Eisai Inc. v. United States Food and Drug Administration ( 2015 )

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
)
EISAI, INC., )
)
)
Plaintiff, )
)
v. ) Civil Action No. 14-cv-1346 (RDM)
)
UNITED STATES FOOD AND DRUG )
ADMINISTRATION, et al., )
)
Defendants. )
)
MEMORANDUM OPINION
A company that obtains approval from the Food and Drug Administration (“FDA”) to
market a drug, no active ingredient of which has been previously approved, is entitled to a five-
year period of exclusivity during which would-be competitors cannot apply for approval of
generic versions of that drug. See 21 U.S.C. § 355(c)(3)(E)(ii). This exclusivity period creates
an incentive for pharmaceutical companies to undertake the lengthy and expensive process of
developing, testing, and obtaining approval of new drugs. On occasion, however, some drugs
effectively receive periods of exclusivity shorter than the five year period authorized by statute.
When the Drug Enforcement Administration (“DEA”) seeks to schedule a new drug under the
Controlled Substances Act, it must request recommendations from the FDA; after the DEA
receives the FDA’s recommendation, it engages in notice-and-comment rulemaking culminating
in a final rule that determines whether and at what level the drug will be scheduled. See 21
U.S.C. § 811. Because the FDA requires applicants for approval of new drugs to commit not to
market those drugs until after the DEA makes its scheduling determination, the scheduling
process can delay the entry of new drugs into the market, sometimes by more than a year after
their FDA approval. The central issue in this case is whether and under what circumstances the
period of time drug manufacturers spend waiting for a final DEA scheduling determination
counts against the five-year exclusivity period.
Plaintiff, Eisai, Inc. (“Eisai”), holds new drug approvals (“NDAs”) for two drugs caught
in this regulatory limbo. It contends—not without force—that its effective loss of months of
market exclusivity while it waited for the DEA to schedule these drugs is at odds with the
balance that Congress struck between incentivizing the development of new drugs and making
affordable medications more broadly available to patients when it enacted the Hatch-Waxman
Amendments in 1984. On the one hand, Congress streamlined the procedure for approval of
typically cheaper generic drugs. On the other, it granted five years of market exclusivity to the
developers of sufficiently innovative drugs, improving the chance that they would see returns on
the significant investments required to bring new drugs to market. As Eisai stresses, requiring
manufacturers of scheduled drugs to lose months—or more—of exclusivity while they await
scheduling determinations might upset this balance and discourage pharmaceutical companies
from pursuing promising drugs that are likely to require scheduling under the Controlled
Substances Act.
This case, however, turns on the meaning, not the wisdom, of an FDA regulation
implementing the Hatch-Waxman Amendments. Under that regulation—the validity of which
Eisai does not challenge—the exclusivity period for a new drug begins when the FDA issues its
letter approving the drug, even if the drug’s manufacturer must await DEA’s scheduling
determination before it can bring the drug to market. The regulation does provide for an
exception under limited circumstances. But the FDA has interpreted that exception narrowly,
2
and the Court is bound to defer to the agency’s reasonable interpretation of its own regulation.
Because Eisai’s drugs do not qualify for the exception under the FDA’s interpretation of its
regulation, the FDA and its co-Defendants are entitled to summary judgment.
I. BACKGROUND
A. The Statutory and Regulatory Regime
The Drug Price Competition and Patent Term Restoration Act, Pub. L. 98-417 (1984),
commonly known as the Hatch-Waxman Amendments, “emerged from Congress’[s] efforts to
balance two conflicting policy objectives: to induce name-brand pharmaceutical firms to make
the investments necessary to research and develop new drug products, while simultaneously
enabling competitors to bring cheaper, generic copies of those drugs to market.” Abbott Labs. v.
Young, 920 F.2d 984, 991 (D.C. Cir. 1990). To achieve the first goal—encouraging investment
in new drugs—the Hatch-Waxman Amendments provided a five-year period of market
exclusivity for new drugs, no active ingredients of which have previously been approved. 21
U.S.C. § 355(c)(3)(E)(ii). Other companies are barred from seeking approval of generic versions
of these new drugs—referred to as “New Chemical Entities” or “NCEs”—during this five-year
window, which begins on “the date of the approval of the [New Chemical Entity] application.”
Id. To achieve the second goal—making it easier for competitors to bring cheap generics to the
market—Congress created the Abbreviated New Drug Approval process, which allows
producers of follow-on drugs to rely on the safety and effectiveness trials conducted by a drug’s
initial developer, streamlining the process of bringing generic drugs to market. See 21 U.S.C.
355(j).
The FDA has promulgated regulations implementing the Hatch-Waxman Amendments.
Tracking the statute, these regulations bar manufacturers from applying for approval of follow-
3
on drugs for “a period of 5 years from the date of approval of the first approved new drug
application.” 21 C.F.R. § 314.108(b)(2). “Date of approval,” in turn, is defined as:
the date on the letter from FDA stating that the new drug application is approved,
whether or not final printed labeling or other materials must yet be submitted as
long as approval of such labeling or materials is not expressly required. “Date of
approval” refers only to a final approval and not to a tentative approval that may
become effective at a later date.
21 C.F.R. § 314.108(a).
The Controlled Substances Act creates five “schedules” for potentially addictive drugs or
drugs that otherwise have “potential for abuse.” 21 U.S.C. § 812. It authorizes the Attorney
General—who, in turn, has delegated this authority to the DEA—to add, remove, or reassign
drugs through notice-and-comment rulemaking. Id. § 811; 28 C.F.R. § 0.100(b). When the FDA
determines that a drug in the approval process “has an abuse potential,” it must forward that
information to the DEA. 21 U.S.C. § 811(g). Before the DEA initiates a rulemaking to schedule
a new drug, it must “request from” the FDA “a scientific and medical evaluation” of the drug, as
well as “recommendations[ ] as to whether such drug . . . should be” scheduled as a controlled
substance. Id. § 811(b). The FDA’s recommendations—which it must provide “within a
reasonable time”—are “binding . . . as to . . . scientific and medical matters.” Id. If the DEA
finds “substantial evidence of potential for abuse,” it “shall initiate proceedings” to schedule the
drug. Id. If the DEA determines that the drug should be scheduled under the Controlled
Substances Act, the manufacturer must update the label with “the controlled substance symbol
designating the schedule in which the controlled substance is listed.” 21 C.F.R. § 201.57(a)(2);
see also id. § 1302.04 (providing further labeling requirements).
The FDA’s approval process for new drugs includes review of the drug’s proposed
labeling. See 21 U.S.C. § 355(b)(1), (d). If the DEA reaches its scheduling determination after
the drug is approved, however, the label must be updated to indicate the drug’s scheduling
4
designation. The parties agree that the label updating process under these circumstances is
governed by 21 C.F.R. § 314.70(b), which requires a manufacturer to supplement its new drug
application with any changes to the controlled substance labeling of its drug and to secure FDA
approval of that supplement before marketing the product. Although the regulations thus require
that a drug manufacturer obtain FDA approval of labeling changes that reflect DEA’s scheduling
of the drug before using the modified labeling, the producer may seek a waiver of that
requirement under 21 C.F.R. § 314.90, and the FDA routinely grants these waiver requests. AR
13 n.71. As a practical matter, such a waiver permits the producer to begin marketing as soon as
the DEA issues its final scheduling determination.
