Sandoz Inc. v. Cochran ( 2022 )

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
SANDOZ INC.,
Plaintiff,
v.
No. 21-cv-600 (DLF)
XAVIER BECERRA, Secretary, United
States Department of Health and Human
Services, 1 et al.,
Defendants.
MEMORANDUM OPINION
Sandoz Inc. challenges a decision of the Food and Drug Administration (FDA) that granted
four years of market exclusivity to the sponsor of a new drug, Aubagio. See Compl. ¶ 1, Dkt. 1.
Sandoz argues that the sponsor was not entitled to exclusivity because the FDA had previously
approved the use of Aubagio’s active ingredient in another drug, Arava. See id. ¶¶ 66–85. Before
the Court is Sandoz’s Motion for Summary Judgment, Dkt. 14, and the government’s Cross-
Motion for Summary Judgment, Dkt. 20. For the reasons that follow, the Court will grant the
government’s motion and deny Sandoz’s motion.
I. BACKGROUND
A. Legal Background
The Food, Drug, and Cosmetic Act (FDCA) prohibits introducing “any new drug” into
interstate commerce without prior approval by the FDA. 21 U.S.C. § 355(a). Pharmaceutical
1
When this complaint was filed, Norris Cochran IV was the Acting Secretary of Health and Human
Services. When Xavier Becerra became Secretary, he was substituted pursuant to Fed. R. Civ. P.
25(d).
companies may obtain that approval in two ways. First, a company may submit a new drug
application (NDA) under § 505 of the FDCA. 21 U.S.C. § 355(b). The FDA may then approve
that application only if the company demonstrates that its drug is safe and effective for its proposed
use—a process that often requires clinical trials. See id. § 355(d); see also id. § 355(b)(1)(A), (d)
(specifying other requirements for NDAs). Alternatively, after the FDA has approved a new drug
and certain rights of that drug’s sponsor have expired, see, e.g., id. § 355(j)(5)(B)(ii), a second
company may apply to market a generic version of the drug by submitting an abbreviated new
drug application (ANDA), id. § 355(j). The FDA may approve an ANDA upon finding that a
generic drug is equivalent to the original, listed drug in several respects—a process that rarely
requires clinical trials. See id. § 355(j)(4); Ipsen Biopharm., Inc. v. Becerra, 2021 WL 4399531,
at *1 (D.D.C. Sept. 24, 2021).
When a company obtains approval to market a new drug, the FDCA may also grant it a
period of market exclusivity. See, e.g., 21 U.S.C. § 355(j)(5)(F)(ii). These periods are “designed
to compensate manufacturers for research and development costs as well as the risk of litigation
from patent holders.” Teva Pharms., USA, Inc. v. Leavitt, 548 F.3d 103, 104 (D.C. Cir. 2008).
Although the FDCA contains multiple provisions that confer exclusivity, see Amarin Pharms.
Ireland Ltd. v. FDA, 106 F. Supp. 3d 196, 199 (D.D.C. 2015), only two are relevant here. 2
First, the FDCA grants at least four years of exclusivity to companies that successfully
submit an NDA for any new drug, “no active ingredient . . . of which has been approved in any
2
Because Sandoz challenges the grant of exclusivity to Aubagio, the Court considers the text of
the FDCA that was operative as of that drug’s approval, September 12, 2021. See A.R. 1456.
2
other [NDA].” 3 21 U.S.C. § 355(j)(5)(F)(ii) (2012). This benefit, which is commonly called new
chemical entity (NCE) exclusivity, prevents other companies from submitting ANDAs that “refer[]
to” the approved drug for at least four years. Id. For this purpose, the term “active ingredient”
refers to “any component 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 body.” 21 C.F.R. § 314.3(b). The FDA determines a drug’s active ingredients
by looking to its chemical structure before it enters the human body, as opposed to after its
metabolization. See A.R. 1580–82; see also Actavis Elizabeth LLC v. FDA, 625 F.3d 760, 764–
66 (D.C. Cir. 2010) (approving this approach).
Separately, the FDCA grants 180 days of exclusivity to the “first applicant” that
successfully submits an ANDA that contains a Paragraph IV “certification.” 21 U.S.C.
§ 355(j)(5)(B)(iv). By way of background, each application to market a generic drug must submit
a certification regarding any patents that claim the associated listed drug. See id.§ 355(j)(2)(A)(vii). This certification may provide that a patent “has expired,” id.
