Astrazeneca Pharmaceuticals Lp v. Food and Drug Administration

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
ASTRAZENECA PHARMACEUTICALS LP,
Plaintiff,
v. Civil Action No. 12-00472 (BAH)
Judge Beryl A. Howell
FOOD AND DRUG ADMINISTRATION,
et al.,
Defendants.
AMENDED MEMORANDUM OPINION
Plaintiff AstraZeneca Pharmaceuticals LP (“AstraZeneca”) has manufactured the drug
quetiapine fumarate (“quetiapine”) under the brand name Seroquel® (“Seroquel”) since 1997
without generic competition. AstraZeneca brought this lawsuit, which presents a question of
statutory interpretation, against the Food and Drug Administration, Margaret A. Hamburg, M.D.,
Commissioner of Food and Drugs, and Kathleen Sebelius, Secretary of Health and Human
Services (collectively, “the FDA”), to challenge the FDA’s approval, on March 27, 2012, of
generic versions of Seroquel. See Complaint, ECF No. 1 (“Compl.”), ¶ 3.
AstraZeneca believes that, under the plain language of 21 U.S.C. § 355(j)(5)(F)(iv),
codifying Section 505(j)(5)(F)(iv) of the Federal Food, Drug, and Cosmetic Act (“FDCA”), it is
entitled to total market exclusivity until December 2, 2012 for the safety information
encapsulated in “Table 2,” which was approved for all Seroquel labels on December 2, 2009 and
must be included on the labels of all generic versions of quetiapine. Based upon this belief,
AstraZeneca seeks a judgment that the FDA’s recent approval of generic versions of quetiapine,
while AstraZeneca retains exclusivity over Table 2, violated AstraZeneca’s exclusivity rights and
was arbitrary, capricious, and contrary to law.
1
Pending before the Court are Cross-Motions for Summary Judgment filed by
AstraZeneca, ECF No. 21, and the FDA, ECF No. 26. For the reasons explained below, the
Court denies AstraZeneca’s Motion for Summary Judgment and grants the FDA’s Motion for
Summary Judgment.
I. BACKGROUND 1
A. STATUTORY AND REGULATORY BACKGROUND
1. New Drug Applications
The pharmaceutical drug approval process for both new and generic drugs is governed by
the FDCA, as amended by, inter alia, the Drug Price Competition and Patent Term Restoration
Act of 1984, Pub. L. No. 98-417, 98 Stat. 1585 (“Hatch-Waxman Amendments”) (codified at 21
U.S.C. §§ 355, 360cc (2000), and 35 U.S.C. §§ 156, 271, and 282 (2000)). 2 The FDA is the
agency charged with approving all new and generic drugs for market. See 21 U.S.C. § 355(a).
Under the FDCA, pharmaceutical drug manufacturers interested in marketing a new
pharmaceutical drug (otherwise known as an “innovator” or “pioneer” drug), such as Seroquel,
must file a new drug application (“NDA”) with the FDA as required by 21 U.S.C. § 355(b)(1),
and must demonstrate, inter alia, the safety and efficacy of the drug. See id.; Compl. ¶ 29.
Pioneer drug companies must file with the FDA “full reports of investigations which have been
made to show whether or not such drug is safe for use and whether such drug is effective in use,”
21 U.S.C. § 355(b)(1)(A), and other information, including “a full statement of the composition
of such drug,” 21 U.S.C. § 355(b)(1)(C), and “specimens of the labeling proposed to be used for
1
This Court provides a brief background below and also incorporates by reference the extensive statutory,
regulatory and case-specific background set forth in AstraZeneca Pharms. LP v. FDA, No. 12-00388, 2012 U.S.
Dist. LEXIS 39611, at *4-24 (D.D.C. Mar. 23, 2012).
2
As noted, section 505 of the FDCA is codified in section 355 of Title 21 of the United States Code. For clarity,
this Opinion refers to the provision by its U.S. Code section number, 355, but maintains 505 when quoting from
parties’ briefs.
2
such drug,” 21 U.S.C. § 355(b)(1)(F). “Once the drug is approved, it is referred to as a ‘listed
drug.’” Sanofi-Aventis U.S. LLC v. FDA, No. 10-1255, 2012 WL 373214, at *1 (D.D.C. Feb. 7,
2012) (citing 21 C.F.R. § 314.3(b)).
The FDA publishes listed drugs in the “Orange Book,” which includes information about
applicable patents and periods of exclusivity. See Orange Book: Approved Drug Products with
Therapeutic Equivalence Evaluations, available at http://www.fda.gov/cder/ob/ (“Orange
Book”). The Orange Book provides notice to generic drug applicants about when drug patents
and periods of exclusivity expire, and when there will be openings to market generic versions of
pioneer drugs. See Defs.’ Mem. in Supp. of Mot. for Summ. J. (“Defs.’ Mem.”) at 4.
2. Abbreviated New Drug Applications
The Hatch-Waxman Amendments to the FDCA allowed manufacturers to seek approval
from the FDA to market generic drugs by filing an abbreviated new drug application (“ANDA”).
See 21 U.S.C. § 355(j). The significance of the Hatch-Waxman Amendments has been aptly
noted by other Judges in this Circuit:
Prior to 1984, all applicants seeking to market pioneer drugs or generic non-
antibiotic drugs had to file [a new drug application (“NDA”)] containing, inter
alia, extensive scientific data demonstrating the safety and effectiveness of the
drug. See 21 U.S.C. § 355(a)-(b); 21 C.F.R. § 314.50. As a result, few generic
non-antibiotic drugs were approved by [the] FDA. See [Glaxo, Inc. v. Heckler,
623 F. Supp. 69, 72 (E.D.N.C. 1985)]. Hatch-Waxman created an abbreviated
approval process for generic non-antibiotic drugs, while retaining incentives for
pioneer drugs, such as marketing exclusivity and patent protections. See 21
U.S.C. § 355(j). The abbreviated new drug application (“ANDA”) process
shortens the time and effort needed for approval of a generic drug by allowing the
applicant to merely demonstrate its product’s bioequivalence to the NDA drug,
without reproducing the entirety of the NDA’s extensive scientific research. See
Eli Lilly and Co. v. Medtronic, Inc., 496 U.S. 661, 676, 110 S. Ct. 2683, 110 L.
Ed. 2d 605 (1990) (describing the ANDA process).
ViroPharma, Inc. v. Hamburg, No. 12-0584, 2012 U.S. Dist. LEXIS 56128, at *6-7 (D.D.C. Apr.
23, 2012) (quoting Allergan, Inc. v. Crawford, 398 F. Supp. 2d 13, 16-17 (D.D.C. 2005)).
3
Unlike applicants for pioneer drugs, applicants for generic drugs are not required to
submit clinical data to demonstrate the safety and efficacy of their product. Instead, according to
the FDA, “if an ANDA applicant establishes that its proposed drug product has the same active
ingredient, strength, dosage form, route of administration, labeling (with certain permissible
differences), and conditions of use as a listed drug, and that it is bioequivalent to that drug, the
applicant” may rely on the FDA’s earlier findings of safety and efficacy for the drug when it was
approved as an NDA. Defs.’ Mem. at 5; see also 21 U.S.C. § 355(j); Compl. ¶ 34.
FDA-approved generic versions of a drug must utilize the “same” labeling as the labeling
approved for the reference-listed drug, except for labeling differences “based on a suitability
petition or because the generic drug and the reference drug are produced or distributed by
different manufacturers.” Compl. ¶ 35; 21 U.S.C. § 355(j)(2)(A)(v) (an ANDA must include
“information to show that the labeling proposed for the new drug is the same as the labeling
approved for the listed drug referred to in clause (i) . . . .”). FDA regulations require that, when a
manufacturer submits an ANDA, “[l]abeling (including the container label, package insert, and,
if applicable, Medication Guide) proposed for the drug product must be the same as the labeling
approved for the reference listed drug,” with certain exceptions not applicable here. 21 C.F.R. §
314.94(a)(8)(iv); see AR 294, 305 (FDA Letter, dated Mar. 27, 2012, to AstraZeneca, explaining
that the “FDA concurs that these portions of the labeling are essential to safe use of a generic
quetiapine product referencing Seroquel for any indication, and the agency would not approve a
quetiapine ANDA referencing Seroquel that omitted them”); Defs.’ Mem. at 3 (noting that “the
safety information in Table 2 is necessary for safe use of the product and therefore cannot be
carved out . . . ”).
4
3. Exclusivity Periods
Since “Congress still wanted to provide incentives for new drug development, alongside
the ANDA process that eased the marketing of generic drugs, Hatch-Waxman entitles an NDA
applicant to a period of market exclusivity (3 or 5 years, depending on the degree of innovation
reflected in the NDA) . . . .” ViroPharma, 2012 U.S. Dist. LEXIS 56128, at *7 (quoting
Allergan, Inc., 398 F. Supp. 2d at 17). During an exclusivity period, the FDA is barred from
approving a generic ANDA for the NDA product. See id. (citing 21 U.S.C. §§ 355(c)(3)(D)(ii)-
(iv), (j)(5)(D)(ii)-(iv)). In this case, for example, following Seroquel’s approval on September
26, 1997, the FDA granted AstraZeneca a five-year period of “new chemical entity” exclusivity
for Seroquel. Defs.’ Mem. at 8.
Pioneer drugs may also be eligible under statutorily prescribed circumstances for
additional periods of exclusivity on the basis of medical studies completed after the drug
approval process. These additional exclusivity periods “provide incentives to pioneer companies
to conduct new clinical investigations [for] previously approved NDAs, including through
‘supplemental’ NDAs (‘sNDAs’).” Compl. ¶ 32 (citation omitted).
The statutory provision at issue in this case, 21 U.S.C. § 355(j)(5)(F)(iv), describes one
such circumstance, for new indications or uses of the already approved pioneer drug. This
section provides:
If a supplement to an application approved under subsection (b) is approved after
the date of enactment of this subsection [enacted Sept. 24, 1984] and the
supplement contains reports of new clinical investigations (other than
bioavailability studies) essential to the approval of the supplement and conducted
or sponsored by the person submitting the supplement, the Secretary may not
make the approval of an [ANDA] . . . for a change approved in the supplement
effective before the expiration of three years from the date of the approval of the
supplement under subsection (b).
5
21 U.S.C. § 355(j)(5)(F)(iv). 3
Three-year exclusivity under this statutory provision is sometimes referred to as “new
indication exclusivity” or “new patient population exclusivity” because it often applies to
applications for approval of the use of an already-approved drug for a new medical indication,
such as to treat a different disorder, or a new population of patients, such as a new age group.
See Defs.’ Mem. at 6 (citation omitted); ViroPharma, 2012 U.S. Dist. LEXIS 56128, at *8-9.
Under FDA regulations, the FDA will not approve an ANDA for three years following the grant
of exclusivity to a pioneer drug if the ANDA “relies on . . . information supporting a change
approved in the supplemental new drug application.” 21 C.F.R. § 314.108(b)(5)(ii). In this case,
for example, the FDA has granted AstraZeneca exclusivity over two pediatric indications for
Seroquel until December 2, 2012. Therefore, any generic version of Seroquel approved before
that date may not be marketed as a drug for the pediatric indications for which AstraZeneca
retains exclusivity. 4
Under section 355(j)(5)(F)(iv), approval for a new use of a drug must be predicated on
new clinical investigations. The FDA defines “new clinical investigation” in its implementing
regulation, 21 C.F.R. § 314.108(a), as “an investigation in humans the results of which have not
been relied on by [the] FDA to demonstrate substantial evidence of effectiveness of a previously
approved drug product for any indication or of safety for a new patient population and do not
duplicate the results of another investigation that was relied on by the agency to demonstrate the
effectiveness or safety in a new patient population of a previously approved drug product.” 21
C.F.R. § 314.108(a). The FDA elaborates that “data from a clinical investigation previously
3
The FDA notes that “[o]ther exclusivities apply to products that are new chemical entities, or for studies
undertaken for original approval. See 21 U.S.C. § 355(j)(5)(F).” Defs.’ Mem. at 5 n.4. The only exclusivity period
at issue here, however, is the three-year period of exclusivity under 21 U.S.C. § 355(j)(F)(iv). See id.
4
An additional six-month period of pediatric exclusivity, until June 2, 2013, applies to Seroquel but that period is
not at issue here. See Defs.’ Mem. at 1.
