UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
_________________________________________
)
ATHENEX INC., et al., )
)
Plaintiffs, )
)
v. ) Case No. 19-cv-00603 (APM)
)
ALEX M. AZAR II, et al. )
)
Defendants. )
_________________________________________ )
MEMORANDUM OPINION
I. INTRODUCTION
In 2013, Congress added Section 503B to the Food, Drug, and Cosmetic Act, creating a
new category of drug manufacturer known as an “outsourcing facility.” Such facilities are
permitted to sell large quantities of compounded drugs directly to hospitals and health care
professionals without receiving premarket approval from the Food and Drug Administration
(“FDA”), so long as the facility satisfies certain statutory requirements. One such requirement
concerns when an outsourcing facility can use a “bulk drug substance”—that is, a drug’s active
ingredient—to compound a drug product. In such cases, FDA must declare there to be a “clinical
need” for the “bulk drug substance” as a precondition to compounding with such substance.
Plaintiffs presently operate an outsourcing facility that produces a drug product using the
bulk drug substance vasopressin. On March 4, 2019, FDA announced that vasopressin is not a
“bulk drug substance for which there is a clinical need” because already available on the market is
an FDA-approved drug that could be used to meet the very patient needs that Plaintiffs claim make
it necessary to bulk compound using vasopressin. FDA’s decision, if valid, forecloses Plaintiffs
from selling their product.
Plaintiffs filed this action challenging FDA’s refusal to place vasopressin on the Section
503B “clinical need” list. They contend that FDA’s inquiry improperly took the availability of a
branded drug product into account when assessing “clinical need.” Instead, Plaintiffs argue, the
“clinical need” inquiry should be confined to simply asking whether there is a therapeutic use for
the bulk drug substance, and vasopressin easily meets that standard. Alternatively, Plaintiffs
maintain that, even under FDA’s reading of “clinical need,” vasopressin qualifies.
The court finds that FDA’s method of determining whether there is a “clinical need” for a
bulk drug substance gives effect to the unambiguously expressed intent of Congress. The court
also concludes that FDA’s exclusion of vasopressin from the “clinical need” list was not arbitrary
and capricious. Accordingly, the court grants Defendants’ Cross-Motions for Summary Judgment
and denies Plaintiffs’ Motion for Summary Judgment.
II. BACKGROUND
A. Factual Background
1. Compounding
In the pharmaceutical drug industry, the term “compounding” refers to the act of
“combining, admixing, mixing, diluting, pooling, reconstituting, or otherwise altering of a drug or
bulk drug substance to create a drug.” 21 U.S.C. § 353b(d)(1). The compounding of drugs
performs “an important role for patients for whom an FDA-approved drug is not appropriate.”
Administrative Record, ECF No. 35 [hereinafter R.], at 89. Take the case of a person who is
allergic to a particular ingredient in an FDA-approved drug. Through compounding, an approved
drug can be reconstituted to eliminate the allergen while still delivering the drug’s therapeutic
benefit. Id. Another example is that certain patient populations may have difficulty ingesting a
2
drug in its approved pill form. Such a drug can be compounded to liquid form for easier
consumption.
Historically, local pharmacists have been responsible for compounding drugs, typically in
small quantities to fill the needs of individual patients with valid prescriptions. That practice still
exists today. FDA has not in the past, nor does it now, require local pharmacists to obtain
regulatory approval before creating and selling a compounded drug product. See generally
21 U.S.C. § 353a (Section 503A of the Food, Drug, and Cosmetic Act).
Over time, however, a new kind of drug manufacturer emerged, one that compounded
drugs in bulk quantities and marketed such drugs directly to hospitals and physicians. Even though
they manufactured on a larger scale, these bulk compounders, like local pharmacists, were not
required to subject their compounded drug products to “premarket review for safety, effectiveness,
and quality” by FDA before bringing them to market. R. at 89.
2. Drug Quality and Security Act
The era of lightly regulated bulk compounding ended following a tragic event. In 2012,
the New England Compounding Center (“NECC”) in Framingham, Massachusetts, produced
contaminated injections that caused a meningitis outbreak, killing more than 60 people and
infecting hundreds more. See id. Within fourteen months of the NECC outbreak, Congress passed
the Drug Quality and Security Act (“DQSA”), which added Section 503B to the Food, Drug, and
Cosmetic Act (“FDCA”). Section 503B created a new category of drug maker called an
“outsourcing facility.” See 21 U.S.C. § 353b. An outsourcing facility may compound drug
products in large quantities without obtaining a prescription for “an identified individual
patient[ ].” Compare 21 U.S.C. § 353b(d)(4)(C) with 21 U.S.C. § 353a. Such facilities are
3
permitted to sell bulk compounded drug products to health care practitioners and hospitals as
“office stock,” for providers to have available and to use on an as-needed basis. See R. at 52.
Section 503B also enhanced FDA’s authority to regulate outsourcing facilities. An
outsourcing facility remains exempt from the FDCA’s premarket approval requirements and
certain labeling and supply-chain requirements, but only if it satisfies eleven statutory criteria.
See 21 U.S.C. § 353b(a) (exempting compounded drugs from 21 U.S.C. §§ 352(f)(1), 355, and
360eee-1). A non-exhaustive list of these criteria include: (1) the manufactured drug is
compounded at a facility that is subject to specified FDA licensing, reporting, fees, and inspection
requirements, id. §§ 353b(a)(1), (a)(9), (a)(11), 353b(b); (2) the ingredients used in the
compounding process meet national standards, id. § 353b(a)(3); (3) FDA has not deemed the drug
to be unsafe, nor difficult to compound, id. §§ 353b(a)(4), (a)(6); (4) “the drug is not essentially a
copy of one or more approved drugs,” id. § 353b(a)(5); (5) the drug is labeled in accordance with
statutory requirements, id. § 353b(a)(10); and (6) the drug will not be sold or transferred by an
entity other than the outsourcing facility, id. § 353b(a)(8).
In addition to the foregoing requirements, Section 503B authorizes two ways in which to
bulk compound a drug. The first is known as “sterile-to-sterile compounding.” Under this
approach, an outsourcing facility starts with an FDA-approved drug, disassembles it, and alters it
into a finished drug product for sale. The second method is known as “bulk compounding.” That
process starts with a “bulk drug substance,” instead of an FDA-approved drug. A “bulk drug
substance” is an ingredient that furnishes pharmacological activity and is incorporated into a
finished drug product to achieve the intended purpose of compounding, such as removal of an
4
allergen. 1 The FDA requires that a drug must be “compounded in an outsourcing facility that does
not compound using bulk drug substances . . . , unless the bulk drug substance appears on a list
established by the Secretary identifying bulk drug substances for which there is a clinical
need . . .,” id. § 353b(a)(2)(A)(i), or “the drug compounded from such bulk drug substance appears
on the drug shortage list,” id. at (A)(ii). Thus, an outsourcing facility’s legal authority to compound
using a bulk drug substance depends in the first instance on a listing determination by FDA.
