Liquidia Technologies, Inc. v. United States Food and Drug Administration

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

LIQUIDIA TECHNOLOGIES, INC., Plaintiff, v. FOOD AND DRUG ADMINISTRATION et al., Civil Action No. 24-2428 (TJK) Defendants, v. UNITED THERAPEUTICS CORPORATION, Intervenor-Defendant.

MEMORANDUM OPINION On August 16, 2024, the U.S. Food and Drug Administration tentatively approved Plaintiff Liquidia Technologies, Inc.’s, New Drug Application for its proposed drug Yutrepia, a treprostinil inhalation powder designed to treat two types of pulmonary arterial hypertension. But that ap- proval was only tentative because FDA determined that Yutrepia’s immediate approval was barred by another drug’s period of temporary exclusivity held by Intervenor-Defendant United Therapeu- tics Corporation. Thus, according to FDA, it cannot finally approve Yutrepia until that exclusivity expires on May 23, 2025. Liquidia now sues FDA, asking the Court to set aside that determination as arbitrary, capricious, and contrary to law. The Court finds that FDA’s decision was proper. So it will deny summary judgment for Liquidia and grant summary judgment in FDA’s and UTC’s favor.

I. Background A. Legal Background The Food, Drug, and Cosmetic Act (“FDCA”) prohibits “introduc[ing] into interstate com- merce any new drug, unless an approval of an application filed pursuant to [the FDCA] is effective with respect to such drug.” 21 U.S.C. § 355(a); Veloxis Pharms., Inc. v. FDA, 109 F. Supp. 3d 104, 107 (D.D.C. 2015). That prohibition requires companies seeking to bring new medicines to market to file a New Drug Application (“NDA”) with the U.S. Food and Drug Administration (“FDA”). AstraZeneca Pharms. LP v. FDA, 872 F. Supp. 2d 60, 62 (D.D.C. 2012), aff’d, 713 F.3d 1134 (D.C. Cir. 2013). Under the FDCA, there are three pathways for an NDA to obtain FDA approval. Veloxis, 109 F. Supp. 3d at 108. The first, known as the “full NDA process,” requires “the manufacturer to submit detailed safety and efficacy data for the drug.” Takeda Pharms., U.S.A., Inc. v. Burwell, 78 F. Supp. 3d 65, 71 (D.D.C. 2015) (Jackson, J.), aff’d in part, vacated in part, 691 F. App’x 634 (D.C. Cir. 2016) (citation omitted); see 21 U.S.C. § 355(b)(1). “This path is used by drug manufacturers for ‘new branded drug[s].’” Id. (quoting Ethypharm S.A. France v. Abbott Lab’ys, 707 F.3d 223, 226 (3d Cir. 2013)).

But not all NDAs have to start from scratch. When applicants seek approval for a generic version of an already-approved drug, they may “choose to file an Abbreviated New Drug Appli- cation (‘ANDA’) pursuant to 21 U.S.C. § 355(j).” Takeda, 78 F. Supp. 3d at 71. Under that pathway, which “facilitates efficient approval of generic versions of pioneer drug products that have already been determined to be safe and effective,” an applicant need not “attempt to demon- strate [the] safety or effectiveness” of its proposed drug. Id. “[I]nstead, the applicant’s only goal is to establish that the generic product is equivalent to another drug that is already known to be safe and effective.” Id. That pathway will not work, however, for applicants seeking approval for a modified version of an already-approved drug. But, fortunately for them, such applicants need not go through the costly full NDA process. Instead, they may file NDAs under 21 U.S.C. § 355(b)(2)— also known as § 505(b)(2) of the FDCA. Braeburn Inc. v. FDA, 389 F. Supp. 3d 1, 6 (D.D.C. 2019). This method, the kind of NDA at issue here, “is a sort of hybrid of the other two pathways.”

Takeda, 78 F. Supp. 3d at 71. In a 505(b)(2) NDA, the applicant is still required to “directly demonstrate that the proposed drug product is safe and effective; however, like the ANDA, a 505(b)(2) applicant can rely on clinical studies that were previously submitted to FDA in support of another drug and that were not conducted or licensed by the 505(b)(2) applicant.” Id. at 72 (citing 21 U.S.C. § 355(b)(2)). That is, these NDAs may “rely on investigations that were ‘not conducted by or for the applicant and for which the applicant has not obtained a right of reference or use from the person by or for whom the investigations were conducted.’” Braeburn, 389 F. Supp. 3d at 6 (quoting 21 U.S.C. § 355(b)(2)). Still, the “505(b)(2) applicant must present infor- mation that bears upon the safety and effectiveness of its drug product in light of the difference between the pioneer drug product and the applicant’s modification of that drug product.” Takeda, 78 F. Supp. 3d at 72.

That said, FDA will not always approve an NDA just because the proposed drug is safe and effective or is equivalent to an already-approved drug. The FDCA also establishes that drugs meeting certain requirements qualify for periods of marketing exclusivity, during which competing drugs falling within the already-approved drug’s zone of exclusivity cannot be approved. Veloxis, F. Supp. 3d at 107–08; Braeburn, 389 F. Supp. 3d at 7. For example, when FDA approves an NDA for a drug “which includes an active moiety” that FDA has already approved and that “con- tains reports of new clinical investigations (other than bioavailability studies) essential to the ap- proval of the application” that were “conducted or sponsored by the applicant,” FDA “may not” approve any other NDAs for the same “conditions of approval of such [prior-approved] drug . . . effective before the expiration of three years from” the first NDA’s approval date.1 21 U.S.C. § 355(c)(3)(E)(iii). In other words, when FDA approves an NDA for a drug containing an already- approved chemical and that includes “reports of new clinical investigations . . . essential to the approval of the” NDA—at least so long as those investigations are not “bioavailability studies”— the company that sponsored those studies and submitted that first NDA receives a three-year period of statutory exclusivity over that NDA’s “conditions of approval.” Id. Thus, determining whether an NDA is barred by a different, already-approved NDA’s ex- clusivity requires answering two questions. First, is the prior NDA eligible for exclusivity under § 355(c)(3)(E)(iii)’s “eligibility clause”? Braeburn, 389 F. Supp. 3d at 7. Under the eligibility clause, an NDA “which includes an active moiety . . . that has been approved in another” NDA must meet four prerequisites to be eligible for exclusivity: (1) it must “contain reports of . . . clin- ical investigations (other than bioavailability studies)”; (2) those clinical investigations must be “new”; (3) they must be “essential to the approval of the” NDA; and (4) the applicant must “con- duct[] or sponsor[]” them. 21 U.S.C. § 355(c)(3)(E)(iii).

Second, if the already-approved NDA is eligible for exclusivity, FDA must then determine whether the pending NDA seeks approval “for the conditions of approval of such [already-ap- proved] drug.” 21 U.S.C. § 355(c)(3)(E)(iii). If it does, FDA “may not” approve the pending NDA. Id. This prohibition has led some courts to refer to this portion of the provision as the “bar clause.” Braeburn, 389 F. Supp. 3d at 7 (quotation omitted). Though neither the statute nor the regulations have defined “conditions of approval,” courts in this district have interpreted it to “set[]

An “active moiety” is “the molecule or ion . . . responsible for the physiological or phar- macological action of the drug substance.” 21 C.F.R. § 314.3(b).

up a ‘logical relationship between the change in the product for which the new clinical investiga- tions were essential to approval of the [already-approved NDA] and the scope of any resulting three-year exclusivity.’” Veloxis, 109 F. Supp. 3d at 120–21 (quoting AstraZeneca, 872 F. Supp. 2d at 80).

