Original Article
`
`Inhaled Treprostinil in Pulmonary Hypertension
`Due to Interstitial Lung Disease
`Aaron Waxman, M.D., Ph.D., Ricardo Restrepo-Jaramillo, M.D.,
`Thenappan Thenappan, M.D., Ashwin Ravichandran, M.D., Peter Engel, M.D.,
`Abubakr Bajwa, M.D., Roblee Allen, M.D., Jeremy Feldman, M.D.,
`Rahul Argula, M.D., Peter Smith, Pharm.D., Kristan Rollins, Pharm.D.,
`Chunqin Deng, M.D., Ph.D., Leigh Peterson, Ph.D., Heidi Bell, M.D.,
`Victor Tapson, M.D., and Steven D. Nathan, M.D.
`
`A BS TR AC T
`
`BACKGROUND
`No therapies are currently approved for the treatment of pulmonary hypertension in
`patients with interstitial lung disease. The safety and efficacy of inhaled treprostinil
`for patients with this condition are unclear.
`
`METHODS
`We enrolled patients with interstitial lung disease and pulmonary hypertension
`(documented by right heart catheterization) in a multicenter, randomized, double-
`blind, placebo-controlled, 16-week trial. Patients were assigned in a 1:1 ratio to
`receive inhaled treprostinil, administered by means of an ultrasonic, pulsed-delivery
`nebulizer in up to 12 breaths (total, 72 μg) four times daily, or placebo. The pri-
`mary efficacy end point was the difference between the two groups in the change
`in peak 6-minute walk distance from baseline to week 16. Secondary end points
`included the change in N-terminal pro–B-type natriuretic peptide (NT-proBNP)
`level at week 16 and the time to clinical worsening.
`
`RESULTS
`A total of 326 patients underwent randomization, with 163 assigned to inhaled
`treprostinil and 163 to placebo. Baseline characteristics were similar in the two
`groups. At week 16, the least-squares mean difference between the treprostinil
`group and the placebo group in the change from baseline in the 6-minute walk
`distance was 31.12 m (95% confidence interval [CI], 16.85 to 45.39; P<0.001).
`There was a reduction of 15% in NT-proBNP levels from baseline with inhaled
`treprostinil as compared with an increase of 46% with placebo (treatment ratio,
`0.58; 95% CI, 0.47 to 0.72; P<0.001). Clinical worsening occurred in 37 patients
`(22.7%) in the treprostinil group as compared with 54 patients (33.1%) in the
`placebo group (hazard ratio, 0.61; 95% CI, 0.40 to 0.92; P = 0.04 by the log-rank
`test). The most frequently reported adverse events were cough, headache, dyspnea,
`dizziness, nausea, fatigue, and diarrhea.
`
`CONCLUSIONS
`In patients with pulmonary hypertension due to interstitial lung disease, inhaled
`treprostinil improved exercise capacity from baseline, assessed with the use of a
`6-minute walk test, as compared with placebo. (Funded by United Therapeutics;
`INCREASE ClinicalTrials.gov number, NCT02630316.)
`
`From Brigham and Women’s Hospital,
`Boston (A.W.); the University of South
`Florida, Tampa (R.R.-J.), and St. Vin-
`cent’s Lung, Sleep, and Critical Care Spe-
`cialists, Jacksonville (A.B.) — both in FL;
`the University of Minnesota, Minneapo-
`lis (T.T.); St. Vincent Medical Group, In-
`dianapolis (A.R.); the Carl and Edyth
`Lindner Research Center at the Christ
`Hospital, Cincinnati (P.E.); University of
`California Davis Medical Center, Sacra-
`mento (R. Allen), and Cedars–Sinai, Los
`Angeles (V.T.); Arizona Pulmonary Spe-
`cialists, Phoenix (J.F.); the Medical Uni-
`versity of South Carolina, Charleston (R.
`Argula); United Therapeutics Corporation,
`Silver Spring, MD (P.S., K.R., C.D., L.P.,
`H.B.); and Inova Fairfax Hospital, Falls
`Church, VA (S.D.N.). Address reprint re-
`quests to Dr. Nathan at the Advanced
`Lung Disease and Lung Transplant Pro-
`gram, Inova Heart and Vascular Institute,
`3300 Gallows Rd., Falls Church, VA
`22042, or at steven . nathan@ inova . org.
`
`This article was published on January 13,
`2021, at NEJM.org.