§ 355(j)(2)(A)(vii)(II), that it “will expire,” id. § 355(j)(2)(A)(vii)(III), or—under Paragraph IV—
that it “is invalid or will not be infringed by the manufacture, use, or sale of the new drug for which
the application is submitted,” id. § 355(j)(2)(A)(vii)(IV). Submitting a Paragraph IV certification
“comes with a risk” because it constitutes a technical act of patent infringement and may spark
“costly litigation.” Teva Pharms. USA, Inc. v. Sebelius, 595 F.3d 1303, 1305 (D.C. Cir. 2010)
(citing 35 U.S.C. § 271(e)(2)). It is for that reason that the FDCA confers a limited period of
exclusivity to certain “first applicant[s].” 21 U.S.C. § 355(j)(5)(B)(iv). The benefit of exclusivity
3
The current version of § 355(j)(5)(F)(ii) replaces the term “active ingredient” with “active
moiety.” See Act to Amend the Federal Food, Drug, and Cosmetic Act, Pub. L. No. 117-9, 135
Stat. 256, 256 (Apr. 2, 2021).
3
seeks to offset the cost of litigation, thereby “expediting the availability of generic equivalents.”
Teva, 595 F.3d at 1305.
As relevant here, the FDCA allows multiple companies to share “first applicant” status.
See 21 U.S.C. § 355(j)(5)(B)(iv)(II)(bb). A company can obtain the status by being the first to
submit a “substantially complete” ANDA that includes a Paragraph IV certification. Id. Other
companies may also obtain the same status by filing “substantially complete” Paragraph IV
ANDAs “on the [same] day” as the initial application. Id. When multiple companies share “first
applicant” status, their periods of exclusivity overlap. See id. § 355(j)(5)(B)(iv)(I). And when any
first applicant benefits from exclusivity, no other company may market a generic version of the
same listed drug. See id. Accordingly, any generic manufacturer that intends to submit a
Paragraph IV ANDA has an incentive to submit that application on the first possible date—i.e.,
four years from the approval of the initial, listed drug. See id. § 355(j)(5)(B)(iv).
B. Factual Background
This case arises from the approval of two drugs—Arava and Aubagio—that contain the
same ingredient—teriflunomide.
The FDA approved Arava in 1998 for the treatment of “adults with active rheumatoid
arthritis.” A.R. 19, 360. That approval identified Arava’s sole active ingredient as leflunomide,
see A.R. 265, 325–326—a chemical that is closely related to teriflunomide. When leflunomide
enters the human body, it “is metabolized . . . to teriflunomide which is responsible for essentially
all of [the former]’s in vivo activity.” A.R. 1455. Leflunomide also tends to degrade into
teriflunomide over time, even before it enters the human body. See A.R. 1445, 1459. For that
reason, the FDA noted that Arava contains trace amounts of teriflunomide. See A.R. 1459.
4
Specifically, the agency designated those amounts as impurities in Arava and provided that they
could be no more than 0.3% of the drug. See A.R. 1459.
The FDA approved the second relevant drug, Aubagio, on September 12, 2012, to treat
“patients with relapsing forms of multiple sclerosis.” A.R. 402, 1456. That approval noted that
teriflunomide was the sole active ingredient in Aubagio. See A.R. 418–19, 1456–57. It also
determined that the FDA had not previously approved another drug with the same active
ingredient. See A.R. 1171, 1457. For that reason, the FDA granted Aubagio at least four years of
NCE exclusivity pursuant to § 355(j)(5)(B)(iv). See A.R. 1457. In doing so, the FDA took the
position that no generic manufacturer could submit a Paragraph IV ANDA that referenced Aubagio
until September 12, 2016. See 21 U.S.C. § 355(j)(5)(B)(iv).
On August 31, 2016, Sandoz wrote to the FDA for the purpose of challenging Aubagio’s
exclusivity. See A.R. 1393. Sandoz argued that the exclusivity was unwarranted because
teriflunomide was “physically present as a bioavailable and physiologically/pharmacologically
active component of” Arava. A.R. 1401. It also argued that the sponsor of both Arava and
Aubagio—Sanofi-Aventis US LLC (Sanofi) 4—was aware of the ingredient’s presence in Arava
and viewed it as beneficial. See A.R. 1396–97. On that issue, Sandoz pointed to several patent
applications and patent infringement suits in which Sanofi claimed positive interactions between
leflunomide and teriflunomide. See A.R. 1396–99; see, e.g., U.S. Patent No. 7,071,222, at [1]
(filed Mar. 7, 1997) (issued July 4, 2006) (noting that a “combination preparation” of leflunomide
and teriflunomide “exhibits surprisingly advantageous immunosuppressive effects”). Finally,
Sandoz insisted that its position did not rest on the fact that “teriflunomide is the active metabolite
4
Sandoz represents that the original sponsor for Arava was Hoechst Marion Roussel, Inc.—a
predecessor in interest to Sanofi. See Pl.’s Mem. in Supp. of Mot. for Summ. J. at 10, Dkt. 16.
5
of leflunomide.” A.R. 1401. In this respect, Sandoz declined to challenge the FDA’s general
approach for identifying active ingredients, which looks to drugs’ structure prior to their
metabolization. See Actavis Elizabeth, 625 F.3d at 764–66.