6
submitted for use in the comprehensive evaluation of the safety of a drug product but not to
support the effectiveness of the drug product would be considered new.” Id. In other words,
data from a previously submitted clinical investigation may nonetheless be considered “new” if
the previous submission was to support the safety of indications already approved in the NDA
and the data is later presented in a supplemental NDA to show the effectiveness of the drug for
new populations or new indications. Moreover, under section 355(j)(5)(F)(iv), the new clinical
investigations must be “essential to the approval of the supplement.” In its regulations, the FDA
explains that “[e]ssential to approval means, with regard to an investigation, that there are no
other data available that could support approval of the application.” 21 C.F.R. § 314.108(a).
According to the FDA, the new indication exclusivity regulation, 21 C.F.R. § 314.108,
when read in context with the definition of “new clinical investigation,” “requires a relationship
between the information from the new clinical investigation, the change to the product or to the
use of the product approved in the supplement, and the scope of any resulting three-year
exclusivity.” Defs.’ Mem. at 7. Thus, “[i]n accordance with the statute and regulation, the scope
of three-year exclusivity depends on the nexus between the subject of the new clinical
investigations and the changes to the product that the investigations support.” Id.; see also
ViroPharma, 2012 U.S. Dist. LEXIS 56128, at *9 (“The FDA has interpreted [§ 355(j)(5)(F)(iv)]
as establishing a relationship between the information obtained from the clinical investigation,
the change approved through the pioneer drug company’s [sNDA], and the scope of the
information relied upon by a generic competitor in a specific ANDA.”) (quoting AstraZeneca,
2012 U.S. Dist. LEXIS 39611, at *3).
The FDA has issued additional regulations on the implementation of 21 U.S.C. §
355(j)(5)(F)(iv). While not defining all types of changes approved in a supplement warranting 3-
7
year exclusivity, the FDA discussed limits on the scope of this statutory provision. For example,
during consideration of proposed implementing regulations, the FDA received comments
requesting clarification “whether a clinical investigation establishing new risks could be eligible
for exclusivity.” Abbreviated New Drug Application Regulations; Patent and Exclusivity
Provisions, 59 Fed. Reg. 50,338, 50,356 (Oct. 3, 1994). The FDA responded that “such studies
would not qualify for exclusivity because ‘protection of the public health demands that all
products’ labeling contain all relevant warnings.’” Id. (quoting preamble to proposed rule
published in 54 Fed. Reg. 28,872, 28,899). The FDA explained that:
Changes that would not warrant exclusivity are, as discussed in the preamble to
the proposed rule, changes in labeling that involve warnings or other similar risk
information that must be included in the labeling of generic competitors.
Applicants obtaining approval for such changes in labeling would, in any event,
have no valid interest in precluding such information from the labeling of other
products.
59 Fed. Reg. 50,338, 50,357.
The FDA further noted that it “does not consider a study to be ‘essential to approval’
simply because the applicant conducted it and submitted the study for agency review . . . .” Id.
Rather, citing the legislative history, the FDA stated that 3-year exclusivity is reserved for
investigations “that are necessary for approval of important innovations,” and require “a
considerable investment of time and money.” Id. at 50,358. According to the FDA, “an
applicant is not entitled to 3-year exclusivity merely because it supplements an approved
application based in part on a clinical investigation or because it certifies to FDA that the clinical
investigation is essential to approval of the application or supplement.” Id. 5 In short, the FDA
5
The FDA has consistently indicated that only “significant changes in already approved drug products, such as a
new use, which require new clinical studies” are covered by the 3-year exclusivity provisions. 54 Fed. Reg. 28,872,
28,896 (July 10, 1989). “Congress understood that the substantial economic rewards of exclusivity might well
encourage drug companies to make minor and unimportant alterations in their marketed drug products or to conduct
additional tests which they could claim provide important new information about a marketed drug product. To
avoid rewarding such behavior, the 3-year provision includes the special criteria intended to restrict eligibility to
8
regulations make clear that 3-year exclusivity is not triggered merely by labeling changes related
to the safety or risks posed by the drug for indications already approved; such changes, if known,
would have been incorporated into the original labeling at the time of the approval of the original
NDA. Nor is a 3-year period of exclusivity triggered by the simple submission of new clinical
investigations or on the applicant’s “say-so.”
B. CASE-SPECIFIC BACKGROUND
AstraZeneca developed and now manufactures the drug Seroquel, which was first
approved by the FDA as an NDA (NDA 20639) on September 26, 1997. See AR 70-81 (FDA
Approval, dated Sept. 26, 1997, of Seroquel, NDA 20639); see also Defs.’ Mem. at 8 (citing
Orange Book). Seroquel is an atypical antipsychotic medication that is used to treat a variety of
psychological disorders, including schizophrenia and bipolar disorder. 6 Although more than 50
generic atypical antipsychotics have FDA approval, Compl. ¶ 39, Seroquel has been marketed
without generic competition for the past fourteen years. See AR 66-67 (Orange Book); see also
Defs.’ Mem. at 8.
Since Seroquel’s approval in 1997, AstraZeneca has filed multiple supplemental
applications (“sNDAs”). Based on the approval of those sNDAs, Seroquel has been approved
for multiple new medical indications and new patient populations. Although originally used
solely for the treatment of schizophrenia, through several sNDAs, Seroquel is now approved for
use with the following medical indications and populations: “(1) in adults and adolescents (ages
13 to 17) to treat schizophrenia; (2) in adults, adolescents and children (ages 10 to 17) for the
significant innovations. See Cong. Rec. H9114, 9124 (daily edition September 6, 1984) (statement of
Representative Waxman); Cong. Rec. S10505 (daily edition August 10, 1984) (statement of Senator Hatch).” Id.
6
“The term ‘atypical antipsychotic’ refers to a class of drugs that, in comparison to the prior generation of
antipsychotic drugs, effectively treats mental disorders while presenting a reduced tendency to cause significant side
effects known as extrapyramidal symptoms—involuntary movement disorders such as tics, tremors and writhing.”
Compl. ¶ 38.
9
acute treatment of manic episodes associated with bipolar I disorder, both as a monotherapy and
as an adjunct to lithium or divalproex; (3) in adults as a monotherapy for the acute treatment of
depressive episodes associated with bipolar disorder; and (4) in adults for the maintenance
treatment of bipolar I disorder, as an adjunct to lithium or divalproex.” Defs.’ Mem. at 9; see AR
66-69 (Orange Book); see also Compl. ¶¶ 21, 42.
AstraZeneca’s labeling has also changed multiple times since Seroquel’s original
approval. The labeling changes fall into two main categories. First, as the FDA has approved
supplements to Seroquel’s NDA, the newly-approved medical indications and patient
populations (“with accompanying three-year exclusivity periods”) have been added to Seroquel’s
labeling. Defs.’ Mem. at 8-9 (citing AR 66-69 (Orange Book)). Second, changes involving new
safety information have been made to the labeling. These additional labeling changes, according
to the FDA, “have not resulted in exclusivity.” Id.
Atypical antipsychotics, such as Seroquel, may have numerous side effects, including
hyperglycemia. To minimize the risk of hyperglycemia in users of atypical antipsychotics, the
FDA has investigated the metabolic changes caused by that class of medications. See Compl. ¶
51. To this end, the FDA has required manufacturers of atypical antipsychotic drugs to provide
data and has mandated labeling changes. See id. ¶¶ 50-54. In 2000, the FDA performed a
“comprehensive review” of pre-clinical, clinical, and post-marketing data to see whether atypical
antipsychotics disturb glucose regulation. Id. ¶ 51. After reviewing the entire class of atypical
antipsychotics, in 2001, the FDA announced that “further study will be needed to elucidate the
potential causality of [diabetes mellitus] by” atypical antipsychotics. Id. In September 2003,
after years researching the issue, the FDA “mandated a class-wide diabetes/hyperglycemia label
change for all atypical antipsychotics,” including Seroquel. Id.
10
The FDA’s concern about the effects of atypical antipsychotics on diabetes and
hyperglycemia has prompted the agency to require AstraZeneca and other manufacturers
continually to update the labels for their atypical antipsychotics with warning information and
data that informs prescribers about possible metabolic effects, including glucose shift data. 7 Id.
¶ 52. The FDA intended to use the glucose shift data to “‘create a universal format for
presentation of metabolic information in the atypical package insert labels,’” thereby
standardizing the hyperglycemia warning across the entire class of atypical antipsychotic
medications. Id. ¶ 53 (quoting the FDA); see also AR 325-39 (FDA Letter, dated Jan. 8, 2008,
to AstraZeneca, referencing original NDAs for Seroquel and Seroquel XR and requesting
metabolic data analyses for the FDA to evaluate the “effects of atypical antipsychotic drugs on
metabolic parameters (e.g., weight, lipids, glucose)”); AR 21 (FDA Email, dated July 22, 2011,
to AstraZeneca, stating that “[t]he Division has been working with sponsors class-wide to create
a universal format for presentation of metabolic information in the atypical package insert
labels.”).
As the FDA considered the metabolic data issue for the entire class of antipsychotic
drugs, AstraZeneca continued to market Seroquel and invest in research and development to find
new indications, patient populations, and forms of Seroquel. In 2007, the FDA approved a new
drug application for Seroquel XR, an extended-release tablet version of Seroquel that only had to
be taken once a day, unlike Seroquel, which had to be taken two to three times per day. Compl.
¶¶ 44-45. 8
7
Glucose shift data describes “the frequency with which patients shift, from beginning to end of treatment, from a
state of normal or borderline glucose levels to a state of hyperglycemia.” Compl. ¶ 52.
8
The Complaint discusses at length Seroquel XR® extended-release tablets (“Seroquel XR”) and seeks the same
declaratory and injunctive relief it seeks with respect to Seroquel. Compl. at 25-26 (seeking, for example, “[a]
permanent injunction prohibiting FDA from issuing final approval of any ANDA for which Seroquel or Seroquel
XR is the reference-listed drug, and vacating and rescinding any final approvals that have issued, until after
11
The FDA continued to focus attention on hyperglycemia safety information related to
both Seroquel and Seroquel XR and the class of atypical antipsychotics. Although Seroquel
already had a label with a general risk statement about hyperglycemia and diabetes, see AR 883,
the FDA was interested in displaying more specific hyperglycemia clinical data. In a letter,
dated January 8, 2008, the FDA requested that AstraZeneca provide tables of data with
summaries of clinical trials related to metabolic parameters for both Seroquel and Seroquel XR.
The FDA requested that this information from various clinical trials “be submitted in stages . . .
as they are completed.” AR 326-38, 338 (FDA letter, dated Jan. 8, 2008, to AstraZeneca,
requesting metabolic data analyses); see also AR 295 (FDA Letter, dated Mar. 27, 2012, to
AstraZeneca, noting that the “FDA requested glucose-related metabolic data for Seroquel by
letter dated January 8, 2008”); Defs.’ Mem. at 9-10.
On June 26, 2008, in response to the FDA’s letter requesting tables summarizing
metabolic data, AstraZeneca submitted the data to the FDA from which Table 2 was eventually
derived, expressly referencing the original NDAs for Seroquel and Seroquel XR. AR 431
(AstraZeneca Letter, dated June 26, 2008, to FDA); see also AR 295 (FDA Letter, dated Mar.
