Amarin Pharmaceuticals Ireland Limited v. Food and Drug AdministrationAmarin Pharmaceuticals Ireland Limited v. Food and Drug Administration
Ann Frances Entwistle, Gerald Cooper Kell, Andrew E. Clark, Cindy Jane Cho, U.S. Department of Justice, Washington, DC, for Defendants.
RANDOLPH D. MOSS, United States District Judge
Amarin Pharmaceuticals Ireland Limited (Amarin) challenges the Food and Drug Administration‘s determination that Amarin‘s new drug, Vascepa (icosapent ethyl) Capsules (Vascepa), is not entitled to a five-year period of market exclusivity under the Federal Food, Drug, and Cosmetic Act (FDCA). That period of exclusivity is available for a new drug, if no active ingredient (including any ester or salt of the active ingredient) of the drug has been approved in any other application for new drug approval. Here, the Food and Drug Administration (FDA or Agency) denied five-year market exclusivity for Vascepa because Vascepa‘s active ingredient—a single molecule—is one component of a mixture that makes up the active ingredient of a previously approved drug. For the reasons set forth below, the Court concludes that the FDA‘s decision must be set aside, and the matter is remanded to the Agency for further proceedings consistent with this decision.
I. BACKGROUND
A. Statutory and Regulatory Background
1. The Hatch-Waxman Amendments
In 1984, Congress enacted the Drug Price Competition and Patent Term Restoration Act,
The two sides of the compromise are codified in separate aspects of the Hatch-Waxman Amendments. On the one hand, to streamline the process for bringing new generic drugs to market, Congress created the abbreviated new drug application (ANDA) process, under which a manufacturer can obtain FDA approval for a generic drug by demonstrating that it has the same active ingredient or active ingredients as a drug previously approved as safe and effective and that the generic drug is otherwise equivalent to that drug. See
On the other hand, in order to maintain incentives for pioneer drug manufacturers to research and invest in new drugs, Congress provided that most drugs with new active ingredients would be entitled to a five-year period of marketing exclusivity. Specifically, Congress provided a five-year period of exclusivity for approved new drugs, no active ingredient (including any ester or salt of the active ingredient) of which has been approved in any other application.
Congress also provided a more limited three-year period of exclusivity for new drugs that contain an active ingredient (including any ester or salt of the active ingredient) that has been approved in another application in certain circumstances where the drug‘s sponsor was required to conduct new research to gain approval. See
2. FDA Regulations And Abbott Labs
Section 314.108 of the FDA‘s regulations implements the five-year exclusivity provision. See
The regulatory history makes clear that the Agency adopted the active moiety approach to address an issue that is not implicated in this case: the availability of exclusivity for multiple closely related forms of the same basic molecule. By defining active moiety to mean the molecule or ion, excluding those appended portions of the molecule that cause the drug to be an ester, salt, ... or other noncovalent derivative (such as a complex, chelate, or clathrate) of the molecule,
When the FDA first proposed regulations implementing this active moiety approach in 1989, it explained that the approach was justified because, in its view, Congress did not intend to confer significant periods of exclusivity on minor variations of previously approved chemical compounds. See Abbreviated New Drug Application Regulations, 54 Fed. Reg. 28,872, 28,898 (proposed July 10, 1989). Almost a year and a half after the FDA issued those proposed regulations, the Court of Appeals addressed the FDA‘s interpretation of the statutory exclusivity provision in Abbott Laboratories v. Young, 920 F.2d 984 (D.C. Cir. 1990) (Abbott Labs). In that case, the FDA argued that it could treat all forms of a molecule that eventually produce the same active moiety alike for purposes of exclusivity.4 See id. at 988. The Agency sought to tether this authority to the statutory text by construing the term including broadly, and arguing that the parenthetical phrase (including any ester or salt of the active ingredient) was merely illustrative. Id. With the statute so construed, the FDA argued that it was permitted to treat other forms of an active ingredient in the same manner as esters and salts. See id.; Abbreviated New Drug Application Regulations; Patent And Exclusivity Provisions, 59 Fed. Reg. 50,338, 50,358 (final rule) (explaining that in Abbott Labs, the FDA interpreted the term active ingredient narrowly to refer to the form of the moiety in the product, but interpreted the parenthetical phrase ‘(including any salt or ester of the active ingredient)’ broadly to include all active ingredients that are different but contain the same active moiety.). The Court of Appeals rejected that interpretation of the exclusivity provision as linguistically infeasible. Abbott Labs, 920 F.2d at 988.
The FDA did not retreat from its active moiety approach. Instead, in 1994 it promulgated the final regulations discussed
B. Factual And Procedural Background
1. Lovaza
In 2004, the FDA approved a new drug application for Lovaza (omega-3-acid ethyl esters) Capsules (Lovaza), an adjunct to diet intended to reduce triglyceride levels in adults with severe hypertriglyceridemia. AR 2-3, AR 144. The sole active ingredient of Lovaza is a mixture that is primarily composed of seven kinds of omega-3 fatty acid ethyl esters. Two of the seven esters—the esters of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)—make up approximately 85% of the mixture.6 The rest of the mixture consists of the other five ethyl esters and other uncharacterized components.