`
`N Engl J Med 2021;384:325-34.
`DOI: 10.1056/NEJMoa2008470
`Copyright © 2021 Massachusetts Medical Society.
`
`n engl j med 384;4 nejm.org
`
`January 28, 2021
`
`325
`
`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on November 9, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2021 Massachusetts Medical Society. All rights reserved.
`
`IPR2021-00406
`United Therapeutics EX2060
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`Precapillary pulmonary hyperten-
`
`sion is defined as an elevation in mean
`pulmonary arterial pressure and pulmo-
`nary vascular resistance.1 In the World Health
`Organization (WHO) classification of pulmonary
`hypertension, precapillary pulmonary hyperten-
`sion due to lung disease is classified as group 3.
`The most common lung diseases associated with
`group 3 pulmonary hypertension are chronic ob-
`structive pulmonary disease and interstitial lung
`disease.
`Pulmonary hypertension has been reported in
`up to 86% of patients with interstitial lung dis-
`ease and is associated with reduced exercise ca-
`pacity, greater need for supplemental oxygen, de-
`creased quality of life, and earlier death.2-4 Despite
`the global prevalence and poor clinical course of
`pulmonary hypertension due to interstitial lung
`disease, there are currently no approved therapies
`for these patients. Although data are limited,
`therapies approved for group 1 pulmonary hyper-
`tension (pulmonary arterial hypertension) have
`been used to treat group 3 pulmonary hyperten-
`sion.5 Previous studies of vasodilator therapies
`have shown conflicting results. The largest trial
`to date evaluated the soluble guanylate cyclase
`stimulator riociguat in a patient population with
`group 3 pulmonary hypertension and was stopped
`early owing to serious harm.6
`Treprostinil is a stable analogue of prostacyclin,
`which promotes direct vasodilation of pulmonary
`and systemic arterial vascular beds and inhibits
`platelet aggregation.7 An inhaled formulation of
`treprostinil was previously shown to improve ex-
`ercise capacity after 12 weeks of therapy in pa-
`tients with group 1 pulmonary hypertension.8
`Data from previously completed pilot studies sug-
`gest that inhaled treprostinil can improve hemo-
`dynamics and functional capacity in patients with
`group 3 pulmonary hypertension.9-12 Therefore,
`the objective of the INCREASE trial was to evalu-
`ate the safety and efficacy of inhaled treprostinil
`in patients with pulmonary hypertension due to
`interstitial lung disease.
`
`Me thods
`
`Trial Design and Oversight
`INCREASE was a multicenter, randomized, double-
`blind, placebo-controlled trial. The steering com-
`mittee (the first author and last two authors), in
`
`collaboration with the trial sponsor (United Ther-
`apeutics), designed the trial and oversaw its con-
`duct. The trial protocol (available with the full
`text of this article at NEJM.org) was approved by
`the institutional review board at each participat-
`ing site. The trial was monitored by an indepen-
`dent data and safety monitoring committee and
`was conducted in accordance with Good Clinical
`Practice guidelines. A full list of trial personnel,
`including the investigators and trial committees,
`is provided in Section S1 in the Supplementary
`Appendix, available at NEJM.org.
`The collection, management, and analysis of
`the data were performed by the sponsor according
`to a prespecified statistical analysis plan (provid-
`ed in the protocol). An independent academic
`statistician reviewed the statistical analysis plan
`and confirmed the primary efficacy analyses.
`Authors had independent access to the data and
`authority to conduct and confirm statistical anal-
`yses. All manuscript drafts were written by the
`steering committee and authors affiliated with
`the sponsor and were reviewed and approved by
`all the authors. The authors assume responsibility
`for the accuracy and completeness of the data, as
`well as for the fidelity of the trial to the protocol.
`
`Trial Population
`The trial population consisted of patients 18 years
`of age or older in whom interstitial lung disease
`was diagnosed on the basis of evidence of diffuse
`parenchymal lung disease on computed tomogra-
`phy of the chest (not centrally adjudicated) per-
`formed within 6 months before randomization.
