Amneal Pharmaceuticals Llc v. Food And Drug Administration et alMOTION for Summary JudgmentD.D.C.March 29, 2017IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA AMNEAL PHARMACEUTICALS LLC, Plaintiff, v. FOOD AND DRUG ADMINISTRATION, et al., Defendants, and LUPIN PHARMACEUTICALS, INC., et al., Intervenor-Defendant. Case No. 1:17-cv-00180 (RDM) PLAINTIFF’S MOTION FOR SUMMARY JUDGMENT Pursuant to Rule 56 of the Federal Rules of Civil Procedure, Plaintiff Amneal Pharmaceuticals LLC (“Amneal”) moves for summary judgment against the federal Defendants (collectively “FDA”) and Intervenors Lupin Ltd. and Lupin Pharmaceuticals, Inc. The Administrative Procedure Act requires the Court to set aside agency action that is arbitrary, capricious, or not in accordance with law. 5 U.S.C. § 706(2)(A). For the reasons set forth in the accompanying Memorandum of Points and Authorities, FDA’s determination that Amneal forfeited 180-day exclusivity under 21 U.S.C. § 355(j)(5)(D)(i)(IV) must be set aside because it is contrary to the unambiguous language of the Federal Food, Drug and Cosmetic Act, fails to provide a reasoned explanation for its departure from the statute’s clear terms and from this Court’s and FDA precedent, and disregards myriad record facts relevant under the statute. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 1 of 50 The Court should enter summary judgment in Amneal’s favor; vacate FDA’s decision that Amneal forfeited 180-day exclusivity; vacate any final approval by FDA of any abbreviated new drug application granted in violation of Amneal’s exclusivity; enjoin FDA from granting final approval to any other such applications until the expiration of Amneal’s 180-day exclusivity; and declare that Amneal has not forfeited 180-day exclusivity. Pursuant to Local Civil Rule 7(f), Amneal respectfully requests oral argument on this Motion. A proposed Order is attached. DATED: March 29, 2017 AMNEAL PHARMACEUTICALS LLC, By its attorneys, /s/ Areta Kupchyk Areta Kupchyk (D.C. Bar No. 483626) FOLEY HOAG LLP 1717 K Street, N.W. Washington, DC 20006 Tel.: 202-223-1200 Dean Richlin (pro hac vice) Marco J. Quina (pro hac vice) FOLEY HOAG LLP 155 Seaport Boulevard Boston, MA 02210 Tel.: 617-832-1000 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 2 of 50 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA AMNEAL PHARMACEUTICALS LLC, Plaintiff, v. FOOD AND DRUG ADMINISTRATION, et al., Defendants, and LUPIN PHARMACEUTICALS, INC., et al., Intervenor-Defendant. Case No. 1:17-cv-00180 (RDM) MEMORANDUM OF POINTS AND AUTHORITIES IN SUPPORT OF PLAINTIFF’S MOTION FOR SUMMARY JUDGMENT Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 3 of 50 i TABLE OF CONTENTS Contents INTRODUCTION......................................................................................................................... 1 STATEMENT OF FACTS ........................................................................................................... 3 A. Legal Background ....................................................................................................3 B. Factual Background .................................................................................................5 1. The Eight-Month Delay in FDA’s Receipt and Commencement of Review of Amneal’s ANDA ........................................................................5 2. FDA’s Demand for Full-Scale Commercial Batch Data Results in an Additional 14-Month Delay .........................................................................6 3. FDA’s 180-Day Exclusivity Forfeiture Ruling..........................................11 4. FDA’s Approval of Amneal’s ANDA and of Third-Party ANDAs ..........12 ARGUMENT ............................................................................................................................... 13 I. FDA’s Rigid, Single-Factor Causation Analysis Fails to Satisfy Chevron and is not the Product of Reasoned Decision Making ............................................................ 13 A. FDA’s Interpretation of “Caused By” as Creating a Rigid, Irrebutable, Single-Factor Causation Test Fails Chevron Step One..........................................14 B. FDA’s Single-Factor Causation Test Fails Chevron Step Two Because It Lacks Any Justification, Conflicts with Prior Precedent without Explanation, and is Self-Contradictory .......................................................................................19 C. FDA’s Decision is not the Product of Reasoned Decision Making Because It Ignores Numerous Facts Relevant to Causation ....................................................23 II. FDA Improperly Concluded Amneal’s ANDA did not Experience a Change in or Review of the Requirements for Approval of the Application ............................... 28 A. Amneal’s ANDA Experienced a Change in or Review of the Requirements for Approval of the Application .............................................................................29 1. FDA’s Commercial Scale-Up Demand .....................................................29 2. FDA’s Review of Whether to Receive the ANDA with a Typographical Error ...................................................................................31 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 4 of 50 - ii - B. FDA’s Requirement that Amneal Show a Change to a “Statutory Requirement” or to FDA Policy Ignores the FDCA’s Text, Failing Chevron Step One .................................................................................................................33 C. FDA’s Limitation of the Statute to Statutory or Policy Change Relies on a Secret FDA Policy and is Inconsistent with Prior Precedent .................................35 III. Amneal is Entitled to a Declaration That It Has Not Forfeited Exclusivity .............. 38 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 5 of 50 - iii - TABLE OF AUTHORITIES* Cases A.H. Phillips, Inc. v. Walling, 324 U.S. 490 (1945) ................................................................................................................. 16 Achernar Broad. Co. v. FCC, 62 F.3d 1441 (D.C. Cir. 1995) ................................................................................................. 24 * Amarin Pharms. Ir. Ltd. v. FDA, 106 F. Supp. 3d 196 (D.D.C. 2015) .................................................................................. passim Am. Maritime Ass’n v. Blumenthal, 590 F.2d 1156 (D.C. Cir. 1978) ............................................................................................... 17 Amerijet Int’l v. Pistole, 753 F.3d 1343 (D.C. Cir. 2014) ............................................................................................... 24 Atchison. v. Wichita Bd. of Trade, 412 U.S. 800 (1973) ................................................................................................................. 22 Bennett v. Donovan, 4 F. Supp. 3d. 5 (D.D.C. 2013) ................................................................................................. 33 Chevron, U.S.A., Inc. v. NRDC, 467 U.S. 837 (1984) .......................................................................................................... passim Christopher v. SmithKline Beecham Corp., 132 S. Ct. 2156 (2012) ....................................................................................................... 36, 37 Donovan v. Stafford Constr. Co., 732 F.2d 954 (D.C. Cir. 1984) ................................................................................................. 39 Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117 (2016) ....................................................................................................... 24, 25 FCC v. American Broadcasting Co., 347 U.S. 284 (1954) ................................................................................................................. 17 George Hyman Constr. Co. v. Brooks, 963 F.2d 1532 (D.C. Cir. 1992) ............................................................................................... 39 * The cases and authorities on which Amneal chiefly relies are marked with asterisks. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 6 of 50 - iv - Greater Bos. Television Corp. v. FCC, 444 F.2d 841 (D.C. Cir. 1970) ..................................................................................................... 22 Indep. Ins. Agents of Am., Inc. v. Hawke, 211 F.3d 638 (D.C. Cir. 2000) ................................................................................................. 34 Jicarilla Apache Nation v. U.S. Dept. of Interior, 613 F.3d 1112 (D.C. Cir. 2010) ................................................................................................ 21 Kachmar v. SunGard Data Sys., Inc., 109 F.3d 173 (3rd Cir. 1997) ................................................................................................... 15 Laffey v. Northwest Airlines, Inc., 567 F.2d 429 (D.C. Cir. 1976) ................................................................................................. 16 Long Island Care at Home, Ltd. v. Coke, 551 U.S. 158 (2007) ................................................................................................................. 37 Martin v. OSHRC, 499 U.S. 144 (1991) ................................................................................................................. 36 *Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983) ....................................................................................................... 20, 24, 25 Mova Pharm. Corp. v. Shalala, 140 F.3d 1060 (D.C. Cir. 1998) ............................................................................................... 18 *Mylan Labs. Ltd. v. FDA, 910 F. Supp. 2d 299 (D.D.C. 2012) .................................................................................. passim Mylan Labs. Inc. v. Thompson, 389 F.3d 1272 (D.C. Cir. 2004) ............................................................................................... 14 Nat’l Black Media Coal. v. FCC, 775 F.2d 342 (D.C. Cir. 1985) ................................................................................................. 22 NLRB v. Wyman, 394 U.S. 759 (1969) ................................................................................................................. 39 Peter Pan Bus Lines, Inc. v. Fed. Motor Carrier Safety Admin., 471 F.3d 1350 (D.C. Cir. 2006) ............................................................................................... 21 Porter v. Cal. Dep’t of Corr., 419 F.3d 885 (9th Cir. 2005) ................................................................................................... 15 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 7 of 50 - v - Purepac Pharm. Co v. TorPharm, Inc., 354 F.3d 877 (D.C. Cir. 2004) ................................................................................................. 24 Qi-Zhuo v. Meissiner, 70 F.3d 136 (D.C. Cir. 1995) ................................................................................................... 34 SEC v. Chenery Corp., 332 U.S. 194 (1947) ........................................................................................................... 38, 39 Sharma v. District of Columbia, 791 F.Supp. 2d 207 (D.D.C. 2011) .......................................................................................... 15 Tcherepnin v. Knight, 389 U.S. 332 (1967) ................................................................................................................. 16 TRW Inc. v. Andrews, 534 U.S. 19 (2001) ................................................................................................................... 34 Williams Gas Processing-Gulf Coast Co. L.P. v. FERC, 475 F.3d 319 (D.C. Cir. 2006) ................................................................................................. 25 Statutory Authorities 21 U.S.C. § 351 ............................................................................................................................. 15 21 U.S.C. § 355(b)(2)(A). ............................................................................................................... 