Although portions of Lovaza‘s label refer to specific components of the mixture, there is no dispute that its sole active ingredient is the mixture as a whole.7 Indeed, as recently as 2014, the FDA rejected a request by Lovaza‘s sponsor to recharacterize Lovaza‘s active ingredients as the separate components of the mixture. In denying that request, the FDA explained that because the Lovaza mixture has not been fully characterized, the FDA has identified the entire fish oil mixture as the active ingredient of Lovaza. FDA, Citizen Petition Response, Docket No. FDA-2013-P-0148 (Feb. 21, 2014) (Lovaza Citizen Petition Response).8 As the Agency explained, when naturally derived mixtures are not sufficiently characterized to precisely identify every molecule that meaningfully con-
This case deals with a collateral consequence of the FDA‘s active moiety approach that arises when a drug‘s active ingredient is made up of multiple molecules and thus potentially multiple active moieties. As explained above, when the FDA determines whether a new drug is eligible for five-year exclusivity, it focuses on the active moiety of the new drug. When comparing single-molecule drugs, the inquiry is straightforward: The FDA takes a new drug‘s single-molecule active ingredient, removes certain appendages (those that make it a salt, ester, or other noncovalent derivative) to identify its active moiety, and then evaluates whether any previously approved single-molecule drug shares that same active moiety. As discussed below, this analysis becomes more complicated when the FDA applies its active moiety approach to mixtures that contain a single active ingredient but multiple active moieties.
2. Vascepa
Amarin is a biopharmaceutical company that focuses on products to improve cardiovascular health. On July 26, 2012, the FDA approved Amarin‘s new drug application for Vascepa. Like Lovaza, Vascepa was approved as an adjunct to diet to reduce triglyceride levels in adults with severe hypertriglyceridemia. The sole active ingredient of Vascepa is a single molecule: icosapent ethyl, the ethyl ester of EPA. EPA, as noted above, is a significant component of the previously approved Lovaza mixture.
When Amarin applied for approval of Vascepa, it had already been in discussions with the FDA regarding both the approval process and the level of exclusivity to which Vascepa would be entitled. On June 14, 2008, representatives from Amarin attended a pre-investigational new drug (Pre-IND) meeting with the FDA to discuss what studies the FDA would require to approve Vascepa. The FDA‘s minutes from that meeting indicate that the parties discussed the FDA‘s preliminary answers to a list of Amarin‘s questions. AR 1044. Two of those questions were whether Vascepa, if approved, would qualify as a new chemical entity based on the fact that it was a single-entity active product that is molecularly distinct from any other approved product, and whether it would accordingly be entitled to five years of exclusivity. AR 1044. The FDA‘s preliminary answer to both questions was Yes. AR 1044. The Agency‘s guidance regarding Vascepa‘s proposed clinical research protocol, moreover, was expressly predicated on the assumption that Vascepa was a new chemical entity. The FDA required Amarin, for instance, to perform extra carcinogenicity studies because it is generally expected that a carcinogenicity study be conducted in two rodent species to support the marketing approval of a new chemical entity for a chronic use indication. AR 1042 (emphasis added).
Amarin submitted its new drug application, along with a request for five-year exclusivity, in September 2011. AR 1032-37. In its request for exclusivity, Amarin emphasized that EPA was not an active ingredient that was approved in the NDA covering Lovaza, because Lovaza was approved as a single active ingredient product and there was no evidence demonstrating whether specific components of the Lovaza mixture were active ingredients. AR 1033. In subsequent filings, Amarin explained that in its view, the key legal issue was whether the prior approval of a drug product, the active ingredient of which is a complex mixture of constituents, constitutes approval of each constituent as an active ingredient so as to preclude NCE exclusivity for a new drug product in which one of those constituents alone is the active ingredient. AR 51.
Amarin‘s new drug application for Vascepa was approved on July 26, 2012, but the FDA did not make a decision regarding five-year exclusivity at that time. See AR 73. Instead, the FDA issued a General Advice letter asking for Amarin‘s response to several precedents that the Agency viewed as consistent with the denial of exclusivity for Vascepa, and Amarin responded to that inquiry. AR 1050. The FDA reached a decision on February 21, 2014, eighteen months after Vascepa was approved. It concluded that Vascepa is not entitled to five-year exclusivity because EPA, the single active moiety in Vascepa, was also an active moiety contained in Lovaza. AR 1.
3. The FDA‘s Decision Denying Exclusivity
In its decision, the FDA explained that the statutory and regulatory authorities focus principally on single component active ingredients, rather than naturally derived mixtures, and that neither the statute nor the regulations expressly address 5-year NCE [new chemical entity] exclusivity in the context of naturally derived mixtures. AR 6. The Agency further noted that its prior decisions and statements have not necessarily resulted in consistent outcomes or used precise terminology in addressing exclusivity in the context of naturally derived mixtures. Id.
For naturally derived mixtures comprising multiple molecules, on the other hand, the FDA explained that the distinction between ‘active ingredient’ and ‘active moiety’ ... becomes crucial. Id. If more than one of the molecules contained in a naturally derived mixture could be responsible for the physiological or pharmacological action of the drug substance, the FDA must decide what constitutes the active ingredient and what constitutes the active moiety or active moieties. AR 6-7. To do so, the Agency decided to treat poorly characterized and well-characterized mixtures differently. For poorly characterized mixtures, where it is difficult to determine with any certainty ... which molecules in the mixture are consistently present or potentially are responsible for the physiological or pharmacological activity of the drug, the Agency treats the entire mixture as the active moiety. AR 7. This result is born of necessity, and each new version of such a naturally derived mixture would be eligible for 5-year NCE exclusivity. Id. In contrast, for well characterized mixtures, where at least some components of the mixture ... are consistently present and active, the FDA decided to treat the entire mixture [as a] single active ingredient, while recognizing that the active ingredient contains more than one component active moiety. Id.
