FTC Comment Before the Food and Drug Administration Concerning Applications for FDA Approval to Market a New Drug (Relating to Abbreviated New Drug Application Procedures)

BEFORE THE
DEPARTMENT OF HEALTH AND HUMAN SERVICES
FOOD AND DRUG ADMINISTRATION
In the Matter of
Applications for FDA Approval to Market a New Drug; Patent Listing Requirements and Application
of 30-Month Stays on Approval of Abbreviated New Drug Applications Certifying That a Patent
Claiming a Drug is Invalid or Will Not be Infringed
Docket No. 02N-0417
Comments of the
United States Federal Trade Commission
December 23, 2002

1 See Generic Drug Entry Prior to Patent Expiration: An FTC Study (Jul. 2002),
available at <http://www.ftc.gov/os/2002/07/genericdrugstudy.pdf>.
2 Department of Health and Human Services, Food and Drug Administration, Applications for
FDA Approval to Market a New Drug; Patent Listing Requirements and Application of 30-
Month Stays on Approval of Abbreviated New Drug Applications Certifying That a Patent
Claiming a Drug is Invalid or Will Not be Infringed, 67 Fed. Reg. 65,448 (Oct. 24, 2002).
2
I. INTRODUCTION
On July 30, 2002, the Federal Trade Commission released a comprehensive study that described
several industry practices that delay FDA approval of generic drug products.1 The FTC Study included
legislative recommendations to address the possibility of future abuses of the generic drug approval process
governed by the Hatch-Waxman Amendments. Chief among these recommendations was a proposed
limitation of only one automatic 30-month stay per drug product per abbreviated new drug application
(ANDA) to resolve infringement disputes over patents listed in the Orange Book prior to the filing date of the
generic applicant’s ANDA. On October 24, 2002, the FDA released this proposed rulemaking to eliminate
the multiple 30-month stays that the FTC Study had identified as most harmful to consumers.
In this proceeding, the Food and Drug Administration (FDA) has requested comment2 on those
proposals to amend its regulations governing the availability of, and triggers for, the 30-month stay provision
of the Hatch-Waxman Amendments as suggested by the FTC Study. Specifically, the FDA proposes: (1)
to amend its existing rules to state that there will be one and only one opportunity for a 30-month stay of
FDA approval of each abbreviated new drug application (ANDA); (2) to clarify the types of patents that must
and must not be listed in the Orange Book; and (3) to revise the declaration statement that new drug
application (NDA) applicants must submit as part of an NDA, an amendment to an NDA, or a supplement to
an NDA.
The FDA's proposal limiting one 30-month stay opportunity per ANDA, although not identical to the
FTC Study’s recommendation, is an important reform that would eliminate a substantial portion of the
potential for unwarranted delay of FDA approval of generic drugs identified by the FTC Study. Consumers
should benefit significantly from earlier market entry of generic versions of brand-name drug products that
are sold at substantial discounts from brand-name drug product prices.
The FDA’s proposals, which clarify the types of patents that can be listed in the Orange Book to
trigger the 30-month stay, provide needed guidance to the industry about patent listing issues. This
comment offers additional suggestions to refine the proposed patent listing and declaration requirements.
These suggestions would have heightened importance should legal challenge render the FDA’s proposed
limitation to one 30-month stay vulnerable while, at the same time, not disturbing the proposed listing
regulations (which could potentially open the door for multiple 30-month stays based on inappropriately
listed patents). In particular, we believe that the proposed requirement that brand-name companies must
list polymorph patents (and certain product-by-process patents, depending upon how the provision is
interpreted) increases, rather than decreases, the potential for delayed market entry by generic drug
products. The comment also suggests how to address the problem of “double patenting” identified in the
FTC Study, but that is not addressed in the FDA’s proposals.
II. BACKGROUND: HOW GENERIC DRUGS OBTAIN FDA APPROVAL

3 21 U.S.C. § 355(b)(1).
4 Id. at § 355(j)(7)(A).
5 Id. at § 355(j)(2)(A)(ii).
6 Id. at § 355(j)(2)(A)(iv).
7 Id. at § 355(j)(2)(A)(vii).
8 Id. at ' 355(j)(2)(A)(vii)(IV).
9 Id. at § 355(j)(2)(B). Although the patent holder and the NDA filer are often the same
person, this is not always the case. Hatch-Waxman requires that all patents that claim the drug
described in an NDA be listed in the Orange Book. Occasionally, this requires an NDA filer to list a
patent that it does not own.
10 Id. at § 355(j)(5)(B)(iii).
11 Id. For example, the statute requires the ANDA applicant to establish bioequivalence. Id. at
§ 355(j)(2)(A)(iv).
3
A brand-name drug manufacturer seeking to market a new drug product must first obtain FDA
approval by filing a New Drug Application (“NDA”). At the time the NDA is filed, the NDA filer must also
provide the FDA with certain categories of information regarding patents that cover the drug that is the
subject of its NDA.3 Upon receipt of the patent information, the FDA is required to list it in an agency
publication entitled “Approved Drug Products with Therapeutic Equivalence,” commonly known as the
“Orange Book.”4
To obtain approval of a generic version of a brand-name drug, Hatch-Waxman requires a generic
applicant to file an Abbreviated New Drug Application (“ANDA”). Under the ANDA process, an applicant
must demonstrate that the active ingredient of the generic drug is “the same” as that of the relevant brand-
name product,5 and also show that the generic drug product is “bioequivalent” to the relevant brand-name
product.6 The ANDA must contain, among other things, a certification regarding each patent listed in the
Orange Book in conjunction with the relevant NDA.7 One way to satisfy this requirement is to provide a
“Paragraph IV certification,” asserting that the patent in question is invalid or not infringed.8
Filing a Paragraph IV certification potentially affects two regulatory provisions – the 30-month stay
provision and the 180-day marketing exclusivity provision. An ANDA filer that makes a Paragraph IV
certification must provide notice to both the patent holder and the NDA filer, including a detailed statement
of the factual and legal basis for the ANDA filer’s assertion that the patent is invalid or not infringed.9 Once
the ANDA filer has provided such notice, a patent holder wishing to take advantage of the statutory stay
provision must bring an infringement suit within 45 days.10 If the patent holder does not bring suit within 45
days, the FDA may approve the ANDA as soon as other regulatory conditions are fulfilled.11 If the patent
holder does bring suit, however, the filing of that suit triggers an automatic 30-month stay of FDA approval of