3. FDA’s Decision Not to List Vasopressin
Plaintiffs Athenex Inc. and its affiliated entities operate an outsourcing facility that
compounds with a bulk drug substance called vasopressin. Vasopressin “increase[s] blood
pressure in adults with vasodilatory shock . . . who remain hypotensive despite fluids and
catecholamines.” R. at 27. As a result of the DQSA, Plaintiffs cannot continue to manufacture
their bulk drug product using vasopressin, unless FDA places it on the “clinical need” list. The
“drug shortage” list is not at issue in this case.
Plaintiffs began the process to get vasopressin listed years ago. In 2013 and 2014, FDA
issued a notice seeking nominations of bulk drug substances to place on the “clinical need” list,
and FDA provided “more detailed information on what it needs to evaluate a nomination.” See R.
at 1, 4. In 2015, FDA again clarified the requirements for a nomination. See id. at 9–13. Fifteen
months later, in January 2017, FDA issued an “interim regulatory policy” for the “clinical need”
list. See id. at 34–45. The interim guidance placed each of the nominated substances into one of
three categories: Category 1 substances signified those under evaluation; Category 2 included
1
A bulk drug substance is defined as “any substance that is intended for incorporation into a finished drug product
and 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. § 207.1 (synonymous with
“active pharmaceutical ingredient”). A finished drug product means “a finished dosage form (e.g., tablet, capsule, or
solution) that contains at least one [bulk drug substance], generally, but not necessarily, in association with other
ingredients in finished package form suitable for distribution to pharmacies, hospitals, or other sellers or dispensers
of the drug products to patients or consumers.” Id.
5
substances that raise significant safety risks; and Category 3 identified substances without
adequate support for listing. See id. at 40–41. FDA classified vasopressin as a Category 1
substance. See id. at 45. For Category 1 substances, like vasopressin, FDA stated that it “d[id]
not intend to take action against an outsourcing facility for compounding a drug using a bulk drug
substance” during the interim review period. See id. at 43. Therefore, Plaintiffs continued to
manufacture their bulk drug product while vasopressin remained under review for “clinical need”
listing.
As of early 2018, FDA had yet to explain how it would determine whether a bulk drug
substance would qualify for the “clinical need” list. That changed in March 2018, when the agency
issued a draft guidance document titled, “Evaluation of Bulk Drug Substances Nominated for Use
in Compounding Under Section 503B of the Federal Food, Drug, and Cosmetic Act.” See id. at
46–67. The draft guidance stated that, if the bulk drug substance is a component of an FDA-
approved drug product, the agency will consider two threshold questions. FDA framed the first
question as follows:
Is there a basis to conclude, for each FDA-approved product that
includes the nominated bulk drug substance, that (i) an attribute
of the FDA-approved drug product makes it medically
unsuitable to treat certain patients for a condition that FDA has
identified for evaluation, and (ii) the drug product proposed to
be compounded is intended to address that attribute?
Id. at 60. FDA explained the reason for asking this first question is that, “[u]nless an attribute of
the FDA-approved drug is medically unsuitable for certain patients, and a drug product
compounded using a bulk drug substance that is a component of the approved drug is intended to
address that attribute, there is no clinical need to compound using that bulk drug substance.” Id.
As to the second threshold inquiry, FDA would ask:
6
Is there a basis to conclude that the drug product proposed to be
compounded must be produced from a bulk drug substance
rather than from an FDA-approved drug product?
Id. at 60. If the answer to either threshold question is “no,” FDA stated, it would not list the bulk
drug substance. See id. If the agency determines the answer is “yes” to both questions—or if the
bulk drug substance is not a component of an FDA-approved drug product—the agency would
make a second inquiry. At this step, FDA would consider various additional factors, including the
“physical and chemical characterization of the substance; any safety issues raised by the use of the
substance in compounding; the available evidence of effectiveness or lack of effectiveness of a
drug product compounded with the substance, if any such evidence exists; and current and
historical use of the substance in compounded drug products.” Id. As to this second-step inquiry,
FDA said that it would “conduct a balancing test . . . under which [it] would consider each factor
in the context of others and to balance them, on a substance-by-substance basis, to determine
whether the substance is appropriate for inclusion” on the “clinical need” list. Id.
In August 2018, FDA announced its intention not to include vasopressin on the “clinical
need” list. See id. at 109. The agency followed the two-step inquiry described in the draft
guidance. FDA decided that vasopressin did not make it past the first step. See id. FDA explained
that the nomination of vasopressin “[did] not identify an attribute of Vasostrict,” the FDA-
approved drug containing vasopressin, “that makes [Vasostrict] medically unsuitable for patients.”
Id. Thus, there was no “clinical need” to bulk compound with vasopressin. FDA also specifically
rejected the need for vasopressin at a higher concentration than the approved product. Id.
Plaintiffs submitted a comment challenging FDA’s intent to exclude vasopressin. Plaintiffs
argued that FDA had misinterpreted Section 503B and that, properly understood, the statute simply
requires the agency to answer whether there is a “clinical need” for the bulk drug substance itself,
7
without reference to the availability and suitability of an FDA-approved product. Id. at 273–78.
Thus, under Plaintiffs’ reading, vasopressin meets the standard of “clinical need” simply because
it has a demonstrated therapeutic value. Id. Alternatively, Plaintiffs maintained that, even under
FDA’s interpretation of Section 503B, vasopressin qualified for listing. Specifically, Plaintiffs
asserted that Vasostrict was medically unsuitable to treat certain patients because: (1) Vasostrict
contains chlorobutanol, an allergen for some patients, and (2) it is not available in ready-to-use
form but must be diluted before use. See id. at 256. Plaintiffs’ product, they insisted, filled these
therapeutic gaps.
Rejecting both arguments, FDA issued a final notice on March 4, 2019, stating that it would
not place vasopressin on the “clinical need” list. See id. at 1600–05. As to Plaintiffs’ statutory
interpretation argument, FDA adhered to its understanding that “Congress intended for the
Agency” as part of the “clinical need” assessment “to evaluate the need for outsourcing facilities
to use the bulk drug substances to compound drug products.” Id. at 69. With respect to Plaintiffs’
alternative position, FDA found lacking the asserted needs for bulk-produced vasopressin. FDA
observed that there is available a chlorobutanol-free formulation of Vasostrict, and Plaintiffs
“fail[ed] to explain why that formulation would be medically unsuitable for patients who have an
allergy to chlorobutanol.” See id. at 1604. The agency also rejected the need for a ready-to-use
product, concluding that Plaintiffs “d[id] not show that [Vasostrict], when not manufactured in the
ready-to-use form, is medically unsuitable for certain patients.” Id. at 1605. Furthermore, even if
Vasostrict were medically unsuitable, the agency found that Plaintiffs did not “establish that drug
products in the relevant concentrations, including ready-to-use products, cannot be prepared from
[Vasostrict]”—and instead must be prepared using vasopressin. Id. Other commenters had noted
that a different outsourcing facility, owned by PharMEDium, compounds a ready-to-use drug with
8
Vasostrict as its starting material. Id. at 198. Consequently, FDA declined to find a “clinical need”
for bulk-compounded vasopressin under Section 503B.