B. Factual Background On January 24, 2020, Liquidia filed a 505(b)(2) NDA for its proposed drug Yutrepia, a treprostinil inhalation powder designed to treat pulmonary arterial hypertension (“PAH”) and pul- monary hypertension associated with interstitial lung disease (“PH-ILD”). J.A. 430–31.2 Liquidia’s NDA relied largely “on the findings of safety and effectiveness for the listed drug Tyvaso.”3 J.A. 430. Full approval of Yutrepia, however, stalled in the face of patent litigation.

J.A. 1146. In the meantime, UTC filed an NDA for its own treprostinil inhalation powder, Tyvaso DPI, which FDA approved on May 23, 2022, for the treatment of PAH and PH-ILD. J.A. 423– 24, 505.

The Tyvaso DPI NDA also largely, but not exclusively, “relied on safety and efficacy data submitted in the Tyvaso NDA.” J.A. 424. Specifically, UTC submitted five studies to accompany its Tyvaso DPI NDA. The first two, known as TRIUMPH I and INCREASE, came from the Tyvaso NDA and proved that inhaled treprostinil was a safe and effective treatment for PAH and PH-ILD. J.A. 424–25. The final three, however, were new to the Tyvaso DPI NDA. The first, the MKC-475-001 study, was a study in which investigators “characterized” the

“J.A.” citations refer to the Joint Appendix the parties filed in accordance with Local Civil Rule 7(n), which contains the relevant portions of the administrative record. See ECF No. (sealed and unredacted); ECF No. 78 (redacted).

Tyvaso, created by United Therapeutics Corporation (“UTC”), is a “treprostinil[] inhala- tion solution for oral inhalation use” also approved to treat PAH and PH-ILD. J.A. 424.

“pharmacokinetics of” Tyvaso DPI and conducted various “safety assessments.” J.A. 425–26.

The second, known as BREEZE, was mainly designed to evaluate Tyvaso DPI’s “safety and tol- erability,” yet it also secondarily assessed its pharmacokinetics “after administration of each treat- ment.” J.A. 426. The final study, called the TIP-PH-102 study, was a “pivotal relative bioavaila- bility study” that conducted “PK [pharmacokinetics] and safety assessments” when “PK samples were collected up to 5 hours post-dose.” J.A. 429.

Liquidia, through two letters submitted to FDA on July 15, 2021, and July 15, 2022, argued that the Tyvaso DPI NDA warranted no exclusivity under the FDCA. J.A. 1–19; J.A. 283–92.

But, on August 16, 2024, FDA determined that Tyvaso DPI “qualifie[d] for 3-year exclusivity” under § 355(c)(3)(E)(iii) because its NDA “include[d] a new clinical investigation (other than a bioavailability study) that was essential to approval and conducted or sponsored by the applicant— the BREEZE study.” J.A. 431. And it further found that “the conditions of approval proposed in the Yutrepia [NDA] are within the scope of Tyvaso DPI’s exclusivity.” Id. FDA determined that “the BREEZE study answered for the first time whether the active moiety treprostinil administered as an inhalation powder is safe and tolerable for chronic use.” J.A. 442. Thus, FDA found that Tyvaso DPI held exclusivity for “the inhalation powder dosage form for the active moiety trepros- tinil for chronic use.” Id. And since “Yutrepia is a proposed treprostinil inhalation powder for chronic use,” FDA found that the three-year period of statutory exclusivity barred immediate ap- proval of Liquidia’s Yutrepia NDA. J.A. 450.

C. Procedural History Liquidia sued FDA and other federal defendants (collectively, “FDA”) on August 21, 2024, claiming that FDA’s failure to fully approve its Yutrepia NDA was arbitrary, capricious, and con- trary to law under the Administrative Procedure Act (“APA”). ECF No. 1 ¶¶ 122–35. Liquidia further moved for a preliminary injunction setting aside that decision. ECF No. 13. UTC also moved to intervene as a defendant to protect its interest in its statutory exclusivity. ECF No. 6.

The Court granted that intervention motion and entered the parties’ jointly proposed briefing schedule. Minute Orders of Aug. 30, 2024. Under that schedule, Liquidia filed a renewed motion for preliminary injunction and motion for summary judgment. ECF Nos. 32–33. FDA and UTC then filed oppositions and their own cross-motions for summary judgment. ECF Nos. 44–45, 47– 48.

Following briefing, the Court held a preliminary-injunction hearing on December 5, 2024.

There, the parties agreed that the Court could “resolve this [case] on the summary judgment mo- tions” and that it should “consolidate[] the [preliminary injunction] with the merits and just grant[] or deny[] the summary judgment motions.” ECF No. 88 at 5, 33. Thus, the Court will consolidate Liquidia’s motion for a preliminary injunction with the pending motions for summary judgment.4 And, for the reasons explained below, it will deny Liquidia’s motion for summary judgment and grant FDA’s and UTC’s motions for the same.

II. Legal Standards Although all parties move for summary judgment, the ordinary summary-judgment stand- ard does not apply. When a plaintiff “seeks review of agency action under the APA,” the court “sits as an appellate tribunal.” Am. Bioscience v. Thompson, 269 F.3d 1077, 1083 (D.C. Cir. 2001).

That is, it has no factfinding role because the case presents “a question of law.” Id. It must ask “whether the agency action is supported by the administrative record and otherwise consistent with the APA standard of review.” Citizens for Resp. & Ethics in Wash. v. SEC, 916 F. Supp. 2d 141, 145 (D.D.C. 2013). On top of its purely procedural requirements, the APA directs courts to “hold

As a result, the Court will deny Liquidia’s motion for preliminary injunction as moot.

See Braeburn, 389 F. Supp. 3d at 6 n.1.

unlawful and set aside agency action . . . found to be . . . arbitrary, capricious . . . or otherwise not in accordance with law.” 5 U.S.C. § 706(2).

When evaluating a claim that agency action is arbitrary or capricious, courts must ensure that the agency has “examine[d] the relevant data and articulate[d] a satisfactory explanation for its action.” FCC v. Fox Television Stations, 556 U.S. 502, 513 (2009) (quotation omitted). That standard does not empower a court to “substitute its judgment for that of the agency.” Id. (quota- tion omitted). In other words, the agency’s “policy choices” are not up for debate—only the “ex- planation it has given.” Id. at 530. That explanation must include a “rational connection between the facts found and the choice made.” Motor Vehicle Mfrs. Ass’n of the U.S., Inc. v. State Farm Mut. Auto Ins. Co., 463 U.S. 29, 43 (1983) (quotation omitted). The agency must consider all “important aspect[s] of the problem”—that is, those Congress considered relevant. Id. An agency also acts arbitrary and capriciously when it departs from its own regulations or when it changes course without recognizing or explaining why. Nat’l Env’t Dev. Assoc.’s Clean Air Project v. EPA, 752 F.3d 999, 1009 (D.C. Cir. 2014); Nuclear Energy Inst., Inc. v. EPA, 373 F.3d 1251, 1296 (D.C. Cir. 2004).