12 Id. at § 355(j)(5)(B)(iii).
13 Id. at § 355(j)(5)(B)(iv).
14 See, e.g., Biovail Corp. and Elan Corp. PLC, Dkt. No. C-4057 (Aug. 20, 2002) (consent
order); Biovail Corp., Dkt. No. C-4060 (Oct. 2, 2002) (consent order); FTC v. Mylan
Laboratories, Inc. et al., 62 F. Supp. 2d 25 (D.D.C. 1999); Roche Holding Ltd., 125 F.T.C. 919
(1998) (consent order); Ciba-Geigy Ltd., 123 F.T.C. 842 (1997) (consent order). See, also, infra
note 18.
15 See supra note 1.
16 Bureau of Economics Staff Report, Federal Trade Commission, The Pharmaceutical
Industry: A Discussion of Competitive and Antitrust Issues in an Environment of Change (Mar.
1999), available at <http://www.ftc.gov/reports/pharmaceutical/drugrep.pdf>; David Reiffen and
Michael R. Ward, Generic Drug Industry Dynamics, Bureau of Economics Working Paper No. 248
(Feb. 2002), available at <http://www.ftc.gov/be/econwork.htm>.
4
the ANDA.12 This provision is referred to as the “30-month stay provision.” During this period, unless the
patent litigation is resolved in the generic applicant’s favor, the FDA cannot approve the generic product.
The second significant component of Hatch-Waxman is the “180-day period of exclusivity.” The
Amendments provide that the first generic manufacturer to file an ANDA containing a Paragraph IV
certification is awarded 180 days of marketing exclusivity, during which the FDA may not approve a
potential competitor’s ANDA.13
III. FTC EXPERIENCE AND INTEREST IN THE PROCEEDING
The Commission has pursued numerous antitrust enforcement actions affecting both brand-name
and generic drug manufacturers.14 In addition, the Commission released a study entitled “Generic Drug
Entry Prior to Patent Expiration” (“FTC Study”) in July 2002. That study examined whether the conduct that
the FTC has challenged represented isolated instances or is more typical of business practices in the
pharmaceutical industry, and whether certain provisions of Hatch-Waxman are susceptible to strategies to
delay or deter consumer access to generic alternatives to brand-name drug products.15 The Commission
has gained expertise regarding competition in the pharmaceutical industry through other means as well.
The Commission staff has conducted empirical analyses of competition in the pharmaceutical industry,
including in-depth studies by the staff of the Bureau of Economics.16
A. The Commission’s Enforcement Actions in the Pharmaceutical Industry
The Commission’s first-generation litigation focused on patent settlement agreements between
brand-name companies and generic applicants that the Commission alleged had delayed the entry of one

17 Abbott Laboratories, Dkt. No. C-3945 (May 22, 2000) (consent order), complaint
available at <http://www.ftc.gov/os/2000/05/c3945complaint.htm>; Geneva Pharmaceuticals, Inc.,
Dkt. No. C-3946 (May 22, 2000) (consent order), complaint available at
<http://www.ftc.gov/os/2000/05/c3946complaint.htm>; Hoechst Marion Roussel, Inc., Dkt. No.
9293 (May 8, 2001) (consent order), complaint available at
<http://www.ftc.gov/os/2000/03/hoechstandrxcomplaint.htm>.
18 The consent order in Abbott Laboratories is available at
<http://www.ftc.gov/os/2000/03/abbot.do.htm>. The consent order in Geneva Pharmaceuticals is
available at <http://www.ftc.gov/os/2000/03/genevad&o.htm>. The consent order in Hoechst/Andrx
is available at <http://www.ftc.gov/os/2001/05/hoechstdo.pdf>. In another matter, Schering-Plough,
the Commission resolved all claims against one of three respondents, American Home Products
(“AHP”), by issuing a final consent order. Schering-Plough Corp., Dkt. No. 9297 (consent order as
to AHP issued Apr. 2, 2002), available at <http://www.ftc.gov/os/2002/04/scheringplough_do.htm>.
The case against the other two respondents is on appeal before the Commission following the
Administrative Law Judge’s initial decision. See Schering-Plough Corp., et al., Dkt. No. 9297 (Initial
Decision) (July 2, 2002), available at <http://www.ftc.gov/os/2002/07/scheringinitialdecisionp1.pdf>.
19 The Commission first raised concerns about the potential anticompetitive impact of improper
Orange Book listings in American Bioscience, Inc. v. Bristol-Myers Squibb Co., et al., Dkt. No.
CV-00-08577 (C.D. Cal. Sept. 7, 2000). See Federal Trade Commission Brief as amicus curiae,
available at <http://www.ftc.gov/os/2000/09/amicusbrief.pdf>. In that case, the parties sought court
approval of a settlement containing a specific factual finding that Bristol-Myers was required to list
American Bioscience’s patent of Bristol-Myers’s branded drug Taxol in the Orange Book. The
Commission was concerned that the court’s approval of the settlement, which would have ordered
Bristol-Myers to maintain the listing and include a judicial finding that the patent met the statutory listing
requirements, would prejudice parties who might later challenge the listing.
5
or more generic applicants.17 Of course, resolving patent infringement litigation through settlement can be
efficient and procompetitive. Certain patent settlements between brand-name companies and generic
applicants, however, drew the Commission’s attention when it appeared that their terms may have reduced
competition through abuses of the Hatch-Waxman regime. The Commission resolved two matters by
consent order.18
A principal focus of the Commission’s second-generation activities has been improper Orange Book
listings.19 Unlike the matters involving settlement agreements, an improper Orange Book listing strategy
involves unilateral abuse of the Hatch-Waxman process itself to restrain trade.
In January of this year, lawsuits relating to Bristol-Myers’s alleged monopolization through improper