B. Procedural Background
On March 4, 2019, the very same day that FDA announced its final decision not to list
vasopressin, Plaintiffs filed suit under the Administrative Procedure Act (“APA”) and moved for
a temporary restraining order and preliminary injunction. See Compl. ECF No. 1 [hereinafter
Compl.]; see also Pls.’ Mot. for Temporary Restraining Order and Preliminary Injunction, ECF
No. 2. FDA agreed to stay any enforcement action with respect to Plaintiffs’ continued bulk
compounding of vasopressin until the court issued a decision on the merits. See Joint Mot. to Enter
Scheduling Order, ECF No. 19. The court therefore denied the motion and entered a scheduling
order for summary judgment briefing. See Order, ECF No. 20; see also Order, ECF No. 22. The
court also permitted Par Sterile Products, LLC and Endo Par Innovation Company, LLC, the
owners of Vasostrict, to intervene as defendants. See Order, ECF No. 20.
Plaintiffs filed their Motion for Summary Judgment on March 18, 2019, see generally Pls.’
Mot., ECF No. 23 [hereinafter Pls.’ Mot.], and the Federal Defendants 2 and Defendants Par Sterile
Products and Endo Par Innovation Company filed their motions on March 29, 2019, see FDA
Def.’s Mot., ECF No. 25 [hereinafter FDA Mot.]; see also Par Defs.’ Mot., ECF No. 27. The court
held oral argument on the motions on April 30, 2019.
III. LEGAL STANDARD
In cases that involve review of final agency action under the APA, the district court “sits
as an appellate tribunal” and “the entire case on review is a question of law.” Am. Biosci., Inc. v.
2
The Federal Defendants are Alex M. Azar, II, in his official capacity as Secretary of Health and Human Services;
the U.S. Department of Health and Human Services; Scott Gottlieb, in his official capacity as Commissioner of FDA;
and FDA.
9
Thompson, 269 F.3d 1077, 1083 (D.C. Cir. 2001) (internal quotations omitted). The court’s review
is limited to the administrative record, and “its role is limited to determining whether or not as a
matter of law the evidence in the administrative record permitted the agency to make the decision
it did.” Philip Morris USA Inc. v. FDA, 202 F. Supp. 3d 31, 45 (D.D.C. 2016) (cleaned up).
When reviewing an agency’s interpretation of a statute, the court employs the two-step
inquiry set out in Chevron U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837 (1984).
See City of Arlington v. FCC, 569 U.S. 290, 296–97 (2013). At step one, the court must determine
whether “Congress has directly spoken to the precise question at issue,” and if Congress has, the
court “must give effect to the unambiguously expressed intent of Congress.” Chevron, 467 U.S.
at 842–43. The court must “employ the traditional tools of statutory construction to interpret the
statute and to resolve any ambiguities.” Dist. of Columbia v. Dep’t of Labor, 819 F.3d 444, 449
(D.C. Cir. 2016). It must analyze “the language itself, the specific context in which that language
is used, and the broader context of the statute as a whole.” Robinson v. Shell Oil Co., 519 U.S.
337, 341 (1997). Furthermore, for a plaintiff to prevail under Chevron step one, it “must do more
than offer a reasonable or, even the best, interpretation; it must show the statute unambiguously
forecloses the [agency’s] interpretation.” Village of Barrington v. Surface Transp. Bd., 636 F.3d
650, 661 (D.C. Cir. 2011) (emphasis in original).
If Congress has not spoken directly to the issue—because “the statute is silent or
ambiguous”—the court, at Chevron step two, must side with the agency if its answer “is based on
a permissible construction of the statute.” Chevron, 467 U.S. at 843. An agency receives
deference at Chevron step two if it has “offered a reasoned explanation for why it chose [its]
interpretation.” Village of Barrington, 636 F.3d at 660.
10
IV. DISCUSSION
A. Chevron Step One
1. The Provision’s Plain Meaning
The court starts with the “language itself.” Robinson, 519 U.S. at 341. The disputed
portion of Section 503B requires “the drug [to be] compounded in an outsourcing facility that does
not compound using bulk drug substances . . . unless the bulk drug substance appears on a list
established by the Secretary identifying bulk drug substances for which there is a clinical
need . . . .” 21 U.S.C. § 353b(a)(i). Plaintiffs interpret the phrase “bulk drug substances for which
there is a clinical need” to require FDA to focus exclusively on the bulk drug substance without
looking to “which branded drugs exist at the time.” See Pls.’ Mot., Pls.’ Mem., ECF No. 23-1
[hereinafter Pls.’ Mem.], at 18. Plaintiffs reason that because “clinical need” modifies “bulk drug
substance,” FDA must ask only whether there is a clinical need for vasopressin, i.e., whether
vasopressin is needed for patient treatment, and nothing more. See id. at 15–17. By that standard,
Plaintiffs maintain, vasopressin easily qualifies, as evidenced by the fact that it is an active
ingredient in an FDA-approved drug. See id. at 18; see also Compl. ¶ 54. Defendants, on the other
hand, understand the provision to require FDA to consider whether a finished bulk drug product
containing vasopressin is necessary relative to FDA-approved products containing vasopressin.
See FDA Mot., FDA Mem., ECF No. 25-1 [hereinafter FDA Mem.], at 17–18.
The term “clinical need” does not lend itself to a straightforward reading. It is not defined
in the FDCA. Nor does it appear anywhere else in the United States Code. Turning to the
dictionary for guidance, “clinical” means “relating to the observation and treatment of actual
patients rather than theoretical or laboratory studies,” OXFORD DICTIONARY, 3 and “need” means
3
https://en.oxforddictionaries.com/definition/clinical (last visited July 31, 2019).
11
“circumstances in which something is necessary; necessity,” OXFORD DICTIONARY. 4 Thus,
Congress required FDA to determine whether a bulk drug substance is “necessary” for the
“treatment of actual patients.” That question, however, begs another one: “Necessary” relative to
what? Whether something is “necessary” cannot be determined in a vacuum. The presence or
absence of “need” must be measured against some point of reference. Is vasopressin “necessary”
as a bulk drug substance in the sense that physicians “need” it to treat patients? The answer is
“yes.” Vasopressin has therapeutic value for patients and therefore it is “needed” for treatment.
On the other hand, is vasopressin “necessary” as a component of a bulk drug product such that,
without it, physicians would be unable to treat patients? The answer is “no.” The FDA-approved
drug Vasostrict can serve the same clinical purpose as Plaintiffs’ product. Thus, whether there is
a “clinical need” for a particular bulk drug substance depends on how one frames the question.