Sandoz subsequently filed two Paragraph IV ANDAs to market generic teriflunomide. See
A.R. 1403–10. Sandoz filed its first application on September 7, 2016—five days before
Aubagio’s exclusivity was set to expire. See A.R. 1403. That application referenced Sandoz’s
challenge letter and asked the FDA, if it accepted the challenge, to “deem [the application]
received” on the date of its submission. Id. Sandoz then filed a second and nearly identical
application on September 12—the first day after Aubagio’s scheduled exclusivity. See A.R. 1407.
The FDA acknowledged the receipt of both applications and advised that it would “take action on
only one” of them. A.R. 1416. It also advised that, if Sandoz prevailed in its exclusivity challenge,
Sandoz could choose which application would proceed. Id.This arrangement positioned Sandoz to benefit from its exclusivity challenge. The FDA
received twenty-one ANDAs on September 12, 2016, that sought to market generic versions of
teriflunomide. See FDA, Paragraph IV Patent Certifications, at 73,
https://www.fda.gov/media/133240/download (last visited July 21, 2022). Sandoz appears to be
the only company, however, that filed a substantially complete ANDA before that date. See Pl.’s
Mem. in Supp. of Mot. for Summ. J. at 16, Dkt. 16 (undisputed). For that reason, if Sandoz
prevailed in its challenge, it would be the only “first applicant” for that generic and thus receive
180 days of market exclusivity. See 21 U.S.C. § 355(j)(5)(B)(iv). In contrast, if its challenge did
not prevail, Sandoz would be one of twenty-one “first applicants” for the generic and would need
to share its exclusivity with those other companies. Id.6
The FDA rejected Sandoz’s exclusivity challenge on July 27, 2018. See A.R. 1436–49.
The agency explained that it did not consider teriflunomide to be an active ingredient in Arava at
the time that it reviewed that drug’s NDA. See A.R. 1445. Instead, it viewed teriflunomide as an
impurity, which was neither “added intentionally” to nor required to be present in the drug. A.R.
1445–46. The agency further reasoned that leflunomide was the “sole active ingredient in Arava”
because it was “the sole component of [that drug] intended to furnish pharmacological activity . . .
in the treatment of adults with active rheumatoid arthritis.” A.R. 1447. Moreover, the agency
reasoned that, “[i]f Arava contained more than one active ingredient, then additional showings
would need to have been made by Sanofi to support [the drug’s] safety and effectiveness.” A.R.
1447. Finally, the agency reasoned that its treatment of Arava was consistent with its past practice
and that Sandoz’s contrary approach would raise administrative difficulties, such as “exponentially
increas[ing] the number of substances that would need to be considered for purposes of the 5-year
exclusivity analysis.” A.R. 1447–48. The FDA thus denied Sandoz’s first ANDA, see A.R. 1465,
which put Sandoz on track to share its period of exclusivity.
Sandoz filed an administrative appeal with the FDA on October 10, 2018. See A.R. 1473–
80. In that appeal, Sandoz argued that teriflunomide was an active ingredient in Arava because
Sanofi “knew that the teriflunomide present in Arava was physiologically active,” regardless of
whether the FDA considered that activity. A.R. 1477. Sandoz further argued that the FDA had
“inherent authority to correct past mistakes” regarding its classification of active ingredients and
had “done so in the past.” A.R. 1477 n.2. Finally, as a rebuttal to the argument that its position
would raise administrative difficulties, Sandoz argued that its case presented unique
circumstances—namely, that Sanofi knew about the presence and effect of teriflunomide at the
time of Arava’s approval. See A.R. 1478.
7
The FDA denied Sandoz’s administrative appeal on February 12, 2021, relying on
substantially the same reasoning as its previous decision. See A.R. 1481–84. The agency also
accepted Sandoz’s second ANDA, such that Sandoz became eligible for “180 days of shared
generic exclusivity.” A.R. 1485–86. Under the FDCA, that period of exclusivity is set to begin
on “the date of the first commercial marketing of [generic teriflunomide] . . . by any first
applicant.” 21 U.S.C. § 355(j)(5)(B)(iv)(I). The parties report that this marketing will not occur
until March 2023 at the earliest. See Joint Status Report of June 29, 2022, Dkt. 39.
In this action, Sandoz challenges the rejection of its first ANDA under the Administrative
Procedure Act, 5 U.S.C. § 551 et seq. Compl. ¶¶ 1, 66–92. Sandoz first argues that the rejection
was inconsistent with 21 U.S.C. § 355(j)(5)(F)(ii). Under its reading of that statute, Aubagio is
ineligible for NCE exclusivity simply because the “FDA approved Arava knowing it contained
teriflunomide,” regardless of whether it approved teriflunomide “as an active ingredient.” Pl.’s
Mem. at 2 (emphasis omitted). In the alterative, Sandoz argues that teriflunomide is an active
ingredient in Arava. See id. at 32–39. And finally, Sandoz argues that awarding NCE exclusivity
to Aubagio was arbitrary and capricious insofar as it was inconsistent with a past agency decision.