20 In re Buspirone Patent Litigation/In re Buspirone Antitrust Litigation, 185 F. Supp. 2d
363 (S.D.N.Y. 2002) (“In re Buspirone”). Some of the same plaintiffs previously had brought suit
under the FDC Act, requesting that the court issue an order compelling Bristol-Myers to de-list the
objectionable patent. Although plaintiffs prevailed at the district court level, the Federal Circuit
reversed that decision, holding that the FDC Act did not provide a private right of action to compel de-
listing of a patent from the Orange Book. See Mylan Pharmaceuticals, Inc. v. Thompson, 268 F.3d
1323, 1331-32 (Fed. Cir. 2001).
21 The Noerr doctrine was first articulated in Eastern R.R. Presidents Conf. v. Noerr Motor
Freight, Inc., 365 U.S. 127 (1961), and United Mine Workers of America v. Pennington, 381 U.S.
657 (1965).
22 15 U.S.C. § 2.
23 Memorandum of Law of Amicus Curiae Federal Trade Commission in Opposition to
Defendant’s Motion to Dismiss, available at <http://www.ftc.gov/os/2002/01/busparbrief.pdf>.
24 In re Buspirone, supra note 20.
25 Biovail Corp., supra note 14.
26 Biovail Corp. and Elan Corp. PLC, supra note 14.
6
listing of a patent on its brand-name drug BuSpar20 presented the Commission with an opportunity to clarify
whether there would be potential antitrust immunity under the Noerr doctrine for improper Orange Book
listings. The Noerr doctrine21 provides antitrust immunity for individuals “petitioning” government.
Specifically, plaintiffs alleged that, through fraudulent filings with the FDA, Bristol-Myers caused that
agency to list the patent in question in the Orange Book, thereby blocking generic competition with its
BuSpar product, in violation of Section 2 of the Sherman Act.22 Bristol-Myers defended its actions by,
among other things, claiming that its activities were immune from antitrust scrutiny under Noerr doctrine.
The Commission filed an amicus brief arguing that Orange Book filings are not “petitioning activity” immune
from antitrust scrutiny.23 On February 14, 2002, the district court issued an opinion denying Bristol-Myers’s
immunity claim.24
In another action, the Commission issued a consent order against Biovail Corporation,25 settling
charges that Biovail illegally acquired an exclusive patent license and wrongfully listed that patent in the
Orange Book for the purpose of blocking generic competition to its brand-name drug Tiazac. This was the
Commission’s first enforcement action to remedy the effects of an allegedly improper, anticompetitive
Orange Book listing.
The Commission also has taken action against alleged anticompetitive agreements between
generic manufacturers. In August, 2002, the Commission issued a consent order against two generic drug
manufacturers to resolve charges that they entered into an agreement that unreasonably reduced
competition in the market for a generic anti-hypertension drug.26
B. The Commission's Industry-Wide Generic Drug Competition Study

27 15 U.S.C. § 46(b).
28 See 65 Fed. Reg. 61334 (Oct. 17, 2000); 66 Fed. Reg. 12512 (Feb. 27, 2001).
29 67 Fed. Reg. at 65454.
7
In light of the questions its various generic drug investigations raised, the Commission proposed an
industry-wide study of generic drug competition in October, 2000. The FTC Study focused solely on the
procedures used to facilitate generic drug entry prior to expiration of the patent(s) that protect the brand-
name drug product – that is, generic entry through the procedures involving Paragraph IV certifications.
In April, 2001, the Commission issued nearly 80 special orders – pursuant to Section 6(b) of the
FTC Act27 – to brand-name companies and to generic drug manufacturers, seeking information about certain
practices that were outlined in the Federal Register notices that preceded OMB clearance to pursue the
study.28 The Commission staff focused the special orders on brand-name drug products that were the
subject of Paragraph IV certifications filed by generic applicants. Only those NDAs in which a generic
applicant notified a brand-name company with a Paragraph IV certification after January 1, 1992, and prior
to January 1, 2001, were included in the FTC Study. The selection criteria resulted in 104 drug products,
as represented by NDAs filed with the FDA, within the scope of the study and included so-called
“blockbuster” drugs such as Capoten, Cardizem CD, Cipro, Claritin, Lupron Depot, Neurontin, Paxil, Pepcid,
Pravachol, Prilosec, Procardia XL, Prozac, Vasotec, Xanax, Zantac, Zocor, Zoloft, and Zyprexa.
The Study made findings with respect to litigation frequency and outcomes that invoked the 30-
month stay and/or the 180-day marketing exclusivity provisions. Based on the these findings, the
Commission recommended changes to the 30-month stay and 180-day marketing exclusivity provisions to
eliminate the problems identified and to restore the balance that Hatch-Waxman initially struck between
encouraging innovation and providing for a streamlined generic drug approval process.
IV. LIMITING BRAND-NAME PHARMACEUTICAL COMPANIES TO A SINGLE 30-MONTH
STAY OPPORTUNITY PER ANDA IS AN EFFECTIVE WAY TO BRING THE ECONOMIC BENEFITS OF
GENERIC DRUGS TO CONSUMERS MORE QUICKLY
The FDA proposes to provide the opportunity for a single 30-month stay of the approval date of each
ANDA with a Paragraph IV certification if the generic applicant is sued for patent infringement. The FDA
states that the statutory language of Hatch-Waxman supports this interpretation, although it is a different
interpretation than the FDA had used previously.29 This proposal is slightly different from the FTC’s
legislative recommendation addressing the problems uncovered by its Study. We recognize that the FDA,
however, may be unable to implement our recommendation without additional legislative authority.
Nonetheless, the FDA proposal is an effective way to bring cheaper, generic copies of brand-name drug
products to the market.
A. The FTC Study Supports Permitting Only One 30-Month Stay Per ANDA
The FTC Study found that certain brand-name pharmaceutical companies have attempted to “game”
the 30-month stay provision by listing patents in the Orange Book after a generic applicant had filed an
ANDA with the FDA. For ease of discussion, patents listed after an ANDA has been filed with the FDA will
be referred to as “later-issued patents.” The FTC Study found that for drug products studied, brand-name
companies have listed patents in the Orange Book after an ANDA has been filed for 8 different brand-name

30 For one of the 7 drug products (BuSpar), the brand-name company listed more than one
later-issued patent in the Orange Book after several generic applicants had filed ANDAs that contained
Paragraph III certifications relating to another patent. The FTC Study did not contain data describing
when each generic applicant amended its pending ANDA to address each of the later-issued patents.
It is likely, however, that most of the ANDAs for BuSpar would not have had the opportunity to be
stayed for a second 30-month period had the proposed rule been in effect.
8
drug products (Platinol, Hytrin (tablets), Paxil, Taxol, BuSpar, Neurontin (tablets and capsules), and Tiazac)
with 6 of the 8 instances occurring since 1998. The delay of FDA approval caused by these later-issued
patents has ranged from 4 to 44 months. The net sales for each of these products ranged from greater than
$100 million per year to greater than $1 billion per year. Thus, the economic impact of the delay caused by
the unwarranted stay of FDA approval of a generic version of the brand-name product can be substantial.
A court found the later-issued patents for Platinol, Hytrin (tablets), Taxol, and BuSpar to be either
invalid or not infringed by the ANDA. Pursuant to a consent agreement with the Commission, the NDA
holder dismissed its patent infringement lawsuit involving the later-issued patent listed for Tiazac. The
infringement litigation involving the later-issued patents for the remaining drug products (Paxil, Neurontin
(tablets and capsules)) is pending.
The FTC Study further explained that the relationship between almost all of the later-issued patents
and the corresponding 8 brand-name products raised issues of whether the patents had been appropriately
listed in the Orange Book. In light of these findings, the Commission recommended that Congress permit
only one automatic 30-month stay per drug product per ANDA to resolve infringement disputes over patents
listed in the Orange Book prior to the filing date of the generic applicant’s ANDA.
B. The FTC Recommendation Would Eliminate Stays Generated From “Later-Issued”
Patents When an ANDA Has Already Been Filed
The FTC proposal does not guarantee the opportunity for one 30-month stay of every ANDA with a
Paragraph IV certification. Rather, FDA approval would be stayed for 30 months based only on those
patents that have been listed in the Orange Book before a generic applicant files an ANDA. The FTC
recommendation eliminates the potential for harm caused by 30-month stays generated by later-issued
patents. For example, if a generic applicant submits an ANDA that seeks FDA approval of a generic drug
product at the expiration of a certain listed patent (i.e., the generic applicant files an ANDA with a
Paragraph III certification) and the brand-name company lists a later-issued patent in the Orange Book for
that brand-name drug product, the FDA would not be stayed for 30 months from approving that previously
filed ANDA. By contrast, under the FDA proposal, approval would be stayed under that scenario.
If the proposed FDA rule had been in place previously, it would have eliminated the second (or
subsequent) 30-month stays of the ANDAs for 7 of the 8 brand-name drug products in which the FTC found
that brand-name companies had listed later-issued patents in the Orange Book.30 In the other instance, the
proposed rule would not have affected the start of the 30-month stay of FDA approval for most of the ANDAs
for the drug product Platinol. In that case, the brand-name company listed later-issued patents in the
Orange Book on the eve of the expiration of the last patent that was blocking FDA approval of the pending
ANDAs.
In sum, permitting only one 30-month stay will eliminate most of the potential for “gaming” the
system to delay FDA approval of generic applicants. It does not, however, completely fix the problem
caused by later-issued patents listed at the eve of expiration of a patent for which generic applicants have