`Confirmation of group 3 pulmonary hypertension
`by right heart catheterization within 1 year be-
`fore randomization was required. Group 3 pul-
`monary hypertension was defined by pulmonary
`vascular resistance of more than 3 Wood units,
`pulmonary capillary wedge pressure of 15 mm
`Hg or lower, and mean pulmonary arterial pres-
`sure of 25 mm Hg or higher. Patients with group
`3 pulmonary hypertension due to connective tis-
`sue disease were also required to have a baseline
`forced vital capacity of less than 70%. Eligible pa-
`tients also had to walk at least 100 m during a
`6-minute walk test. Patients receiving drug treat-
`ment (i.e., pirfenidone or nintedanib) for their
`underlying lung disease were required to have
`been receiving a stable dose for at least 30 days
`before undergoing randomization. Patients re-
`
`326
`
`n engl j med 384;4 nejm.org
`
`January 28, 2021
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on November 9, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2021 Massachusetts Medical Society. All rights reserved.
`
`

`

`Inhaled Treprostinil in Pulmonary Hypertension
`
`ceiving approved therapy for pulmonary arterial
`hypertension within 60 days before randomiza-
`tion were not eligible for enrollment. A complete
`list of trial enrollment criteria is provided in Sec-
`tion S2. Written informed consent was obtained
`from all the patients.
`
`weeks 8 and 16 (or at early discontinuation) after
`the patients recovered from the 6-minute walk
`test. The St. George’s Respiratory Questionnaire
`(SGRQ), a quality-of-life measure, was completed
`at baseline and week 16 or at the time of early
`discontinuation.
`
`Trial Procedures
`Within 30 days after screening, eligible patients
`were randomly assigned in a 1:1 ratio to receive
`inhaled treprostinil (Tyvaso, United Therapeutics)
`or placebo in a double-blind manner. Random-
`ization, based on permuted blocks, was stratified
`by baseline 6-minute walk distance (≤350 m vs.
`>350 m) and was implemented through an inter-
`active Web-response system.
`Inhaled treprostinil (0.6 mg per milliliter) was
`administered by means of an ultrasonic, pulsed-
`delivery nebulizer at 6 μg per breath. Placebo
`was administered similarly as a visually identical
`solution. The first dose of trial drug (3 breaths)
`was administered in the clinic, followed by at
`least a 1-hour observation period. The dose of
`treprostinil or placebo was adjusted, with dose
`escalation (an additional 1 breath four times daily)
`occurring as often as every 3 days, with a target
`dose of 9 breaths four times daily and a maxi-
`mum dose of 12 breaths four times daily. Investi-
`gators adjusted the dose on an individual patient
`basis to achieve the maximum tolerated dose
`leading to functional improvement.
`
`Trial Assessments
`The 6-minute walk test was performed and labo-
`ratory data were obtained at baseline and at
`weeks 4, 8, 12, and 16, or at the time of early
`discontinuation of treprostinil or placebo. Each
`6-minute walk test was performed 10 to 60 min-
`utes after the most recent dose of active drug or
`placebo, which is the time of peak plasma
`treprostinil exposure. (A description of the pro-
`cedure for the 6-minute walk test is provided in
`Section S3.) A trough test was performed at
`week 15 at least 4 hours after the participant
`received a dose of treprostinil or placebo and at
`least 24 hours before the week 16 test. Pulse
`oximetry was performed immediately before,
`during, and after each 6-minute walk test. Mea-
`surement of N-terminal pro–B-type natriuretic
`peptide (NT-proBNP) levels and pulmonary func-
`tion tests were performed at baseline and at
`
`Outcome Measures
`The primary end point of the trial was the dif-
`ference between the two groups in the change in
`peak 6-minute walk distance from baseline to
`week 16. Secondary efficacy end points were
`analyzed in the following hierarchical testing
`order: the change in NT-proBNP level from base-
`line to week 16, the time to clinical worsening,
`the change in 6-minute walk distance at peak
`plasma treprostinil level at week 12, and the
`change in 6-minute walk distance at trough
`treprostinil level at week 15. The time to clinical
`worsening was evaluated from the time of ran-
`domization until the patient’s withdrawal from
`the trial and was defined as the time until the
`occurrence of any one of the following events:
`hospitalization for a cardiopulmonary indication,
`a decrease in 6-minute walk distance greater
`than 15% from baseline that was directly related
`to the disease under study at two consecutive
`visits and at least 24 hours apart, death from any
`cause, or lung transplantation.