Neither the Food, Drug, and Cosmetic Act nor the FDA’s regulations prohibit a producer
from marketing an approved drug before that drug is scheduled. The FDA, however, requires
new drug applicants to agree not to market drugs that the “FDA has proposed for scheduling
under the Controlled Substances Act . . . until the Drug Enforcement Administration makes a
final scheduling decision.” U.S. Food and Drug Administration, FDA Form 356h, available at
http://www.fda.gov/downloads/AboutFDA/ReportsManualsForms/Forms/UCM082348.pdf (last
visited Sept. 28, 2015). The upshot is that manufacturers of drugs that have been proposed for
scheduling cannot market their drugs until, at the earliest, the DEA issues a final scheduling
determination, even if the FDA approved the drugs at an earlier time. And if the FDA calculates
the five-year exclusivity period beginning on the date of its approval, the producer is unable to
market the drug during a portion of its exclusivity period, stripping that asset of some of its
value.
B. Belviq and Fycompa
Plaintiff produces two recently approved drugs that lost portions of their window of
market exclusivity while awaiting DEA scheduling. The first, Belviq, is a weight-loss drug that,
5
according to the complaint, “took fourteen years and cost over $300 million” to develop. Dkt. 1
¶ 36. The FDA issued a letter approving Belviq on June 27, 2012. Id. ¶ 37. In the letter, which
bore the caption “NDA APPROVAL,” the agency reminded the drug manufacturer that it had
“agreed not to market [Belviq] until the Drug Enforcement Administration has made a final
scheduling decision.” AR 68. Moreover, the agency advised:
[W]hen the scheduling is finalized, you will need to make appropriate revisions to
the package insert, the patient package insert and the carton and immediate-
container labels through supplementation of your NDA. This would include the
statements detailing the scheduling of Belviq in the labeling, as required under 21
C.F.R. §§ 201.57(a)(2) and (c)(10)(i).
Id. The FDA subsequently listed Belviq in Approved Drug Products with Therapeutic
Equivalence Evaluations—the agency publication commonly known as the “Orange Book”—and
determined that Belviq’s five-year exclusivity period began on June 27, 2012, when the approval
letter was issued.
The FDA submitted its scheduling recommendation to the DEA for Belviq on June 25,
2012, just two days before it approved the drug. The DEA issued its final rule scheduling Belviq
almost a year later, on May 8, 2013. See Schedules of Controlled Substances: Placement of
Lorcaserin Into Schedule IV, 78 Fed. Reg. 26701-02 (May 8, 2013).
The other drug at issue, Fycompa, followed a similar path. Fycompa is used to treat
seizures in patients suffering from epilepsy. Dkt. 1 ¶ 11. The FDA issued a letter approving
Fycompa on October 22, 2012. Id. ¶ 47. In language materially identical to that used in the
Belviq approval, that letter stated:
[Y]ou agreed not to market this drug until the Drug Enforcement Administration
has made a final scheduling decision. We further note that, when the scheduling
is finalized, you will need to make appropriate revisions to the package insert, the
patient package insert and the carton and immediate-container labels through
supplementation of your NDA. This would include statements detailing the
scheduling of Fycompa in the labeling, as required under 21 § C.F.R. 201.57(a)(2)
and (c)(10)(i).
6
AR 76-77. The FDA subsequently included Fycompa in the Orange Book and determined that
its five-year exclusivity period began on the date of the approval letter.
The FDA recommended that the DEA schedule Fycompa on January 28, 2013—about
three months after Fycompa was approved and the exclusivity period began. The DEA issued a
final rule scheduling Fycompa on January 2, 2014—over fourteen months after the FDA issued
its approval letter. After the DEA scheduled Belviq and Fycompa, Eisai had to submit for FDA
approval revised labeling incorporating the scheduling. 21 C.F.R. § 314.70(b). As is typical,
Eisai received waivers pursuant to 21 C.F.R. § 314.90 of the requirement that these changes be
approved before the drugs could be marketed with the revised labeling. Arg. Tr. at 6:24-7:5.
Thus, Eisai was authorized to begin marketing Belviq in May 2013 and Fycompa in January
2014.
C. Eisai’s Petition and Lawsuit
On July 25, 2013, Eisai filed a petition with the FDA challenging the agency’s
calculation of the exclusivity periods for Belviq and Fycompa. See AR 20-143. In the petition,
Eisai argued that the FDA’s determination that the exclusivity periods for Belviq and Fycompa
began before the company could market the drugs was arbitrary and capricious and violated the
FDA’s regulations. AR 31-46. In Eisai’s view, exclusivity for new drugs subject to scheduling
under the Controlled Substances Act should be “triggered only when FDA-approved labeling
incorporating the final schedule permits commercial marketing of the products.” AR 33.
The FDA denied Eisai’s petition on April 30, 2014. AR 1-19. The agency read the
petition to “ask FDA to decide that there are two approval dates for their drugs: (1) when FDA
has completed its review of the [New Drug Application] and issues an approval letter, and (2)
when DEA has completed its scheduling process, with only the latter being considered for
purposes of 5-year NCE exclusivity.” AR 14. It rejected Eisai’s argument that Belviq and
7
Fycompa qualified for an exception to the general rule that a drug is approved on the date of the
FDA’s approval letter. Id. at 17-18.
Eisai filed this lawsuit on August 8, 2014. The company seeks a declaration that the
FDA’s determination of the exclusivity periods for Belviq and Fycompa violates the
Administrative Procedure Act, 5 U.S.C. § 706(2), (“APA”), and an order requiring the FDA to
recognize those periods as beginning on the date when the company was actually permitted to
bring the drug to market. Dkt. 1 at 25. Eisai subsequently moved for summary judgment,
arguing that the FDA violated 21 C.F.R. § 314.108(a), departed from its earlier practice, unfairly
treated Eisai’s drugs differently than similar products, and failed to provide a reasonable basis
for its actions. Dkt. 14. The FDA filed a cross-motion for summary judgment, arguing that
Belviq and Fycompa did not qualify for an exception to the FDA’s general rule that five-year
exclusivity begins to run on the date of the agency’s approval letter, that the agency’s practice
has been consistent, and that the agency treats all drugs identically by beginning their exclusivity
periods on the date of the FDA’s approval letter. Dkt. 15. The Court held oral argument on the
cross-motions on September 1, 2015.