III. Analysis Liquidia argues that FDA’s decision not to immediately approve Yutrepia is arbitrary, ca- pricious, and contrary to law. First, Liquidia argues that FDA’s decision finding Tyvaso DPI eligible for exclusivity was erroneous. Second, even assuming Tyvaso DPI is eligible for exclu- sivity, it argues that FDA was wrong to conclude that Yutrepia fell within the scope of that exclu- sivity. The Court disagrees on both fronts.

A. Tyvaso DPI Is Eligible for Exclusivity Recall that for NDAs involving active moieties that FDA has already approved, an ap- proved NDA is eligible for exclusivity when it “contains reports of . . . clinical investigations

(other than bioavailability studies)” that are “new,” “essential to the approval of the” NDA, and “conducted or sponsored” by the applicant. 21 U.S.C. § 355(c)(3)(E)(iii). FDA found that Tyvaso DPI was eligible for exclusivity based on UTC’s BREEZE study. J.A. 431. While the parties do not dispute that BREEZE was conducted or sponsored by UTC, see ECF No. 55 at 13 n.4, Liquidia argues that it was (1) not a “clinical investigation[] (other than [a] bioavailability stud[y]),” (2) not “new,” and (3) not “essential to the approval of the” Tyvaso DPI NDA. It is wrong across the board.

1. BREEZE Is a Clinical Investigation Other Than a Bioavailability Study Liquidia’s first argument—that BREEZE is a bioavailability study unable to support ex- clusivity—is itself two-fold. First, Liquidia argues that FDA acted contrary to law in concluding that BREEZE was not a bioavailability study because its secondary endpoints (or measured out- comes) looked at bioavailability and pharmacokinetics. Second, Liquidia argues that FDA acted arbitrarily and capriciously in deciding that Tyvaso DPI qualified for exclusivity when it departed, without explanation, from its prior determination that it did not. The Court disagrees.

a. FDA’s Decision That BREEZE Was a Clinical Investigation Other Than a Bioavailability Study Was Not Contrary to Law Liquidia first argues that FDA acted contrary to law in deciding that BREEZE was a clin- ical investigation other than a bioavailability study. Its position is straightforward. According to Liquidia, the FDCA prohibits granting exclusivity under § 355(c)(3)(E)(iii) unless the NDA was supported by “clinical investigations (other than bioavailability studies).” It accepts that FDA’s regulations—which define a “clinical investigation” as “any experiment other than a bioavailabil- ity study in which a drug is administered or dispensed to, or used on, human subjects” and a bioa- vailability study as “a study to determine the bioavailability or the pharmacokinetics of a drug”— accurately define those statutory terms. 21 C.F.R. § 314.108(a); ECF No. 41 at 30–31. So it concludes that BREEZE, which secondarily studied Tyvaso DPI’s pharmacokinetics, must be a bioavailability study.

FDA and UTC also accept that FDA’s regulations correctly reflect the FDCA’s text. ECF No. 56 at 29; ECF No. 55 at 24. But they disagree with Liquidia’s conclusion that BREEZE is a “bioavailability study” incapable of supporting exclusivity just because one of its minor goals was to study Tyvaso DPI’s pharmacokinetics. Instead, in its decision, FDA noted that BREEZE’s “primary endpoint was safety and tolerability.” J.A. 432. Thus, even though bioavailability was relevant for some secondary endpoints, BREEZE was not “solely a bioavailability study.” J.A.

432. UTC takes a similar position, arguing that a study is a bioavailability study only if its “pri- mary purpose” is to study bioavailability. ECF No. 56 at 31. Thus, the parties’ disagreement is over how central the concept of bioavailability must be to a study before it should be considered a “bioavailability study.”

Ultimately, the Court agrees with FDA and UTC that BREEZE is a clinical investigation other than a bioavailability study. Liquidia’s arguments that the inclusion of any bioavailability data in a study effectively taints the entire study as a “bioavailability study” are unpersuasive. 5 To begin, the regulations, which define a “bioavailability study” as “a study to determine the bioavailability or the pharmacokinetics of a drug,” confirm that this is a purpose-based inquiry. 21 C.F.R. § 314.108(a) (emphasis added). The use of the infinitive here “denotes purpose.” See Allison Engine Co. v. United States ex rel. Sanders, 553 U.S. 662, 668 (2008). And common

FDA and UTC also argue that Liquidia has either waived or is estopped from arguing that BREEZE is a bioavailability study because that argument conflicts with those it made before FDA.

Because the Court agrees with FDA on the merits, it need not resolve that waiver or estoppel issue.

But the Court notes that, during administrative proceedings, Liquidia argued what the Court holds here: that investigations should be characterized based on their “primary endpoint[s]” alone and that “secondary endpoints a[re] irrelevant.” J.A. 10.

English usage reflects that the use of a noun to modify or complement another noun often denotes the second noun’s primary purpose. For example, if an employee uses her lunch break to eat her lunch and run an errand or check her personal email, it is still her lunch break—even if it also served those secondary purposes. Similarly, a grocery store is still referred to as such even if it also happens to sell things other than groceries—like batteries, or t-shirts, or greeting cards. In other words, that a grocery store also sells these other items does not transform it into a battery store, or a t-shirt store, or a greeting card store, because selling them is not the store’s primary purpose. Consistent with this common usage, a bioavailability study is a study that is primarily intended to study bioavailability. And that a study might have a secondary, bioavailability-related purpose does not transform it into a “bioavailability study.”

The rest of the FDCA’s text and structure supports this conclusion, even if it takes some unpacking to see why. As the Supreme Court has recognized, “[a] provision that may seem am- biguous in isolation is often clarified by the remainder of the statutory scheme.” United Sav. Ass’n of Tex. v. Timbers of Inwood Forest Assocs., Ltd., 484 U.S. 365, 371 (1988). For example, where “only one of the permissible meanings produces a substantive effect that is compatible with the rest of the law,” courts should interpret that phrase accordingly. Id. Here, read in its entirety, the FDCA makes clear that investment-backed, innovative NDAs supported by the applicant’s own studies that prove the safety or effectiveness of the innovative drug warrant exclusivity. On the other hand, non-innovative NDAs that rely on studies proving only that a proposed drug is bioe- quivalent to an already-approved drug do not. So studies like BREEZE, which mainly study the safety and efficacy of an innovative form of a drug, neatly fit within the category of studies that support exclusivity under the FDCA. That the study also looked at bioavailability as a secondary endpoint is not enough to make it a “bioavailability study,” thereby preventing it from supporting exclusivity. To conclude otherwise would “produce[] a substantive effect that is [in]compatible with the rest of the law.” Id. To understand why, it helps to appreciate when § 355(c)(3)(E)(iii) applies. Consider that the provision expressly limits exclusivity eligibility to “applications submitted under subsection (b).” Thus, ANDAs submitted under subsection (j) are ineligible for exclusivity under § 355(c)(3)(E)(iii).6 Two important factors distinguish subsection (b) NDAs and subsection (j) ANDAs. The first is the amount of innovation. As discussed above, NDAs submitted under sub- section (b) seek approval for either entirely new drugs or drugs that modify an already-approved drug. Takeda, 78 F. Supp. 3d at 71–72. So subsection (b) NDAs seek approval for drugs that are not “simply a generic version of a branded drug.” Id. at 72 (quotation omitted). On the other hand, the subsection (j) ANDA process is designed only to prove that the proposed drug “is equivalent to another drug that is already known to be safe and effective.” Id. at 71. That is, as a prerequisite to applying under subsection (j), an ANDA’s proposed drug generally cannot include significant innovation. The second characteristic that distinguishes subsection (b) NDAs from subsection (j) ANDAs is whether the application itself must “show whether [the proposed] drug is safe for use and whether such drug is effective in use.” 21 U.S.C. § 355(b)(1)(A)(i). As explained above, subsection (b) NDAs must meet this requirement (even if they rely on the research of others to do so). See id. § 355(b)(2). Subsection (j) ANDAs do not. Id. § 355(j)(2)(A)(vi).