`Exploratory end points were the changes in
`peak 6-minute walk distance at weeks 4 and 8,
`quality of life as measured with the use of the
`SGRQ at week 16, and the distance–saturation
`product (calculated by multiplying the total dis-
`tance walked by the lowest oxygen saturation
`measurement during the 6-minute walk) at
`week 16. Safety end points included adverse
`events, abnormal laboratory results, oxygenation
`as measured by pulse oximetry (Spo2) and sup-
`plemental oxygen requirement, changes in pul-
`monary function test results, hospitalization for
`a cardiopulmonary indication, and investigator-
`reported exacerbations of underlying lung disease,
`defined as acute, clinically significant respiratory
`deterioration characterized by evidence of new
`widespread alveolar abnormality. A full list of
`trial end points is provided in Section S4.
`
`Statistical Analysis
`Original estimates suggested that with 266 pa-
`tients randomly assigned in a 1:1 ratio to receive
`
`n engl j med 384;4 nejm.org
`
`January 28, 2021
`
`327
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on November 9, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2021 Massachusetts Medical Society. All rights reserved.
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`inhaled treprostinil or placebo, the trial would
`have at least 90% power at a significance level of
`0.05 (two-sided) to detect a between-group differ-
`ence of 30 m in the change in peak 6-minute
`walk distance from baseline at week 16, assuming
`a standard deviation of 75 m. To account for ap-
`proximately 15% of participants discontinuing the
`trial, 314 patients would need to be enrolled.
`For the primary efficacy analysis, the change
`in 6-minute walk distance was analyzed by mixed-
`model repeated-measures methods, under the
`assumption that missing data were missing at
`random. The model included the change from
`baseline to peak 6-minute walk distance as the
`dependent variable, with treatment, week, and
`treatment-by-week interaction as fixed effects,
`and the baseline 6-minute walk distance as a
`covariate. A sensitivity analysis for the primary
`end point was performed by means of a multiple
`imputation approach with a multivariate normal
`imputation model according to the Markov
`chain Monte Carlo method. The imputation mod-
`el included treatment group, all scheduled visits,
`the patient’s sex, and the patient’s age at ran-
`domization. If the result for the primary efficacy
`end point was significant, secondary efficacy end
`points were to be evaluated according to a hier-
`archical testing procedure. Confidence intervals
`have not been adjusted for multiplicity and can-
`not be used to infer definitive treatment effects
`for secondary efficacy end points. Additional de-
`tails of the statistical methods are provided in
`Section S5.
`
`R esults
`
`Patients
`Of 462 patients screened for eligibility, 326 were
`enrolled at 93 centers from February 3, 2017,
`through August 30, 2019, and were randomly as-
`signed to receive placebo (163 patients) or inhaled
`treprostinil (163 patients) (Fig. 1). Reasons for
`screening failure for the 136 patients who were
`excluded are shown in Table S1. Baseline charac-
`teristics were similar in the two groups (Table 1).
`The mean age of the patients was 66.5 years,
`46.9% were female, and the most common diag-
`nosis was idiopathic interstitial pneumonia (in
`44.8%). Baseline test data are provided in Table S2.
`At baseline, the mean 6-minute walk distance was
`259.6 m, the mean pulmonary vascular resistance
`
`was 6.2 Wood units, and the mean NT-proBNP
`level was 1832.9 pg per milliliter.
`
`Exposure and Follow-up
`Patients in the treprostinil group took a median
`of 11 breaths from the inhaler (66 μg) at each of
`four daily sessions at week 12 and 12 breaths (72
`μg) per session at week 16. The percentage of
`patients in this group who took 10 to 12 breaths
`(60 to 72 μg) per session was 57.0% at week 12
`and 57.8% at week 16. Patients in the placebo
`group took a median of 12 breaths from the
`inhaler per session at weeks 12 and 16.
`The date of the database lock was February 18,
`2020. Forty patients assigned to receive inhaled
`treprostinil (24.5%) and 38 assigned to placebo
`(23.3%) discontinued the assigned regimen pre-
`maturely. These patients were encouraged to re-
`main in the trial and complete assessments
`through week 16; 33 patients in the treprostinil
`group and 35 in the placebo group discontinued
`participation in the trial. The reasons for discon-
`tinuation are shown in Figure 1.
`
`Primary End Point
`Mean within-group changes in the 6-minute walk
`distance are shown in Figure 2. Mixed-model
`repeated-measures analysis showed that the least-
`squares mean difference between the treprostinil
`group and the placebo group in the change from
`baseline in peak 6-minute walk distance was
`31.12 m (95% confidence interval [CI], 16.85 to
`45.39; P<0.001) (Table 2 and Fig. S1). Similar ef-
`fects were observed across subgroups, including
`subgroups defined by disease cause and severity
`(as measured by baseline 6-minute walk distance),
`baseline hemodynamics, and dose group (Fig. S2).