The FDA then proceeded to lay out, for the first time, a three-part framework for identifying the active moiety or moieties of such mixtures. AR 6. Under that framework, the FDA generally considers component molecules of a mixture to be previously approved active moieties for purposes of determining a subsequent drug‘s eligibility for five-year exclusivity where (1) specific molecules in the mixture have been identified; (2) those specific molecules are consistently present in the mixture; and (3) those molecules are responsible at least in part for the physiological or pharmacological action of the mixture, based on a finding that they make a meaningful contribution to the activity of the mixture. AR 7-8. The determination of whether a particular molecular component of a previously approved mixture meets these criteria is based on technological tools and scientific concepts available at the time the FDA evaluates the exclusivity of a new drug—not the understanding that the FDA had when it approved the mixture in the first place. AR 8, n. 36.
The FDA thus identified two different approaches to the five-year exclusivity analysis. When dealing with single molecule drugs, the FDA applies a one-to-one approach. That is, the drug product contains a single active ingredient and a single active moiety, so there is a simple, one-to-one relationship between the active ingredient and the active moiety. AR 6. [O]f necessity, the FDA applies this same approach to poorly characterized
Applying this framework, the FDA concluded that Lovaza is a well-characterized mixture with respect to its omega-3 acid components, and it thus applied its one-to-many approach. AR 16. The FDA further concluded that (1) the EPA in the Lovaza mixture has been identified as a specific molecule present in the mixture, (2) it is consistently present in the mixture, and (3) the available evidence establishes that EPA has meaningful pharmacological activity in lowering serum triglyceride levels, the approved indication for both Lovaza and Vascepa. Id. Because EPA is the active moiety in Vascepa, the FDA concluded that both drugs contain the same active moiety and that Vascepa, accordingly, does not qualify for five-year exclusivity.
The FDA acknowledged that it has not always clearly set out its rationale for its determinations in the past; that neither the Agency nor regulated industry has used consistent terminology in this context; and that past [FDA] exclusivity determinations have not always been consistent. AR 22. Nonetheless, it concluded that in light of the relevant authorities, applicable scientific principles and past Agency action, its newly announced three-part test best harmonizes the relevant authorities and the outcomes of relevant prior Agency actions. Id. Although Amarin invested in the studies necessary to obtain market approval for Vascepa, in the FDA‘s view, [t]he amount of research that a sponsor invests in a drug is not determinative of that drug‘s eligibility for 5-year NCE exclusivity. AR 23. Investment of that type, instead, is a central factor in whether a drug is eligible for 3-year exclusivity. Id. Here, because Vascepa and Lovaza contain the same active moiety (EPA), the FDA concluded that Vascepa is entitled only to 3-year exclusivity.
II. STANDARD OF REVIEW
Amarin‘s challenge to the FDA‘s decision denying five-year exclusivity to Vascepa implicates the Agency‘s interpretation of the Hatch-Waxman Amendments and its reasoning in its administrative decision denying Amarin‘s request for exclusivity, AR 1-24. The case thus involves at least two distinct but potentially overlapping standards of APA review. Fox v. Clinton, 684 F.3d 67, 74 (D.C. Cir. 2012).
First, to the extent the FDA‘s denial of five-year exclusivity for Vascepa was governed by its interpretation of the Hatch-Waxman Amendments, its decision is subject to review under the two-step framework established in Chevron U.S.A. v. Natural Resources Defense Council, 467 U.S. 837, 104 S. Ct. 2778, 81 L. Ed. 2d 694 (1984). Under the first step of Chevron, the Court must consider whether Congress has directly spoken to the precise question at issue. Id. at 842, 104 S. Ct. 2778. If so, the Court must give effect to the unambiguously expressed intent of Congress. Id. at 843, 104 S. Ct. 2778. In making this determination, the Court applies the traditional tools of statutory construction, id. at 843, 104 S. Ct. 2778 n.9, including looking to the text, structure, and the overall statutory scheme, as well
Second, more generally, the APA precludes agency action that is arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.
These standards overlap and are, at times, intertwined. Most significantly, because Chevron‘s second step asks whether an agency‘s interpretation is arbitrary or capricious in substance, Judulang v. Holder, 565 U.S. 42, 132 S. Ct. 476, 483 n. 7, 181 L. Ed. 2d 449 (2011) (citing Mayo Found. for Med. Educ. & Research v. United States, 562 U.S. 44, 131 S. Ct. 704, 711, 178 L. Ed. 2d 588 (2011)), the analysis required pursuant to Chevron [s]tep [t]wo, and that required under the arbitrary and capricious standard enunciated in State Farm overlap, EchoStar Satellite L.L.C. v. FCC, 704 F.3d 992, 1001-02 (D.C. Cir. 2013). Similarly, mirroring the arbitrary and capricious standard, a court will affirm an agency‘s interpretation at Chevron step two only if the agency has offered a reasoned explanation for why it chose that explanation. Vill. of Barrington, Ill. v. Surface Transp. Bd., 636 F.3d 650, 660 (D.C. Cir. 2011). Ultimately, under either standard of review, the relevant question is whether the FDA‘s decision represents the result of a reasonable exercise of its authority.
III. DISCUSSION
Under the text of the Hatch-Waxman Amendments, a drug manufacturer is entitled to five-year exclusivity if its newly-approved drug does not contain an active ingredient that was previously approved in another drug application.
In its administrative decision, the FDA espouses two conflicting definitions of active ingredient. At times, the Agency asserts that active ingredient and active moiety mean different things, and, indeed, the FDA criticizes Amarin for conflat[ing] the two concepts. AR 6. In a footnote, however, the FDA quotes from the Final Rule adopting the five-year exclusivity regulations, which asserts that, for relevant purposes, active ingredient means active moiety. Id. Unsurprisingly, the FDA does not now rely on the first of these interpretations in its effort to reconcile the statutory reference to a drug‘s active ingredient with the regulatory reliance on the drug‘s active moiety. If these concepts are distinct, as the administrative decision suggests, the statutory language must control, and the decision must be set aside. As Amarin correctly observes, the one thing that can be said with certainty about the meaning of the statute is that active ingredient cannot logically mean not active ingredient.