II. LEGAL STANDARDS
The APA precludes agency action that is “arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). An agency’s decision,
accordingly, must be the product of “reasoned decisionmaking.” Motor Vehicle Mfrs. Ass’n of
U.S. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 52 (1983). An agency action will normally
be set aside as “arbitrary and capricious” if the agency “has relied on factors which Congress has
not intended it to consider, entirely failed to consider an important aspect of the problem, offered
an explanation for its decision that runs counter to the evidence before the agency, or is so
8
implausible that it could not be ascribed to a difference in view or the product of agency
expertise.” Id. at 43.
An agency must also, of course, “adhere to its own regulations.” Brock v. Cathedral
Bluffs Shale Oil Co., 796 F.2d 533, 536 (D.C. Cir. 1986). An agency has wide latitude to
interpret its regulation, however, and its interpretation is “controlling unless plainly erroneous or
inconsistent with the regulation.” Auer v. Robbins, 519 U.S. 452, 461 (1997) (citations and
quotation marks omitted); see also Bowles v. Seminole Rock & Sand Co., 325 U.S. 410, 414
(1945). Thus, the Court must defer to the FDA’s interpretation of its regulation “unless an
alternative reading is compelled by the regulation’s plain language or by other indications of the
[agency’s] intent at the time of the regulation’s promulgation.” Thomas Jefferson Univ. v.
Shalala, 512 U.S. 504, 512 (1994) (citations and quotation marks omitted). And, although an
agency enjoys significant leeway when interpreting its regulation, that interpretation must bear
the mark of consistency, for “agency action is arbitrary and capricious if it departs from agency
precedent without explanation.” Ramaprakash v. FAA, 346 F.3d 1121, 1124 (D.C. Cir. 2003).
III. DISCUSSION
Eisai argues that the FDA violated the APA in multiple respects. First, the company
alleges that the agency acted arbitrarily and capriciously by triggering the start of the five-year
exclusivity period based on the date in the NDA approval letters. Eisai argues that the agency
should have instead recognized that the NDA approval letters it issued fell within the regulatory
exception. Even if they did not, the company contends, the FDA’s interpretation of its own
regulation is inconsistent with the text and both regulatory and congressional intent. Eisai also
alleges that failing to recognize that it fell within the exception resulted in a departure from past
9
agency practice. Finally, Eisai asserts that the FDA violated the APA by treating Belviq and
Fycompa differently than other, similarly situated products
A. Application of the FDA’s Interpretation to Belviq and Fycompa
The Court will first consider Eisai’s argument that, even accepting the FDA’s
interpretation of the “approval date” regulation, the company should prevail because the letters
approving Belviq and Fycompa contained language that should have brought the drugs within
the exception to the regulation and delayed the FDA’s official approval date. To summarize the
argument: The regulation provides that the “date of approval” is the date on the NDA approval
letter “as long as approval of [final printed] labeling or materials is not expressly required,” 21
C.F.R. § 314.108(a); the FDA has interpreted that exception as applying only when the
“express[] require[ment]” that the FDA approve the final labeling appears in the actual approval
letter that the FDA sends; and Eisai contends that even accepting that interpretation, the letters at
issue here contained such a requirement and so Belviq and Fycompa should have fallen within
the exception and received a later approval date.
If Eisai is correct about that, then the Court could resolve this case on the facts and avoid
issuing a broader ruling on the FDA’s interpretation of its regulation. The Court concludes,
however, that the FDA reasonably concluded that the letters do not themselves expressly require
subsequent label approval.
There is no question that Eisai had to finalize the labels on the packaging for Belviq and
Fycompa after DEA scheduled the drugs. FDA regulations require that the label of any
scheduled drug reflect the Controlled Substances Act schedule on which the drug is listed. 21
C.F.R. § 201.57(a)(2); see also id. § 201.57(c)(10)(i) (requiring the a drug’s full prescribing
information to include “the schedule in which [a drug] is controlled”). The proposed labels for
10
Belviq and Fycompa did not include any such designation because the FDA had not yet referred
the drugs for DEA scheduling when Eisai submitted the NDAs, including the proposed labels,
for the drugs. To cover this situation, FDA regulations set out the requirements for making
various changes to drug applications and provide that “[a]ny change to the information required
by § 201.57(a)” is a “major change” that “requir[es] supplement submission and approval prior
to distribution of the product made using the change.” 21 C.F.R. § 314.70(b), (b)(2)(v)(C). At
oral argument, the FDA conceded that § 314.70(b) applied to the labeling changes that added
scheduling information for Belviq and Fycompa. Arg. Tr. at 65:16-23. Thus, at the time Eisai
received its NDA approval letters, FDA regulations mandated that Eisai update its labels once
DEA made a scheduling determination and that it secure FDA approval of those updated labels.
The question here is whether the NDA approval letters themselves “expressly required”
that approval. See 21 C.F.R. § 314.108(a). If so, then the drugs should have fallen within the
exception in § 314.108(a), even as the FDA construes the regulation, and should have received
an official approval date that reflected that later approval. This is a close question on the facts.
The NDA approval letter for Belviq included a reminder to Eisai that the company would “need
to make appropriate revisions to the package insert, the patient package insert and the carton and
immediate-container labels through supplementation of [its] NDA.” AR 68. The letter for
Fycompa contained identical language. AR 76-77. As the Court reads the letters, both expressly
require Eisai to revise the labels, but neither expressly requires that the FDA approve those edits.
It is true that the letters require the sort of label revision that the FDA must presumptively
approve under its regulations. 21 C.F.R. § 314.70(b)(2)(v)(C). But only an express requirement
of FDA “approval” triggers the exception in 21 C.F.R. § 314.108(a), at least as the FDA
11
construes that exception, and the letters say nothing about approval. Although this distinction is
a fine one, it is not without significance.
Even though the letters facially appear to trigger a sequence of events that should end
with FDA approval of the updated labels before the drugs could go to market, the record reveals
that the agency’s practice is quite different. FDA regulations allow the agency to waive the
requirement that it approve of labeling changes before products incorporating those changes can
be brought to market. See 21 C.F.R. § 314.90(a) (“An applicant may ask the Food and Drug
Administration to waive under this section any requirement that applies to the applicant under
§§ 314.50 through 314.81.”). In its administrative petition to the FDA, Eisai recognized that the
FDA “routinely approves” those waivers and allows companies to amend their labels to reflect
DEA scheduling through a “changes-being-effected (CBE) supplement after a determination by
the [FDA’s] review division that a prior-approval supplement is not necessary.” AR 34
(emphasis added). In other words, the FDA’s practice has been that, once it approves an NDA, a
company may obtain a waiver and simply file a supplement reflecting the addition of the DEA
scheduling information without obtaining FDA approval before going to market. See Norwich
Eaton Pharm., Inc. v. Bowen, 808 F.2d 486, 492 (D.C. Cir. 1987) (noting that “‘[s]upplements’
effect changes in applications that have already been approved” (emphasis added)).