31 21 U.S.C. § 355(b)(1).
32 21 C.F.R.§ 314.53(b)
9
filed a Paragraph III certification. Thus, it is important to tighten the patent listing requirements as
suggested by the FTC Study and discussed below to result in fewer improper listings. To do otherwise only
preserves an aspect of the generic drug approval process that has a demonstrable history of abuse.
V. THE FDA SHOULD TIGHTEN ITS PATENT LISTING REQUIREMENTS TO BE
CONSISTENT WITH THE RECOMMENDATIONS IN THE FTC STUDY
The FDA has proposed to clarify the types of patents that must and must not be listed in the
Orange Book. Specifically, the FDA has proposed that patents claiming packaging, metabolites, and
intermediates must not be listed in the Orange Book. By contrast, it proposes that product-by-process
patents must be listed in the Orange Book because in such claims the “patented invention is the product
(as opposed to the process used to make the product).” In addition, patents claiming a different form of the
drug substance must be listed in the Orange Book.
Hatch-Waxman’s listing provisions contain a 2-prong test that must be met before brand-name
companies can list patents in the Orange Book. For ease of discussion, the listing provision will be referred
to as either the “listing statute” or section 505(b)(1). An NDA filer shall file with the new drug application the
patent number and the expiration date of any patent:
[1] which claims the drug for which the applicant submitted the application or which claims
a method of using such drug and
[2] with respect to which a claim of patent infringement could reasonably be asserted if a
person not licensed by the owner engaged in the manufacture, use, or sale of the drug.”31
The FDA’s existing regulations interpret this 2-prong statutory requirement to mean that listable patents
consist of drug substance (active ingredient) patents, drug product (formulation and composition) patents,
and method of use patents. Process patents are not covered by the listing regulation.32
We support FDA’s proposal to prohibit brand-name companies from listing patents in the Orange
Book claiming packaging, metabolites, and intermediates because such patents do not meet both prongs
of the test in the listing statute. We urge the FDA to refine its approach to the listing of product-by-process
patents based upon our examination of the product-by-process patents that have been listed in the Orange
Book. The FDA should also reconsider its reasons for requiring the listing of patents in the Orange Book
that claim a different form of the drug substance than the one approved by the FDA for any given drug
product base. Such a requirement is contrary to the unambiguous language of the listing statute and it
could leave unchecked problems the Commission documented in the FTC Study.
Although the single 30-month stay proposal will address a significant portion of the problems the
FTC Study documented, it is important to clarify which patents may be listed in the Orange Book to ensure
that the listing statute is not used to delay generic drugs from coming onto the market. As noted above,
the one 30-month stay proposal does not completely address the problem of brand-name companies listing
later-issued patents in the Orange Book. Thus, it is critical to ensure that the 30-month stay provision does
not continue to have the potential to be “gamed.” Moreover, the Commission is aware of patents in the
Orange Book that provided the basis for the first 30-month stay of a generic company’s ANDAs that do not

33 See FTC Study at v, Appendix H at A-40 (discussing the patents listed for Prilosec).
34 67 Fed. Reg. at 65451.
35 Two examples are Prilosec (omeprazole, Patent No. 4,636,499, claiming a metabolite
compound) and BuSpar (buspirone Patent No. 6,150,365, claiming use of a metabolite compound).
The generic applicants in the omeprazole litigation moved for summary judgment that they did not
contribute to or induce infringement of the ‘499 patent. The district court granted that motion based, in
part, on the argument of the generic applicants that the patent could not cover a patient’s ingesting and
metabolizing omeprazole because that activity was prior art to the patent. In re Omeprazole Patent
Litigation, 2001 WL 585534 (S.D.N.Y. 2001). In a third more recent case involving a metabolite
patent for the drug product Claritin (loratadine, Patent No. 4,659,716), the district court held that ‘716
patent was invalid because it was inherently anticipated by prior art, namely the original drug substance
patent, Patent No. 4,282,233. Schering Corp. v. Geneva Pharms. et. al., 2002 U.S. Dist. LEXIS
14587 (D. N.J. Aug. 8, 2002).
36 In re Omeprazole Patent Litigation, 2001 WL 585534 (S.D.N.Y. 2001); see also, Zenith
Labs. v. Bristol-Myers Squibb, 10 F.3d 1418, 1422 (Fed. Cir. 1994) (a patient may directly infringe a
patent on a metabolite when the administered drug product is converted in vivo into the claimed
metabolite product). Section 271 of the Patent Act gives a patentee the option of suing a direct
infringer (i.e., the patient) and/or a contributory/inducing infringer (i.e., the generic drug company). 35
U.S.C. § 271(b), (c) (one who induces or contributes to direct infringement of a patent is liable as an
infringer).
10
appear to satisfy Hatch-Waxman’s patent listing standards.33 Indeed, the worst outcome would be for a
court to strike down FDA’s one 30-month stay proposal while, at the same time, upholding FDA’s
regulations requiring the listing of patents in the Orange Book that do satisfy the listing statute, thus
opening the door for multiple 30-month stays based on inappropriately listed patents.
A. Patents claiming packaging, metabolites, and intermediates do not meet the listing
requirements and, therefore, should not be listed in the Orange Book
As FDA describes, a metabolite is the chemical compound into which a patient’s body metabolizes
or converts the active ingredient of a drug product.34 Often the metabolite, rather than the active ingredient
itself, produces the drug’s therapeutic effect in the body.
Of the drug products for which the FTC Staff examined listed patents in the FTC Study, there are at
least two instances where brand-name companies have listed and sued generic companies for infringement
of metabolite patents.35 Typically, the patentee charges that the generic applicant will induce or contribute
to the infringement of the metabolite patent by selling its drug to patients who then directly infringe the
patent by ingesting the drug and metabolizing it.36
One district court explicitly has held that a brand-name company may not list a metabolite patent
in the Orange Book, because the metabolite patent does not “claim the drug,” as required by the listing

