William H. Carson, M.D. President & CEO

Otsuka Pharmaceutical Development & Commercialization, Inc.

July 13, 2015

Division of Dockets Management

FOOD AND DRUG ADMINISTRATION

Department of Health and Human Services

5630 Fishers Lane

Room 1061, HFA-305

Rockville, Maryland 20852

CITIZEN PETITION

Dear Madam/Sir:

Pursuant to section 505(q) of the Federal Food, Drug, and Cosmetic Act (“FFDCA”) and 21

C.F.R. § 10.30, Otsuka Pharmaceutical Development & Commercialization, Inc., for itself and

on behalf of its parent company, Otsuka Pharmaceutical Co., Ltd., and its affiliates (collectively

“Otsuka”), submits this citizen petition. This petition requests that the Commissioner of the Food

and Drug Administration (“FDA”) refuse to approve or delay approval of the New Drug

Application (“NDA”) submitted by Alkermes plc (“Alkermes”) for aripiprazole lauroxil (the

“Alkermes NDA”).

As detailed below, this petition rests on two grounds. First, FDA should delay or withhold final

approval of the Alkermes NDA at least until the expiration of Otsuka’s applicable Hatch-

Waxman exclusivity period on December 5, 2017. Second, the Alkermes NDA should not be

approved because it does not satisfy the FFDCA’s substantial evidence requirement of two

clinical trials, and it does not meet FDA’s strict standards for deviating from the sound general

rule of requiring this minimum quantum of evidence of two clinical trials.

Petitioner Otsuka holds an approved NDA for Abilify® (aripiprazole),1 an atypical antipsychotic

indicated for treatment of schizophrenia and several other indications. Otsuka also holds an

approved NDA for Abilify Maintena® (aripiprazole),2 a long-acting formulation for treatment of

schizophrenia. On August 25, 2014, Alkermes announced that it had submitted the Alkermes

1 NDA No. 021436. 2 NDA No. 202971.

2

NDA, which Alkermes proposes as a long-acting formulation for treatment of schizophrenia.3

On September 9, 2014, Otsuka submitted a citizen petition (Docket No. FDA-2014-P-1354)

requesting that FDA refuse to file or refuse to accept for substantive review the Alkermes NDA

because the application was (and is) facially legally insufficient.4 FDA denied that petition on

February 3, 2015, as premature.5 The issues and arguments raised in that prior citizen petition

are now ripe and, accordingly, Otsuka here adopts and incorporates the legal arguments from its

previous petition and relies upon them (in addition to new arguments raised below) to support its

claims and requests for relief in this petition.

A. ACTIONS REQUESTED

FDA should delay or withhold final approval of the Alkermes NDA at least until Otsuka’s three-

year exclusivity for the conditions of approval of aripiprazole expires on December 5, 2017.

FDA should also refuse to approve the Alkermes NDA in its entirety because the NDA does not

satisfy the substantial evidence of effectiveness requirement prescribed in Section 505(b)(1)(A)

of the FFDCA. Contrary to that requirement, the Alkermes NDA is supported with only one

adequate and well-controlled clinical trial (“AWCT”) of the drug (aripiprazole lauroxil) for

which Alkermes seeks approval.

B. STATEMENT OF GROUNDS

I. Summary of Argument

FDA’s decision in the Envarsus® matter, which was affirmed recently by the United States

District Court for the District of Columbia in Veloxis Pharmaceuticals, Inc. v. FDA,6 applies here

and compels FDA to deny or delay any final approval of the Alkermes NDA. As FDA’s

decision in that matter makes clear, exclusivity under Sections 505(c)(3)(E)(iii) and (iv) of the

FFDCA is “triggered by an overlap in the conditions of approval between the first-in-time 505(b)

drug and the second-in-time 505(b)(2) NDA.”7 Otsuka’s first-in-time NDA/sNDAs for Abilify

Maintena were approved and, based on the new clinical investigations submitted in support of

the applications, the drug received three-year exclusivity for a long-acting, once-monthly,

injectable formulation of aripiprazole in the treatment of schizophrenia as demonstrated by

improvements of symptoms in acutely relapsing patients and maintaining symptom control in

stabilized patients (maintenance or relapse patients).

3 Press Release, Alkermes Submits New Drug Application to FDA for Aripiprazole Lauroxil for Treatment of

Schizophrenia (Aug. 25, 2014), available at http://phx.corporate-ir.net/phoenix.zhtml?c=92211&p=irol-

newsArticle&ID=1960579&highlight (hereinafter “Alkermes News Release, Aug. 25, 2014”). 4 Letter from William H. Carson, M.D., President & CEO, Otsuka Pharmaceutical Development &

Commercialization, Inc., to Division of Dockets Management, Food and Drug Administration (Sept. 9, 2014)

(hereinafter “Otsuka Petition”). A copy of the original petition is attached as Exhibit 1. 5 Letter from Janet Woodcock, M.D., Director, Center for Drug Evaluation and Research, to William H. Carson,

M.D., President & CEO, Otsuka Pharmaceutical Development & Commercialization, Inc. (Feb. 3, 2015). 6 No. 14-2126, 2015 U.S. Dist. LEXIS 77559 (D.D.C. June 12, 2015). 7 Id. at *30-31.

3

If the Alkermes NDA is approved, that approval would contradict the logic of FDA’s decision in

the Envarsus matter because the conditions of approval of the Alkermes NDA would overlap

with the conditions of approval for Abilify Maintena. The clinical trials conducted by Alkermes,

similar to Otsuka’s, were designed as placebo-controlled studies in an acutely relapsed

population to show the effectiveness of a long-acting, once-monthly, injectable formulation that

releases aripiprazole for treatment of schizophrenia. Alkermes describes aripiprazole lauroxil as

having the same key characteristics approved in the Abilify Maintena NDA and sNDA labeling:

(1) “the active moiety is aripiprazole”8 in (2) a “long-acting,”9 (3) “injectable”10 (4) “once-

monthly”11 formulation for (5) “treatment of schizophrenia”12 in “patients experiencing [an]

acute exacerbation.”13 Even if aripiprazole lauroxil is deemed a new active moiety and a New

Chemical Entity (“NCE”), it is nonetheless true, as Alkermes expressly concedes, that “[o]nce in

the body, aripiprazole lauroxil converts to aripiprazole.”14 Because Otsuka’s Abilify Maintena

and the Alkermes NDA share common conditions of approval, final approval of the Alkermes

NDA must be delayed until Otsuka’s three-year exclusivity expires.

FDA should also deny the Alkermes NDA in its entirety because it is legally and factually

insufficient to constitute substantial evidence of effectiveness. Assuming FDA determines the

Alkermes NDA is for an NCE, the sound general rule is that an NDA for an NCE must be

supported by, among other things, at least two AWCTs. That rule applies here. Alkermes,

however, undeniably fails to satisfy this requirement as it has submitted only one AWCT.

Alkermes attempts to avoid the two AWCT requirement by improperly seeking to invoke the

abbreviated Section 505(b)(2) pathway and relying on the aripiprazole tablet. This attempted

reliance on aripiprazole’s record is permissible only if FDA deems the active moieties of

aripiprazole and aripiprazole lauroxil as essentially the same compound (and, even then,

Otsuka’s three-year exclusivity acts as a bar to approval pending the expiration of that

exclusivity). Otsuka argued in its first petition that this approach on the part of Alkermes

violates FDA’s bright-line rule regarding changes to molecules with non-covalent bonds. Those

prior arguments are adopted here and incorporated by reference.

FDA should reject the transparent and improper gamesmanship being attempted here. Although

Alkermes appears to maintain that a single AWCT of its NCE and reliance on the aripiprazole

tablet are sufficient to meet the effectiveness requirements of Section 505(b)(1), Alkermes

8 See Richard Pops, Chairman and CEO, Alkermes’ CEO Presents at Goldman Sachs Healthcare Conference

(June 13, 2013), available at http://seekingalpha.com/article/1500922-alkermes-ceo-presents-at-goldman-sachs-

healthcare-conference-transcript (hereinafter “Pops Goldman Sachs Healthcare Conference”). 9 Press Release, Alkermes’ New Drug Application for Aripiprazole Lauroxil for Treatment of Schizophrenia

Accepted for Filing by U.S. FDA (Oct. 22, 2014), available at http://phx.corporate-

ir.net/phoenix.zhtml?c=92211&p=irol-newsArticle&ID=1980168 (hereinafter “Alkermes News Release, Oct. 22,

2014”). 10 Id. 11 Id. 12 Id. 13 Press Release, Alkermes Announces Positive Topline Results from Pivotal Phase 3 Study of Aripiprazole

Lauroxil for Treatment of Schizophrenia (Apr. 8, 2014), available at

http://investor.alkermes.com/mobile.view?c=92211&v=203&d=1&id=1916604 (hereinafter Alkermes News

Release, Apr. 8, 2014). 14 Alkermes News Release, Aug. 25, 2014.

4

nevertheless seeks to avoid the blocking effect of Otsuka’s three-year exclusivity for a long-

acting, injectable, monthly aripiprazole formulation (marketed as Abilify Maintena).15 This

attempted end-run, if permitted, would circumvent Otsuka’s lawful exclusivity covering certain

overlapping conditions of use for aripiprazole. Alkermes is not entitled to game the system and

avoid the bar of Otsuka’s exclusivity while, at the same time, avoiding the burden of the two-trial

requirement.

II. Statement of Facts

Otsuka holds an NDA for Abilify (aripiprazole) for treatment of schizophrenia.16 Multiple drug

products are marketed under this NDA, including different formulations, dosage forms, routes of

administration, and indications.17 On February 28, 2013, FDA approved an NDA for Abilify

Maintena (aripiprazole), a once-monthly, long-acting, injectable formulation of aripiprazole for

the treatment of schizophrenia.18 That NDA was supported by a new clinical investigation of

Abilify Maintena in one longer-term, double-blind, placebo-controlled, randomized withdrawal

(maintenance) trial with appropriate reference to all of Otsuka’s existing proprietary aripiprazole

data. An sNDA for Abilify Maintena was later approved on December 5, 2014, for the treatment

of schizophrenia based on clinical data from one short-term (twelve-week), randomized, double-

blind, placebo-controlled trial in acutely relapsed adults. Otsuka received three-year marketing

exclusivity under Section 505(c)(3)(E)(iii) and (iv) for the February 28 and December 5

approvals.

On August 25, 2014, Alkermes announced that it had submitted an NDA seeking FDA approval

of aripiprazole lauroxil.19 By its own admission, Alkermes claims that aripiprazole lauroxil is

essentially the same drug as aripiprazole, treating the same patient population, for the same

indication(s), with the same expected clinical benefit as Abilify. Alkermes says that, “[o]nce in

the body, aripiprazole lauroxil converts to aripiprazole, which is commercially available under

the name ABILIFY®.”20 Alkermes further claims that, despite the “more complicated” nature of

the aripiprazole lauroxil molecule, the safety and effectiveness of aripiprazole lauroxil can be

and is supported by FDA’s findings regarding Abilify (aripiprazole).21 Alkermes also describes

15 See, e.g., Press Release, Alkermes plc Reports First Quarter 2015 Financial Results (Apr. 30 2015), available

at http://investor.alkermes.com/mobile.view?c=92211&v=203&d=1&id=2042120 (“Aripiprazole lauroxil, our long-

acting atypical antipsychotic for schizophrenia, is moving toward FDA approval and launch later this year…”). 16 NDA No. 021436. 17 See NDA Nos. 021436, 021713, 021729, and 021866, and related supplements. 18 NDA No. 202971. 19 Alkermes News Release, Aug. 25, 2014. 20 Id. 21 See, e.g., Pops Goldman Sachs Healthcare Conference (“What [sic] so, cool about Aripiprazole Lauroxil,

9070 is that, because the active moiety is aripiprazole. We inject this once a month into the muscle as a prodrug, it

metabolizes into aripiprazole, we track in the blood stream aripiprazole levels . . . . We know that we’re delivering

therapeutic concentrations of aripiprazole.”); Richard Pops, Chairman and CEO, Alkermes CEP Presents at Citi

Global Healthcare Conference Transcript (Feb. 25, 2013), available at http://seekingalpha.com/article/1222541-

alkermes-ceo-presents-at-citi-global-helathcare-conference-transcript (“Once in the body this more complicated

molecule antibody clips down to Aripiprazole, for the active moiety in the blood stream of these patients for the

month and time is Aripiprazole, and that way we can build off of a huge clinical foundation of safety and efficacy of

this molecule.” (emphasis added)); Jim Frates, Senior VP and CFO, Alkermes’s Management Presents at Deutsche

Bank 38th Annual dbAccess Health Care Conference Transcript (May 29, 2013), available at

5

aripiprazole lauroxil as having the same key characteristics approved in the Abilify Maintena

NDA and sNDA labeling: (1) “the active moiety is aripiprazole”22 in (2) a “long-acting,”23 (3)

“injectable” 24 (4) “once-monthly” 25 formulation for (5) “treatment of schizophrenia” 26 in

“patients experiencing [an] acute exacerbation.”27

On October 22, 2014, Alkermes announced that FDA had accepted the Alkermes NDA for filing,

with a target action date of August 22, 2015.28 According to Alkermes, its NDA is supported by

only a single AWCT. 29 Indeed, Alkermes later sent Otsuka a Paragraph IV certification,

notifying Otsuka that Alkermes had submitted a 505(b)(2) containing data from studies

conducted on the aripiprazole tablet.

III. Legal Overview

A. Exclusivity Under Section 505(c)(3)(E)(iii) & (iv)

Sections 505(c)(3)(E)(iii) and (iv) of the FFDCA provide that NDAs and sNDAs are eligible for

three years of marketing exclusivity for “conditions of approval” (in the case of an NDA) or a

“change” in the conditions of approval (in the case of an sNDA) where the application contains

reports of “new clinical investigations” that were “essential to the approval” of the application.