On the whole, then, § 355(c)(3)(E)(iii) reflects a desire to protect and encourage—through a grant of exclusivity—innovative NDAs that themselves prove the safety and efficacy of proposed drugs without granting a commensurate exclusivity to copycat ANDAs that seek approval for

Indeed, they are ineligible for exclusivity under any provision in § 355. 21 U.S.C. § 355(c)(3)(E)(ii)–(v), (j)(5)(F)(ii)–(v).

generic drugs. For this reason, it makes sense to read the phrase “bioavailability studies”—in accordance with common English usage of a noun complementing another noun—as not including studies that were mainly intended to prove the safety or efficacy of a medical innovation and that only secondarily include bioavailability data.

Finally, this conclusion is bolstered by the way the phrase “bioavailability” is used through- out the FDCA: mainly in the context of providing companies a short-cut to gain approval for ge- neric drugs that are the bioequivalents of already-listed drugs. For example, “bioavailability” was first added to the FDCA in the Drug Price Competition and Patent Term Restoration Act of 1984, also known as the Hatch-Waxman Amendments. See Pub. L. No. 98-417, 98 Stat. 1585; Veloxis, 109 F. Supp. 3d at 107. And, in that Act, it most often appeared in subsection (j), which allows for the filing of ANDAs for new generic drugs with the same “conditions of use,” “active ingredi- ent” “route of administration, . . . dosage form, and . . . strength” as an already listed drug. See 21 U.S.C. § 355(j)(1), (2)(A); 98 Stat. at 1585–92. To be approved, an ANDA must submit, among other things, proof that the proposed drug is “bioequivalent” to another drug that FDA has already approved. 21 U.S.C. § 355(j)(2)(A)(ii)(I), (iv). The statute proceeds to establish that drugs are “bioequivalent” when “the rate and extent of absorption of the [proposed] drug do not show a significant difference from the rate and extent of absorption of the [already-approved] drug.” Id. § 355(j)(8)(B)(i). That is much like the statute’s definition of “bioavailability”: “the rate and ex- tent to which the active ingredient or therapeutic ingredient is absorbed from a drug.” Id. § 355(j)(8)(A)(i). Thus, drugs are “bioequivalents” when they have the same “bioavailability.”

This is why the statute often refers to “bioavailability and bioequivalence studies” together. See, e.g., id. § 355(j)(2)(D)(iii)(3)(B).

The strong textual connection between the concept of bioavailability and the subsection (j)

ANDA process further underscores that the best way to read the phrase “bioavailability studies” is to include only those studies primarily intended to study a proposed drug’s bioavailability, es- pecially in comparison with an already-approved drug’s bioavailability. Said differently, the FDCA as a whole seems to assume that “bioavailability studies” are those studies primarily con- cerned with proving that a proposed drug is a bioequivalent (or has a similar bioavailability) to an already-approved drug.

BREEZE, on the other hand, primarily studied the safety and tolerability of Tyvaso DPI and only secondarily studied its pharmacokinetics and bioavailability. Thus, labeling it a bioavail- ability study, and rendering it unable to support exclusivity, makes little sense. Studies such as BREEZE advance both the FDCA’s goals of promoting innovative drug-safety studies while re- ducing unnecessary human studies and FDA’s aim that “no unnecessary human research should be done.” Veloxis, 109 F. Supp. 3d at 107–08; 21 C.F.R. § 320.25(a)(1). Those goals would be severely undermined if studies like BREEZE could not support exclusivity.

For these reasons, the Court concludes that a bioavailability study is a study whose primary purpose is to study a drug’s bioavailability. And since the BREEZE study’s primary purpose—as shown by its primary endpoint—was “to evaluate the safety and tolerability” of treprostinil inha- lation powder, with only a secondary aim of “evaluat[ing] the systemic exposure and pharmaco- kinetics (PK) of treprostinil,” J.A. 774, the Court agrees with FDA and UTC that the BREEZE study was a clinical investigation other than a bioavailability study.7

Though the Court concludes that the phrase “bioavailability study” incorporates a pri- mary-purpose test, it need not decide how to treat a study with multiple primary purposes. Further, the Court sees no distinction between this standard and FDA’s such that a remand under the Chenery doctrine would be required. See SEC v. Chenery Corp., 332 U.S. 194, 196 (1947). That is because, even though FDA did use the word “solely” in finding that BREEZE was not “solely a bioavailability study,” in reaching that conclusion, FDA noted that BREEZE’s “primary endpoint

b. FDA Did Not Depart Without Explanation from a Prior Deter- mination That BREEZE Was a Bioavailability Study Liquidia next argues that, even if BREEZE is not a bioavailability study, FDA acted arbi- trarily and capriciously in departing from a prior decision that BREEZE was such a study and in failing to recognize or explain why it changed its position. Specifically, Liquidia points to a May 23, 2022, document, titled “Exclusivity Summary,” in which Liquidia claims FDA determined that BREEZE was a bioavailability study. See J.A. 453. In response, FDA asserts that this 2022 Ex- clusivity Summary was merely an internal document that did not reflect an official, final FDA position. ECF No. 55 at 20–22. Regardless, the Court sides with FDA.

First, even assuming the 2022 Exclusivity Summary did reflect FDA’s official position, it is not clear that FDA determined in it that BREEZE was a bioavailability study for exclusivity purposes. That document states, in relevant part, only that “[t]he basis for approval is the safety, tolerability, and bioavailability established in two studies. The safety and tolerability study pro- vided confirmatory efficacy information only.” J.A. 453. So although the Exclusivity Summary says that two studies “established” “safety, tolerability, and bioavailability,” it does not say that BREEZE established bioavailability. Indeed, the next sentence in the document implies that one study (which the parties all agree refers to BREEZE) was a “safety and tolerability study”—not a bioavailability study. Id. Second, even assuming FDA did determine that BREEZE was a bioavailability study in the 2022 Exclusivity Summary, it did not act arbitrarily and capriciously in departing from that conclusion. Far from it. First, contrary to Liquidia’s argument, FDA expressly recognized that its was safety and tolerability.” J.A. 432. FDA thus concluded that BREEZE was not a bioavailability study because its primary purpose was to study safety and tolerability—the same conclusion the Court reaches here.

position had changed. Indeed, in its 2024 Exclusivity Summary, FDA stated, “This Exclusivity Summary supersedes the original one dated May 23, 2022.” J.A. 498 (emphasis added). The Court can think of no clearer way to recognize a change in position than in identifying the prior decision and expressly noting that it was being “superseded.” Moreover, FDA adequately ex- plained in its nearly 40-page 2024 decision why it found that Tyvaso DPI was eligible for exclu- sivity. J.A. 413–52. And, as discussed above, it explained that BREEZE was not a bioavailability study because its “primary endpoint was safety and tolerability.” J.A. 432. Thus, even if FDA did decide in 2022 that BREEZE was a bioavailability study, it both recognized and explained why it took a different position in 2024.8 Any change in position on its part was not close to arbitrary and capricious.

