Graphical abstract

Overview of the study. #: patients were stratified by baseline treatment with pulmonary hypertension (PH)-specific therapies (endothelin receptor antagonists, phosphodiesterase-5 inhibitors or soluble guanylyl cyclase stimulators; stratified as one therapy versus two therapies versus treatment-naïve) and chronic thromboembolic pulmonary hypertension (CTEPH) subpopulation (inoperable (with/without balloon pulmonary angioplasty (BPA)) versus persistent/recurrent after pulmonary endarterectomy (PEA) (including PEA followed by BPA)). ¶: the haemodynamic set comprised the first 91 patients randomised, who were planned to undergo right heart catheterisation (RHC) at week 20; the non-haemodynamic cohort comprised the remaining patients for which no week 20 RHC was required. +: secondary and exploratory end-points: time to death or PH-related hospitalisation, improvement in World Health Organization Functional Class (WHO FC) at week 26, change from baseline to week 26 in PAH-SYMPACT questionnaire cardiopulmonary and cardiovascular symptoms domains, and change from baseline to week 26 in Borg Dyspnoea Index or Borg Category-Ratio 10 scale. 6MWD: 6-min walk distance; SoC: standard of care; PVR: pulmonary vascular resistance; TTCW: time to clinical worsening; GMR: between-treatment geometric least squares mean ratio.

Abstract

Background SELECT was the first global randomised controlled trial of selexipag with standard of care in patients with inoperable or persistent/recurrent chronic thromboembolic pulmonary hypertension.

Methods SELECT was a multicentre, randomised, double-blind, placebo-controlled, parallel-group, group-sequential, phase 3 study (ClinicalTrials.gov: NCT03689244). Adults aged ≤85 years in World Health Organization Functional Class I–IV, with a 6-min walk distance of 100–450 m, were randomised (1:1) to receive selexipag (200–1600 µg twice daily titration until individual maximum tolerated dose)+standard of care or placebo+standard of care. Patients were recruited into the haemodynamic set (first 91 randomised patients to undergo right heart catheterisation (RHC); week 20) or non-haemodynamic cohort (remaining patients, no RHC required). The primary end-point was percent of baseline pulmonary vascular resistance (PVR; week 20). Safety was also assessed.

Results Of 321 patients screened, 128 were randomised (haemodynamic set n=91 (selexipag n=47; placebo n=44)). In the haemodynamic set, 29 (31.9%) patients had previous pulmonary endarterectomy (PEA), 20 (22.0%) balloon pulmonary angioplasty (BPA), and 14 (15.4%) both PEA and BPA; 28 (30.8%) were inoperable. The independent data monitoring committee recommended to stop the study for futility as no statistically significant difference was observed for the primary end-point (between-treatment geometric least squares mean ratio of PVR: 0.95, 95% CI 0.84–1.07; p=0.412). Adverse events were reported in 63 (98.4%) and 53 (82.8%) patients for selexipag and placebo, respectively.

Conclusions SELECT was discontinued for futility, as no treatment effect on the primary end-point (PVR) was observed. Safety data were consistent with the established safety profile of selexipag, with no new safety signals identified.

Shareable abstract

SELECT was a large global RCT investigating selexipag treatment for inoperable or persistent/recurrent CTEPH in patients with diverse treatment backgrounds. SELECT was discontinued for futility, as per pre-planned analysis of the primary end-point (PVR). https://bit.ly/4fAr446

Introduction

Chronic thromboembolic pulmonary hypertension (CTEPH) is defined as pulmonary hypertension (PH) in the presence of organised, chronic, thromboembolic material and altered vascular remodelling in pulmonary vasculature, as determined by imaging modalities [1, 2]. This results in elevated pulmonary vascular resistance (PVR), and ultimately right heart failure and death [3]. The incidence of CTEPH after acute pulmonary embolism (PE) has been reported as 0.56% for all patients with PE and 2.8–3.2% for PE survivors, based on data from a systematic review and meta-analysis [4].

The 2022 European Society of Cardiology (ESC)/European Respiratory Society (ERS) guidelines recommend multimodal treatment, including surgical/interventional procedures and medical therapy [1, 2]. Pulmonary endarterectomy (PEA) is a potentially curative surgery in patients with operable CTEPH [5]; however, ≤40% of patients are inoperable due to either distal pulmonary artery obstructions, haemodynamics or comorbidities [6, 7]. Balloon pulmonary angioplasty (BPA) is available for patients with inoperable or persistent/recurrent CTEPH [6]; however, there is limited access to specialist centres with the expertise to perform PEA and BPA [8]. Only three medical therapies have been approved for CTEPH. The first, riociguat, a soluble guanylate cyclase (sGC) stimulator, is approved globally for the treatment of inoperable or post-operative persistent CTEPH [9]; selexipag is approved in Japan [10, 11] and treprostinil in Europe [12–14].