FDA interprets both provisions to afford protection to “conditions of approval.”30 The new

clinical investigations submitted in support of the first-in-time 505(b) NDA determine the scope

of the conditions of approval and, thus, the resulting exclusivity. FDA has been explicit that

“[t]he statute sets up a relationship between the ‘new clinical investigations’ that are ‘essential to

the approval of the [application],’ and the scope of exclusivity.”31

http://seekingalpha.com/article/1467941-alkermes-management-presents-at-deutsche-bank-38th-annual-dbaccess-

health-care-conference-transcript (“[W]hat we are trying to do is deliver Aripiprazole, native Aripiprazole, over the

course of a month.”); Richard Pops, Chairman and CEO, Alkermes’s CEO Presents at Bank of America Merrill

Lynch Smid Cap Conference Transcript (May 8, 2013), available at http://seekingalpha.com/article/1415361-

alkermes-ceo-presents-at-bank-of-america-merrill-lynch-smid-cap-conference-transcript (“Our product is a prodrug,

the prodrug of Aripiprazole designed specifically to be an injectable product once a month. Once it’s injected, it fits

comfortably in the muscle for a long period of time and it tubulizes and releases Aripiprazole.”); James Frates, CFO,

Alkermes’ Management Presents at Credit Suisse 2012 Healthcare Conference Transcript (Nov. 14, 2012), available

at http://seekingalpha.com/article/1009251-alkermes-management-presents-at-credit-suisse-2012-healthcare-

conference-transcript (“And one of the things – one of the questions we don’t have to answer in the clinical program

is whether ABILIFY actually treats schizophrenia.”). 22 Pops Goldman Sachs Healthcare Conference. 23 Alkermes News Release, Oct. 22, 2014. 24 Id. 25 Id. 26 Id. 27 Alkermes News Release, Apr. 8, 2014. 28 Alkermes News Release, Oct. 22, 2014. 29 Id. 30 See Letter from Janet Woodcock, M.D., Director, Center for Drug Evaluation and Research, to Alessandra C.

Ravetti and Emily Marden, Pfizer Inc., at 3-4 (Aug. 31, 2010) (hereinafter “Xalatan Response”). 31 Letter from Keith O. Webber, Ph.D., Deputy Director, Office of Pharmaceutical Science, Center for Drug

Evaluation and Research, to Kevin McKenna, Ph.D., c/o Peter O. Safir, Covington & Burling LLP, at 8 (Mar. 27,

2012) (hereinafter “Seroquel Letter”). FDA’s regulations confirm this reading of the statutory language, by

6

Where, as here, there is any “overlap” in the conditions of approval between the approved first-

in-time 505(b) application and the second-in-time 505(b)(2) application, the exclusivity

protecting those overlapping conditions of approval of the first-in-time 505(b) blocks FDA

approval of the second-in-time 505(b)(2) application.32 That block in full force remains until the

exclusivity attached to the first-in-time 505(b) application expires. This is true even if the

second-in-time 505(b)(2) relies on clinical investigations other than those relied upon in the first-

in-time 505(b) NDA.33

In the case of Veloxis Pharmaceuticals, Inc. v. FDA, plaintiff Veloxis challenged FDA’s

withholding of final approval of Veloxis’s Envarsus XR for the prophylaxis of organ rejection in

de novo kidney transplant patients under Section 505(c)(3)(E)(iii). The court rejected the

challenge, affirming FDA’s action in withholding final approval, because the Envarsus XR

conditions of approval overlapped with the Astagraf XL NDA, which was filed and received

exclusivity first. The same is true here.

The background of the Veloxis matter was that Astellas Pharma US, Inc. submitted, and FDA

approved, a 505(b)(1) NDA for a tacrolimus formulation in 1994, under the trade name

Prograf.34 Prograf is a twice-daily, immediate release capsule used for the “prophylaxis of organ

rejection in patients receiving” kidney transplants.35 FDA approved a 505(b)(1) NDA for a once

daily tacrolimus formulation for the same indication under the trade name Astagraf XL in

2013.36 The Astagraf XL NDA relied on two clinical studies that demonstrated the drug could

be used for the “prophylaxis of organ rejection in patients receiving de novo kidney

transplants.”37 FDA determined that Astagraf XL was entitled to three-year exclusivity for these

conditions of approval.38

In December 2013, Veloxis submitted a 505(b)(2) NDA for a tacrolimus formulation, under the

trade name Envarsus XR. 39 Envarsus is a once daily, extended-release tablet for the

“prophylaxis of organ rejection in both de novo and conversion kidney transplant patients.”40

The Envarsus XR NDA relied on independent clinical studies, as well as clinical studies relied

on for the approval of the Prograf NDA.41 Despite the fact that Veloxis did not “rely on” the

Astagraf XL clinical studies, FDA determined that Astagraf XL’s three-year exclusivity

prevented the agency from granting complete and final approval to Envarsus XR.42

providing that FDA cannot issue final approval to “a 505(b)(2) [NDA] . . . for the conditions of approval of the

[first-in-time 505(b) NDA].” 21 C.F.R. § 314.108(b)(4)(iv). 32 Veloxis, 2015 U.S. Dist. LEXIS 77559, at *30-31. 33 Id. at *30-34. 34 Id. at *12. 35 Id. 36 Id. at *12-13. 37 Id. at *13 (internal quotation marks omitted). 38 Id. 39 Id. at *14. 40 Id. (internal quotation marks omitted). 41 Id. 42 Id. at *14-15.

7

Veloxis filed suit, challenging FDA’s withholding of complete and final approval. The district

court affirmed FDA’s final agency action. The court first held that exclusivity under Section

505(c)(3)(E)(iii) “is triggered by an overlap in the conditions of approval between the first-in-

time 505(b) drug and the second-in-time 505(b)(2) NDA.”43 The court dismissed Veloxis’s

argument that three-year exclusivity only acts as a bar where a 505(b)(2) applicant “relies upon”

the first-in-time 505(b) NDA, concluding that there does not need to be “an overlap between the

‘new clinical investigations’ supporting the first-in-time 505(b) NDA and second-in-time

505(b)(2) NDA.”44 The court then held that Envarsus XR and Astagraf XL shared conditions of

approval. 45 The Astagraf XL “conditions of approval” for a once-daily, extended release

tacrolimus formulation for prophylaxis of organ rejection in de novo kidney transplant patients

were “share[d]” with Envarsus XR.46 The court decided that other “‘clinically meaningful’

differences” – which Veloxis argued included different dosage forms, formulations, and clinical

and pharmokinetic profiles – did not change this conclusion because “[t]he effect of marketing

exclusivity in [Section 505(c)(3)(E)(iii)] turns on whether a second-in-time 505(b)(2) NDA

shares any conditions of approval with the first-in-time 505(b) drug granted exclusivity.”47

B. Adequate And Well Controlled Clinical Trials

Section 505(b)(1) of the FFDCA sets forth the information requirements to support approval of

an NDA. Section 505(b)(1) requires NDA applicants to include “full reports of investigations”

of safety and effectiveness.48 FDA’s guidance on the subject is consistent with this plural

statutory language: “With regard to quantity [of evidence required to show that there is

‘substantial evidence’ of effectiveness], it has been FDA’s position that Congress generally

intended to require at least two adequate and well-controlled studies, each convincing on its own,

to establish effectiveness. FDA’s position is based on the language in the statute and the

legislative history of the 1962 amendments.”49

Section 505(d), which sets out FDA’s approval requirements, confirms that a single AWCT

generally does not suffice. Section 505(d)(1) provides that FDA shall refuse to approve an NDA

where “the investigations, reports of which are required to be submitted to the Secretary pursuant

to subsection (b) of this section, do not include adequate tests by all methods reasonably

applicable to show whether or not such drug is safe for use under the conditions prescribed,

recommended, or suggested in the proposed labeling.”50 Further, if there is a lack of “substantial

evidence that the drug will have the effect it purports or is represented to have under the

conditions of use prescribed, recommended, or suggested in the proposed labeling,” the

43 Id. at *30-31. 44 Id. 45 Id. at *41-45. 46 Id. at *44. 47 Id. (emphasis added); Veloxis, Compl. ¶¶ 92, 95-106, ECF No. 1. 48 21 U.S.C. § 355(b)(1)(A) (emphasis added). 49 Guidance for Industry: Providing Clinical Evidence of Effectiveness for Human Drug and Biological

Products, at 3 (May 1998) (emphasis added; citation and footnote omitted) (hereinafter “Clinical Effectiveness

Guidance”). 50 21 U.S.C. § 355(d)(1) (emphasis added).

8

Secretary shall refuse to approve the NDA.51 Section 505(d) defines “substantial evidence” as

“evidence consisting of adequate and well-controlled investigations, including clinical

investigations.”52

Congress enacted Section 505(d) of the FFDCA, which allows for approval based on a single

AWCT and “confirmatory evidence” shortly after the release of FDA’s draft guidance. FDA

subsequently finalized the draft guidance without significant changes, indicating the Agency’s

belief that the new provision in Section 505(d) did not change FDA’s filing or approval

standards requiring two AWCTs. The statutory exception to the two AWCT requirement

described in 505(d) does not provide additional exceptions to those rare situations detailed in the

Clinical Effectiveness Guidance; it only clarified that they exist.53

Likewise, FDA’s regulations provide that an NDA must include “[a] section describing the

clinical investigations of the drug”; that FDA “may refuse to approve an application [because] . .

. [t]he investigations required under section 505(b) of the act do not include adequate tests by all

methods reasonably applicable to show whether or not the drug is safe for use under the

conditions prescribed, recommended, or suggested in its proposed labeling” or “[t]here is a lack

of substantial evidence consisting of adequate and well-controlled investigations”; and that

“[r]eports of adequate and well-controlled investigations provide the primary basis for

determining whether there is ‘substantial evidence’ to support the claims of effectiveness for new

drugs.”54

There are only two circumstances where a single, new AWCT may be sufficient to show

“substantial evidence” of effectiveness. The first is where “a single study of a new use, with

independent substantiation from study data in related uses, could provide evidence of

effectiveness.”55 That is, evidence from one or more AWCTs of a previously-approved drug

product with the same active moiety can count toward the two-study requirement for a proposed

new use of that same active moiety.56 The second is where a single AWCT meets one or more

factors related to the size of the study, multiple endpoints, consistent results across population

subsets, and strength of statistical results.57 However, reliance on a single study “will generally

be limited to situations in which a trial has demonstrated a clinically meaningful effect on

mortality, irreversible morbidity, or prevention of disease with potentially serious outcome and

confirmation of the result in a second trial would be practically or ethically impossible.”58

While generally an NDA sponsor must conduct all clinical and nonclinical studies necessary to

support a finding of safety and effectiveness, or obtain a right to reference data owned by others,

51 Id. § (d)(5). 52 Id. § (d) (emphasis added). 53 Clinical Effectiveness Guidance, at 3-4 (“In making this clarification, Congress confirmed FDA’s

interpretation of the statutory requirements for approval and acknowledged the Agency’s position that there has been

substantial progress in the science of drug development in higher quality clinical trial data.”). 54 21 C.F.R. § 314.50(b)(5); 314.125(b)(2)(5), (d)(5); 314.126(a) (emphases added). 55 Clinical Effectiveness Guidance, at 8. 56 Id. at 7-12. 57 Id. at 12-16. 58 Id. at 13.

9

section 505(b)(2) allows an NDA sponsor to satisfy some portion of the 505(b)(1) evidentiary

requirements with investigations conducted by other parties and to which the sponsor did not

have a right of reference, provided that the NDA sponsor submits certifications to any patents

listed for the reference listed drug. Although an NCE may be approved under Section 505(b)(2),

FDA states that data to support such an application “is likely to be derived from published

studies, rather than FDA’s previous finding of safety and effectiveness of a drug.” 59 This

reference to use of supporting literature is based on the potential that an NCE “may have been

studied by parties other than the applicant and published information may be pertinent to the new

application.”60 An NCE would thus be suitable for approval under 505(b)(2) when it is based on

independent published studies rather than prior FDA findings of effectiveness because the

required investigations must be evaluations of the active moiety contained in the proposed new

drug product.

IV. Discussion

a. The Alkermes NDA, if deemed otherwise approvable, must be delayed

pending the expiration of applicable Hatch-Waxman exclusivity protecting

long-acting, monthly injectable formulations of aripiprazole.

Otsuka has three-year exclusivity rights for certain conditions of approval of aripiprazole, as

evidenced by listings in the Orange Book for Abilify Maintena. Specifically, the new

investigations in support of the NDA and sNDA for Abilify Maintena supported approval for a

long-acting, once-monthly, injectable formulation of aripiprazole. One study established the

effectiveness of aripiprazole (Abilify Maintena) in a longer-term maintenance study. The second

study established effectiveness of aripiprazole (Abilify Maintena) in a short-term study of adult

patients with schizophrenia experiencing an acute relapse.61 FDA awarded exclusivity in both

cases.

That exclusivity protects certain conditions of approval for which Alkermes is seeking approval:

treatment of schizophrenia using a once-monthly, long-acting, injectable formulation of

aripiprazole. Because this exclusivity was granted prior to any determination regarding approval

of the Alkermes 505(b)(2) NDA, FDA is barred from approving the Alkermes NDA for these

protected conditions of approval until this exclusivity expires. As confirmed by the district court

in Veloxis Pharmaceuticals, Inc. v. FDA, it is irrelevant that the Alkermes NDA “relies on” the

Abilify tablet, and not Abilify Maintena. The dispositive point is that the conditions of approval

for Abilify Maintena overlap with the conditions of approval for the Alkermes NDA. That

overlap bars FDA from granting final approval to the Alkermes NDA. Nor is it of any

consequence that the Alkermes NDA is an NCE.

i. Alkermes cannot avoid Otsuka’s three-year exclusivity by relying on

Otsuka’s Abilify tablet and ignoring the relevant conditions of

approval for Abilify Maintena.

59 Guidance for Industry: Applications Covered by Section 505(b)(2), Draft Guidance, at 3 (Oct. 1999). 60 Id. at 5. 61 See NDA No. 202971.

10

Alkermes cannot avoid the three-year exclusivity protecting Abilify Maintena’s conditions of

approval simply by “referring to” the Abilify tablet. It is immaterial for exclusivity purposes

whether a 505(b)(2) application specifically “refers to” a protected drug product (i.e., Abilify

Maintena) for that drug’s exclusivity to apply.

Under the plain language of Section 505(c)(3)(E)(iii), entitlement to three-year exclusivity

requires (1) submission of a 505(b) NDA; (2) a drug that contains an active ingredient approved

after 9/1984; and (3) at least one new clinical investigation, excluding bioavailability studies,

that is essential to the conditions of approval of the 505(b) drug. Under the plain language of

Section 505(c)(3)(E)(iv), entitlement to three-year exclusivity requires (1) submission of a

supplement to a 505(b) NDA that is approved after 9/1984, and (2) at least one new clinical

investigation, excluding bioavailability studies, that is essential to the conditions of approval of

the 505(b) drug.

Once exclusivity is granted to the first-in-time 505(b) NDA because those preconditions are met,

such as in the case of the Abilify Maintena NDA/sNDA, FDA “may not make the approval of an

application submitted under subsection (b) of this section for the conditions of approval of such

drug in the approved subsection (b) application” or “for a change approved in the supplement,”

unless the later NDA is a full standalone 505(b)(1) NDA.