Thus, for purposes of this action, the FDA focuses its argument on the interpretation it offered in the footnote, under which active ingredient and active moiety mean the same thing, and it argues that this definition survives scrutiny under the two-step framework in Chevron. Following this framework, the FDA first argues that the statutory phrase active ingredient is ambiguous, and that Congress, accordingly, left a gap for the FDA to fill. It then maintains that its decision to adopt the active moiety approach in the face of this ambiguity was reasonable and that it is entitled to the Court‘s deference.
A. Chevron Step One
Under Chevron step one, the Court must consider whether Congress has directly spoken to the precise question at issue. 467 U.S. at 842, 104 S. Ct. 2778. If so, the Court must give effect to the unambiguously expressed intent of Congress. Id. at 843, 104 S. Ct. 2778. In making this determination, the Court applies the traditional tools of statutory construction, id. at 843 n. 9, 104 S. Ct. 2778, including looking to the text, structure, and the overall statutory scheme, as well as the problem Congress sought to solve, Fin. Planning Ass‘n, 482 F.3d at 487; see also Cal. Indep. Sys. Operator Corp., 372 F.3d at 400. Here, the precise question at issue is whether, for purposes of exclusivity, the Hatch-Waxman Amendments permit the FDA to interpret active ingredient that has been approved to mean any active moiety in a previously approved drug.
1. Abbott Labs and Actavis Elizabeth
The FDA starts by arguing that the phrase active ingredient is ambiguous, leaving room for the Agency to read it to mean active moiety for purposes of the five-year exclusivity provision. According to the FDA, the Court of Appeals has twice held that the phrase active ingredient, as used in the exclusivity provision, is ambiguous—first in Abbott Labs, 920 F.2d at 987-88, and again in Actavis Elizabeth LLC v. FDA, 625 F.3d 760, 764 (D.C. Cir. 2010).
The FDA‘s reliance on Actavis is better placed but also fails to advance the Agency‘s position. In Actavis, the plaintiff argued that the phrase active ingredient unambiguously refers to the post-ingestion form of a single-molecule active ingredient, rather than to the ingredient‘s form before it is ingested. The Court of Appeals rejected that argument and concluded that the phrase active ingredient was in relevant respects ambiguous, and that the FDA had acted within the scope of its delegated authority when it interpreted active ingredient to refer to the pro-drug form of a molecule.9 Actavis, 625 F.3d at 764-65. But even though Actavis makes clear that the term active ingredient is ambiguous at least to the extent that it may refer to either a pre- or post-ingestion form of a drug, that conclusion does not resolve whether Congress has directly spoken to the precise question at issue here, Chevron, 467 U.S. at 842, 104 S. Ct. 2778. A judicial decision concluding that a statutory term admits of some ambiguity does not open the door at Chevron step one for purposes of all interpretations.
2. The FDA‘s Interpretation Of Active Ingredient
Because the Court of Appeals has not previously addressed whether the Hatch-Waxman Amendments encompass the active moiety interpretation at issue here, the Court must consider whether the statute unambiguously forbids the Agency‘s interpretation. Barnhart v. Walton, 535 U.S. 212, 218, 122 S. Ct. 1265, 152 L. Ed. 2d 330 (2002). The question is not whether the phrase is, in some abstract sense, ambiguous, or whether it may be susceptible to various interpretations in another context, but whether, read in context and using the traditional tools of statutory construction, [it] encompasses the agency‘s construction. Cal. Indep. Sys. Operator Corp., 372 F.3d at 400; see also Indep. Ins. Agents, 211 F.3d at 644-45. This analysis may sound like Chevron step two because the term active ingredient is ambiguous in some applications, but, under Chevron step one, where the text and reasonable inferences from it give a clear answer against the government ... that ... is the end of the matter. Cal. Indep. Sys. Operator Corp., 372 F.3d at 401 (quotation marks omitted). In any event, whether analyzed under Chevron step one or two, the answer is the same; the statute‘s text, structure, and purpose do not encompass or permi[t] the con-
a. The Canon Against Surplusage
First, the contention that active ingredient means active moiety is at odds with the canon against surplusage. See Indep. Ins. Agents, 211 F.3d at 644-45 (rejecting agency‘s interpretation at Chevron step one based on the tandem canons of avoiding surplusage and expressio unius). It is a cardinal principle of statutory construction that a statute ought, upon the whole, to be construed in manner that ensures that no clause, sentence, or word shall be superfluous, void, or insignificant. TRW Inc. v. Andrews, 534 U.S. 19, 31, 122 S. Ct. 441, 151 L. Ed. 2d 339 (2001) (citation omitted). The FDA‘s contention that active ingredient means active moiety, however, would render the parenthetical clause in the exclusivity provisions either redundant or incomprehensible.