27, 2012, to AstraZeneca); AR 883 (Internal FDA Consultative Review, dated Mar. 27, 2012, to
December 2, 2012”). Yet, Seroquel XR is not at issue here. Contrary to the allegations in the Complaint that, on
March 27, 2012, the FDA granted final approval to “one or more abbreviated new drug applications (‘ANDAs’) for
generic versions of Seroquel and Seroquel XR,” id. ¶ 3, the FDA expressly stated that its decision to grant approvals
to ANDAs relates only to Seroquel. See AR 303 at n.20 (FDA Letter, dated Mar. 27, 2012, to AstraZeneca,
explaining that “[t]oday’s decision relates only to Seroquel. For Seroquel XR, there continue to be multiple
overlapping exclusivities, two of which expire on April 8, 2012, as well as patent protections. These protections
could have implications for carve-out decisions made by a line-by-line review of product labeling which involves
consideration of issues beyond Table 2.”); see also Defs.’ Mem. at 12 n.12 (“The agency did not issue any ANDA
approvals or decision for Seroquel XR [on March 27, 2012].”). Indeed, the parties’ briefs make clear that only
exclusivity for Seroquel is at issue in this case and is the focus of AstraZeneca’s motion for summary judgment. See
generally Pl.’s Mem; Pl.’s Reply at 2 n.1 (noting that “because FDA has only granted final approvals for generic
versions of Seroquel, AstraZeneca focuses here on the December 2, 2009 approval for Seroquel.”). Moreover,
AstraZeneca only asks the Court to “(a) vacate FDA’s approvals of any ANDA for which Seroquel is the reference-
listed drug; and (b) declare that FDA could not have lawfully granted approval of any ANDA for which Seroquel is
the reference-listed drug prior to December 3, 2012.” Pl.’s Mem. at 23; Pl.’s Reply at 23. Accordingly, this Court
addresses AstraZeneca’s claims with respect to the exclusivity of Table 2 only for Seroquel. Nevertheless, to the
extent that AstraZeneca seeks in the Complaint to bar FDA approval for ANDAs for Seroquel XR on the basis of its
marketing exclusivity for labeling containing Table 2, those claims are bound by the ruling here.
12
the FDA Office of Generic Drugs (“OGD”), analyzing information in Table 2). In its letter,
AstraZeneca noted that the data “follows the criteria as specified in the January 2008 letter, and
subsequent clarifications” from the FDA. AR 431. The FDA explains that this data, submitted
to the FDA in a letter by AstraZeneca, was “coded by FDA as general correspondence, not as a
prior approval supplement (PAS) or a Changes Being Effected (“CBE”) supplement to the
Seroquel NDA.” AR 295 (FDA Letter, dated Mar. 27, 2012, to AstraZeneca).
Although the data submitted by AstraZeneca in June 2008 are not part of the
Administrative Record, 9 the parties do not dispute that this data consisted of fifteen clinical
trials, three of which were conducted with Seroquel alone for the treatment of bipolar depression
or disorder; two were conducted with both Seroquel and Seroquel XR for schizophrenia; and ten
were conducted with Seroquel XR alone, including six trials for the treatment of bipolar disorder
and major depressive disorder (“MDD”). AR 309. None of these clinical trials were conducted
on pediatric patients or for purposes of generating data for Table 2. AR 301; Defs.’ Mem. at 10.
AstraZeneca describes only seven of the trials as “new” because they had not been previously
submitted to the FDA. Pl.’s Mem. at 3 (Statement of Facts, ¶ 4).
Almost four months after submission of the data related to metabolic changes “in Patients
Receiving Quetiapine,” AR 431, which led to Table 2, on October 28, 2008, AstraZeneca sought
two new pediatric indications for Seroquel by submitting formal supplemental applications,
referred to as “S-045” and “S-046,” respectively, for: (1) adolescents (ages 13 to 17) to treat
schizophrenia, and (2) adolescents and children (ages 10 to 17) for the acute treatment of manic
episodes associated with bipolar I disorder. AR 95. These sNDAs appear to have been the
subject of multiple years of study. As early as 2001, AstraZeneca proposed a pediatric study
9
The table of contents for the Supplemental Administrative Record notes, in connection with AstraZeneca’s June
26, 2008 letter, that the “06/13/2008 data submission of 1152 pages [is] omitted.” ECF No. 14-1 at 3.
13
request to the FDA and, in 2003, the agency requested that the results “from trials in pediatric
patients with (1) schizophrenia, and with (2) acute mania, as part of bipolar disorder” be
submitted within 5 years. See AR 311-24, 312 (FDA Letter, dated Feb. 11, 2003, to
AstraZeneca, requesting pediatric clinical trial information and making “[r]eference . . . to
[AstraZeneca’s] Proposed Pediatric Study Request submitted on March 2, 2001, to
[AstraZeneca’s] New Drug Application for Seroquel (quetiapine fumarate) tablets.”).
When AstraZeneca submitted its formal applications for FDA approval to market
Seroquel for two new pediatric indications, AstraZeneca expressly requested 3-year exclusivity
for those pediatric indications. See AR 433-34 (AstraZeneca’s claim (undated) for three-year
exclusivity for supplemental new drug application for pediatric indications, stating: “The new
clinical investigation(s) provide safety and efficacy data regarding the use of Seroquel . . .
Tablets for the treatment of bipolar mania in pediatric patients ages 10-17 and schizophrenia in
pediatric patients ages 13-17 that could not be gleaned from published information.
Accordingly, these new clinical investigations are essential to the approval of this supplemental
new drug application.”). By contrast, the Administrative Record contains no explicit request
from AstraZeneca for 3-year exclusivity for the addition of Table 2 to Seroquel labeling.
The FDA considered AstraZeneca’s formal supplemental applications for pediatric
indications of Seroquel while simultaneously continuing to refine the labeling of antipsychotic
drugs, including Seroquel, with respect to metabolic effects for all approved indications. Until
the end of 2008, the FDA and AstraZeneca addressed the sNDAs for the approval of pediatric
indications and the class-wide metabolic data separately in their communications. AstraZeneca’s
applications for the pediatric indications, as noted, had been submitted as formal supplements to
the Seroquel NDA (an “sNDA”) on October 28, 2008 and AstraZeneca made eleven subsequent
14
submissions particularly regarding the pediatric supplements. See AR 95 (FDA Letter, dated
Dec. 2, 2009, to AstraZeneca approving sNDAs 045 and 046 and summarizing submissions
relevant to approval). Moreover, in its earlier request for exclusivity for the pediatric
indications, AstraZeneca discussed clinical trials it believed were essential to approval, but did
not mention Table 2. AR 433-34.
While AstraZeneca and the FDA addressed in the same correspondence multiple pending
issues related to Seroquel and Seroquel XR, the agency viewed the labeling changes prompted
by metabolic data and the new pediatric indications as distinct and separate issues, which were
being considered independently. Nothing in the Administrative Record suggests that the
exclusivity periods for which AstraZeneca had applied for the pediatric indications would be
extended to include other changes being negotiated between the agency and the pharmaceutical
company.
For example, shortly after submission of the sNDAs for the new pediatric indications, the
FDA sent a letter, dated December 18, 2008, to AstraZeneca directing the company to include
additional data regarding glucose levels in the labeling for Seroquel and Seroquel XR and to
“elevat[e]” the data for “glucose changes . . . from the clinical trials [from the Adverse Reactions
section] to the Warnings/Precautions section of labeling.” AR 11; Pl.’s Mem. at 16. The FDA
further directed AstraZeneca to use the label of another antipsychotic drug, Zyprexa, as a model
for the correct formatting of the label. AR 11. The FDA commented that it was “currently
reviewing [AstraZeneca’s] metabolic data submission and the pediatric efficacy supplements
submitted under this NDA (S-045 and S-046),” evidently treating them as separate submissions,
and noting that the agency would “be providing further labeling comments . . . .” AR 12; see
also, e.g., AR 706-09 (FDA Pediatric Exclusivity Determination Checklist, dated Jan. 21, 2009,
15
referring only to the pediatric indications, supplements #045 and #046, and not discussing Table
2); AR 710 (AstraZeneca Letter, dated Feb. 12, 2009, to FDA, updating tables with changes in
metabolic parameters in response to the FDA’s February 2009 request); AR 711-844, 718 (FDA
Clinical Review on Metabolic Parameters (Hyperglycemia, Hyperlipidemia, and Weight Gain),
dated Mar. 26, 2009, relating to adult data and noting that the pediatric data is being reviewed
separately); AR 845-52, 849 (NDA Regulatory Filing Review, dated Apr. 22, 2009, noting as
background in a “Memo of Filing Meeting,” with no mention of Table 2, that “[t]hese
supplements (S045 / S046) include data to support the use of Seroquel (molecular entity was
approved in 1997) for the following pediatric indications, schizophrenia (13-17 years of age) and
bipolar mania (10-17 years of age). The supplements are in response to a written request issued
on 2/11/2003.”); AR 853-65, 860 (FDA Memorandum, dated Aug. 13, 2009, recommending
approval of NDA supplements for pediatric indications and noting as a “Comment” in the text
that “[t]he Division also requested that the sponsor conduct an analysis of all clinical trials to
study [metabolic effects] . . . . The sponsor has recently submitted these data for both pediatric
and adult population. Further modifications to product labeling will be made based on our
review of these submitted data (refer to separate metabolic reviews).”). Thus, the
communication between the agency and AstraZeneca in no way suggested that AstraZeneca was
eligible for exclusivity for its response to the agency’s requests for metabolic data.
On October 16, 2009, the FDA requested that AstraZeneca “[p]lease include a table
summarizing the shift changes from normal to high fasting glucose and from borderline to high
fasting glucose for the short-term, placebo-controlled clinical trials in adults.” AR 15-19, 18
(FDA Email, dated Oct. 16, 2009, to AstraZeneca, forwarding attachment with labeling
changes). The agency also instructed AstraZeneca to “[p]lease refer to current Zyprexa labeling
16
for examples of these tables.” Id. AstraZeneca responded, in a letter on November 18, 2009,
stating that it was “providing a response to the proposed label and Medication Guide,” and
forwarding an “Amendment to a Pending Application.” AR 1969. As part of this submission,
AstraZeneca included draft labeling for Seroquel that included not only metabolic data in Table 2
but also referred to many other labeling changes. Tellingly, in track changes, the FDA referred
to labeling changes in connection with multiple pending applications, not just sNDAs 045 and
046, the supplemental applications for the pediatric indications. See AR 1973 (Apparent FDA
note in draft labeling referring to “your labeling changes submitted under S-042, 044, 045, 046
and 048. In this version, we have made additional modifications in several sections. We have
included bracketed comments to note these changes or request additional revisions where
needed.”). Thus, it appears that the FDA was communicating with AstraZeneca about numerous
proposed changes all at once, with no suggestion that exclusivity was at stake for all of the many
changes proposed in track changes in the draft label.
As noted, one of the many proposed changes in the draft label was Table 2. See AR
1991. Next to Table 2, AstraZeneca noted that it “[a]dded table summarizing the shift changes
from normal to high fasting glucose and from borderline to high fasting glucose as requested by
FDA in the 10/16/09 FDA communication.” Id. The “Source” of Table 2 is listed as “NDA 20-
639, Metabolic Response, submitted 26 June 2008, Table #339.” Id. Thus, the November 2009
draft labeling proposal from AstraZeneca demonstrates that the addition of Table 2 was derived
from its submission to the FDA on June 26, 2008, almost four months before it submitted
separately its two formal supplemental applications for the pediatric indications.
Less than a month after AstraZeneca transmitted to the FDA track changes in the
proposed labeling, the FDA approved, on December 2, 2009, the pediatric supplemental
17
applications for Seroquel as well as AstraZeneca’s proposed labeling changes, including the
addition of Table 2. AR 95-99 (FDA Letter, dated Dec. 2, 2009, to AstraZeneca, approving
supplemental new drug applications for pediatric indications) (“Approval Letter”). The FDA’s
decision to approve Table 2 for inclusion in the labeling and its decision to approve the pediatric
indications for Seroquel were two separate actions. See id. The FDA, however, communicated
its decisions to approve both the display of class-wide safety information, including Table 2, and
the approval of the pediatric sNDAs in the same letter. Id.
The Approval Letter primarily focused on the approval of the two pediatric sNDAs, for
which AstraZeneca earned 3 years of exclusivity, stating that:
These supplemental new drug applications provide for the use of Seroquel
(quetiapine fumarate) tablets for the treatment of schizophrenia in adolescents 13
to 17 years of age and the treatment of bipolar mania in children and adolescents
10 to 17 years of age. We have completed our review of these applications. They
are approved….
AR 95.
The Approval Letter also noted, however, that the supplemental drug applications “are
approved, effective on the date of this letter, for use as recommended in the enclosed, agreed-
upon labeling text.” Id. (emphasis added). The letter directed that the content of the labeling
shall be formatted in structured product labeling (SPL) format and “[f]or administrative purposes
. . . designate[d] [as] ‘SPL for approved NDA 020639/S-045/S-046.’” AR 95-96.