Simply put, FDA may not approve a second-in-time 505(b)(2) NDA that shares “conditions of

approval” with the first-in-time 505(b) drug. The “change” referred to in 505(c)(3)(E)(iv) is

simply a “change” in the “conditions of approval.” 62 The district court in Veloxis

Pharmaceuticals, Inc. v. FDA recognized this point, analogizing 505(c)(3)(E)(iii) to

505(c)(3)(E)(iv).63

As a matter of law, it is irrelevant to this analysis upon what drug the second-in-time 505(b)

applicant relies. The court in Veloxis rejected this very argument based on the text of the

exclusivity statute. Veloxis argued that Astagraf XL’s three-year exclusivity did not apply

because “the Envarsus XR NDA ‘did not rely upon any of the studies or data supporting

approval of Astagraf XL or upon [the] FDA’s prior findings that Astagraf XL is safe and

effective.’”64 The court held that FDA’s withholding of final approval to the Envarsus XR NDA

was proper because the Envarsus XR NDA overlapped with the Astagraf XL conditions of

62 See Xalatan Response, at 3-4 (“Of particular relevance to this response is section 505(c)(3)(E)(iv) of the Act,

which grants 3-year exclusivity to an applicant if an sNDA is approved and contains reports of new clinical

investigations (other than bioavailability studies) that are essential to the approval of the sNDA and were conducted

or sponsored by the applicant. If 3-year exclusivity is granted, FDA may not approve for 3 years a 505(b)(2)

application or an abbreviated new drug application (ANDA) for the same conditions of approval as those approved

in the sNDA.” (emphasis added)); see also 54 Fed. Reg. 28872, 28897 (July 10, 1989) (proposed rule) (“[T]he two

provisions that confer exclusivity on changes in already approved drugs delay the effective date of approval of all

ANDA’s and 505(b)(2) applications that have the same ‘conditions of approval’ as the innovator’s drug . . . .”); id. at

28901 (“The exclusivity provisions of section 505(c)(3)(D)(iii) and (iv) [now (E)] of the act delay the effective date

of approval of any 505(b)(2) application that is for the conditions of use of a previously approved application . . . .”). 63 Veloxis, 2015 U.S. Dist. LEXIS 77559, at *40 & n.15. 64 Id. at *26.

11

approval, even though the Envarsus XL relied on Prograf.65 The court held that the plain text of

the statute compelled this conclusion: “Exclusivity under [Section 505(c)(3)(E)(iii)] is triggered

by an overlap in the conditions of approval between the first-in-time 505(b) drug and the second-

in-time 505(b)(2) NDA . . . .” 66 As the court noted, FDA’s regulations confirm this

interpretation.67

The three-year exclusivity statutory provisions do not contain any language requiring a sponsor

to “refer to” a specific drug product in order for that product’s exclusivity to apply. And FDA

itself has confirmed that a second-in-time 505(b) applicant need not “rely on” a specific drug

product (i.e., “a finished dosage form, for example, [a] tablet . . ., that contains a drug

substance,” such as the Abilify tablet or Abilify Maintena injectable)68 to be barred by a first-in-

time 505(b)’s three-year exclusivity. FDA interprets the term “drug” in the bar clause of

statutory exclusivity provisions to mean drug substance – i.e., “an active ingredient that is

intended to furnish pharmacological activity or other direct effect in the diagnosis, cure,

mitigation, treatment, or prevention of disease or to affect the structure or any function of the

human body,”69 such as aripiprazole.70

FDA has concluded that, “when exclusivity attaches to an active moiety [i.e., aripiprazole] or to

an innovative change in an already approved drug [i.e., the Abilify Maintena NDA/sNDA], the

submission or effective date of approval of . . . 505(b)(2) applications for a drug with that active

moiety or innovative change will be delayed until the innovator’s exclusivity has expired,

whether or not FDA has approved subsequent versions of the drugs entitled to exclusivity, and

regardless of the specific listed drug product to which the . . . 505(b)(2) application refers.71

65 Id. at *30-31. Regarding similarly structured orphan drug exclusivity provisions, the United States District

Court for the District of Columbia recently ruled that the relevant statutory provisions were structured formulaically

and thus acted as a “limitation on the agency’s authority” and “a restriction of the FDA’s ability to approve the

marketing of other such drugs for the same rare disease or condition.” Depomed, Inc. v. U.S. Dep’t of Health &

Human Servs., No. 12-cv-1592 (KBJ), 2014 U.S. Dist. LEXIS 126235, *30-31 (D.D.C. Sept. 5, 2014). Due to their

similar structure, these observations apply equally to the three-year exclusivity provisions, which provide for no

“reference” or “reliance” on the part of the subsequent applicant. 66 Veloxis, 2015 U.S. Dist. LEXIS 77559, at *30-31. 67 Id. at *33-34. The use of the term “relies on” in 21 C.F.R. § 314.108(b)(4)(iv) can only refer to an ANDA

submitted pursuant to a suitability petition that “relies on” the information in the original application. Id. 68 21 C.F.R. § 314.3(b). 69 Id. 70 See, e.g., 54 Fed. Reg. at 28897. If “drug” in the bar clause were interpreted to mean drug product, three-year

exclusivity, which does not require reference to a specific product and thus blocks ANDAs or 505(b)(2) applications

that do not reference the drug product, would be broader than five-year exclusivity, which does require reference

and thus would block only those applications that reference the product. The agency has said, “FDA does not

believe Congress intended “the exclusivity provisions to operate inconsistently.” Id. Consequently, three-year

exclusivity covers “changes in non-new chemical entities rather than covering only specific drug products.” 71 Id. (emphasis added). FDA’s interpretation about the irrelevance of reliance to exclusivity determinations is

not inconsistent with FDA’s interpretation about patent certification being dependent on reliance at the discretion of

the sponsor. See, e.g., Letter from Janet Woodcock, M.D., Director, Center for Drug Evaluation and Research, to

David B. Clissold, Hyman, Phelps & McNamara, P.C. (Sept. 18, 2013) (“Suboxone Response”); see also Letter

from Steven K. Galson, M.D., M.P.H., Acting Director, Center for Drug Evaluation and Research, to Donald O.

Beers, Arnold & Porter, LLP, and William F. Cavanaugh, Jr., Patterson, Belknap, Webb & Tyler, LLP, at 7-8 (Nov.

30, 2004) (“Patent certification obligations . . . are linked to identification of the listed drug or drugs on which the

application relies.”). Regarding patents, FDA has explained that it is appropriate for sponsors to make their own

12

FDA reinforced this interpretation by explaining:

The exclusivity provisions of sections 505(c)(3)([E]) (iii) and (iv) of the act delay

the effective date of approval of any 505(b)(2) application that is for the

conditions of use of a previously approved application that contained new clinical

investigations essential for approval. Consequently, if two 505(b)(2) applications

are under review at the same time and one is approved before the other, the

effective date of approval of the second application to be approved will be

delayed, regardless of the date of submission, if the first contained new clinical

investigations essential for approval and thereby qualified for exclusivity.72

The agency’s recent guidance reflects this point.73 By reiterating that the agency interprets the

term “drug” to mean “drug substance” in the bar clauses of both the three- and five-year

exclusivity provisions, FDA has recently affirmed that the applicability of three-year exclusivity

does not require reference to or reliance on a specific drug product.74

Alkermes’s attempt to rely on the Abilify tablet NDA and not on data from Otsuka’s Abilify

Maintena NDA and sNDA makes no difference. Otsuka was the first to obtain three-year

exclusivity protecting long-acting, monthly injectable formulations of aripiprazole for the

treatment of schizophrenia with conditions of approval for both maintenance and acutely

relapsing patients, and exclusivity has attached to an “active moiet[y] in [a] new chemical

entit[y] or [a] change[] in [a] non-new chemical entit[y].” Though the Alkermes NDA may not

specifically refer to Abilify Maintena, Alkermes admits that aripiprazole lauroxil converts to

aripiprazole in the body. 75 Even assuming Alkermes can independently demonstrate

effectiveness of its own drug product with one separate clinical investigation, this does not

diminish Otsuka’s exclusivity rights nor reduce or eliminate that exclusivity as a bar to approval.

That bar is in effect until December 5, 2017.

ii. Exclusivity of aripiprazole encompassing long-acting, monthly,

injectable formulations for treatment of schizophrenia protecting

conditions of approval in both maintenance and acutely relapsed

patients blocks approval of aripiprazole lauroxil.

FDA has interpreted “conditions of approval” entitled to three-year exclusivity under Section

505(c)(3)(E)(iii) and (iv) to be derived from the new clinical investigations conducted in support

reliance/certification decisions because, ultimately, sponsors of RLDs have a potential remedy through patent

litigation. 72 54 Fed. Reg. at 28901. 73 Guidance for Industry: New Chemical Entity Exclusivity Determinations for Certain Fixed-Combination

Drug Products, at 6 (Oct. 2014) (“[T]he Agency interpreted the term drug in the bar clause to mean drug

substance.”). 74 Id. 75 Poster from Alkermes Presentation on Results of Phase 3 Study for Aripiprazole Lauroxil in Schizophrenia at

the American Society of Clinical Pyschopharmacology Annual Meeting (June 17, 2014).

13

of the NDA or sNDA. 76 The court in Veloxis affirmed this interpretation. Under this

interpretation, the Alkermes NDA is seeking approval for conditions of approval protected by

three-year exclusivity, and therefore, final approval to that NDA should be withheld.

The new clinical investigations that were essential to the FDA’s approval of Abilify Maintena

and that led to three-year exclusivity were (1) a short-term, randomized, double-blind, placebo-

controlled trial in acutely relapsed adults (Protocol 31-12-291), and (2) a longer-term, double-

blind, placebo-controlled, randomized withdrawal (maintenance) trial in adults (Protocol 31-07-

246). The new investigations supported approval for a long-acting, once-monthly, injectable

formulations of aripiprazole for treatment of schizophrenia as demonstrated in both a longer-term

maintenance clinical trial and in a short-term trial in acutely relapsed patients. Those conditions

of approval have each received independent three-year exclusivity.

Alkermes seeks approval for the protected conditions of approval. Alkermes describes

aripiprazole lauroxil as having the same key characteristics approved in the Abilify Maintena

NDA and sNDA labeling: (1) “the active moiety is aripiprazole”77 in (2) a “long-acting,”78 (3)

“injectable” 79 (4) “once-monthly” 80 formulation for (5) “treatment of schizophrenia” 81 in

“patients experiencing [an] acute exacerbation.”82 The short-term clinical trials conducted by

Alkermes and Otsuka to support the applications discussed in this petition were remarkably

similar. Both studies were designed as placebo-controlled studies in an acutely relapsed

population to show the effectiveness of a long-acting, once-monthly, injectable formulation that

releases aripiprazole for treatment of schizophrenia. Notably, according to clinicaltrials.gov,

Alkermes has not conducted a longer-term, double-blind, placebo-controlled randomized

withdrawal (maintenance) trial.

A comparison of the study designs, patient populations and outcomes are shown in the table

below. These studies are analogous to the standard six-week studies that are the standard

protocols for FDA approval of oral medications. The goal is to demonstrate that the

investigational drug can improve symptoms in patients with an exacerbation of their

schizophrenia to a statistically significantly greater extent than can placebo. The studies below

were twelve weeks in duration to allow for three monthly injections in these patients; other long-

acting, injectable formulations of antipsychotic products used similar designs in the past (e.g.,

Invega Sustenna (9-13 week studies) and Risperdal Consta (12 week studies)).

76 Veloxis, 2015 U.S. Dist. LEXIS 77559, at *39 (“The parties agree that the scope of Astagraf XL’s marketing

exclusivity is limited to the ‘conditions of approval’ based upon the ‘new clinical investigations’ that were

conducted in support of the Astagraf XL NDA.”); FDA’s Reply Mem. In Supp. of Defs.’ Mot. For Summ. J. at 3,

AstraZeneca Pharms. LP v. FDA, 872 F. Supp. 2d 60 (D.D.C. 2012) (“FDA interprets [Section 505(c)(3)(E)(iv)] to

require a relationship between the new clinical investigation, the change to the product or to the use of the product

approved in the supplement, and the scope of any resulting three-year exclusivity.”); Seroquel Letter (the scope of

marketing exclusivity is determined by the nexus between the clinical trials and the change approved in the

application). 77 See Richard Pops at Goldman Sachs Healthcare Conference. 78 Alkermes News Release, Oct. 22, 2014. 79 Id. 80 Id. 81 Id. 82 Alkermes News Release, Apr. 8, 2014.

14

Aripiprazole Once-Monthly in the

Acute Treatment of Schizophenia

Aripiprazole Lauroxil in Acute

Exacerbation of Schizophrenia

Study design Randomized, double-blind, placebo-

controlled

Randomized, double-blind, placebo-

controlled

Study duration 12 Weeks 12 Weeks

Patient population Patients with schizophrenia

experiencing an acute psychotic episode

at screening and baseline

Patients with schizophrenia currently

experiencing an acute relapse

PANSS* entry criteria Total score >= 80 Total score of 70-120

PANSS baseline score 103-104 92-94

CGI-S** baseline score 5.2 4.9

Primary clinical endpoint Change from baseline in PANSS total

score

Change from baseline in PANSS total

score

Number of

patients/experimental

treatment

168-172 207-208

Drug vs. placebo

improvement in PANSS total

score

Drug = 27

Placebo = 13

Drug = 21

Placebo = 10

References Kane et al.

Aripiprazole once-monthly in the acute

treatment of schizophrenia: Findings

from a 12-week, randomized, double-

blind, placebo-controlled study.

J Clin Psychiatry 2014;75(11):1254–

1260

Meltzer et al.

A randomized, double-blind, placebo-

controlled trial of aripiprazole lauroxil in

acute exacerbation of schizophrenia.

J Clin Psychiatry On-line publication.

* PANSS is the Positive and Negative Syndrome Scale, a 30-item scale that measure the symptoms of schizophrenia with a 7-

point assessment (ranging from a score of 1 (symptom absent) to a score of 7 (symptom extreme)); PANSS total score can range

from 30-210, with a score of 90-100 indicating moderately severe symptomatology.

** CGI-S is the Clinical Global Impressions – Severity of Illness scale, which is a 7-point scale to measure the overall severity of

illness in a patient; Scores include: 1-not ill, 2-minimally or borderline ill, 3-mildly ill, 4-moderately ill, 5-markedly ill, 6-

severely ill, 7-extremely ill.