This practice makes sense. A label change to reflect DEA scheduling is often as simple
as adding a small logo to the label with a Roman numeral that reflects the drug’s schedule and a
sentence that states the same information. It is such a simple and non-discretionary edit that the
FDA generally treats it as ministerial and not worth further agency review. As a result of the
FDA’s “routine[]” practice of accepting these supplements, Eisai recognized that “the day the
CBE supplement is submitted with the necessary label changes is the day the sponsor can
12
commercially market the product . . . .” AR 34. And, indeed, Eisai received such waivers for
both Belviq and Fycompa. 1 Thus, the NDA approval letters Eisai received were, in reality, the
final FDA approval necessary for market entry, and the DEA’s scheduling determination was the
only hurdle that remained. Because those letters did not facially contain an “express[]
require[ment]” that the FDA approve of additional labeling requirements and because Eisai
understood that, in fact, the FDA would not need to approve of any additional labeling
requirements before the company could take the drugs to market, the Court concludes that the
FDA’s decision to consider the date of approval for both Belviq and Fycompa as the date found
on their NDA approval letters was not arbitrary or capricious. See Emily’s List v. FEC, 581 F.3d
1, 22 n.20 (D.C. Cir. 2009) (“Agencies generally do not violate the APA’s deferential arbitrary-
and-capricious standard when they employ bright-line rules for reasons of administrative
convenience, so long as those rules fall within a zone of reasonableness and are reasonably
explained.”).
B. The FDA’s Interpretation of the 21 C.F.R. § 314.108(a) Exception
The next question is whether the FDA has reasonably construed the exception to its
regulation defining “date of approval,” 21 C.F.R. § 314.108(a). This is a critical issue in this
case because the agency initiates the five-year exclusivity period for a drug on “the date of
approval of the . . . new drug application.” 21 C.F.R. § 314.108(b)(2). Section 314.108 tracks
the statutory language, which prohibits submission of applications for generic versions of a drug
“before the expiration of five years from the date of approval” of the original new drug
1
Only one of these waivers is in the record, see AR 107, but Eisai avers, and the FDA has not
disputed, that both waivers were granted after the approval letters that triggered the start of the
exclusivity periods here.
13
application. 21 U.S.C. § 355(c)(3)(E)(ii) (emphasis added). The statute does not define “date of
approval,” but an FDA regulation does:
Date of approval means the date on the letter from FDA stating that the new drug
application is approved, whether or not final printed labeling or other materials
must yet be submitted as long as approval of such labeling or materials is not
expressly required. “Date of approval” refers only to a final approval and not to a
tentative approval that may become effective at a later date.
21 C.F.R. § 314.108(a). Thus, in run-of-the-mill cases, the date of approval is simply the “date
on the letter from the FDA stating that the new drug application is approved.” Id. There is little
room for the agency to interpret this clear rule, and its interpretation is not at issue here.
There is room, however, to interpret the exception to the general rule, which applies
when “approval of [final printed] labeling or other materials” is “expressly required.” Id. The
regulation is entirely silent as to who must “expressly require[ ]” labeling approval, where that
requirement must appear, or even what constitutes an “express require[ment].” The FDA argues
that the phrase “expressly required” refers only to the letter the agency issues advising that the
new drug application is approved, and absent an express requirement of subsequent approval in
that letter, the approval letter triggers the five-year exclusivity period. The Court must defer to
that interpretation unless it is “plainly erroneous or inconsistent with the regulation.” Auer, 519
U.S. at 461. 2
1. Text of the Regulation
Eisai argues that “there is no requirement in the regulation’s text that the ‘express
requirement’ be in the approval letter, and the FDA cannot read such a requirement into the
2
Auer has been the target of skepticism in recent years. See Decker v. Nw. Envtl. Def. Ctr., 133
S. Ct. 1326, 1338-39 (2013) (Roberts, C.J., concurring in part and dissenting in part); id. at 1339-
1342 (Scalia, J., dissenting); Perez v. Mortg. Bankers Ass’n, 135 S. Ct. 1199, 1210-1211 (2015)
(Alito, J., concurring in part and concurring in the judgment); id. at 1213-1225 (Thomas, J.,
concurring in the judgment). This Court, however, is bound to follow Auer unless and until the
Supreme Court modifies the relevant standard.
14
regulation.” Dkt. 17 at 4. To prevail on this argument, Eisai must show that it would be “plainly
erroneous or inconsistent with the regulation” for the FDA to interpret its regulation in this
manner. Auer, 519 U.S. at 461; see also Thomas Jefferson Univ., 512 U.S. at 512 (“[The
Court’s] task is not to decide which among several competing interpretations best serves the
regulatory purpose. Rather, the agency’s interpretation must be given controlling weight unless
it is plainly erroneous or inconsistent with the regulation.” (internal quotation marks omitted)).
The fact that 21 C.F.R. § 314.108(a) does not specify which document must “expressly
require[]” subsequent approval renders the regulation ambiguous. It does not, however, render
the FDA’s reading plainly erroneous.
The FDA’s interpretation of the regulatory text, moreover, is a reasonable one. Eisai
reads too much into the fact that the phrase “expressly required” is not followed by a reference to
the source of that “require[ment].” The full sentence that contains the exception reads: “Date of
approval means the date on the letter from FDA stating that the new drug application is
approved, whether or not final printed labeling or other materials must yet be submitted as long
as approval of such labeling or materials is not expressly required.” 21 C.F.R. § 314.108(a)
(emphasis added). It is not plainly erroneous or otherwise inconsistent with the regulation’s text
to read the phrase “expressly required” to refer back to the phrase “letter from FDA stating that
the new drug application is approved,” which appears in the very same sentence.
Although the regulation’s text could have been clearer, the FDA’s interpretation is a
plausible one, and Eisai’s “alternative” construction is not “compelled by the regulation’s plain
language.” Thomas Jefferson Univ., 512 U.S. at 512.
15
2. The Agency’s Intent Behind the Regulation
Eisai also argues that the FDA’s interpretation of its regulation is inconsistent with the
agency’s intent at the time it proposed the regulation. The company points to the preamble to the
Federal Register notice that proposed 21 C.F.R. § 314.108, which stated, in relevant part:
The “date of approval” of the application as used in these provisions means the
date on the approval letter sent by FDA to the applicant. A requirement in the
approval letter for submission (but not for approval) of final printed labeling or
other material that might delay the actual initiation of marketing of the product is
not relevant to a determination of the date of approval, so long as the product
could be legally marketed.
54 Fed. Reg. 28872, 28898 (July 10, 1989). Eisai asks the Court to make an inference from this
explanation: If a requirement for submission of final printed labeling is “not relevant” to the date
of approval “so long as the product could be legally marketed,” then it stands to reason, the
company suggests, that the date on the approval letter should not be the official “date of
approval” if the product cannot be legally marketed at that time. This argument is not without
merit. But there are several reasons why the preamble is ultimately insufficient to compel the
Court to cast aside the high level of deference that Auer otherwise requires.