To address this ambiguity, the court looks to the statutory purpose of the term “clinical
need.” United States v. Cordova, 806 F.3d 1085, 1099 (D.C. Cir. 2015). Congress plainly thought
that there are some bulk drug substances for which there is a “clinical need” and others for which
there is not. Otherwise, why else would Congress direct FDA to answer the question? Congress
therefore meant for the “clinical need” inquiry to perform a sorting function. FDA’s understanding
of “clinical need” carries out this purpose; Plaintiffs’ does not. This becomes clear when a bulk
drug substance is thought of as an active pharmaceutical ingredient. See 21 C.F.R. § 207.1
(defining “bulk drug substance” the same as “active pharmaceutical ingredient”). An “active
pharmaceutical ingredient” is “any substance that is intended for incorporation into a finished drug
product and 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
4
https://en.oxforddictionaries.com/definition/need (last visited July 31, 2019).
12
the body.” Id. Under Plaintiffs’ reading, there always will be a “clinical need” for any
compoundable active pharmaceutical ingredient because, by definition, such an ingredient is one
that “furnishes pharmacological activity” or effects “diagnosis, cure, mitigation, treatment, or
prevention of disease.” In other words, it delivers some therapeutic benefit. So, for example,
Plaintiffs say, there is a “clinical need” for vasopressin because it is used to treat life-threatening
and emergency conditions, such as vasodilatory shock, diabetes insipidus, and gastrointestinal
bleeding. See Compl ¶¶ 3, 54. By the same logic, there is a “clinical need” for ibuprofen, because
it is used to reduce fever and treat pain or inflammation. But reading “clinical need” in this way
does not create a category of active pharmaceutical ingredients for which there is not a “clinical
need.” Only when the question considers the actual way in which the active pharmaceutical
ingredient supplies a therapeutic benefit—by its administration as a finished drug product—does
the inquiry produce the kind of categorization that Congress surely envisioned.
Plaintiffs balk at this analysis, arguing that the agency’s framing of “clinical need”
improperly replaces “bulk drug substance,” as that term appears in Section 503B(a)(2)(A), with
the term “compounded drug product.” See Pls.’ Reply Mem. Supporting Pls.’ Mot. and Opp’n to
Defs.’ Cross-Mots., ECF No. 30 [hereinafter Pls.’ Reply], at 9–11. More concretely, Plaintiffs
argue that FDA’s approach improperly focuses on vasopressin as incorporated into a drug product,
and not strictly as a bulk drug substance. This construction, Plaintiffs argue, is not faithful to
Section 503B’s plain text. Id.
To be sure, FDA’s approach to “clinical need” involves considering the bulk drug
substance as a finished product, and the term “finished product” does not appear in Section
503B(a)(2)(A). The court, however, must read a statute to give its words the effect that Congress
intended. See United States v. Braxtonbrown-Smith, 278 F.3d 1348, 1352 (D.C. Cir. 2002) (stating
13
that the court “must avoid an interpretation that undermines congressional purpose considered as
a whole when alternative interpretations consistent with the legislative purpose are available”)
(citing United States v. Am. Trucking Ass’ns, 310 U.S. 534, 543 (1940) (“[E]ven when the plain
meaning did not produce absurd results but merely an unreasonable one ‘plainly at variance with
the policy of the legislation as a whole’ this Court has followed that purpose, rather than the literal
words.”) (citation omitted)). Here, as discussed, Plaintiffs’ reading would produce the
unreasonable result that there is a “clinical need” for every bulk drug substance because, by
definition, a bulk drug substance is one “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. § 207.1. Plaintiffs’ interpretation thus draws no
actual distinction among bulk drug substances—there is a “clinical need” for all. Only when
“clinical need” is assessed against the availability and suitability of an approved drug does the
term perform the classifying function that Congress intended. Defendants’ reading therefore
faithfully adheres to Section 503B’s text, while Plaintiffs’ does not.
2. The Provision in Context
Viewing the term “clinical need” in its statutory context lends further support to
Defendants’ reading of Section 503B, and further frustrates Plaintiffs’ interpretation. See King v.
Burwell, 576 U.S. __, __, 135 S. Ct. 2480, 2489 (2015) (stating “oftentimes the meaning—or
ambiguity—of certain words or phrases may only become evident when placed in context”) (citing
FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 132 (2000)) (internal quotations
omitted).
14
a. Bulk compounding as an exception within an exception
Critical to the court’s interpretation of “clinical need” is how the subsection containing that
provision, 503B(a)(2)(A)(i), lines up with other parts of Section 503B. The very first sentence of
Section 503B exempts outsourcing facilities from the FDCA’s new-drug approval process, certain
labeling requirements, and certain supply chain requirements. See 21 U.S.C. § 353b(a) (exempting
compounded drugs from 21 U.S.C. §§ 352(f)(1), 355, and 360eee-1). Products made by
outsourcing facilities are thus an exception to the FDCA’s usual requirements for new drug
products. Next, Section 503B(a)(2) requires outsourcing facilities to compound “in an outsourcing
facility that does not compound using bulk drug substances,” generally, subject only to enumerated
exceptions. 21 U.S.C. § 353b(a)(2) (emphasis added). Section 503B(a)(2)(A)(i)–(ii) specifies the
exceptions, authorizing compounding with a bulk drug substance—but only if FDA lists the bulk
drug substance on the “clinical need” list, or the drug compounded from the bulk drug substance
appears on FDA’s “drug shortage” list.
As the above recitation shows, Congress viewed compounding with a bulk drug substance
as an exception within an exception. Section 503B’s initial exception exempts from the FDCA’s
new-drug requirements compounded drug products that start with an FDA-approved product. The
exception to this exception is compounding with a bulk drug substance. See 21 U.S.C.
§ 353b(a)(2)(A)(i)–(ii) (“The drug is compounded in an outsourcing facility that does not
compound using bulk drug substances . . . unless . . .) (emphasis added). Outsourcing facilities
may resort to compounding with a bulk drug substance only if: FDA determines that there is a
“clinical need” for the bulk drug substance, or FDA places the compounded drug product on its
“drug shortage” list. As the terms “need” and “shortage” imply, Congress meant for FDA to
approve compounding with a bulk drug substance not as a matter of course. See Gustafson v.
15
Alloyd Co., Inc., 513 U.S. 561, 575 (1995) (stating the principle of construction that a “word is
known by the company it keeps”). FDA’s approach to assessing “clinical need” is consistent with
this design.
Under Plaintiffs’ interpretation, however, “clinical need” loses its character as an exception
within an exception. According to Plaintiffs, a “clinical need” exists for a bulk drug substance if
it has some therapeutic value. See Pls.’ Mem. at 18. But, if that were so, then the term “clinical
need” would not connote a modest “list” but an extensive catalog of active ingredients. According
to Plaintiffs, that catalog would include every active ingredient contained in an FDA-approved
drug plus active ingredients that are “not components of FDA-approved drugs . . . if their clinical
need is shown some other way[.]” Id. 5 FDA could keep only “unproven, fringe ingredients” off
the list. Id. Such a reading, however, swallows Congress’s general directive that an outsourcing
facility “not compound using bulk drug substances . . . .” 21 U.S.C. § 353b(a)(2) (emphasis added).
And it would mean that Congress authorized large-scale compounding with a wide array of bulk
drug products through the “clinical need” list. If Congress had intended to sanction such an
expansive practice, the court would have expected it to say so more clearly. See Whitman v.