2. BREEZE Was a New Clinical Investigation FDA also properly determined that BREEZE was a new clinical investigation. Again, the parties’ dispute is not over what the relevant regulations mean or whether they are consistent with the statutory framework. The regulations establish that a “new clinical investigation” is an investigation in humans the results of which have not been relied on by FDA to demonstrate substantial evidence of effectiveness of a previously approved drug product for any indication or of safety for a new patient population and do not du- plicate the results of another investigation that was relied on by the agency to

Liquidia also argues that FDA acted arbitrarily and capriciously in treating BREEZE dif- ferently than another study submitted with the Tyvaso DPI NDA—the TIP-PH-102 study, which FDA called “the pivotal relative bioavailability study.” See J.A. 429. But FDA never said in the discussion section of its memorandum whether TIP-PH-102 was in fact a clinical investigation other than a bioavailability study sufficient to support exclusivity. See J.A. 443–50. Because FDA found that BREEZE supported exclusivity eligibility, it did not need to decide that question and expressly reserved on it. J.A. 436 n.86. Furthermore, while BREEZE focused on safety and tol- erability, any “safety assessments” conducted under TIP-PH-102 were secondary to its “eval- uat[ion of] the systemic exposure and [pharmacokinetics] of treprostinil administered as treprosti- nil inhalation powder (Tyvaso DPI) and treprostinil inhalation solution (Tyvaso).” J.A. 429. That is different from BREEZE, so FDA did not fail to treat “similar products in the same way.” See Bracco Diagnostics, Inc. v. Shalala, 963 F. Supp. 20, 28 (D.D.C. 1997).

demonstrate the effectiveness or safety in a new patient population of a previously approved drug product. 21 C.F.R. § 314.108(a). Thus, to be “new,” a “clinical investigation” must meet two elements.

First, FDA cannot have previously relied on the results of that investigation when considering a prior drug product. And second, the investigation cannot merely “duplicate the results” of a prior investigation that FDA had previously relied upon in approving another drug product. The parties do not dispute that FDA did not rely on BREEZE before the Tyvaso DPI NDA, so only the second element is at issue.

a. BREEZE Did Not Duplicate Another Investigation Liquidia argues that BREEZE is not a new clinical investigation because “its results merely duplicated findings from prior studies on which FDA had already relied to approve certain drugs.”

ECF No. 41 at 34. Liquidia argues that, since TRIUMPH I and INCREASE—studies which FDA had previously used to approve Tyvaso9 for PAH and PH-ILD patients—proved that inhaled treprostinil was safe and effective, “BREEZE merely confirmed these results.” Id. Additionally, Liquidia contends that BREEZE was also merely “confirmatory of prior studies showing the safety and tolerability of dry powder formulations containing the excipient [fumaryl diketopiperazine, also known as] FDKP,” which is the powder excipient used in Tyvaso DPI. Id. at 34–35.

BREEZE did not duplicate the results of any prior investigation. The regulation states that a clinical investigation is not new if it “duplicate[s] the results of another investigation”—using the singular form of the word. 21 C.F.R. § 314.108(a) (emphasis added). But to advance this argument, Liquidia points to at least three different investigations to claim that BREEZE, because its results are similar in some respects to theirs, merely duplicates their results. Not so. The

As described above, Tyvaso, rather than Tyvaso DPI, is administered through an inhaled nebulous solution, not a dry powder. J.A. 424.

BREEZE study was materially different from them all because none involved dry-powder inhaled treprostinil.

The first two studies Liquidia cites—TRIUMPH I and INCREASE—did not involve dry- powder. Instead, they studied PAH and PH-ILD patients who were being treated with placebos or inhaled-solution Tyvaso. So those studies could not answer whether treprostinil was safe and tolerable in the dry-powder dosage form. And the other study that Liquidia cites, which demon- strated the safety and tolerability of the dry-powder excipient FDKP, did so in relation to a drug that uses a different active moiety. See J.A. 242 (noting that Afrezza, the drug at issue in the FDKP studies, is a form of insulin). Thus, no other single study showed that treprostinil was safe and tolerable in the dry-powder dosage form—BREEZE was the first. Because it did not “duplicate” the results of any prior investigation, BREEZE is a “new clinical investigation” under the regula- tions. 21 C.F.R. § 314.108(a).

b. FDA Did Not Act Arbitrarily and Capriciously in Determining That BREEZE Was a New Clinical Investigation Liquidia still argues that—even if BREEZE is a “new clinical investigation”—FDA acted arbitrarily and capriciously in so concluding because it failed to explain why or even acknowledge that it reached a different conclusion in the 2022 Exclusivity Summary. But, again, the 2022 Ex- clusivity Summary does not do the work Liquidia claims, even assuming it represented a final, official FDA position on Tyvaso DPI’s exclusivity.

First, as explained above, FDA did acknowledge that its 2024 decision departed from and “supersede[d]” the 2022 Exclusivity Summary. J.A. 498. And second, FDA fully explained why it determined that BREEZE was a new clinical investigation. In its memorandum on exclusivity, FDA’s Center for Drug Evaluation and Research Exclusivity Board clarified that “[t]he BREEZE study answered, for the first time, a unique clinical question as to whether treprostinil as an inhalation powder is safe and tolerable for chronic use.” J.A. 460. And in its letter to Liquidia, FDA reiterated that BREEZE “addressed a specific safety question, the tolerability of an active moiety (treprostinil) in a new inhalation powder dosage form.” J.A. 439. It emphasized that “[s]uch information was not available from any other source,” meaning BREEZE’s results did not duplicate any prior investigation’s results. Id. All this more than adequately explains why FDA departed from its prior determination that BREEZE “provided confirmatory efficacy information only.” J.A. 453. Because BREEZE provided data that answered a question that no single prior study had, FDA acted reasonably in departing from any contrary conclusion in the 2022 Exclusiv- ity Summary.

Undeterred, Liquidia argues that, even if FDA explained its departure, it still acted arbi- trarily since that decision conflicts with longstanding FDA policy. And, true, “an ‘[u]nexplained inconsistency’ in agency policy is ‘a reason for holding an interpretation to be an arbitrary and capricious change from agency practice.’” Encino Motorcars, LLC v. Navarro, 579 U.S. 211, 222 (2016) (quoting Nat’l Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 981 (2005)). But FDA’s decision to recognize exclusivity here is neither unexplained nor inconsistent with any preexisting policies.

According to Liquidia, FDA will not consider a clinical investigation to be “new” unless it provides “new safety and efficacy findings” that “permit broader use” of a previously approved drug. ECF No. 41 at 36 (citing 54 Fed. Reg. 28872, 28899 (July 10, 1989)). Liquidia claims that FDA had, by 2017, already concluded that preexisting data supported approving a dry-powder form of treprostinil (meaning BREEZE had no new safety and efficacy findings) and that, since BREEZE did not study an “expanded patient population” beyond those already taking Tyvaso, BREEZE could not permit a broader use of treprostinil than was already approved. ECF No. 41 at 36.

Liquidia’s arguments come up short. First, Liquidia is just plain wrong that FDA had de- termined by 2017 that it had all the information it needed to approve a dry-powder dosage form of treprostinil. Instead,

J.A. 689 (emphasis added) (sealed).

Id. J.A. 690 (sealed).

Id.

J.A. 691. Thus, it blinks reality to say that it already had all the safety and tolerability findings it needed to allow a dry-powder form of treprostinil.