First, the preamble sheds little light on the issue that is central to Eisai’s case, which is
where or how the FDA must “expressly require” approval of labeling materials to trigger the
exception in 21 C.F.R. § 314.108(b)(2). As explained above, the FDA’s interpretation of the
regulation as reaching only “express require[ments]” for future approval of labeling materials in
the NDA approval letter is itself consistent with the plain text of the regulation, see Part III.B.1,
and neither approval letter at issue here contained such an express requirement, see Part III.A.
The preamble speaks principally to when labeling changes required in NDA approval letters
should be deemed more than ministerial (i.e., when they prevent the product from being legally
16
marketed), but it is of less help in settling whether such requirements found outside of approval
letters should alter the official date of approval.
Second, both the preamble and the regulation might raise potential concerns about how
the FDA approval process and DEA’s scheduling process interact, but they are not the concerns
at issue in this litigation. A company may only avail itself of the exception in 21 C.F.R.
§ 314.108 when the FDA expressly requires that the company obtain approval for labeling
changes before taking the drug to market. The preamble is fully consistent with the regulation’s
text, suggesting that the date on the NDA approval letter should be the final date of approval
unless the FDA delays when the applicant could legally market the drug by requiring some
additional FDA approval of labeling. But the approval of the drugs’ labeling was not the source
of the delay for either Belviq or Fycompa. Instead, as explained above, the FDA is in the habit
of exercising its discretion to waive under 21 C.F.R. § 314.90 any requirement that it approve
ministerial changes to labeling that reflect DEA scheduling. See AR 13 & n.71. Eisai
recognized during its administrative appeal that the FDA “routinely” grants these waivers, and
Eisai received waivers for both Belviq and Fycompa. See id.; see also Arg. Tr. at 6:24-7:5.
Thus, at the time Eisai received its NDA approval letter from the FDA for both drugs, the
company understood that there was only one actual impediment to taking the drugs to market.
That barrier was DEA finalizing the drugs’ scheduling, not FDA approval of any subsequent
labeling changes. Both the regulation and the preamble, however, address only the latter and say
nothing about the former when defining “date of approval.”
Once again, Eisai raises a fair point, but it is one that is insufficient to overcome the
FDA’s interpretation of its own regulation. The preamble to the regulation may support Eisai’s
broader policy argument that a drug should not be deemed “approved” for exclusivity purposes
17
until it can legally go to market. But, the preamble is not sufficiently clear or definite—or, for
that matter, binding on the agency—to demonstrate that the FDA’s interpretation of its otherwise
ambiguous regulation is “plainly erroneous or inconsistent with the regulation.” Auer, 519 U.S.
at 461.
3. The Statutory Purpose
Eisai also contends that there is tension between the FDA’s interpretation of § 314.108
and Congress’s intent to provide an effective incentive to producers of new drugs through the
provision of a five-year period of market exclusivity. The company asserts that beginning the
five-year exclusivity period before a drug could be legally marketed would impermissibly
“change the incentive structure adopted by Congress” and reflect the agency’s “arrogat[ion] for
itself the power to deprive a NCE sponsor of its earned exclusivity.” Dkt. 14-1 at 12 n.5
(quotation marks and alterations omitted). This argument is more atmospheric than formal in
Eisai’s briefing, and Eisai has explicitly disclaimed any attack on § 314.108 itself as inconsistent
with the governing statute, 21 U.S.C. § 355. See Arg. Tr. at 31:17-22 (noting that Chevron
deference does not apply because “nobody is challenging the regulation”). The Court takes Eisai
to be suggesting that the statutory purpose is so clearly contrary to the FDA’s interpretation of its
regulation that the interpretation does not even satisfy the highly deferential standard in Auer.
Eisai cites two cases that are relevant to this question: Ranbaxy Labs. Ltd. v. Leavitt, 469
F.3d 120, 121-22 (D.C. Cir. 2006) and Teva Pharm. USA, Inc. v. Sebelius, 595 F.3d 1303, 1316
(D.C. Cir. 2010). Both concerned an agency regulation that affected the operation of 21 U.S.C.
§ 355(j)(2)(A)(vii), which provides that in certain circumstances, the first generic drug
manufacturer to challenge successfully a patent held by the developer of a new drug is entitled to
a 180-day period of limited exclusivity during which the FDA may not approve any other
generic drug manufacturer’s version of the drug. The FDA’s policy under attack in both cases,
18
however, allowed drug developers to exploit a loophole that prevented generic manufacturers
from obtaining 180-day exclusivity: After an application for approval of a generic version was
filed, the drug’s patent holder could request that the FDA “delist” the challenged patent from the
developer’s new drug application, and the agency would grant the request as long as the
developer had not already filed a lawsuit to protect its patent. Ranbaxy Labs., 469 F.3d at 122-
23. Because the manufacturer no longer claimed patent protection, the first generic manufacturer
whose application was approved would not receive 180-day exclusivity. In both cases, the D.C.
Circuit struck down the FDA’s policy because it was “inconsistent with the structure of the
statute” by “diminish[ing] the incentive for a manufacturer of generic drugs to challenge a [listed
patent.” Ranbaxy Labs., 469 F.3d at 125-26; see also Teva, 595 F.3d 1318.
Eisai would analogize this case to Ranbaxy Labs. and Teva. In its view, the FDA’s
policy of beginning the five-year exclusivity period for drugs before the drugs can be legally
marketed interferes with the incentive Congress intended to create for developers of new drugs.
This is a substantial argument. There are, however, important ways in which this case is unlike
Ranbaxy Labs. and Teva. For one thing, the FDA policies in those cases allowed drug
manufacturers to manipulate the exclusivity system in direct conflict with Congress’s purpose,
see Teva, 595 F.3d at 1317 (“[G]iven the incentives for the brand manufacturer, [strategic
delisting] will be used only where its impact on Congress’s scheme is most destructive.”
(emphasis omitted)). There does not appear to be a similar concern here. Although bureaucratic
delay within the FDA or the DEA or other innocent factors might hamper a drug manufacturer’s
use of its full five-year exclusivity period, Eisai has not suggested that the FDA’s policy opens
the door to intentional and strategic manipulation by private third-party competitors.
19
More significantly, the Court must consider the actual impediment that Eisai faced in
marketing Belviq and Fycompa after the approval letters were issued. One possible impediment
was FDA Form 356h, which required Eisai to commit that it would refrain from marketing the
drugs before the DEA made its scheduling determination. See U.S. Food and Drug
Administration, FDA Form 356h, available at http://www.fda.gov/downloads/AboutFDA/
ReportsManualsForms/Forms/UCM082348.pdf (last visited Sept. 28, 2015). Eisai, however, has
not separately challenged the lawfulness of that requirement—nor does it argue that Form 356h
has anything to do with the FDA’s “date of approval” regulation, which creates an exception
only for “such labeling or other materials” where FDA approval is “expressly required,” 21
C.F.R. § 314.108(a). FDA Form 356h might be an obstacle to Eisai (and other similarly situated
companies) receiving a full five-year window of exclusivity, but that obstacle has little to do with
whether the agency’s interpretation of its “date of approval” regulation is “plainly erroneous or
inconsistent with the regulation.”