See id. at 39–43. For those reasons, Sandoz asks this Court to set aside both the rejection of its
first ANDA and the preceding grant of NCE exclusivity to Aubagio. See Compl. at 40; see also 5
U.S.C. § 706(2).
II. LEGAL STANDARD
A court will grant summary judgment if the moving party “shows that there is no genuine
dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R.
Civ. P. 56(a); see also Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247–48 (1986). A “material”
fact is one with potential to change the substantive outcome of the litigation. See Liberty Lobby,
8
477 U.S. at 248; Holcomb v. Powell, 433 F.3d 889, 895 (D.C. Cir. 2006). And a dispute is
“genuine” if a reasonable jury could determine that the evidence warrants a verdict for the
nonmoving party. See Liberty Lobby, 477 U.S. at 248; Holcomb, 433 F.3d at 895.
In cases arising under the Administrative Procedure Act, summary judgment “serves as the
mechanism for deciding, as a matter of law, whether the agency action is supported by the
administrative record and otherwise consistent with the APA standard of review.” Sierra Club v.
Mainella, 459 F. Supp. 2d 76, 90 (D.D.C. 2006). The Court will thus “hold unlawful and set aside”
agency action that is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance
with law,” 5 U.S.C. § 706(2)(A), “in excess of statutory jurisdiction, authority, or limitations, or
short of statutory right,” id. § 706(2)(C), or “unsupported by substantial evidence,” id. § 706(2)(E).
III. ANALYSIS
A. Sandoz has Article III standing
Before reaching the merits of this case, the Court must determine whether Sandoz has
Article III standing. See Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 93–95 (1998). To
establish standing, Sandoz must demonstrate that it has suffered an “injury in fact” that is “concrete
and particularized” and “actual or imminent, not conjectural or hypothetical.” Lujan v. Defenders
of Wildlife, 504 U.S. 555, 560–61 (1992) (internal quotation marks omitted). It must also establish
that there is “a causal connection between the injury and the conduct complained of” and that it is
“likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision.”
Id. (internal quotation marks omitted). Each of these elements “must be supported in the same
way as any other matter on which the plaintiff bears the burden of proof.” Id. at 561.
Sandoz has adequately established Article III standing. The D.C. Circuit has previously
held that the denial of market exclusivity is an injury in fact because it exposes plaintiffs to
9
“allegedly unlawful competition.” Teva, 595 F.3d at 1312. Sandoz falls within that holding
because it is slated to receive shared market exclusivity instead of sole market exclusivity, which
will expose it to a similar competitive harm. See supra section I.B. Sandoz has also shown that
the FDA caused its injury. And finally, Sandoz has shown that its injury is redressable through
setting aside the FDA’s denial of its first ANDA, see 5 U.S.C. § 706(2)—a result that would make
Sandoz the only “first applicant” for generic teriflunomide, 21 U.S.C. § 355(j)(5)(B)(iv)(II), and
grant it sole exclusivity for 180 days following that drug’s “first commercial marketing,” id.§ 355(j)(5)(B)(iv)(I). For those reasons, and because the Court otherwise has subject-matter
jurisdiction under 28 U.S.C. § 1331, the Court will reach the merits of the parties’ dispute.
B. NCE Exclusivity Turns on Whether the Active Ingredient of the New Drug
Was Approved as an Active Ingredient in a Prior Drug
To determine whether Aubagio is entitled to NCE exclusivity, the Court must determine
the standard for assigning NCE exclusivity in general. That standard is set forth in 21 U.S.C.
§ 355(j)(5)(F)(ii), which provides:
If an application submitted under subsection (b) of this section for a drug, no active
ingredient . . . of which has been approved in any other application under subsection
(b) of this section, is approved . . . , no application may be submitted under this
subsection which refers to the drug for which the subsection (b) application was
submitted before the expiration of [at least four years].
21 U.S.C. § 355(j)(5)(F)(ii). The phrase “subsection (b)” refers to 21 U.S.C. § 355(b), which
governs the process for submitting and evaluating NDAs. See id. § 355(b). Accordingly,
§ 355(j)(5)(F)(ii) confers exclusivity upon the approval of an NDA “for a drug, no active
ingredient of which has been approved in any other” NDA. Id. § 355(j)(5)(F)(ii).
The parties offer competing interpretations of this provision, and especially of its adjectival
phrase, “no active ingredient . . . of which has been approved.” Id. The FDA argues that the
10
language confers exclusivity when the “active ingredient[s]” of a new drug have not previously
“been approved” as active ingredients in a prior drug. See Gov’t’s Mem. in Supp. of Summ. J. at
12, Dkt. 22. Under that view, Aubagio would be entitled to exclusivity unless the FDA both
considered teriflunomide to be an active ingredient in Arava and approved it as such. In contrast,
Sandoz argues that § 355(j)(5)(F)(ii) confers exclusivity when the “active ingredient” of a new
drug has not previously “been approved” in any capacity, including as an impurity in a prior drug.