`In addition, the between-group difference in the
`change from baseline in peak 6-minute walk
`distance at week 16 was significant when ana-
`lyzed with multiple imputation according to the
`Markov chain Monte Carlo method (30.97 m;
`95% CI, 16.53 to 45.41; P<0.001) (Fig. S3).
`
`Secondary and Exploratory End Points
`Patients assigned to inhaled treprostinil, as
`compared with those assigned to placebo,
`showed significant improvements in each of the
`secondary end points (Table 2). The NT-proBNP
`level decreased 15% from baseline with inhaled
`treprostinil and increased 46% from baseline with
`
`328
`
`n engl j med 384;4 nejm.org
`
`January 28, 2021
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on November 9, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2021 Massachusetts Medical Society. All rights reserved.
`
`

`

`Inhaled Treprostinil in Pulmonary Hypertension
`
`462 Patients were assessed for eligibility
`
`136 Failed screening
`
`326 Underwent randomization
`
`163 Were assigned to receive inhaled
`treprostinil
`
`163 Were assigned to receive placebo
`
`40 Discontinued treprostinil prematurely
`16 Had an adverse event
`6 Died
`6 Had progressive disease
`3 Had protocol violation
`7 Withdrew
`2 Had other reason
`
`38 Discontinued placebo prematurely
`13 Had an adverse event
`5 Died
`10 Had progressive disease
`9 Withdrew
`1 Had other reason
`
`33 Discontinued study
`participation
`7 Had an adverse event
`8 Died
`4 Had progressive disease
`2 Had protocol violation
`10 Withdrew consent
`2 Had other reason
`
`35 Discontinued study
`participation
`3 Had an adverse event
`10 Died
`1 Was lost to follow-up
`7 Had progressive disease
`13 Withdrew consent
`1 Had other reason
`
`130 Completed week 16 of study
`assessment
`
`128 Completed week 16 of study
`assessment
`
`Figure 1. Screening, Randomization, and Follow-up.
`Of 462 patients screened for eligibility, 326 patients underwent randomization and received at least one dose of
`the assigned treprostinil or placebo (included in the intention-to-treat and safety populations). Reasons for screen-
`ing failure (136 patients) are shown in Table S1. Of the patients who underwent randomization, 40 patients in the
`treprostinil group and 38 in the placebo group discontinued the assigned regimen prematurely. These patients were
`not withdrawn from the trial but were encouraged to remain and complete assessments through week 16; 33 patients
`in the treprostinil group and 35 in the placebo group discontinued trial participation before week 16.
`
`placebo, as assessed by the least-squares mean
`for the log-transformed ratio to the baseline level
`at week 16 (treatment ratio, 0.58; 95% CI, 0.47 to
`0.72; P<0.001) (Fig. S4). Clinical worsening oc-
`curred in 37 patients (22.7%) in the treprostinil
`group, as compared with 54 patients (33.1%) in
`the placebo group (hazard ratio, 0.61; 95% CI,
`0.40 to 0.92; P = 0.04 by the log-rank test) (Fig. S5).
`The least-squares mean change from baseline to
`week 12 in peak 6-minute walk distance was
`31.29 m greater in the treprostinil group than in
`the placebo group (P<0.001), and the change
`
`from baseline to week 15 in trough 6-minute
`walk distance was 21.99 m greater in the trepro-
`stinil group (P = 0.004). There was no significant
`between-group difference in patient-reported qual-
`ity of life as assessed with the SGRQ or in the
`distance–saturation product at week 16 (Tables
`S3 and S4).
`
`Safety End Points
`The most frequently reported adverse events were
`cough, headache, dyspnea, dizziness, nausea,
`fatigue, and diarrhea (Table 3). Most of these
`
`n engl j med 384;4 nejm.org
`
`January 28, 2021
`
`329
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on November 9, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2021 Massachusetts Medical Society. All rights reserved.