15

Because Abilify Maintena and the Alkermes NDA share conditions of approval based on their

similar clinical trials, complete and final approval of the Alkermes NDA must be delayed until

Otsuka’s three-year exclusivity expires. Unlike the Veloxis case, there is nothing else left for the

Alkermes NDA to be immediately approved for prior to the expiration of Otsuka’s exclusivity.83

iii. It is of no matter if FDA determines the Alkermes NDA is an NCE.

Any conclusion that the Alkermes NDA involves a NCE is irrelevant here. Under the plain text

of the statute, an NCE is subject to a first-in-time 505(b)’s three-year exclusivity when it shares

“conditions of approval.” As explained above, once exclusivity is granted to the first-in-time

505(b) NDA, FDA “may not make the approval of an application submitted under subsection (b)

of this section for the conditions of approval of such drug in the approved subsection (b)

application” (505(c)(3)(E)(iii)) or “of an application submitted under subsection (b) of this

section for a change approved in the supplement” (505(c)(3)(E)(iv)).

Plainly, Section 505(c)(3)(E)(iv) does not require that a second-in-time 505(b)(2) application

contain the same active moiety as a product protected by three-year exclusivity to be barred by

that NDA’s exclusivity. Instead, the provision prohibits FDA from approving a 505(b)(2)

application for “a change” (i.e., condition of approval) approved in an sNDA.

Nor does Section 505(c)(3)(E)(iii). While that provision suggests that there must be a

relationship between “such drug in the [first-in-time 505(b) application]” and the drug in the

subsequent 505(b)(2) application for it to be barred from approval, there is no requirement that

the first-in-time 505(b) application contain the same active moiety as the second-in-time

505(b)(2) application. As explained above, FDA has interpreted “drug” to mean drug substance,

and FDA cannot approve a second-in-time 505(b)(2) application “for the conditions of approval

of such drug in the approved subsection (b) application.” In other words, FDA cannot approve a

second-in-time 505(b)(2) application for the protected conditions of approval of aripiprazole.

Here, Alkermes relies on the safety and effectiveness findings of aripiprazole in its 505(b)(2)

application and has admitted aripiprazole lauroxil ultimately converts to aripiprazole in the

body.84

Indeed, a drug for which investigations are used to meet the requirements of Section 505(b)(1)

must be the same drug for determining applicable exclusivity under Sections 505(c)(3)(E)(iii)

and (iv).85 Studies on drugs used to support an NDA must be the same drug as the drug in the

83 In Veloxis, the agency could approve Envarsus XR for an indication that would cover conversion patients

because Astagraf XL did not have exclusivity for that condition of approval. 84 See, e.g., 54 Fed. Reg. at 28897 (exclusivity covers “changes in non-new chemical entities rather than

covering only specific drug products”). 85 This is consistent with FDA’s Proposed Rule regarding exclusivity, where FDA stated that: “[a 505(b)(2)

application] for a new chemical entity would not be subject to any patent protection or exclusivity accorded a

previously approved drug, because, by definition, there will be no applicable previously approved drug.” 54 Fed.

Reg. at 28890. At the time of this statement FDA had contemplated only literature-based NCE 505(b)(2)

applications for which there would be no applicable previously approved product. Accordingly, now that FDA has

permitted the submission of NCE 505(b)(2) applications that rely upon investigations of previously approved drugs,

so too must the exclusivity protections provided in the Hatch-Waxman amendments apply. Indeed, patent

16

NDA and must be the same drug for purposes of exclusivity. Those provisions of the FFDCA

must be read in harmony.86

Section 505(b)(2), which Alkermes relies on here, instructs that “[a]n application submitted

under paragraph (1) for a drug for which the investigations described in clause (A) of such

paragraph and relied upon by the applicant for approval” must include certain certifications or

statements. The clear antecedent in 505(b)(2) of “the investigations described in clause (A)” is

the drug in the current application. The use of singular “drug” means that there is only one

“drug,” and the definition includes any other “drug for which the investigations described in

clause (A)” were submitted. In this case, since studies on both aripiprazole and aripiprazole

lauroxil were submitted with Alkermes’s 505(b)(2) NDA, they are both considered to be the

same “drug” for the purposes of 505(b)(2).

“Such drug” for purposes of 505(c)(3)(E)(iii) must, in turn, be the same drug. The language of

Section 505(c)(3)(E)(iii) instructs that FDA cannot approve a second-in-time 505(b) applicant

“for the conditions of approval of such drug in the [first-in-time 505(b) application].” A

505(b)(2) applicant that relies on a “drug” for purposes of meeting the two AWCT requirement

must be the same “such drug” for purposes of exclusivity. The contrary interpretation goes

against the underlying goals of the law, here by allowing immediate approval of a drug like

aripiprazole lauroxil. The law operates consistent with its underlying goals only if a drug

substance is subject to the exclusivity of the similar drug on which it can rely for approval,

without loopholes. Therefore, the definition of “drug” must incorporate aripiprazole and

aripiprazole lauroxil.

Even if FDA technically deems aripiprazole lauroxil to be an NCE by its bright-line structural

rule (but allows Alkermes to rely on Otsuka’s clinical data), this fact cannot mean that the

Alkermes NDA is immune from the blocking effects of Otsuka’s three-year exclusivity for

conditions of use that are the same in all material respects.

This interpretation is consistent with prior FDA precedent. FDA has interpreted the applicability

of three-year exclusivity to mean that proposed drug products with similar, though not

chemically identical, molecular structures may be blocked by three-year exclusivity. 87 The

Alkermes NDA should not be approved until Otsuka’s three-year exclusivity expires.

certifications have been submitted for different NCE 505(b)(2) applications. See, e.g., NDA No. 022399 (Horizant)

and NDA No. 021667 (Nutrestore). 86 In King v. Burwell, for example, the Supreme Court looked to the larger context of the Patient Protection and

Affordable Care Act to understand the meaning of the phrase “an Exchange established by the State.” No. 14-114,

slip op. at 9-10 (U.S. June 25, 2015). Specifically, the Court looked to the underlying purpose and mechanics of the

law as well as to the provision establishing the Federal Exchange, which uses the term “such Exchange,” to

conclude that the phrase “an Exchange established by the State” includes Exchanges established by the federal

government. Id. (“In other words, State Exchanges and Federal Exchanges are equivalent—they must meet the

same requirements, perform the same functions, and serve the same purposes.”). 87 FDA’s decision set forth in the Xalatan response confirms that an NCE may be blocked by another product’s

exclusivity despite having a different active moiety. FDA made clear that the three-year exclusivity awarded to

Pfizer for new clinical studies establishing that Pfizer’s Xalatan (latanoprost) product could be used as a “first-line

indication” precluded the marketing of Lumigan (bimatoprost) and Tratavan (travoprost) for a first-line indication.

According to FDA, during Xalatan’s period of three-year exclusivity, “only Pfizer could label its product for first-

17

b. The Alkermes NDA does not satisfy the requirements for demonstration of

clinical effectiveness.

i. The Alkermes NDA includes only a single AWCT and, therefore, it

cannot be approved for it fails to satisfy the requirements for

demonstrating clinical effectiveness.

Otsuka’s first petition, which is here adopted and incorporated by reference, requested that FDA

not accept the Alkermes NDA for scientific review. That request, like the request in the present

petition, was based on the fact that the Alkermes NDA is not eligible for approval as a matter of

law because the NDA is not supported by two AWCTs of the active moiety for which approval is

sought (aripiprazole lauroxil), nor does it meet any of the established exceptions to the statutorily

prescribed demonstration of effectiveness by two AWCTs.88

The safety/effectiveness standard for approving an NDA is higher than the standard for accepting

an NDA for review, which was the action requested in the first petition.89 While Otsuka objected

to FDA’s acceptance of the Alkermes NDA for review because it failed to satisfy the two AWCT

requirement (and Otsuka persists in that objection), acceptance for review and final approval are

qualitatively different. The former gets a product “in the door” for regulatory review while the

latter is the FDA “stamp of approval” that allows the product on the market accompanied by the

attendant risk if the product is not demonstrably safe.

Sections 505(b)(1)(A) and 505(d) speak in plurals, indicating that a single AWCT generally does

not suffice.90 FDA has made clear that it shares this view of what the statute requires: “FDA’s

position [is] that Congress generally intended to require at least two [AWCTs].”91 The scientific

policy reason for adherence to this requirement is, as FDA’s guidance makes clear, strong: the

risk of unreliability – for reasons of bias or otherwise – of a single, unreplicated study. “The

usual requirement for more than one adequate and well-controlled investigation reflects the need

for independent substantiation of experimental results.”92 Accordingly, the exceptions to this

line use for reduction of elevated IOP [inter-ocular pressure] in patients with open-angle glaucoma or ocular

hypertension.” Xalatan Response, at 25. FDA took this position even though Travatan and Lumigan were approved

as NCEs. 88 Ex. 1, at 15 (“This petition formally requests only that FDA refuse to file the Alkermes NDA. However, in

the event that FDA accepts the Alkermes NDA for filing and thereby implicitly denies this citizen petition, the

Alkermes NDA must not be approved for all of the reasons reviewed herein, which are incorporated here by

reference.”). 89 Compare 21 U.S.C. § 355(d)(1) (“If the Secretary finds . . . that the investigation, reports of which are

required to be submitted to the Secretary pursuant to subsection (b) of this section, do not include adequate tests by

all methods reasonably applicable to show whether or not such drug is safe for use under the conditions prescribed

recommended or suggested in the proposed labeling thereof . . . he shall issue an order refusing to approve the

application.” (emphasis added)), with 21 C.F.R. § 314.101(d)(3) (“FDA may refuse to file an application . . . if . . .

[t]he application . . . is incomplete because it does not on its face contain information required under section 505(b) .

. . .”). 90 21 U.S.C. § 355(b)(1)(A), (d). The limited exception to this rule does not apply here. 91 Clinical Effectiveness Guidance, at 3. 92 Id. at 4.

18

rule must be very narrowly construed and applied only in very rare cases, which FDA has

indicated is its policy.93

As explained in Otsuka’s first citizen petition, the Alkermes NDA must contain reports from at

least two AWCTs of the proposed active moiety. Under FDA’s bright-line structural rule

defining an NCE, due to the potential for unexpected clinical effects from the chemical

differences between two molecules based on covalent bonds, aripiprazole lauroxil and

aripiprazole contain different active moieties and aripiprazole lauroxil is deemed to be an NCE

by FDA.94 As a matter of law, an NDA for an NCE must generally contain full reports of two

clinical trials, and a single AWCT of an NCE cannot and should not suffice as independent

substantiation for filing or approval purposes.95

The process that the drug aripiprazole lauroxil undergoes before emerging as aripiprazole is

complex and not well-characterized with intermediates formed along the way.96 This complex

process is entirely absent when simply injecting aripiprazole into patients, and how it affects

safety and efficacy requires rigorous independent investigation. It is exactly for such reasons

that FDA’s bright-line rule exists, and why drugs on aripiprazole lauroxil’s side of this bright-

line rule are deemed to be NCEs with all attendant benefits and burdens. One such burden is the

requirement that, other than in extraordinary circumstances not applicable here, a sponsor of an

application for an NCE must demonstrate independent substantiation of experimental results

through at least two AWCTs investigating the NCE.

Otsuka’s first citizen petition also explained why the Alkermes NDA does not meet either of the

two exceptions that would permit approval on the basis of a single AWCT.97 First, aripiprazole

lauroxil is not a new use of an active moiety contained in a previously approved drug product.

Second, the single AWCT of aripiprazole lauroxil demonstrated symptomatic benefit rather than

reduced morbidity/mortality or prevention of disease where repetition of trial results would pose

ethical concerns. Nor can safety and effectiveness of this NCE be demonstrated under 505(b)(2),

because data to support an NCE 505(b)(2) application “is likely to be derived from published

studies, rather than FDA’s previous finding of safety and effectiveness of a drug,” and there have

been no other clinical trials of aripiprazole lauroxil independently reported in published

literature.98

93 See id. at 7-16. These exceptions are when there is confirmatory evidence regarding the effectiveness of the

specific drug substance in the application or when the study meets certain factors regarding size or trial design that

reduce the need for confirmatory evidence. This second exception is reserved for those rare circumstances when the

“trial has demonstrated a clinically meaningful effect on mortality, irreversible morbidity, or prevention of a disease

with a potentially serious outcome and confirmation of the result in a second trial would be practically or ethically

impossible.” Id. at 13 (emphasis added). 94 Ex. 1, at 9-10. 95 Id. at 10-11. 96 See Morten Rohde et al., Biological Conversion of Aripiprazole Lauroxil - An N-acyloxymethyl Aripiprazole

Prodrug, Results in Pharma Sciences 4:19-25 (2014). 97 Ex. 1, at 12-13. 98 Ex. 1, at 6, 12; Guidance for Industry: Applications Covered by Section 505(b)(2), Draft Guidance, at 3 (Oct.

1999).

19

For all of the reasons stated above and in the first petition, the Alkermes NDA should not be

approved because it is based on reports of only a single AWCT.

ii. Assuming, without conceding, that FDA could, under the facts of this

matter, approve the Alkermes NDA based upon a single AWCT, FDA

is without authority to take that action absent adoption of a regulation

in accordance with the rulemaking requirements of the

Administrative Procedure Act.

If FDA now proposes to deviate from its settled interpretations regarding the necessity of two

AWCTs and to adopt a new interpretation of the statutory requirement, FDA would first need to

comply with the notice-and-comment rulemaking requirements of the Administrative Procedure

Act because that “interpretation” would be a legislative rule. Were FDA to apply an

interpretation of its statute and regulations to the Alkermes NDA that permits its approval with

only a single AWCT, FDA’s action would be unlawful as it would constitute the adoption of a

new legislative rule – one that has “the force and effect of law” and “creates new law or imposes

new rights or duties” 99 – that was adopted without complying with notice-and-comment

rulemaking.100

V. Conclusion

For the reasons and based on the legal authority cited above and in its Otsuka’s first petition,

FDA should grant Otsuka’s request and refuse approval of or delay approval of the Alkermes

NDA.

C. ENVIRONMENTAL IMPACT

Petitioner claims a categorical exclusion under 21 C.F.R. § 25.31.

D. ECONOMIC IMPACT

Petitioner will, upon request by the Commissioner, submit economic impact information in

accordance with 21 C.F.R. § 10.30(b).