See Pl.’s Mem. at 22. Under that view, the fact that the FDA considered teriflunomide at all in the
course of approving Arava would prevent Aubagio from receiving exclusivity. For the reasons
that follow, the Court agrees with the FDA’s interpretation of § 355(j)(5)(F)(ii).
To begin, the plain text of § 355(j)(5)(F)(ii) favors the FDA’s position. On its face, that
provision refers both to one new NDA, which resulted in the approval of a new drug, and to all
previously approved NDAs. See 21 U.S.C. § 355(j)(5)(F)(ii). The provision then asks whether
any “active ingredient” of the new drug was “approved” in the NDA for any prior drug. Id. It is
true that this language does not elaborate on what that approval entails. But because the provision
expressly refers to the approval of active ingredients, its most natural reading considers their
approval as active ingredients. See Hibbs v. Winn, 542 U.S. 88, 101 (2004) (counseling “that
statutory language must be read in context [since] a phrase gathers meaning from the words around
it”).
Sandoz’s contrary reading of the text strains the ordinary meaning of “approved.” To
approve something means “to accept [it] as satisfactory,” Merriam Webster’s Collegiate
Dictionary (10th ed. 1997), or “to give [it] formal sanction,” Black’s Law Dictionary (11th ed.
2019). Here, Sandoz submits that § 355(j)(5)(F)(ii) contemplates not only the approval of drugs
and their active ingredients, but also the approval of what the FDA defines as those drugs’
11
impurities. See Pl.’s Mem. at 21, 26. But the FDA defines an impurity as a drug component that
is “added [un]intentionally . . . as a result” of the drug’s “manufacturing [or] storage.” A.R. 1459.
The FDA also notes that it does not require drugs to contain a “minimum level” of any impurity
and may instead limit the amount of impurities that drugs may contain. See id.; see also 21 U.S.C.
§ 355(d) (requiring the FDA to assess whether the process for manufacturing, processing, and
packing a drug is adequate to “preserve its . . . purity”). That regulatory approach is not naturally
described as deeming impurities to be satisfactory, let alone giving them formal sanction. So
although the FDA plainly considers impurities in approving drugs, an ordinary English speaker
would not say that it approves those impurities themselves.
By way of example, consider the subset of drugs that contain carcinogenic impurities.
Although the FDA may approve those drugs on the ground that they contain only trace amounts
of their impurities, no ordinary person would say that the agency thereby “approved” the impurities
themselves. 21 U.S.C. § 355(j)(5)(F)(ii). Instead, that person would say that the agency approved
the drugs despite their impurities—or instead that the agency simply tolerates those impurities.
This example weighs against reading § 355(j)(5)(F)(ii)’s “approved” language to cover any
impurity, as the provision lacks a textual basis for distinguishing harmful impurities from harmless
ones. Indeed, Sandoz concedes that the provision has “no textual basis for distinguishing”
impurities from inactive ingredients. Pl.’s Reply at 13, Dkt. 27. The best reading of
§ 355(j)(5)(F)(ii) thus refers only to approval of active ingredients as active ingredients.
That conclusion is consistent with the use of the verb “approve” throughout § 505 of the
FDCA. See Henson v. Santander Consumer USA, 137 S. Ct. 1718, 1722–23 (2017) (applying the
presumption that “identical words used in different parts of the same statute carry the same
meaning” (internal quotation marks omitted)). The full text of that section refers to the approval
12
of “application[s],” 21 U.S.C. § 355(a); “supplement[s]” to applications, id. § 355(j)(5)(F)(iv);
“drug[s],” id. § 355(b)(4)(A); “method[s] of using” drugs, id. § 355(b)(1)(viii)(II); and the
“labeling” for drugs, id. § 355(j)(10)(A)(i). It also refers to the approval of “active ingredient[s].”
Id. § 355(j)(5)(F)(ii). But the statute never refers the approval of other drug components, as
distinct from the approval of new drugs themselves, see id. § 355(d) (asking whether a “drug is
safe for use” and requiring an assessment of “its identity, strength, quality, and purity”). And
because the FDCA never describes the individualized review of those drug components, this Court
will not read that review to be implied within § 355(j)(5)(F)(ii).
This Court’s conclusion is also consistent with the role of active ingredients elsewhere in
the Hatch-Waxman Act, which has been codified in 21 U.S.C. § 355(j). See Hatch-Waxman Act,
Pub. L. No. 98-417, 98 Stat 1585 (Sept. 24, 1984). That Act governs not only the standard for
NCE exclusivity for new drugs, see 21 U.S.C. § 355(j)(5)(F)(ii), but also the submission and
approval of ANDAs, see id. § 355(j)(2), (4). And as relevant here, the Act allows approving an
ANDA only if the proposed generic contains the same active ingredients as a listed drug. See id.