In a separate section of the Approval Letter, captioned “Risk Evaluation and Mitigation
Strategy Requirements, the FDA explained that “[s]ince Seroquel (quetiapine fumarate) was
approved on September 26, 1997, [the FDA has] become aware of additional clinical trial data
and postmarketing safety data that show a risk of hyperglycemia, hyperlipidemia and weight gain
associated with all forms of Seroquel (quetiapine fumarate) in all patient populations. [The FDA
18
considers] this information [in Table 2] to be ‘new safety information’ as defined in section 505-
1(b) of FDCA.” AR 96 (Approval Letter). Immediately following that statement, the letter
states that AstraZeneca’s proposed REMS (Risk Evaluation and Mitigation Strategy), “submitted
on October 22, 2009 . . . is approved.” Id. Attached to the Approval Letter is the approved
labeling showing Table 2 in the “Adults” rather than the “Children and Adolescents” section of
the document. AR 112-13.
Although the FDA expressed approval for both the new pediatric uses and the Table 2
labeling change in a single letter, each change was approved on its own merit. The FDA notes
that it “consolidated a number of . . . actions” referred to in the approval letters sent to
AstraZeneca for the two new pediatric indications (S-45, S-46). See AR 295 n.9 (FDA Letter,
dated March 27, 2012, to AstraZeneca). Support for this assertion is found, inter alia, in an
internal FDA Memorandum, noting that “the addition of Table 2 was not related to approval of
these new indications. It is not unusual for DPP [the Division of Psychiatry Products] to bundle
actions together . . . . [M]odifications to product labeling usually occur in concert with other
actions being taken.” AR 883 (Internal FDA Consultative Review, dated Mar. 27, 2012, to OGD
regarding, inter alia, quetiapine ANDAs and information included in Table 2); see also AR 866-
68 (FDA Internal Memorandum, dated Dec. 2, 2009, recommending approval of the pediatric
sNDAs based on pediatric clinical trial results and the opinion of the Psychopharmacologic
Drugs Advisory Committee that the efficacy and safety of Seroquel had been established, and
discussing labeling changes but with no mention of Table 2); see also AR 869-76, 873 (FDA
Exclusivity Summary, dated Dec. 11, 2009, for Seroquel Tablets, noting the three “clinical
investigations submitted in the [pediatric] application that are essential to approval,” none of
which are Table 2 or contributed to the creation of Table 2).
19
The FDA’s communications following the December 2, 2009 approval of the pediatric
indications and proposed labeling only reinforce that the addition of Table 2 was a decision
distinct from the decision to grant the sNDAs for pediatric indications and related exclusivity,
and was part of the agency’s broader efforts with respect to the provision of safety information
related to the metabolic effects of atypical antipsychotics. In an email from the FDA to
AstraZeneca, on July 22, 2011, for example, the FDA explained that the agency “has been
working with sponsors class-wide to create a universal format for presentation of metabolic
information in the atypical package insert labels.” AR 21 (Email from FDA, dated July 22,
2011, to AstraZeneca). The FDA referenced AstraZeneca’s pending supplements (NDA 20639,
S-053, NDA 22047, S-026) and requested that AstraZeneca “please submit revised labeling to
these supplements that incorporates this new format - Hyperglycemia and Diabetes
Mellitus/Dyslipidemia/Weight Gain as separate headings under one warnings and precaution
section, with the same introductory paragraph preceding the headings, as in the Latuda/Invega
labels.” Id. Thus, the FDA’s efforts to improve metabolic data in the labeling of atypical
antipsychotics continued, apart from the indications for which AstraZeneca was granted
exclusivity on December 2, 2009.
The crux of the instant dispute is that AstraZeneca believes that it is entitled to a 3-year
period of exclusivity for use of Table 2 in Seroquel labeling because Table 2 was a change based
on clinical trials approved as part of a supplemental application. AstraZeneca contends that the
FDA’s approval of ANDAs for generic versions of Seroquel violates its right to exclusive use of
the labeling, and was arbitrary, capricious, and contrary to law. The FDA does not dispute that
Table 2 is essential information for all quetiapine labels, but believes that Table 2 is not entitled
to exclusivity in part because it was not “essential” to approval of the pediatric supplement under
20
21 U.S.C. § 355(j)(5)(F)(iv). See AR 294 (FDA Letter, dated Mar. 27, 2012, to AstraZeneca,
noting that the addition of Table 2 is “essential to safe use of a generic quetiapine product
referencing Seroquel for any indication, and the agency would not approve a quetiapine ANDA
referencing Seroquel that omitted them”); see also AR 882-85 (Internal FDA Consultative
Review, dated Mar. 27, 2012, noting that the “effects of Seroquel on glucose are not specific to a
particular indication and that the data in Table 2 are relevant to any population receiving
quetiapine”).
C. PROCEDURAL BACKGROUND
Under 21 C.F.R. § 10.25(a), “[a]n interested person” may petition the FDA “to issue,
amend, or revoke a regulation or order, or to take or refrain from taking any other form of
administrative action.” See AR 277-92 (Guidance for Industry: Citizen Petitions and Petitions
for Stay of Action Subject to Section 505(q) of the Federal Food, Drug, and Cosmetic Act, dated
June 2011). In September of 2011, AstraZeneca filed two “Citizen Petitions” with the FDA
requesting that the agency not grant final approval to any ANDA based on Seroquel or Seroquel
XR unless the labeling of the ANDA includes the labeling that the FDA has required of
AstraZeneca, including Table 2. See AR 1-21 (AstraZeneca Citizen Petition regarding Seroquel,
dated Sept. 2, 2011) (“Seroquel Citizen Petition”); AR 28-59 (AstraZeneca Citizen Petition
regarding Seroquel XR, dated Sept. 2, 2011) (“Seroquel XR Citizen Petition”); see also Compl. ¶
6. 10 More to the point, AstraZeneca argued in its Citizen Petitions that Table 2 was entitled to
exclusivity based on clinical trials that AstraZeneca performed for Seroquel XR that were
essential to the approval of the addition of Table 2 to labeling for both Seroquel and Seroquel
10
AstraZeneca also asked the FDA not to approve generic quetiapine fumarate tablets if the labeling omitted two
other warnings regarding increased suicidality in children, adolescents, and young adults and the clinical worsening
of depression and the risk of suicide. Neither of these warnings are at issue here.
21
XR. See AR 2, 4-6 (Seroquel Citizen Petition) (“Because these new clinical investigations with
Seroquel XR for the treatment of bipolar disorder and MDD [major depressive disorder] were
essential to the approval of the labeling supplement for Seroquel, the labeling is entitled to three
years of exclusivity in accordance with Section 505(j)(5)(F)(iv) of the FDCA.”); see also Defs.’
Mem. at 3. 11
The FDA denied AstraZeneca’s Citizen Petitions “without comment” on March 7, 2012,
the last day of the 180-day period in which the FDA was statutorily required to respond. AR 23-
27 (FDA Letter, dated March 7, 2012, to Covington and Burling LLP, noting that “[w]e have
carefully considered the information submitted in the Petitions. For the reasons stated below, the
Petitions are denied without comment on whether we will take the actions that you request.”)
(“Citizen Petition Denial”); see also 21 U.S.C. § 355(q); Defs.’ Mem. at 11. Although the FDA
acknowledged AstraZeneca’s assertion of exclusivity periods over the labeling, the agency
ducked this issue, merely stating that “[t]he periods of exclusivity described above for Seroquel
and Seroquel XR may or may not apply or be relevant to the Agency’s final decisions with
respect to any individual application and its labeling depending on the particulars of an ANDA
and the timing of its approval.” AR 26 (Citizen Petition Denial); Compl. ¶¶ 7-8 (citation
omitted). Since the FDA had not yet approved any generic versions of Seroquel, the FDA noted
that it would not “be appropriate” to address the merits of AstraZeneca’s claims. AR 26.
11
As the FDA points out, the “theory for exclusivity that AstraZeneca raised in its citizen petition, which was that
Table 2 was entitled to three-year exclusivity based on clinical trials that AstraZeneca had performed for a different
drug product [i.e. Seroquel XR],” differs from the position it asserts now. Defs.’ Mem. at 3; see also AR 6
(Seroquel Citizen Petition, arguing that “the data and information in Table 2 cannot be included in the labeling of a
generic version of Seroquel until the exclusivity periods for the Seroquel XR bipolar disorder and MDD indications
have expired”). Rather than rely on Seroquel XR’s sNDA for new indications of MDD and bipolar disorder,
AstraZeneca is now hitching its proverbial wagon to the Seroquel sNDAs for new pediatric indications, as well as
the approval of Table 2 itself, as the bases for exclusive use for 3 years of Table 2.
22
The FDA conveniently conflated the legal issues raised by the Citizen Petitions: namely,
(1) whether the agency would approve ANDAs for generic versions of Seroquel without the
identical labeling required to be used by AstraZeneca, and (2) whether the agency concurred in
AstraZeneca’s legal judgment that Table 2 was derived from “protected data” that was subject to
exclusivity periods, which did not expire until December 2, 2012. While the former issue was
dependent upon the specific ANDA application pending before the FDA, the second issue could
have been addressed even if no ANDA were ever approved.
Less than one week later, following the FDA’s denial of AstraZeneca’s Citizen Petitions,
AstraZeneca filed a Complaint together with a Motion for a Preliminary Injunction in the District
Court for the District of Columbia. AstraZeneca sought to prevent the FDA from granting final
approval to ANDAs to manufacture generic forms of Seroquel. See AstraZeneca Pharms. LP v.
FDA, No. 12-388, 2012 U.S. Dist. LEXIS 39611 (D.D.C. Mar. 23, 2012). The case was assigned
to another Judge on this court and, on March 12, 2012, that Judge denied the Motion for
Preliminary Injunction and dismissed the case as unripe because the FDA had not yet decided to
grant approval to any ANDA for a generic version of Seroquel. See id. at *55-56.
Only four days after AstraZeneca’s case was dismissed, on March 27, 2012, the FDA
granted final approval to eleven ANDAs for generic versions of Seroquel. Pl.’s Mem. at 3 ¶ 6;
Compl. ¶ 11; Defs.’ Mem. at 12. 12 On the same day, the FDA also issued a letter to AstraZeneca
explaining its reasoning for granting approval to the ANDAs, stating in relevant part:
12
The approval letters for the eleven ANDAs, all approved on March 27, 2012, may be found in the Administrative
Record. See AR 978-81 (Accord Healthcare Inc.); AR 1010-13 (Apotex Corp.); AR 1085-88 (Aurobindo Pharma
USA, Inc.); AR 1121-24 (Caraco Pharmaceutical Labs, Ltd.); AR 1171-74 (Dr. Reddy’s Laboratories, Inc.); AR
1241-44 (Mylan Pharmaceuticals Inc.); AR 1253-56 (Lupin Pharmaceuticals Inc.); AR 1321-24 (Roxane
Laboratories, Inc.); AR 1358-61 (Roxane Laboratories, Inc.); AR 1377-80 (Torrent Pharma, Inc.); AR 1429-32
(Teva Pharmaceuticals USA). Certain of these ANDAs had been pending before the FDA with tentative approvals
granted several years before the final approval. See, e.g., AR 1429 (FDA Letter, dated Mar. 27, 2012, to Teva
Pharmaceuticals USA, approving ANDA for quetiapine fumarate tablets and referring to “the tentative approval
letter issued by this office on December 22, 2008” and “your amendments dated August 2, and December 23,
23
In sum, FDA concurs that these portions of the labeling [including Table 2] are
essential to safe use of a generic quetiapine product referencing Seroquel for any
indication, and the agency would not approve a quetiapine ANDA referencing
Seroquel that omitted them. FDA does not concur, however, that an ANDA
referencing Seroquel is precluded from including Table 2 or the suicidality
warnings by virtue of AstraZeneca’s 3-year exclusivity on certain indications for
Seroquel XR.
AR 305-06 (FDA letter, dated Mar. 27, 2012, to AstraZeneca). The FDA provided several
explanations for finding that Table 2 was not protected by a period of exclusivity, including,
inter alia, (1) that the data was not specific to any indication and generally changes in labeling
that involve the addition of warnings are not entitled to 3-year exclusivity; (2) Table 2 does not
include data from indications for which Seroquel has 3-year exclusivity and does not include any
pediatric data; and (3) the “coincidental” timing of the addition of Table 2 being approved at the
same time as the pediatric supplements does not mean that the labeling change merits a period of
exclusivity. See AR 301-03.