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`Table 1. Characteristics of the Patients at Baseline.*
`
`Characteristic
`
`Female sex — no. (%)
`
`Inhaled Treprostinil
`(N = 163)
`
`85 (52.1)
`
`Placebo
`(N = 163)
`
`68 (41.7)
`
`All Patients
`(N = 326)
`
`153 (46.9)
`
`Mean age at randomization (range) — yr
`
`65.6 (26–90)
`
`67.4 (36–85)
`
`66.5 (26–90)
`
`Age distribution — no. (%)
`
`<65 yr
`
`65 to <80 yr
`
`≥80 yr
`
`Race or ethnic group — no. (%)†
`
`White
`
`Black or African American
`
`American Indian or Alaska Native
`
`Asian
`
`Multiple
`
`Unknown
`
`Hispanic or Latino ethnic group — no. (%)†
`
`Yes
`
`No
`
`Data missing
`
`Mean time since diagnosis — yr
`
`Cause of lung disease — no. (%)
`
`64 (39.3)
`
`83 (50.9)
`
`16 (9.8)
`
`112 (68.7)
`
`41 (25.2)
`
`2 (1.2)
`
`7 (4.3)
`
`0
`
`1 (0.6)
`
`11 (6.7)
`
`152 (93.3)
`
`0
`
`0.54±1.16
`
`48 (29.4)
`
`100 (61.3)
`
`15 (9.2)
`
`126 (77.3)
`
`30 (18.4)
`
`1 (0.6)
`
`5 (3.1)
`
`1 (0.6)
`
`0
`
`16 (9.8)
`
`146 (89.6)
`
`1 (0.6)
`
`0.54±1.31
`
`112 (34.4)
`
`183 (56.1)
`
`31 (9.5)
`
`238 (73.0)
`
`71 (21.8)
`
`3 (0.9)
`
`12 (3.7)
`
`1 (0.3)
`
`1 (0.3)
`
`27 (8.3)
`
`298 (91.4)
`
`1 (0.3)
`
`0.54±1.23
`
`Idiopathic interstitial pneumonia
`
`Chronic hypersensitivity pneumonitis
`
`Occupational lung disease
`
`Combined pulmonary fibrosis and emphysema
`
`Connective tissue disease
`
`Other
`
`Idiopathic interstitial pneumonia subcategory — no. (%)
`
`Idiopathic pulmonary fibrosis
`
`Idiopathic nonspecific interstitial pneumonia
`
`Respiratory bronchiolitis associated with interstitial lung
`disease
`
`Desquamative interstitial pneumonia
`
`Acute interstitial pneumonia
`
`Unclassified idiopathic interstitial pneumonia
`
`Use of supplemental oxygen — no. (%)
`
`Background therapy — no. (%)
`
`None
`
`Pirfenidone only
`
`Nintedanib only
`
`65 (39.9)
`
`10 (6.1)
`
`5 (3.1)
`
`42 (25.8)
`
`40 (24.5)
`
`1 (0.6)
`
`37 (22.7)
`
`21 (12.9)
`
`2 (1.2)
`
`0
`
`0
`
`5 (3.1)
`
`119 (73.0)
`
`133 (81.6)
`
`19 (11.7)
`
`11 (6.7)
`
`81 (49.7)
`
`146 (44.8)
`
`9 (5.5)
`
`1 (0.6)
`
`40 (24.5)
`
`32 (19.6)
`
`0
`
`55 (33.7)
`
`16 (9.8)
`
`0
`
`1 (0.6)
`
`1 (0.6)
`
`8 (4.9)
`
`19 (5.8)
`
`6 (1.8)
`
`82 (25.2)
`
`72 (22.1)
`
`1 (0.3)
`
`92 (28.2)
`
`37 (11.3)
`
`2 (0.6)
`
`1 (0.3)
`
`1 (0.3)
`
`13 (4.0)
`
`114 (69.9)
`
`233 (71.5)
`
`119 (73.0)
`
`25 (15.3)
`
`19 (11.7)
`
`252 (77.3)
`
`44 (13.5)
`
`30 (9.2)
`
`* Plus–minus values are means ±SD. Additional patient characteristics at baseline are provided in Table S2 in the
`Supplementary Appendix. Percentages may not total 100 because of rounding.
`† Race and ethnic group were reported by the patient.
`
`330
`
`n engl j med 384;4 nejm.org
`
`January 28, 2021
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on November 9, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2021 Massachusetts Medical Society. All rights reserved.