§ 355(j)(2)(A)(ii); see also id. § 355(j)(2)(A)(ii)(III) (noting one exception). That point goes to
the overall “design” of the Hatch-Waxman Act, Wolf Run Min. Co. v. FMSHRC, 659 F.3d 1197,
1200 (D.C. Cir. 2011): Because the Act uses the overlap of active ingredients to determine ANDA
eligibility, there are good reasons to think that the Hatch-Waxman Congress intended that same
variable to foreclose NCE exclusivity. For one, there is a logical symmetry between granting
exclusivity to certain new drugs and authorizing generic versions of certain established drugs. And
because Congress concluded that an overlap of active ingredients was relevant to identifying the
latter, it presumably deemed the same metric to be relevant in identifying the former. Another
consideration is the efficient use of agency resources. Because the overlap of active ingredients is
13
relevant to assessing ANDAs, see 21 U.S.C. § 355(j)(2)(A)(ii), the Hatch-Waxman Congress
presumably expected the FDA to track those ingredients over time. The Congress would have
accordingly assumed that the FDA could also that same information to assess NCE exclusivity. In
contrast, Sandoz does not identify any other provision of the FDCA that would require the FDA
to track “impurities controlled in approved NDAs,” Gov’t’s Mem. at 22, and the agency does not
track them at present, see id. (undisputed). The design of the Hatch-Waxman Act thus supports
the agency’s broader argument that Sandoz’s position would create a substantial administrative
burden. See Gov’t’s Mem. at 22.
Finally, the Court need not resolve the parties’ dispute regarding the exact purpose of
§ 355(j)(5)(F)(ii). Both parties agree that Congress enacted the provision to promote the creation
of certain novel drugs. See Pl.’s Mem. at 27–28; Gov’t’s Mem. at 16–19. From there, the FDA
argues that Congress wanted to promote the creation of drugs that are the first to use a given
substance as an active ingredient, see Gov’t’s Mem. at 17–19, whereas Sandoz argues that
Congress wanted to promote drugs that are the first to use a new substance that qualifies an active
ingredient, see Pl.’s Mem. at 27–28; Pl.’s Reply at 7. But neither party has provided a compelling
reason to prefer its theory on this discrete issue. Indeed, there are good reasons why Congress
might want to promote either of the above subsets of drugs, and Sandoz’s reliance on legislative
history sheds no light on the question.5 Regardless, it is appropriate to “begin and end with the
plain” and unambiguous text of the FDCA, United States v. Novo A/S, 5 F.4th 47, 54 (D.C. Cir.
2021). See Eagle Pharms., Inc. v. Azar, 952 F.3d 323, 340 (D.C. Cir. 2020). And that text makes
5
As a general matter, “legislative history is not the law.” Azar v. Allina Health Servs., 139 S. Ct.
1804, 1814 (2019) (citation omitted). And here, Sandoz gains nothing from the proposition that
§ 355(j)(5)(F)(ii) promotes the development of “new chemical entities,” Pl.’s Mem. at 27 (quoting
130 Cong. Rec. 24,425 (Sept. 6, 1984) (statement of Rep. Waxman)), as this entire case concerns
what that provision recognizes to be “new.”
14
clear that NCE exclusivity turns on whether the active ingredient of the new drug was previously
approved as an active ingredient.
Sandoz’s remaining arguments do not persuade.
First, both of Sandoz’s cases are inapposite. See Pl.’s Mem. at 24 (citing Baker Norton
Pharms., Inc. v. FDA, 132 F. Supp. 2d 30 (D.D.C. 2001), and Pharmanex v. Shalala, 221 F.3d
1151 (10th Cir. 2000)). Baker Norton states that the NDA process “consider[s] the safety of all
parts of any new drug,” but it does so only to contrast the scope of § 505 of the FDCA from that
of the Orphan Drug Act. 132 F. Supp. 2d at 35. The case did not address whether the FDA
individually approves each ingredient and impurity in its approved drugs. See id. Moreover,
although Pharmanex rejects as “too simple” the suggestion that “ingredients are in no sense
‘approved’ in the new drug approval process,” its only proffered nuance is that the “approval of
active ingredients is integral to the overall new drug approval process.” 221 F.3d at 1156 (citing
21 CFR § 314.50(d)(1)). The case thus says nothing about whether the FDA can be said to
“approve[]” drug components other than active ingredients. 21 U.S.C. § 355(j)(5)(F)(ii).