3 21 U.S.C. § 355(j) ........................................................................................................................... 3 21 U.S.C. § 355(j)(5)(D). .............................................................................................................. 4 21 U.S.C. § 355(j) (2)(vii) .............................................................................................................. 3 21 U.S.C. § 355(j)(5)(B)(iv) ........................................................................................................... 4 21 U.S.C. § 355(j)(5)(D)(i)(IV) ............................................................................................. passim 21 U.S.C. § 355(j)(7)(A)(i) ............................................................................................................. 3 21 U.S.C. § 355(j)(7)(A)(iii) ........................................................................................................... 3 35 U.S.C. § 271(e)(2)(A) ................................................................................................................ 3 Pub. L. No. 98-417, 98 Stat. 1585, amended by Medicare Prescription Drug, Improvement and Modernization Act of 2003, Pub. L. No. 108-173, 117 Stat. 2066 ............... 3 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 8 of 50 - vi - Regulations 21 CFR 314.101(b)(1) ................................................................................................................... 33 Additional Authorities T. Edward Damer, Attacking Faulty Reasoning (2012) ............................................................... 23 Dan B. Dobbs et al., The Law of Torts § 189 (2d ed. 2011) ......................................................... 26 Letter from FDA to Anchen Pharmaceuticals, Inc. (March 13, 2013), http:// www.accessdata.fda.gov/drugsatfda_docs/appletter/2013/ 091476Orig1s000ltr.pdf ........................................................................................................... 19 Letter from FDA to Nycomed U.S., Inc. (Feb. 25, 2010), http://www.accessdata.fda.gov/ drugsatfda_docs/appletter/2010/078548s000ltr.pdf.................................................................. 38 Letter from FDA to Sandoz, Inc. (Sept. 20, 2011), https://www.regulations.gov/ contentStreamer?documentId=FDA-2010-P-0632-0017&attachmentNumber=1 . .................. 38 Letter from FDA to Synthon Pharmaceuticals, Inc. (Nov. 26, 2010), http:// www.accessdata.fda.gov/drugsatfda_docs/appletter/2010/090229s000ltr.pdf ......................... 38 Letter from FDA to Teva Pharmaceuticals USA (March 30, 2012), http://www.accessdata.fda.gov/drugsatfda_docs/appletter/2012/077159s000ltr.pdf . ............ 38 Patent Challenge Provisions of the Medicare Reform Bill, S. Cong. Rec. S16105 (December 8, 2003) .................................................................................................................. 17 Restatement (Third) of Torts § 27 (Am. Law Inst. 2010) ............................................................. 26 U.S. Food & Drug Admin., Guidance for Industry – 180-Day Exclusivity: Questions and Answers (Jan. 2017), http://www.fda.gov/downloads/Drugs/Guidance ComplianceRegulatoryInformation/Guidances/UCM536725.pdf ..................................... passim U.S. Food & Drug Admin., Guidance for Industry: ANDAs: Stability Testing of Drug Substances and Products (June 2013), https://www.fda.gov/downloads/ drugs/guidancecomplianceregulatoryinformation/guidances/ucm320590.pdf ........................... 7 U.S. Food & Drug Admin., Guidance for Industry: ANDAs: Stability Testing of Drug Substances and Products: Questions and Answers (May 2014), http://www.fda.gov/ downloads/drugs//guidances/ucm366082.pdf ............................................................................. 7 U.S. Food & Drug Admin., Guidance for Industry: Q1A(R2) Stability Testing of New Drug Substances and Products (Nov. 2003), http://www.fda.gov/downloads/drugs// guidances/ucm073369.pdf .......................................................................................................... 7 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 9 of 50 - vii - U.S. Food & Drug Admin., Manual of Policies and Procedures 5225.1, Guidance on the Packaging of Test Batches (Sept. 2012) ..................................................................................... 7 Wikipedia, Denying the antecedent, https://en.wikipedia.org/wiki/Denying_the_antecedent (last visited Mar. 29, 2017) ............................................................................................................... 23 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 10 of 50 1 INTRODUCTION This case involves Amneal’s challenge to FDA’s (or “the Agency’s”) unlawful, arbitrary, and capricious decision that Amneal forfeited 180-day exclusivity to its generic version of Namenda XR® (Memantine Hcl extended release capsules). Congress amended the Federal Food, Drug, and Cosmetic Act (the “FDCA”) to incentivize generic drug competition by providing 180 days of exclusivity to any first applicant submitting an Abbreviated New Drug Application (“ANDA”) challenging the patents of a non- generic drug. Congress created this exclusivity to reward ANDA applicants for challenging possibly invalid or non-infringed patents and assuming the high risk and cost of patent litigation. A first generic applicant can “forfeit,” or lose, its exclusivity if it “fails to obtain tentative approval. . . within 30 months after the date on which the application is filed, unless the failure is caused by a change in or review of the requirements for approval of the application imposed after the date on which the application is filed.” 21 U.S.C. § 355(j)(5)(D)(i)(IV). FDA erroneously concluded that such a forfeiture occurred here. Amneal contends that two changes in or reviews of the requirements for approval of its ANDA caused Amneal’s failure to obtain timely tentative approval. First, Amneal’s ANDA was delayed for 14 months after FDA demanded that Amneal produce data from commercial-scale production batches, rather than from the far smaller pilot-scale batches that FDA’s guidance states are sufficient. Second, Amneal’s application experienced an eight-month delay resulting from FDA’s review of whether it would receive Amneal’s ANDA. With respect to FDA’s scale-up demand, the Agency first concluded that it did not cause the failure to obtain tentative approval based on a single fact: it was no longer an active issue as of the forfeiture date. FDA addressed no other relevant facts such as the nature of the change, its consistency with prior requirements, its onerousness, or how much it disrupted review. In the Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 11 of 50 - 2 - alternative, FDA concluded that its scale-up demand was not a change in or review of approval requirements because there had been no change to statutory requirements or to FDA policy. FDA’s decision is arbitrary, capricious, and contrary to law. FDA’s rigid, overly simplistic, single-factor causation analysis falls far short of what the FDCA and fundamental principles of administrative law require. As this Court has already held, “caused by” under the FDCA unambiguously requires a fact-intensive analysis, and cannot be reduced to a single fact based on timing of a change. Even if the statute were ambiguous – which it is not – FDA’s causation analysis is arbitrary and capricious. FDA has provided no justification for distilling causation – normally a complex factual inquiry – into a simple, dispositive, yes-or-no question, based on the existence of a single fact, on a single day, in a 30-month period. Moreover, FDA’s single-factor test conflicts with FDA precedent and is self-contradictory. And FDA’s analysis is not the product of reasoned decision making: it disregards myriad facts relevant to causation and ignores how various events interacted to thwart tentative approval. With respect to the delay in receiving the ANDA, FDA’s failure to consider that delay (either alone or in combination with the commercial scale-up demand) is an independent basis for vacating FDA’s decision. FDA’s conclusion in the alternative that its demand for commercial-scale data was not a “change in or review of the requirements for approval of the application,” because it did not involve a statutory or policy change, also fails. FDA’s focus on statutory or policy change ignores the phrase “of the application” in the statute. FDA’s analysis is also arbitrary and capricious because it relies upon a secret, undisclosed policy and departs from FDA precedent without explanation. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 12 of 50 - 3 - STATEMENT OF FACTS A. Legal Background The Drug Price Competition and Patent Term Restoration Act of 1984 (the “Hatch- Waxman Amendments”) amended the FDCA to create an abbreviated approval pathway to encourage generic drug competition. Pub. L. No. 98-417, 98 Stat. 1585, amended by Medicare Prescription Drug, Improvement and Modernization Act of 2003, Pub. L. No. 108-173, 117 Stat. 2066. Prior to the enactment of the Hatch-Waxman Amendments, burdensome regulations often stifled generic market entry, requiring submission of a full New Drug Application (“NDA”). This hampered access to cost-effective generic medicines. The Hatch-Waxman Amendments codified the ANDA pathway for expedited approval of generic drugs. See 21 U.S.C. § 355(j). Upon approval of an NDA for a non-generic drug, generally designated a reference listed drug or “RLD,” FDA lists the drug in what is commonly referred to as the “Orange Book.” 21 U.S.C. § 355(j)(7)(A)(i). The Orange Book also lists patents identified by the NDA holder that claim the reference drug or a method of using it (“listed patent”). 21 U.S.C. § 355(j)(7)(A)(iii). See also 21 U.S.C. § 355(b)(2)(A). Congress required each ANDA applicant to include one of four patent certifications for each listed patent. 21 U.S.C. § 355(j)(2)(vii). One of these is the “Paragraph IV” certification that the listed patents are invalid, unenforceable, or will not be infringed by the proposed generic drug. Id. Filing an ANDA with a Paragraph IV certification is, by statute, a technical act of infringement and often provokes a patent infringement lawsuit. 35 U.S.C. § 271(e)(2)(A). Congress created incentives to reward generic manufacturers who challenge listed patents and assume the risk and expense of litigation. The FDCA provides the “first applicant” that “submitted an [ANDA] containing [a Paragraph IV] certification” with a 180-day exclusivity Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 13 of 50 - 4 - period against generic competitors during which FDA cannot approve competing ANDAs for the same reference drug. 21 U.S.C. § 355(j)(5)(B)(iv). A first applicant can “forfeit” (or lose) 180-day exclusivity if certain events occur or fail to occur. Id. at § 355(j)(5)(D). Relevant here, a first applicant can forfeit its 180-day exclusivity if it “fails to obtain tentative approval of the application within 30 months after the date on which the application is filed, unless the failure is caused by a change in or a review of the requirements for approval of the application imposed after the date on which the application is filed.” 21 U.S.C. § 355(j)(5)(D)(i)(IV) (emphases added). The causation analysis under the FDCA is flexible and fact-intensive. FDA should not impose forfeiture if, as a factual matter, there is any causal connection between the failure to obtain tentative approval and a change in or review of approval requirements for the application. In fact, FDA decisions and court precedent have made clear “but for causation is not required” in order to show causation under the statute. Mylan Labs. Ltd. v. FDA, 910 F. Supp. 2d 299, 311 n.7 (D.D.C. 2012). FDA has explained that “this interpretation of the statute best effectuates the policy underlying the exception. It does not penalize applicants for FDA’s reviews of, or changes in, approval requirements imposed on applicants after their ANDAs are filed that are a cause of the failure to obtain. . . approval within 30 months. This interpretation also continues to incentivize ANDA applicants to challenge patents by preserving in many instances the opportunity to obtain 180-day exclusivity.”1 1 U.S. Food & Drug Admin., Guidance for Industry – 180-Day Exclusivity: Questions and Answers” at 22 (Jan. 2017), http://www.fda.gov/downloads/Drugs/GuidanceCompliance RegulatoryInformation/Guidances/UCM536725.pdf (hereinafter “180-Day Exclusivity Guidance”). Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 14 of 50 - 5 - As long as “one of the causes of failure to obtain tentative or final approval by the 30- month forfeiture date was a change in or review of the requirements for approval imposed after the application was filed, an applicant will not forfeit eligibility for exclusivity notwithstanding that there may have been other causes for failure to obtain tentative approval. . . .” Id. As this Court has explained: “[A]n applicant need only show that acceptability of one aspect of the ANDA (e.g., chemistry) was delayed due to a change in or review of the requirements for approval, irrespective of what other elements may also have been outstanding at the 30-month date. Mylan, 910 F. Supp. 2d at 302. B. Factual Background 1. The Eight-Month Delay in FDA’s Receipt and Commencement of Review of Amneal’s ANDA On June 10, 2013, Amneal submitted ANDA No. 205825 seeking approval to market a generic version of Namenda XR in all four approved strengths. FDA 0051-57. Namenda XR (Memantine HCl extended release capsules) contains Memantine Hydrochloride and is indicated for the treatment of moderate to severe dementia of the Alzheimer’s type. FDA 0978. Amneal’s ANDA also included Paragraph IV certifications to the unexpired Orange-Book-listed patents for Namenda XR. FDA 0109. Amneal’s application was the first-submitted ANDA that included at least one Paragraph IV certification to at least one of the listed patents for Namenda XR. FDA 1041, FDA 1043-44. Accordingly, Amneal was eligible for 180-day exclusivity. On November 5, 2013, nearly five months after the application was submitted, FDA sent a letter to Amneal that it was refusing to receive the ANDA because Amneal allegedly “failed to produce accelerated stability data which encompasses at least 84 days in the accelerated stability chamber for the 30 count and the 500 counts for all strengths.” FDA 0063-65. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 15 of 50 - 6 - Almost immediately after receiving FDA’s letter, Amneal told FDA that Amneal had, in fact, provided the Agency with all of the requisite data. FDA 0072-73. The ANDA, FDA was told, simply contained a typographical error in a summary table of the data that incorrectly listed the stability testing initiation date as November 27, 2012, instead of November 21, 2012, when the testing had actually been started. FDA 0072-73. The data within the application, however, were complete. FDA 0072-73. At FDA’s request, Amneal emailed FDA on November 11, 2013 to confirm that Amneal initiated stability testing on November 21, 2012. FDA 0072-73. In addition, even though the requisite data were in the ANDA as originally filed, Amneal also provided documentation requested by FDA. FDA 0076-93. Following this, Amneal repeatedly contacted FDA for updates, including on November 18 and again on December 4. FDA 0070-75. On December 4, at FDA’s request, Amneal resubmitted the information previously provided on November 11. FDA 0076-93. Two months later FDA finally rescinded its refusal to receive the ANDA. FDA 0109-12. On February 18, 2014, FDA informed Amneal that “the Office of Generic Drugs has decided to rescind [FDA’s] ‘Refuse to Receive’ letter dated November 5, 2013” and deemed Amneal's ANDA “ACCEPTABLE FOR FILING” as of the June 10, 2013 date of submission. FDA 0109- 12. It was only then, more than eight months after the submission of Amneal’s original and complete ANDA, that FDA finally received and began substantive review of the ANDA. 2. FDA’s Demand for Full-Scale Commercial Batch Data Results in an Additional 14-Month Delay FDA’s longstanding guidance to ANDA applicants has been to submit ANDAs with pilot-size batches and corresponding production data as part of the ANDA’s Chemistry, Manufacturing, and Controls (“CMC”) information. For example, FDA’s refuse-to- Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 16 of 50 - 7 - receive guidance outlines “deficiencies that may cause FDA to refuse to receive an ANDA. . . [because they are] not sufficiently complete to permit a substantive review.” “ANDA Submissions – Refuse-to-Receive Standards Guidance for Industry,” available at FDA 0243 (“RTR Guidance”). Citing prior guidance,2 the RTR Guidance states that “The applicant should provide three pilot-scale batches or two pilot-scale and one small-scale batch. . . .” FDA 0252. As this guidance shows, while ANDA sponsors may seek commercial-scale production on the order of 1,000,000 units, FDA permits and encourages applicants to file such ANDAs using data from pilot-scale batches a tenth of that size, or 100,000 units. Prior to the ANDA at issue in this case, Amneal – across a portfolio of nearly 100 approved ANDAs – had never previously been required to produce a commercial-scale batch as a condition of approval. See FDA 0252. See also FDA 0749-50. FDA concedes that its public guidance and general policy “do not describe” the circumstances under which FDA will require more than pilot-scale batches. FDA 1056. The Agency also concurs that it is not “surprising 2 The RTR guidance cites U.S. Food & Drug Admin., Guidance for Industry: ANDAs: Stability Testing of Drug Substances and Products: Questions and Answers at 3, 8 (May 2014) (emphasis original) (stating that “the applicant should submit data from three pilot scale batches or should submit data from two pilot scale batches and one small scale batch. This applies to all dosage forms” and “two of the three batches should be of at least 10 percent of the proposed [commercial] production batch or 100,000 finished dosage units, whichever is greater (i.e., pilot scale batches). The third batch can be smaller. . .”), http://www.fda.gov/downloads/drugs// guidances/ucm366082.pdf; U.S. Food & Drug Admin., Guidance for Industry: ANDAs: Stability Testing of Drug Substances and Products at 2 (June 2013) (same), http://www.fda.gov/ downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM320590.pdf; U.S. Food & Drug Admin., Guidance for Industry: Q1A(R2) Stability Testing of New Drug Substances and Products at 8 (November 2003) (“Two of the three batches should be at least pilot scale batches, and the third one can be smaller if justified.”), http://www.fda.gov/ downloads/drugs//guidances/ucm073369.pdf. See also U.S. Food & Drug Admin., Manual of Policies and Procedures 5225.1, Guidance on the Packaging of Test Batches at 5 (Sept. 2012) (“The minimum amount to be packaged is 100,000 units.”). Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 17 of 50 - 8 - that Amneal believes it has not been previously asked to provide a commercial-scale test batch to support approval.” FDA 1056. Consistent with FDA practice and guidance, Amneal’s ANDA submission included data from pilot-size batches of well over 100,000 units. See FDA 0488. Amneal’s proposed one million unit commercial-scale batch size, as described in the ANDA submission, was over six times larger than its pilot-size batches. See FDA 0488. On September 10, 2014, the Agency issued a complete response letter informing Amneal that FDA “cannot approve this ANDA in its present form.” FDA 0231. While most of the deficiencies were minor, FDA’s letter also raised a series of questions regarding the composition of Amneal's product. To address these questions, FDA’s letter informed Amneal that the company would have to provide certain information. Most importantly, the letter informed Amneal that FDA had determined the application “requires the review of the resulting data from the manufacture of a commercial size lot [i.e., batch] before application approval. In this regard, please submit all the required CMC information for the production of a 7 mg and 28 mg Memantine HCl lot of the size intended for commercial distribution.” FDA 0233. Amneal immediately and diligently began work on a submission responsive to this new requirement for commercial-scale information. But this was not a trivial undertaking; providing commercial-size batch data was a necessarily time-consuming and onerous process. To comply, Amneal had to order, and its suppliers had to produce, hundreds of kilograms of the bulk active pharmaceutical ingredient (“API”) (along with the required excipients and other components), and Amneal had to process that API. FDA 0301-02, FDA 0649-50. This process alone took several months to complete, which occurred in late January and early February 2015. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 18 of 50 - 9 - FDA 0301-02, FDA 0649-50. The process of producing an intermediate,3 and producing and packaging two distinct commercial-size lots of finished product took longer still: until mid-April 2015. FDA 0301-02, FDA 0649-50. Completing the analysis for FDA’s newly-required commercial-size lots took yet another month. See FDA 0301-02, FDA 0649-50. On May 12, 2015, roughly eight months after FDA first informed Amneal that the company’s application would require the production and analysis of two complete commercial- size lots as a condition of approval, Amneal submitted an amendment to its ANDA with full CMC data and analysis from the scaled-up lots, including over half a gigabyte of supporting data. FDA 0265-0389. Amneal's submission also answered each of the related questions that prompted FDA to require the manufacture and analysis of commercial-size lots. Id. On August 25, 2015, FDA sent Amneal a series of follow-on questions related to the scaled-up data that again focused on Amneal’s proposed composition of its product based on characteristics of the reference listed drug. FDA 0418-19. Amneal responded on September 15, 2015 by (among other things) providing further information regarding the reference listed drug and the production of commercial-size lots. FDA 0433-68. In mid-October 2015, while FDA was still considering Amneal’s commercial-scale data, FDA issued an Import Alert against Amneal’s proposed API manufacturer, Megafine Pharma (P) Limited (“Megafine”). FDA’s alert effectively necessitated a change in API supplier by Amneal. FDA 0486, FDA 0527, FDA 0668. 3 Amneal’s product is produced by blending extended and immediate release API pellets to create an intermediate blend, and then packaging that blend into capsules for the final commercial product. See FDA 0302. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 19 of 50 - 10 - Shortly thereafter, on October 22, 2015, FDA asked Amneal to “scale down the proposed commercial batch size” because of the “complexities” introduced by the blending step of Amneal’s manufacturing process. FDA 0483. On November 9, 2015, fourteen months after FDA’s original demand for commercial-scale information, Amneal responded and agreed to FDA’s request. Amneal stated: As per the Agency’s recommendation, Amneal will scale down the proposed commercial batch size from 1,000,000 capsules to 150,000 capsules (same as ANDA submission batch size) for all strengths of Memantine Hydrochloride Extended Release Capsules (i.e. 7 mg, 14 mg, 21 mg and 28 mg). FDA 0488. Amneal also proposed a new API supplier to replace Megafine. FDA 0486, 0490. On November 16, 2015, Amneal submitted an amendment to its ANDA that both changed the API supplier and scaled down its proposed commercial batch size from 1,000,000 capsules to 150,000 capsules. FDA 0526-28. In support of both changes, Amneal submitted data from new pilot-scale batches (which also now served as commercial-scale batches). See FDA 0526-32. Amneal’s submission had to undergo chemistry review evaluating both changes less than a month before the forfeiture date. Id. On December 7, 2015 – just three days before the thirty-month forfeiture date on December 10 – FDA, by letter, “acknowledge[d] that [Amneal] will not scale up and batch sizes will remain as per the exhibit [i.e. pilot] lot size.” FDA 0714. The letter also raised deficiencies related to Amneal’s November 16, 2015 submission. FDA 0713-16. The deficiencies related to Amneal’s proposed change in API supplier; however, some deficiencies also involved questions relating to Amneal’s overall process, including regarding the blending used to create the intermediate and regarding characteristics of the encapsulated product. For example, deficiencies 4 and 5 included issues relating to the blend of Amneal’s intermediate, deficiency 6 involved questions regarding the content uniformity of Amneal’s capsules, and the Agency also Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 20 of 50 - 11 - recommended that Amneal conduct specific dissolution testing. Id. The chemistry issues raised in this letter were still outstanding and being addressed on the forfeiture date, three days later. See FDA 0717-23. 