To the extent Eisai challenges the underlying impediment resulting from the DEA’s delay
in scheduling the drugs, moreover, that policy concern goes well beyond the type of concern at
issue in Ranbaxy Labs. and Teva. Unlike the challenges to the FDA’s actions in Ranbaxy Labs.
and Teva, consideration of the DEA’s delay is difficult to reconcile with the statutory text, which
simply provides that the exclusivity period runs for “five years from the date of the approval of
the [new drug] application.” 21 U.S.C. § 355(c)(3)(E)(ii). If Congress had intended for the FDA
to take into account reasons that marketing of a drug might be delayed that are beyond the
FDA’s control, it is unlikely that it would have focused on the date of the FDA’s approval of the
NDA, rather than the date on which the drug was permitted to be marketed or the date on which
all regulatory requirements were satisfied. Indeed, before the Hatch-Waxman Amendments were
20
adopted, the FDA had “specified” by regulation “that the applicant shall be notified in writing
that the application is approved and the application shall be approved on the date of the
notification,” and the D.C. Circuit has “found absolutely no reason to believe that Congress
intended the term ‘approved’ in the Hatch-Waxman Amendments to mean anything other than
what the FDA understood it to mean.” Mead Johnson Pharm. Grp. v. Bowen, 838 F.2d 1332,
1336 (D.C. Cir. 1988). Put simply, the FDA has always looked to the NDA approval letter when
determining a new drug’s approval date, and the D.C. Circuit has endorsed that approach as
consistent with Congress’s intent. That the FDA also tethers the exception in § 314.108(a) to the
approval letter appears perfectly consistent with Congress’s intent.
The only other impediment that Eisai might have faced was the requirement that it
include the DEA scheduling designation in its final label. Eisai might rely, for example, on the
fact that approval of a drug’s labeling is a part of the FDA’s approval process. See 21 U.S.C. §
355(b)(1) (requiring new drug applicants to submit “specimens of the labeling proposed to be
used for such drug”); 21 C.F.R. § 314.105(c) (“FDA will approve an application after it
determines that the drug meets the statutory standards for . . . labeling”). If approval requires a
determination that the drug’s labeling is adequate, then it is reasonable to argue that the FDA has
not granted “approval” until it makes that determination. And for controlled substances like
Belviq and Fycompa, the FDA does not approve the final labeling until after the DEA makes its
scheduling decision. See 21 C.F.R. §§ 201.57(a)(2), 314.70(b)(2)(v)(C).
The FDA, however, has drawn a distinction between minor labeling changes (such as the
addition of a Controlled Substances Act schedule designation) and more significant ones in this
regard. Its regulations explicitly contemplate that the agency may “approve an application and
issue the applicant an approval letter on the basis of draft labeling” as long as the “only
21
deficiencies” are “editorial or other similar deficiencies in the draft labeling.” 21 C.F.R.
314.105(b). That FDA’s position is both understandable and reasonable. Where the agency
requires that an applicant make only minor, mechanical changes to its labeling, the agency has
concluded its approval of the application. The fact that the FDA may be required to perform the
ministerial function of reviewing minor changes to draft labeling after it formally approves the
drug—and, with a waiver, see 21 C.F.R. § 314.90, after the drug has already been marketed—
does not render the formal approval in any way illusory. Moreover, as explained above, the
FDA has reasonably explained that it routinely grants waivers—as it did for Belviq and
Fycompa—of any pre-approval requirement for label changes required after the DEA makes a
scheduling designation. See AR 13 & n.71; AR 107.
Ultimately, then, to the extent any FDA action impeded Eisai from taking advantage of
the full five years of market exclusivity, it was the use of Form 356h—which was not a labeling
rule, which did not require any post-approval FDA action, and which is not separately challenged
in this action. The existence of that impediment has no relevance to the Court’s interpretation of
the FDA’s entirely separate “date of approval” regulation, and, in particular, its exception for
“labeling or other materials” where further FDA approval is “expressly required.”
Moreover, although Eisai correctly characterizes the policy considerations that underlie
the Hatch-Waxman Amendments, its position is difficult to reconcile with the statutory text.
Most significantly, as the FDA noted, adopting Eisai’s interpretation of § 314.108 would have
the anomalous effect of creating different “approval” dates for different purposes under the
Hatch-Waxman Amendments. For example, as Eisai acknowledges, the FDA does and will
continue to determine which of two competing drugs was first approved—and therefore entitled
to five-year exclusivity in the first place—by reference to the date of the approval letter. Arg. Tr.
22
at 28:18-23; see also 21 U.S.C. § 355(c)(3)(E)(i) (providing five-year exclusivity for any drug
“no active ingredient . . . of which has been approved in any other application”) (emphasis
added). It is unlikely that Congress intended to refer to different triggering events when it used
the phrase “has been approved” early in the first sentence of § 355(c)(3)(E)(i) and the phrase
“date of approval” later in that same sentence. But, in any event, the FDA hardly exceeded the
scope of its interpretive authority when it concluded that the phrase should be given a consistent
meaning.
Despite the ambiguity noted above with respect to the “expressly required” language,
§ 314.108 speaks with relative clarity about the criteria for determining the “date of approval”
for a new drug. The date on which the drug may be legally marketed is not one of those criteria.
Instead, the regulation focuses on when the FDA has taken the last action it must take that
constitutes “approval.” See 21 C.F.R. § 314.108(a) (“Date of approval means the date on the
letter from the FDA stating that the new drug application is approved . . . .” (emphasis added));
see also id. (providing no exception where “final printed labeling or other materials must yet be
submitted as long as approval of such labeling or materials is not expressly required” (emphasis
added)). This case illustrates that the date on which the FDA “approv[es]” a drug is not always
the date on which the drug may be marketed. Although this disparity may reflect a flaw either in
the FDA’s regulation or in the statute, it does not reflect any deficiency in the agency’s
interpretation of its regulation. 3 Because § 314.108(a) does not, by its terms, admit of an
interpretation that triggers five-year exclusivity at the time a drug may be legally marketed,
3
The Court notes that legislation has been introduced in Congress that would explicitly define
the “date of approval” of a drug subject to scheduling as a controlled substance as “the date of
issuance of the interim final rule controlling the drug.” H.R. 639, Improving Regulatory
Transparency for New Medical Therapies Act. The bill was passed by the House of
Representatives on March 16, 2015, and is awaiting action in the Senate.
23
rather than when the FDA takes an action that constitutes “approval” of the drug, the Court
cannot plausibly interpret the regulation to avoid the fundamental statutory concerns Eisai has
raised. And because Eisai has limited its challenge to an attack on the FDA’s interpretation of its
regulation, the Court need not resolve the question whether § 314.108, as reasonably interpreted
by the FDA, is inconsistent with 21 U.S.C. § 355.