`
`

`

`Inhaled Treprostinil in Pulmonary Hypertension
`
`Observed
`
`MMRM
`
`MCMC
`
`125
`
`Inhaled
`treprostinil
`
`121
`
`132
`
`148
`
`148
`
`4
`
`131
`
`121
`
`120
`
`Placebo
`
`8
`Week
`
`12
`
`16
`
`30
`25
`20
`15
`10
`5
`0
`−5
`−10
`−15
`−20
`
`ChangefromBaseline(m)
`
`0
`
`Figure 2. Mean Change from Baseline in Peak 6-Minute Walk Distance
`through Week 16.
`Shown are mean (±SE) changes from baseline (dashed line) in peak 6-min-
`ute walk distance over the 16-week trial period. The data shown are for pa-
`tients with available data (observed) as well as for the results of two analy-
`sis methods used to account for missing data. The values shown at each
`data point indicate the number of patients assessed at that time point.
`The primary analysis used mixed-model repeat-measurement (MMRM)
`methods, with the assumption that missing data were missing at random.
`The model included the change from baseline to peak 6-minute walk dis-
`tance as the dependent variable, with treatment, week, and treatment-by-
`week interaction as fixed effects, and the baseline 6-minute walk distance
`as a covariate. A sensitivity analysis for the primary end point was per-
`formed with the use of a multiple imputation approach with a multivariate
`normal imputation model using the Markov chain Monte Carlo (MCMC)
`method. The imputation model included treatment group, all scheduled
`visits, patient’s sex, and patient’s age at randomization. The confidence
`intervals have not been adjusted for multiplicity and cannot be used to infer
`definitive treatment effects.
`
`could worsen ventilation–perfusion matching in
`patients with group 3 pulmonary hypertension.
`Inhaled agents have the advantage of preferen-
`tially redirecting blood flow to the best-ventilated
`lung units, thus reducing the risk of ventilation–
`perfusion mismatching.9,14 Indeed, a retrospec-
`tive study of inhaled treprostinil in patients with
`group 3 pulmonary hypertension showed that
`such patients had improvements in functional
`class and 6-minute walk distance without any ad-
`verse effect on peripheral oxygen saturation, rein-
`forcing the concept of unchanged or even im-
`proved ventilation–perfusion matching with
`inhaled treprostinil.10 Similarly, in the current
`trial, we found no evidence of worsened oxygen-
`ation, which further allays concerns about venti-
`lation–perfusion mismatching.
`The INCREASE trial was not without its limi-
`
`events were of mild-to-moderate intensity. Seri-
`ous adverse events occurred in 23.3% of the pa-
`tients who received inhaled treprostinil and in
`25.8% of those who received placebo (Table S5).
`No serious adverse events were reported signifi-
`cantly more frequently in the treprostinil group
`than in the placebo group. A full list of serious
`adverse events is provided in Table S5.
`Significantly fewer patients in the treprostinil
`group than in the placebo group had exacerba-
`tions of underlying lung disease (43 [26.4%] vs.
`63 [38.7%]; P = 0.02 by Fisher’s exact test). Fewer
`patients in the treprostinil group than in the
`placebo group had a first occurrence of clinical
`worsening that involved hospitalization for a
`cardiopulmonary indication (18 [11.0%] vs. 24
`[14.7%]; P = 0.41). Inhaled treprostinil had no
`deleterious effect on any pulmonary function
`test variable during the trial (Table S6). There
`were no significant treatment-related changes in
`pulse oximetry or supplemental oxygen use in ei-
`ther group over the trial period (Tables S7 and S8).
`
`Discussion
`
`Pulmonary hypertension frequently complicates
`the treatment of patients with interstitial lung
`disease and is associated with worse functional
`status, greater need for supplemental oxygen, and
`worse outcomes.3,13 In the INCREASE trial, pa-
`tients treated with inhaled treprostinil had sig-
`nificant improvements in exercise capacity, as
`evidenced by changes in the 6-minute walk dis-
`tance. Treatment with inhaled treprostinil was
`also associated with a lower risk of clinical wors-
`ening than that in patients who received placebo,
`as well as reductions in NT-proBNP levels and
`fewer exacerbations of underlying lung disease,
`over the 16-week treatment period. The safety
`profile of inhaled treprostinil observed in this
`vulnerable patient population was similar to that
`reported in previous studies. The most frequent-
`ly reported adverse events were cough, headache,
`dyspnea, dizziness, nausea, fatigue, and diarrhea.
`The use of inhaled treprostinil was not associated
`with any decrement in lung function.