3. FDA’s 180-Day Exclusivity Forfeiture Ruling In December 2015, counsel for Amneal wrote FDA seeking confirmation that Amneal would not forfeit 180-day exclusivity. FDA 0640-59. FDA responded by letter on September 28, 2016 (the “Letter Decision”). FDA 1041-58. The Agency began its response by recognizing that the inquiry mandated by the FDCA is whether the “failure to obtain a tentative approval or approval at the 30-month date is caused by [a] change in or review of approval requirements.” FDA 1042 (emphasis original). FDA also correctly noted that under the Agency’s precedent, “‘but for’ causation is not required.” Id. The Agency then recognized that it has made available a favorable presumption to ANDA filers, stating that “FDA generally will presume that the failure to obtain tentative approval or approval was caused by a change in or review of approval requirements if, at the 30- month date, the evidence demonstrates that the sponsor was actively addressing the change in or review of approval requirements (or FDA was considering such efforts), and these activities precluded tentative approval (or approval) at that time. [If the evidence does not so demonstrate,] FDA generally does not presume that the failure was caused by a change or review of approval requirements.” Id. (emphases added). FDA observed that, as of FDA’s calculated forfeiture date on December 10, 2015, Amneal had agreed to FDA’s request that Amneal scale down its proposed commercial manufacturing to the pilot-scale batch size. FDA reasoned that because FDA’s scale-up demand was “resolved before the forfeiture date of December 10, 2015. . . it could not preclude tentative Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 21 of 50 - 12 - approval at that time.” FDA 1054-55. This constituted the totality of the Agency’s factual analysis and reasoning with respect to the existence of causation. In particular, FDA did not analyze the factual record before it with respect to numerous facts relevant to causation. FDA did not analyze the length of the 14-month delay, its proximity to the forfeiture date, how disruptive the change was, how much effort was required for compliance, how much effort was required by FDA to review the change, how it may or may not have affected the timing of approval, or how the change interacted with other delays, such as the delay in receipt, to prevent timely tentative approval. FDA also concluded, in the alternative, that “the request to scale up and complete a commercial-size batch before approval was [not] a change in or review of the requirements for approval.” FDA 1055. FDA reasoned that neither a “statutory requirement” nor FDA’s “specific requirements for certain complex products” had changed. Id. FDA claimed that it had acted pursuant to an FDA policy requiring commercial-scale information for “certain” products. The Agency failed, however, to identify any instance, available to Amneal or to the public, in which FDA had articulated such a policy. The Agency stated it could not provide that information as it was confidential. FDA 1055 n.43. See also FDA 1032 n.59 (redacting information from administrative record). Thus, the Letter Decision conceded that “FDA’s general policies on manufacturing test batches do not describe [the circumstances under which FDA might require commercial-scale data], nor is it surprising that Amneal believes it has not been previously asked to provide a commercial-scale batch to support approval.” FDA 1056. 4. FDA’s Approval of Amneal’s ANDA and of Third-Party ANDAs FDA completed chemistry review of Amneal’s ANDA in September 2016. FDA 0971, 1011-1012. Amneal’s ANDA was granted tentative approval on September 28, 2016 and final approval on October 12, 2016. FDA 1037-1040, FDA 1059-1063. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 22 of 50 - 13 - As a consequence of its forfeiture ruling, FDA granted final approval to several third- party ANDA applications for generic versions of Namenda XR® at all four strengths, including ANDA 206028, sponsored by Defendant-Intervenor Lupin Limited (“Lupin”). Absent forfeiture, Amneal qualifies for 180-day exclusivity and FDA would not have been permitted by statute to grant these approvals. ARGUMENT I. FDA’s Rigid, Single-Factor Causation Analysis Fails to Satisfy Chevron and is not the Product of Reasoned Decision Making An ANDA sponsor forfeits its 180-day exclusivity if it “fails to obtain tentative approval of the application within 30 months after the date on which the application is filed unless the failure is caused by a change in or a review of the requirements for approval of the application imposed after the date on which the application is filed.” 21 U.S.C. §355(j)(5)(D)(i)(IV) (emphasis added). FDA found no causation here based on a causation test that turns on a single fact: whether Amneal was still addressing the change in approval requirements on the forfeiture date. Specifically, FDA reasoned that its “request that Amneal manufacture and submit information on commercial size lots for the 7 mg and 28 mg strengths was. . . resolved before the forfeiture date of December 10, 2015. Therefore, it could not preclude tentative approval at that time.” FDA 1055 (emphasis added).4 This overly simplistic, rigid test – that exalts the circumstances existing at one point in time above all other facts – is foreclosed by the FDCA under traditional canons of construction, 4 FDA applied a similar rigid and simplistic analysis with respect to a labeling change to Amneal’s ANDA, which FDA conceded constituted a change in requirements for approval. FDA’s analysis with respect to the labeling revision turned exclusively on the fact that the proposed labeling was found acceptable just one month “[b]efore the 30-month date of December 10, 2015.” FDA 1023-24. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 23 of 50 - 14 - as well as by this Court’s and FDA precedent. Even if FDA had applied the analysis required by the statute, its decision here was not the product of reasoned decision making. A. FDA’s Interpretation of “Caused By” as Creating a Rigid, Irrebutable, Single-Factor Causation Test Fails Chevron Step One The FDCA addresses causation in clear and plain terms: a failure to obtain timely tentative approval is excused if “the failure is caused by a change in or a review of the requirements for approval of the application imposed after the date on which the application is filed.” 21 U.S.C. § 355(j)(5)(D)(i)(IV) (emphasis added). This Court reviews FDA’s interpretation under Chevron, U.S.A., Inc. v. NRDC, 467 U.S. 837 (1984). Under Chevron, the Court first asks whether Congress has spoken to the precise question at issue. If the intent of Congress is clear, the Court must give effect to the unambiguously expressed intent. Chevron, 467 U.S. at 842-43. If the statute is silent or ambiguous, the Court must ask whether the agency’s position rests on a permissible construction of the statute. Mylan Labs., Inc. v. Thompson, 389 F.3d 1272, 1280 (D.C. Cir. 2004) citing Chevron, 467 U.S. at 842-43. Ultimately, the question under step two is “whether the FDA’s decision represents the result of a reasonable exercise of its authority.” Amarin Pharms. Ir. Ltd. v. F.D.A., 106 F. Supp. 3d 196, 206 (D.D.C. 2015). As this Court has explained, “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.” See id. at 208. “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.” Id. There is no ambiguity in the statute and, more importantly, there is absolutely no indication that Congress intended causation – one of the most fact-intensive concepts in law – to Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 24 of 50 - 15 - mean a single, dispositive, yes-or-no question, based on the existence or nonexistence of a single fact, on a single day, and to the exclusion of all other facts. Cf. Porter v. Cal. Dep’t of Corr., 419 F.3d 885, 895 (9th Cir. 2005) (reducing causation to “a mechanically applied criterion” in the form of a temporal test “would be unrealistically simplistic”); Kachmar v. SunGard Data Sys., Inc., 109 F.3d 173, 178 (3rd Cir. 1997) (“[T]he absence of immediacy between the cause and effect does not disprove causation.”); Sharma v. District of Columbia, 791 F. Supp. 2d 207, 219-20 (D.D.C. 2011) (“[A] number of factors, including but not limited to temporal proximity, are relevant to evaluating causation.”). This Court has already held as much. In Mylan Laboratories Ltd. v. FDA, this Court held that the causation analysis under precisely the forfeiture provision at issue here requires more than a bright-line test that merely considers the timing of a change in approval requirements. In that case, this Court upheld FDA’s decision to grant exclusivity when a change in the USP monograph5 for the drug caused an ANDA applicant to miss the 30-month forfeiture deadline. 910 F. Supp. 2d at 307. This Court refused to reverse the Agency’s grant of exclusivity because, contrary to the plaintiff’s arguments, the Agency in evaluating forfeiture had not applied “a bright-line test” based merely on the timing of the change. Id. at 307. And this Court was emphatic in its rejection of such a simplistic causation analysis, stating that “[t]he Court will not give the forfeiture exception [such an] interpretation.” Id. at 309. “Causation in this context,” this Court wrote, “requires a showing of something more than just the fact and timing of” the change. Id. FDA’s analysis satisfied the statute because the Agency had conducted a fact- 5 The USP, or United States Pharmacopoeia, is an official compendium of drug information recognized by FDA. An ANDA applicant seeking approval of drug for which there is a USP monograph must generally show that the proposed product’s strength, quality and purity meet the standards in the monograph. See 21 U.S.C. § 351. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 25 of 50 - 16 - specific, multi-factor analysis that considered the timing of the change as well as myriad other facts equally important to causation under the statute. Id. at 306, 309. Indeed, in Mylan, FDA specifically argued that the causation analysis under the statute was “a very fact-specific question” involving “a number of potential factors.” Id. at 306. Specifically, the Agency enumerated no fewer than seven non-exclusive factors relevant to the analysis. These “included” but were “not limited to,” 1) the finality of the change; 2) “the timing of [the change] in relation to a particular 30-month forfeiture date;” 3) whether compliance was required by FDA; 4) whether compliance was required by the statute; 5) “the consistency of the [change] with pre-existing approval requirements;” 6) “the nature and timing of the sponsor’s efforts to comply;” and 7) “the nature and timing of FDA’s review of such efforts.” Id. at 306. A multi-factor analysis, like that conducted by FDA in Mylan and approved by this Court, comports with the “text, structure, and purpose” of the Hatch-Waxman Amendments. See Amarin, 106 F. Supp. 3d at 208-09. Forfeiture of exclusivity is an exception to the statute’s default grant of 180-day exclusivity. See Mylan, 910 F. Supp. 2d at 307 (describing forfeiture as an “exception” to exclusivity). Under traditional canons of construction, exceptions to a statute must be narrowly construed in view of the statute’s overall goal. See A.H. Phillips, Inc. v. Walling, 324 U.S. 490, 493 (1945) (“Any exemption from such. . . remedial legislation. . . must therefore be narrowly construed, giving due regard to the plain meaning of statutory language and the intent of Congress. To extend an exemption to other than those plainly and unmistakably within its terms and spirit is to abuse the interpretative process and to frustrate the announced will of the people.”); Tcherepnin v. Knight, 389 U.S. 332, 336 (1967) (“Remedial legislation should be construed broadly to effectuate its purposes.”); Laffey v. Nw. Airlines, Inc., 567 F.2d 429, 462 (D.C. Cir. 1976) (Remedial statutes should be liberally construed in favor of Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 26 of 50 - 17 - beneficiaries). Similarly, statutes that impose forfeitures, or which are penal in nature, generally must be construed narrowly. Am. Mar. Ass’n v. Blumenthal, 590 F.2d 1156, 1165 (D.C. Cir. 1978) (“[S]tatutes imposing a forfeiture. . . are to be construed strictly.”); see also FCC v. Am. Broad. Co., 347 U.S. 284, 296 (1954) (“[P]enal statutes are to be construed strictly.”). Congress created the 180-day exclusivity period and its exception, the forfeiture provision, as part of an overall scheme to promote generic competition in the prescription drug marketplace. Mylan, 910 F. Supp. 2d at 311. Forfeiture was adopted in response to concerns that ANDA applicants were purposely delaying entry into the market, so called “exclusivity parking” where a first filer and brand company would agree to a delayed market entry. See e.g. Patent Challenge Provisions of the Medicare Reform Bill, S. Cong. Rec. S16105 (December 8, 2003). By limiting the reach of the forfeiture exception to exclude failures to obtain tentative approval caused by changes in approval requirements, Congress avoided punishing applicants for post- filing changes outside of their control. This preserved the statute’s overall goal of encouraging generic competition. FDA’s crabbed, rigid, single-factor application of causation in Amneal’s case frustrates that purpose. FDA has recognized that the purpose and structure of the Hatch-Waxman Amendments require a broad, flexible application of causation. Just months ago, in January 2017, FDA published guidance that “describes the Agency’s current thinking” on 180-day exclusivity. 180- Day Exclusivity Guidance, supra note 1.6 In that guidance, FDA acknowledges that the statute 6 Although the 180-Day Exclusivity Guidance was issued after FDA’s decision in this case, it explains that it is intended to summarize FDA’s past precedent in an accessible and transparent manner. See 180-Day Exclusivity Guidance at 1. The need for the guidance was because “as a general matter, [FDA] has implemented these statutory provisions within the context of application-specific decisions”. Id. However, only “some FDA decisions have been made Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 27 of 50 - 18 - does not even require “but-for” causation. All the record need show is that “at least one aspect of the ANDA (e.g., bioequivalence) was delayed. . . irrespective of what other review issues may have been outstanding at the 30-month date.” Id. at 22. Accord Mylan, 910 F. Supp. 2d at 302. This analytical flexibility, FDA has explained, is compelled by the statute’s policy and structure: “FDA has determined that this interpretation of the statute best effectuates the policy underlying the exception. It does not penalize applicants for FDA’s reviews of, or changes in, approval requirements imposed on applicants after their ANDAs are filed. . . [and it] continues to incentivize ANDA applicants to challenge patents by preserving in many instances the opportunity to obtain 180-day exclusivity.” 180-Day Exclusivity Guidance, supra note 1, at 22. FDA’s rigid application of causation in Amneal’s case, by contrast, penalizes applicants, discourages investment, and slows generic entry into the market: the opposite of what Congress intended. Instead of promoting generic competition, FDA has created a nonsensical standard that leads to absurd results. Cf. Mova Pharm. Corp. v. Shalala, 140 F.3d 1060, 1068 (D.C. Cir. 1998) (it is a “long-standing rule that a statute should not be construed to produce an absurd result”). FDA’s single-factor causation test perversely incentivizes sponsors to stretch the resolution of a change of requirements past the forfeiture date if at all possible.7 More troublingly, FDA’s single-factor causation test arbitrarily punishes applicants who have onerous changes in approval requirements imposed early in the process. By contrast, it provides relief to applicants who face far less burdensome changes that fortuitously span the forfeiture date based publicly available. . .” Id. Thus, the recent guidance is intended to “enhance transparency” by disclosing how FDA has developed the law in both public and nonpublic determinations. Id. 7 Although FDA requires an applicant to be “actively addressing” the change of review of approval requirements, even a few days can matter, as will be shown below. FDA incentivizes applicants to be “active,” but not so active that they resolve issues before the forfeiture date. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 28 of 50 - 19 - on mere happenstance. Congress could not have intended such absurd and arbitrary results. See Mylan, 910 F. Supp. 2d at 31. This absurdity is not hypothetical, but has played out in just this way to another ANDA sponsor subjected to FDA’s single-factor causation test. Anchen Pharmaceuticals was the first filer on two separate ANDAs for Fluvoxamine Maleate Extended-release capsules, with one application for a 100 mg dose and a second for the 150 mg dose. Letter from FDA to Anchen Pharmaceuticals, Inc. at 2 and n.4 (March 13, 2013), http://www.accessdata.fda.gov/ drugsatfda_docs/appletter/2013/091476Orig1s000ltr.pdf. The 150 mg dose had a forfeiture date of October 13 and the 100 mg dose had a forfeiture date of October 20, one week later. Id. Anchen was required to respond to a change in labeling requirements on both applications, which was resolved on October 18. Id. This was right in between the two forfeiture dates. That meant that, on 150 mg forfeiture date, the labeling issue was still being addressed but, on the 100 mg forfeiture date (just a week later), it had already been resolved. As a result, FDA found no forfeiture on the 150 mg dose but imposed a forfeiture on the 100 mg dose. Thus, in Anchen’s case, the sole, exclusive, and dispositive fact determining the outcome of the causation analysis was, quite literally, the day of the week on which an official at FDA sent a letter resolving the labeling issue. It is difficult to imagine a causation test as unmoored from the statute’s text and purpose, and with more arbitrary and capricious results. B. FDA’s Single-Factor Causation Test Fails Chevron Step Two Because It Lacks Any Justification, Conflicts with Prior Precedent without Explanation, and is Self-Contradictory Even if the Court concludes that the word “caused” in the FDCA is ambiguous, FDA’s irrebuttable single-factor causation rule is an impermissible interpretation of the statute and fails Chevron’s second step. See Amarin, 106 F. Supp. 3d at 206 (“Chevron’s second step asks Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 29 of 50 - 20 - whether an agency’s interpretation is arbitrary or capricious in substance”). Indeed, it is “apparent that the decision is both procedurally and substantively flawed.” Id. at 218. First, FDA has failed to provide any justification supporting the rigid, single-factor causation analysis it has applied. FDA’s apparent interpretation of causation “must be upheld, if at all, on the basis articulated by the agency itself.” Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 50 (1983). But here, FDA “does not offer—or even attempt— any reasoned explanation for how” the causation required by the statute is supportably distilled to a single-factor dispositive test. See Amarin, 106 F. Supp. 3d at 218. FDA “makes no attempt to explain how its approach” to causation in this case “furthers Congress’s purposes or is otherwise a reasonable policy choice.” See id. FDA makes no effort to make sensible the nonsensical. Why should causation be defined by a single, highly circumstantial factor such as the timing of resolution of a change when, in all other circumstances, causation involves a highly contextual, factual analysis? How can FDA justify anomalous and arbitrary results, like the Anchen case, where the entire causation analysis turns on what day of the week a letter was sent? Why should a brief delay that just happens to span the forfeiture date be more significant than a months-long, highly disruptive one that occurs earlier in review? Causation is not so simplistic. Detaining a sprinter at the starting blocks can cause him to fail to run a five-minute mile, just as much as stopping him just before he crosses the finish line. Amneal and the public are entitled, at the very minimum, to a reasoned explanation from FDA justifying such a rigid, anomalous, and unconventional view of causation. But FDA, and the record, are silent on all of these critical questions. If anything, FDA’s guidance undercuts, rather than supports, FDA’s rigid, single-factor causation analysis. 180-Day Exclusivity Guidance, supra note 1, at 22 (statute does not require Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 30 of 50 - 21 - “but for” causation as it effectuates purpose of statute and does not “penalize applicants for FDA’s reviews of, or changes in, approval requirements imposed on applicants after their ANDAs are filed. . . [and it] continues to incentivize ANDA applicants to challenge patents by preserving in many instances the opportunity to obtain 180-day exclusivity.”). Second, FDA’s decision failed to find that “caused by” in the statute was even ambiguous, which renders FDA’s decision fatally flawed from the start. See Peter Pan Bus Lines, Inc. v. Fed. Motor Carrier Safety Admin., 471 F.3d 1350, 1354 (D.C. Cir. 2006) (“Chevron step 2 deference is reserved for those instances when an agency recognizes that the Congress’s intent is not plain from the statute’s face.”). Third, FDA’s single-factor causation test is inconsistent, without explanation, with FDA’s prior precedent. As discussed above, in Mylan FDA asserted that causation was “a very fact-specific question” involving “a number of potential factors.” 910 F. Supp. 2d at 306. Indeed, it enumerated no fewer than seven nonexclusive relevant factors including timing, the nature of the change, its onerousness, its consistency with pre-existing requirements, and more. FDA’s 180-Day Exclusivity Guidance, which summarizes past FDA precedent, likewise, explains that the Agency will consider other factors: “When evaluating causation for purposes of determining forfeiture under this provision, FDA will take into account how close the change. . . occurred to the 30-month date as well as the amount of effort needed to respond to the change.” 