Following the FDA’s approval of the eleven ANDAs, on March 28, 2012, AstraZeneca
again filed a Complaint in the District Court for the District of Columbia, as well as a Motion for
a Temporary Restraining Order, seeking to (1) vacate the FDA’s final approval of ANDAs for
which Seroquel and Seroquel XR are the reference-listed drugs and (2) enjoin the FDA from
granting any other such final approvals pending the Court’s resolution of AstraZeneca’s Motion
for a Preliminary Injunction. Pl.’s Mot. for Temp. Restraining Order (TRO) (“Pl.’s Mot. for
TRO”), ECF No. 3, at 1; Compl. ¶ 85. AstraZeneca argues that “[a]bsent immediate relief from
2011”); AR 1171 (Letter, dated Mar. 27, 2012, from FDA to Dr. Reddy’s Laboratories, referring to “the tentative
approval for your Quetiapine Fumarate Tablets . . . issued by this office on June 17, 2009 . . . .”).
24
the Court, FDA’s final approval of these generics before they are lawfully subject to final
approval could cost AstraZeneca in the range of $2 billion in lost revenues.” Compl. ¶ 18.
This Court denied Plaintiff’s motion for a Temporary Restraining Order on March 28,
2012, finding that AstraZeneca had not shown a likelihood of success on the merits. See
Astrazeneca Pharms. LP v. FDA, No. 12-472, 2012 U.S. Dist. LEXIS 54863, at *7-10 (D.D.C.
Mar. 28, 2012). 13
Following this Court’s denial of AstraZeneca’s Motion for a Temporary Restraining
Order, the parties supplemented the Administrative Record and briefed the pending Cross-
Motions for Summary Judgment. 14 The parties’ Cross-Motions for Summary Judgment are now
before this Court.
II. STANDARD OF REVIEW
A. SUMMARY JUDGMENT
Pursuant to Rule 56 of the Federal Rules of Civil Procedure, summary judgment shall be
granted “if the movant shows that there is no genuine dispute as to any material fact and the
movant is entitled to judgment as a matter of law.” FED. R. CIV. P. 56(a); Anderson v. Liberty
13
The Court also observed that it appeared from the record that the FDA had made “‘tactical decision[s]’ to prevent
the plaintiff ‘from seeking judicial review of FDA’s legal position,’” including denying AstraZeneca’s Citizen
Petition without comment on the merits on the last day of the comment period and advocating for denial of
AstraZenca’s Motion for a Preliminary Injunction on ripeness grounds when the FDA, only four days following the
court’s denial of the Motion for Preliminary Injunction, gave final approval to the ANDAs for generic versions of
Seroquel. See Astrazeneca Pharms. LP, 2012 U.S. Dist. LEXIS 54863, at *10-12 (quoting Pl.’s Mot. for TRO). The
FDA had the opportunity both at the time of responding to AztraZeneca’s Citizen Petitions and the Motion for
Preliminary Injunction to reveal the agency’s legal position that Table 2, as incorporated in the labeling for
Seroquel, did not warrant a 3-year exclusivity period. Accord CollaGenex Pharms., Inc. v. Thompson, No. 03-1405,
2003 U.S. Dist. LEXIS 12523, at *16-17 (D.D.C. July 22, 2003) (rejecting FDA argument that company’s effort to
prevent approval of ANDA, which was “not quite final,” was unripe because company’s challenge was to agency’s
determination that drug at issue was an antibiotic and not entitled to protections from generic drugs available under
the Hatch-Waxman Amendments rather than an attack on the possible ANDA itself) (emphasis omitted).
14
The Administrative Record in this case consists of 101 documents totaling 2,070 pages. When AstraZeneca filed
its motion for a Temporary Restraining Order before this Court, it filed the identical Administrative Record,
consisting of 38 documents totaling 292 pages, that had been filed before Judge Kollar-Kotelly when she decided
AstraZeneca’s earlier Motion for a Preliminary Injunction. Pl.’s Mot. for TRO, ECF No. 3, Ex. A, at 8-11 (Index of
Administrative Record); AstraZeneca Pharms. LP v. FDA, No. 12-388, 2012 U.S. Dist. LEXIS 39611 (D.D.C. Mar.
23, 2012).
25
Lobby, Inc., 477 U.S. 242, 247 (1986); Estate of Parsons v. Palestinian Auth., 651 F.3d 118, 123
(D.C. Cir. 2011); Tao v. Freeh, 27 F.3d 635, 638 (D.C. Cir. 1994). Summary judgment is
properly granted against a party who “after adequate time for discovery and upon motion . . .
fails to make a showing sufficient to establish the existence of an element essential to that party’s
case, and on which that party will bear the burden of proof at trial.” Celotex Corp. v. Catrett,
477 U.S. 317, 322 (1986). The burden is on the moving party to demonstrate that there is an
“absence of a genuine issue of material fact” in dispute. Id. at 323.
In ruling on a motion for summary judgment, the Court must draw all justifiable
inferences in favor of the nonmoving party, and shall accept the nonmoving party’s evidence as
true. Anderson, 477 U.S. at 255; Estate of Parsons, 651 F.3d at 123; Tao, 27 F.3d at 638. The
Court is only required to consider the materials explicitly cited by the parties, but may on its own
accord consider “other materials in the record.” FED. R. CIV. P. 56(c)(3). For a factual dispute to
be “genuine,” Estate of Parsons, 651 F.3d at 123, the nonmoving party must establish more than
“the mere existence of a scintilla of evidence” in support of its position, Anderson, 477 U.S. at
252, and cannot simply rely on allegations or conclusory statements, Greene v. Dalton, 164 F.3d
671, 675 (D.C. Cir. 1999). Rather, the nonmoving party must present specific facts that would
enable a reasonable jury to find in its favor. See Anderson, 477 U.S. at 250. If the evidence “is
merely colorable, or is not significantly probative, summary judgment may be granted.” Id. at
249-50 (citations omitted). An agency is “entitled to summary judgment if the path of its
reasoning is sufficiently discernible in light of the record.” Settles v. U.S. Parole Comm’n, 429
F.3d 1098, 1108 (D.C. Cir. 2005).
26
B. ADMINISTRATIVE PROCEDURE ACT
Under the Administrative Procedure Act (“APA”), the reviewing Court must set aside
those agency actions that are in excess of an agency’s statutory jurisdiction, authority, or
limitations. 5 U.S.C. § 706(2)(C). In order to determine whether an agency has acted in excess
of its statutory authority, this Court must engage in a two-step inquiry set out in Chevron U.S.A.
Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837 (1984); see also Fox v. Clinton, No. 11-
5010, 2012 U.S. App. LEXIS 11852, at *22 (D.C. Cir. June 12, 2012) (“As a general matter, an
agency’s interpretation of the statute which that agency administers is entitled to Chevron
deference.”).
In Chevron Step One, the Court first asks whether “Congress has directly spoken to the
precise question at issue.” Bhd. of R.R. Signalmen v. Surface Transp. Bd., 638 F.3d 807, 811
(D.C. Cir. 2011) (quoting Chevron, 467 U.S. at 842). If so, the Court “‘must give effect to the
unambiguously expressed intent of Congress.’” Id. (quoting Chevron, 467 U.S. at 843). In
deciding whether a statute is ambiguous, the Court has “a duty to conduct an ‘independent
examination’ of the statute in question.” Martini v. Fed. Nat’l Mortg. Ass’n, 178 F.3d 1336,
1345 (D.C. Cir. 1999) (citation omitted).
“If Congress has not directly addressed the precise question at issue, the reviewing court
proceeds to Chevron Step Two.” Ass’n of Private Sector Colls. & Univs. v. Duncan, Nos. 11-
5174, 11-5230, 2012 U.S. App. LEXIS 11269, at *23 (D.C. Cir. Jun. 5, 2012) (quoting HARRY T.
EDWARDS & LINDA A. ELLIOTT, FEDERAL STANDARDS OF REVIEW — REVIEW OF DISTRICT
COURT DECISIONS AND AGENCY ACTIONS 141 (2007)). In Chevron Step Two, “the question for
the court is whether the agency’s answer is based on a permissible construction of the statute.”
Chevron, 467 U.S. at 843; see also Ne. Hosp. Corp. v. Sebelius, 657 F.3d 1, 4-5 (D.C. Cir. 2011).
27
“The whole point of Chevron is to leave the discretion provided by the ambiguities of a statute
with the implementing agency.” Ass’n of Private Sector Colls. & Univs., 2012 U.S. App. LEXIS
11269, at *24 (citations omitted). In conducting its Chevron Step Two analysis, the Court will
“defer to the agency’s permissible interpretation, but only if the agency has offered a reasoned
explanation for why it chose that interpretation.” Vill. of Barrington, Ill. v. Surface Transp. Bd.,
636 F.3d 650, 660 (D.C. Cir. 2011).
III. DISCUSSION
AstraZeneca contends that the FDA’s final approval of generic versions of Seroquel was
in contravention of 21 U.S.C. § 355(j)(5)(F)(iv), violating AstraZeneca’s exclusivity rights over
Table 2 as established by that section, and was arbitrary, capricious, and contrary to law. Pl.’s
Mem. in Supp. of Mot. for Summ. J. (“Pl.’s Mem.”), ECF No. 21, at 1. Specifically,
AstraZeneca argues that Table 2 is subject to exclusivity under the plain language of 21 U.S.C. §
355(j)(5)(F)(iv) because this table was derived from new clinical investigations and was
“essential” to the changes approved by the FDA as part of the sNDAs approved on December 2,
2009. AstraZeneca asks that the Court enter summary judgment in its favor, vacate the
approvals of generic drugs, and permanently enjoin the FDA from granting any other final
approvals of generic versions of Seroquel before December 3, 2012. Pl.’s Mot. for Summ. J.
(“Pl.’s Mot.”), ECF No. 21, at 1. Since there is no dispute of material fact, and the dispute
centers around a purely legal question of statutory interpretation, it is appropriate to resolve the
case on summary judgment. 15
15
AstraZeneca has requested oral argument on the pending motions for summary judgment. See Pl.’s Mot. at 1.
Having carefully considered the briefs and administrative record, as well as having heard argument by the parties on
the Motion for a Temporary Restraining Order, the Court exercises its discretion to decide the motions on the
papers. LCvR 7(e).
28
AstraZeneca argues in support of its Motion for Summary Judgment that the FDA’s
approval of the ANDAs was unlawful, for two distinct reasons based on the plain and
unambiguous language of 21 U.S.C. § 355(j)(5)(F)(iv). First, AstraZeneca argues that “the
sNDAs approved in December 2009 were supported by new clinical investigations on pediatric
use essential to approval of the sNDAs, and Table 2 was a ‘a change approved in th[os]e
supplement[s].’” Pl.’s Reply at 1. Second, AstraZeneca argues that “separately, considered by
itself, Table 2 contains reports of new clinical investigations essential to approval of the labeling
change supplement required by FDA and approved in December 2009—the addition of a table of
glucose shift data.” Id. (emphasis in original). Since AstraZeneca’s overarching claim, and the
two distinct arguments in support of that claim, focus on statutory interpretation, the Court “must
begin with the first step of the two-part framework announced in Chevron . . . and ask whether
Congress has directly addressed the precise question at issue.” Nat’l Auto. Dealers Ass’n v.
FTC, No. 11-1711, 2012 U.S. Dist. LEXIS 70831, at *12 (D.D.C. May 22, 2012) (citations and
quotation marks omitted). “When determining whether Congress has spoken to the precise
question at issue, courts must first exhaust the traditional tools of statutory construction.” Mylan
Pharms., Inc. v. Sebelius, No. 12-524, 2012 U.S. Dist. LEXIS 56178, at *27 (D.D.C. Apr. 23,
2012) (citations and internal quotation marks omitted). “If, however, the statute is silent or
ambiguous on the specific issue, ‘the question for the court is whether the agency’s answer is
based on a permissible construction of the statute.’” Univ. Med. Ctr., Inc. v. Sebelius, No. 11-
260, 2012 U.S. Dist. LEXIS 53395, at *27 (D.D.C. Apr. 17, 2012) (quoting Chevron, 467 U.S. at
843). “When the agency’s construction of a statute is challenged, its interpretation need not be
the best or most natural one by grammatical or other standards . . . . Rather [it] need be only
reasonable to warrant deference.” Id. (citations and internal quotation marks omitted).