E. CERTIFICATION

I certify that, to my best knowledge and belief: (a) this petition includes all information and

views upon which the petition relies; (b) this petition includes representative data and/or

99 Jerri’s Ceramic Arts, Inc. v. Cons. Prod. Safety Comm’n, 874 F.2d 205, 207 (4th Cir. 1989) (quoting Nat’l

Latino Media Coalition v. FCC, 816 F.2d 785, 788 (D.C. Cir. 1987); Southern California Edison Co. v. FERC, 770

F.2d 779, 783 (9th Cir. 1985); General Motors Corp. v. Ruckelshaus, 742 F.2d 1561, 1562 (D.C. Cir. 1984)); Scenic

Am., Inc. v. U.S. Dep’t of Transp., 49 F. Supp. 3d 53, 61 (D.D.C. 2014) (a rule is substantive, among other things,

“[i]f in the absence of the rule there would not be an adequate legislative basis for enforcement action or other

agency action to confer benefits or ensure the performance of duties”) . 100 Courts have not hesitated to invalidate FDA’s actions when the agency failed to comply with its rulemaking

obligations. See, e.g., Syncor Int’l Corp. v. Shalala, 127 F.3d 90, 95-96 (D.C. Cir. 1997).

information known to the petitioner which are unfavorable to the petition; and (c) I have takenreasonable steps to ensure that any representative data and/or information which are unfavorableto the petition were disclosed to me. I further certify that the information upon which I havebased the action requested herein first became known to the party on whose behalf this petition issubmitted on or about the following dates: July 26, 2012 (Alkermes announcement of initiationof Phase 3 clinical trial of aripiprazole lauroxil), August 25, 2014 (Alkermes announcement ofsubmission of NDA to FDA based on single Phase 3 clinical trial), December 5, 2014 (approvalof sNDA for Abilify Maintena), January 16, 2015 (date of Orange Book update reflecting sNDAthree-year exclusivity), and June 16, 2015 (filing of memorandum opinion in VeloxisPharmaceuticals, Inc. v. United States Food and Drug Administration). If I received or expectto receive payments, including cash and other forms of consideration, to file this information orits contents, I received or expect to receive those payments from the following persons ororganizations: Otsuka Pharmaceutical Development &Commercialization, Inc. I verify underpenalty of perjury that the foregoing is true and correct as of the date of the submission of thispetition.

Respectfully submitted,

OTSUKA PHARMACEUTICAL CO., LTD.

By: William H. Carson, M.D.President &CEOOtsuka Pharmaceutical Development &Commercialization, Inc.

Cc: Elizabeth H. Dickinson, Esq., Chief Counsel to the FDAJanet Woodcock, M.D., Director, Center for Drug Evaluation and Research, FDA

20

21

EXHIBIT 1

Otsuka

September 9, 2014

Division of Dockets ManagementFOOD AND DRUG ADMINISTRATION

Department of Health and Human Services5630 Fishers LaneRoom 1061, HFA-305Rockville, Maryland 20852

Dear Madam or Sir:

William H. Carson, M.D.President &CEO

Otsuka Pharmaceutical Development &Commercialization, Inc.

CITIZEN PETITION

Otsuka Pharmaceutical Development &Commercialization, Inc., on behalf of its parentcompany, Otsuka Pharmaceutical Co., Ltd. ("Otsuka"), hereby submits this petition undersection 505(q) of the Federal Food, Drug, and Cosmetic Act ("FFDCA"), in accordance with 21C.F.R. § 10.30, to request that the Commissioner of Food and Drugs refuse to accept forsubstantive review a New Drug Application ("NDA") submitted by Alkermes plc ("Alkermes")for approval of aripiprazole lauroxil because the application, based on public representations byAlkermes, is deficient on its face.

Otsuka holds an approved NDA for Abilify~ (aripiprazole),~ an atypical antipsychotic indicatedfor treatment of schizophrenia, and several other psychiatric indications. Otsuka also holds anapproved NDA for Abilify Maintena~ (aripiprazole),Z along-acting formulation for treatment ofschizophrenia. On August 25, 2014, Alkermes announced the submission of an NDA foraripiprazole lauroxil (the "Alkermes NDA"~, which Alkermes proposes as along-actingformulation for treatment of schizophrenia. Alkermes seeks FDA approval based only on resultsfrom a purported single adequate and well-controlled clinical trial ("AWCT") of aripiprazolelauroxil despite the fact that aripiprazole lauroxil is a New Chemical Entity ("NCE") accordingto an FDA bright-line rule, which establishes when an active moiety is deemed to be an NCEbased on chemical structure.4

~ NDA No. 021436.2 NDA No. 202971.3 Press Release, Alkermes Submits New Drug Application to FDA for Aripiprazole Lauroxil for Treatment of

Schizophrenia (Aug. 25, 2014), available at http://phx.corporate-ir.net/phoenix.zhtml?c=92211&p=irol-newsArticle&ID=1960579&highlight (hereinafter "Alkermes News Release, Aug. 25, 2014").

4 As used throughout this petition, a New Chemical Entity or NCE shall have the meaning set forth in 21 C.F.R.§ 314.108(a), as interpreted and applied by FDA and as described below in Section Il.b.

Otsuka Pharmaceutical Development Sr Commercialization, Inc. 11 University Square Drive, Suite 500 •Princeton, NJ 08540 •Phone: 609-452-5663 •Email: william.carson~otsuka-us.com

A. ACTION REQUESTED

Otsuka respectfully requests that FDA refuse to accept for substantive review the AlkermesNDA, seeking approval for a new active moiety not contained in any previously-approved drugproduct, because the single AWCT reported in the application cannot satisfy the substantialevidence of effectiveness requirement prescribed in Section 505(b)(1)(A) of the FFDCA.

B. STATEMENT OF GROUNDS

I. Summary of Argument

The Alkermes NDA should not be accepted for filing because it does not satisfy the informationrequirements of Section 505(b) and thus cannot be ultimately approved. Specifically, theAlkermes NDA is deficient on its face because it cannot meet the statutory requirement fordemonstration of substantial evidence of effectiveness. First, FDA generally requires that NDAsprovide evidence of effectiveness based on at least two AWCTs of the proposed drug product toensure that the results are independently reproducible. However, the Alkermes NDA containsfull reports of only a single, new AWCT of axipiprazole lauroxil.

Second, the Alkermes NDA does not meet any of FDA's criteria for approving an NDA based ononly a single new AWCT. FDA has approved NDAs with reports of a single AWCT of a new useof an already-approved drug. However, according to FDA's bright-line rule, aripiprazole lauroxilis deemed to contain a new active moiety that has never been contained in a previously approveddrug product. Consequently, the single study of aripiprazole lauroxil is the only available studyof a drug with a new active moiety rather than a new use of an approved drug. In accordancewith the applicable FDA rule, having only a single study is not sufficient in this case to triggeracceptance for substantive review.

FDA has also approved new drugs based on a single new AWCT of a drug with a new activemoiety. However, FDA has narrowly limited this exception to proposed drug products with asingle AWCT that have demonstrated certain types of effects (such as reduction in mortality)where it would be unethical to require a second, confirmatory study. The Alkermes NDA,according to Alkermes's public statements, does not meet these criteria.

A decision that a single AWCT is sufficient to support approval of the Alkermes NDA in thisinstance would require an arbitrary and capricious break with prior precedent and a sharpdeparture from FDA's well-established interpretations of relevant statutes and regulations. Thepresent application does not warrant commitment of agency resources for a substantive reviewuntil the application is resubmitted with the legally necessary information as defined under theAct. Furthermore, this petition presents a clear legal question that does not require substantivescientific review for resolution. As described below, FDA can —and should —exercise itsdiscretion to refuse to accept for substantive review the Alkermes NDA because no amount ofsubstantive review can remedy the failure of the Alkermes NDA to meet applicable statutory andregulatory requirements.

2

II. Legal Background

a. Standard for Refusing to Accept Application For Review

FDA may refuse to accept an NDA for review if, among other reasons, the application "isincomplete because on its face it does not contain the information required under section505(b)."5 According to FDA, this includes: "Failure to include evidence of effectivenesscompatible with statute and regulations."6 One such failure is: "Presentation of a single adequateand well-controlled trial without adequate justification of why the single trial should be regardedas fulfilling the statutory requirement for substantial evidence of effectiveness."~

According to FDA, "RTF [Refusal-to-file] is an important regulatory tool to help CDER avoidunnecessary review of incomplete applications," because "[i]ncomplete applications can lead tomultiple-cycle reviews and inefficient use of CDER resources."g Moreover, "CDER alsobelieves an RTF action can allow an applicant to begin repair of critical deficiencies in theapplication far sooner than if these were identified much later in a complete response action andmay lead to more rapid approval of safe and effective drug products."9 Consequently, substantivereview of a deficient application would be a waste of time and resources for both the agency andthe applicant.

b. Standard for Distinguishing New Active Moieties

FDA defines an NCE as "a drug that contains no active moiety that has been approved by FDAin any other application submitted under section 505(b) of the act."10 An active moiety isdefined, in turn, as:

[T]he molecule or ion, excluding those appended portions of the molecule thatcause the drug to be an ester, salt (including a salt with hydrogen or coordinationbonds), or other non-covalent derivative (such as a complex, chelate, or clathrate)of the molecule, responsible for the physiological or pharmacological action ofthe drug substance. 1

According to this bright-line definition of active moiety, NCE status is determined exclusivelyon the basis of chemical structure rather than on a case-by-case analysis of pharmacologicalactivity. Thus, "an ester, salt ... , or other noncovalent derivative ... of [a] molecule" is deemedto be the same molecule (and thus same active moiety), whereas a molecule with anon-estercovalent bond is deemed to be a different molecule (and thus a new active moiety) — "even if the

5 21 C.F.R. § 314.101(d)(3).6 Manual of Policies and Procedures, Center for Drug Evaluation and Research, Office of New Drugs: Good

Review Practice: Refuse to File (Oct. 11, 2013) (hereinafter "RTF MAPP"), at 20.Id.

g Id., at 2.91d.10 21 C.F.R. § 314.108(a).~ 1 Id

molecule includes a covalent bond to a molecule that was itself previously an active moiety."12

In sum, "[i]f the drug substance is composed of non-ester covalently bonded molecules, thecovalently bonded molecule is considered the active moiety."13

FDA has provided the following scientific justification for its bright-line rule that treats covalentstructural changes to a molecule as an NCE:

[I]t has been FDA's longstanding experience that even minor covalent structuralchanges are capable of producing not only major changes in the activity of thedrug but changes that are not readily predicted. Because of their potentialsignificance, FDA has always identified changes in covalent structure, includingminor changes ... as sufficient to create a new "active moiety," and thereby tocreate a new chemical entity.

The potential significance of modifications in covalent structure, even wherepreviously approved drugs contain the same "active site," is reflected in theamount and kind of data required for approval of such changes. Such a changerequires submission of an amount of data comparable to that required for anentirely new molecule.la

FDA provides the following justification for treating the formation of a salt or ester of amolecule, or other non-covalent changes to a molecule, as the same active moiety:

In contrast to most changes in the covalent structure of a molecule, the formationof a salt or a complex, or of an ester, is not intended to, and generally cannot, alterthe basic pharmacologic or toxicologic properties of the molecule.ls

c. Standard for Demonstrating Clinical Safety and Effectiveness

NDA sponsors must submit "full reports of investigations which have been made to showwhether or not such drug is safe for use and whether such drug is effective in use."16 Section505(b)(1)(A) contemplates that the required reports of investigations of safety and effectivenesssupporting the approval of "such drug" will have been conducted by the sponsor or by anotherentity from whom the sponsor has obtained a right of reference. In contrast, Section 505(b)(2)makes reference to "an application ... for a drug for which the investigations described in clause(A) of such paragraph and relied upon by the applicant for approval of the application were not

lZ Letter from Gary J. Buehler, Office of Generic Drugs, FDA, to Chad A. Landmon (Oct. 23, 2009)(hereinafter "Vyvanse Citizen Petition Response"), at 12.

13 1d., at 11.la Id at 14 (quoting a July 26, 1989, Citizen Petition Response, Docket No. 1987P-0339, at 11-12).'S ld.16 FFDCA § 505(b)(1)(A). Although the term "such drug" is not precisely defined in section 505(b), it refers to

the particular "drug product" for which approval is sought (i.e., a finished dosage form (such as a tablet or capsule))that contains an active ingredient "intended to furnish pharmacological activity or other direct effect" on the humanbody. 21 C.F.R. § 314.3(b). It is clear from a plain reading of the statute that the antecedent of the term "such drug"in 505(b)(1)(A) is the term "any drug" in 505(b)(1) for which an application is filed.

4

conducted by or for the applicant and for which the applicant has not obtained a right ofreference or use from the person by or for whom the investigations were conducted."17 Undersection 505(b)(2), the source of the "full reports of investigations" may differ from thatcontemplated under section 505(b)(1). However, in both cases, the quantity and quality of therequired investigations of safety and effectiveness must meet the statutory standards of section505(b)(1)(A).

FDA shall not approve an application submitted under 505(b) if the investigations required under505(b)(1)(A) are inadequate to support a finding of safety, or provide evidence that the proposeddrug is not safe.18 Section 505(d)(5) of the FFDCA also establishes a "substantial evidence"standard for determining whether "such drug ...will have the effect it purports or is representedto have under the conditions of use prescribed, recommended, or suggested in the proposedlabeling." If these tests of safety and effectiveness are not met, the NDA cannot be approved.19

The "substantial evidence" standard for demonstrating effectiveness has been defined as"adequate and well-controlled investigations" to enable qualified experts to conclude "that thedrug will have the effect it purports."20 The phrase "adequate and well-controlled investigations"has generally been interpreted to mean at least two individual adequate and well-controlledhuman clinical trials, which FDA has deemed necessary to satisfy a basic principle of scientificinvestigations by ensuring that clinical trial results are reproducible.21

Although the "substantial evidence" approval standard generally requires at least two well-controlled clinical trials, a 1997 amendment to section 505(d) provides that "data from oneadequate and well-controlled clinical investigation and confirmatory evidence (obtained prior toor after such investigation)" may constitute "substantial evidence" of effectiveness in somecases.22 According to FDA, this statutory amendment merely confirmed prior FDA practice: "Inmaking this clarification, Congress confirmed FDA's interpretation of the statutory requirementsfor approval and acknowledged the Agency's position that there has been substantial progress inthe science of drug development resulting in higher quality clinical trial data."23

FDA has consistently identified only two circumstances in which a single, new AWCT may besufficient to show "substantial evidence" of effectiveness.24 The first is where "a single study ofanew use, with independent substantiation from study data in related uses, could provideevidence of effectiveness.i25 In other words, evidence from one or more AWCTs of a

" FFDCA § 505(b)(2).18 FFDCA § 505(d)(1)-(4).19 FFDCA § 505(d)(5).20 FFDCA § 505(d)(7).21 See, e.g., Guidance for Industry: Providing Clinical Evidence of Effectiveness for Human Drug and

Biological Products, May 1998, (hereinafter "Clinical Effectiveness Guidance") at 4 ("The usual requirement formore than one adequate and well-controlled investigation reflects the need for independent substantiation ofexperimental results. A single clinical experimental finding of efficacy, unsupported by other independent evidence,has not usually been considered adequate scientific support for a conclusion of effectiveness.").