37 Mylan Pharmaceuticals, Inc. v. Thompson, 139 F. Supp. 2d 1, 19-21 (D.D.C. 2001),
rev’d on other grounds, 268 F.3d 1323 (Fed. Cir. 2001).
38 109 F.3d 756 (Fed. Cir. 1997).
39 67 Fed. Reg. at 65452.
40 Id.
41 In re Thorpe, 777 F.2d 695, 697 (Fed. Cir. 1985) (“Product-by-process claims are not
specifically discussed in the patent statute. The practice and governing law have developed in response
to the need to enable an applicant to claim an otherwise patentable product that resists definition by
other than the process by which it is made”).
42 In re Bridgeford, 357 F.2d 679, 682 (C.C.P.A. 1966). The use of product-by-process
claims, however, appears to be rare in the patenting of pharmaceutical products and drug substances
11
statute.37 The court looked to the precedent, Hoechst-Roussel Pharms., Inc. v. Lehman,38 which
interpreted the term “claims” in the Patent Term Restoration portion of the Hatch-Waxman Amendments at
35 U.S.C. § 156(a) and concluded that a metabolite patent does not “claim” the approved drug product. In
light of this interpretation, metabolite patents should not be listed in the Orange Book because they do not
claim the drug as required by the listing statute’s 2-prongs.
Likewise, “intermediate” patents listed in the Orange Book present a category that also do not
literally claim the approved drug product. An intermediate patent claims a chemical compound that is used
during the production of an active ingredient, but is not present in the final, marketed form of the drug
product. The claimed compound is an “intermediate” on the pathway to the approved drug. The FDA notes
that under its regulations, intermediates are “in-process materials” rather than drug substances or even drug
components. Thus, patents that claim intermediates do not claim the approved drug product and fail the
first prong for listing.39
B. Patents Reciting a Known Product, but a Novel Process Drafted in the Product-By-
Process Format, Do Not Claim the Drug Product And Should Not Be Listed in the
Orange Book
The FDA proposes to clarify that patents containing product-by-process claims are to be listed in
the Orange Book because in such claims the “patented invention is the product (as opposed to the process
used to make the product).”40 Product-by-process patents must be listed in the Orange Book if they meet
the 2-prong test of the listing statute. We suggest that the FDA refine its approach to safeguard against
the listing of claims reciting a known product and a novel process that are drafted in the product-by-process
format. Such claims do not claim a product and, therefore, do not meet the 2-prong test of the listing
statute.
Product-by-process claims typically are used when a novel product cannot be adequately identified
or described by its physical characteristics.41 In these situations, the courts have recognized that “the right
to a patent on an invention is not to be denied because of the limitations of the English language, and in a
proper case, a product may be defined by the process of making it.”42 The format of a product-

governed by Hatch-Waxman because it is possible to identify and claim such products by their physical
characteristics rather than by the process of making the product.
43 Union Carbide Chemicals & Plastics Technology Co. v. Shell Oil Co., 2002 WL
31094845 at * 7 (Fed. Cir. 2002) (using standard patent claim construction principles to determine
whether a claim is a product-by-process claim); Mentor Corp. v. Coloplast, Inc., 998 F.2d 992, 997
(Fed. Cir. 1993) (interpreting claim as a method of use claim rather than product-by-process claim,
where the claim language did not refer to a process of making or manufacturing).
44 67 Fed. Reg. at 65452. See also U.S. Patent and Trademark Office, “Manual of Patent
Examining Procedure” § 2173.05(p) (8th ed. 2001) (A product-by-process claim “is a product claim
that defines the claimed product in terms of the process by which it is made.”).
45 In re Brown, 459 F.2d 531, 535 (CCPA) 1972) (“[I]t is the patentability of the product
claimed and not of the recited process steps which must be established.”); see also Atlantic
Thermoplastics Co., Inc. v. Faytex Corp., 974 F.2d 1279, 1282 (Fed. Cir. 1992) (Newman, J.,
dissent from denial of reh’g en banc) (stating that in “‘true’ product-by-process claims . . . patentability
and validity depend[] on the novelty and unobviousness of the product . . .”).
46 Cochrane v. Badische Anilin & Soda Fabrik, 111 U.S. 293, 311, 4 S.Ct. 455, 465
(1884) (“It was an old article. While a new process for producing it was patentable, the product itself
could not be patented, even though it was a product made artificially for the first time”); see also In re
Thorpe, 777 F.2d at 697; In re Luck, 476 F.2d 650, 652 (C.C.P.A. 1973); see also, Atlantic
Thermoplastics Co., Inc., 974 F.2d at 1284 (Newman, J., dissent from denial of reh’g en banc)
(stating that such claims “are not properly called ‘product-by-process’ claims, if that term is used with
precision.”).
47 In re Ewert, 77 F.2d 498, 499 (C.C.P.A. 1935) (when the product is old and only the
processs is new, a “claim may extend only to the process for making it.”).
12
by-process claim ordinarily is as follows: “[substance x] made by the process of [steps (a) and (b)]” or the
“the product when prepared by [steps (a) and (b)].” Nevertheless, the format of a claim is not determinative
of whether the claim is a product-by-process claim, or rather some other type of claim.43
Rather, as the FDA has stated, product-by-process claims are those in which the patented
invention is the product,44 as opposed to the process used to make the product. A new and patentable
process cannot make a known product, which results from that process, patentable. The most essential
requirement for a product-by-process claim is that the end product of the process be new and patentably
distinct from prior products.45 For ease of discussion, we refer to claims that meet this requirement as
“true” product-by-process claims. When a claim is drafted in product-by-process format, but only relies on
the recited process for patentability, it is not a “true” product-by-process claim.46 To the extent that such a
patent claim is valid, it can only extend to the process recited in the claim.47
Because the product in a true product-by-process claim is a novel invention, the first prong of the

