Alliance Hippocratic Medicine v. FDA ( 2023 )

United States Court of Appeals
for the Fifth Circuit
United States Court of Appeals
Fifth Circuit
____________ FILED
April 12, 2023
No. 23-10362 Lyle W. Cayce
____________ Clerk
Alliance for Hippocratic Medicine; American
Association of Pro-Life Obstetricians &
Gynecologists; American College of Pediatricians;
Christian Medical & Dental Associations; Shaun
Jester, D.O.; Regina Frost-Clark, M.D.; Tyler Johnson,
D.O.; George Delgado, M.D.,
Plaintiffs—Appellees,
versus
Food & Drug Administration; Robert M. Califf,
Commissioner of Food and Drugs; Janet Woodcock, M.D., in her
official capacity as Principal Deputy Commissioner, U.S. Food and Drug
Administration; Patrizia Cavazzoni, M.D., in her official capacity as
Director, Center for Drug Evaluation and Research, U.S. Food and Drug
Administration; United States Department of Health and
Human Services; Xavier Becerra, Secretary, U.S. Department of
Health and Human Services,
Defendants—Appellants,
versus
Danco Laboratories, L.L.C.,
Intervenor—Appellant.
No. 23-10362
______________________________
Appeal from the United States District Court
for the Northern District of Texas
USDC No. 2:22-CV-223
______________________________
UNPUBLISHED ORDER
Before Haynes, ∗ Engelhardt, and Oldham, Circuit Judges.
Per Curiam:
For the reasons given below, IT IS ORDERED that defendants’
motions for a stay pending appeal are GRANTED IN PART. At this
preliminary stage, and based on our necessarily abbreviated review, it appears
that the statute of limitations bars plaintiffs’ challenges to the Food and Drug
Administration’s approval of mifepristone in 2000. In the district court,
however, plaintiffs brought a series of alternative arguments regarding
FDA’s actions in 2016 and subsequent years. And the district court
emphasized that its order separately applied to prohibit FDA’s actions in and
after 2016 in accordance with plaintiffs’ alternative arguments. As to those
alternative arguments, plaintiffs’ claims are timely. Defendants have not
shown that plaintiffs are unlikely to succeed on the merits of their timely
challenges. For that reason, and as more fully explained below, defendants’
motions for a stay pending appeal are DENIED IN PART. Defendants’
alternative motions for an administrative stay are DENIED AS MOOT.
Plaintiffs’ motion to dismiss the appeal is DENIED. The appeal is
EXPEDITED to the next available Oral Argument Calendar.
_____________________
∗
Judge Haynes concurs only in part: she concurs in the grant of the expedited
appeal and the denial of the motion to dismiss. With respect to the request for a stay of the
district court’s order, as a member of the motions panel, she would grant an administrative
stay for a brief period of time and defer the question of the stay pending appeal to the oral
argument merits panel which receives this case.
2
No. 23-10362
I.
A.
Congress delegated to the Food and Drug Administration (“FDA”)
the responsibility to ensure that “new drugs” are “safe and effective.” 21
U.S.C. §§ 321(p), 355; see also id. § 393(b)(2)(B). When making its approval
determination, FDA evaluates whether a new drug application (“NDA”)
includes scientific evidence demonstrating that the drug is safe and effective
for its intended uses. Id. § 355(d); see also 21 C.F.R. §§ 314.50, 314.105(c).
Similarly, when a sponsor submits a supplemental new drug application
(“SNDA”) proposing changes to the conditions of approval for a drug (such
as changes to a drug’s labeling or FDA-imposed restrictions), FDA reviews
the scientific evidence to support the changes. See 21 C.F.R. § 314.70. To
approve a generic version of a previously approved drug, FDA reviews
whether an abbreviated new drug application (“ANDA”) contains
information showing that the proposed generic drug is materially the “same”
as the approved drug. 21 U.S.C. § 355(j)(2).
In 1992, FDA promulgated the so-called “Subpart H” regulations.
Subpart H accelerates approval of drugs “that have been studied for their
safety and effectiveness in treating serious or life-threatening illnesses and
that provide meaningful therapeutic benefit to patients over existing
treatments (e.g., ability to treat patients unresponsive to, or intolerant of,
available therapy, or improved patient response over available therapy).” 21
C.F.R. § 314.500. Originally, Subpart H was intended to promote rapid
approval for life-saving HIV-AIDS drugs. But given that Subpart H approvals
were accelerated, FDA recognized that it would need post-approval safety
measures. These post-approval safety measures would “assure safe use” of
the quickly approved Subpart H drugs. Id. § 314.520. In 2007, Congress
ratified these post-approval safety measures as “risk evaluation and
3
No. 23-10362
mitigation strategies” (“REMS”), which “ensure that the benefits of the
drug outweigh the risks.” 21 U.S.C. § 355-1(a)(1)–(2).
B.
In 2000, FDA approved mifepristone to be marketed with the brand
name Mifeprex under Subpart H (the “2000 Approval”). See 21 C.F.R.
§ 314.500; FDA Add. 181. 1 In the 2000 Approval, FDA concluded that
pregnancy is a “life-threatening illness,” triggering an accelerated approval
of mifepristone under Subpart H. FDA Add. 186. FDA also concluded that a
variety of post-approval restrictions on Mifeprex were required “to assure
safe use.” 21 C.F.R. § 314.520. As noted in the previous section, today we
call such post-approval restrictions “REMS.” The 2000 Approval imposed
several REMS, including: (1) limiting the drug to pregnant women and girls
for use through 49 days gestation; (2) requiring three in-person office visits,
the first to administer mifepristone, the second to administer misoprostol,
and the third to assess any complications and ensure there were no fetal
remains in the womb; (3) requiring the supervision of a qualified physician;
and (4) requiring the reporting of all adverse events from the drugs. FDA
Add. 181–91. FDA granted Danco Laboratories, LLC, an exclusive license to
manufacture, market, and distribute Mifeprex in the United States. FDA
Add. 109.
In 2002, two of the plaintiff associations in this case filed a citizen
petition challenging the 2000 Approval (the “2002 Citizen Petition”). See
21 C.F.R. § 10.25(a); PI App. 280–375. Roughly fourteen years later, FDA
denied the 2002 Citizen Petition (the “2016 Petition Denial”). FDA Add.
_____________________
1
Citations to the addendum to FDA’s emergency motion for a stay pending appeal
are denoted “FDA Add.” Citations to the appendix to plaintiffs’ motion for a preliminary
injunction are denoted “PI App.”
4
No. 23-10362
804–36. And on the very same day in March 2016, FDA approved several
major changes to mifepristone’s approved conditions of use, including its
REMS. Specifically, FDA removed four of the original safety restrictions by
(1) increasing the maximum gestational age at which a woman can use the
drug from 49 to 70 days; (2) reducing the number of required in-person office
visits from three to one; (3) allowing non-doctors to prescribe and administer
the chemical abortions drugs; and (4) eliminating the requirement for
prescribers to report non-fatal adverse events from chemical abortion (the
“2016 Major REMS Changes”). FDA Add. 777–802.
In March 2019, one of the plaintiff associations filed a second citizen
petition challenging the 2016 Major REMS Changes (the “2019 Citizen
Petition”). FDA Add. 192–217. That petition asked FDA to “restore” the
2000 Approval’s REMS and “retain” a requirement that mifepristone be
dispensed to patients in person. FDA Add. 192.
In April 2019, FDA approved GenBioPro, Inc’s ANDA for a generic
version of mifepristone (the “2019 Generic Approval”). PI App. 694–708.
GenBioPro’s generic version of mifepristone has the same labeling and
REMS requirements as Danco’s Mifeprex.
In April 2021, FDA announced that it would “exercise enforcement
discretion” to allow “dispensing mifepristone through the mail . . . or
through a mail-order pharmacy” during the COVID-19 pandemic (the “2021
Mail-Order Decision”). PI App. 713–15. FDA took this action in response to
a letter from the American College of Obstetricians and Gynecologists and
the Society for Maternal-Fetal Medicine. PI App. 710–11.
Later that year, in December 2021, FDA denied almost all of the 2019
Citizen Petition (the “2021 Petition Denial”). FDA Add. 837–76. In
particular, FDA expressly rejected the 2019 Citizen Petition’s request to
keep the in-person dispensing requirements and announced that the agency
5
No. 23-10362
had concluded that “the in-person dispensing requirement is no longer
necessary.” FDA Add. 842.
Finally, in January 2023, FDA approved a modified REMS for
mifepristone lifting the in-person dispensing requirement. See REMS Single
Shared System for Mifepristone 200 mg (Jan. 2023), https://perma.cc/MJT5-
35LF (the “2023 Mail-Order Decision”). 2
C.
