation and the USP DI may no longer be
available.
    If the AMA's Drug Evaluation and/or the USP DI become unavailable,
FDA will stop using them as evidence that a new use is the standard of
care.
    78. One comment noted that there are diverse opinions in the
medical community about what standard of care means. Another noted that
``consistent with sound medical practice'' is not the same as
``standard of care'' and that an unapproved treatment may be considered
to be sound medical practice but should still be studied. Several
comments noted that FDA should take care in how it interprets
``standard medical treatment or therapy.'' These comments noted that
manufacturers should not be allowed to take advantage of a situation of
their own creation. In other words, standard medical treatment should
not be interpreted as meaning treatment that is regularly used because
physicians have no other choice because to do so would eliminate the
requirements for completing any pediatric research.
    FDA agrees that just because a certain treatment is consistent with
sound medical practice does not mean that it is the standard of care.
FDA has stated that whether a medical specialty society that is
represented in or recognized by the Council of Medical Specialty
Societies (or is a subspecialty of such society) or is recognized by
the American Osteopathic Association has found that a new use is
consistent with sound medical practice will be considered as evidence
that it is the standard of care. Moreover, just because an unapproved
use of a drug or device is the standard of care, does not mean that it
is automatically exempt from the requirement to conduct the study
needed to submit a supplemental application.
    79. Several comments noted that it is almost inconceivable that the
study of a new use for children could be viewed as unethical.
    FDA will make this determination on a case-by-case basis.
    80. Several comments argued for making the exemption process
public. One comment said that all information should be made public as
soon as a manufacturer requests an exemption and that if an exemption
is granted all information should remain in the public domain so that
interested parties will be able to play a role in keeping FDA informed
as to when it should be revoked. Another suggested that prior to
granting any exemption, FDA should hold a meeting of the appropriate
advisory committee so that the public has the opportunity to review and
comment upon the request.
    As set forth previously, FDA declines to adopt a notice and comment
process for considering exemption requests. The information will be
made available to the public consistent with FOIA and FDA's
regulations. FDA has the option of consulting advisory committees about
exemption requests, when appropriate.
4. Subpart D--FDA Action on Submissions, Requests, and Applications
    a. Agency action on a submission (Sec. 99.301). Proposed
Sec. 99.301 described the range of FDA's actions when it receives a
submission. For example, under the proposal, FDA could determine that a
manufacturer's submission does not comply with the regulatory
requirements, request additional information or documents to assist the
agency in determining whether the information to be disseminated
complies with applicable requirements, or determine that the
information fails to provide data, analyses, or other written matter
that is objective and balanced. The proposal also described FDA actions
in response to a manufacturer's submission when the manufacturer is
committing to submit a supplement on completed studies or is agreeing
to conduct the necessary studies and then submit a supplement.
    81. Proposed Sec. 99.301(a) provided that, within 60 days, FDA may
determine that a submission does not comply with the requirements of
the proposal or that it needs more information. A number of comments
objected to the proposal because they believed that FDA would use it to
extend the 60-day time period. The concern was that FDA would, on day
59, advise a manufacturer that their submission was not complete and
therefore the 60-day time period had not begun. The comments said that
Congress meant for FDA to give a final answer within the 60-day time
period. Some comments argued that FDA should let manufacturers know if
their submission is complete within a short period of time, e.g.,
within 15 days of receiving the submission.
    In response to these comments, FDA has eliminated proposed
Sec. 99.301(a)(2) so that manufacturers will have a final decision
within 60 days. Within the 60-day period, FDA will either notify a
manufacturer that it has not met the requirements set forth in the law
or allow the dissemination to go forward. FDA is not adopting the
comment's suggestion that it advise sponsors as to whether their
submissions are complete within a certain number of days (e.g., 15).
The 60-day statutory timeframe is too short for the agency to make a
commitment to provide such advice.
    82. One comment stated that FDA should be required to notify the
manufacturer promptly if it approves a submission in less than 60 days.
    There is no requirement in the statute that FDA notify a
manufacturer unless it intends to stop the dissemination of information
under this part. Therefore, FDA is not revising the regulation as
suggested. The agency will, however, make an effort to notify
manufacturers promptly if it approves a submission in less than 60
days.
    83. One comment requested that FDA change the ``may'' in proposed
Sec. 99.301(a) to ``shall'' and to clarify that a sponsor may begin to
disseminate material if it has not heard from FDA within 60 days.
Another comment suggested that FDA clarify Sec. 99.301 to indicate that
FDA will review an IND or IDE and will notify the manufacturer of the
IND or IDE approval and that, until such notification, the manufacturer
cannot disseminate the information.
    FDA declines to change the ``may'' to ``shall'' in Sec. 99.301(a).
FDA is not required to do any of the