Second, Sandoz gets no milage out of § 355(j)(5)(F)(iii). See Pl.’s Mem. at 25–26; Pl.’s
Reply at 7–9. That provision confers three years of exclusivity, even when a drug’s active
ingredient already “has been approved” in a prior NDA, if the application “contains reports of new
clinical investigations . . . [that were] essential to the approval of the application and conducted or
sponsored by the applicant.” 21 U.S.C. § 355(j)(5)(F)(iii). The provision thus confers a shorter
term of exclusivity for producing new and valuable clinical investigations, rather than for relying
on new active ingredients (however defined). See id. In this respect, the provision does not support
either party’s interpretation of § 355(j)(5)(F)(ii), as both enable some set of drugs to obtain
exclusivity under § 355(j)(5)(F)(iii). And contrary to Sandoz’s objection, the FDA’s interpretation
15
“is defining the line between three- and five-year exclusivity, not eviscerating it.” Actavis
Elizabeth, 625 F.3d at 765.
Finally, this Court’s interpretation of § 355(j)(5)(F)(ii) will not give any company a
“windfall,” Pl.’s Mem. at 27. Sandoz laments that the FDA’s reading of § 355(j)(5)(F)(ii) would
“extend exclusivity to those who simply adapt existing chemicals in new ways,” even if those
chemicals had long been in use as inactive ingredients. Pl.’s Reply at 8. But Congress could have
reasonably intended that outcome, on the thought that adapting existing chemicals to be active
ingredients is a significant, scientific contribution. See A.R. 1443 (noting that teriflunomide was
“not an approved drug product anywhere in the world” at the time of Aubagio’s approval). And
as discussed above, Sandoz has failed to show that Congress shared its particular account of the
statute’s purpose, which closely resembles the FDA’s alternative. See supra.
For those reasons, the Court holds that NCE exclusivity turns on whether the active
ingredient of the new drug was approved as an active ingredient in any prior drug. Accordingly,
Aubagio is entitled to that exclusivity unless the FDA previously approved teriflunomide as an
active ingredient.
C. Teriflunomide Was Not Previously Approved as an Active Ingredient
The FDA did not previously approve teriflunomide as an active ingredient. As discussed
above, the agency approved Arava on the understanding that leflunomide was its only active
ingredient. See A.R. 265 (listing Arava’s ingredients); A.R. 326 (describing Arava’s mechanism
of action); A.R. 1454–56 (discussing Arava’s approval). Moreover, in approving Arava, the
agency assessed teriflunomide only as an impurity. See A.R. 264 (noting that Arava contains the
impurity “A77 1726”); A.R. 271 (discussing the accumulation of “A77 1726” in the drug product);
A.R. 1455 n.28 (noting that “A77 1726” refers to teriflunomide). Finally, Sandoz has not identified
16
any other drug, prior to Aubagio, in which the FDA approved teriflunomide as an active ingredient.
For those reasons, the FDA properly concluded that “no active ingredient of [Aubagio] ha[d] been
approved in any other” NDA. 21 U.S.C. § 355(j)(5)(F)(ii).
In addition, Sandoz has failed to show that teriflunomide is an active ingredient in Arava.
See Pl.’s Mem. at 32–39; Pl.’s Reply at 16–24. The FDA defines “active ingredient” to mean “any
[drug] component 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 body.” 21 C.F.R. § 314.3(b). Under that definition, which Sandoz does not
challenge in this case, a drug component qualifies as an active ingredient only if it is “intended”
to play a pharmacological role. Id. And here, Sandoz has failed to show that Sanofi intended
teriflunomide to play a pharmacological role in Arava. Sandoz invokes patent materials that claim
a positive interaction between leflunomide and teriflunomide in general terms. See Pl.’s Mem. at
33–34. But it is undisputed that the teriflunomide is present in Arava only due to the degradation
of leflunomide. See A.R. 1459–60. Sanofi does not add teriflunomide to Arava deliberately. See
id. Nor did Sanofi tout teriflunomide’s purported pharmacological benefits in applying for Arava’s
approval. And although Sandoz argues that the presence of teriflunomide in a Arava results from
Sanofi’s chosen method of manufacturing, there is no indication that Sanofi adopted that method
with the “inten[t]” for degraded teriflunomide to “furnish pharmacological activity”—as 21 C.F.R.
§ 314.3(b) requires.
The combination of two patents that Sanofi submitted before the approval of Aubagio—
Canadian Patent Application No. 2,201,040 and United States Patent No. 7,071,222—does not
suggest the necessary intent. See Pl.’s Mem. at 33–34; Pl.’s Reply at 23–24. Sandoz reads the
Canadian patent to show that Sanofi “knew some portion of a leflunomide drug product would
17
inevitably degrade into teriflunomide during manufacturing and storage,” and the United States
patent to show that Sanofi “regarded the supplementation of leflunomide with small amounts of
teriflunomide as therapeutically beneficial.” Pl.’s Reply at 23–24. It accordingly argues that the
“patents together demonstrate Sanofi’s intent” to harness degraded teriflunomide for a beneficial
effect. Id. at 24 (emphasis omitted). But the Canadian application concerned a technique for
slowing the degradation of leflunomide during its storage, on the ground that such degradation
“make[s] difficult an accurately dosed, constant, controlled administration of [that substance].”