In November 2022, plaintiffs (physicians and physician organizations)
filed this suit against FDA, HHS, and a several agency heads in the official
capacities. Plaintiffs first challenged FDA’s 2000 Approval of the drug. But
they also requested multiple grounds of alternative relief for FDA’s
subsequent actions. Immediately after filing, plaintiffs moved for a
preliminary injunction ordering FDA to withdraw or suspend (1) FDA’s
2000 Approval and 2019 Generic Approval, (2) FDA’s 2016 Major REMS
Changes, and (3) FDA’s 2021 Mail-Order Decision and its 2021 Petition
Denial of the 2019 Citizen Petition. If that’s confusing, we hope this chart
helps:
_____________________
2
Danco suggests the 2023 Mail-Order Decision moots part of plaintiffs’ claims.
See Danco Stay App. 22. We disagree. The Supreme Court has explicitly instructed this
court to review a new agency action finalized after litigation commenced and while the
appeal was pending because this decision was a “final agency action” for purposes of 5
U.S.C. § 704. Biden v. Texas, 142 S. Ct. 2528, 2544-45 (2022) (quotation omitted).
6
No. 23-10362
Event Citation Description
Approved mifepristone with
these REMS: (1) pregnancies
under 50 days gestation;
2000 Approval FDA Add. 181–91 (2) three in-person office visits;
(3) supervision of a qualified
physician; and (4) reporting of
all adverse events
2002 Citizen Plaintiffs’ challenge to 2000
PI App. 280–375
Petition Approval
2016 Petition FDA denial of 2002 Citizen
FDA Add. 804–36
Denial Petition
FDA changed four of the 2000
Approval’s REMS:
(1) increased maximum
gestational age to 70 days;
2016 Major (2) reduced required in-person
FDA Add. 768, 777–802
REMS Changes office visits to one; (3) allowed
non-doctors to prescribe and
administer mifepristone; and
(4) eliminated reporting of non-
fatal adverse events
2019 Citizen Plaintiffs’ challenge to 2016
FDA Add. 192–217
Petition Major REMS Changes
FDA ANDA Approval Letter
2019 Generic
PI App. 694–708 for mifepristone generic to
Approval
GenBioPro, Inc.
FDA announces “enforcement
discretion” to allow
2021 Mail-Order
PI App. 713–15 mifepristone to be dispensed
Decision
through the mail during
COVID-19
FDA denial of almost all of the
2019 Citizen Petition, including
2021 Petition
FDA Add. 837–76 plaintiffs’ request to keep the
Denial
in-person dispensing
requirements
2023 Mail-Order https://perma.cc/MJT5- FDA permanently removed the
Decision 35LF in-person dispensing REMS
7
No. 23-10362
On April 7, 2023, the district court entered an order staying the
effective date of the 2000 Approval and each of the subsequent challenged
actions. 3 The district court stayed its own order for seven days to allow the
defendants time to appeal.
II.
FDA and Danco (“stay applicants” or “applicants”) ask us to stay
the district court’s order pending appeal. Our power to grant a stay is
inherent. See In re McKenzie, 180 U.S. 536, 551 (1901); Scripps-Howard Radio
v. FCC, 316 U.S. 4, 10–14 (1942). It’s also statutory. See Fed. R. App. P.
8; 28 U.S.C. § 1651; 5th Cir. R. 27.3; see also 16A Charles Alan
Wright & Arthur R. Miller, Federal Practice &
Procedure § 3954 (5th ed. Apr. 2022 update).
But we grant stays “only in extraordinary circumstances.” Williams v.
Zbaraz, 442 U.S. 1309, 1311 (1979) (Stevens, J., in chambers); see also Graves
v. Barnes, 405 U.S. 1201, 1203 (1972) (Powell, J., in chambers) (same);
Ruckelshaus v. Monsanto Co., 463 U.S. 1315, 1316 (1983) (Blackmun, J., in
chambers) (same). This rule reflects the fact that “a stay is not a matter of
right, even if irreparable injury might otherwise result.” Virginian Ry. Co. v.
United States, 272 U.S. 658, 672 (1926). Instead, a stay requires “an exercise
of judicial discretion.” Ibid. A “decree creates a strong presumption of its
own correctness,” which often counsels against a stay. Id. at 673.
_____________________
3
As both parties recognize, this order would have the practical effect of an
injunction because it would remove mifepristone from the market. Plaintiffs filed a motion
to dismiss applicants’ appeal on the theory that § 705 stays are not sufficient to trigger our
interlocutory appellate jurisdiction under 28 U.S.C. § 1292(a). We disagree. See Abbott v.
Perez, 138 S. Ct. 2305, 2319–20 (2018) (explaining that the “practical effect” test of 28
U.S.C. §§ 1292(a)(1) and 1293 “prevents [the] manipulation” that could occur “if the
availability of interlocutory review depended on the district court’s use of the term
‘injunction’”).
8
No. 23-10362
The Supreme Court has prescribed “four traditional stay factors”
that govern this equitable discretion in most civil cases. Ala. Ass’n of Realtors
v. HHS, 141 S. Ct. 2485, 2487 (2021) (quotation omitted); see also Hilton v.
Braunskill, 481 U.S. 770, 776–77 (1987); Rose v. Raffensperger, 143 S. Ct. 58,
59 (2022) (reversing stay of an injunction after the court of appeals failed to
analyze the traditional stay factors). Those factors are:
(1) whether the stay applicant has made a strong showing that
he is likely to succeed on the merits; (2) whether the applicant
will be irreparably injured absent a stay; (3) whether issuance
of the stay will substantially injure the other parties interested
in the proceeding; and (4) where the public interest lies.
Nken v. Holder, 556 U.S. 418, 426 (2009) (quoting Hilton, 481 U.S. at 776);
see also Whole Woman’s Health v. Jackson, 141 S. Ct. 2494, 2495 (2021).
Although no factor is dispositive, the likelihood of success and irreparable
injury factors are “the most critical.” Nken, 556 U.S. at 434. Success on
either factor requires that the stay seeker make a strong not merely
“possib[le]” showing. Ibid.In these respects, stays might appear identical to preliminary
injunctions. Similar factors govern both and both require an “extraordinary”
deployment of judicial discretion. Winter v. Nat. Res. Def. Council, 555 U.S.
7, 24 (2008). But the two are not “one and the same.” Nken, 556 U.S. at 434.
A stay “operates upon the judicial proceeding itself,” not on the conduct of
a particular actor. Id. at 428. And, once one party has won an injunction,
proof burdens reverse. It is the enjoined party who seeks a stay, or FDA and
Danco here, who must carry the burden of proving that the Nken factors
command us to issue one. See Landis v. N. Am. Co., 299 U.S. 248, 255 (1936).
If the stay applicants show that circumstances require a stay of some
but not all of the district court’s order, we may, in our discretion, “tailor a
stay so that it operates with respect to only some portion of the proceeding.”
9
No. 23-10362
Trump v. Int’l Refugee Assistance Project, 137 S. Ct. 2080, 2087 (2017) (per
curiam) (quoting Nken, 556 U.S. at 428).
We find that FDA and Danco succeed only in part.
III.
Regarding likelihood to succeed on the merits, the stay applicants
raise four arguments. They contend (A) plaintiffs are unlikely to defend the
district court’s stay because they lack standing. They next contend
(B) plaintiffs’ claims are untimely. Then they claim (C) plaintiffs’ claims are
unexhausted. Finally, applicants contend (D) FDA’s actions are not
arbitrary, capricious, or otherwise contrary to law. We consider each in turn.
A.
We begin with Article III standing. To bring their claims in federal
court, plaintiffs must satisfy the familiar tripartite test: they must show they
suffered an injury in fact, that’s fairly traceable to the defendants, and that’s
likely redressable by a favorable decision. See Lujan v. Nat’l Wildlife Fed’n,
497 U.S. 871 (1990). Importantly, only one plaintiff needs to have standing to
present a valid case or controversy. See Rumsfeld v. Forum for Acad. &
Institutional Rts., Inc., 547 U.S. 47, 52 n.2 (2006).
Plaintiffs and the district court offered numerous theories of standing.
At this preliminary, emergency stage, we are unpersuaded by applicants’
contentions that all of these theories fail to create a justiciable case or
controversy. We need only consider two: (1) injuries to doctors and
(2) injuries to the plaintiff medical associations. 4
_____________________
4
We are cognizant of the fact that the Supreme Court has disavowed the theories
of third-party standing that previously allowed doctors to raise patients’ claims in abortion
10
No. 23-10362
1.
First, it appears that the individual plaintiffs and doctors in plaintiff
associations have standing to challenge FDA’s actions.
To allege an injury in fact, these doctors must show they have suffered
an “invasion of a legally protected interest” that is both “concrete and
particularized” and “actual or imminent, not conjectural or hypothetical.”