180-Day Exclusivity Guidance, supra note 1, at 23. Here, by contrast, FDA, without explanation, considered timing only. FDA’s “failure to come to grips with conflicting precedent constitutes an inexcusable departure from the essential requirement of reasoned decision making.” Jicarilla Apache Nation vs. U.S. Dept. of Interior, 613 F.3d 1112, 1120 (D.C. Cir. 2010) (“[W]e have never approved an agency’s decision to Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 31 of 50 - 22 - completely ignore relevant precedent. . . .”) As the D.C. Circuit has explained, it is “a clear tenet of administrative law that if the agency wishes to depart from its consistent precedent it must provide a principled explanation for its change of direction.” Nat’l Black Media Coal. v. FCC, 775 F.2d 342, 355 (D.C. Cir. 1985). Accord Atchison v. Wichita Bd. of Trade, 412 U.S. 800, 808 (1973) (agency has a “duty to explain its departure from prior norms”). At the very least, “an agency changing its course must supply a reasoned analysis indicating that prior policies and standards are being deliberately changed, not casually ignored, and if an agency glosses over or swerves from prior precedents without discussion it may cross the line from tolerably terse to intolerably mute.” Greater Bos. Television Corp. v. FCC, 444 F.2d 841, 852 (D.C. Cir. 1970). Fourth, FDA’s own description of its analysis is inconsistent with the dispositive single- factor test it ultimately applied. See Amarin, 106 F. Supp. at 218 (agency’s analysis was “both minimal and confounding,” and “conflicts with other portions of the Agency’s decision”). FDA has consistently described its analysis as applying a favorable presumption of the existence of causation. In its Letter Decision to Amneal, FDA stated that it “will presume [causation]. . . if, at the 30-month date, the evidence demonstrates that the sponsor was actively addressing the change in or review of approval requirements.” FDA 1042 (emphasis added). Indeed, FDA’s guidance also suggests that in situations where the presumption is not satisfied FDA will consider other facts. See 180-Day Exclusivity Guidance, supra note 1, at 22-23. FDA nowhere says that it has created an irrebutable adverse presumption of no causation simply based on timing. But that, in effect, is the test FDA applied in Amneal’s case. What FDA has overlooked is that it does not logically follow that a failure to satisfy a favorable presumption of causation means there is no causation. There may be other reasons for Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 32 of 50 - 23 - causation. Indeed, FDA’s error here is so basic it is a formally named logical fallacy: denying the antecedent.8 FDA applied its presumption as follows: 1. If at the forfeiture date the sponsor is addressing the change (A), then there is causation (B). 2. At the forfeiture date, the sponsor was no longer addressing the change (not A). 3. Therefore, there was no causation (not B). This violates Logic 101, as the following example makes clear: 1. If you are a truck driver (A), then you have a job (B). 2. You are not a truck driver (not A). 3. Therefore, you have no job (not B). See T. Edward Damer, Attacking Faulty Reasoning at 82 (2012) (“The reason such an argument is fallacious is that in denying the antecedent. . . the arguer fails to recognize that A is not the only reason that could bring about B.”). C. FDA’s Decision is not the Product of Reasoned Decision Making Because It Ignores Numerous Facts Relevant to Causation FDA may attempt to argue that it applied a robust, multi-factor causation analysis in Amneal’s case after all. But the Agency’s written decision belies that contention: The Agency’s causation analysis, in its entirety, was: “The Agency’s request that Amneal manufacture and submit information on commercial size lots. . . was. . . resolved before the forfeiture date. . . [and] [t]herefore it could not preclude tentative approval or approval at that time.” FDA 1055. 8 See Wikipedia, Denying the antecedent, https://en.wikipedia.org/wiki/Denying_the_antecedent (last visited Mar. 29, 2017). Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 33 of 50 - 24 - FDA may point to the conclusory sentence that follows as evidence that FDA implicitly considered other potential factors: “Nor has Amneal demonstrated that the request in the September 10, 2014 complete response letter [that Amneal provide commercial-scale data] caused it to fail to obtain tentative approval or approval by December 10, 2105.” FDA 1055. That conclusory statement, however, fails to satisfy basic standards of reasoned decision making. As the Supreme Court has recently reemphasized: [A]n agency must give adequate reasons for its decisions. The agency must examine the relevant data and articulate a satisfactory explanation for its action including a rational connection between the facts found and the choice made. That requirement is satisfied when the agency’s explanation is clear enough that its path may reasonably be discerned. But where the agency has failed to provide even that minimal level of analysis, its action is arbitrary and capricious. . . . Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125 (2016) citing State Farm, 463 U.S. at 38-43. See also Mylan, 910 F. Supp. 2d at 306; Amarin, 106 F. Supp. 3d at 206. It is “a fundamental requirement of administrative law. . . that an agency set forth its reasons for decision; an agency’s failure to do so constitutes arbitrary and capricious agency action . . . . At bottom, an agency must explain why it chose to do what it did. And to this end, conclusory statements will not do; an agency's statement must be one of reasoning.” Amerijet Int'l v. Pistole, 753 F.3d 1343, 1350 (D.C. Cir. 2014) (emphasis original). An agency also may not “entirely fail[] to consider an important aspect of the problem, offer[] an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” State Farm, 463 U.S. at 43. See Purepac Pharm. Co v. TorPharm, Inc., 354 F.3d 877, 885 (D.C. Cir. 2004) (“FDA may not ignore. . . evidence before it. . . .”); Achernar Broad. Co. v. FCC, 62 F.3d 1441, 1446 (D.C. Cir. 1995) (“Failure to weigh the entire record would constitute reversible error. . . .”). Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 34 of 50 - 25 - FDA’s analysis here falls far short of this standard. FDA provided no reasoning other than that the change in approval requirements had been resolved as of the forfeiture date. See Mylan, 910 F. Supp. 2d at 306. It considered no other facts or evidence, did not offer an explanation for why its decision runs counter to that evidence, and is implausible in view of the record. And it is too late for the Agency to correct the flaws in its analysis. “Counsel’s explanation to this court cannot substitute for reasoned decisionmaking at the agency level.” Williams Gas Processing-Gulf Coast Co. L.P. v. FERC, 475 F.3d 319, 329 (D.C. Cir. 2006) (“We do not ordinarily consider agency reasoning that appears nowhere in the [agency’s] order.”) (emphasis original). See also State Farm, 463 U.S. at 50 (“It is well established that an agency’s action must be upheld, if at all, on the basis articulated by the agency itself.”); Encino, 136 S. Ct. at 2127 (“It is not the role of the courts to speculate on reasons that might have supported an agency’s decision.”). FDA’s analysis in this case looks especially threadbare when compared against the “very fact-specific” analysis involving “a number of potential factors” that FDA championed, and this Court approved, in Mylan. 910 F. Supp. 2d at 306. With respect to FDA’s commercial scale-up demand many of the Mylan factors support causation, yet all were ignored by FDA here: • timing (encompassing 14-months of the 30-month forfeiture period, and resolved a mere month before the forfeiture date, with FDA chemistry review of scaled- down commercial batches continuing well past the forfeiture date); • mandatory nature (FDA required compliance for application approval); • consistency with pre-existing requirements (inconsistent with pilot-scale guidance); • the nature and timing of the sponsor’s efforts (substantial, and months-long); and Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 35 of 50 - 26 - • nature, timing, and impact on review (also substantial, and months-long, taking up almost half of the 30-month forfeiture period, and effectively holding up chemistry review during that time). Similarly, FDA did not address the causative impact of the months-long delay when FDA reviewed whether it would receive Amneal’s ANDA despite the existence of a typographical error. In effect, that delay – during which Amneal’s ANDA itself was not being reviewed at all – eliminated over a quarter of Amneal’s 30-months for review, reducing it to 22 months. FDA also failed to address how these delays interrelated, possibly amplifying their effects, or affected other issues still extant as of the forfeiture date. Even tort law – which imposes the far more stringent “but for” causation test – will consider as factual causes individual events that combine to form a sufficient cause – the proverbial railway fires that combine to create a conflagration. This is true even if those individual events alone would not have been sufficiently causative. As the Restatement of Torts explains: In some cases, tortious conduct by one actor is insufficient. . . to cause the plaintiff’s harm. Nevertheless, when combined with conduct by other persons, the conduct overdetermines the harm, i.e., is more than sufficient to cause the harm. . . . The fact that an actor’s conduct requires other conduct to be sufficient to cause another’s harm does not obviate [causation-in-fact.] Restatement (Third) of Torts § 27, (Am. Law Inst. 2010), cmt. f. See also Dan B. Dobbs, et al., The Law of Torts § 189 (2nd 3d. 2011). (“[T]he conduct of all defendants as a group is aggregated and considered as a whole. The but-for test is then applied to their conduct taken as a unit or set. If the combined conduct is a but-for cause of the plaintiff’s harm, then cause is established.”). As one of the Restatement’s examples teaches, three actors negligently leaning on a car, whose individual weights alone are insufficient to push the car over a cliff, but whose combined weight is, “are each a factual cause of the destruction of [the] car.” Restatement (Third) of Torts § 27, cmt. f. The Restatement adopts these independently insufficient causes as Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 36 of 50 - 27 - factual causes because “we recognize them as such in our common understanding of causation, even if the but-for standard does not. . . . [It] comports with deep-seated intuitions about causation and fairness in attributing responsibility.” Id., cmt. c. In Amneal’s case, taken together, the impact of the delays here is striking. As shown by Figure 1, below, these delays took up 22 out of the 30 months – almost 75% – of the forfeiture period. Figure 1 Eight months of delay (shown in purple) related to the delay in receipt of the ANDA. Fourteen months of delay (shown in red) related to delay resulting from FDA’s demand for commercial- scale data. A mere 25% of the 30-month period is not consumed by these delays. Yet nowhere does FDA consider or address the vast amount of Amneal’s 30 months that was expended by the cumulative effect of these delays.9 9 Even if one of these delays is not a change in or review of approval requirements, it still occurred, and cannot simply be ignored as if it never had. For example, if the delay in receipt was not, itself, a change in or review of approval requirements, such a substantial delay is still relevant in considering the impact and effect of the delay due to FDA’s demand for commercial- scale data. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 37 of 50 - 28 - Similarly, FDA should have considered whether issues that existed on the forfeiture date would have existed if not for the delays above. Both delays held up Amneal’s chemistry review for months. But on the forfeiture date, FDA still had unresolved chemistry questions regarding the blending of Amneal’s intermediate product and regarding the fill of Amneal’s capsules. See FDA 0713-16.10 FDA, at least, should have considered the possibility that, but for the delays, these questions could have been resolved sooner. Likewise, FDA should have considered whether, in the absence of these delays, tentative approval could have been granted before issues preventing forfeiture on December 10, 2015 popped up. Amneal’s ANDA was not in a condition for approval as of the forfeiture date in part due to Amneal’s need to change API suppliers after its proposed supplier was subject to an import ban in October 2015. This was just two months before the forfeiture date. Shifting the application’s review forward just two months – far less than the delays here – could have meant tentative approval before that issue even arose. II. FDA Improperly Concluded Amneal’s ANDA did not Experience a Change in or Review of the Requirements for Approval of the Application FDA further erred in its consideration of whether Amneal’s application experienced a “change in or a review of the requirements for approval of the application imposed after the date on which the application is filed.” 