48 See FTC Study at A-42- A-44.
49 67 Fed. Reg. at 65452.
50 Patents claiming a chemical compound that differ by water-of-hydration or that form a
crystalline structure different from the active ingredient are referred to as “polymorphs.” Under the
proposed change, an NDA holder who had obtained FDA approval of the anhydrate form of a drug
substance (having no water) would be required to list a patent, if it obtained one, claiming the
monohydrate form of the same drug substance (having one water molecule in its crystalline structure for
each molecule of the drug substance) or any patent it obtains that claims another hydrated form of the
13
listing statute requires these claims to be listed in the Orange Book (provided they also satisfy the other
criteria of the listing statute). When a claim relies solely on a novel process for patentability, even if such a
claim is drafted in product-by-process format, it is not a product claim, and therefore, does not satisfy the
first prong. Neither the listing statute nor FDA’s proposal allows the listing of patents based on claims in
which the patentee only relied on the process as the novel invention.
The Commission in its Study identified several patents listed in the Orange Book based on claims
drafted in a product-by-process format for which, according to the patent itself, the novel aspect of the
invention was the process, not the product. For example, the FTC Study identified that certain patents
listed for the drug product Paxil contained only process claims and claims drafted in the product-by-
process format. The latter claims recited an admittedly known drug substance made according to a
purportedly novel process.48 Such claims are not “true” product-by-process claims of the type identified in
the FDA’s proposal.
The FDA has “invite[d] comment on ways to ensure that only appropriate product-by-process
patents are listed, while maintaining that act’s restriction against listing process patents.”49 We make two
suggestions to ensure that only appropriate product-by-process patents are listed in the Orange Book.
First, as an additional component of its enhanced patent declaration, the FDA may wish to require a
question on product-by-process patents. If an NDA holder relies on claims drafted in the product-by-
process format to support the listing of a patent, then the FDA should require the NDA holder to certify that
the novelty of the claim is the product, not the solely recited process. In addition, the NDA holder should
certify that the claimed product has been approved through the NDA, as required by the FDA’s proposed
rule and additional declaration requirement. Section VI below contains suggested language for the
declaration.
Second, the FDA should revise the text of the proposed regulation to reflect the fact that only
product-by-process claims in which the product is novel should be listed. Thus, the language “(in which the
product is novel)” could be added immediately following the mention of product-by-process patents in the
proposed regulation at § 314.53(b).
C. The Plain Language of the Listing Statute and the Underlying Purpose of Hatch-
Waxman Neither Require Nor Support the Listing of Patents in the Orange Book
That Claim a Different Form of a Drug Substance Than That Approved Through the
NDA
The FDA proposes to amend its listing regulations so that patents claiming a different form of a
drug substance than that approved through the NDA must be listed in the Orange Book.50 The plain

drug substance.
51 67 Fed. Reg at 65452.
52 21 U.S.C. § 355(j)(2)(A)(ii). For ease of discussion, this section will be referred to as the
“ANDA submission section.”
53 67 Fed. Reg. at 65452
54 Association of American Physicians and Surgeons, Inc. v. FDA, 2002 U.S. Dist. LEXIS
19689, 709 (D.D.C. 2002) citing Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc.
467 U.S. 837, 842-43 (1984).
14
language of the listing statute and the underlying purpose of Hatch-Waxman support retaining the FDA’s
existing regulations that do not allow listing patents in the Orange Book that claim a different form of a drug
substance than that approved through the NDA.
As discussed above, the first prong of the listing statute requires that a patent be listed in the
Orange Book that “claims the drug or a method of using the drug that is the subject of the new drug
application or amendment.” The FDA proposes to add to the language in its regulation implementing the
first prong (which is nearly identical to the listing statute) the phrase “or that claim a drug substance that is
the same as the active ingredient that is subject of the approved or pending application within the meaning
of section 505(j)(2)(A)(ii) of the act.”51
Section 505(j)(2)(A)(ii) requires generic applicants to show that the active ingredient of the new drug
(i.e., the generic version) is “the same” as that of the listed drug (i.e., the previously approved brand-name
drug).52 The FDA states that under the ANDA submission section it has allowed generic applicants to
submit ANDAs for generic drugs that contain different forms of the active ingredient than that approved for
the brand-name drug. Thus, in this proposal the FDA seeks to harmonize the listing statute with the ANDA
submission section.
The FDA recognizes that allowing NDA holders to submit such patent information contradicts its
“longstanding position that the patent must claim the approved drug or the drug product that is the subject
of the application” (i.e., the listing statute’s requirements).53 The proposal now asserts that because a
generic drug must be therapeutically equivalent to the brand-name drug product and this requirement can be
shown by a generic drug containing a drug substance having a different physical form than the brand-name
drug product, it is consistent to require that patents for drug substances be listed in the Orange Book even
if those drug substances have different physical forms than that approved through the NDA.
1. The Listing Statute Does Not Contemplate the FDA’s Proposal
In our view, the listing statute prohibits the proposed policy change. The FDA’s proposed
interpretation is contrary to the unambiguous language of the listing statute, in particular the first prong that
requires any listed patent claim the drug (either the drug substance or the drug product) that is the subject
of the NDA. Congress has spoken to the precise question at issue.54
When the FDA approves an NDA, that approval covers one form or polymorph of a drug substance.
The NDA holder may not sell a version of the drug product containing an unapproved form of the drug

55 Serono Labs, Inc., v. Shalala, 158 F.3d 1313, 1320-21 (D.C. Cir. 1998).
56 Indeed, in the discussion of why metabolites patents do not satisfy the requirements of the
listing statute, the FDA cites with approval the notion that the term “claims” as used in the first prong of
the listing statute is based on patent law concepts. See 67 Fed. Reg. at 65451. There is no reason
why the listing statute would require a different analysis for determining whether metabolite patents and
polymorph patents should be listed in the Orange Book.
57 But see Zenith Laboratories, Inc. v. Abbott Laboratories, Inc., 1997 LEXIS U.S.Dist.
23954 (D. NJ 1997).
15
substance. The first prong of the listing statute requires the NDA holder to list any patents claiming the
approved form in the Orange Book. Conversely, the plain language of the statute does not allow the listing
of patents that claim an unapproved form of the drug substance.
The requirements of the ANDA submission section do not change this analysis because they do
not alter the plain language of the listing statute. The FDA allows generic drug products to contain a drug
substance that differ from that approved through the NDA because the FDA looks to principles of
pharmaceutical equivalence in defining the “same active ingredient” requirement of the ANDA submission
section.55 Critically, the term “same” does not appear in the listing statute.
No harmonization of the two section is necessary because the scope of drug substances to be
considered the “same” in the ANDA submission section is broader than the requirements governing the
patents to be listed in the Orange Book. Indeed, the very structure of the Hatch-Waxman system works
precisely because Congress took advantage of the fact that drugs can be pharmaceutically and
therapeutically equivalent without being identical (infringing) in a patent law sense. It is this lack of patent
law identity that allow generic drugs to be marketed under Hatch-Waxman before the relevant patents have
expired.
The analysis required to determine whether a patent must be listed in the Orange Book is a patent
law analysis (whether the patent “claims” NDA’s drug),56 whereas the analysis required to determine
whether a generic product contains the same active ingredient as that of a listed drug is a pharmaceutical
analysis.57 As explained below, the two analyses will give different results.
For example, to obtain a later polymorph patent from the PTO, the patentee typically
demonstrates how the polymorph is patentably distinct from the FDA-approved drug substance. If the later
polymorph were “the same” in a patent sense, then the patent claiming it would not issue in the first
instance. Thus, by virtue of obtaining the later patent, the NDA holder often explicitly takes the position
that the polymorph patent does not satisfy the listing statute (i.e., that it does not claim the drug substance
that is the subject of the NDA). Nevertheless, the FDA-approved drug substance and the later polymorph
may well be pharmaceutically equivalent and, therefore, the “same” in the sense of the ANDA submission
section of Hatch-Waxman.
The drug substance patents listed for the drug product Paxil illustrates the distinction between a
patent law analysis and a pharmaceutical analysis. The original drug substance patent (Patent No.
4,007,196) covering all forms of paroxetine hydrochloride has expired. The NDA holder obtained another
drug substance patent (Patent No. 4,721,723) that claims paroxetine hydrochloride hemihydrate, the form of