Spokeo, Inc. v. Robins, 578 U.S. 330, 339 (2016) (quotation omitted). Plaintiffs
must identify specific injuries that go beyond “general averments” or
“conclusory allegations.” Friends of the Earth, Inc. v. Laidlaw Env’t Servs.
(TOC), Inc., 528 U.S. 167, 184 (2000) (quoting Lujan, 497 U.S. at 888).
Where a plaintiff seeks prospective relief and hence points to future injuries,
the Supreme Court has emphasized that “threatened injury must be certainly
impending to constitute injury in fact, and that allegations of possible future
injury are not sufficient.” Clapper v. Amnesty Int’l USA, 568 U.S. 398, 409
(2013) (quotation omitted).
Here, FDA-approved the “Patient Agreement Form,” which is part
of the REMS for mifepristone, provides:
_____________________
cases. See Dobbs v. Jackson Women’s Health Org., 142 S. Ct. 2228, 2275 & n.61 (2022). So
we express no opinion on plaintiffs’ third-party standing theories.
11
No. 23-10362
2023 Mail-Order Decision at 10. FDA thus cannot deny that serious
complications from mifepristone are certainly impending. Those
complications are right there on the “Patient Agreement Form” that FDA
itself approved and that Danco requires every mifepristone user to sign.
According to the applicants, more than 5,000,000 women have taken this
drug since the 2000 Approval. FDA Stay App. 1. That means that, again
according to the applicants’ own information, between 100,000 (2%) and
12
No. 23-10362
350,000 (7%) of mifepristone users had unsuccessful chemical abortions and
had to “talk with [their] provider[s] about a surgical procedure to end [their]
pregnanc[ies].” 2023 Mail-Order Decision at 10. And where did those
hundreds of thousands of women go for their “surgical procedures”? Again,
we need not speculate because the 2016 Major REMS Changes, the 2021
Petition Denial, and the 2023 Mail-Order Decision all allow non-doctors to
prescribe mifepristone. The women who use this drug cannot possibly go
back to their non-doctor-prescribers for surgical abortions, so again, as the
“Patient Agreement Form” itself says, they must instead seek “emergency
care” from a qualified physician.
The plaintiff emergency room doctors have a concrete, particularized
injury since they have provided—and with certainty will continue to
provide—the “emergency care” that applicants specified in the “Patient
Agreement Form.” PI App. 167, 169, 194, 206. Mifepristone users who
present themselves to the plaintiffs have required blood transfusions,
overnight hospitalization, intensive care, and even surgical abortions. PI App.
205–06. As one doctor testified:
For example, in one month while covering the emergency
room, my group practice admitted three women to the hospital.
Of the three women admitted in one month due to chemical
abortion complications, one required admission to the
intensive care unit for sepsis and intravenous antibiotics, one
required a blood transfusion for hemorrhage, and one required
surgical completion for the retained products of conception
(i.e., the doctors had to surgically finish the abortion with a
suction aspiration procedure).
PI App. 206.
Another doctor testified:
[O]ne of my patients had obtained mifepristone and
misoprostol from a website, without an in-person
13
No. 23-10362
visit. . . . After taking the chemical abortion drugs, she began
having very heavy bleeding followed by significant abdominal
pain and a fever. When I saw her in the emergency room, she
had evidence of retained pregnancy tissue along with
endometritis, an infection of the uterine lining. She also had
acute kidney injury, with elevate creatinine. She required a
dilation and curettage (D&C) surgery to finish evacuating her
uterus of the remaining pregnancy tissue and hospitalization
for intravenous (IV) antibiotics, IV hydration, and a blood
transfusion. I spent several hours with her the day of her
surgery/hospital admission, keeping me from my primary
patient responsibilities in the labor and delivery unit and
requiring me to call in an additional physician to help cover
those responsibilities.
PI App. 194–95. As a result of FDA’s failure to regulate this potent drug,
these doctors have had to devote significant time and resources to caring for
women experiencing mifepristone’s harmful effects. This harm is sufficiently
concrete.
A second independent injury from the adverse effects of mifepristone
is the “enormous stress and pressure” physicians face in treating these
women. PI App. 215. One doctor said the strain “is some of the most
emotionally taxing work I have done in my career.” PI App. 880. Thus, this
is an independent injury because FDA’s actions “significantly affect[]” the
doctors’ “quality of life.” Sierra Club v. Morton, 405 U.S. 727, 734–35 (1972).
The doctors offered specific facts to explain this stress. Women who
take these drugs are susceptible to “torrential bleeding.” PI App. 170, 215. In
fact, “the risk of severe bleeding with chemical abortion is five times higher
than from surgical abortion.” PI App. 879. And these situations can quickly
go from bad to worse. As one doctor testified:
One of my patients, who was about nine weeks pregnant, had
previously been treated by hospital staff for a pulmonary
14
No. 23-10362
embolism with anti-coagulants. She was advised that she could
not seek a chemical abortion because it was contraindicated
due to the medications; yet the woman left the hospital and
sought an abortion at Planned Parenthood of Indiana. The
woman was given mifepristone by the doctor at Planned
Parenthood and took the drug. The woman called an Uber for
a ride home from Planned Parenthood. The woman began to
experience bleeding and other adverse effects from the
mifepristone. The woman’s Uber driver did not take her home
because she was so ill and instead brought her to the hospital’s
emergency department. At the hospital, the woman came
under my care. The woman had not yet taken the second
abortion drug, misoprostol. I treated the patient for the adverse
effects she suffered and told her not to take the misoprostol
given to her by Planned Parenthood because of the grave risk
that she could bleed out and die.
PI App. 216–17. Another doctor recounted an experience where he treated a
patient—who “suffered from two weeks of moderate to heavy bleeding, and
then developed a uterine infection”—by providing her “with intravenous
antibiotics” and performing a D&C procedure. PI App. 886. If the patient
waited a few more days to go to the hospital, the doctor predicted that “she
could have been septic and died.” PI App. 886. Another doctor testified that
he has encountered “at least a dozen cases of life-threatening complications”
from these drugs, and the frequency of these emergency situations has only
increased over time. PI App. 865.
The risks are only exacerbated for women who have ectopic
pregnancies. PI App. 207. This occurs in approximately two percent of
pregnancies. PI App. 539. As one doctor explained:
Chemical abortion drugs will not effectually end an ectopic
pregnancy because they exert their effects on the uterus, which
leaves women at risk of severe harm from hemorrhage due to
tubal rupture, in need of emergent surgery or potentially at risk
15
No. 23-10362
of death. Failure to perform an ultrasound prior to prescribing
abortion drugs will cause some women to remain undiagnosed
and at high risk for these adverse outcomes.
PI App. 208. The risks are greater under FDA’s relaxed standards. That is
because “without an in-person examination, it is impossible to rule out an
ectopic pregnancy,” placing a woman “at an increased risk of rupture or even
death.” PI App. 886.
The doctors also face an injury from the irreconcilable choice between
performing their jobs and abiding by their consciences. These doctors
structured their careers so they would not have to administer abortions. And
yet, because women often come to hospitals when they experience
complications from these drugs, these doctors sometimes have no other
choice but to perform surgical abortions. As one doctor testified:
The FDA’s expansion of chemical abortions also harms my
conscience rights because it could force me to have to surgically
finish an incomplete elective chemical abortion. I object to
abortion because it ends a human life. My moral and ethical
obligation to my patients is to promote human life and health.
But the FDA’s actions may force me to end the life of a human
being in the womb for no medical reason.
PI App. 209–10. And this harm is not speculative. Several doctors confirmed
that they have had to surgically complete an abortion or remove an unborn
child. PI App. 886, 205. As one doctor testified: “In my practice, I have cared
for at least a dozen women who have required surgery to remove retained
pregnancy tissue after a chemical abortion. Sometimes this includes the
embryo or fetus, and sometimes it is placental tissue that has not been
completely expelled.” PI App. 205. That same doctor described how she had
to “perform[] a suction aspiration procedure” on one patient who took the
pill but needed surgery to complete the abortion. PI App. 206. Others have
seen it firsthand. One doctor recounted a time where a woman came to the
16
No. 23-10362
emergency room “with heavy vaginal bleeding and unstable vital signs as a
result of taking chemical abortion drugs.” PI App. 195. When the woman
arrived in the emergency room, the baby in her womb was not dead; the
doctors were “able to detect a fetal heartbeat.” PI App. 195. But due to the
mother’s unstable condition, the doctors “had no choice but to perform an
emergency D&C.” PI App. 196. The doctor testified that her colleague “felt
as though she was forced to participate in something that she did not want to
be a part of—completing the abortion.” PI App. 196.