If active ingredient means active moiety, and active moiety is defined as a molecule excluding (among other things) those portions that render the molecule a salt or an ester,
The regulatory history highlights why the FDA‘s active moiety approach cannot be reconciled with the statutory text. Although the parenthetical clause expressly refers only to any ester or salt of the active ingredient,
b. The Presumption Of Consistent Usage
The second problem with the Agency‘s interpretation is that it requires the Agency to interpret the phrase active ingredient differently for purposes of the ANDA and exclusivity provisions of the Act. In the ordinary course, however, identical words used in different parts of the same Act are intended to have the same meaning. Sorenson v. Sec‘y of Treasury, 475 U.S. 851, 860, 106 S. Ct. 1600, 89 L. Ed. 2d 855 (1986) (quotation marks omitted); see also, e.g., Powerex Corp. v. Reliant Energy Servs., Inc., 551 U.S. 224, 232, 127 S. Ct. 2411, 168 L. Ed. 2d 112 (2007) (A standard principle of statutory construction provides that identical words and phrases within the same statute should normally be given the same meaning.); IBP, Inc. v. Alvarez, 546 U.S. 21, 34, 126 S. Ct. 514, 163 L. Ed. 2d 288 (2005) (noting that identical words used in different parts of the same statute are generally presumed to have the same meaning). Absent good reason, it is safe to assume that Congress intended active ingredient to have the same meaning when it used that term in different, but closely related, places in the same statute.
In the present context, there are strong reasons to conclude that the presumption of consistent usage applies. First, the Hatch-Waxman Amendments were designed to strike a compromise between the interests in bringing generic drugs to market quickly and inexpensively and providing a period of market exclusivity to innovators. See Abbott Labs, 920 F.2d at 985; see also, e.g., 130 Cong. Rec. 23058-59 (1984) (statement of Rep. Synar) (This bill is an important compromise that improves research and development and increases price competition in the drug marketplace.). Although not conclusive, this suggests that Congress intended, at least at a general level, to treat the pro-generic and pro-branded provisions of the law as flip-sides of the same coin, a coherent set of incentives for stakeholders with competing interests. Cf. Teva Pharm. Indus. Ltd. v. Crawford, 410 F.3d 51, 54 (D.C. Cir. 2005) (Because the balance struck in the Hatch Waxman Act is quintessentially a matter for legislative judgment, the court must attend closely to the terms in which the Congress expressed that judgment.); Ranbaxy Labs. Ltd. v. Leavitt, 469 F.3d 120, 126 (D.C. Cir. 2006) (The FDA may not ... change the incentive structure adopted by the Congress.).
Congress, moreover, enacted the relevant references to active ingredients at the same time, in the same amendments, and inserted the language into the same subsection of the FDCA. See
As the FDA correctly observes, the presumption of consistent usage is not unrebuttable. It will give way, for example, where application of the presumption creates a genuine conflict with the purpose or text of the statute or where Congress uses a term profligate[ly] in an Act, including in contexts where a consistent definition is incompatible with the overall statutory structure. See Util. Air Regulatory Grp. v. EPA, 134 S. Ct. 2427, 2441-42, 189 L. Ed. 2d 372 (2014). At oral argument, the FDA further argued that the meaning of the phrase active ingredient turns on context. Tr. Oral Arg. 68. When Congress provided that a generic drug may take advantage of the ANDA process if its active ingredient(s) are the same as the active ingredient(s) contained in a previously approved and listed drug, Congress intended to refer to entire molecules or mixtures, and not simply the active moiety. See
In Abbott Labs, the Court of Appeals confirmed that Congress may, in some circumstances, use the same word to mean different things within the same statute. There, as here, the FDA argued that Congress was using the term active ingredient loosely in the five-year exclusivity provision, possibly as a virtual synonym for active moiety. Abbott Labs, 920 F.2d at 987. But in Abbott Labs, the Court declined to consider that theory because the agency had not employ[ed] it in its administrative decision. Id.
As in Abbott Labs, 920 F.2d at 987, the Court does not foreclose the possibility that the Agency might reasonably construe and apply the phrase active ingredient to have a different meaning in different contexts or when dealing with different kind[s] of drugs, Serono Labs., Inc. v. Shalala, 158 F.3d 1313, 1319 (D.C. Cir. 1998) (upholding FDA‘s interpretation of statutory requirement that active ingredient in ANDA be the same as previously approved active ingredient to require clinical identity, rather than chemical identity, in context of category of drugs for which slight natural variations in molecular structure were expected). But, as discussed above, this not a circumstance where Congress has used the phrase loosely or where a consistent definition is incompatible with the overall statutory structure or purpose, see Util. Air Regulatory Grp., 134 S. Ct. at 2441-42. The ANDA and exclusivity provisions may serve different functions, but they are part of a single statutory structure, and Congress added a
The FDA‘s own practices, moreover, demonstrate that adopting a consistent interpretation of active ingredient would not result in any incompatibility with the statutory structure or purpose. To the contrary, in other contexts, the FDA has made its exclusivity determination based on an entire mixture, rather than the mixture‘s active moieties. See, e.g., AR 413 (noting that the same ‘active’ components were present in different quantities in InfaSurf and Survanta, but explaining that the agency would treat the entire mixture as the active moiety); AR 71 (noting that the FDA considered the entire mixture to be the active moiety for Orphan Drug exclusivity purposes). Indeed, the FDA concedes that its approach in some prior cases is irreconcilable with its new framework. See Dkt. 13-1 at 21; AR 20. And even under the framework the FDA announced in this case, for poorly characterized mixtures, the FDA‘s exclusivity analyses will continue to focus on mixtures as a whole.
Notably, the statute contemplates that the FDA will approve drugs containing multiple active ingredients, and, with respect to at least one naturally derived mixture, the FDA has taken precisely this approach. See AR 15-16; Tr. Oral Arg. 77-78 (discussing the FDA‘s characterization of Menotropins as mixtures composed of multiple active ingredients). The FDA‘s approach with respect to Lovaza stands in contrast: Not only did the FDA treat the Lovaza mixture as a single active ingredient for purposes of approval, it also denied a Citizens’ Petition requesting that the mixture be re-characterized as containing multiple active ingredients on the same day that it issued its decision in the instant dispute. See Lovaza Citizen Petition Response. Ultimately, the FDA is free to determine whether any particular naturally derived mixture is better understood as containing one or multiple active ingredients. To the extent that the FDA is concerned that granting five-year exclusivity for different mixtures will unduly allow pharmaceutical companies to obtain exclusivity for components of mixtures that were already well-understood, it can take precisely that approach.