Am. Trucking Ass’ns, 531 U.S. 457, 468 (2001) (“Congress, we have held, does not alter the
fundamental details of a regulatory scheme in vague terms or ancillary provisions—it does not,
one might say, hide elephants in mouseholes.”).
5
Plaintiffs never identify how many different bulk drug substances would appear on the “clinical need” list, if the
court were to accept their reading. When FDA invited nominations for the “clinical need” list, commenters submitted
2,000 bulk drug substances for consideration. See R. at 38. Defendants referred the court to the “Orange Book,” a
1,500-page document listing every approved drug product under Section 505 of the Food, Drug, and Cosmetic Act.
See U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES, APPROVED DRUG PRODUCTS (2019),
https://www.fda.gov/media/71474/download (last visited July 31, 2019). Whatever number of active ingredients the
Orange Book recognizes, the total surely exceeds the number that Congress had in mind.
16
Plaintiffs also point to the three additional criteria adjoined with the “clinical need” list:
that the bulk drug substance must comply with a monograph, 6 if applicable; it must be
manufactured at a properly registered establishment; and it must be accompanied by a valid
certificate of analysis. Plaintiffs argue that the “clinical need” list, like the three additional criteria
adjoining it, should focus exclusively on the bulk drug substance. See Pls.’ Mem. at 20 (citing 21
U.S.C. § 353b(a)(2)(B)-(D)). This argument carries little weight because, like the other statutory
criteria, the agency’s interpretation places the bulk drug substance at the center of its “clinical
need” analysis by determining whether the bulk drug substance will fill a gap in a clinical setting.
That the agency takes the qualities of approved products into account to make that assessment does
not result in a conflict between FDA’s interpretation of “clinical need” and the other three criteria
contained in Section 503B(a)(2)(B)–(D).
b. FDA’s involvement in making the “clinical need” list
Another feature of Section 503B that supports Defendants’ interpretation is how Congress
directed FDA to go about creating the “clinical need” list. FDA must: (1) “publish[ ] a notice in
the Federal Register proposing bulk drug substances to be included on the list, including the
rationale for such proposal”; (2) “provide[ ] a period of not less than 60 calendar days for comment
on the notice”; and (3) “publish[ ] a notice in the Federal Register designating bulk drug substances
for inclusion on the list.” 21 U.S.C. § 353b(a)(2)(A)(i)(I)–(III). These procedural prerequisites
6
A “drug monograph” is a “publication that specifies for a drug (or class of related drugs) the kinds and amounts of
ingredients it may contain, the conditions and limitations for which it may be offered, directions for use, warnings,
and other information that its labeling must contain. The monograph may contain important information concerning
interactions with other drugs.” https://medical-dictionary.thefreedictionary.com/drug+monograph (last visited July
31, 2019).
17
make clear that Congress intended for FDA to apply its expertise in determining whether a bulk
drug substance fills a “clinical need.” Defendants’ reading is consistent with Congress’s intent.
Plaintiffs’ construction of Section 503B, on the other hand, would make the agency’s
expertise largely irrelevant to the listing process. As Plaintiffs would have it, FDA must classify
any active ingredient in any approved drug as a bulk drug substance for which there is a “clinical
need.” See Pls.’ Mem. at 18. If that is what Congress intended, then why would it have instructed
FDA to announce its “rationale” for each proposed listing and subject the proposed listing to public
scrutiny? Such notice and comment would be a meaningless exercise for the active pharmaceutical
ingredients found in FDA-approved drugs. Congress clearly wanted FDA to play a substantive
role in the listing process. Only Defendants’ approach allows FDA to perform that function.
Plaintiffs acknowledge that, under their reading, the “clinical need” listing of most bulk
drug substances would not require FDA’s expertise. See Hr’g Tr., ECF No. 36 [hereinafter Hr’g
Tr.], at 26–27 (arguing that FDA will do “some work that’s difficult and . . . some work that’s
easy”); see also Pls.’ Reply at 27. Nevertheless, Plaintiffs contend that “[t]he fact that some cases
are easy because the rationale is immediately apparent, and some are not does not make it pointless
to use notice and comment for the whole enterprise, when that procedure can accommodate both
without disrupting patient care.” Pls.’ Reply at 27. But that reading seems implausible. Congress
could not have meant for FDA to proceed through notice and comment when listing all bulk drug
substances, when that time-consuming and resource-intensive process would serve a meaningful
purpose only for the rare, “difficult” case of a bulk drug substance that is not found in an FDA-
approved drug product. Id. The court must assume that Congress did not impose a burdensome
procedural requirement simply for the sake of imposing it. Cf. Franchise Tax Bd. of Cal. v. U.S.
Postal Service, 467 U.S. 512, 524 (1984) (stating that there is “no reason to believe that Congress
18
intended to impose a meaningless procedural requirement . . .”); Prof’l Air Traffic Controllers
Org. v. Fed. Labor Relations Auth., 685 F.2d 547, 563–64 (D.C. Cir. 1982) (“Congress did not
intend to erect meaningless procedural barriers to effective agency action.”).
Moreover, a related provision suggests that if Congress really wanted all active ingredients
found in FDA-approved products to automatically appear on the “clinical need” list, it passed up
a straightforward way to express that intent. Section 503B’s neighboring provision, Section 503A,
permits traditional compounders, like licensed pharmacists, to compound “for an identified
individual patient.” 21 U.S.C. § 353a(a). For that practice, Congress required traditional
compounders to use a “bulk drug substance” that “compl[ies] with the standards of an applicable”
monograph; if such monograph does not exist, then the compounder may use “drug substances
that are components of drugs approved by the Secretary”; and, if no monograph exists and the drug
substance is not a component of an FDA-approved product, then the drug substance must “appear
on a list developed by the Secretary through regulations issued by the Secretary . . . .” Id.
§ 353a(b)(1)(A)(i)(I)–(III). Section 503A thus spells out three different categories of bulk drug
substances that can be compounded for an individual patient, only one of which requires FDA pre-
approval.
As this provision demonstrates, Congress knows how to permit compounding with drug
substances found in safe and effective FDA-approved products, and it also knows how to define
substances that must “appear on a list developed by the Secretary.” Section 503B reflects the latter
approach but not the former. Congress did not include in Section 503B, as it did in Section 503A,
an allowance for outsourcing facilities to compound using bulk drug substances that either comply
with the standards of an applicable monograph or “are components of drugs approved by the
Secretary.” Instead, Congress mandated an FDA listing before use of a bulk drug substance for
19
mass compounding. This fundamental difference between Sections 503A and 503B supports
Defendants’ interpretation of “clinical need” as requiring more than a mere showing of a bulk drug
substance’s therapeutic value.
Plaintiffs dismiss the significance of the different text in Section 503A. They argue that
the general absence of a pre-production notice-and-comment process in Section 503A (except in
Section 503A(b)(1)(A)(i)(III)) makes sense because it would be nonsensical to require a local
pharmacist to await a rulemaking process by FDA before compounding a drug. By contrast,
Plaintiffs insist, in the context of Section 503B “there is no reason not to use the default notice and
comment procedure for all bulk drug substances in the context of 503B.” Pls.’ Reply at 27.