And not only have these doctors suffered injuries in the past, but it’s
also inevitable that at least one doctor in one of these associations will face a
harm in the future. Cf. City of Los Angeles v. Lyons, 461 U.S. 95 (1983). Here,
the plaintiff-doctors have “‘set forth’ by affidavit or other evidence ‘specific
facts’” that they are certain to see more patients. Clapper, 568 U.S. at 411(quoting Lujan v. Defs. of Wildlife, 504 U.S. 555, 561 (1992)). That’s because
FDA has removed almost all of mifepristone’s REMS and thus enabled
women to (1) get the drug without ever talking to a physician, (2) take the
drug without ever having a physical exam to ensure gestational age and/or an
ectopic pregnancy, and (3) attempt to complete the chemical abortion
regimen at home; FDA has also (4) directed the hundreds of thousands of
women who have complications to seek “emergency care” from the plaintiffs
and plaintiffs’ hospitals. Several doctors testified that they have seen an
increasing number of women coming to the emergency room with
complications from chemical abortions due to FDA’s virtual elimination of
controls on the dispensing and administration of the drugs. PI App. 194, 205,
215, 866. And given how many women these doctors have seen in emergency
departments in the past, these doctors quite reasonably know with statistical
certainty—again, a statistic estimated on Mifeprex’s own “Patient
Agreement Form”—that women will continue needing plaintiffs’
“emergency care.” See PI App. 205, 215, 868. The crisis is “concededly
17
No. 23-10362
ongoing.” Friends of the Earth, 528 U.S. at 184. Accordingly, plaintiffs face a
“substantial risk” of recurrence. Susan B. Anthony List v. Driehaus, 573 U.S.
149, 158 (2014) (quotation omitted).
And even if one of the named doctors never sees another patient, it’s
inevitable that one of the thousands of doctors in plaintiff associations will.
For example, one of the plaintiff associations, the American Association of
Pro-Life Obstetricians & Gynecologists, “is the largest organization of pro-
life obstetricians and gynecologists” and has “more than 7,000 medical
professionals nationwide.” PI App. 165. The Christian Medical and Dental
Association has “more than 600 physicians and approximately 35
OBGYNs.” PI App. 179. The American College of Pediatricians has a
membership of “more than 600 physicians and other healthcare
professionals.” PI App. 187. These associations presented affidavits from
individual members, elucidating the various harms discussed herein. See
Friends of the Earth, 528 U.S. at 183–84. Thus, they have associational
standing to sue on behalf of their members. See N.Y. State Club Ass’n, Inc. v.
City of New York, 487 U.S. 1, 9 (1988); Hunt v. Wash. State Apple Advertising
Comm’n, 432 U.S. 333, 343 (1977). That means that so long as one doctor
among the thousands of members in these associations faces an injury,
Article III is satisfied. See Rumsfeld, 547 U.S. at 52 n.2.
The doctors can also show that these injuries are traceable to FDA
regulations and redressable by this court. See Defs. of Wildlife, 504 U.S. at560–61. That’s because the 2016 Major REMS Changes, the 2021 Petition
Denial, and the 2023 Mail-Order Decision all empower non-doctors to
prescribe mifepristone and thus shift the costs of the drug onto the plaintiff
physicians who must manage the aftermath. See, e.g., PI App. 218 (“I spent a
significant amount of time that day working to save her life from unnecessary
complications due to the irresponsible administration and use of mifepristone
and misoprostol. As a result of the significant time that I devoted to that
18
No. 23-10362
patient, my time and attention was taken away from other patients, who also
need my care.”); PI App. 867 (“Because more women [who take
mifepristone] are unnecessarily presenting in the emergency department,
more of my time and attention is taken away from other patients who need
it.”). In this way, “[t]he FDA’s actions have created a culture of chaos for
emergency room physicians.” PI App. 867. And we’re capable of redressing
plaintiffs’ injuries by restoring the 2000 Approval’s REMS. Accordingly, at
this stage, applicants have not shown that all of the plaintiffs lack standing.
We hasten to emphasize the narrowness of this holding. We do not
hold that doctors necessarily have standing to raise their patients’ claims. See
supra n.4. We do not hold that doctors have constitutional standing whenever
they’re called upon to do their jobs. And we do not hold that doctors have
standing to challenge FDA’s actions whenever the doctor sees a patient
experiencing complications from an FDA-approved drug. Rather, we hold
that on the record before us applicants know that hundreds of thousands of
women will—with applicants’ own statistical certainty—need emergency
care on account of applicants’ actions. And because applicants chose to cut
out doctors from the prescription and administration of mifepristone,
plaintiff doctors and their associations will necessarily be injured by the
consequences. This is an exceedingly unusual regime. In fact, as far as the
record before us reveals, FDA has not structured the distribution of any
comparable drug in this way.
FDA’s principal contention to the contrary is that mifepristone is
comparable to “ibuprofen.” FDA Stay App. 1. The theory appears to be that
we cannot recognize plaintiffs’ standing here without opening a pandora’s
box in which doctors have standing to litigate everything at all times,
including the banalities of over-the-counter Advil.
19
No. 23-10362
We disagree because FDA’s own documents show that mifepristone
bears no resemblance to ibuprofen. In the 2000 Approval, FDA imposed a
“Black Box” warning on mifepristone. FDA requires “Black Box” warnings
when a drug “may lead to death or serious injury.” 21 C.F.R. § 201.57(c)(1).
In its 2000 Approval, FDA conditioned its approval of mifepristone on the
inclusion of this “Black Box” warning:
FDA Add. 182. The 2016 Major REMS Changes relaxed many of the
requirements for marketing and using mifepristone. But it retained this
“Black Box” warning:
20
No. 23-10362
https://perma.cc/R56J-BHW4.
Ibuprofen’s label, which FDA helpfully provided in its stay
addendum, obviously bears no resemblance to the “Black Box” warning on
mifepristone’s label. FDA Add. 465–68. To the contrary, FDA has a special
regulation regarding ibuprofen so all manufacturers of that over-the-counter
medicine include the same information on their labels. See 21 C.F.R.
§ 201.326. It says nothing about REMS, surgery, emergencies, Emergency
Rooms, or death.
In sum, applicants’ own documents—from the “Patient Agreement
Form” to the “Black Box” warning that have accompanied mifepristone
21
No. 23-10362
ever since the 2000 Approval up to and including today—prove that
emergency room care is statistically certain in hundreds of thousands of
cases. Plaintiff doctors have provided that emergency room care and are
statistically certain to provide it in the future.
2.
Second, the associations have standing. As previously discussed, they
have associational standing to sue on behalf of their members. See N.Y. State
Club Ass’n, Inc., 487 U.S. at 9; Hunt, 432 U.S. at 343. The associations
presented affidavits from individual member doctors who have suffered
harms. See Friends of the Earth, 528 U.S. at 183–84. Accordingly, they have
standing to sue on their members’ behalf.
Plaintiff associations have also suffered independent injuries because
FDA’s actions have frustrated their organizational efforts to educate their
members and the public on the effects of mifepristone. See Havens Realty
Corp. v. Coleman, 455 U.S. 363, 379 (1982) (holding that housing non-profit
had standing to challenge racial steering practices that impaired its ability “to
provide counseling and referral services for low-and-moderate-income
homeseekers”). As a result, plaintiff associations have expended “time,
energy, and resources to compensate for this lack of information by
conducting their own studies and analyses of available data” to “the
detriment of other advocacy and educational efforts.” PI App. 174. The
Supreme Court has previously stated that such a “concrete and
demonstrable injury to the organization’s activities—with the consequent
drain on the organization’s resources—constitutes far more than simply a
setback to the organization’s abstract social interests,” Havens, 455 U.S. at
379, even where the organizational interest is purely “noneconomic,” id. at
379 n.20. Rather, under these circumstances, “there can be no question that
the organization has suffered an injury in fact.” Id. at 379.
22
No. 23-10362
This injury is also traceable to FDA’s elimination of non-fatal adverse
events in the 2016 Major REMS Changes. And it’s redressable by an order
vacating those changes. Accordingly, these associations also have standing.
B.
Next we turn to timeliness.
Everyone acknowledges that 28 U.S.C. § 2401(a)’s six-year
limitations period applies to all of this case’s challenged actions. And
plaintiffs’ right of action against the lion’s share of the challenged actions are
squarely within the six-year window. That includes all of plaintiffs’
alternative arguments challenging the 2016 Major REMS Changes, the 2019
Generic Approval, the 2021 Mail-Order Decision, and the 2021 Petition
Denial of the 2019 Citizen Petition.
True, FDA’s March 2016 Major REMS Changes were promulgated
more than six years before plaintiffs filed suit in November 2022. But Section
2401(a) instructs that the six-year period begins when “the right of action
first accrues.” “And ‘[t]he right of action first accrues on the date of the final
agency action.’” Texas v. Biden, 20 F.4th 928, 951 n.3 (5th Cir. 2021), rev’d
on other grounds, 142 S. Ct. 2528 (2022) (quoting Wash. All. of Tech. Workers
v. DHS, 892 F.3d 332, 342 (D.C. Cir. 2018)). Though FDA promulgated the
Major REMS Changes in 2016, the Agency didn’t respond to plaintiffs’ 2019
Petition challenging those changes until December 16, 2021. So plaintiffs’
right of action against FDA’s final decision first accrued in December of
2021. See 21 C.F.R. § 10.45. That’s less than a year before plaintiffs sued,
which is well within the limitations period.