Although the FDA did not discuss the issue in its administrative process, it argues before this Court that the exclusivity and ANDA provisions serve different purposes, because the exclusivity provision is designed to promote novelty while the ANDA provisions require the FDA to ascertain whether generic drugs are safe and effective. Tr. Oral Arg. 55; id. at 7. But, as explained above, while the two provisions do, of course, play different roles, they are part of a unified statutory scheme intended to strike a balance between fostering innovation and promoting access to affordable medications. When the phrase is interpreted to mean the same thing, each provision still performs its function within the statute.
In any event, the Agency‘s interpretation cannot be squared with what the Agency actually said in its decision. Rather than positing that, when it asserted that Lovaza‘s active ingredient was the entire mixture, it was using the term active ingredient in a different manner than it is used in the exclusivity provision, the FDA said just the opposite. Throughout the decision, it is clear that the FDA is using the phrase as it is used in the exclusivity provision. The decision says under FDA‘s regulations, a drug‘s active ingredient is distinct from its active moiety, and, at least in the case of a naturally derived mixture, a single active ingredient can have multiple active moieties. AR 23 (emphasis added). The Agency‘s use of
To be sure, an agency need not set forth its reasoning in each individual adjudication, where it has done so in prior adjudication or in a rulemaking that supplies a generally applicable rationale or explanation that is then simply applied in the adjudication. See WLOS TV, Inc. v. F.C.C., 932 F.2d 993, 995 (D.C. Cir. 1991) (Where an agency merely implements prior policy, an explanation that allows this court to discern ‘the agency‘s path’ will suffice.) (quoting Hall v. McLaughlin, 864 F.2d 868, 872-73 (D.C. Cir. 1989)). But the Agency still must demonstrate that it has given reasoned consideration to all the material facts and issues. See Ventura Broad. Co v. F.C.C., 765 F.2d 184, 189 (D.C. Cir. 1985) (quotation marks omitted). Yet, at no point does the FDA appear to have considered how its new application of the active moiety approach to naturally derived mixtures can be reconciled with the text, structure, or purposes of the Hatch-Waxman Amendments. Although the Agency‘s Final Rule does state that the phrase active ingredient should be construed to mean active moiety, 59 Fed. Reg. at 50,358, there is no evidence that the Agency ever considered, much less explained, how this approach might sensibly apply in the context of naturally derived mixtures or other drugs where the Agency has construed active ingredient and active moiety to refer to significantly different substances. To the contrary, the administrative decision candidly observes that the regulations do not address 5-year NCE exclusivity in the context of naturally derived mixtures but rather focus principally on single component ingredients. AR 6.
c. The Statutory Focus On An Active Ingredient That Has Been Approved
The third (and related) difficulty with the Agency‘s approach is that its focus on a drug component that was never the subject of the FDA‘s approval is also inconsistent with the statutory text, which considers whether the new drug contains an active ingredient which has been approved in [a prior] application.
Thus, although there is no dispute that at the time the NDA for Lovaza was approved, the entire mixture of omega-3 acid ethyl esters was list[ed] as the drug‘s active ingredient, AR 2, the FDA applied its three-step test to determine the active moieties in Lovaza more than seven years later, at the time Vascepa was approved. See AR 7-8. In 2014, the Agency assessed (1) whether the mixture is well-characterized, (2) whether one or more specific molecules ... are consistently present, and (3) whether those molecules are responsible at least in part for
The FDA does not argue that an active moiety contained in a naturally derived mixture is ever approved in any sense of the word. At oral argument, the FDA explained that it approves drugs and not active ingredients. Tr. Oral Arg. 62-63. It thus argues that the statutory reference to an ingredient that has been approved cannot carry significant weight. Put differently, the FDA argues that Congress made a mistake when it referred to the FDA‘s approval of an active ingredient, and that, as a result, it is reasonable for the Agency to consider any component of a drug which [the Agency] has approved, Tr. Oral Arg. 63, when it later identifies the constituent moieties of the approved drug. According to the FDA, this reasoning applies even where the moiety is a molecule that was not independently evaluated by the FDA.
It is not correct, however, to say that the FDA does not approve active ingredients when it approves drugs or drug products. In at least one other case, the FDA has affirmatively argued—and the Court agreed—that the FDA approves active ingredients as well as finished drug products. Pharmanex v. Shalala, 221 F.3d 1151, 1154-56 (10th Cir. 2000) (accepting FDA‘s argument that the phrase an article that is approved as a new drug includes active ingredients, and explaining that [i]t is evident from § 355 that approval of active ingredients is integral to the overall new drug approval process). At times, the FDCA refers to the approval of drugs, see, e.g.,
Moreover, even if Congress could have been more precise, its intent is evident. It makes no material difference whether it is the active ingredient that is approved in an NDA—as Congress wrote—or whether, slightly more literally, it is the NDA that
Determining which molecules of previously approved drugs are relevant for exclusivity determinations based on information that was not before the FDA when it approved those drugs is also inconsistent with the statutory purpose of the exclusivity provision, which was intended to provide incentives for drug manufacturers to invest in innovation. See Actavis Elizabeth, 625 F.3d at 764. Under the FDA‘s approach, a would-be innovator would not know whether its new drug would qualify for market exclusivity until after the drug was fully developed and approved by the FDA. At that point in time, the FDA would then consider whether a component of a previously approved drug was consistently present in that drug and whether it was responsible at least in part for the physiological or pharmacological action of the mixture. AR 7-8. As this case illustrates, that approach would inject substantial uncertainty into the process and would detract from the incentives for innovation that Congress intended to provide. The Agency has not even hinted at a statutory purpose or policy rationale that would render that approach a reasonable interpretation of the statute.