58 Patent law allows the patenting of a species compound (i.e., the anhydrate of the ‘423
patent) even when a broad category, called a genus, of compounds (i.e., all paroxetine compounds of
the ‘196 patent) has been described in an earlier publication. See In re Kaplan, 789 F.2d 1574, 1578
(Fed. Cir. 1986).
16
the active ingredient approved through the NDA. The NDA holder then obtained a third drug substance
patent (Patent No. 5,900,423) claiming paroxetine hydrochloride anhydrate form A. To prevent the ‘723
hemihydrate patent from being considered invaliding prior art to the ‘423, the patentee distinguished the
anhydrate of the ‘423 patent from the hemihydrate form claimed in the ‘723 patent.58 (The relationship of the
scope of coverage of the three patents is illustrated in Figure 1.) Although the hemihydrate and anhydrate
forms both fall within the broader category of all paroxetine compounds covered by the ‘196 patent, the ‘723
and ‘423 patents are necessarily patentably distinct from each other. They cannot be the “same” in a
patent analysis. If they were the “same,” then the ‘723 patent would invalidate the ‘423 patent. In spite of
the fact that the anhydrate and hemihydrate forms are patentably distinct; however, they may react
identically in a pharmaceutical sense and, therefore, be pharmaceutically equivalent.

59 67 Fed. Reg. at 65453. The FDA states that not listing all patents claiming various
polymorphs of an approved drug substance may mislead potential generic applications into submitting
ANDAs containing a polymorph for which a brand-name company has a patent. It could also waste
FDA resources that were used to review an ANDA for a drug that is covered by an unlisted patent.
60 See FTC Study at 51-52.
61 The listing of additional patents that could form the basis of an unwarranted 30-month stay of
FDA approval could have the effect of delaying FDA approval and slowing the market entry of generic
drugs. This result “would thwart Congress’s central goal, in enacting the Hatch-Waxman Amendments,
to bring generic drugs onto the market as rapidly as possible.” Mova Pharmaceutical Corp. v.
Shalala, 140 F.3d. 1060, 1068 (D.C. Cir. 1998).
17
2. Listing Patents That Do Not Claim the Approved Drug Product Undermines
the Purpose of Hatch-Waxman
Several other considerations counsel against changing the current listing policy with respect to
patents claiming a different form of the drug substance. First, one of the goals of the 180-day exclusivity
provision in Hatch-Waxman was to encourage generic applicants to “design-around” existing patents. Many
generic applicants often do design bioequivalent products having formulations that fall outside a brand-name
company’s patents on its own approved formulation. Only those patents claiming the NDA holder’s
approved formulation, and not every patent claiming a bioequivalent formulation, may be listed in the Orange
Book. A generic applicant’s ability to formulate a bioequivalent drug product using a drug substance that
differs from that of the NDA by either water-of-hydration or crystal form should be viewed as a mechanism
by which the generic applicant designs around patents covering the NDA-approved product, and therefore,
listed in the Orange Book. Those design-arounds are possible only because Congress recognized -- and
codified through Hatch-Waxman -- the difference between therapeutic equivalence and sameness in a
patent law sense.
Second, the FDA’s proffered justification that listing more patents in the Orange Book may
conserve agency and industry resources may be theoretically possible, but is not borne out by the facts.59
For example, when the polymorph patent issues and is listed after a generic applicant files an ANDA, the
listing of the polymorph patent can delay generic competition, but it cannot conserve resources that have
already been expended. This is the fact situation for the Paxil product.60
Third, the notice function served by Orange Book listings (i.e., providing generic applicants notice of
what patents to design around) is not substantial because: (1) we have learned through our investigations
that generic applicants actively monitor patenting activities by the brand-name companies on their own; and
(2) the Orange Book does not inform generic applicants of all the patents they need to design-around (e.g.,
process patents are not required to be listed). Thus, relying on the notice function of the Orange Book for
the change in policy appears to be misplaced.
Fourth, the potential harm from listing inappropriate patents in the Orange Book can be great if
such listings provide the basis for a 30-month stay. NDA holders can always litigate patent infringement
claims regardless of whether the patent is listed in the Orange Book. As the FTC Study documented,
however, there can be substantial harm to consumers if generic drugs are denied FDA approval because of
30-month stays based on improperly listed patents.61 The FTC Study identified listed patents claiming