21 U.S.C. §355(j)(5)(D)(i)(IV) (emphasis added). Amneal’s application experienced two changes in or reviews of approval requirements that caused its failure to receive tentative approval: 1) FDA’s commercial scale-up demand, and 2) FDA’s review of whether it would receive Amneal’s application notwithstanding a typographical error. 10 Thus, even FDA’s assertion that the issues related to its scale-up demand were resolved by the forfeiture date is not supported by the record. As of the forfeiture date, FDA was still conducting chemistry review of Amneal’s proposed scaled-down commercial batches. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 38 of 50 - 29 - FDA’s letter decision failed to even consider whether the delay in receipt resulted from a change in or review of approval requirements. As for the commercial scale-up demand, FDA only challenged whether its demand for commercial-scale data constitutes a change in a “requirement for approval of the application.” FDA concluded that it did not because there had been no change to a “statutory requirement” or a change to FDA policy. See FDA 1055. That analysis is not supported by the statute and is arbitrary and capricious. A. Amneal’s ANDA Experienced a Change in or Review of the Requirements for Approval of the Application The statute imposes only three requirements, set out in plain, clear English. There must be: 1) A change in or review of; 2) Requirements for approval of the application; 3) Imposed after the date on which the application is filed. 21 U.S.C. §355(j)(5)(D)(i)(IV). These requirements are met by FDA’s demand for commercial- scale data as well as by FDA’s review of whether to receive Amneal’s ANDA notwithstanding a typographical error. 1. FDA’s Commercial Scale-Up Demand FDA’s commercial scale-up demand meets all of the requirements of the statute. First, the scale-up demand was imposed in September 2015, well over a year after the ANDA’s June 2013 submission date. FDA 0233. Requiring commercial-scale, rather than pilot-scale data was also a “change.” FDA customarily allows ANDA filers to submit, and FDA routinely receives and approves, ANDA’s supported by pilot-scale data that seek approval for commercial-scale manufacture. FDA 0252. See also, supra, note 3 (citing guidance). The demand that Amneal provide commercial-scale Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 39 of 50 - 30 - data was a departure – i.e. a change – from that guidance. It also involved a post-filing change to the expectations for the application from those that existed at the time of filing – which is what the statute expressly sets as the point of comparison. As FDA explains in its Letter Decision, the determination to require commercial-scale data can only be made “during the substantive review of the ANDA by the chemistry reviewers. Merely having a complex process is not, by itself, typically sufficient to warrant a request for the manufacture of commercial batches pre- approval.” FDA 1055 (emphasis added). In other words, at the time of the ANDA’s submission FDA does not require submission of commercial-scale data and only makes that determination later in the review process. That’s a change by any definition. It is also a change in a “requirement for approval of the application.” Consider the state of affairs when Amneal filed its application: FDA has stated it will generally receive and approve applications containing only pilot-scale data. And it indeed does so. The requirement for approval for such an application is, at that point, submission of pilot-scale data. If commercial-scale data was a requirement for approval of an application, then FDA’s policy of permitting submission with pilot-scale batch data would be nonsensical. FDA would be encouraging applicants to submit futile applications that could never be approved without additional commercial-scale data. Thus, for all applications, at the time of filing, submission of commercial-scale data is not a requirement for approval. All that is required is submission of pilot-scale data. Yet, in Amneal’s application, something changed that made that application not approvable despite the fact that it included pilot-scale data. Thus, there was a change to a requirement for approval of the application. If there is any doubt, one need only look at the words FDA itself used. FDA told Amneal that the Agency was going to “require[] the review of Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 40 of 50 - 31 - the resulting data from the manufacture of a commercial size lot before application approval.” FDA 0233. Cf. 21 U.S.C. §355(j)(5)(D)(i)(IV) (“requirements for approval of the application”). As demonstrated by Figure 2, FDA’s demand for commercial-scale data almost exactly tracks the statutory text. Figure 2 FDA’s demand for commercial-scale data meets the three requirements of the statute, and FDA erred in concluding otherwise.11 2. FDA’s Review of Whether to Receive the ANDA with a Typographical Error FDA’s consideration of whether to receive Amneal’s ANDA notwithstanding a typographical error was also a change in or review of the requirements for approval of the 11 It is of no moment that FDA ultimately granted Amneal approval at the smaller pilot-scale batch size. Amneal’s application, as submitted, sought manufacture of product at a commercial- scale of 1,000,000 units and did not seek approval of manufacture at the much smaller pilot-scale batch size. FDA required commercial-scale batch data to approve that application as filed. FDA’s statement that it could not approve the application “in its present form” confirms as much. FDA 0231. Amneal was required to amend its application to reduce its proposed commercial batch size. FDA 0526-28. It was only then that FDA withdrew its demand for scaled-up data. But that was because it no longer applied to the amended application, not because it was never a requirement for approval. Moreover, the amendment did not nullify the substantial delay and other ongoing effects of FDA’s scale-up demand. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 41 of 50 - 32 - application after the date on which it was filed. FDA did not even consider this possibility, and that is an independent basis for vacating the Agency’s forfeiture decision. FDA initially refused to receive Amneal’s ANDA because, FDA believed, the ANDA did not include stability data for a sufficient period of time. FDA 0063-65. It turned out that the requisite data was there, but a summary table of the data included a typographical error listing an incorrect start date for the testing. FDA 0072-73. After receiving FDA’s refusal to receive letter, Amneal informed FDA that the ANDA simply contained a typographical error and that the requisite data was included. FDA 0072-73; FDA 0076-93. FDA then undertook a months-long review to determine, ultimately, that the Agency would rescind its refusal to receive the application. FDA 0109-12. These events occurred “after the date on which the application [was] filed”. See 21 U.S.C. § 355(j)(5)(D)(i)(IV). FDA concluded that Amneal’s ANDA was “filed” for purposes of the tentative approval forfeiture statute on June 10, 2013 when Amneal submitted a substantially complete ANDA. FDA 1047-49. Thus, by necessity, all of the activity regarding whether the Agency would receive the application occurred after filing. And this activity involved both a “change in or review of” and a “requirement[] for approval of the application.” See 21 U.S.C. § 355(j)(5)(D)(i)(IV). FDA has never said why it requested further information from Amneal or what the Agency was doing during the many months it took for the Agency to rescind its refusal to receive the ANDA. But, presumably, the Agency was doing what it should have been doing: reviewing whether it would accept an application that contained a typographical error but nevertheless included all of the requisite stability data. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 42 of 50 - 33 - And this review involved a requirement for approval. The Agency’s receipt of an ANDA is by necessity a condition for approval, as it is a condition for any review at all. See, e.g., FDA 0244 (“The receipt of an ANDA means that FDA made a threshold determination that the ANDA is sufficiently complete to permit a substantive review.”) citing 21 CFR 314.101(b)(1). If the Agency refuses to receive the application, it is not even reviewed by the Agency, much less approved. Thus, by logical extension, FDA’s receipt of the application is a requirement for approval. B. FDA’s Requirement that Amneal Show a Change to a “Statutory Requirement” or to FDA Policy Ignores the FDCA’s Text, Failing Chevron Step One FDA found no change in or review of approval requirements in connection with FDA’s demand for commercial-scale data because, the Agency reasoned, there was no change to a “statutory requirement” or to “FDA’s specific requirements for certain complex products” as, apparently, embodied in a policy FDA had never made public. FDA 1055-56. FDA’s focus on “statutory” changes, or changes to its policies as applied to classes of applications is inconsistent with the plain language of the FDCA. See Bennett v. Donovan, 4 F. Supp. 3d. 5, 13 (D.D.C. 2013) (setting aside agency action that was inconsistent with statute). The FDCA does not refer to changes in the statutory standards, or to FDA’s policies or guidance. Rather, the statute clearly states that forfeiture is excused if the failure to obtain tentative approval is “caused by a change in or a review of the requirements for approval of the application imposed after the date on which the application is filed.” 21 U.S.C. §355(j)(5)(D)(i)(IV) (emphases added). The statute specifically mandates that the inquiry be directed to changes applied to particular applications and is not limited to broader statutory or policy changes. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 43 of 50 - 34 - FDA’s requirement that the record show a statutory or policy change renders a significant portion of the statutory text, specifically, the words “of the application” meaningless and thus violates the canon against surplusage. See Qi-Zhuo v. Meissner, 70 F.3d 136, 139 (D.C. Cir. 1995) (“[A]ll words in a statute are to be assigned meaning, and. . . nothing therein is to be construed as surplusage.”); Indep. Ins. Agents of Am., Inc. v. Hawke, 211 F.3d 638, 644-45 (D.C. Cir. 2000) (refusing to give Chevron deference to an agency interpretation that violated the surplusage canon). As this Court recently observed in rejecting similar attempts by FDA to gloss over express statutory language, “It is a cardinal principle of statutory construction that a statute ought, upon the whole, to be construed in a manner that ensures that no clause, sentence, or word shall be superfluous, void, or insignificant.” Amarin, 106 F. Supp. 3d at 209 quoting TRW Inc. v. Andrews, 534 U.S. 19, 31 (2001). “[O]f the application” gives the statutory text only one possible meaning: any and all requirements for approval that apply to that particular application, even application-specific requirements. “By defining” the statute to encompass only changes to all applications or classes of applications, “FDA has deprived [the words ‘of the application’] of any coherent meaning.” See id. FDA’s Letter Decision states: “The determination as to whether manufacture of a commercial-scale batch is necessary is application specific, and is made during substantive review of the ANDA by the chemistry reviewers.” FDA 1055. Thus, FDA concedes that its policy involves an application-specific change to the approval requirements for the application after it is filed. That is all the statute requires, and FDA erred in concluding otherwise. Further, FDA’s interpretation of what constitutes a change in or review of approval requirements is contrary to the Hatch-Waxman Amendments’ goal to promote generic entry into the prescription drug market. See Amarin, 106 