Next, applicants claim that plaintiffs’ primary challenges to the 2000
Approval and FDA’s 2016 Petition Denial to their 2002 Citizen Petition are
time-barred. Though admittedly a close question, we ultimately agree with
applicants at this preliminary juncture.
23
No. 23-10362
Plaintiffs’ right of action against the 2000 Approval and 2016 Petition
Denial first accrued on March 29, 2016—the date FDA issued its final
decision rejecting their 2002 Petition challenging the 2000 Approval. See 21
C.F.R. § 10.45. But plaintiffs didn’t file suit until November 18, 2022, more
than six months beyond the statute of limitations. The district court
nevertheless found timely the plaintiffs’ challenges to the 2000 Approval and
the 2016 Petition Denial. How? First, the district court held that FDA
“reopened” those decisions in 2016 and 2021, thus restarting the statute of
limitations. Second—and alternatively—the district court decided plaintiffs
were entitled to equitable tolling.
We consider each justification in turn.
First, reopening. “The reopen[ing] doctrine allows an otherwise
untimely challenge to proceed where an agency has—either explicitly or
implicitly—undertaken to reexamine its former choice.” Nat’l Biodiesel Bd.
v. EPA, 843 F.3d 1010, 1017 (D.C. Cir. 2016) (quotation omitted). Put simply,
the purpose of the reopening doctrine is “to pinpoint an agency’s final action
in cases where the agency has addressed the same issue multiple times.”
Texas v. Biden, 20 F.4th at 951. The limitations period runs from the agency’s
earlier decision unless the later decision “opened the issue up anew.” Ibid.
(quotation omitted). This makes good sense: Because a key step in the
timeliness inquiry is determining when an agency action became final, it’s
sometimes necessary to determine whether an agency’s subsequent action
“actually reconsidered” its former action, Growth Energy v. EPA, 5 F.4th 1,
21 (D.C. Cir. 2021) (per curiam) (quotation omitted), or merely
“reaffirm[ed] its prior position,” Sierra Club v. EPA, 551 F.3d 1019, 1024
(D.C. Cir. 2008) (quotation omitted); see also Texas v. Biden, 20 F.4th at 951
(“If the agency opened the issue up anew, and then reexamined and
reaffirmed its prior decision, the agency’s second action (the reaffirmance)
is reviewable. . . . But if the agency merely reaffirmed its decision without
24
No. 23-10362
really opening the decision back up and reconsidering it, the agency’s initial
action is the only final agency action to review.” (quotation omitted)).
Courts have articulated various tests for determining whether an
agency has reopened a prior decision. These tests fall into two general
categories.
Under the first, courts look “to the entire context of the [relevant
agency action] including all relevant proposals and reactions of the agency to
determine whether an issue was in fact reopened.” Pub. Citizen v. Nuclear
Regul. Comm’n, 901 F.2d 147, 150 (D.C. Cir. 1990); see also, e.g., id. at 150–
53; Growth Energy, 5 F.4th at 21–22; Nat’l Ass’n of Reversionary Prop. Owners
v. Surface Transp. Bd., 158 F.3d 135, 141–46 (D.C. Cir. 1998). An agency can
reopen an earlier decision in many ways, but the quintessential example of
this type of reopening is when an agency “hold[s] out [its prior rule] as a
proposed regulation, offer[s] an explanation for its language, solicit[s]
comments on its substance, and respond[s] to the comments in promulgating
the regulation in its final form.” Am. Iron & Steel Inst. v. EPA, 886 F.2d 390,
397 (D.C. Cir. 1989). Under the second reopening category, courts consider
whether an agency “constructively reopened” its prior decision. Kennecott
Utah Copper Corp. v. DOI, 88 F.3d 1191, 1214–15 (D.C. Cir. 1996). They do
so by evaluating whether “the revision of accompanying regulations
significantly alters the stakes of judicial review as the result of a change that
could have not been reasonably anticipated.” NRDC v. EPA, 571 F.3d 1245,
1266 (D.C. Cir. 2009) (quotation omitted).
Although a close call, we are unsure at this preliminary juncture and
after truncated review that FDA reopened the 2000 Approval in its 2016
Major REMS Changes and its 2021 Petition Denial.
As for the first reopening test, neither the 2016 Major REMS Changes
nor the 2021 Petition Denial appears to “substantive[ly] reconsider[]”
25
No. 23-10362
FDA’s 2000 Approval. Growth Energy, 5 F.4th at 21. FDA’s 2016 decision to
relax many of the REMS was issued in response to Danco’s supplemental
application requesting as much. See PI App. 615–52. And FDA’s 2021
Petition Denial was issued in response to plaintiffs’ 2019 Citizen Petition
asking FDA to “restore” the pre-2016 REMS—not revoke or reconsider
FDA’s underlying 2000 Approval. See PI App. 667–93. Therefore neither of
the “relevant proposals” prompted FDA to reopen and reconsider its 2000
Approval. Pub. Citizen, 901 F.2d at 150.
That said, the district court correctly noted that FDA nevertheless
“undertook a full review of the Mifepristone REMS Program” when it
reviewed plaintiffs’ 2019 Citizen Petition—even though the plaintiffs only
asked FDA to restore the pre-2016 status quo ante. See PI App. 735–76; FDA
Add. 22. In FDA’s words:
In 2021, FDA also undertook a full review of the Mifepristone
REMS Program. In conducting this review, FDA reviewed
multiple different sources of information, including published
literature, safety information submitted to the Agency during
the COVID-19 PHE, FDA Adverse Event Reporting System
(FAERS) reports, the first REMS assessment report for the
Mifepristone REMS Program, and information provided by
advocacy groups, individuals, and the Plaintiffs in ongoing
litigation, as well as information submitted by the sponsors of
the NDA and the ANDA[.]
PI App. 735. And after conducting this unrequested “full review” of the
REMS Program, FDA (inter alia) added two modifications to the REMS
Program that plaintiffs never even mentioned in their 2019 Citizen Petition,
including “a requirement that pharmacies that dispense the drug be specially
certified.” PI App. 736; see also id. at 735 n.11 (acknowledging that “this was
not raised in your Petition”). All of this suggests FDA went back to the
beginning, including its very first REMS report, and conducted an
26
No. 23-10362
independent review that far exceeded the issues raised in the 2019 Citizen
Petition.
Especially because the dangerousness of a drug is grounds to withdraw
its approval, see 21 U.S.C. § 355(e)—and REMS are required to “ensure that
the benefits of the drug outweigh the risks,” id. § 355-1(a)(1)–(2)—plaintiffs
reasonably argue that FDA’s 2021 “full review” of the entire REMS
Program was in effect a reconsideration of FDA’s 2000 Approval. Indeed,
plaintiffs might very well prevail on that claim later in this litigation. But at
this early juncture—and in light of our necessarily truncated review—we are
not yet confident enough to say that viewed in “the entire context,” FDA
“has undertaken a serious, substantive reconsideration of the [2000
Approval]” rather than “incremental adjustments to existing regulations.”
Texas v. Biden, 20 F.4th at 952–93 (quotation omitted).
The result is the same under the second reopening test. Recall that
under the second test, “[a] constructive reopening occurs if the revision of
accompanying regulations significantly alters the stakes of judicial review as
the result of a change that could have not been reasonably anticipated.”
Sierra Club, 551 F.3d at 1025 (quotation omitted).
Sierra Club is the seminal case. In 1994, EPA adopted a rule that
exempted major sources of air pollution from the Clean Air Act’s emission
standards during startups, shutdowns, and malfunctions (the “SSM
exemption”). Id. at 1022. But the 1994 rule also required sources to develop
an SSM plan in order to receive the benefit of the SSM exemption. Ibid. An
SSM plan required “the source to demonstrate how it will do its reasonable
best to maintain compliance with the standards, even during SSMs.” Ibid.(quotation omitted). SSM plans were publicly available and were
incorporated into the sources’ permits under Title V of the Clean Air Act.
Ibid.27
No. 23-10362
In a series of rulemakings between 2002 and 2006, EPA substantially
weakened the requirement that sources maintain and follow an SSM plan in
order to benefit from the SSM exemption. It removed the requirement that a
source’s Title V permit incorporate its SSM plan; it stopped making SSM
plans publicly available; and it ultimately retracted the requirement that
sources implement their SSM plans during SSM periods. Id. at 1023.