In addition, BREEZE did in fact provide new safety data permitting broader use of trepros- tinil. As laid out above, BREEZE, for the first time, provided the necessary safety and tolerability data that permitted treprostinil to be approved in a dry-powder dosage form. And, as FDA ex- plains, an active moiety’s use is broadened when it is approved in a new dosage form. J.A. 436 n.87. When an NDA “represent[s] an additional treatment option for patients,” such as by taking a preexisting active moiety and presenting it in a more portable or convenient form, that new dos- age form “permit[s] broader use of the drug” by opening it up to those who might not have been able to use the prior, less convenient dosage form. J.A. 437. This explains why, even in the policy

Liquidia relies on, FDA recognized that “those changes in an approved drug product that affect its . . . dosage form . . . would be granted exclusivity.” 54 Fed. Reg. at 28899. Because BREEZE was a study that led to the approval of treprostinil in the novel dry-powder dosage form, it permit- ted the broader use of treprostinil and falls within the class of studies that FDA, under the policy Liquidia identifies, routinely regards as qualifying for exclusivity. See id. at 28898–99.

Finally, that BREEZE did not study an “expanded patient population” is beside the point.

Indeed, Liquidia does not explain where this purported requirement comes from. At best, Liquidia pulls it from AstraZeneca Pharmaceuticals LP v. FDA, where the court noted that the FDCA grants exclusivity when an NDA seeks approval for “new indications or uses of the already approved pioneer drug” to treat new patient populations. 872 F. Supp. 2d at 64. Presumably, applicants seeking to expand the use of a drug to new indications in new patient populations would need to study the drug’s safety and efficacy in those populations. But AstraZeneca does not say that only new indications or uses designed to treat new patient populations support exclusivity. See id. (noting that the FDCA provides for exclusivity in various “circumstances,” and describing exclu- sivity for new indications or uses as merely “one such circumstance”). So it does not hold that only studies involving “new patient populations” can be considered “new clinical investigations.”

And, again, long-standing FDA policy recognizes that “changes in an approved drug product that affect its . . . dosage form . . . would be granted exclusivity.” 54 Fed. Reg. at 28899. Nothing suggests that policy does not extend to situations when the new dosage form only permits broader use of a drug within the population for which the drug was already approved.

For these reasons, BREEZE presented new findings that allowed FDA to broaden the use of treprostinil by approving a dry-powder form of the drug. This is consistent with the longstand- ing policy Liquidia identifies. So FDA’s decision that BREEZE was a new clinical investigation for exclusivity purposes was not arbitrary and capricious.

3. BREEZE Was Essential to Tyvaso DPI’s Approval In its last argument about Tyvaso DPI’s eligibility for exclusivity, Liquidia argues that the BREEZE study was not “essential to the approval of” Tyvaso DPI’s NDA. ECF No. 41 at 37 (quoting 21 U.S.C. § 355(c)(3)(E)(iii)). Again, despite framing FDA’s decision as contravening the text of the FDCA, Liquidia does not dispute that FDA’s regulations accurately reflect the Act’s standards. Liquidia agrees with FDA that an investigation is “essential to approval” when “there are no other data available that could support approval of the NDA.” Id. (quoting 21 C.F.R. § 314.108(a)). And Liquidia argues that, since BREEZE only confirmed safety and efficacy data produced by the other studies discussed above, FDA could have approved Tyvaso DPI without BREEZE.

Liquidia’s argument faces a high hurdle from the start: FDA concluded that “[w]ithout the BREEZE study, [it] would not have had sufficient information regarding the safety and tolerability of multiple doses of the new dosage form of treprostinil to support approval.” J.A. 435. In other words, FDA represents that, without BREEZE, it would not have approved Tyvaso DPI’s NDA.

And importantly, the FDCA delegates substantial discretion to FDA to determine whether the in- vestigations and tests submitted as part of an NDA “do not show that [a] drug is safe for use.” 21 U.S.C. § 355(d). That is, the FDCA primarily delegates to FDA the authority to say whether an NDA has sufficient safety data or not. Thus, Liquidia faces a difficult task to show that FDA did not reasonably conclude that it would not have approved Tyvaso DPI but for BREEZE.

Liquidia’s first tack is familiar: it argues that FDA could have approved Tyvaso DPI rely- ing solely on information FDA already had. In particular, Liquidia argues that FDA should have been able to conclude that a dry-powder form of treprostinil was safe, tolerable, and effective based on TRIUMPH I, INCREASE, and the Afrezza NDA. But, as discussed above, none of those studies looked at the interaction of the active moiety treprostinil with the dry-powder dosage form.

And, as FDA explained, it was specifically concerned with “the tolerability of the inhalation pow- der dosage form of treprostinil.” J.A. 433 (footnote omitted). FDA was not concerned with the safety or tolerability of inhaled treprostinil or dry-powder excipients independently, but with the mixture of the two together, as “the inhalation powder dosage form could present new or worse tolerability issues than those observed with Tyvaso and other approved treprostinil products.” Id. And since neither TRIUMPH I, INCREASE, nor the studies in the Afrezza NDA could answer whether the treprostinil moiety with a dry-powder dosage form would be less tolerable than other dosage forms of treprostinil, BREEZE was essential to answer that question.10 Liquidia responds by asserting that FDA “manufactured” these safety concerns in its 2024 exclusivity decision. ECF No. 41 at 39. But the strong presumption of administrative propriety and regularity works against it. United Steelworkers of Am., AFL-CIO-CLC v. Marshall, 647 F.2d 1189, 1217 (D.C. Cir. 1980). And in any event, FDA’s “manufactured” concern is supported by the record.

J.A. 691 (partially sealed).

At the time it made this recommendation, FDA already had access to all the information present in the TRIUMPH I and the Afrezza studies, so it is reasonable to conclude that FDA’s concern addressed something that those studies would not have answered—like the impact a dry-powder dosage form would have on treprostinil’s tolerability.

FDA also reasonably found that the two other, shorter investigations submitted as part of the Tyvaso DPI NDA could not support a finding of tolerability for chronic use as they involved only “single-dose experience[s].” J.A. 433.

J.A. 433. J.A.

827 (sealed).

Liquidia’s reference to an FDA letter saying that “little beyond demonstrating bioavaila- bility was necessary for Tyvaso DPI” is not to the contrary. See J.A. 913. Indeed, it supports FDA’s position. By saying that “little beyond” bioavailability data was required for Tyvaso DPI’s approval, FDA recognized that there was still something beyond bioavailability that it needed be- fore it could approve Tyvaso DPI. And the sentence immediately following that quotation clarifies what that “little” bit of information was: the “[s]afety of Tyvaso DPI . . . supported by” the BREEZE study. Id. Moreover, that BREEZE did not study PH-ILD patients or that its main phase lasted only three weeks does nothing to alter this conclusion. Liquidia argues that, since BREEZE only stud- ied PAH patients and was only “designed to study short term safety and tolerability,” it could not have been essential to Tyvaso DPI’s approval because that approval went well beyond the specific parameters of the BREEZE study. ECF No. 41 at 40. But Liquidia gives no reason why FDA could not take the results from the BREEZE study and extrapolate those results onto a broader patient population—which is what FDA did. J.A. 433–35, 881.