ZZ 21 U.S.C. § 355(d)(7).Z3 Clinical Effectiveness Guidance, at 4.24 See generally Clinical Effectiveness Guidance.zs Id. , at 8.

previously-approved drug product with the same active moiety can count toward the two-studyrequirement for a proposed new use of that same active moiety (e.g., in a different dosageform).26 The second is where a single AWCT meets one or more factors related to the size of thestudy, multiple endpoints, consistent results across population subsets, and strength of statisticalresults (i.e., large and complex trials that provide a level of evidence comparable to two or moreindependent studies).27 However, FDA has stated that this second exception "will generally belimited to situations in which a trial has demonstrated a clinically meaningful effect on mortality,irreversible morbidity, or prevention of a disease with potentially serious outcome andconfirmation of the result in a second trial would be practically or ethically impossible."28

d. Demonstrating Safety and Effectiveness of NCEs Under Section 505(b)(2)

FDA distinguishes between applications that are submitted under Section 505(b)(2) as amodification to a previously approved drug and those submitted as an NCE.29 For an applicationseeking approval of a modification to a previously approved drug, the sponsor may rely, in part,on FDA's finding of safety and effectiveness of the previously approved product, "coupled withthe information needed to support the change from the approved product."30 In other words,reliance on previously-approved NDAs to support a finding of safety and effectiveness isappropriate where the new drug product and reference drug product contain the same activemoieties.

Although an NCE may also be approved under Section 505(b)(2), FDA states that data to supportsuch an application "is likely to be derived from published studies, rather than FDA's previousfinding of safety and effectiveness of a drug."31 This reference to use of supporting literature isbased on the potential that an NCE "may have been studied by parties other than the applicantand published information may be pertinent to the new application."32 An NCE would thus besuitable for approval under Section 505(b)(2) when it is based on independent published studiesrather than prior FDA findings of effectiveness because the required investigations must beevaluations of the active moiety contained in the proposed new drug product.33

Zb ld., at 7-12.Z' Id., at 12-16.28 1d., at 13.Z9 Guidance for Industry: Applications Covered by Section 505(b)(2), Draft Guidance, October 1999,

(hereinafter "505(b)(2) Draft Guidance"), at 3. FDA has defended its interpretation of FFDCA Section 505(b)(2) asset forth in the Draft Guidance. See generally Letter from Janet Woodcock, M.D., Director, Center for DrugEvaluation and Research, to Kathleen M. Sanzo and Lawrence S. Ganslaw, Morgan, Lewis & Bockius, LLP; JeffreyB. Chasnow, Pfizer Inc.; Stephen E. Lawton and Gillian R. Woollett, Biotechnology Industry Organization; andWilliam R. Rakoczy, Lord, Bissell &Brook LLP (Oct. 14, 2003) (hereinafter "Consolidated 505(b)(2) CitizenPetition Response").

30 505(b)(2) Draft Guidance, at 3.s I Id32 Id., at 5.33 The 505(b)(2) Draft Guidance also includes a general statement that "[i]n some cases, data on a drug with

similar pharmacologic effects could be considered critical to approval." 505(b)(2) Draft Guidance, at 5. The DraftGuidance does not elaborate further on the circumstances in which such data may be relied upon, nor on what thedefinition of critical is in this context. However, the potential relevance of information from another drug productwith a different active moiety than is contained in the proposed drug product does not change the default

G~

III. Factual Background

Otsuka holds an NDA for Abilify (aripiprazole), an atypical antipsychotic initially approved intablet form by FDA on November 15, 2002, for the treatment of schizophrenia.34 Subsequent tothe initial approval, FDA has approved multiple new applications for different formulations,dosage forms and routes of administration of Abilify, as well as several additional indications forthe product.35 Most recently, on February 28, 2013, FDA approved an NDA for Abilify Maintena(aripiprazole), along-acting, intra-muscular injection for the treatment of schizophrenia.36

Each of the approved formulations and dosage forms of Abilify, including Abilify Maintena, isbased on the same active moiety, aripiprazole, the chemical structure of which is depicted inFigure 1.

On August 25, 2014, Alkermes announced that it had submitted to FDA an NDA seekingapproval of a compound called aripiprazole lauroxil, a "once-monthly, long-acting injectableatypical antipsychotic for the treatment of schizophrenia."37 The chemical structure ofaripiprazole lauroxil is depicted in Figure 2.

According to Alkermes "[o]nce in the body, aripiprazole lauroxil converts to aripiprazole, whichis commercially available under the name ABILIFY~."38 However, this conversion processproduces an intermediate molecule, N-hydroxymethyl aripiprazole. According to Alkermes:

Conversion of aripiprazole lauroxil to aripiprazole after administration isgoverned by a process, which involves dissolution of the prodrug from theinjection site, subsequent enzyme-mediated cleavage that generates anintermediate form (N-hydroxymethyl aripiprazole), and dissociation of acovalently bonded hydroxymethyl group that ultimately results in release of thearipiprazole active moieTy.39

The chemical structure of the intermediate form, N-hydroxymethyl aripiprazole is depicted inFigure 3.

requirement that two AWCTs must be conducted on a drug product containing a new active moiety to satisfy thesubstantial evidence of clinical effectiveness requirement.

34 NDA No. 021436.3s See NDA Nos. 021436, 021713, 021729, and 021866, and related supplements.36 NDA No. 202971.37 Alkermes News Release, Aug. 25, 2014.sa Id.39 Poster from Alkermes Presentation on Results of Phase 3 Study for Aripiprazole Lauroxil in Schizophrenia at

the American Society of Clinical Psychopharmacology Annual Meeting (June 17, 2014) (hereinafter "AlkermesPoster, June 17, 2014") (attached as Exhibit A).

Figure 1: Abilify (aripiprazole) and Abilify Maintena (aripiprazole)

G CI

N~N \ /

O

O N \ /

Figure 2: Aripiprazole Lauroxil

Figure 3: N-Hydroxymethyl aripiprazole

Alkermes seeks FDA approval based only on results from asingle AWCT—"the pivotal phase 3study assessing the efficacy and safety of aripiprazole lauroxil, in which aripiprazole lauroxildemonstrated significant improvements in schizophrenia symptoms, compared to placebo."4o

There are no independent published literature reports of studies of aripiprazole lauroxil thatsupport a finding that aripiprazole lauroxil is safe and effective.

IV. Discussion

Alkermes seeks approval for an NCE supported by a single AWCT without adequatejustification for submitting results from a single AWCT. Because the Alkermes NDA does not —and cannot —satisfy the information requirements of Section 505(b), the application should notbe accepted for substantive review.

a. The aripiprazole lauroxil NDA must contain reports from at least two AWCTs ofthe proposed active moiety unless it meets established exceptions permitting only asingle new AWCT

Under FDA's bright-line rule, aripiprazole lauroxil and aripiprazole containdifferent active moieties

FDA has established abright-line rule, based exclusively on chemical structure, for definingwhether the active moiety of a prodrug such as aripiprazole lauroxil is the whole molecule (i.e.aripiprazole lauroxil) prior to the bioconversion process or the previously-approved moleculeultimately released through a bioconversion process (i.e. aripiprazole). Under FDA's bright-linerule, it is irrelevant that aripiprazole lauroxil "ultimately" "convert[s]"41 into aripiprazole whendetermining whether the active moiety is aripiprazole lauroxil or aripiprazole. The decidingfactor is the chemical structure prior to bioconversion, and, specifically, whether there arecovalent structural modifications. The chemical structures of Abilify (aripiprazole) andaripiprazole lauroxil, as depicted in Figures 1 and 2, respectively, are not identical. Aripiprazolelauroxil consists of the aripiprazole moiety with a covalently bonded lauroxil group,42 whichmakes aripiprazole lauroxil an NCE under FDA's bright-line rule.

Alkermes suggests that aripiprazole lauroxil is essentially the same drug as aripiprazole becausearipiprazole lauroxil undergoes a conversion process "that ultimately results in release of thearipiprazole active moiety."43 However, Alkermes also acknowledges that the conversion processincludes "enzyme-mediated cleavage that generates an intermediate form (N-hydroxymethylaripiprazole), and dissociation of a covalently bonded hydroxymethyl group."44 These twostatements suggest that Alkermes is trying to benefit from both sides of FDA's bright-line rule.

ao Alkermes News Release, Aug. 25, 2014.41 Alkermes Poster, June 17, 2014.42 1d. See also Press Release, Alkermes Unveils Long-Acting Injectable Proprietary Platform for Treatment of

Schizophrenia and Other CNS Disorders (Feb. 4, 2010), available athttp://investor.alkermes.com/phoenix.zhtml?c=92211&p=irol-newsArticle&ID=1383098&hi h~li~=.

43 Alkermes Poster, June 17, 2014.~ Id.

D

On one hand, Alkermes's published reference to "dissociation of a covalently bondedhydroxymethyl group" squarely and unequivocally places aripiprazole lauroxil on one side ofFDA's bright-line rule as an active moiety that is not the same as aripiprazole. In fact,Alkermes's acknowledgment of an intermediate produced during the bioconversion process,which published data suggest may remain present for extended periods followingadministration,45 is consistent with the rationale FDA has provided for its bright-line ruleregarding covalently-bonded molecules.46 On the other hand, Alkermes's published reference torelease of "the aripiprazole active moiety" from axipiprazole lauroxil suggests that aripiprazoleand aripiprazole lauroxil contain the same active moiety.47 Alkermes apparently tries to use thisrepresentation to justify its submission of results from only a single AWCT and dependence onFDA's findings of safety and effectiveness of Abilify to support approval of aripiprazolelauroxil. Alkermes cannot have it both ways, and FDA should not permit Alkermes to treataripiprazole lauroxil selectively as the same active moiety as aripiprazole and a new activemoiety when FDA has established abright-line rule.

2. As a matter of established law, an NDA for an NCE must generally contain fullreports of two clinical trials

Section 505(b) of the FFDCA requires that NDAs must contain full reports of "investigations"sufficient to establish the effectiveness of an NCE. FDA has long, and consistently, interpretedthis provision to require at least two adequate and well-controlled clinical trials to establish"substantial evidence" of effectiveness except where limited exceptions are satisfied.48 FDA

45 See Morten Rohde et al., Biological Conversion of Aripiprazole Lauroxil - An N-acyloxymethyl AripiprazoleProdrug, Results in Pharma Sciences 4:19-25 (2014).

ab See supra notes 13-15 and accompanying text.47 See, e.g., Richard Pops, Chairman and CEO, Alkermes's CEO Presents at Goldman Sachs Healthcare

Conference Transcript (June 11, 2013), available at http://seekin,~alnha.com/article/1500922-alkermes-ceo-presents-at-goldman-sachs-healthcare-conference-transcript ("What [sic] so, cool about Aripiprazole Lauroxil, 9070 is that,because the active moiety is aripiprazole. We inject this once a month into the muscle as a prodrug, it metabolizesinto aripiprazole, we track in the blood stream aripiprazole levels .... We know that we're delivering therapeuticconcentrations of aripiprazole."); Richard Pops, Chapman and CEO, Alkermes CEP Presents at Citi GlobalHealthcare Conference Transcript (Feb. 25, 2013), available at http://seekingalpha.com/article/1222541-alkermes-ceo-presents-at-citi-global-helathcare-conference-transcript ("Once in the body this more complicated moleculeantibody clips down to Aripiprazole, for the active moiety in the blood stream of these patients for the month andtime is Aripiprazole, and that way we can build off of a huge clinical foundation of safety and efficacy of thismolecule."); Jim Frates, Senior VP and CFO, Alkermes's Management Presents at Deutsche Bank 38`~ AnnualdbAccess Health Care Conference Transcript (May 29, 2013), available at http://seekingalpha.com/article/1467941-alkermes-mana ement-presents-at-deutsche-bank-38th-annual-dbaccess-health-care-conference-transcript ("[W]hatwe are trying to do is deliver Aripiprazole, native Aripiprazole, over the course of a month."); Richard Pops,Chairman and CEO, Alkermes's CEO Presents at Bank of America Merrill Lynch Smid Cap Conference Transcript(May 8, 2013), available at http://seekingaluha.com/article/1415361-alkermes-ceo-presents-at-bank-of-america-merrill-lynch-smid-cap-conference-transcrikt ("Our product is a prodrug, the prodrug of Aripiprazole designedspecifically to be an injectable product once a month. Once it's injected, it fits comfortably in the muscle for a longperiod of time and it tubulizes and releases Aripiprazole."); James Frates, CFO, Alkermes' Management Presents atCredit Suisse 2012 Healthcare Conference Transcript (Nov. 14, 2012), available athttp://seekin~pha.com/artic 1 e/ 1009251-alkermes-management presents-at-credit-suisse-2012-healthcare-conference-transcript ("And one of the things —one of the questions we don't have to answer in the clinical programis whether ABILIFY actually treats schizophrenia.").

48 See generally, Clinical Effectiveness Guidance.

10

Guidance states that "it has been FDA's position that Congress generally intended to require atleast two adequate and well-controlled studies, each convincing on its own, to establisheffectiveness" for a new drug. a9

The text of the FFDCA and FDA's regulations make it clear that the investigations required todemonstrate substantial evidence of effectiveness must be investigations of the proposed newdrug. While FDA generally defines "drug" to mean a particular "drug product," i.e., a specificformulation of a product containing one or more active ingredients, investigation of a particulardrug product necessarily requires investigation of the active moiety contained in the proposednew drug product. Put simply, the default two-study rule to demonstrate clinical effectivenesscannot be satisfied by evidence from the findings of effectiveness for a drug product containingan active moiety that is not contained in the proposed new drug.so

Alkermes appears to recognize that its application, with reports of only a single AWCT ofaripiprazole lauroxil, is inadequate to support approval under Section 505(b) based only on datafrom its own application. This fact is evidenced by Alkermes's acknowledgment that approval ofits application depends on additional data from clinical trials supporting the Abilify label.sl

However, clinical trials evaluating the effectiveness of Abilify cannot simply be deemed byAlkermes to represent a clinical trial of the same active moiety as that contained in aripiprazolelauroxil when FDA's bright-line rule clearly establishes that aripiprazole lauroxil contains adifferent active moiety than Abilify. Neither prior studies of Ability nor any other publishedstudies apart from Alkermes's single trial provide any clinical data showing the effectiveness ofaripiprazole lauroxil and its intermediate form, N-hydroxymethyl aripiprazole. Therefore, thereare no external data that Alkermes can use to satisfy the general rule requiring at least twoclinical investigations of the proposed new drug.

a9 1d., at 2.so This conclusion is consistent with FDA's guidance on the necessary evidence to demonstrate clinical

effectiveness. As reviewed below, infra note 53 and accompanying text, one exception to the two-study rule iswhere studies of other formulations involving the same active moiety can be used to support approval based only ona single new AWCT of the proposed new drug product. This suitability of using evidence from another drug productcontaining the same active moiety as a proposed new drug product underscores the fact that the investigations toshow substantial evidence of effectiveness must be conducted on the active moiety contained in the proposed newdrug product. It is also consistent with FDA's justification for treating covalently bonded molecules as NCEs, whichare then treated as requiring "submission of an amount of data comparable to that required for an entirely newmolecule." Vyvanse Citizen Petition Response, at 12.