The Sierra Club filed suit in 2007. But the Sierra Club did not
challenge the changes to the SSM plan requirements that EPA had adopted
in its 2002, 2003, and 2006 rulemakings. Instead, it challenged the legality of
the SSM exemption itself. Id. at 1024. EPA had adopted that exception in
1994 and had not considered rescinding it in any of its rulemakings during the
2000s. Rather, those rulemakings had treated the SSM exemption as a
given—in fact, they had strengthened it by weakening the SSM plan
requirements. See id. at 1022–23.
The D.C. Circuit nonetheless held that the Sierra Club’s challenge to
the SSM exemption was timely. Even though EPA had not expressly
reopened its decision to create a SSM exemption, it had constructively
reopened that decision “by stripping out virtually all of the SSM plan
requirements that it created to contain that exemption.” Id. at 1025(quotation omitted). Because EPA had allegedly abandoned these “necessary
safeguards” limiting the SSM exemption, its rulemakings had “changed the
calculus for petitioners in seeking judicial review and thereby constructively
reopened consideration of the exemption.” Id. at 1025–26 (quotation
omitted).
Sierra Club thus establishes that an agency can constructively reopen
a decision if it removes essential safeguards that had previously limited or
contained the impact of that decision. In making this determination, the D.C.
Circuit looks to the extent to which the agency has “alter[ed] th[e] regulatory
28
No. 23-10362
framework” and whether the agency has “work[ed] a change that [plaintiffs]
could not have reasonably anticipated.” Nat’l Biodiesel Bd., 843 F.3d at 1017.
Under Sierra Club and its progeny, FDA’s 2016 Major REMS
Changes and 2021 Petition Denial seemingly reopened its 2000 Approval
decision. Of course, FDA did not expressly reconsider its mifepristone
approval. But it eliminated the “necessary safeguards,” Sierra Club, 551 F.3d
at 1025, that had accompanied and limited the impact of that approval for two
decades. The in-person dispensing requirement, for example, was critical to
FDA’s initial approval of mifepristone in 2000, which relied on the in-person
dispensing requirement to dismiss concerns about provider qualifications,
improper use, illicit distribution, and detection of adverse events. See PI App.
519–23. And the in-person dispensing requirement was also the cornerstone
of the REMS for mifepristone that FDA approved in 2011 and then relied on
in its 2016 rejection of plaintiffs’ 2002 Citizen Petition. See PI App. 578–82,
605, 608.
Thus FDA’s elimination of the in-person distribution requirement—
not to mention various other REMS—arguably worked a “sea change” in the
legal framework governing mifepristone distribution that plaintiffs “could
not have reasonably anticipated” and that “significantly alters the stakes of
judicial review.” Nat’l Biodiesel Bd., 843 F.3d at 1017 (quotation omitted).
That’s because the in-person dispensing requirement was FDA’s primary
tool for ensuring the safe distribution and use of mifepristone, so plaintiffs
arguably had little reason to anticipate this important change before 2021.
FDA does not argue otherwise, appearing to concede that its 2021
announcement was a stark departure from previous regulatory approaches.
And because this change eliminates a major safeguard against complications
and adverse effects arising from improper mifepristone use, it can be said to
“significantly alter[] the stakes of judicial review” for plaintiff doctors who
treat patients with these complications. Ibid. (quotation omitted).
29
No. 23-10362
Even so, we ultimately hold at this early and emergency stage that
these alterations didn’t constructively reopen the 2000 Approval for review.
That’s because there’s at least a colorable argument that plaintiffs “could
have . . . reasonably anticipated” changes like those in 2016 and 2021 by dint
of the statutorily defined supplemental application process and other similar
revision mechanisms. NRDC v. EPA, 571 F.3d at 1266 (quotation omitted);
see, e.g., 21 C.F.R. § 314.71(b). We also recognize that it’s somewhat of a
strain to say that the 2016 Major REMS Changes and 2021 Petition Denial
(and related changes) altered the regulatory landscape to such a degree that
the prior rule is only now “worth challenging” when it otherwise might “not
have been.” Sierra Club, 551 F.3d at 1025–26 (quotation omitted). After all,
plaintiffs did challenge the 2000 Approval well before the 2016 and 2021
changes were even proposed. But again, plaintiffs could very well prevail on
this reopening claim.
In the alternative, the district court held that plaintiffs were entitled
to equitable tolling of the statute of limitations. FDA Add. 23–25. We are
unpersuaded. “[A] litigant is entitled to equitable tolling of a statute of
limitations only if the litigant establishes two elements: ‘(1) that he has been
pursuing his rights diligently, and (2) that some extraordinary circumstance
stood in his way and prevented timely filing.’” Menominee Indian Tribe of Wis.
v. United States, 577 U.S. 250, 255 (2016) (quoting Holland v. Florida, 560
U.S. 631, 649 (2010)). Here, no “extraordinary circumstance” prevented
plaintiffs from filing within six years of FDA’s 2016 Petition Denial. The
district court is of course correct that FDA took “13 years, 7 months, and 9
days” to render that March 2016 ruling, FDA Add. 24, but that delay had no
impact on the length of the statute-of-limitations period or plaintiffs’ capacity
to challenge the 2016 Petition Denial.
30
No. 23-10362
C.
Next exhaustion. Stay applicants contend they are likely to succeed on
the merits because plaintiffs failed to exhaust their claims before FDA. We
disagree.
“As a general rule, claims not presented to the agency may not be
made for the first time to a reviewing court.” Wash. Ass’n for Television &
Child. v. FCC, 712 F.2d 677, 680 (D.C. Cir. 1983); cf. United States v. L.A.
Tucker Truck Lines, 344 U.S. 33, 37 (1952). For challenges to FDA actions,
the general administrative exhaustion requirement is codified at 21 C.F.R.
§ 10.45(b). Section 10.45(b) states that a “request that the [FDA]
Commissioner take or refrain from taking any form of administrative action
must first be the subject of a final administrative decision based on a petition
submitted under § 10.25(a).” See id. § 10.25(a) (“An interested person may
petition the [FDA] Commissioner to issue, amend, or revoke a regulation or
order, or to take or refrain from taking any other form of administrative
action.”).
No one disputes that every argument the plaintiffs raised in their 2019
Citizen Petition is exhausted. That includes all of plaintiffs’ challenges to the
2016 Major REMS Changes and everything fairly embraced by those
challenges. For example, the 2019 Citizen Petition argued explicitly that
FDA should “[c]ontinue limiting the dispensing of Mifeprex to patients in
clinics, medical offices, and hospitals.” FDA Add. 193, 209–16. When FDA
rejected that request in the 2021 Petition Denial, it expressly reaffirmed its
commitment to mail-order abortion drugs. As such, plaintiffs have properly
exhausted their challenge to FDA’s by-mail distribution regime by raising it
in the 2019 Citizen Petition.
Even if plaintiffs failed to exhaust their claims, courts retain
“discretion to waive exhaustion” where one of the “traditionally
31
No. 23-10362
recognized” exceptions applies. Wash. Ass’n for Television & Child., 712 F.2d
at 681–82. Two exceptions are relevant here: futility and administrative abuse
of process.
Start with futility. Plaintiffs need not exhaust claims where they can
demonstrate “the futility or inadequacy of administrative review.” Gardner
v. Sch. Bd. Caddo. Par., 958 F.2d 108, 112 (5th Cir. 1992); see also Honig v.
Doe, 484 U.S. 305, 327 (1988). The futility exception applies when
exhaustion would be “clearly useless” and “it is certain [a] claim will be
denied.” Tesoro Refin. & Mktg. Co. v. FERC, 552 F.3d 868, 874 (D.C. Cir.
2009) (quotation omitted); see also Carr v. Saul, 141 S. Ct. 1352, 1361 (2021)
(“[T]his Court has consistently recognized a futility exception to exhaustion
requirements.”).
Given FDA’s 2016 Petition Denial and its 2021 Petition Denial, it
would have been futile for plaintiffs to include a challenge to the 2000
Approval in their 2019 Citizen Petition. FDA rejected this exact challenge in
its 2016 Petition Denial. So it would have been “clearly useless” to raise the
precise challenge again in the 2019 Citizen Petition. Further, this exact
reasoning applies with equal force to plaintiffs’ challenge to the 2019 Generic
Approval because it’s entirely dependent on the underlying 2000 Approval.
Thus, plaintiffs’ challenges to the 2000 Approval and the 2019 Generic
Approval are not barred by exhaustion.
Next, administrative abuse of process. It’s well-established that where
an agency fails to follow its own regulations, exhaustion may not be required.
See Way of Life Television Network, Inc. v. FCC, 593 F.2d 1356, 1359–60 (D.C.