Selexipag (Uptravi) is an oral IP prostacyclin receptor agonist approved for the treatment of CTEPH in Japan [11]. In a phase 3 randomised, placebo-controlled study of selexipag in 78 Japanese patients with inoperable or persistent/recurrent CTEPH, significant improvements from baseline to week 20 were observed for change in PVR (primary end-point; mean difference −93.5 (95% CI −156.8– −30.3) dyn·s·cm−5; p=0.006), Borg Dyspnoea Index (BDI)/Borg Category-Ratio 10 (CR10) scale score and cardiac index compared with placebo. However, no improvements in exercise capacity or World Health Organization Functional Class (WHO FC) were observed [10].

The present study, SELECT (ClinicalTrials.gov: NCT03689244), was the first global study of CTEPH in the era of expanded availability and expertise with multimodal treatments. SELECT was designed to investigate the efficacy and safety of selexipag as an add-on to standard of care in patients with inoperable or persistent/recurrent CTEPH after PEA and/or BPA.

MethodsStudy design and participants

SELECT was a multicentre, randomised, double-blind, placebo-controlled, parallel-group, group-sequential, phase 3 study (figure 1a). As there has been rapid evolution in the management of CTEPH, the study design involved stratification of treatment allocation by baseline PH-specific therapies and CTEPH subpopulations (based on history of BPA and/or PEA) and included a pre-planned testing hierarchy that was powered for the primary end-point as well as both key secondary end-points. Patients were randomised to receive selexipag or placebo, with a titration phase of up to 12 weeks (figure 1a).

FIGURE 1

a) Overall study design of SELECT and b) design flow of decision making for analysis conducted at sequential time-points. The double-blind treatment period started on the day of randomisation when patients received their first dose. The double-blind period ended on the day of the last dose with the end of double-blind treatment (EDBT) visit. The EDBT visit was planned to occur within 4 weeks after the announcement of the end of the double-blind period, for all subjects who had not discontinued treatment early. As per protocol, the end of the double-blind period could have been announced either when the overall target number of clinical worsening events had been reached, or earlier following recommendation of the independent data monitoring committee (IDMC) or sponsor's decision. It was announced following the recommendation of the IDMC to stop the study for futility. The open-label period started with the first dose of the open-label study intervention in the evening of the day of the last dose of double-blind study intervention; this period was only applicable to a small number of patients due to change in study design. #: patients were required to have pulmonary vascular resistance (PVR) at rest ≥5 WU for the haemodynamic set and ≥3.75 WU for the non-haemodynamic cohort; diagnosis and inoperability confirmed by an independent adjudication committee. ¶: the 12-week titration phase involved an incremental increase in study treatment from 200 μg twice daily to the individual maximum tolerated dose, up to 1600 μg twice daily (if well tolerated, incremental increase was 200 μg twice daily per week from week 0 to week 7). +: only patients who completed the 52-week period according to the initial study protocol were enrolled into the open-label extension. PH: pulmonary hypertension; 6MWD: 6-min walk distance; TTCW: time to clinical worsening; QoL: quality of life; EOLT: end of open-label treatment; EOS: end of study; CTEPH: chronic thromboembolic pulmonary hypertension; PEA: pulmonary endarterectomy; BPA: balloon angioplasty; WHO FC: World Health Organization Functional Class; ERA: endothelin receptor antagonist; PDE5: phosphodiesterase-5.

Eligible patients were adults aged ≤85 years with a diagnosis of inoperable (with/without BPA) or persistent/recurrent CTEPH after PEA and/or BPA (assessed by independent reviewers, as per the 2015 ESC/ERS guidelines [15]), who were in WHO FC I–IV with a 6-min walk distance (6MWD) of 100–450 m and were receiving full anticoagulation therapy for ≥90 days. Patients were recruited in two sequential cohorts: the haemodynamic set (first 91 patients randomised who were planned to undergo right heart catheterisation (RHC) (and left heart catheterisation, if needed) at week 20, in addition to overall study assessments; the remaining patients comprised the non-haemodynamic cohort (no post-baseline haemodynamic assessments). All patients had mean pulmonary arterial pressure ≥25 mmHg and PVR ≥5 WU (haemodynamic set) or ≥3.75 WU (non-haemodynamic cohort) collected ≥90 days after last PEA or BPA, as applicable. The full eligibility criteria are shown in supplementary table S1.

Patients were randomised (1:1) to receive oral selexipag or placebo. Treatment allocation was stratified by treatment with PH-specific therapies (endothelin receptor antagonists (ERAs), phosphodiesterase-5 (PDE5) inhibitors or sGC stimulators; stratified as one therapy versus two therapies versus treatment-naïve) and CTEPH subpopulation (inoperable (with/without BPA) versus persistent/recurrent after PEA (including PEA followed by BPA)). Doses of selexipag were titrated from 200 to 1600 µg twice daily in weekly increments of 200 µg until the individual maximum tolerated dose was reached. Patients with moderate hepatic impairment or who were concomitantly receiving a moderate CYP2C8 inhibitor had their dosing frequency adjusted to once daily.