This argument reflects Liquidia’s confusion over the role BREEZE played in FDA’s ap- proval process. As FDA has explained, it would not have approved Tyvaso DPI without some direct evidence that the active moiety treprostinil was tolerable when administered in the dry- powder dosage form. The Court sees no reason why FDA’s worry would be more pronounced in the PAH patient population than in the PH-ILD patient population, and Liquidia offers none. For such a cross-cutting concern, FDA’s decision to extrapolate BREEZE’s tolerability results to all the relevant patients—PAH and PH-ILD—was reasonable.11 Liquidia also does not explain why FDA could not extrapolate from a 3-week, multi-dose study like BREEZE to conclude that any long-term use of dry-powder treprostinil would be safe and effective. Indeed, BREEZE even in- cluded an extension phase to “evaluate[] the long-term safety and tolerability of” of dry-powder treprostinil, making that extrapolation even more reasonable. J.A. 760.

For these reasons, FDA reasonably concluded that BREEZE was essential to its approval of the Tyvaso DPI NDA. That investigation answered a question that no other investigation or study had: whether treprostinil was safe and tolerable when mixed with a dry-powder excipient.

FDA reasonably concluded that this question had to be answered before it could approve Tyvaso DPI for any indication or patient population. Thus, the Court cannot say that FDA’s conclusion that it needed BREEZE to do so was arbitrary and capricious.

* * * For all these reasons, the Court finds no error with FDA’s conclusion that Tyvaso DPI was eligible for exclusivity under 21 U.S.C. § 355(c)(3)(E)(iii).

B. Yutrepia Falls Within the Scope of Tyvaso DPI’s Exclusivity Still, Liquidia argues that, even assuming Tyvaso DPI is eligible for exclusivity, FDA was wrong to conclude that Yutrepia fell within Tyvaso DPI’s zone of exclusivity. As explained below, the Court disagrees.

Liquidia makes much of a letter FDA sent to UTC asking it to “[p]rovide justification for the extrapolation of pulmonary safety (e.g. bronchospasm risk) for Tyvaso DPI from the PAH . . . population to the PH-ILD . . . population.” J.A. 916. But to say that FDA, at one point, had concerns about making such an extrapolation is not to say that extrapolation would be unreasona- ble. Furthermore, after considering that FDA eventually concluded that there was no “notable difference in respiratory adverse events (AEs) between” Tyvaso DPI and Tyvaso and that Tyvaso itself was safe and tolerable for PH-ILD patients, J.A. 435, 878, it was reasonable for it to conclude that Tyvaso DPI would also be safe and tolerable for PH-ILD patients.

Recall that the FDCA limits the scope of an NDA’s exclusivity to that NDA’s “conditions of approval.” 21 U.S.C. § 355(c)(3)(E)(iii). Though neither the statute nor the regulations define that phrase, courts in this district have recognized that “[t]he FDCA sets up a ‘logical relationship between the change in the product for which the new clinical investigations were essential to ap- proval of the [NDA] and the scope of any resulting three-year exclusivity.” Veloxis, 109 F. Supp. 3d at 120–21 (quoting AstraZeneca, 872 F. Supp. 2d at 80). That is, “three-year exclusivity can only be as broad as the conditions of approval that were based upon the new clinical investigations identified in the . . . NDA.” Id. at 121. For this reason, Liquidia argues that exclusivity under § 355(c)(3)(E)(iii) “can be no broader than the innovations presented to the FDA in the new clinical investigations.” ECF No. 41 at 43 (quoting Veloxis, 109 F. Supp. 3d at 121 n.16).

On this, FDA and UTC agree with Liquidia. ECF No. 55 at 33; ECF No. 56 at 43. But the parties disagree about whether FDA properly applied that standard in finding exclusivity for “the inhalation powder dosage form of the active moiety treprostinil for chronic use.” J.A. 450.

Liquidia argues that this grant of exclusivity “violates the FDCA because there is no logical con- nection between this sweeping condition of approval and the change in the product that BREEZE studied.” ECF No. 41 at 43. Instead, in Liquidia’s view, Tyvaso DPI’s exclusivity should be limited to the circumstances of the BREEZE study: specifically, it should be limited to short-term PAH patients switching from stable doses of Tyvaso, and it should be limited to treprostinil inha- lation powders that use Tyvaso DPI’s excipient, FDKP.12 Once again, Liquidia’s argument seems to stem from a misunderstanding of BREEZE’s

Before advancing this argument, Liquidia cursorily contends that BREEZE presented no innovations because it did not study any new indications or patient populations and because FDA could have concluded that Tyvaso DPI was safe, tolerable, and effective based on TRIUMPH I, INCREASE, and the Afrezza NDA studies. The Court has already explained why it rejects those arguments.

role in FDA’s approval process. As FDA has maintained, Tyvaso DPI’s “innovation” was its dry- powder dosage form. E.g., J.A. 442. FDA has long recognized that changes to an active moiety’s dosage form generally warrant exclusivity. 54 Fed. Reg. at 28899. And BREEZE, for the first time, presented that innovation to FDA and proved that it was tolerable—or at least, no less toler- able than any other treprostinil product. J.A. 442. Thus, contrary to Liquidia’s assertion, the relationship between Tyvaso DPI’s exclusivity and the innovation BREEZE studied was logical: BREEZE studied the safety and tolerability of dry-powder treprostinil, and FDA recognized a commensurate scope of exclusivity. In other words, the scope of exclusivity was “no broader than the innovations presented to the FDA in the new clinical investigations.” ECF No. 41 at 43 (em- phasis omitted) (quoting Veloxis, 109 F. Supp. 3d at 121 n.16).

Still, Liquidia argues that FDA arbitrarily and capriciously departed from policies reflected in two prior exclusivity decisions when it failed to limit exclusivity to the specific patient popula- tion BREEZE studied. Not so.

In the first decision, an applicant sought approval of its NDA for Envarsus XR, a drug designed to prevent organ rejection in de novo kidney-transplant patients (patients with newly transplanted kidneys) and conversion patients (patients who had already received a kidney trans- plant but who replaced one of the immunosuppressive drugs in their treatments with another drug).

J.A. 1023, 1033. But FDA concluded that exclusivity for a prior NDA—the Astagraf XL NDA— blocked Envarsus XR’s approval for the de novo patient population. J.A. 1058. But it also con- cluded that the Astagraf XL NDA did not block approval for conversion patients. Id. Liquidia argues that, since the studies essential to Astagraf XL’s approval in that case involved only de novo patients, FDA has a policy of limiting exclusivity only to those patient populations evaluated in the essential studies.

The problem is that FDA did not limit Astagraf XL’s exclusivity to the de novo patient population for the reason Liquidia suggests—i.e., because that was the population involved in the NDA’s studies. Instead, FDA declined to find exclusivity for conversion patients because Astagraf XL was not approved for “conversion patients and thus its exclusivity c[ould] not extend to block approval for this population.” J.A. 1058. Indeed, the Astagraf XL NDA did not even seek approval for conversion patients, and in any event, FDA also concluded that its NDA failed to show that Astagraf XL was safe, tolerable, or effective for that patient population. J.A. 1058–59. As FDA explained, because of the important medical differences between de novo and conversion patient populations, “[s]eparate studies are needed to support approval in de novo patients and conversion patients.” J.A. 1023. In other words, the results of a study designed to evaluate the de novo population would not automatically extend to the conversion population. And since the Astagraf XL NDA’s investigations only studied de novo patients, FDA did not approve it for conversion patients.

As such, FDA’s decision on the Envarsus XL NDA does not reflect a policy of limiting exclusivity only to the patient populations evaluated in an essential investigation. Instead, it re- flects the commonsense conclusion that a drug cannot have exclusivity over a patient population or indication for which that drug itself is not approved.