This argument is unpersuasive for two reasons. First, it misses the point of the textual
difference between the two provisions. If Congress wanted outsourcing facilities to manufacture
using bulk drug substances already recognized in a drug product approved for safety and efficacy,
it had an easy way to do so—it could have simply used the same text found in Section 503A in
Section 503B. Instead, it created a new “clinical need” standard. The court must presume that
this textual difference reflects a purposeful choice by Congress not to equate “clinical need” with
Section 503A’s standards. Second, Plaintiffs’ argument does not explain why Congress would
insist on a time-consuming notice-and-comment process for bulk drug substances found in drug
products whose safety and efficacy are established. As discussed, no good purpose is served by
requiring FDA to certify hundreds, if not thousands, of bulk drug substances already contained in
approved drugs. Congress presumably would not require an agency to commit its resources to a
perfunctory rulemaking process. The textual differences between Sections 503A and 503B
therefore support Defendants’ reading of “clinical need.”
20
c. Plaintiffs’ interpretation renders the “drug shortage list” superfluous
Yet another contextual feature of Section 503B bolsters Defendants’ interpretation: the
“drug shortage list.” Congress allowed outsourcing facilities to compound with a bulk drug
substance, without undergoing notice and comment, if the drug compounded from the bulk drug
substance appears on the “drug shortage list in effect under section 356e of this title at the time of
compounding, distribution, and dispensing.” See 21 U.S.C. § 353b(a)(2)(A)(ii). The drug shortage
list predates the DQSA, and its requirements are codified in 21 U.S.C. § 356e. That provision
directs the Secretary to maintain a list of drugs that are in short supply in the United States for
various listed reasons, including regulatory delay, shortage of an active or inactive ingredient,
delay in shipping of the drug, or demand increase for the drug. See id. § 356e. The list must
include the name of the drug in shortage, its manufacturer, the reason for the shortage, and the
estimated duration of the shortage. Id. § 356e(b).
Plaintiffs’ interpretation of “clinical need,” if accepted, would violate a basic rule of
statutory construction: “[E]very word and every provision is to be given effect and that none
should needlessly be given an interpretation that causes it to duplicate another provision or to have
no consequence.” Nielsen v. Preap, 586 U.S. ___, ___, 139 S. Ct. 954, 969 (2019) (cleaned up
and citation omitted). Recall, under Plaintiffs’ understanding of “clinical need,” FDA would be
required to list every bulk drug substance found in an FDA-approved drug. That list would be
long. If the “clinical need” list were to contain hundreds of active ingredients, then the drug
shortage list would have little or “no consequence.” Id. Not because there would be no drug
shortages. But because any active ingredient contained in a drug placed on the shortage list already
is likely to appear on the “clinical need” list because of its therapeutic value. Stated differently,
FDA rarely, if ever, would need to authorize compounding based on a drug shortage, because the
21
active ingredient contained in such a scarce drug already will be on the “clinical need” list and
greenlighted for use. Thus, the statutory allowance for using a bulk drug substance based on drug
shortage would serve little or no function under Plaintiffs’ interpretation.
Plaintiffs respond that, under their reading of “clinical need,” the drug shortage list still
would have utility. For instance, there could be future drug product innovations where the
component bulk drug substance is not yet placed on the clinical need list due to the required notice-
and-comment period, but there is an immediate need for the drug itself due to shortage. See Hr’g
Tr. at 30–33. Is this scenario possible? Maybe. But it is doubtful that Congress intended for the
drug-shortage listing to cover only outlier scenarios, such as when a drug manufacturer comes to
market with a new FDA-approved product but does not produce enough of it to meet demand
during the time period the product’s active ingredient is under review for placement on the “clinical
need” list. When Congress passed the DQSA, it understood the importance of the drug shortage
list. Congress could not have not intended for the “clinical need” list to fully eclipse the drug
shortage list, leaving the latter as a mere “backstop” to the former. See id. at 33.
d. The “essentially a copy” provision
To support their interpretation, Plaintiffs point to a provision that limits outsourcing
facilities to producing a “[compounded] drug [that] is not essentially a copy of one or more
approved drugs.” See 21 U.S.C. § 353b(a)(5). The statute defines “essentially a copy of an
approved drug” to mean:
(A) a drug that is identical or nearly identical to an approved drug, or a
marketed drug not subject to section 353(b) of this title and not
subject to approval in an application submitted under section 355 of
this title, unless, in the case of an approved drug, the drug appears
on the drug shortage list in effect under section 356e of this title at
the time of compounding, distribution, and dispensing; or
22
(B) a drug, a component of which is a bulk drug substance that is a
component of an approved drug or a marketed drug that is not
subject to section 353(b) of this title and not subject to approval in
an application submitted under section 355 of this title, unless there
is a change that produces for an individual patient a clinical
difference, as determined by the prescribing practitioner, between
the compounded drug and the comparable approved drug.
21 U.S.C. § 353b(d)(2). As a result of the “essentially a copy” provision, unless the drug product
appears on the shortage list, outsourcing facilities can compound using a bulk drug substance that
is a component of an approved drug only if the bulk-compounded drug supplies to “an individual
patient a clinical difference” from the approved drug. Id. Congress left it to medical practitioners
to make the “clinical difference” determination. Id.
Plaintiffs argue that FDA’s “clinical need” interpretation makes redundant the “essentially
a copy” provision. They contend that the “essentially a copy” definition ensures that outsourcing
facilities will not compound drug products that rival FDA-approved drugs, so the clinical need
provision cannot fulfill the same purpose. See Pls.’ Mem. at 22–26. By drawing a comparison to
approved drugs during the “clinical need” inquiry, Plaintiffs maintain, FDA “disturbs the balance
Congress struck when it made the policy decision of how best to account for the competing interest
in encouraging the use of the drug approval process.” See id. at 23. Defendants, on the other hand,
assert that the “essentially a copy” provision is consistent with its “clinical need” interpretation.
Their reading, they say, ensures that health practitioners will only use the compounded drug for
patients who need the compounded drug to fill therapeutic gaps left by an FDA-approved drug.
See Hr’g Tr. at 64.
Plaintiffs are correct that the “essentially a copy” provision works to protect the new drug
approval process and, by extension, provides a market advantage to FDA-approved drugs over
those produced by outsourcing facilities. Compare R. at 108 (FDA justifying “clinical need”
23
interpretation to protect drug approval process) with FOOD AND DRUG ADMINISTRATION,
COMPOUNDED DRUG PRODUCTS THAT ARE ESSENTIALLY COPIES OF APPROVED DRUG
PRODUCTS UNDER 503B OF THE FEDERAL FOOD, DRUG, AND COSMETIC ACT: GUIDANCE FOR
INDUSTRY 4 (January 2018), https://www.fda.gov/media/98964/download (last visited July 31,
2019) [hereinafter Essential Copy Guidance] (FDA stating that the essentially a copy provision
“protects the integrity and effectiveness of the new drug . . . approval process[]”). But the fact that
FDA’s method of determining “clinical need” furthers that same purpose is not fatal to the
agency’s interpretation. “In some cases, redundancy may reflect the broad purpose of a
congressional statute.” Nat’l Ass’n of Clean Water Agencies v. EPA, 734 F.3d 1115, 1126 (D.C.