`Patients with group 3 pulmonary hyperten-
`sion are often treated with systemic pulmonary
`vasodilators, which are currently approved only
`for treatment of group 1 pulmonary hyperten-
`sion. However, there is concern that such agents
`
`n engl j med 384;4 nejm.org
`
`January 28, 2021
`
`331
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on November 9, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2021 Massachusetts Medical Society. All rights reserved.
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`Table 2. Summary of Primary and Secondary End Points.*
`
`End Point
`
`Primary end point
`
`Inhaled Treprostinil
`(N = 163)
`
`Placebo
`(N = 163)
`
`Treatment Effect
`(95% CI)
`
`P
`Value
`
`Change in peak 6-minute walk distance from baseline
`to wk 16 — m†
`
`21.08±5.12
`
`−10.04±5.12
`
`31.12±7.25
`(16.85 to 45.39)‡
`
`<0.001
`
`Secondary end points§
`
`Change in plasma concentration of NT-proBNP from
`baseline to wk 16¶
`
`Mean (±SD) change — pg/ml
`
`−396.35±1904.90
`
`1453.95±7296.20
`
`Median — pg/ml
`
`Range — pg/ml
`
`Ratio to baseline
`
`Occurrence of clinical worsening — no. (%)
`
`Any event
`
`Hospitalization for cardiopulmonary indication
`
`Decrease in 6-minute walk distance of >15% from
`baseline
`
`Death from any cause
`
`Lung transplantation
`
`−22.65
`
`20.65
`
`−11,433.0 to 5373.1
`
`−5483.3 to 87,148.3
`
`0.85±0.06
`
`1.46±0.11
`
`0.58±0.06 (0.47 to 0.72)‖
`
`<0.001
`
`0.61 (0.4 to 0.92)**
`
`0.04
`
`37 (22.7)
`
`18 (11.0)
`
`13 (8.0)
`
`4 (2.5)
`
`2 (1.2)
`
`54 (33.1)
`
`24 (14.7)
`
`26 (16.0)
`
`4 (2.5)
`
`0
`
`Least-squares mean change in peak 6-minute walk
`distance from baseline to wk 12 — m†
`
`Least-squares mean change in trough 6-minute walk
`distance from baseline to wk 15 — m
`
`18.77±4.99
`
`−12.52±5.01
`
`9.3±5.5
`
`−12.7±5.5
`
`31.29±7.07
`(17.37 to 45.21)‡
`
`21.99±7.7
`(6.85 to 37.14)‡
`
`<0.001
`
`0.005††
`
`*
`
`†
`
`‡
`§
`
`¶
`
` Plus–minus values are means ±SE, unless otherwise indicated. For secondary end points, the confidence intervals (CIs) have not been
`adjusted for multiplicity and cannot be used to infer definitive treatment effects. NT-proBNP denotes N-terminal pro–B-type natriuretic
`peptide.
` The effect of inhaled treprostinil as compared with placebo on the change in 6-minute walk distance was evaluated with the use of a
`mixed-model repeat measurement with the change from baseline in peak 6-minute walk distance as the dependent variable; treatment,
`week, and treatment-by-week interaction as the fixed effects; baseline 6-minute walk distance as the covariate; and subject as the random
`effect. Results are shown in Figures S1 and S3.
` This is a least-squares mean difference between the groups.
` The effect of inhaled treprostinil as compared with placebo on the change in log-transformed NT-proBNP was evaluated with the use of
`a mixed-model repeat measurement with the change from baseline in log-transformed NT-proBNP as the dependent variable; treatment,
`week, and treatment-by-week interaction as the fixed effects; and log-transformed baseline NT-proBNP as the covariate. Ratio to baseline
`is the least-squares mean of the change from baseline in log-transformed data.
` The change in plasma concentration of NT-proBNP from baseline to week 16 was assessed in 156 patients in the treprostinil group and
`160 in the placebo group.
` This is the treatment ratio, which is the ratio of ratios between two treatment groups.
`‖
`** This is a hazard ratio, calculated from a Cox proportional-hazards model. The P value was calculated with the use of a log-rank test strati-
`fied by the baseline 6-minute walk distance category.
`†† The P value was obtained from 100 multiple imputations with Markov chain Monte Carlo estimation with the use of analysis of covariance
`(ANCOVA) modeling, with the change from baseline in peak 6-minute walk distance as the dependent variable, treatment as a