A.R. 389–90. And the United States patent concerned the benefits of deliberately adding fixed
amounts of teriflunomide to fixed amounts of leflunomide, as opposed to the benefits of allowing
leflunomide to degrade into teriflunomide over time. See A.R. 398, 400. The combination of
those patents does not suggest that Sanofi intended the degradation of leflunomide in Arava to
have a beneficial effect. See 21 C.F.R. § 314.3(b). To the contrary, the Canadian application
suggests that Sanofi may have intended to slow that degradation altogether. 6
Finally, the fact that the leflunomide in Arava metabolizes to teriflunomide in vivo is
inapposite. See Pl.’s Mem. at 26. The FDA has elected to determine drugs’ active ingredients by
looking to their chemical structure ex vivo. See A.R. 1580–82. The D.C. Circuit has previously
approved that approach. See Actavis Elizabeth, 625 F.3d at 764–66. And Sandoz declined to
challenge the approach in its administrative proceedings. See A.R. 1401 (declining to argue that
“NCE exclusivity for Aubagio should be rescinded merely because teriflunomide is the active
6
Sandoz also invokes two efforts by Sanofi to enforce foreign patents: one before a New Zealand
court in 2009 and the other before a German court in 2011. See Pl.’s Mem. at 34. But it is hard to
see how patent litigation in those years could bear on Sandoz’s intent either during the
development of Arava in the 1990s, 21 C.F.R. § 314.3(b), or when the drug was “approved” in
1998, 21 U.S.C. § 355(j)(5)(F)(ii).
18
metabolite of leflunomide”). Any challenge to the approach thus falls outside the scope of this
case.
For those reasons, the Court holds, first, that the FDA did not approve teriflunomide as an
active ingredient in Arava and, second, that teriflunomide is not an active ingredient in that drug. 7
Aubagio thus was entitled to NCE exclusivity. See 21 U.S.C. § 355(j)(5)(F)(ii).
D. The FDA’s Decision Was Neither Arbitrary Nor Capricious
The remainder of this case is straightforward. Because Aubagio was entitled to NCE
exclusivity, no company could “submit[]” an ANDA that “refer[red] to” it within four years of its
approval. Id. The FDA approved Aubagio on September 12, 2021. A.R. 1456. Sandoz filed its
first ANDA for teriflunomide, however, on September 7, 2016—five days before Aubagio’s
exclusivity expired. See A.R. 1403. The FDCA thus required the FDA to reject Sandoz’s first
ANDA as untimely. See 21 U.S.C. § 355(j)(5)(F)(ii). And as such, the agency’s decision was not
“arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C.
§ 706(2)(A); see Compl. ¶¶ 66–92.
The Court need not resolve Sandoz’s remaining argument regarding the separate drug,
Zegerid. See Pl.’s Mem. at 39–43; Pl.’s Reply at 24–26. It is true that the FDA must “treat similar
cases in a similar manner unless it can provide a legitimate reason for failing to do so.” Kreis v.
Secretary of the Air Force, 406 F.3d 684, 687 (D.C. Cir. 2005) (citation omitted). But on Sandoz’s
7
Because the Court concludes that teriflunomide is not an active ingredient in Arava, the Court
need not decide whether § 355(j)(5)(F)(ii) requires the FDA to reconsider that question over
twenty years after Arava’s approval. See 21 U.S.C. § 355(j)(5)(F)(ii) (using the present perfect
tense to ask whether an active ingredient “has been approved” in a prior NDA). In addition, the
Court need not decide whether the patent materials in this case suffice to show that degraded
teriflunomide has “pharmacological activity” in Arava, 21 C.F.R. § 314.3(b). Compare Gov’t’s
Mem. at 26–27 (arguing that the FDA “has not been presented with scientific evidence” of that
activity), with Pl.’s Reply at 23 (arguing that the agency “cannot ignore scientific disclosures just
because they appear in patents rather than journals”).
19
own account, Zegerid is only similar to Arava if the latter contains two active ingredients, see Pl.’s
Mem. at 40–41, which it does not. For that reason, the Court does not “consider it necessary to go
into the details of” the Zegerid decision. Actavis Elizabeth, 625 F.3d at 766.
CONCLUSION
For the above reasons, this Court will deny Sandoz’s Motion for Summary Judgment, Dkt.
14, and grant the government’s Cross Motion for Summary Judgment, Dkt. 20. A separate order
consistent with this decision accompanies this memorandum opinion.
________________________
DABNEY L. FRIEDRICH
United States District Judge
July 22, 2022
20