Cir. 1979); see also Wash. Ass’n for Television & Child., 712 F.2d at 681. That’s
especially true “where the obvious result would be a plain miscarriage of
justice.” Hormel v. Helvering, 312 U.S. 552, 558 (1941). Here, FDA was
required by its own regulations to respond to citizen petitions within 180
32
No. 23-10362
days. See 21 C.F.R. § 10.30(e)(2). Instead of timely responding, FDA
responded to plaintiffs’ first petition fourteen years after it was filed. And it
responded to the second petition over two years after it was filed. FDA
plainly and repeatedly refused to follow its own regulations here. Even
assuming any of plaintiffs’ challenges were unexhausted and that it wasn’t
futile to raise them before FDA, FDA’s repeated failure to follow its own
regulations indicates that the district court did not abuse its “discretion to
waive exhaustion.” Wash. Ass’n for Television & Child., 712 F.2d at 681.
D.
As applicants recognize, FDA’s actions are constrained by the APA’s
arbitrary-and-capricious standard. See 5 U.S.C. § 706(2)(A). Under that
standard, “the agency must examine the relevant data and articulate a
satisfactory explanation for its action including a rational connection between
the facts found and the choice made.” Motor Vehicle Mfrs. Ass’n of U.S., Inc.
v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (quotation omitted);
see also Sw. Elec. Power Co. v. EPA, 920 F.3d 999, 1013 (5th Cir. 2019) (judicial
review of agency action “is not toothless”). We must “consider whether the
decision was based on a consideration of the relevant factors and whether
there has been a clear error of judgment.” State Farm, 463 U.S. at 43(quotation omitted). An agency’s action is “arbitrary and capricious” if it
“entirely failed to consider an important aspect of the problem, offered an
explanation for its decision that runs counter to the evidence before the
agency, or is so implausible that it could not be ascribed to a difference in
view or the product of agency expertise.” Ibid.When an agency acts, it must “reasonably consider[] the relevant
issues and reasonably explain[]” its actions. FCC v. Prometheus Radio Project,
141 S. Ct. 1150, 1158 (2021); see also ibid. (“The APA’s arbitrary-and-
capricious standard requires that agency action be reasonable and reasonably
33
No. 23-10362
explained.”); Michigan v. EPA, 576 U.S. 743, 750, 752 (2015) (“[A]gency
action is lawful only if it rests on a consideration of the relevant factors” and
“important aspect[s] of the problem.” (quotation omitted)). Of course, we
cannot “substitute” our “own policy judgment for that of the agency.”
Prometheus, 141 S. Ct. at 1158. We nonetheless must still carefully ensure that
“the agency has acted within a zone of reasonableness and, in particular, has
reasonably considered the relevant issues and reasonably explained the
decision.” Ibid. The upshot is that we “must set aside any action premised
on reasoning that fails to account for ‘relevant factors’ or evinces ‘a clear
error of judgment.’” Univ. of Tex. M.D. Anderson Cancer Ctr. v. HHS, 985
F.3d 472, 475 (5th Cir. 2021) (quoting Marsh v. Or. Nat. Res. Council, 490
U.S. 360, 378 (1989)).
Here, applicants have failed to carry their burden at this preliminary
stage to show that FDA’s actions 5 were not arbitrary and capricious. We have
two principal concerns in that regard. First, FDA failed to “examine the
relevant data” when it made the 2016 Major REMS changes. State Farm, 463
U.S. at 43. That’s because FDA eliminated REMS safeguards based on
studies that included those very safeguards. FDA Add. 59, 122–23, 171. Imagine
that an agency compiles studies about how cars perform when they have
passive restraint systems, like automatic seatbelts. See State Farm, 463 U.S.
at 34–36. For nearly a decade, the agency collects those studies and continues
studying how cars perform with passive safety measures. Then one day the
agency changes its mind and eliminates passive safety measures based only on
existing data of how cars perform with passive safety measures. Cf. id. at 47–
_____________________
5
Here we limit our discussion to FDA’s decisions in the 2016 Major REMS
Changes and its subsequent agency actions. As described above in Part III.B, it appears at
this preliminary juncture that plaintiffs’ challenges to the 2000 Approval and 2016 Petition
Denial are untimely.
34
No. 23-10362
49. That was obviously arbitrary and capricious in State Farm. And so too
here. The fact that mifepristone might be safe when used with the 2000
Approval’s REMS (a question studied by FDA) says nothing about whether
FDA can eliminate those REMS (a question not studied by FDA).
True, FDA studied the safety consequences of eliminating one or two
of the 2000 Approval’s REMS in isolation. But it relied on zero studies that
evaluated the safety-and-effectiveness consequences of the 2016 Major
REMS Changes as a whole. This deficiency shows that FDA failed to consider
“an important aspect of the problem” when it made the 2016 Major REMS
Changes. Michigan v. EPA, 576 U.S. at 752 (quotation omitted).
Second, the 2016 Major REMS Changes eliminated the requirement
that non-fatal adverse events must be reported to FDA. After eliminating that
adverse-event reporting requirement, FDA turned around in 2021 and
declared the absence of non-fatal adverse-event reports means mifepristone
is “safe.” See, e.g., FDA Add. 861–76 (explaining that FDA’s FAERS
database, which collates data on adverse events, indicated that the 2016
Major REMS Changes hadn’t raised “any new safety concerns”). This
ostrich’s-head-in-the-sand approach is deeply troubling—especially on a
record that, according to applicants’ own documents, necessitates a REMS
program, a “Patient Agreement Form,” and a “Black Box” warning. See
supra Part III.A. And it suggests FDA’s actions are well “outside the zone of
reasonableness.” Prometheus, 141 S. Ct. at 1160. It’s unreasonable for an
agency to eliminate a reporting requirement for a thing and then use the
resulting absence of data to support its decision.
These actions make it unlikely that plaintiffs’ arbitrary-and-capricious
challenges will fail on the merits, at least as far as they challenge FDA’s
decisions including and following the 2016 Major REMS Changes.
35
No. 23-10362
IV.
Beyond likelihood of success on the merits, we also must consider the
other three factors for granting a stay. Those are “[A] whether the applicant
will be irreparably injured absent a stay; [B] whether issuance of the stay will
substantially injure the other parties interested in the proceeding; and [C]
where the public interest lies.” Nken, 556 U.S. at 434 (quotation omitted).
We address each in turn. And we (D) discuss how the Comstock Act, 18
U.S.C. §§ 1461, 1462 affects the stay inquiry. Outside of the 2000 Approval,
we find that the applicants fail to make a strong showing on any of these
factors for a stay.
A.
Of the remaining three factors, irreparable injury matters most. See
Nken, 556 U.S. at 434. FDA argues that the plaintiffs fail to show irreparable
injury. But the irreparable injury factor asks whether “the [stay] applicant will
be irreparably injured” absent a stay, not whether the plaintiff would be
irreparably injured absent an injunction. Ibid. (emphasis added) (quotation
omitted). Similarly, FDA’s assertion that the district court’s injunction will
harm pregnant women or other members of the public does not speak to the
irreparable injury factor (although it may speak to other factors), because
those persons are not stay applicants in this case.
Since FDA does not articulate any irreparable harm that FDA will
suffer absent a stay, it makes no showing on this “critical” prong. Ibid. We
may not need to address the merits of the applicants’ stay request any further,
because failure to show irreparable injury often “decides the [stay]
application.” Whalen v. Roe, 423 U.S. 1313, 1318 (1975) (Marshall, J., in
chambers).
Danco by contrast does claim it will suffer irreparable injury, albeit in
just one paragraph. Danco notes that mifepristone is its sole product and
36
No. 23-10362
argues that it may have to shut down absent relief. We have held that
catastrophic financial losses “may be sufficient to show irreparable injury.”
Wages & White Lion Investments, LLC v. FDA, 16 F.4th 1130, 1142 (5th Cir.
2021) (emphasis added) (quotation omitted). Of course, irreparable injury
alone does not entitle Danco to a stay. See Virginian Ry. Co., 272 U.S. at 672.
And even if it did, neither FDA nor Danco articulates why this, or any
other, injury would require a stay of all of the district court’s order, rather
than only part. Recall that we may narrowly “tailor a stay” to impact “only
some portion of the proceeding.” Int’l Refugee Assistance Project, 137 S. Ct. at
2087 (quotation omitted). The applicants’ arguments suggest, at best, that
they require relief only from the district court’s treatment of the 2000
Approval. They make no argument as to why the district court’s treatment
of the 2016 Major REMS Changes and later FDA activity irreparably harms
anyone.
Applicants’ forfeiture of this contention is understandable because
the world operated under the 2000 Approval for sixteen years, apparently
without problems. And neither applicant contends that it’ll be irreparably
injured without a stay so long as the 2000 Approval and its associated REMS
remain in effect. Thus, the irreparable injury factor counsels against a stay.
B.
