Tyvaso® and Tyvaso DPI® (treprostinil)
Case Name: United Therapeutics Corp. v. Liquidia Techs., Inc., No. 23-975-RGA, 2026 WL 2935925 (D. Del. Sept. 30, 2026) (Andrews, J.)
Drug Product and Patent(s)-in-Suit: Tyvaso® and Tyvaso DPI® (treprostinil); U.S. Patent No. 11,826,327 (“the ’327 patent”)
Nature of the Case and Issue(s) Presented: UTC) markets Tyvaso and Tyvaso DPI, inhaled formulations of treprostinil. The ’327 patent claims methods of treating pulmonary hypertension associated with interstitial lung disease (“PH-ILD”) with inhaled treprostinil. Liquidia sought FDA approval to market Yutrepia®, a competing dry-powder formulation of treprostinil, and UTC sued alleging infringement of the ’327 patent. Liquidia stipulated to direct and indirect infringement of independent claim 1 and dependent claim 14, leaving validity as the principal dispute concerning those claims.
Claim 1 recites a method of improving exercise capacity in a PH-ILD patient by administering via inhalation an effective amount of at least 15 micrograms of treprostinil up to a maximum tolerated dose, with at least 6 micrograms per breath. Claim 14 further requires administration through a dry-powder inhaler. Claims 5, 6, 9, and 17 add specified clinical outcomes to claim 1, including reductions in NT-proBNP, reductions in ILD exacerbations, improvements in forced vital capacity, and an increase in six-minute walk distance. Liquidia challenged the claims on several grounds: the clinical-outcome limitations of claims 5, 6, 9, and 17 were merely intended results and therefore lacked patentable weight; the claims were obvious based principally on earlier studies of inhaled treprostinil in PH-ILD; commercial off-label use of Tyvaso before the critical date triggered the on-sale bar; and that a 2017 protocol for the INCREASE clinical trial inherently anticipated the claimed treatment. After a bench trial, the court agreed with Liquidia only as to patentable weight and otherwise rejected its invalidity defenses.
Why UTC Prevailed: Regarding the patentable weight issue, UTC’s own infringement theory and expert testimony established that prescribing Yutrepia in accordance with claim 1 would “automatically” infringe the dependent claims, without requiring any change in how the claimed treatment was performed. Applying the Federal Circuit’s 2025 decision in Bayer v. Mylan, the court concluded that the claimed clinical outcomes did not have a functional relationship with or transform the administration method of claim 1. Rather, they reflected clinical benefits observed in the INCREASE trial.
Regarding obviousness, Liquidia failed to prove that the prior art showed a reasonable expectation of successfully treating PH-ILD with inhaled treprostinil. Before April 2020, multiple clinical trials involving pulmonary arterial hypertension drugs had produced negative or inconclusive results in PH-ILD, including trials presenting significant safety concerns. Although a small number of highly experienced physicians had used Tyvaso off-label in PH-ILD patients, Judge Andrews found that those physicians possessed skill substantially beyond that of a POSA. Therefore, the art at the time would have discouraged, not encouraged, a POSA from treating PH-ILD with inhaled treprostinil. Liquidia relied principally on the Faria-Urbina retrospective study, which reported results from 22 patients with Group 3 pulmonary hypertension treated with inhaled treprostinil. The court found the study insufficient to establish a reasonable expectation of success. Among other problems, its dataset was small and selected, the data did not permit reliable conclusions specifically concerning PH-ILD patients, and the results included potentially concerning reductions in forced vital capacity and oxygen saturation. The study itself recommended further prospective investigation, and the court credited testimony characterizing its results as merely “hypothesis-generating.”
Notwithstanding its findings that physicians prescribed Tyvaso off-label to PH-ILD patients using dosing that later appeared in claim 1before the critical date, and that those patients experienced improved exercise capacity, the court rejected Liquidia’s on-sale bar defense. The court distinguished a commercial sale of Tyvaso from a commercial sale of the claimed method. UTC had not itself offered to perform the claimed treatment method for consideration, and Tyvaso did not substantially embody the method because it had other uses, most notably its approved use for PAH. The fact that physicians independently purchased and prescribed Tyvaso for PH-ILD therefore did not establish that the claimed method itself had been “on sale.”
Finally, the court rejected Liquidia’s inherent-anticipation argument based on UTC’s publicly available 2017 INCREASE protocol. The 2017 protocol contemplated a minimum dose of 6 micrograms, whereas the actual INCREASE trial began at 18 micrograms and claim 1 required at least 15 micrograms. Because the later clinical trial did not follow the dosing procedure disclosed in the prior-art protocol, its successful results could not establish that those results were necessarily present in, or the natural result of, practicing the earlier protocol.