Cir. 2013). Here, multiple provisions within Section 503B reflect Congress’s intent that bulk
compounded drugs be used primarily to fill gaps left by FDA-approved drugs. Perhaps the clearest
such expression is the general rule, discussed above, that outsourcing facilities “not compound
using bulk drug substances . . . .” 21 U.S.C. § 353b(a)(2). That rule creates a clear market
advantage for approved drugs. Congress also required outsourcing facilities to sell their drug
products directly to hospitals and physicians for use as “office stock,” and expressly prohibited
them from using wholesalers to make sales. Id. § 353b(a)(8). Wholesalers are critical to the drug
supply chain and cutting off that path of distribution limits an outsourcing facility’s ability to sell
its product. In short, like the “essentially a copy” provision, these other conditions constrain the
commercial activities of outsourcing facilities relative to FDA-approved drug manufacturers. That
FDA’s understanding of “clinical need” has a similar market effect therefore does not conflict with
Congress’s overall design.
Moreover, FDA’s method of determining “clinical need” easily co-exists with the
“essentially a copy” provision. As a general rule, the “essentially a copy” provision prohibits
24
outsourcing facilities from producing a compounded drug “a component of which is a bulk drug
substance that is a component of an approved drug.” Id. § 353b(d)(2)(B). In other words, a
compounded drug cannot be made with a bulk drug substance if such substance already is
contained in an approved product. The exception to that general rule is if the compounded drug
entails a change from the approved product “that produces for an individual patient a clinical
difference, as determined by the prescribing practitioner . . . .” Id. FDA’s “clinical need”
determination mirrors, but does not duplicate, the “essentially a copy” provision. It asks whether
the bulk drug product fills a gap of medical unsuitability left by an approved drug and, if so,
whether that gap cannot be filled by compounding with the approved drug. Thus, both the
“essentially a copy” provision and FDA’s “clinical need” inquiry are directed at identifying
whether the compounded product is one that fills a therapeutic purpose unmet by the approved
drug. The two provisions therefore complement each other in furthering the statutory purpose.
Plaintiffs complain that reading Section 503B in this way is impermissible because,
through its pre-market “clinical need” inquiry, FDA is in effect making the “clinical difference”
determination that the statute leaves to treatment providers. The agency is thus impermissibly
regulating the practice of medicine, say Plaintiffs. See Pls.’ Mem. at 27. Not so. Take the case
of vasopressin. FDA’s decision not to recognize a “clinical need” for vasopressin does not
interfere with a physician’s decision whether to treat a patient with the approved product,
Vasostrict, or a compounded version of Vasostrict that meets a patient’s individual need. FDA’s
vasopressin decision only regulates the type of drug that reaches the marketplace. Such a decision
rests well within FDA’s regulatory authority under the FDCA, and it does not intrude on the
25
practice of medicine. 7 See United States v. Evers, 643 F.2d 1043, 1048 (5th Cir. 1981) (“[W]hile
the [FDCA] was not intended to regulate the practice of medicine, it was obviously intended to
control the availability of drugs for prescribing by physicians.”).
* * *
“A provision that may seem ambiguous in isolation is often clarified by the remainder of
the statutory scheme . . . because only one of the permissible meanings produces a substantive
effect that is compatible with the rest of the law.” United Sav. Ass’n of Tex v. Timbers of Inwood
Forest Assocs., Ltd., 484 U.S. 365, 371 (1988). Here, the broader structure of Section 503B, as
well as its relationship to Section 503A, compel the court to conclude that FDA’s interpretation of
“clinical need” is the only interpretation “compatible with the rest of the law.”
e. Legislative History
In addition to a statute’s text and structure, legislative history is a relevant consideration at
Chevron step one. See Sierra Club v. EPA, 551 F.3d 1019, 1027 (D.C. Cir. 2008) (“Although
Chevron step one analysis begins with the statute’s text, the court must . . . exhaust the traditional
tools of statutory construction, including examining the statute’s legislative history . . . .”) (internal
quotation marks omitted). The legislative history for the DQSA is sparse. Neither chamber
produced a committee report. Yet, no one disputes that the event that motivated Congress to enact
the DQSA was the deadly meningitis outbreak caused by the contaminated injections produced by
the New England Compounding Center (“NECC”). See The Fungal Meningitis Outbreak: Could
it have been Prevented? Before the Subcomm. on Oversight and Investigations of the H. Comm.
on Energy and Commerce, 112th Cong. 181 (2012) [hereinafter House Hearing] (Chairman Stearns
7
Relatedly, Plaintiffs argue that the agency’s determination of “medical unsuitab[ility]” is an unlawful regulation of
medicine. See Pls.’ Reply at 33. But the inquiry focuses on whether the approved drug is unsuitable for a patient
population; it does not make a clinical determination about a specific patient.
26
stating, “we convene this hearing . . . to examine the recent outbreak . . . linked to contaminated
products made by [NECC] . . . the first question we all ask is, Could this have been prevented? . .
. the answer appears to be yes”); see also Hr’g Tr. at 42; see also Compl. ¶ 17. NECC, and other
compounders like it, had grown beyond the traditional notion of a local pharmacist compounding
individual prescriptions. See House Hearing (Margaret Hamburg, former FDA Commissioner,
stating “the industry has evolved well beyond the neighborhood pharmacist”). By amending the
FDCA, Congress wished to strengthen federal regulatory authority over bulk compounders that
produced drugs on a large scale that were not subject to the FDA’s new-drug efficacy and safety
standards.
FDA’s method of determining “clinical need” better comports with the DQSA’s primary
objective of protecting public safety. Plaintiffs’ interpretation, if accepted, would substantially
constrain FDA’s judgment and expertise with respect to the “clinical need” list, and concomitantly
would enable outsourcing facilities to compound using a wide array of bulk drug substances to
create drug products that are not subject to any premarket efficacy or safety review. FDA’s view,
by contrast, grants the agency a more prominent role before a bulk compounded drug hits the
market, and favors compounding with an FDA-approved product that has undergone rigorous
premarket efficacy and safety review. FDA’s approach better achieves the public health and safety
objective that Congress had in mind when it passed the DQSA. Cf. Thompson v. W. States Med.
Ctr., 535 U.S. 357, 369 (2002) (“Preserving the effectiveness and integrity of the FDCA’s new
drug approval process is clearly an important governmental interest, and the Government has every
reason to want as many drugs as possible to be subject to that approval process. The Government
also has an important interest, however, in permitting the continuation of the practice of
27
compounding so that patients with particular needs may obtain medications suited to those
needs.”).