Second, Liquidia points to FDA’s exclusivity decision for the drug MorphaBond. That drug, an extended-release morphine sulfate tablet, was approved with “labeling describing intrana- sal and intravenous [abuse-deterrent] properties.” J.A. 1073, 1083. While MorphaBond’s intra- venous abuse-deterrent properties were supported by non-clinical investigations incapable of sup- porting exclusivity, J.A. 1084 n.62, its intranasal abuse-deterrent properties were supported by a new clinical investigation essential to its approval as an intranasal-abuse-deterrent morphine drug,

J.A. 1083–84. Thus, FDA concluded that MorphaBond’s exclusivity was limited only to intranasal abuse deterrence. J.A. 1085.

Id. (sealed). What’s more, FDA noted that MorphaBond was not labeled for general abuse deterrence but only for intranasal and intravenous abuse deterrence. J.A. 1085–86.

Liquidia argues that, since the MorphaBond investigations evaluated only one form of abuse deterrence, this decision reflects an FDA policy that exclusivity can go no further than “what the new clinical investigation was actually intended to evaluate.” ECF No. 41 at 46. And, Liquidia posits, since “BREEZE was never intended to study dry powder inhalation form of treprostinil for chronic use as a general matter, but rather only for use in PAH patients,” granting exclusivity beyond the narrow class of patients studied in BREEZE would “conflict[]” with the MorphaBond precedent. Id. at 47. Liquidia misunderstands FDA’s reasoning in the MorphaBond decision. Despite some admittedly impre- cise language in the decision, what mattered in MorphaBond was that the relied-upon investigation only provided data that would have supported approval for an intranasal abuse deterrence. That is why FDA reemphasized that intranasal and intravenous abuse were the only “routes of abuse that the product ha[d] been demonstrated to deter.” J.A. 1086. And it also explains why FDA noted that ” J.A. 1084 (sealed). Thus, FDA concluded that “the scope of Morphabond’s [sic] exclusivity is limited to the condition of approval supported by [the study]: labeling describing the expected reduction of abuse . . . by the intranasal route.” Id.

, FDA’s decision to recognize the scope of exclusivity it did does not conflict with either the Astagraf XL decision or the Mor- phaBond decision. Unlike with Astagraf XL, Tyvaso DPI was approved to treat both PAH and PH-ILD patients. The same is true with MorphaBond, since MorphaBond was not approved for a general abuse-deterrence label. What’s more, while FDA explained that it refused to extrapolate Astagraf XL’s exclusivity to conversion patients because of the material medical differences be- tween those two patient populations, Liquidia has given the Court no reason to doubt the propriety of the FDA’s extrapolation here. As already explained, the concern FDA had with the tolerability of mixing treprostinil with a dry-powder excipient would logically apply equally to each patient population, regardless of any other differences in those populations. And FDA further concluded that Tyvaso and Tyvaso DPI had nearly identical risk profiles. J.A. 435. Thus, it was reasonable for FDA to conclude that Tyvaso DPI would be safe and tolerable for Tyvaso’s approved popula- tions.

Indeed, FDA noted in its letter to Liquidia that “a particular clinical investigation may be more limited in scope or more specific than the conclusions (and thus the scope of exclusivity) that can be drawn from it.” J.A. 423. FDA continued, “[t]he scope of a product’s innovation similarly might not be defined by specific characteristics of its clinical studies.” Id. And it then concluded that the results of the BREEZE study—that dry-powder treprostinil was not any less tolerable than other forms of the moiety—could be extrapolated from the small, short-term patient population studied in BREEZE to all chronic PAH and PH-ILD patients. J.A. 447–48. Nothing about the particular patient population studied would have limited the results purely to that population. Id. For these reasons, FDA did not act arbitrarily, capriciously, or contrary to law in failing to limit Tyvaso DPI’s exclusivity only to PAH patients switching from regular Tyvaso.

Finally, Liquidia argues that FDA should have limited Tyvaso DPI’s exclusivity to its spe- cific formulation, which includes FDKP. This argument fails for similar reasons. Liquidia again contends that FDA’s decision departs from FDA precedent because, in a prior decision about the drug Adzenys ER, FDA limited exclusivity to a drug’s specific formulation. Even putting aside that, in the MorphaBond decision, FDA refused to limit exclusivity to the drug’s specific formu- lation (thereby demonstrating that FDA has no general policy limiting exclusivity to a drug’s for- mulation), J.A. 1086, FDA’s Adzenys ER decision says nothing about this case. There, FDA decided that neither Dyanavel XR nor Mydayis, two prior approved NDAs, blocked approval for Adzenys ER, an amphetamine drug meant to treat ADHD. J.A. 1093. Instead, FDA found that those two approved drugs’ scopes of exclusivity were limited to their specific formulations. Id. But it did so because those formulations materially altered the drugs by giving each a different and unique “drug release profile.” Id. And the innovations presented in the studies those two NDAs relied on were limited to each’s specific drug release profile. J.A. 1100. Thus, because Adzenys ER’s formulation resulted in a drug release profile different from those of both Dyanavel XR and Mydayis, FDA concluded that it fell outside the scope of the innovation presented by either prior NDA. Id. Accordingly, FDA’s Adzenys ER decision did not establish a general FDA policy to limit exclusivity to an approved NDA’s specific formulation. Instead, it reflects FDA’s view that, for exclusivity purposes, a drug’s chemical composition matters when that composition affects how the innovative portions of the drug operate. And Liquidia does not contend that Tyvaso DPI’s formulation impacts how it operates as a dry-powder inhaler. That is, Liquidia does not argue that its own excipient causes Yutrepia to perform, in relation to the innovation presented, any differ- ently than Tyvaso DPI. At best, Liquidia argues that Yutrepia does exactly what Tyvaso DPI does, only more safely (as FDKP allegedly has certain safety risks). Of course, FDA “disagree[s] with th[at] premise.” J.A. 445, 49. But even assuming Yutrepia is safer than Tyvaso DPI, that is beside the point. Under Liquidia’s own standard, exclusivity extends to “the innovations presented to the FDA in the new clinical investigations.” ECF No. 41 at 43 (emphasis omitted) (quoting Veloxis, 109 F. Supp. 3d at 121 n.16). As already explained, the “innovation” presented to FDA in BREEZE was the dry-powder dosage form in general—not just Tyvaso DPI’s formulation. So unlike in the Adzenys ER case—where the innovations presented were the drug release profiles tied to each drug’s specific formulation—the innovation studied and presented in BREEZE was not limited to Tyvaso DPI’s chemical formulation.

For all these reasons, Liquidia has failed to show that FDA had a policy or practice of limiting an NDA’s scope of exclusivity either to the particular patient population in a study or the drug’s specific chemical formulation. So it has failed to show that FDA arbitrarily and capriciously departed from any such practice in concluding that Yutrepia falls within the scope of Tyvaso DPI’s zone of exclusivity.

IV. Conclusion For all the above reasons, the Court will grant FDA’s and UTC’s Cross-Motions for Sum- mary Judgment and deny Liquidia’s Motion for Summary Judgment. A separate order will issue.

The Court intends to promptly resolve the pending motions to dismiss UTC’s cross claims.

/s/ Timothy J. Kelly TIMOTHY J. KELLY United States District Judge Date: February 27, 2025