51 See, e.g., Richard Pops, Chairman and CEO, Alkermes CEP Presents at Citi Global Healthcare ConferenceTranscript (Feb. 25, 2013), available at http://seekingalnha.com/article/1222541-alkermes-ceo-presents-at-citi-~lobal-helathcare-conference-transcrikt ("Once in the body this more complicated molecule antibody clips down toAripiprazole, for the active moiety in the blood stream of these patients for the month and time is Aripiprazole, andthat way we can build off of a huge clinical foundation of safety and efficacy of this molecule.").

b. The Alkermes NDA does not meet either of the two exceptions that would permitapproval on the basis of a single AWCT

1. Aripiprazole lauroxil is not a new use of an active moiety contained in a previouslyapproved drug product

It is well established that "a single study of a new use" of a previously approved drug productcan provide evidence of effectiveness.s However, this exception to the general rule requiringtwo new studies depends on availability of "independent substantiation from study data in relateduses" of the same active moiety.53 In such cases, at least two clinical trials have been conductedon the relevant active moiety (thereby satisfying the underlying concern about reproducibility)even though only a single study was conducted for the proposed new use of a previouslyapproved active moiety. Such was the case with Abilify Maintena. Only a single new AWCTwas conducted to support approval of Abilify Maintena. However, taken together with multipleAWCTs of Abilify, which contains the exact same active moiety as Abilify Maintena, multipleAWCTs supported the approval of Abilify Maintena.

Because aripiprazole lauroxil is deemed to be a new active moiety rather than a new use of thesame active moiety in a previously approved drug, there is no available "independentsubstantiation" of the effectiveness of aripiprazole lauroxil based on previously-approved drugproducts.54 Consequently, the first exception to the two-trial default rule cannot be applicable toaripiprazole lauroxil.ss

2. The single AWCT of aripiprazole lauroxil demonstrated symptomatic benefit ratherthan reduced morbidity/mortality or prevention of disease where repetition of trialresults would pose ethical concerns

The second exception to the default two-trial rule is reserved for a large, well-designed studywith strong statistical results that provides evidence of effectiveness comparable to two or more

SZ See Clinical Effectiveness Guidance, at S (emphasis added). See generally Clinical Effectiveness Guidance,8-12.

s3 Id., at 8.sa Although an NCE that contains a covalently bonded chain to the active moiety of a previously approved drug

may be considered "related" to that drug, prior precedent indicates that drug products satisfying the bright-line rulefor an NCE are required to conduct at least two clinical studies to show effectiveness. See, e.g., the NDAs forVyvanse (NDA No. 021977), Colazal (NDA No. 020610), Cerebyx (NDA No. 020450), Lusedra (NDA No.022244), and Neurontin (NDA No. 020235). In one case FDA approved an NDA for a prodrug with a covalent bondwithout requiring two AWCTs. However, in that case FDA later conceded it had erred in failing to classify Emendas an NCE, leaving open the question of what FDA would have required at the time of approval if Emend wasproperly classified as an NCE. See Memorandum from Gary Buehler, Director, Office of Generic Drugs, Center forDrug Evaluation and Research, to NDA 022023 —Emend for Injection, Merck and Co. Inc., Emend ExclusivityDetermination (Dec. 1, 2009) ("Emend for Injection under FDA's interpretation of 21 CFR § 314.108 should havebeen classified at the time of approval as an NCE."), available athttp://www.accessdata.fda. og v~~satfda docs/NDA/2008/022023s000 AdminCorres P2.~df.

ss As contemplated by FDA's 505(b)(2) guidance, it might be possible that an NCE could be approved under505(b)(2) if there was independent published data regarding the new active moiety. 505(b)(2) Draft Guidance, at 3,5. However, there have not been any other clinical trials of aripiprazole lauroxil independently reported in publishedliterature.

12

trials (thereby satisfying the underlying concern about reproducibility). However, this exceptionwill "generally be limited to situations in which a trial has demonstrated a clinically meaningfuleffect on mortality, irreversible morbidity, or prevention of a disease with potentially seriousoutcome and confirmation of the result in a second trial would be practically or ethicallyimpossible."56 FDA has been consistent in limiting ap~rovals of NCEs based on reports of asingle AWCT to NDAs that meet these strict criteria.5

Nothing in Alkermes's public statements suggests that Alkermes believes it meets this strictstandard. Alkermes claims only that its single AWCT of aripiprazole lauroxil demonstratedsymptomatic improvement from baseline — not a clinically meaningful effect on mortality orirreversible morbidity.sg Nor has Alkermes suggested that there might be an ethical considerationto prevent a second clinical tria1.59 Thus, confirmation of the first trial's result in a second trial isnot "practically or ethically impossible."60 Accordingly, there is no basis for Alkermes tocontend that a single AWCT is sufficient to provide "substantial evidence" of effectiveness.

c. FDA should not accept for substantive review an NDA that is deficient on its faceand cannot be approved without resubmission of the NDA

1. Substantive review of the aripiprazole lauroxil NDA would be a waste of limitedagency resources

FDA has ample authority to refuse to accept for filing an NDA if, among other things, it "isincomplete because it does not on its face contain information required under section 505(b)."61

Because the Alkermes NDA is based on a single AWCT of aripiprazole lauroxil, and becausearipiprazole lauroxil is deemed to be an NCE under FDA's bright-line rule, the Alkermes NDAis deficient on its face and, thus, cannot be approved in its present form. The deficiencies in the

sb Clinical Effectiveness Guidance, at 13.57 See, e.g., the NDAs for Brilinta (NDA No. 022433) (indicated for the reduction of thrombotic cardiovascular

events in patients with acute coronary syndrome based on one very large adequate and well-controlled clinical trial);Jevtana (NDA No. 201023) (indicated in combination with prednisone for the treatment of hormone refractorymetastatic prostate cancer previously treated with a docetaxel-containing treatment regimen, a patient populationwithout any treatment options offering a survival benefit); Folotyn (NDA No. 022468) (indicated for the treatmentof patients with relapsed or refractory peripheral T-cell lymphoma, a patient population without any approvedtreatment options); Tyzeka (NDA No. 022011) (indicated for the treatment of chronic hepatitis B in patients withevidence of viral replication and active liver inflammation based on one trial that was designed to be effectively twotrials in one); and Iplex (NDA No. 021884) (indicated for the treatment of the extremely rare condition of growthfailure in children with severe primary IGF-1 deficiency or with growth hormone gene deletion who have developedneutralizing antibodies to growth hormone).

58 Alkermes Poster, June 17, 2014 ("The primary outcome measure is the change from baseline to Day 85 in thePositive and Negative Syndrome Scale (PANSS) total score."). As FDA has clearly stated, "[r]epetition of positivetrials showing only a symptomatic benefit would generally not present the same ethical concerns" as a drug showinga clinically meaningful effect on mortality or irreversible morbidity, and this weighs strongly in the "matter ofjudgment" of the review staff. Clinical Effectiveness Guidance, at 13.

s9 Such a claim would be dubious given the current market availability of Abilify Maintena, the ability forAlkermes to conduct an active control trial against Abilify Maintena, and the fact that Alkermes chose to conduct aplacebo study for its trial.

bo Clinical Effectiveness Guidance, at 13.61 21 C.F.R. § 314.101(d)(3)

13

Alkermes aripiprazole lauroxil NDA cannot be resolved during the review cycle because asecond trial is required under applicable law and regulation. Moreover, the necessity to conduct asecond trial on aripiprazole lauroxil presents a legal question that does not require scientificexpertise to substantiate an agency final decision. Therefore, substantive review of theapplication as submitted would be a waste of time and agency resources.

2. Substantive review of the Alkermes NDA is inappropriate because approval ofaripiprazole lauroxil based on a single AWCT would be an arbitrary and capriciousdeparture from agency policy

An RTF decision is particularly warranted in the case of the aripiprazole lauroxil NDA because,for all of the reasons discussed above, there is no reasonable basis upon which FDA couldultimately approve the NDA following a complete review. As submitted, the Alkermes NDAcould only be approved with data from a single AWCT involving an active moiety never beforecontained in an approved drug product. Such an approval of aripiprazole lauroxil on a singleAWCT would upend the "delicate balance" enshrined in the Hatch/Waxman amendments.Section 505(b)(2), under which Alkermes appears to seek an abbreviated approval, was part ofan attempt to balance the importance of innovation (through grants of exclusive marketingrights) and lower-cost generic drugs (supported by abbreviated approval pathways enablingsponsors to rely on FDA findings of safety and effectiveness of approved drug products and thusavoid conducting AWCTs).62

Alkermes essentially wants to be considered the same as Abilify to avail itself of an abbreviatedpathway reserved for drug products with the same active moiety and yet different than Abilify toobtain five years of marketing exclusivity reserved for NCEs. According to Alkermes, "themolecule we have developed is a new chemical entity called aripiprazole lauroxil,"63 whichwould thus be expected to qualify for five years of FDA marketing exclusivity granted to allNCEs.64 However, Alkermes also claims that, because of the ultimate conversion into thearipiprazole moiety, aripiprazole lauroxil is essentially the same drug, treating the same patientpopulation, for the same indication(s), with the same expected clinical benefit as Abilify, and

bZ Public statements by Alkermes suggest that Alkermes is poised to market aripiprazole as a high-value branddrug rather than as a low-cost generic copy of the previously-approved drug product, Abilify. See, e.g., MarkStejbach, Chief Commercial Officer, Alkermes Plc CEO Hosts R&D Day Transcript (July 17, 2013), available athttp://seekin~pha.com/article/1557512-alkermes-nlc-alks-ceo-hosts-r-and-d-day-transcript?part=since ("If you aregoing to pick a next one —and Aripiprazole obviously is the next one with Maintena recently being approved — itmakes sense as the market leader. So, this is sales on a world-wide basis. This is all indication, so a big market.But the point here is Abilify is a $7 billion drug; iYs the market leader .... If you look at Abilify Maintena, whatthis tells you is, this is still a premium product market. It is not a matter of price cutting to gain share in acommodity-like market."). Approval of aripiprazole lauroxil would, therefore, not further the Hatch/Wa~cmanpurpose of supporting lower-cost generic drugs.

63 Richard Pops, Chairman and CEO, Alkermes's CEO Presents at J.P. Morgan Healthcare ConferenceTranscript (Jan. 8, 2013), available at http://seekin~alpha.com/article/1100871-alkermes-ceo-presents-at j_p-morgan-healthcare-conference-transcript?part=single.

~ See, e.g., Memorandum from Gary Buehler, Director, Office of Generic Drugs, Center for Drug Evaluationand Research, to NDA 022023 —Emend for Injection, Merck and Co. Inc., Emend Exclusivity Determination (Dec.1, 2009) (indicating that a prodrug that includes a covalent bond should be awarded five-year exclusivity as an NCEdespite the speed with which it is converted to apreviously-approved active moiety after administration), availableat http://www.accessdata.fda. o~ v/dru satfda docs/NDA/2008/022023s000 AdminCorres P2.pdf.

14

that the safety and effectiveness of aripiprazole lauroxil can therefore be supported by FDA'sfindings regarding Abilify (aripiprazole).65 Drug sponsors should not be able to have it bothways. FDA appears to recognize that this type of outcome cannot be countenanced.66

d. If accepted for substantive review, the Alkermes NDA must be denied

This petition formally requests only that FDA refuse to file the Alkermes NDA. However, in theevent that FDA accepts the Alkermes NDA for filing and thereby implicitly denies this citizenpetition, the Alkermes NDA must not be approved for all of the reasons reviewed herein, whichare incorporated here by reference.

e. Conclusion

For the reasons, and on the basis of the legal authority, cited above, FDA should grant Otsuka'srequest that the agency refuse to accept for substantive review the NDA submitted by Alkermesfor aripiprazole lauroxil, an acknowledged NCE under the FDA's bright-line rule, because thesingle AWCT upon which the NDA is based is not sufficient to provide "substantial evidence ofeffectiveness" for the proposed indication, as required by Section 505(b)(1)(A).

C. ENVIRONMENTAL IMPACT

Petitioner claims a categorical exclusion under 21 C.F.R. § 25.31.

D. ECONOMIC IMPACT

Petitioner will, upon request by the Commissioner, submit economic impact information inaccordance with 21 C.F.R. § 10.30(b).

bs See supra note 47.66 See, e.g., Consolidated 505(b)(2) Response, at 34 ("FDA is particularly interested in examining the use of

505(b)(2) applications to obtain approval of drug products for which the only difference from the listed drug is in theform of the active ingredient, such as a change in salt .... T'his use of section 505(b)(2) does not result in theapproval of an innovative drug product that offers a new therapeutic benefit or alternative .... [n]or do suchchanges in the active ingredient generally represent an improvement in terms of safety or effectiveness. FDA isconcerned that, in addition to not representing innovating drug development, this use of 505(b)(2) applications mayhave undesirable policy and public health consequences:

1. It may undermine current incentives for development of promising new active moieties that Congressincluded in the Hatch-Waxman Amendments.

2. It may lead to proliferation of pharmaceutical alternative drug products, with resulting confusion in themarketplace.

3. It may divert resources, otherwise available for innovative drug reseazch, to the development andpatenting of alternative active ingredients").