* * *
Although the extent to which the current regulatory scheme comports with the incentive
structure that the Hatch-Waxman Amendments put in place raises difficult issues, the question
presented here is a narrow one: Is the FDA’s interpretation of the exception to its “date of
approval” regulation “plainly erroneous or inconsistent with the regulation.” The answer to that
question is relatively straightforward. The FDA has reasonably construed the exception to mean
that the requirement for further FDA approval must appear in the letter itself.
This leads to one final interpretive question: What rational basis might plausibly exist to
confer different effective periods of exclusivity on new drugs based on whether a requirement for
further FDA approval of labeling or other materials is expressly mentioned in the approval letter
itself or simply mandated by statute or regulation. As explained above, the answer is that the
approval letters signify completion of the FDA’s process for reviewing and approving a
manufacturer’s NDA. The letters do, after all, bear the caption “NDA APPROVAL.” Some
requirements that may remain after the FDA sends such a letter are ministerial and do not affect
the finality of that approval. In those circumstances—including when the manufacturer will have
to finalize a label to reflect DEA scheduling—the FDA does not need to consider further any
aspect of the NDA and so it treats its “approval” as final. In other cases, however, the
requirements that remain are substantial and will require further substantive review and approval.
24
In those cases, the agency expressly requires further action in its approval letter as a signal that
the FDA has concluded that its process is not yet completed. Although not the only method that
the FDA might use to differentiate applications that require no further substantive review from
those that still demand some significant approval, the approach the FDA has adopted is a
reasonable one. The operative statute makes it necessary for the FDA uniformly to define the
“date of approval,” and the agency drew that line in a reasonable place. The Court, accordingly,
concludes that the FDA’s interpretation of 21 C.F.R. § 314.108(a) is not “plainly erroneous or
inconsistent with the regulation,” Auer, 519 U.S. at 461, and it thus defers to that interpretation.
C. Razadyne and the FDA’s Prior Practice
Eisai also argues that the FDA’s interpretation of its “date of approval” regulation
constitutes an “‘unexplained reversal’” that constitutes “‘the height of arbitrary and capricious
decision making.’” Dkt. 14-1 at 21 (quoting Purepac Pharm. Co. v. Thompson, 354 F.3d 877,
884 (D.C. Cir. 2004)). As Eisai acknowledges, “agency action is arbitrary and capricious if it
departs from agency precedent without explanation.” Ramaprakash v. FAA, 346 F.3d 1121,
1124 (D.C. Cir. 2003). Here, Eisai’s argument turns on the FDA’s previous treatment of
Razadyne ER (“Razadyne”), a drug used to treat Alzheimer’s disease.
The record establishes that Razadyne was approved in an FDA letter dated December 22,
2004. Dkt. 1-1. Unlike Belviq and Fycompa, however, Razadyne was approved “without a
tradename.” Id. at 3. Instead, “because of recent medication errors associated with” the drug
under its former appellation and a similarly-named drug, the producer had agreed to “adopt a
new name for the [drug] prior to its marketing,” and, critically, “to submit the new proposed
proprietary names to the Agency for [its] review prior to implementation.” Id. Following the
general rule articulated in § 314.108, the FDA initially determined that Razadyne’s exclusivity
25
period began on December 22, 2004—the date of the letter approving the drug. See Dkt. 1-6.
The producer, however, requested that the the FDA “revise the approval date” to April 1, 2005.
Dkt. 1-4. Its request was “based upon the fact that it was not until the later date that FDA and
[the producer] agreed upon the new trade name . . . for the product.” Id. The FDA granted the
request, stating it was “reasonable to conclude that Razadyne . . . was not approved until April 1,
2005, when the Agency completed its review of the proposed trade name, found it acceptable,
and conveyed this information to” the drug’s producer. Id. Although the FDA’s letter approving
the request does not cite § 314.108, the parties appear to agree that the agency’s decision to
change the approval date for Razadyne reflected a determination that the drug qualified for the
exception to the approval-date rule in that section. See Dkt. 15 at 16; Dkt. 17 at 9.
Eisai contends that the FDA’s decision to change the approval date for Razadyne is
inconsistent with the narrow construction of § 314.108 the agency now offers. It reasons that
“[t]here was no explicit statement [in the FDA’s initial approval letter] to the effect that” the
sponsor of Razadyne “must obtain FDA approval of a trade name for its product before
marketing could begin,” so Razadyne’s approval date could not have been changed under the
standard the agency now asserts. Dkt. 17 at 9 (internal quotation marks omitted).
The Razadyne letter stated that Razadyne had made a “commitment to adopt a new name
for [Razadyne] prior to its marketing, and to submit the new proposed proprietary names to the
[FDA] for [its] review prior to implementation.” Dkt. 1-1 at 3. Eisai makes something of the
fact that this was a “voluntary” commitment—rather than being “expressly required” to do
anything, the manufacturer simply received the FDA’s recitation of voluntary commitments it
had made during the approval process. Dkt. 14-1 at 20 n.7. But the question of whether a
26
voluntary commitment of this type gives rise to a “require[ment]” for purposes of the “date of
approval” regulation has nothing to do with the issue presented in this case.
Instead, the relevant distinction the FDA draws for present purposes is between its
statement that Eisai must “make appropriate revisions” to certain labeling components—which
does not refer at all to FDA review or approval—and its statement to the manufacturer of
Razadyne that that company had committed to “submit the new proposed proprietary names to
the [FDA] for [its] review prior to implementation.” Dkt. 1-1 at 3 (emphasis added). Eisai
argues that this language does not expressly require “approval.” Dkt. 17 at 5. True, the
Razadyne letter does not explicitly say that the FDA must approve Razadyne’s new trade name
before the drug is marketed; it says only that the producer must “submit” the name for “review
prior to implementation.” But by expressly requiring “review prior to implementation,” the
Razadyne letter comes closer to expressly requiring “approval” of the subsequent submission
than did the Belviq or Fycompa letters, which did not expressly require any further agency
review. And it seems reasonable for the FDA to have concluded that requiring the manufacturer
of Razadyne to submit the new name for agency “review prior to implementation” carried with it
an understanding that FDA approval was necessary, particularly since the FDA had rejected the
prior name as confusing and as posing potential health concerns.
The Razadyne letter thus reveals where the FDA draws the line for the § 314.108
exception. Its interpretation is strict enough that merely mentioning regulatory obligations that,
in turn, give rise to a requirement to obtain approval of a label change is insufficient to trigger
the exception. But it is accommodating enough that language expressly referring to an
obligation to submit labeling-related materials for “review” qualifies for the exception. The
agency did not need to draw the line in this place—it could have adopted an interpretation of the
27
exception under which both Eisai and the sponsor of Razadyne qualified, and it likely could have
adopted an interpretation under which neither qualified. The conclusion it actually reached,
however, reasonably relies on a real—even if modest—textual difference in the letters the FDA
issued, and it warrants deference. See Rollins Envtl. Servs. (NJ) Inc. v. EPA, 937 F.2d 649, 652
(D.C. Cir. 1991) (“[I]n a competition between possible meanings of a regulation, the agency’s
choice receives substantial deference.”); cf. Universal City Studios LLLP v. Peters, 402 F.3d
1238, 1243 (D.C. Cir. 2005) (“An agency’s strict construction of a general rule in the face of
waiver requests is insufficient evidence of an abuse of discretion.” (internal quotation marks
omitted)).