29
As explained above, the threshold question is whether the statute is ambiguous, or
instead, by its plain terms, compels the result urged by AstraZeneca. If the statute is ambiguous,
then the Court must defer to the FDA’s determination, which was well within the agency’s
expertise, so long as its decision was not arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with law. See, e.g., ViroPharma, Inc., 2012 U.S. Dist. LEXIS
56128, at *37 (“It bears emphasis that [i]n an area as complex as the regulatory system for
pharmaceuticals, the agency Congress vests with administrative responsibility must be able to
exercise its authority to meet changing conditions and new problems.”) (citations and internal
quotation marks omitted and alteration in original). Upon review of the administrative record,
and for the reasons explained below, the Court concludes that 21 U.S.C. § 355(j)(5)(F)(iv) is
ambiguous. The FDA has reasonably interpreted and applied the applicable statute, and
therefore its final approval of the ANDAs was not arbitrary, capricious, or an abuse of discretion.
A. CHEVRON STEP ONE
Under Chevron Step One, “[f]irst, always, is the question whether Congress has directly
spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the
matter; for the court, as well as the agency, must give effect to the unambiguously expressed
intent of Congress.” Chevron, 467 U.S. at 842-43; see also NRDC v. EPA, 643 F.3d 311, 322
(D.C. Cir. 2011) (“We begin with the statute.”). As noted, the relevant provision of the FDCA
subject to this Court’s statutory interpretation and Chevron analysis, and at the heart of the
dispute, is 21 U.S.C. § 355(j)(5)(F)(iv).
AstraZeneca’s position is that the Court’s analysis should end at Chevron Step One
because the plain language of 21 U.S.C. § 355(j)(5)(F)(iv) sets forth a condition that, if met,
leads to a specific result. AstraZeneca interprets the statute to mean that “if (1) there is an
30
approved sNDA containing reports of new clinical investigations sponsored or conducted by the
applicant that were essential to the approval of the sNDA, then (2) the sNDA applicant has three
years of exclusivity over a ‘change approved in the supplement.’” Pl.’s Mem. at 1. Since the
sNDAs for Seroquel approved on December 2, 2009 included reports of clinical investigations
that were essential to approval of the sNDAs, then, AstraZeneca argues, the labeling change (i.e.,
the inclusion of Table 2) approved in the supplement, is entitled to exclusivity by the plain words
of the statute. As noted, AstraZeneca bases its exclusivity rights on “two separate types of new
clinical investigations”: (1) “the new clinical investigations conducted to establish the efficacy of
Seroquel to treat schizophrenia and bipolar disorder in adolescents,” and (2) “the new clinical
investigations that provided glucose shift data for Table 2.” Pl.’s Reply at 2; see also id. at 3
(Table illustrating alleged bases for AstraZeneca’s statutory exclusivity rights). The question,
then, AstraZeneca argues, is “whether Section 505(j)(5)(F)(iv) ‘means what it says.’” Pl.’s
Mem. at 7 (quoting Landstar Express Am., Inc. v. Fed. Maritime Comm’n, 569 F.3d 493, 498
(D.C. Cir. 2009)). If it does mean what it says, AstraZeneca argues, then AstraZeneca has
exclusivity over Table 2 through December 2, 2012. See Pl.’s Mem. at 7.
The FDA, on the other hand, has interpreted the statutory provision at issue as requiring a
relationship between the subject of the “new clinical investigations,” the change to the product or
use of the product, and the scope of the three-year exclusivity, explaining as follows:
The statute sets up a relationship between the “new clinical investigations” that
are “essential to the approval of the supplement,” and the scope of the exclusivity.
That is, if an applicant submits a supplement and gets 3-year exclusivity for a
change in the use of the drug product supported by new clinical investigations, the
FDA may not approve an ANDA referencing that drug product for the “change
approved in the supplement” during that 3-year exclusivity period. Because the
change in the drug product or use of the drug product that was approved in the
supplement was based at least in part on the new clinical investigations, it
naturally follows that the scope of any exclusivity also will relate to the scope of
those new clinical investigations.
31
AR 299 (FDA Letter, dated Mar. 27, 2012, to AstraZeneca).
Although the FDA, in its briefs, framed its argument as a “reasonable interpretation” of
the statute, the Court believes that the FDA’s interpretation is closest to the plain meaning of the
statute. Indeed, the Court agrees with the FDA that the statute sets up a “logical relationship
between the change in the product for which the new clinical investigations were essential to
approval of the supplement, and the scope of any resulting three-year exclusivity.” Defs.’ Reply
at 3-4. Interpreting the plain meaning of the statute in this way means that only changes to
labeling derived from clinical studies related to the changes approved in the supplement may be
entitled to exclusivity. See id. at 3-4. Thus, AstraZeneca is not automatically entitled to
exclusivity for Table 2 by virtue of unrelated clinical studies supporting the pediatric indications
approved on December 2, 2009.
AstraZeneca also argues, however, that Table 2 deserves exclusivity on its own merit,
independent of the pediatric supplements. See Pl.’s Mem. at 17. Section 355 (j)(5)(F)(iv),
however, requires that a “supplement to an application” contain reports of “new clinical
investigations” and be “essential to the approval of the supplement.” Here, the administrative
record demonstrates, and AstraZeneca admits, that Table 2 was an “amendment” to the pediatric
supplemental new drug applications, not a supplement itself. See Pl.’s Reply at 14. Nor was
Table 2 a part of AstraZeneca’s three supplements that were “superseded” by the December 2009
approval letter. Defs.’ Reply at 7, n.1. 16 Moreover, as noted, the data from which Table 2 was
derived was submitted by AstraZeneca in 2008 pursuant to general correspondence between the
16
The December 2, 2009 Approval Letter stated that previous labeling supplements “have been superseded by this
approval action.” AR 95. The supplements referred to are dated July 19, 2007, September 11, 2008, and December
4, 2008. According to the FDA, “[n]one of those supplements have submission dates that relate to Table 2; nor does
AstraZeneca assert that those supplements are for Table 2.” Defs.’ Reply at 7 n.1.
32
parties. See AR 883, 295; Defs.’ Mem. at 9. It does not appear “plain” to the Court that data
submitted in correspondence is a “supplement” or that the data submitted for Table 2 constitutes
a “new clinical investigation” for purposes of the statute.
Although the statute’s plain meaning favors the FDA, since plausible plain meaning
arguments cut both ways, the Court believes the better reading of the statute is that section
355(j)(5)(F)(iv) is ambiguous. See ViroPharma, 2012 U.S. Dist. LEXIS 56128, at *46 (“[T]he
fact ‘[t]hat a statute is susceptible of one construction does not render its meaning plain if it is
also susceptible of another, plausible construction[.]’”) (citation omitted) (finding ambiguous 21
U.S.C. § 355(v)(3)(B), which exempts certain drugs from the exclusivity provided for under 21
U.S.C. § 355(j)(5)(F)(iv), and deferring to agency’s interpretation of the statute).
Section 355(j)(5)(F)(iv) is ambiguous because key phrases in the statutory provision are
undefined and their meaning disputed. The parties dispute, for example, whether or not Table 2
is even a “supplement” within the meaning of the section. Compare Pl.’s Reply at 4 (“The new
clinical data in Table 2 . . . is an approved supplement to the Seroquel NDA.”), with Defs.’ Mem.
at 9 (“The data from which Table 2 was derived were [not] submitted . . . as a supplement.”).
This dispute over whether Table 2 is a “supplement” is due to the ambiguities in the terms used
to define the scope of which supplements trigger the 3-year marketing exclusivity under section
355(j)(5)(F)(iv). The statutory provision at issue does not define “new clinical investigation,” or
what makes a particular investigation “essential to approval,” or what it means when a change is
approved at the same time as a supplement is approved when that change is not part of the
original supplemental application. While a lack of definitions does not automatically mean that
a statute is ambiguous, see Goldstein v. SEC, 451 F.3d 873, 878 (D.C. Cir. 2006), without these
key definitions, “nothing about ‘the specific context in which [the phrase] is used’ or ‘the
33
broader context of the statute as a whole’ is likely to compel the conclusion that the phrase has a
definite meaning.” ViroPharma, 2012 U.S. Dist. LEXIS 56128, at *43 (quoting Blackman v.
District of Columbia, 456 F.3d 167, 176 (D.C. Cir. 2006)). Each of these key phrases is
examined in more detail below.
First, the Court turns to the phrase “new clinical investigation.” It is an established
principle of statutory construction that a provision’s context should be used to assist in
determining whether a statute commands a certain interpretation or has a plain meaning. See,
e.g., Samantar v. Yosuf, 130 S. Ct. 2278, 2289 (2010) (“[W]e do not … construe statutory
phrases in isolation; we read statutes as a whole.”) (quoting United States v. Morton, 467 U.S.
822, 828 (1984)); see also Dolan v. U.S. Postal Serv., 546 U.S. 481, 486 (2006) (“Interpretation
of a word or phrase depends upon reading the whole statutory text, considering the purpose and
context of the statute[.]”). In the context of this provision, the meaning of “new clinical
investigation” is not “plain.” It is not immediately apparent that AstraZeneca’s clinical studies
underlying Table 2 are “new clinical investigations” for purposes of the statute. In 2008,
AstraZeneca submitted to the FDA the clinical studies that are the basis for Table 2 in
AstraZeneca’s unrelated effort to have new indications of Seroquel XR approved. See Defs.’
Mem. at 9; Defs.’ Reply at 15. AstraZeneca was later asked to reanalyze this data to produce
Table 2. Defs.’ Mem. at 10. Since the same clinical investigations were used to support
unrelated applications, it is not clear that these studies are “new” with respect to Table 2.
Furthermore, if these investigations are deemed “new” to support exclusivity for Table 2, are
they then ineligible to count again as “new” to support exclusivity for any of the unrelated
applications for which AstraZeneca originally submitted them? In other words, the statute does
34
not describe the circumstances when clinical investigations are considered “new.” Thus, the
Court is not persuaded that “new clinical investigations” is unambiguous.
Second, the phrase “essential to approval” is ambiguous. Congress provides no guidance
or criteria under which reports of new clinical investigations may be judged “essential” versus
merely persuasive or noteworthy. See Upjohn Co. v. Kessler, 938 F. Supp. 439, 444 (W.D.
Mich. 1996) (noting that a determination of what data is “essential to approval” is “squarely
within the ambit of the FDA’s expertise and merit[s] deference . . . ”) (citations and internal
quotation marks removed). Furthermore, AstraZeneca seems to interpret the FDA’s judgment
that Table 2 contains essential safety data as automatically making Table 2 “essential to
approval” of the supplement containing unrelated pediatric data for new pediatric indications.
See Pl.’s Mem. at 2. The words of the statute cannot bear this overbroad interpretation.
Third, AstraZeneca argues that the plain meaning of the phrase “a change approved in the
supplement” means that any change approved in the supplement is entitled to exclusivity. See
Pl.’s Mem. at 8. The clear implication of this reading is that any change approved in the
supplement, whether or not it is related to the “new clinical investigation,” or to the new uses
approved in the supplement, is entitled to exclusivity. See id. AstraZeneca argues that this
reading is compelled by Congress’s use of the phrase “a change” rather than “the change.” See
id. This reading is unpersuasive for two reasons. First, it is not unusual for supplements to
contain more than one change. See Defs.’ Mem. at 23-24. By its use of “a” rather than “the,”
Congress provided the flexibility for the FDA to approve multiple changes within one
supplement, rather than requiring a separate supplement for each change. Second, AstraZeneca’s
interpretation ignores the implication of the phrase “in the supplement.” As noted, the most
plausible meaning is that there must be a connection between the change and the supplement
35
beyond mere coincidence of appearing in the same approval letter. See id. at 24. Congress could
not have intended that the FDA’s decision to report on completely unrelated regulatory actions in
the same letter for the sake of efficiency would confer on drug manufacturers additional periods
of exclusivity when that exclusivity was not otherwise merited. Thus, even if this phrase has a
plain meaning, it favors the FDA rather than AstraZeneca. At the very least, however, this
phrase is ambiguous. 17
Finally, the subsection read as a whole is ambiguous. AstraZeneca proposes one
interpretation that goes thusly: (1) Any change to a pioneer drug application counts as a
supplement under this statute; (2) This makes Table 2 a supplement; (3) The data to generate
Table 2 is essential to the approval of Table 2; (4) Therefore, Table 2 is entitled to exclusivity.