15

E. CERTIFICATION

I certify that, to my best knowledge and belief: (a) this petition includes all information andviews upon which the petition relies; (b) this petition includes representative data and/orinformation known to the petitioner which are unfavorable to the petition; and (c) I have takenreasonable steps to ensure that any representative data and/or information which are unfavorableto the petition were disclosed to me. I further certify that the information upon which I havebased the action requested herein first became known to the party on whose behalf this petition issubmitted on or about the following dates: July 26, 2012 (Alkermes announcement of initiationof Phase 3 clinical trial of aripiprazole lauroxil) and August 25, 2014 (Alkermes announcementof submission of NDA to FDA based on single Phase 3 clinical trial). If I received or expect toreceive payments, including cash and other forms of consideration, to file this information or itscontents, I received or expect to receive those payments from the following persons ororganizations: Otsuka Pharmaceutical Development &Commercialization, Inc. I verify underpenalty of perjury that the foregoing is true and correct as of the date of the submission of thispetition.

Respectfully submitted,

OTSUKA PHARMACEUTICAL CO, LTD.

By: William H. Carson, M.D.President &CEOOtsuka Pharmaceutical Development &Commercialization, Inc.

Cc: Elizabeth H. Dickinson, Esq., Chief Counsel to the FDAJanet Woodcock, M.D., Director, Center for Drug Evaluation and Research, FDA

16

Exhibit A

17

Safety and Efficacy of Aripiprazole Lauroxil: Results from a Phase 3, Multicenter, Randomized, Double-blind, Placebo-controlledStudy in Patients with Acute Exacerbation of Schizophrenia

Srdjan Stankovic, MD, MSPH; Robert Risinger, MD; Marc de Somer, MD, MPH; Yangchun Du, PhD; Jacqueline Zummo, MPH, MBA; Lisa Corey, MS; Bernard Silverman, MD; Elliot Ehrich, MDClinical Development and Medical Affairs, Alkermes, Inc., Waltham, MA, USA

orb. me.h~4a~~y oi.n a~,. m.a~.uoo. m~ m. v..o~.m a«nmanen., mw of p.ce~m nmwe aiwn s ya.~s.~rn. mop wmmo~ aye. d ~.mw• ~ v..aa wnuowaN..a rte, ean.m~c.

meaK.~o~=Lo~pac4np njecbtlp (lAls) npasaM an mporlenl ogpn la treelvpu~¢opMene. aM e sgnifiwnl baGy of ovHerce suppe9a Mal lAls ollar anoption for tlwVMK Nab appoa~s to rmprovo aMororce maMeEly, tlwab/~asWOry ̂ Detler oNcomes br pxponls sulle~inp Isom sclumplvein.?•TMwn w I«mulaoom a some arypra~ anocnrnoles eye eveiYbk, mwIwmuYEom Ilnl allae tie po4Nal to panEe pneW~ IMxp~y la patie~Ys eMpovW n n terms of Eav eM irlarval, imgovaE plurmemkine0c (PK) poRpaM impo~eE ease alapplvton cie rwaGE.Aryipem4laurori (AL) 5 an nmva4ve UI alygcal aligsKMUc'inEavebpnenl ~w tM Vubm~t W xlvmplraw. TM IarmWEon aM kNmlopy~o~a w,z aaw•a m.iww ~a ~m arsaowo~.~a sic nd~w wncnenwa ~tinemy wa w ore.ur nary wm~ ~.sp.n m ao,..~a ao~~q~~«w.rn. ogwn. oimn swrw.s ~o...mm m. ar ry, m1a~Y,aM bknEiYly ofrcw~monWy a~ipipezop 4~oul w. pbwEo in a ripe. nnEomloU. xall

eoe~mlhtl i~ism~Eorel Gniul bNl n p~IpMnwlh xM[opl~ronw.

F~g~ra ~. scar o.s~9n

riraa wins ~ rwe.~vwra

inpex~tMm R~~

~Ea~mm~ mp/MbmrFap~uO Mba

s~.m o w>r n wu w.~u u.r~n

• Mu4cenkr. ia~gpm¢eO. UauOM-OIM.pl~nboianhWYU. 1]-walk stWynpeleNs evpOnBM~ry stile eualMtlgl olaCNmpM~IY~.ElgiCb pat~rts rnra nMwniaEln~1:1'.l rtlbb WMr: ~rppexoM lauroulBB] mp IM. anPP~V Vura~il W I mp IM. a m~GYp placebo IM Ib'x a hp~~a~~,.i

Patents abo eeeaM0on19u0y trap ll5 mp ~~pp~xob or geeaEo)Niry

.~~»KPeWnp racarveG lM inptlon on DAY 1. COY ~.~b dYS]'.tMf Itrp~mengrwE en4uYon vif~ oa~rtW al DAY BSTM1ara rnra 1 vlery lolbw.~q viols mnd 1M on Deye 113 enE t41 (.e.. t aM2 montlro all~r DaY ~I toe paEems wtw tlk rw~ Amer Me opm~bMletlemgn~Y IFpun 1).

• t • ~

Mnr.~. wo.<u:SHa7 PODulaoon: all paue~wlw nurveE xt Eose of ~WpY a~N IanppemNmamas of geuea).Full Malyfu 5e11F/S) EIfi[sy Papulauon'. au raMomimO suE~x:ls xlw«....a zi m:. a m,ar a~q,~a r.a ai w.,+~ i w~mary ancaq.ssesvnamane~ aanmm roe of may any.

r~m+ry e~avdm air,..:CM1ayalmm WSNrwb WyB5ln Pov4ve ane Napapve SyrM~omeSob(MN55) Wbl score uinp an analyse d cove~iarce ~MM.OVA) wY~ a b9oeaarmanurrea ~awara M1~CFl ~vMwcn

TM MICOVA moEN imJutleE shMy repanaM Nutmeat pmuD es legas.eM Wxlna MNaS bdl score es a wnrulo.TM Ibmmel procMun was ~eaE Im Na muYgla ~ypodieeis NaM1p.

EK~[xyvnf eboerol~zatl uiege aaC moEel lo~rapoleE manures ~MAlitl.~

s~o a.n.m+n.mry*~,.

~a.n e~avdma,.iy.~.:GNI[el Gbbal lmyembn-Imgowmam (CGl~I) econs al Oay BS ainp a rwn~panmotrk Wikoxon nnk wn bn ueinp the LOCF epyoxM1.

• m.wmm.iw«w.,.r...~..a.i. nw.wa.icn.4»ioi oos oay. oomcomwnmm Inyn awe w. Wubo.ro bw tloes w. WmDol waro sbMacelry~Iprvfini for Na Amery ~I/ury anEpon.

Owr~ll Ra~pon~</u+aly~lf:• 0.Rroe as PUJ55 tohl icon z~p96 Muaaw (mpmmrrt M fympoms) lmm

e~~Nm «CGi.i xore of z 1m~ imc~owa) or 1(vary micn impowa~.SWry utl TaMneility

e..`rra m.a a, v.au~.~..~w,v~.awn. «mro (r~eq em omen..r.~ym..wn.

Flpure 2. PaHaM Disposition

65%o1 M4rbwln~wrvM ~Tp~ob bu~mII8B2 nq vq 8T6Wpahorb wro ~ecaiv0 aryp~zaM Yvonl Ht mp wnpMMd OouONdYnOo-.,~~i ~, asxawwm~.roo,«...e w~+ao

Table'1. Demographics and Baseline Characteristics,Rantlomiietl Population

PY[tlo ~L N1 mp ~L W3 mpn~]OB

!U 395 11.&5 JB9 tO 1J NJ 1106M~Mr• mwnn ~%)

Mek 138 (fi8.B) tat (8811 ta318081

~. ~ Ixl

&xx o~uiun.~. K,~

e+(asxlMIa0 ~1xeuae~

ee (m BlB~(tl.11~a c~~ei

se N].1~8113891zeius~

n.Wun.nlxlNOM NrcrceEwopv

aM

1021f 0)BO (]B 5~ZB 12.5

1N Ii 9)B1 (]p 1)Yl 111

1031.0)]8 (!) 5)]B 1J5

man DO iB 911888 GO )! 1)Bd BO ZB 18 J1

~r lkW^~9. m~ri 18.HS IsaB7 x)ee 1s 331 n]L IS.en

~ PeexM vino>a0omneMN mar 16Bx1.M tM nroeu~aoo was ao yoarso.,y.. less ri.ni.

Flpure JA. Change trom Baseline in PAN83 Total Bcoreby Ysit, ANCOVA LOCF

0

-10

1 B 16 Y1 ]B ]8 ~3 60 61 M 71 18 86

VISIT Day

Figure 3B. CMnpa hom Batelina in PANES T0419toreby Visk, MMRM

10 -

-~ ~w]mY .16 ti-.LLMmp

20 ••

26 .OD1 ••

1 B 16 33 3Y ]e ~] W 61 M 71 ]8 B6

Vlslt Day

SmEeEolly nM dinralty apn~fiam improwmaMs vnlh CON doses of engqazab bwMil Hen Eemon~trabd ae early as Day B arM continuetlNrouph b the and o(douEb-0IiIW treatrnont (ANCOVA. LOCF, antl MMRM analyses) (Flpun 7A 61B).At bemlina. maen (50) PAN55 bbl wores were 93.8 (17.3), 92.8 (10.2), aM 92.0 (10.7) br pWceCo, aripiproiole Iaumb1401 rtp and Ynpproi0lelaumnl 882 rtp proupa, rnpocOwly Lean pwro msm (SE) W npe from baeellne b tM eiM of tlx EouEk-blind treabnent hom tM ANCOVA modalwas -8.811.4) with P~a~. -p.8 (1.II. arW -21.8 (1.4) with anppromb Ieuroxi1401 mp antl 882 rtp dome. reapocEvoly.

Table 2. Summary of CGI-I at End of Double-btlndTreatmentCGIa Seor~ %aubo

(n~7Y8)

~ 1x)

AI Mt mp

In•198)

~ ~x>

AL 882 mp

(n=I0~)

~ ~v.~Vary nxch i~rymvsd 15 (7.~ 27113.8) ?51723)

Much imgowtl 33116.e) 8B (317) at (38.T

Minirtully improvetl !3X21.9) r5 (23.0) 52 (25.5)

No cna~ge 42121.1) ffi(18.3) 24(11.8)

Mircmally zone 37 (18.8) 17 (5.8) 18 (7.B)

Much wore 23(17.7) 12 (8.1) 5 (2.5)

Very m!h wince 3 (7.5) 1 (0.5) 1 (0.5)

P-wlus - W.001 W.007

ean.rowuer ~~ma eam»n vow+a~.o-e.a w~rc.~mr nero~CGIa fcom n. pYcaUo.TN poporlon of pW~Txtw xera lery muN mpore0' or'muc~ imposedw.a ea~vnr iwv win. (v~o.roi. ~ ~+a.~ ~wew,eNv.a nv ~~viM uippnmM Yuoul N1 mp aM upipamY bwwi 883 mp w. pYcaEo at~~tl~ ~sapwmnl lipn C1y 15 b aM of bulmerl.

Figure 4. Percent Overall Response by Visk

W%

~ ..~

... ...

~o1c

~x rc~

5 ]0%K

10%

0%B 16 ]3 ]Y A 86

sbm. roan.-romm~ Vblt Day

TM p~rnntpa of o~arell nspoMers (PAt155 improvement 'f0%w CGI-14~enl wM IM pim~ry Mary fiMinps.

iTe ra p W weraY response n OOM tlw e~gqruaN Yuonl611 mp enCaraw~ow v~ro,~ aez ma uro~,w w,s u.~~iN.~,a cin~N w~r~meiOay 15 aM wrtnwE tl~eaup~ b tlw eM d th buNiarY p~noE (Fqure ~~.

Table 3. TEAEs RaponeU in >6% oT Patients in Aripiprazola Lauroxil Groups, Safety Population

TEAEs Plxaho(nm207)n~%~

AL M1 mp(n~Z07)n~%)

AL 882 mg(n=208)n~%)

Insomnia 24X11.6) 20(9.7) 25(12.0)Akathisia 9 4.3) 24X11.6) 24(11.5)HeaOache 17 (8.2) 17 (82) 18 (8.7)Anxiety 14 (6.8) 6 (2.9) 11 (5.3)

TwM M~„a i~W~YO ~1 YW w~ f~mu Marx w~ti (~). 11 J~Ygtlr h~~h~rX pecE (4 in IM O~ VUV. ] in Ow aTp~w bvwi Mt mp ppp~M~ Ntlr upp~xeY Yu~o~I8B2 mO Wa+V) ~M t U~VW~~ Ywzi 88t nq YrouV) Evinp IM1s bbxuV G~ba. rcW~V oM OMN (NNO M MTHO~) IM1M OCWn~O'H tlNpr+bc Pout/ No MvNul54E was roGo~ h' may tlrn 1 pINrY.

P75

This was a fixeC-tlose study wdh no opportunKy for doseatljustment; therefore it is possible that some patientsmight have benefdetl from fieuble dosing.

• Both doses of aripiprazole lauroxildemonstrated robust efficacy with clinicallymeaningful and statistically signficanlimprovements evident as early as Day 8, whichpersisted throughout the treatment period.• The observed effect of aripiprazole lauroxil

was evident and consistent across allprimary and secondary endpoints.

• Both doses of aripiprazole lauroxil were well-tolerated overall.

• The results of this study suppoA the conclusionthat aripiprazole lauroxil may represent a newtreatment option for patients withschizophrenia.

• Further evaluation of potential benefits relatedto flexibility in dosing schedule andadminisVation of aripiprazole lauroxil isongoing.

1. LeucH 5, Tally M, Komosu K,MeI. N4psycMYc Erupa ~aiaua pleubo forrampaa y.wtiwn in acNmprraNe. ~ sy~mma0c revsw am mem-an~~yis.Laxet. 2013:3]8(88311 ~-p)1.

2. LNEemm~yar P,L'u~5N1~~H, Nuiam PM, eI aI. ANtlu4on nanctlM~arca enEn VNwtYa wiN fMimMreno o~ x~awllxEw UnotUa~wiN

subo7mal Viw iaeG~~. J GF PaK'wdY. ~OOB'.TN~1:880.BB6.a ic.~.~r.c.,~..ww~c. canriwnwn.,«omm.m.w~. i«.mwv~ma

bnpagip npcOma. BrJ PaK.'IYby SugY. 3008:53'58}SB].0. Wplen G. GaoY J.Zunmo J.ImPp bbrpectigin~ac4bM ~nEps~cM1olcsm

msduEon WMroree ~M cirvpl, IuMnrel. eM acawne oulremn eluM1mgrane. PeEeiYPNwA~v.2013:t3(n'.11f1.1180.

m.: sn,cr w.. iwa.a.~w ~onanea Mnuerma:, i~~.AY eult~ors aia ompbyvos b Rlkarmes.lrc

ClkeYmes~Patrent inspired'

Srd~an. Stankovic@alkermes. coin