More relevant to the FDA’s consistency is the fact that it has always treated drugs
awaiting DEA scheduling in the same way it treated Belviq and Fycompa here. The record
reveals 11 drugs approved since 2005 that were subject to DEA scheduling determinations. See
AR 157. In none of those cases did the FDA delay the beginning of the exclusivity period until
after the drug had been scheduled. Id. And the FDA has offered a reasonable explanation for its
decision to apply the § 314.108(a) exception to Razadyne, which is not a scheduled drug. The
evidence thus suggests that the FDA is actually consistent in how it treats drugs referred for
scheduling and that Razadyne was a unique case that received an exception in light of the
agency’s reasonable understanding that its Razadyne approval letter called for further FDA
action. Such a reasoned explanation backed by relevant evidence is not arbitrary and capricious.
See Emily’s List, 581 F.3d at 22 n.20.
D. Eisai’s Other Arguments
In addition to its contention that the FDA misinterpreted its own regulation, Eisai argues
that the agency arbitrarily treated its drugs differently from (1) drugs that do not require
scheduling under the Controlled Substances Act and (2) other drugs that do require such
28
scheduling. The APA, Eisai notes, mandates that “an agency must treat similar cases in a similar
manner unless it can provide a legitimate reason for failing to do so.” Indep. Petroleum Ass’n of
Am. v. Babbitt, 92 F.3d 1248, 1258 (D.C. Cir. 1996).
The first of these arguments has it backwards. The FDA, as noted above, treated Belviq
and Fycompa precisely the same way it treats drugs not subject to controlled substance
scheduling—it began their five-year exclusivity periods on the date of the FDA’s letters
approving the drugs. Eisai asks the agency to treat its drugs differently from drugs that need not
await scheduling before they can be marketed. But the agency’s refusal to do so is manifestly
not a failure to “treat similar cases in a similar manner.” Babbitt, 92 F.3d at 1258. And although
Eisai argues that the FDA has “entirely failed to consider an important aspect of the problem” in
this case by refusing to tie the beginning of the exclusivity period to the date on which a drug
may be marketed, State Farm, 463 U.S. at 43, that argument is directed at the validity of
§ 314.108 itself. It is the language of the regulation—not the FDA’s interpretation of it—that
omits any reference to the date on which a drug may be legally marketed.
The second argument fares no better. It is true that the FDA’s interpretation of § 314.108
gives the agency significant discretion regarding the start date for five-year exclusivity of
scheduled drugs. By simply including a detailed statement of the particular labeling
requirements facing a drug manufacturer, including the requirement to obtain subsequent FDA
approval of updated labels reflecting scheduling information, the FDA can create a letter that
qualifies for the exception and confers a substantial benefit on the recipient company. This
discretion could be abused, but there is no evidence the FDA has done so here. Instead, as noted
above, the FDA has not delayed the approval date until after DEA scheduling for any of the 11
drugs the agency has considered since 2005 that required DEA scheduling. See AR 157.
29
To the extent Eisai suggests, moreover, that the FDA treats similarly situated drugs
differently because different companies end up with different exclusivity windows based on the
time it takes DEA to schedule the drugs (and the time it takes FDA to refer the drugs for
scheduling), Dkt. 14-1 at 18-19, Eisai might in theory have a claim against either agency on that
ground. But Eisai is not suing to obtain a prompt scheduling determination—it is suing to
change the effect that delays in scheduling determinations have on drug manufacturers’
exclusivity periods. As explained above, however, all scheduled drugs are treated alike in this
respect: None of them obtain deferred approval dates pending completion of the DEA the
scheduling process.
Finally, Eisai notes in two sentences at the end of its brief that the FDA’s response to the
company’s petition is “devoid of any consideration of FDA’s disparate treatment of NCEs that
require [Controlled Substances Act] scheduling.” Dkt. 14-1 at 25. This argument sounds in the
oft-repeated rule that agencies, not courts, should decide issues in the first instance, for courts
“cannot exercise their duty of review unless they are advised of the considerations underlying
[agency] action.” SEC v. Chenery Corp., 318 U.S. 80, 94 (1943). Eisai has a point here. The
FDA’s response to this argument, which Eisai raised below, see AR 38-39, is minimal at best.
Nevertheless, the Court concludes that the agency’s explanation is just barely enough to satisfy
the requirement of Chenery.
The FDA’s decision survives because of the final footnote of its response to Eisai’s
petition. There, the agency explained that it “underst[ood] the equitable arguments” that Eisai
raised and was “actively considering whether it should change its approach going forward,
perhaps to an approach of issuing complete response letters to drugs subject to scheduling rather
than approval letters in appropriate circumstances.” AR 18 n.96. The sentence in the text
30
accompanying that footnote explained that “under the existing statutory framework, there is only
a single date of approval, and an exclusivity period begins on that date.” AR 18. And two
paragraphs earlier, the FDA explained that “by arguing that the approval letter that [it] received
is not really an approval of [its] NDAs,” Eisai was in reality “asserting that [it] . . . should have
received a form of ‘complete response letter’” that would reject the application “‘in its present
form’” and “explain[] what additional information must be submitted before approval can be
granted.” Id. (quoting 21 C.F.R. § 314.110). But, the agency noted, Eisai “did not seek such a
response.” Id.
This explanation responds, albeit obliquely, to Eisai’s complaints about unequal
treatment. The agency recognized that there were “equitable” concerns with the process that cost
Eisai some of its exclusivity period for both Belviq and Fycompa. It also recognized that there
might be an existing solution to this equitable concern under a separate regulatory provision—
the “complete response letter.” But, the FDA explained, Eisai had not asked for a complete
response letter. The company instead went through the standard NDA process and received an
approval. And like all other companies that receive an NDA approval letter, Eisai’s five-year
exclusivity window was triggered on the date that letter was issued. Because the company chose
to pursue that course, and because the “existing statutory framework” allowed for “only a single
date of approval,” the agency was unwilling to entertain Eisai’s argument. That explanation,
although far from the clear response agencies should strive to provide, sufficiently addressed the
company’s argument that it was receiving unequal treatment.
IV. CONCLUSION
For the foregoing reasons, Eisai’s motion for summary judgment is DENIED and the
FDA’s motion for summary judgment is GRANTED. An appropriate Order accompanies this
Memorandum Opinion.
31
/s/ Randolph D. Moss
RANDOLPH D. MOSS
United States District Judge
Date: September 30, 2015
32