The FDA properly rejects this circular reading of the statute as omitting the proper emphasis on
the nexus between the new clinical investigations and new uses required to trigger the additional
period of exclusivity. See Defs.’ Reply at 3-4.
While AstraZeneca is correct that “an agency’s own regulations cannot create or give rise
to ambiguity when the statutory language is unambiguous[,]” Pl.’s Reply at 6 (citing Chevron,
467 U.S. at 843 n.9 (“The judiciary is the final authority on issues of statutory construction and
must reject administrative constructions which are contrary to clear congressional intent.”)), in
this case the parties’ different interpretations of the statutory provision highlight the statute’s
ambiguity.
Accordingly, the Court proceeds to Chevron Step Two.
17
The FDA could in the future minimize misunderstanding about which approved change warrants an exclusivity
period, prompted by the ambiguity in the statute, by more carefully and precisely delineating at the time of the
approval -- in separate communications if necessary -- those changes in an sNDA that warrant an exclusivity period
and those changes that do not.
36
B. CHEVRON STEP TWO
As this Court found in denying AstraZeneca a Temporary Restraining Order, the FDA
has reasonably interpreted 21 U.S.C. § 355(j)(5)(F)(iv) in denying exclusivity for Table 2. The
FDA’s determination that the approval of Table 2 for Seroquel labeling at the same time as the
approval of the sNDA for pediatric indications “was only coincidental” is, itself, a determination
“well within the agency’s expertise . . . .” AstraZeneca Pharms. LP v. FDA, No. 12-472, 2012
U.S. Dist. LEXIS 54863, at *9 (D.D.C. Mar. 28, 2012). Therefore, the Court will defer to the
FDA’s conclusion here unless it was unreasonable. Following a review of the administrative
record, the Court concludes that the agency’s interpretation of the statute — that a substantive
relationship between new clinical studies and changes in the supplement, not the format of a
submission, dictates what changes receive exclusivity — is reasonable for several reasons. First,
the administrative record shows that the pediatric supplements were approved on their own
merits based upon clinical investigations unrelated to the Table 2 labeling change, which
standing alone does not entitle AstraZeneca to exclusivity. Second, the FDA’s interpretation of
the statute is largely consistent with past practice, and therefore not arbitrary and capricious.
Third, the FDA’s interpretation is consistent with the legislative history.
1. The Administrative Record Shows that the Pediatric Supplements
Were Approved on Their Own Merit, and Table 2 Standing Alone
Does Not Entitle AstraZeneca to Exclusivity.
AstraZeneca posits two bases for exclusivity for Table 2 arising from two separate sets of
clinical investigations: the clinical investigations supporting pediatric indications and the clinical
investigations that yielded the data for Table 2. According to AstraZeneca, “each separately and
independently establishes AstraZeneca’s exclusivity rights under the plain, unambiguous
language of the statute.” Pl.’s Reply at 3. Since the Court finds the statute to be ambiguous,
37
however, analysis of these two arguments turns on whether the FDA’s interpretation of the
statute was reasonable, as explained below. See Chevron, 467 U.S. at 843.
a. Clinical Investigations Supporting Pediatric Indications Do
Not Support Exclusivity For Table 2.
The administrative record shows that the pediatric supplements approved on December 2,
2009 were approved on their own merit, and the addition of Table 2 was not a factor in the
evaluation of the safety and efficacy of Seroquel for pediatric indications. See AR 866-68 (FDA
Internal Memorandum, dated Dec. 2, 2009, referring to pediatric indication issues only, not
Table 2).
While AstraZeneca states repeatedly that “the addition of Table 2 to the label was part of
the supplements approved on December 2, 2009,” see, e.g., Pl.’s Reply at 20-21, AstraZeneca
has not established that the approval of Table 2 was essential to the approval of the pediatric
sNDAs, which were supported by entirely different clinical trials. None of AstraZeneca’s
citations to the Administrative Record show that the FDA mandated changes to Table 2 “as a
condition of approval of the sNDAs.” Pl.’s Reply at 1; see also Pl.’s Mem. at 16. An example
of evidence that conceivably supports AstraZeneca’s position in the Administrative Record is an
internal FDA Memorandum, dated August 13, 2009. See AR 853-64. In the memorandum, an
FDA official stated that, “[w]e should be negotiating labeling changes with the sponsor prior to
approval of these NDA supplements” and “I recommend the Division should consider approval
of this set of NDA supplements provided that an agreement is reached between the sponsor
[AstraZeneca] and the Agency regarding the language in the labeling.” Id. at 863-64. Since
Table 2 was not related to the pediatric supplements, however, which were approved on their
own merit, this internal recommendation, without more, does not render Table 2 “essential to
approval” of the pediatric supplements.
38
Moreover, the fact that the FDA said it would not approve generics without Table 2 does
not mean that it would not have approved the pediatric indications for Seroquel without Table 2.
Indeed, it appears to the Court that the addition of Table 2 merely coincided with, but was not
essential to, the sNDA approvals on December 2, 2009. See AR 303 (FDA Letter, dated Mar.
27, 2012, to AstraZeneca, stating that “[t]his data does not qualify for any protection solely by
virtue of the timing of FDA’s approval of the supplement, including Table 2. Rather, the scope
of exclusivity must relate to the new clinical investigations that were conducted.”); AR 883
(Internal FDA Consultative Review, dated Mar. 27, 2012, noting that “the addition of Table 2
was not related to approval of these new [pediatric] indications. It is not unusual for [the
Division of Psychiatry Products] to bundle actions together - modifications to product labeling
usually occur in concert with other actions being taken.”).
As the FDA notes, AstraZeneca’s reliance on approval of the pediatric indications to
support a period of exclusivity for the unrelated addition of Table 2 seems to be an attempt “to
bootstrap AstraZeneca’s exclusivity for the pediatric indications into exclusivity for Table 2
because those separate changes were approved at the same time.” Defs.’ Mem. at 21. The FDA,
however, has reasonably construed the phrase “a change approved in the supplement” to mean “a
change relating to ‘new clinical investigations.’” Id. at 24; see also Univ. Med. Ctr., Inc., 2012
U.S. Dist. LEXIS 53395, at *27 (finding that the agency’s “construction of a statute . . . need not
be the best or most natural one . . . . Rather [it] need be only reasonable to warrant deference.”)
(citations and internal quotation marks omitted). “[T]he FDA was within its discretion to apply a
limiting principle so that Hatch-Waxman’s exclusivity provisions do not apply to all approved
changes that are ‘new’ . . . . [T]he general exclusivity period provided in § 355(j)(5)(F)(iv) . . . is
39
itself limited to that which is ‘new’ about the given drug.” ViroPharma, 2012 U.S. Dist. LEXIS
56128, at *50.
As noted, the FDA has also interpreted the statute to require a “nexus between the
subject of the new clinical investigations and the changes to the product that the investigations
support.” Defs.’ Mem. at 7. Here, AstraZeneca has not shown any nexus between the approval
of the supplements for pediatric indications and the labeling change consisting of the addition of
Table 2. First, it is undisputed that Table 2 contains no data from the new clinical investigations
related to the pediatric indications. See AR 301; Defs.’ Mem. at 10; Pl.’s Reply at 4, 11-12.
Second, the data from which Table 2 was derived was submitted months before the data for the
pediatric indications, and was submitted as correspondence from AstraZeneca to the FDA rather
than as a formal supplemental application. See AR 431 (AstraZeneca Letter, dated June 26,
2008, to the FDA). Third, Table 2 was approved for addition to the “Adults” section of labeling
and is thus explicitly unrelated to data from the pediatric clinical investigations. See AR 112
(Seroquel Current Labeling, dated Nov. 2011). AstraZeneca has therefore not shown that the
clinical investigations supporting its sNDAs regarding pediatric indications provide a basis for
exclusivity for labeling changes in the form of the addition of Table 2 to Seroquel labeling.
b. Clinical Investigations From Which Table 2 Is Derived Are
Not A Basis For Exclusivity.
AstraZeneca also argues that it is entitled to exclusivity for Table 2 because the FDA had
not previously relied on seven of the fifteen clinical investigations from which Table 2 was
derived. See Pl.’s Reply at 18-19. This argument is unpersuasive for two reasons. First, it
seems to depend on the assumption that Table 2 was approved as a formal “supplement.” The
FDA has made clear, however, that the approval of Table 2 was merely coincidental to the
40
approval of the pediatric indication supplements approved in December 2009, and that “Table 2
was not submitted pursuant to a supplement.” Defs.’ Mem. at 31.
Second, examining Table 2 separately from its coincidental inclusion with the pediatric
supplement, the FDA’s interpretation that Table 2 is not entitled to independent exclusivity is
reasonable. The FDA’s interpretation of the statute as only granting exclusivity for significant
innovations is reasonable given the statute’s careful balance between providing exclusivity rights
to promote innovation and making generic alternatives available to patients. See Defs.’ Mem. at
33-34; Defs.’ Reply at 15; see also Upjohn Co., 938 F. Supp. at 441 (citing Abbott Labs. v.
Young, 920 F.2d 984, 985 (D.C. Cir. 1990)) (“At the same time that it expedited approval of
generic drugs, Congress recognized the need to protect the interests of the original drug
manufacturers and to provide incentives for the invention of new products.”). The FDA’s
reading of the statutory provision is also reasonable given that the provision itself tethers the
“change” approved in the supplement to subsection (b), and thereby imparts to the meaning of a
formal supplement that it mark a sufficiently significant change that would have warranted an
additional use or indication of the drug if it had been submitted with the original NDA under
subsection (b). See 21 U.S.C. § 355(b); 21 U.S.C. § 355(j)(5)(F)(iv).
AstraZeneca’s interpretation, by contrast, “would effectively provide for whole-product
exclusivity whenever FDA determined that data from a clinical trial would have relevance to the
safety of the product for all indications, even if the data were insufficient to support approval
(and exclusivity) for a new indication.” Defs.’ Mem. at 35. Accordingly, it is reasonable for the
FDA not to grant exclusivity here because there would be no valid interest in withholding this
safety data from consumers. Furthermore, the Court is not persuaded by AstraZeneca’s
argument that the existence of a “carve out” exception, whereby generics may receive approval
41
so long as they “carve out” minor labeling changes that have received exclusivity, see Pl.’s
Reply at 16-17, renders the FDA’s interpretation of the statute unreasonable. As previously
discussed, the FDA’s interpretation need only be reasonable to be accorded deference, and it is
reasonable.
2. The FDA Did Not Act Arbitrarily or Capriciously.
The FDA’s interpretation of the statute is consistent with its past practice. 18 The FDA
has emphasized that generally applicable safety information in labeling should not be subject to
exclusivity. See Abbreviated New Drug Application Regulations; Patent and Exclusivity
Provisions, 59 Fed. Reg. 50,338, 50,357 (Oct. 3, 1994) (Supplementary Information
Accompanying Final Rule, dated Oct. 3, 1994, noting that innovators have “no valid interest in
precluding such information from the labeling of other products”). While the FDA declined in
its final implementing regulations to define the nature of supplemental applications, which, “if
supported by clinical investigations, would warrant 3-year exclusivity,” the FDA noted that
“[c]hanges that would not warrant exclusivity are . . . changes in labeling that involve warnings
or other similar risk information that must be included in the labeling of generic competitors.”
Id. (emphasis added).
18
The Court does not find convincing AstraZeneca’s argument that the FDA should be “judicially estopped” due to
the agency’s “litigation tactics” from arguing that it has a longstanding practice of rejecting exclusivity rights in
situations such as this one. Pl.’s Mem. at 11; Pl.’s Reply at 9-10. As noted, supra at note 13, the FDA should have
been more forthcoming about its determination regarding AstraZeneca’s lack of exclusivity rights to Table 2 in
response to the company’s Citizen Petitions and when this matter was before Judge Kollar-Kotelly. By waiting until
March 27, 2012 to explain its reasoning for not recognizing exclusivity for Seroquel labeling, the FDA needlessly
delayed consideration of the merits of the agency’s determination, which could have been considered when
AstraZeneca filed its first complaint for injunctive relief and caused the plaintiff and this Court to participate in a
fire drill to resolve these significant legal issues in the context of a Temporary Restraining Order on March 28, 2012.