The next Nken factor asks whether “issuance of the stay will
substantially injure the other parties interested in the proceeding.” 556 U.S.
at 434 (quoting Hilton, 481 U.S. at 776); see also Ala. Ass’n of Realtors, 141 S.
Ct. at 2487 (same); Planned Parenthood v. Abbott, 134 S. Ct. 506, 506–08
(2013) (mem.) (opinions of seven Justices using the same standard). This
language again focuses on harm from the stay, not the injunction. Cf. Whole
Woman’s Health, 141 S. Ct. at 2495 (using less specific “balance of the
equities” language). To succeed on this prong, applicants must show that the
37
No. 23-10362
requested stay will not harm the opposing appellees or other interested
parties.
Applicants discuss at length their view that the district court’s order
might harm various persons, but mostly decline to address the apposite
question, which is why the requested stay would not harm relevant persons.
What points the applicants do make on this relevant question distill down to
two arguments.
First, applicants briefly argue that the injuries the plaintiffs would
suffer from a stay are speculative or minimal. But we have already addressed
why plaintiffs’ injuries are non-speculative. See supra Part III.A. We have also
addressed the specific risks impacting women and the plaintiffs that stem
from the 2016 Major REMS Changes and other post-2016 FDA decisions
that the district court enjoined. See supra Part III.A, D. The applicants’
abbreviated argument focuses on consequences flowing from the district
court’s treatment of the 2000 Approval and largely ignores plaintiffs’
alternative arguments regarding the 2016 Major REMS Changes and what
followed.
Second, the applicants argue that the plaintiffs’ failure to bring
litigation sooner undercuts any contention that they would be harmed from a
stay. That contention is untenable given FDA’s fourteen-year delay in
adjudicating the 2002 Citizen Petition. But, even setting aside FDA’s own
delays, the applicants do not explain why the plaintiffs’ alleged
procrastination warrants a stay of the entirety of the district court’s order,
rather than just the portion of the order impacted by long litigation delay (the
2000 Approval).
To the extent applicants make any showing that the third Nken factor
favors a stay, they do so only with respect to the 2000 Approval and do not
address plaintiffs’ alternative arguments.
38
No. 23-10362
C.
The last Nken factor asks “where the public interest lies.” 556 U.S. at
434 (quotation omitted). The stay applicants make three principal
arguments.
First, the applicants argue that “procedural irregularity” in the court
below favors relief. But the applicants do not explain why any specific alleged
irregularity necessarily speaks to public (versus their own private) interest.
Even if we assume away that problem, it is not clear to us, on our accelerated
review, that any litigation below was irregular. And even if we assume, which
we do not, that the district court or the plaintiffs departed from acceptable
procedure, it’s unclear on this record that applicants have embraced “the
principles of equity and righteous dealing” in the twenty-one years since the
filing of the 2002 Citizen Petition. Binh Hoa Le v. Exeter Fin. Corp., 990 F.3d
410, 416 (5th Cir. 2021) (quotation omitted) (noting that a party’s own
imperfect conduct can prejudice their request for equitable relief).
Second, Danco argues that avoidance of “judicial conflict” warrants a
stay given the order of an out-of-circuit district court. Comity between
federal courts is a cognizable interest. See Def. Distrib. v. Platkin, 55 F.4th
486, 495–96 (5th Cir. 2022). We have every respect for fellow federal courts.
But we cannot embrace an argument that would, in effect, allow the decision
of an out-of-circuit district court to impel us towards “extraordinary” relief
that would be otherwise inappropriate. Williams, 442 U.S. at 1311 (quotation
omitted).
Third, the stay applicants warn us of significant public consequences
should the district court’s order result in the withdrawal of mifepristone from
the market. These consequences, the applicants say, include injury to
pregnant women, to public healthcare systems, and to the sense of order that
governs FDA drug approvals. But these concerns center on the district
39
No. 23-10362
court’s removal of mifepristone from the market. The applicants make no
arguments as to why the 2016 Major REMS Changes, the 2019 Generic
Approval, or the 2021 and 2023 Mail Order Decisions are similarly critical to
the public even though they were on notice of plaintiffs’ alternative requests
for relief. And it would be difficult for applicants to argue that the 2016 Major
REMS Changes and subsequent FDA activity were so critical to the public
given that the Nation operated—and mifepristone was administered to
millions of women—without them for sixteen years following the 2000
Approval.
The applicants have made some showing that the public interest
warrants equitable relief from the district court’s treatment of the 2000
Approval. Motivated in part by the accelerated posture of our review, we
credit their showing.
D.
The parties vehemently dispute how their competing interpretations
of the Comstock Act of 1873 might impact the validity of the district court’s
order. The Comstock Act prohibits the carriage in interstate commerce of
“any drug, medicine, article, or thing designed, adapted or intended for
producing abortion.” 18 U.S.C. § 1462. It similarly prohibits the mailing of
any “article, instrument, substance, drug, medicine, or thing which is
advertised or described in a manner calculated to lead another to use or apply
it for producing abortion.” Id. § 1461.
Both statutory provisions specify a mens rea of “knowingly.” Id.
§§ 1461–62. The plain text does not require that a user of the mails or
common interstate carriage intend that an abortion actually occur. Rather, a
user of those shipping channels violates the plain text merely by knowingly
making use of the mail for a prohibited abortion item.
40
No. 23-10362
The applicants’ principal defense against the Comstock Act is that
FDA was not required to consider it. After all, say the applicants, 21 U.S.C.
§§ 355 and 355-1 guide FDA’s discretion over drug approval and REMS, and
those statutes do not explicitly require consideration of other statutes like 14
U.S.C. § 1462.
Even assuming that’s true, however, the Comstock Act nevertheless
undermines applicants’ showing on the final three Nken factors. For
example, if the Comstock Act is construed in-line with its literal terms, then
Danco cannot say it is irreparably harmed by the district court’s order,
because Danco has no interest in continuing to violate the law, which (under
a plain view of the Act) it does every time it ships mifepristone. For further
example, if the Comstock Act is strictly understood, then applicants may lose
the public interest prong entirely, because there is no public interest in the
perpetuation of illegality. See Louisiana v. Biden, 55 F.4th 1017, 1035 (5th Cir.
2022).
The applicants raise other defenses. For example, they argue that the
Food and Drug Administration Amendments Act, Pub. L. No. 110-85, 121Stat. 823 (2007) (“FDAAA”) sub silentio repealed the Comstock Act, at least
where mifepristone is concerned. That’s because the FDAAA in 2007
created a statutory framework governing REMS and drugs with then-existing
distribution restrictions. See id. § 909(b). Mifepristone was one such drug.
So, say applicants, the FDAAA acted to legalize shipment of mifepristone,
regardless of what the Comstock Act might say. But “repeals by implication
are not favored.” Maine Cmty. Health Options v. United States, 140 S. Ct.
1308, 1323 (2020) (quotation omitted). We regard each of Congress’s
statutes as effective unless either “intention to repeal” one of them is “clear
and manifest” or the two laws are “irreconcilable.” Ibid. (quotation
omitted). Section 909(b) did not expressly legalize mifepristone; agency
action (not statute) did that. Section 909(b)’s brief text makes no mention of
41
No. 23-10362
mifepristone at all. So, there is no “irreconcilable” conflict. And we hesitate
to find “clear and manifest” intention to repeal a 150-year-old statute that
Congress has otherwise repeatedly declined to alter in the far reaches of a
single section of the cavernous FDAAA.
Failing all else, the applicants argue that the Comstock Act does not
mean what it says it means. Or rather, that judicial gloss and lax enforcement
over the past century act to graft relevant exceptions onto it. The applicants
rely on a memo authored by the Office of Legal Counsel to press this position.
See FDA Add. 258–78. That memo’s thorough exploration of this topic notes
that a variety of aging out-of-circuit opinions and a single footnote within one
Supreme Court dissent favor the applicants’ position. FDA Add. 262–68).
The speed of our review does not permit conclusive exploration of this
topic. To the extent the Comstock Act introduces uncertainty into the
ultimate merits of the case, that uncertainty favors the plaintiffs because the
applicants bear the burden of winning a stay. See Landis, 299 U.S. at 255.
Since plaintiffs already prevail on most Nken factors concerning most of the
agency items effectively enjoined by the district court’s order, we need not
definitively interpret the Comstock Act to resolve this stay application.
* * *
As the stay applicants, defendants bear the burden of showing why
“extraordinary circumstances” demand that we exercise discretion in their
favor. To the extent the defendants make any such showing, they do so only
with respect to the 2000 Approval—not the plaintiffs’ alternative arguments
challenging FDA’s 2016 Major REMS Changes and all subsequent actions.
Our decision to grant partial relief does not reflect our view on any merits
question. The defendants’ motions to stay the district court’s order are
GRANTED IN PART and DENIED IN PART. The appeal is
EXPEDITED to the next available Oral Argument Calendar.
42