62 This prohibition is rooted in 35 U.S.C. § 101, which provides that an inventor “may obtain a
patent” for a new invention. 35 U.S.C. § 101; see also In re Hallman, 655 F.2d 212, 216 (C.C.P.A.
1981). The courts have interpreted the word “a” in this provision to mean that only one patent may
issue for a single scientific advance. In re Vogel, 422 F.2d 438, 441(C.C.P.A. 1970).
63 See In re Vogel, 422 F.2d at 441.
64 See U. S. Patent and Trademark Office, “Manual of Patent Examining Procedure,” Section
804.02, available at <http://www.uspto/web/offices/pac/mpep/documents/0431.htm>.
18
different polymorphs than those approved for Hytrin, Paxil, and Neurontin that have provided the basis of 30-
month stays on FDA approval of generic versions of these drug products.
Fifth, the proposed change to require the listing of patents that claim a different form of the drug
substance does not remedy a documented abuse of Hatch-Waxman. This proposal is unlike the proposal
to provide the opportunity for a single 30-month stay that responds, in part, to documented abuses of
Hatch-Waxman that have caused significant delay of FDA approval of generic drugs. Indeed, requiring NDA
holders to list patents that do not claim the approved drug product increases, rather than decreases, the
potential for delayed market entry by generic drug products.
In light of these reasons, we continue to believe that patents that claim a different physical form of
the active ingredient of an approved brand-name drug product should not be listed in the Orange Book.
D. Patents Rejected for Obviousness-Type Double Patenting in Light of the Claims of
an Earlier Listed Patent, as Identified Through the Use of a Terminal Disclaimer in
the Later Patent, Should Not Be Listed in the Orange Book
The FDA proposal does not address the problem of double patenting identified in the FTC Study.
The FTC Study recommended that patents that claim subject matter that is obvious in light of the claims of
the NDA holder’s earlier listed patent, as identified through the use of a terminal disclaimer in the later
patent, should not be listed in the Orange Book. The FDA may wish to safeguard against inclusion of
double patents in the Orange Book as they may form the basis for a 30-month stay.
When a patent applicant obtains a second patent claiming subject matter that is either the same
as,62 or obvious in light of, the claims of an earlier patent issued to the same applicant, it is called “double
patenting.” There are two types of double patenting: statutory and judicially created obviousness-type.
Generally, the patent statute and the doctrine of obviousness-type double patenting render such patents
invalid. The purpose of these rules is to prevent an inventor from extending the term of patent exclusivity by
the subsequent patenting of variations that are not patentably distinct from the first-patented invention.63
Unlike statutory double patenting which cannot be cured, obviousness-type double patenting can
be cured only if the applicant files what is called a “terminal disclaimer.” When the later patent claims the
identical subject matter as the earlier patent, as opposed to obvious subject matter, the statutory bar on
double patenting cannot be overcome with a terminal disclaimer. The later patent is invalid. 64
A terminal disclaimer acts to disclaim the term of the later patent that extends beyond the term of

65 See id. at 4 (discussing why a terminal disclaimer is required to overcome judicially created
double patenting rejections in applications filed on or after June 8, 1995).
66 We note that although this problem may be limited if there is only one 30-month stay, it is
important to ensure that only patents that meet the requirements of the listing statute are included in the
Orange Book. See supra text accompanying note 33.
67 In the Fosamax case, the NDA-holder dropped the later-filed suit shortly after filing,
eliminating the potential for the successive 30-month stay to block generic approval. See FTC Study at
A-44.
19
the original patent, so that both patents expire on the same day.65 It is deemed an effective remedy
because it restricts the patentee’s exclusivity to the term of the original patent. Theoretically, the patentee
receives no additional exclusionary term from the later patent.
When an NDA holder lists the later patent in the Orange Book, the patent holder does receive an
additional term of exclusivity if it generates a second, later-expiring 30-month stay.66 An NDA holder’s
ability to list a patent with a terminal disclaimer effectively vitiates the remedial effect of the terminal
disclaimer. The NDA holder essentially obtains an extension of the exclusionary effect of its patent by
listing it in the Orange Book, contrary to the intent of the patent laws. This problem is not a theoretical one.
The FTC Study identified that in the case of Paxil and Fosamax, the NDA-holder had listed and sued on
both an earlier-listed patent and a later-issued, double patent that contained a terminal disclaimer.67
The FDA may wish to consider prohibiting the listing of patents having a terminal disclaimer over an
already-listed patent and requiring an additional question concerning these patents in its enhanced
certification procedure. Specific proposed language for the certification is set forth in Section VI. below.
These measures are appropriate because only the earlier-listed patent provides the extent of protection
contemplated by the Hatch-Waxman statute for the claimed subject matter.
VI. MORE RIGOROUS PATENT CERTIFICATION REQUIREMENTS CAN ASSIST IN
REDUCING POTENTIAL ABUSE OF THE PATENT LISTING PROCESS
The FDA has redesigned the declaration that brand-name companies use to seek a listing of a
patent in the Orange Book. This redesign is a significant improvement, but even more stringent certification
requirements will help ensure that patent listings are appropriate.
We have several suggestions based on our enforcement experience and on the results of the FTC
Study. First, the certification should require that the person attesting to the certification is either senior
patent counsel with the NDA holder or an outside patent counsel specifically designated to act as the NDA
holder’s agent. Our experience indicates that certifications have been made by individuals without
knowledge of the regulatory requirements or familiarity with the pertinent patent issues. Second, the FDA
may wish to consider adding a knowledge requirement to the certification along the following lines: “I attest
that I am familiar with the requirements of 21 C.F.R. § 314.53 and with the scope of the claims of patent
number _________.”
Third, the FDA should consider adding two additional declarations on patents with product-by-
process claims and terminal disclaimers. To address the product-by-process issue, FDA may wish to add

20
subpart F. It could read as follows:
F. For each drug substance or drug product claim that was (1) identified as listable in
subparts B and C and (2) is drafted in product-by-process format, please provide the
following information:
1. Is the product of the recited process novel? [If the answer to question F.1 is “no,” stop.
The patent cannot be listed. If yes, please identify the claim(s) by number.]
To address the terminal disclaimer issue, the FDA may wish to add the following questions to
subpart A of the proposed declaration:
5. Does this patent contain a terminal disclaimer over a patent that has been listed in the Orange
Book? [If the answer to question A.5 is yes, stop here. The patent may not be listed in the
Orange Book. If the answer is no, proceed to subpart B.]

21
Finally, in its proposed declaration, the FDA included a new emphasis on the specific claims of the
listed patent, rather than solely on the patent as a whole. This is a desirable step that should add clarity to
the patent listing process. Nevertheless, it is unclear whether the FDA’s regulation now treats individual
claims of the patent to be listable, or whether the patent itself is listable given the presence of a single
listable claim. For example, a single patent may include both claims to a process or an intermediate,
which are not listable, as well as listable claims to an a product formulation. In that case, an NDA holder
may argue that litigation solely on a process claim might generate a 30-month stay. In light of the
proposed changes to the patent certification requiring identification of specific claims supporting listing, we
suggest the newly requested information be recorded in the Orange Book. Accurate patent listings, along
with identification of the claims that support the listings, will ensure that the FDA invokes the 30-month stay
only when a suit is brought over a claim that supports a listing in the Orange Book.
VII. CONCLUSION
The FDA has proposed to clear away unnecessary roadblocks to the approval of generic drug
products. The FDA's important action addressing the competitive problems existing in the approval process
for generic drugs, if promulgated and upheld, will an effective way to bring the economic benefits of generic
drugs to consumers more quickly. The Commission urges the FDA, however, to make the proposed
reforms even more effective by tightening its patent listing requirements. This additional step will prevent
unnecessary delays in generic drug approvals by ensuring that patents are not listed in the Orange Book
that do not meet the listing statute’s requirements.
By direction of the Commission.
Donald S. Clark
Secretary