While the Court faults the FDA for contributing to a needless waste of resources for the parties and the Court on this
matter by not making its views plain earlier, AstraZeneca has shown no reason why the Court should prevent the
FDA from relying on its past precedent in responding to a challenge to its final decision to approve generic drugs
referencing Seroquel. Furthermore, AstraZeneca has raised new arguments in its Motion for Summary Judgment,
and the FDA may respond in turn by including citations to its relevant past decisions.
42
Notably, the FDA has also denied exclusivity for similar labeling changes to other drugs
in Seroquel’s class of antipsychotics. See AR 303 n.21 (FDA Letter, dated Mar. 27, 2012, to
AstraZeneca, stating that “[t]he agency’s conclusion that Table 2 is not protected and its
implications for generic quetiapine products is consistent with the agency’s treatment of other
second-generation antipsychotics for which data regarding metabolic changes, including Table-2
like data, have been made.”). According to the FDA, in fact, “[s]even drugs [in the same
antipsychotic class as Seroquel] (Invega, Invega Sustenna, Abilify, Risperdal, Risperdal Consta,
Fanapt, and Latuda) have included class-wide labeling changes for metabolic data, and none of
them has received exclusivity for those changes.” Defs.’ Mem. at 36 n.22.
AstraZeneca cites the FDA’s handling of four drugs (Meridia, Travatan, Rapamune, and
Colcrys) as examples to support the proposition that the “FDA has an established practice of
granting exclusivity to labeling information, even when such information is relevant to all
indications of use for a drug, and even when such information appears in the ‘warnings’ section
of a drug’s label.” Pl.’s Mem. at 20. These examples are readily distinguishable. For each of
these four drugs, the FDA granted exclusivity to the labeling because the FDA found the labeling
changes “substantively related to the new clinical investigations that were essential to the
approval of a clinical efficacy supplement.” Defs.’ Reply at 16; see also Defs.’ Mem. at 39
(labeling changes were “directly related to a clinical efficacy supplement for the protected
indication.”).
In the case of Meridia, according to the FDA, the labeling change for which exclusivity
was granted was “substantively related to the new clinical investigations that were essential to
the approval of a clinical efficacy supplement.” Defs.’ Reply at 16; see also Defs.’ Mem. at 37-
38. Similarly, in the case of Travatan, the FDA granted exclusivity to a labeling change on the
43
basis of clinical data that “expanded the patient population” by replacing a statement that
Travatan had not been studied in patients with renal or hepatic impairment with clinical findings
showing that “no clinically relevant changes were observed” when those patients took the drug.
Defs.’ Mem. at 38.
In the cases of Rapamune and Colcrys, information that was originally used to support
new efficacy supplements was also considered necessary safety labeling information. With
respect to Rapamune, the FDA determined that Rapamune was entitled to exclusivity for “a
clinical efficacy supplement for cyclosporine withdrawal procedures in patients at low to
moderate risk for rejection.” Defs.’ Mem. at 39 (citation omitted). The FDA concluded,
however, that the labeling information exclusivity should be extended to the population at high
risk of rejection because that “labeling information might help raise physicians’ awareness of the
risks of cyclosporine.” Declaration of Benjamin C. Block (“Block Decl.”), ECF No. 21, Ex. 2
(quoted in Defs.’ Mem. at 39). Likewise, for Colcrys, 3-year exclusivity was originally granted
to dosing information essential to treating acute gout flares. AR 957 (FDA Letter, dated May 25,
2011, to Sidley Austin LLP regarding the approval of Colcrys). When the FDA concluded that
the same dosing information was necessary for the related indication of prophylaxis of gout
flares, exclusivity was granted to the prophylaxis indication, even though it was previously not
protected. See id. at 977; Defs.’ Mem. at 39.
The critical difference in circumstances between the protected labeling information for
the drugs Rapamune and Colcrys and unprotected Table 2 for Seroquel is that, in the cases of the
former drugs, the labeling changes were tied to the approval of efficacy supplements, whereas
the approval of Table 2 was never essential to the approval of any efficacy supplement for
Seroquel. See, e.g., AR 11-12 (FDA Letter, dated Dec. 18, 2008, to AstraZeneca, making clear
44
that metabolic data submissions, which led to creation of Table 2, are separate from the pediatric
efficacy supplements under consideration). Table 2 did not contribute to the approval of the
pediatric or any other efficacy supplement, and has not yet led to any other changes in efficacy,
such as a new dosage or prescribing regimen or a new indication or use in a new patient
population. In other words, if the safety data reflected in Table 2 had been known at the time of
the FDA’s approval of the Seroquel NDA for its original indications, it would have been
included in the original labeling. Thus, although the FDA considers Table 2 “essential” safety
information, the FDA’s decision not to grant Table 2 exclusivity is consistent with its past
decisions and distinguishable from the instances where the FDA has granted exclusivity for
changed safety information in labeling.
Furthermore, the FDA provided a reasoned explanation for its decision to approve
ANDAs for generic versions of Seroquel. On March 27, 2012, the FDA issued a seventeen-page
letter to AstraZeneca explaining the agency’s rationale for finding that Table 2 in Seroquel was
not protected by a three-year exclusivity period. See AR 293-310; see also Defs.’ Mem. at 12-
13. Since this letter explains, inter alia, that generally-applicable safety information of the type
in Table 2 is not subject to protection, the agency’s decision to deny Table 2 exclusivity is not
arbitrary and capricious. See Motor Vehicles Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto
Ins. Co., 463 U.S. 29, 43 (1983) (quoting Burlington Truck Lines v. United States, 371 U.S. 156,
168 (1962), for the proposition that “the agency must examine the relevant data and articulate a
satisfactory explanation for its action including a ‘rational connection between the facts found
and the choice made.’”).
Therefore, because the FDA provided a reasoned explanation for its decision, the FDA’s
decision to deny AstraZeneca exclusivity for Table 2 was not arbitrary and capricious.
45
The potential implications of AstraZeneca’s broad reading of the statute also lend support
to the FDA’s interpretation. As noted, the FDA has undertaken an effort to implement updated
safety labeling for drugs in the same class as Seroquel with metabolic data similar to what is
displayed in Table 2. Defs.’ Reply at 11. AstraZeneca’s interpretation of this statutory provision
would seriously impede the FDA’s initiative to improve the safety data available for this class of
drugs, which includes Invega, Invega Sustenna, Abilify, Risperdal, Risperdal Consta, Fanapt,
and Latuda. 19 Id. Though all of the safety data added to the labels of these drugs is derived from
clinical investigations, the FDA claims that none of these investigations provided the basis for
three-year exclusivity. Id.; see also AR 303-04 n.21 (FDA Letter, dated Mar. 27, 2012, to
AstraZeneca). A reading of the statutory provision at issue that could prevent the FDA from
requesting from drug manufacturers any class-wide safety labeling changes involving clinical
trials without triggering exclusivity periods is untenable and, as detailed below, plainly
inconsistent with the legislative history.
3. The FDA’s Interpretation of the Statute is Consistent with Legislative
History.
The Court finds the FDA’s interpretation of the statute consistent with the statute’s
legislative history. As noted, a central purpose of the Hatch-Waxman Amendments, as reflected
in its title “The Drug Price Competition and Patent Term Restoration Act,” was to promote
competition by providing a mechanism through which generic drugs were more easily approved.
See 21 U.S.C. § 355(j); see also ViroPharma, 2012 U.S. Dist. LEXIS 56128, at *7 (“The
abbreviated new drug application (‘ANDA’) process shortens the time and effort needed for
19
Since certain safety information, such as Table 2, “is necessary for safe use of the product,” it would not be
subject to the “carve out” exception for ANDAs, see Defs.’ Mem. at 3, whereby ANDA applicants may receive
FDA approval so long as they “carve out” protected portions of labeling. See, e.g., Bristol-Myers Squibb v. Shalala,
91 F.3d 1493, 1500 (D.C. Cir. 1996). By contrast, for example, the FDA determined that certain protected pediatric
use information for Seroquel “may be safely carved out of generic quetiapine fumerate [sic] tablets labeling and
replaced with appropriate legal disclaimers.” AR 880 (FDA Memorandum to File, dated May 23, 2011).
46
approval of a generic drug by allowing the applicant to merely demonstrate its product’s
bioequivalence to the NDA drug, without reproducing the entirety of the NDA’s extensive
scientific research.”) (citations omitted).
Congress had dual purposes, however, and enacted the abbreviated approval process for
generic drugs while simultaneously retaining incentives, such as marketing exclusivity, to
encourage innovation and development from pioneer drugmakers. See 21 U.S.C. § 355(j); see
also ViroPharma, 2012 U.S. Dist. LEXIS 56128, at *7 (“Because Congress still wanted to
provide incentives for new drug development, alongside the ANDA process that eased the
marketing of generic drugs, Hatch-Waxman entitles an NDA applicant to a period of market
exclusivity . . . which bars FDA approval of a generic ANDA for the NDA product . . . . Thus,
pursuant to Hatch-Waxman’s provisions, pioneer drug companies are entitled to certain periods
of marketing exclusivity during which they are protected from generic competition.”) (quotation
marks and citations omitted). As the FDA notes, the Act struck a “careful balance between
exclusivity and generic entry” into the market. Defs.’ Reply at 11; see also Abbott Labs., 920
F.2d at 985 (“Congress struck a balance between expediting generic drug applications and
protecting the interests of the original drug manufacturers.”) (citing H.R.REP. NO. 98-857, pt. 1,
at 15 (1984), reprinted in 1984 U.S.C.C.A.N. (98 Stat. 1585) 2647, 2648).
While AstraZeneca is thus correct to point out that Congress provided exclusivity rights
“as an incentive for pioneer companies to engage in expensive clinical research,” Pl.’s Reply at
1, granting exclusivity to Table 2 would disrupt the “careful balance” Congress crafted.
Adopting AstraZeneca’s interpretation, where every change approved in a supplemental
application receives exclusivity, would increase the potential for companies to receive whole-
product exclusivity and bar generic competition altogether for labeling changes unrelated to
47
innovations in drug use. See Defs.’ Mem. at 36. As the FDA notes, AstraZeneca’s interpretation
would result in an “unwarranted evergreening of exclusivity,” Defs.’ Reply at 10, allowing
AstraZeneca to retain a monopoly over production of a drug by periodically updating safety
information in their labeling. This risk is particularly worrisome given the FDA’s practice that
“modifications to product labeling usually occur in concert with other [regulatory] actions being
taken.” AR 883 (FDA Consultative Review, dated Mar. 27, 2012, to OGD). AstraZeneca’s
interpretation would create a perverse incentive for pharmaceutical companies to drag out their
presentation of vital safety data to the FDA in order to bar generic competition beyond the
periods determined acceptable by Congress. While Congress was no doubt concerned that
pharmaceutical manufacturers have incentives to continue research and development in order to
discover vital new drugs, Congress plainly did not intend for these manufacturers to retain
exclusivity into perpetuity.
The legislative history makes clear that the sponsors of the Hatch-Waxman Amendments,
Senator Orrin Hatch and Representative Henry Waxman, envisioned that three-year exclusivity
would be granted for significant changes, such as approvals for new therapeutic uses and new
patient populations. See 130 CONG. REC. 24,425 (1984) (Statement of Rep. Henry Waxman)
(three-year exclusivity is intended to “encourage drugmakers to obtain FDA approval for
significant therapeutic uses of previously approved drugs”); see also 130 CONG. REC. 23,766
(1984) (Statement of Senator Orrin Hatch) (three-year exclusivity is intended to protect “some
changes in strength, indications, and so forth . . . .”). These sponsors’ statements support the
FDA’s decision to consistently grant exclusivity for changes resulting from studies supporting
efficacy supplements, such as in the cases of Meridia, Travatan, Rapamune, and Colcrys, but not
48
to grant exclusivity for Table 2, which involves general “warnings or other similar risk
information” applicable to all indications for the drug. See AR 1529 (59 Fed. Reg. 50,357).
The legislative history thus supports the FDA’s reasonable conclusion that Table 2 is not
entitled to exclusivity under 21 U.S.C. § 355(j)(5)(F)(iv).
IV. CONCLUSION
For the foregoing reasons, AstraZeneca’s Motion for Summary Judgment, ECF No. 21, is
DENIED, and the FDA’s Motion for Summary Judgment, ECF No. 26, is GRANTED. An
Order consistent with this Memorandum Opinion was issued on June 28, 2012.
DATED: July 5, 2012
/s/ Beryl A. Howell
BERYL A. HOWELL
United States District Judge
49