For their part, Plaintiffs rely on a comment by Senator Lamar Alexander that the DQSA
was not intended to “limit access to quality compounded drugs for providers and patients or alter
the practice of medicine.” Pls.’ Mem. at 19 (citing 159 Cong. Rec. S8072 (daily ed. Nov. 18,
2013)). That general statement does not, however, carry the weight Plaintiffs attribute to it. In
fact, right before making this statement, Senator Alexander said that “I want to make clear that all
involved on this legislation have no intent of limiting patient or provider access to quality
compounded drugs that fill a clinical need.” 159 Cong. Rec. S8072 (emphasis added). The
Senator’s statement leads the court right back to where it began—when is there a “clinical need”
for a drug compounded from a bulk drug substance?
* * *
For the foregoing reasons, the court finds that Defendants prevail at Chevron step one.
FDA’s method of determining “clinical need” for a bulk drug substance by referring to the
approved product is supported by the text, structure, and purposes of Section 503B.
B. Chevron Step Two
In the interest of completeness, the court proceeds to Chevron step two. Defendants also
would prevail at Chevron step two, if the statute were to remain ambiguous after employing the
traditional tools of statutory construction. For the reasons already discussed, FDA’s interpretation
of “clinical need” is “based on a permissible construction of the statute.” Chevron, 467 U.S. at
843.
Further, as required to receive Chevron deference, the agency “offered a reasoned
explanation for why it chose [its] interpretation.” Village of Barrington, 636 F.3d at 660. In its
28
final notice excluding vasopressin, FDA explained that it formulated its methodology to avoid
“unnecessarily expos[ing] patients to the risks associated with drug products that do not meet the
standards applicable to FDA-approved drug products for safety, effectiveness, quality, and
labeling and would undermine the drug approval process.” R. at 108. FDA also explained that it
sought to determine whether compounding must occur with a bulk drug substance, as opposed to
the FDA-approved product, because the statute requires the FDA to “determine that there is a
clinical need for outsourcing facilities to compound a drug product using the bulk drug substance.”
Id. (emphasis in original omitted). These explanations suffice to show that the agency had
reasonable grounds for its construction of “clinical need.”
Plaintiffs offer no new arguments at Chevron step two that they did not raise at Chevron
step one. See Pls.’ Mem. at 28–29. Those arguments did not persuade the court at the first step,
and they fare no better at the second step.
C. FDA’s Decision to Exclude Vasopressin from the “Clinical Need” List was Not
Arbitrary and Capricious
Alternatively, Plaintiffs challenge FDA’s decision to exclude vasopressin from the
“clinical need” list as arbitrary and capricious, even under the agency’s approach to assessing
“clinical need.” See Pls.’ Mem. at 30–37. Under the “narrow” arbitrary and capricious standard
of review, the court may not “substitute its judgment for that of the agency,” but must instead
determine whether the agency “examine[d] the relevant data and articulate[d] a satisfactory
explanation for its action including a rational connection between the facts found and the choice
made.” Motor Vehicle Mfrs. Ass'n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43
(1983) (internal quotation marks and citation omitted). An agency action is “arbitrary and
capricious” and will be set aside if the agency “has relied on factors which Congress has not
intended it to consider, entirely failed to consider an important aspect of the problem, offered an
29
explanation for its decision that runs counter to the evidence before the agency, or is so implausible
that it could not be ascribed to a difference in view or the product of agency expertise.” Id.
Plaintiffs contend that their vasopressin drug product meets a “clinical need” even under
FDA’s interpretation for two reasons. First, Plaintiffs argue that Vasostrict is medically unsuitable
for certain patients because it contains the allergen chlorobutanol, thereby giving rise to a “clinical
need.” See Pls.’ Mem. at 35–37. But in denying listing, FDA identified a chlorobutanol-free
version of Vasostrict. R. 1604. It noted that vasopressin’s sponsor had “fail[ed] to explain why
that formulation would be medically unsuitable for patients who have an allergy to chlorobutanol.”
Id. Plaintiffs do not appear to dispute this agency finding, as they do not contest it in their reply
brief. See generally Pls.’ Reply. FDA’s explanation satisfies the arbitrary and capricious standard.
Second, Plaintiffs assert that their product is available in a ready-to-use, intravenous
version, which offers safety benefits for patients relative to Vasostrict. To support this contention,
Plaintiffs point out that vasopressin has been designated a “high alert” drug that presents a serious
risk of patient harm from preparation or administration error. See Pls.’ Mem. at 30–35; see also
R. at 464. Their ready-to-use version, Plaintiffs claim, is safer than Vasostrict, which requires
dilution before intravenous delivery. As Plaintiffs put it, their drug product “giv[es] emergency
rooms an option other than [Vasostrict’s] 16-step process . . . resembling compounding on the fly.”
See Pls.’ Mem. at 31.
FDA considered Plaintiffs “ready-to-use” claim but rejected it. 8 First, the agency found
that commenters did not “identify a basis to conclude that there is an attribute of [Vasostrict] that
8
Plaintiffs incorrectly assert that FDA did not consider the “high alert” designation of vasopressin. Pls.’ Mem. at 31,
33. FDA may not have used the words “high alert,” but it certainly recognized and addressed Plaintiffs’ argument
that its ready-to-use product fills a “clinical need.” See R. at 1598–99 (“[T]he comments contended that [ready-to-
use] products compounded from bulk drug substances can reduce the risk of medication errors or contamination,
decrease access times, and increase efficiencies in healthcare facility operations. They state these advantages are
particularly important for drugs in emergency settings.”).
30
makes [it] unsuitable such that patients need a compounded drug product.” R. at 1599. As the
agency put it, manufacturing a drug that is ready-to-use to “improve[] efficiency” or to “address
the possibility that the approved drug might be mishandled by a medical professional” is not
addressing clinical need. Id. at 1605 (emphasis added). Instead, the ready-to-use capability just
provides an “advantage[], relative to the approved drug product[].” Id. at 1599. Even though
Plaintiffs may have demonstrated that ready-to-use is advantageous, that does not mean that
Vasostrict—because it is not ready-to-use—is medically unsuitable. The FDA’s view on this issue
is not “so implausible that it could not be ascribed to a difference in view or the product of agency
expertise.” State Farm, 463 U.S. at 43.
Second, and as importantly, the agency found that the commenters did not show that the
proposed drug product “must be made from a bulk drug substance rather than by diluting the
approved drug.” R. at 1599. Another drug supplier, PharMEDium Services, operates an
outsourcing facility that compounds Vasostrict into a ready-to-use formulation. See id. at 1599;
see also id. at 198. Plaintiffs have not asserted that their ready-to-use bulk compounded product
meets a clinical need that PharMEDium’s sterile-to-sterile compounded product does not.
See Pls.’ Mem; see also Pls.’ Reply (not mentioning PharMEDium’s product). FDA’s rejection
of a “clinical need” for vasopressin therefore was not arbitrary and capricious.
V. CONCLUSION
For the reasons set forth above, the court denies Plaintiffs’ Motion for Summary Judgment,
ECF No. 23, and grants Defendants’ Cross-Motions for Summary Judgment, ECF Nos. 25 and 27.
A final order accompanies this Memorandum Opinion.
Dated: August 1, 2019 Amit P. Mehta
United States District Court Judge
31