F. Supp. 3d at 198. By granting exclusivity to an Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 44 of 50 - 35 - applicant whose delay was caused by a change in the review requirements “of the application,” Congress sought to avoid penalizing applicants for delays beyond their control. See 180-Day Exclusivity Guidance, supra note 1, at 22 (not penalizing applicants for changes in approval requirements continues to incentivize ANDA applicants to challenge patents). FDA undermines the statute’s goals when it excludes post-filing changes to specific applications unless there has been a broader statutory or policy change. If FDA is correct, and a substantial change in expectations like a post-filing demand for commercial-scale data is not a change sufficient to trigger the statute, then the aims of the statute will be thwarted. An ANDA sponsor hoping to ensure 180-day exclusivity now has a significant reason not submit an ANDA unless and until it has fully scaled up to commercial production and obtains all of the requisite commercial-scale data. Otherwise, it proceeds at the risk that FDA might change the rules midstream, and require commercial-scale data after the 30-month clock has already started. Such a construction of the statute delays generic entry into the market and creates substantial barriers to entry for applicants who cannot afford to invest in full commercial-scale production before they even have an ANDA on file. C. FDA’s Limitation of the Statute to Statutory or Policy Change Relies on a Secret FDA Policy and is Inconsistent with Prior Precedent FDA’s construction also fails Chevron step two because it is arbitrary and capricious in substance, see Amarin, 106 F. Supp. 3d at 206. FDA’s interpretation is arbitrary and capricious because (1) it encompasses secret, nonpublic policies; and (2) it is inexplicably inconsistent with FDA’s prior precedent. Application of Secret Policy. Amneal submitted its ANDA in accordance with FDA’s public requirements and guidance for like applications, and relied on those public statements in determining what might be required and how best to put itself in a position to satisfy Congress’s Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 45 of 50 - 36 - 30-month forfeiture deadline. As discussed above, FDA’s public guidance permits, even encourages, the submission of pilot-size batches and corresponding production data in support of an ANDA. Amneal’s submission was consistent with these requirements. FDA claims that there was no change in the approval requirements because its demand for commercial-scale information was pursuant to an FDA policy requiring commercial-scale information for “certain” products on an application-specific basis. FDA 1055. FDA, however, did not identify a single instance available to Amneal or to the public, in which FDA has articulated such a policy. In fact, the Letter Decision expressly states FDA is unable to disclose when it has applied this policy due to confidentiality obligations under FDA’s disclosure regulations. FDA 1055 n. 43. Indeed, portions of the administrative record in this case relating to those instances have likewise been redacted by FDA. FDA 1032. FDA admits that this policy is not evident from the public record available to applicants prior to filing an ANDA, stating that “FDA’s general policies on manufacturing test batches do not describe [the circumstances under which FDA might require commercial-scale data].” FDA 1056. This Court should not give deference to FDA’s interpretation of statutory language when it results in ANDA applicants being punished by the Agency without fair-notice of FDA’s requirements. See Christopher v. SmithKline Beecham Corp., 567 U.S. 142, 156 (2012) (“To defer to the agency’s interpretation. . . would seriously undermine the principle that agencies should provide regulated parties fair warning of the conduct a regulation prohibits or requires.”); see also Martin v. OSHRC, 499 U.S. 144, 158 (1991) (identifying “adequacy of notice to regulated parties” as one factor relevant to the reasonableness of the agency’s interpretation). The Supreme Court has “long warned” against agencies enacting “unfair surprise[s]” against Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 46 of 50 - 37 - regulated parties. Christopher, 132 S. Ct. at 2167 quoting Long Island Care at Home, Ltd. v. Coke, 551 U.S. 158, 170-171 (2007). Amneal invested in an ANDA application and took the risk of provoking a patent infringement suit for the anticipated reward of 180-day exclusivity. See Mylan, 910 F. Supp. 2d at 311 (withholding exclusivity would deprive an ANDA sponsor of its “anticipated reward” for “stick[ing] out [its] neck[]”). Given the stakes and that investment, it is arbitrary and capricious for the Agency to then find a forfeiture of exclusivity where the failure to obtain approval was caused by application of a secret policy that could not have been anticipated, and which was disclosed after the application had already been filed. Had Amneal been aware of the potential need to provide commercial-scale data, it may have been able to do so before the ANDA’s submission and avoided a substantial delay after the 30-month clock had already been started. Unexplained Inconsistency with FDA Precedent. FDA’s focus on “statutory requirements” and “specific requirements for certain complex products” in this case is also inconsistent with FDA’s prior precedent. In FDA’s 180-Day Exclusivity Guidance, FDA provided several product-specific examples of what has constituted a “change or review of the requirements for approval” including: changes in a reference drug’s labeling or formulation that require an ANDA applicant to respond, and changes in approval requirements related to drug quality that require an applicant to conduct additional testing. 180-Day Exclusivity Guidance, supra note 1, at 23. These are not statutory changes or changes to policy; rather, they are changes to the requirements for specific products and applications. For example, in 2010, FDA excused Nycomed’s failure to obtain timely tentative approval for Imiquimod Cream, 5%, on the basis of the agency’s “ongoing review of the requirements for approval of Imiquimod Cream, Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 47 of 50 - 38 - 5%.”12 Similarly, FDA excused Teva’s failure to obtain tentative approval due to a change “for Teva’s proposed product” that “required. . . additional testing and. . . an additional drug substance specification.”13 Similarly, changes in the labeling or formulation of specific products have been deemed sufficient.14 In each of these examples, there was no general change to statutory requirements or to FDA policy. FDA nevertheless found those changes still qualified. It was arbitrary and capricious for FDA to have evaluated Amneal under a different standard without any explanation. III. Amneal is Entitled to a Declaration That It Has Not Forfeited Exclusivity Given FDA’s errors outlined above, Amneal is entitled to at least vacatur of FDA’s forfeiture decision, vacatur of FDA’s improper approval of third-party ANDAs contrary to Amneal’s exclusivity, and a remand for the Agency to issue a decision that is not arbitrary, capricious, and contrary to law. However, this case presents one of the rare circumstances in which the Court should enter a declaration that Amneal has not forfeited exclusivity, rather than merely remanding to the agency for further proceedings. While the Supreme Court has directed that courts ordinarily should vacate and remand erroneous agency decisions, see SEC v. Chenery Corp., 332 U.S. 194, 196 (1947), this “ordinary 12 Letter from FDA to Nycomed U.S., Inc. at 2 (Feb. 25, 2010), http://www.accessdata.fda.gov/ drugsatfda_docs/appletter/2010/078548s000ltr.pdf. 13 Letter from FDA to Teva Pharmaceuticals USA at 2 n.1 (March 30, 2012), http://www.accessdata.fda.gov/drugsatfda_docs/appletter/2012/077159s000ltr.pdf 14 See e.g. Letter from FDA to Synthon Pharmaceuticals, Inc. at 2, n.1 (Nov. 26, 2010) (change to labeling of reference drug), http://www.accessdata.fda.gov/drugsatfda_docs/appletter/2010/ 090229s000ltr.pdf; Letter from FDA to Sandoz, Inc. (Sept. 20, 2011) (formulation change to reference drug), https://www.regulations.gov/contentStreamer?documentId=FDA-2010-P-0632- 0017&attachmentNumber=1. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 48 of 50 - 39 - remand rule” is not absolute. As the D.C. Circuit has explained, “On occasion. . . we find that a remand would be futile on certain matters as only one disposition is possible as a matter of law. In such cases, we retain and decide the issue. Chenery and its progeny, while emphasizing the importance of agency explanations, do not require that we convert judicial review of agency action into a ping-pong game.” George Hyman Constr. Co. v. Brooks, 963 F.2d 1532, 1539 (D.C. Cir. 1992) quoting NLRB v. Wyman, 394 U.S. 759, 766-67 n.6 (1969). Thus, a court may decline to remand, and enter a declaration, where “no rational factfinder conscientiously applying [the law]” would find otherwise. George Hyman, 963 F.2d at 1539. See also Donovan v. Stafford Constr. Co., 732 F.2d 954, 961 (D.C. Cir. 1984) (“remand. . . would serve no purpose” where under all the record evidence “only one conclusion would be supportable”). That is the case here. As outlined above, the delays at issue here were substantial, highly onerous, extremely disruptive of review, consumed Amneal’s ANDA for months, and continued until nearly the eve of the forfeiture date. No rational factfinder, applying the proper causation test mandated by the FDCA, could conclude that these delays did not cause the failure to obtain tentative approval. Likewise, no rational factfinder, applying the application-specific analysis mandated by the FDCA, could conclude that these delays were not due to changes in or reviews of approval requirements. In view of the overwhelming record in this case, “[t]o remand would be an idle and useless formality.” Wyman, 395 U.S. at 767-67 n.6. This Court should enter a declaration that Amneal has not forfeited exclusivity. CONCLUSION For the foregoing reasons, Amneal respectfully requests that the Court enter an order granting summary judgment to Amneal and granting the relief sought in Amneal’s Complaint. Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 49 of 50 - 40 - DATED: March 29, 2017 AMNEAL PHARMACEUTICALS LLC, By its attorneys, /s/ Areta Kupchyk Areta Kupchyk (D.C. Bar No. 483626) FOLEY HOAG LLP 1717 K Street, N.W. Washington, DC 20006 Tel.: 202-223-1200 Dean Richlin (pro hac vice) Marco J. Quina (pro hac vice) FOLEY HOAG LLP 155 Seaport Boulevard Boston, MA 02210 Tel.: 617-832-1000 Case 1:17-cv-00180-RDM Document 25 Filed 03/29/17 Page 50 of 50 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA AMNEAL PHARMACEUTICALS LLC, Plaintiff, v. FOOD AND DRUG ADMINISTRATION, et al., Defendants, and LUPIN PHARMACEUTICALS, INC., et al., Intervenor-Defendant. Case No. 1:17-cv-00180 (RDM) [PROPOSED] ORDER Having considered Plaintiff’s Motion for Summary Judgment, the related memoranda filed by the parties, and the Administrative Record, the Court hereby GRANTS Plaintiff’s Motion. IT IS HEREBY ORDERED THAT: 1. Plaintiff’s Motion is GRANTED. 2. The Court VACATES FDA’s decision that Plaintiff forfeited 180-day exclusivity under 21 U.S.C. § 355(j)(5)(D)(i)(IV). 3. The Court VACATES any final approval by FDA of any abbreviated new drug application, other than Plaintiff’s, for Memantine hydrochloride extended-release capsules, 7 mg, 14 mg, 21 mg, or 28 mg strengths. 4. The Court ENJOINS FDA from granting final approval to any abbreviated new drug application, other than Plaintiff’s, for Memantine hydrochloride extended-release Case 1:17-cv-00180-RDM Document 25-1 Filed 03/29/17 Page 1 of 2 capsules, 7 mg, 14 mg, 21 mg, or 28 mg strengths, until the expiration of Amneal’s 180-day exclusivity. 5. The Court DECLARES that Plaintiff has not forfeited 180-day exclusivity under 21 U.S.C. § 355(j)(5)(D)(i)(IV). Entered this ____ day of _____ , 2017. Hon. Randolph D. Moss United States District Judge Case 1:17-cv-00180-RDM Document 25-1 Filed 03/29/17 Page 2 of 2