Of course, as the FDA points out, there is inherent uncertainty as to whether a particular drug will be entitled to five-year exclusivity because there is always a risk that a would-be innovator will be beaten to the punch by another manufacturer of a drug with the same active ingredient. But the FDA‘s approach multiplies this uncertainty. Innovators would not only be required to race against each other, they would not know whether they were destined to lose even before beginning the race because the FDA might conclude that a long-ago-approved mixture contained a disqualifying active moiety. The FDA has offered no basis to justify its decision to construe the statute‘s reference to an active ingredient that has been approved as referring to any active moiety that the FDA might someday find is consistently present and responsible at least in part for the mixture‘s pharmacological action.
3. The FDA‘s Alternative Arguments
In an effort to avoid at least some of these difficulties, the FDA intermittently argues that it interprets the entire clause active ingredient (including any ester or salt of the active ingredient)—rather than just the phrase active ingredient—to mean active moiety. See, e.g., Dkt. 29 at 2-3; Tr. at 88 (explaining that the surplusage problem is what FDA was getting at when they said in the regs that this was interpreting the entire phrase with the parenthetical). But that interpretation is not the one stated in the administrative decision or the Final Rule. The FDA was clear: The agency has concluded that the term ‘active ingredient,’ as used in the phrase ‘active ingredient (including any ester or salt of the active ingredient),’ means active moiety. AR 6, n.31 (quoting 59 Fed. Reg. at 50,358) (emphasis added). Although the FDA was defining active ingredient as used in the context of the broader clause, the term that the Agency was defining was active ingredient. It is on this basis that the Agency must defend its decision. See Williams Gas Processing-Gulf, 475 F.3d at 329 (We do not
Moreover, even if the Agency had attempted to define the entire clause to mean active moiety, that approach would run into problems of its own, stemming from the Court of Appeals’ holding in Abbott Labs. In Abbott Labs, the FDA argued that the parenthetical clause expressed Congress‘s intention to treat all forms of a molecule that reduce to the same moiety—not just salts and esters—alike for purposes of exclusivity. 920 F.2d at 988. The Court of Appeals rejected the government‘s unconvincing attempts to employ the ‘including’ clause to cover all possible permutations of active ingredient. Id. To the extent that the FDA now relies on the entire clause, including the parenthetical, rather than just the term active ingredient, it encounters the same difficulty that the Court of Appeals identified in Abbott Labs. If [i]t is simply not plausible to read ‘including any salt or ester’ as merely illustrative, to mean including any form that eventually produces the same active moiety, id. it seems equally implausible to read the entire phrase active ingredient (including any ester or salt of the active ingredient) to have that meaning. It is difficult to discern what this variation adds to the argument rejected in Abbott Labs.
More narrowly, the FDA suggests that, even in the absence of a relevant textual ambiguity in the statute, Congress may, at a minimum, have left a gap for the Agency to fill when dealing with naturally derived mixtures. Chevron, 467 U.S. at 843, 104 S. Ct. 2778. The Agency notes that the statute does not expressly address 5-year NCE exclusivity in the context of naturally derived mixtures and that the relevant statutory and regulatory authorities on 5-year NCE exclusivity appear to focus principally on single component active ingredients. AR 6; see also Dkt. 13-1 at 16 ([T]he fact that Lovaza and Vascepa are made of a naturally derived mixture offers another reason to reject Amarin‘s suggested interpretation of ‘active ingredient’ here.); Oral Arg. Tr. 56, 72-73. Because there is no evidence that Congress considered this unique context, the FDA submits that it is left with discretion to fashion a reasonable solution for the problem.
This argument misunderstands both Chevron and the problem with the Agency‘s interpretation of the exclusivity provisions. Chevron does not ask whether Congress affirmatively considered each context in which the statute might be applied. Rather, it requires that courts apply the traditional tools of statutory construction and, only after doing so, ask whether the statute is in relevant respects plain. Here, even if Congress did not consider how the law might apply to mixtures, it did not leave a relevant statutory gap for the FDA to fill.
The problem with the FDA‘s interpretation, moreover, is not that Congress failed to delegate to the FDA some discretion to address how to identify the active ingredient(s) of a naturally derived mixture; there may well be circumstances, for example, where the Agency must exercise its discretion, and expertise, to determine whether a particular mixture contains one active ingredient or several. Instead, the problem is that the statutory text and structure foreclose the exclusivity inquiry that the Agency undertook in these circumstances. Even assuming that Congress lacked a particular intent with respect to how the Agency should identify a mixture‘s active ingredient(s), the FDA‘s authority stemming from that gap extends only to answering that question—it does not confer an authority on the Agen-
The Court does not doubt that there are other contexts in which the FDA is entitled to deference in its interpretation of the Hatch-Waxman Amendments, particularly where that interpretation turns on scientific or medical expertise. See, e.g., A.L. Pharma, Inc. v. Shalala, 62 F.3d 1484, 1490 (D.C. Cir. 1995) ([C]ourts give a high level of deference to an agency‘s evaluations of scientific data within its area of expertise.). There may be times, moreover, where the meaning of the statutory phrase may vary based on the context of the kind of drug at issue. Cf. Serono, 158 F.3d at 1320-21 (interpreting same as in context of menotropins, which naturally vary such that an overly restrictive definition of variability would make it impossible to demonstrate sameness for ANDA purposes). But even where Congress has established an ambiguous line, the agency can go no further than the ambiguity will fairly allow. City of Arlington, Tex. v. F.C.C., 569 U.S. 290, 133 S. Ct. 1863, 1874, 185 L. Ed. 2d 941 (2013). In this case, the Court concludes that Congress has exceeded the bounds of its statutory authority to interpret the exclusivity provision, and that its interpretation, accordingly, fails at Chevron ‘s first step.
B. Chevron Step Two and Arbitrary and Capricious Review
Even if the statute were in relevant respects ambiguous, the FDA‘s interpretation would still fail at Chevron‘s second step, which requires the Court to determine whether the FDA has permissibly exercised its delegated authority. At Chevron step two, the Court defer[s] 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, 636 F.3d at 660. This analysis overlaps substantially with the APA‘s arbitrary and capricious inquiry, because [w]hether a statute is unreasonably interpreted is close analytically to the issue whether an agency‘s actions under a statute are unreasonable. Am. Fed‘n of Gov‘t Employees, AFL-CIO, Local 446 v. Nicholson, 475 F.3d 341, 355 (D.C. Cir. 2007) (quotation marks omitted); see also, e.g., Agape Church, Inc. v. F.C.C., 738 F.3d 397, 410 (D.C. Cir. 2013) (The analysis ... under Chevron Step Two and arbitrary and capricious review is often the same, because under Chevron step two, the court asks whether an agency interpretation is arbitrary or capricious in substance.) (quotation marks and brackets omitted).
An agency‘s action, moreover, must be upheld, if at all, on the basis articulated by the agency itself. State Farm, 463 U.S. at 50, 103 S. Ct. 2856. Although lack of clarity alone is not a sufficient basis to strike down an agency decision, Bowman Transp., Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 285-86, 95 S. Ct. 438, 42 L. Ed. 2d 447 (1974), neither counsel nor the Court may amend or supply the agency‘s rationale, see Riffin v. Surface Transp. Bd., 592 F.3d 195, 198 (D.C. Cir. 2010). In order to survive judicial review ..., an agency action must be supported by ‘reasoned decision-making.’ Tripoli Rocketry Ass‘n v. Bureau of Alcohol, Tobacco, Firearms and Explosives, 437 F.3d 75, 77 (D.C. Cir. 2006) (citations omitted). The result must be logical and rational, id. as well as adequately explained and coheren[t]. Fox, 684 F.3d at 75 (citations and internal quotation marks omitted).
The decision‘s sole acknowledgment of the apparent divergence between the statutory and regulatory inquiries is both minimal and confounding. In the course of its 24-page, single spaced administrative decision, the FDA refers to the statutory text only twice—in a background recitation of the governing legal framework, and in a parenthetical in a footnote. Quoting from the Final Rule adopting the five-year exclusivity regulations, that parenthetical states: The agency has concluded that the term ‘active ingredient,’ as used in the phrase ‘active ingredient (including any salt or ester of an active ingredient),’ means active moiety. AR 6, n.31. The footnote, however, raises more questions than it answers.
Most importantly, the footnote conflicts with other portions of the Agency‘s decision. The decision, for example, criticizes Amarin for conflat[ing] the terms active ingredient and active moiety. AR 6; id. at n.29. It stresses that the distinction between active ingredient and active moiety is crucial in the context of naturally derived mixtures, AR 6, and emphasizes more generally that a drug‘s active ingredient is distinct from its active moiety, AR 23. The decision‘s entire one-to-many analysis turns on the premise that some drugs may have one active ingredient, but many active moieties. AR 5, AR 19; AR 23. And, it repeatedly concludes that the active ingredient of Lovaza is the Lovaza mixture as a whole, and not EPA or any other active moiety. E.g. AR 19 (the Agency agrees that the active ingredient of Lovaza is the Lovaza mixture as a whole). Thus, although the footnote appears to treat the phrases active ingredient and active moiety as synonymous for purposes of the five-year exclusivity determination, the remainder of the decision is emphatic in concluding that the terms have distinct meanings, at least in the context of naturally derived mixtures.
Moreover, although an administrative decision can rely on an agency‘s prior consideration of an issue, the FDA has never addressed why the phrase active ingredient should be given different meanings in different provisions of the Act, let alone explained how the regulatory focus on active moiety can apply where the active ingredient and active moiety refer to different substances. The Agency makes no attempt to explain how its approach furthers Congress‘s purposes or is otherwise a reasonable policy choice, especially in light of the clear interest in providing notice to potential innovators of the exclusivity to which they might eventually be entitled. And the FDA‘s regulations do
The Agency‘s ultimate conclusion that Vascepa, a drug no active ingredient of which ... has been approved in a previous NDA, was not entitled to exclusivity, is contrary to the statute‘s plain meaning. Rather than explaining this discrepancy, the administrative decision only adds to the problem by emphasizing the divergence between the Agency‘s regulatory inquiry and the statutory requirement. Whether the problems with the FDA‘s decision are characterized as failures under Chevron step one, step two, or the APA‘s requirement of reasoned decision-making, the Agency‘s decision must be set aside.11
IV. CONCLUSION
For the reasons set forth above, Plaintiff‘s Motion for Summary Judgment is GRANTED and Defendants’ Cross-Motion for Summary Judgment is DENIED. The decision denying Amarin‘s claim for exclusivity vacated and the matter is remanded to the FDA for proceedings consistent with this Opinion.
Carlos LOUMIET, Plaintiff, v. UNITED STATES of America, et al, Defendants.
Civil Action No. 12-1130 (CKK)
United States District Court, District of Columbia.
Signed May 28, 2015