Study end-points

The primary end-point was PVR at week 20, assessed at rest within 2–5 h post-dose, expressed as percent of baseline PVR (haemodynamic set). The secondary, exploratory and safety end-points are presented in supplementary table S2.

Statistical analysis

The sample size was calculated by t-test for superiority on the log-transformed percent of baseline PVR at week 20 using EAST version 6.4.0.1 (www.cytel.com) for ratio of means. A sample size of 90 patients in the haemodynamic set was sufficient to provide a power of >90% for the treatment effect of a 30% relative improvement in geometric mean compared with placebo with a coefficient of variation of 35–50% and a two-sided significance level of 5%. If patient enrolment was not terminated for futility or efficacy at analysis time-points 1 or 2, the total sample size would have been 280 patients.

The planned analysis time-points of SELECT, including stopping rules, are presented in figure 1b. At analysis time-point 1, the final analysis of PVR (primary end-point) was performed when 91 patients had performed the week 20 RHC. The testing strategy for primary and secondary end-points is presented in supplementary figure S1. When the first 91 randomised subjects had undergone RHC at week 20 or prematurely discontinued the study, the PVR end-point was tested on the haemodynamic set at two-sided α of 5%. The PVR end-point was tested by means of an ANCOVA model on the loge-transformed percent of baseline PVR at week 20 (including imputed values). Model covariates included randomised treatment, stratification factors as per the interactive web response system (PH-specific therapies and CTEPH population) and the loge-transformed baseline PVR value. Failure to meet the primary end-point was to lead to termination of the study for futility, as no remaining α would be available to formally test secondary end-points. The subsequent analysis time-points shown in figure 1b were dependent on meeting the primary end-point. The statistical analyses methods are presented in supplementary table S3.

The full analysis set (FAS) comprised all patients assigned to a study treatment. The efficacy evaluable set for week 26 efficacy end-points comprised patients in the FAS who reached the week 26 visit (according to the analysis time window for the specific week 26 efficacy end-point) or who had discontinued prior to week 26 (except for sponsor decision). Safety end-points were assessed in the safety set (SAF; all patients who received ≥1 doses of study treatment).

Finally, a post-hoc analysis was conducted to assess PVR by geography, operability and prior therapies.

Ethics approval

The trial protocol was reviewed and approved by the institutional review board or ethics committee at each participating centre. SELECT was conducted in accordance with the Declaration of Helsinki, and all patients provided written informed consent before participation.

ResultsPatients

In total, 321 patients were screened at 117 study centres across 31 countries; patients were randomised in 65 of these centres in Asia, Eastern Europe, Latin America, North America, Western Europe and Australia. The study period was from 23 January 2019 to 7 June 2022. Overall, 128 patients were randomised (FAS: selexipag n=64; placebo n=64) and received study treatment (SAF: selexipag n=64; placebo n=64). The first 91 of these patients were assigned to the haemodynamic set (selexipag n=47; placebo n=44). The disposition of patients is shown in figure 2. The independent data monitoring committee recommended to stop the study for futility; therefore, all ongoing patients prematurely discontinued the study (sponsor decision).

FIGURE 2

Disposition of patients in SELECT. EDBT: end of double-blind treatment. #: patients were screened at 117 study centres across 31 countries; ¶: patients were randomised in 65 study centres.

Baseline demographic and disease characteristics were generally well balanced between the selexipag and placebo groups in all cohorts (table 1). Mean±sd N-terminal pro-brain natriuretic peptide (NT-proBNP) at baseline within the FAS was higher in the selexipag group (951.0±1193.2 ng·L−1) compared with placebo (693.1±814.8 ng·L−1). In the selexipag and placebo groups, respectively, 26 (40.6%) and 25 (39.1%) patients had inoperable CTEPH in the FAS compared with 18 (38.3%) and 17 (38.6%) in the haemodynamic set. The median (range) time since diagnosis was 4.0 (0–24) and 6.0 (0–11) years for the selexipag and placebo groups (haemodynamic set), respectively. The medical history of patients is presented in supplementary table S4. The proportions of patients with respiratory/thoracic/mediastinal disorders and cardiac disorders were higher in the selexipag group (85.9% and 59.4%, respectively) compared with placebo (65.6% and 45.3%, respectively).

TABLE 1

Demographics and baseline characteristics of SELECT study participants

Efficacy

There was no statistically significant difference observed in the primary end-point (PVR), which led to the decision to stop the study for futility, as per the predefined interim analysis (figure 3). The between-treatment geometric least squares mean ratio of PVR (week 20/baseline) in the haemodynamic set was 0.95 (95% CI 0.84–1.07) (p=0.412). Mean±sd change in PVR from baseline to week 20 in the haemodynamic set was −92.4±163.4 and −82.8±179.0 dyn·s·cm−5 for selexipag and placebo, respectively.

FIGURE 3

Primary end-point in SELECT study: pulmonary vascular resistance (PVR) at week 20, assessed at rest, within 2–5 h post-dose, expressed as ratio of week 20 to baseline PVR. Data presented as geometric least squares mean ratio with 95% confidence interval. An ANCOVA model was applied on the loge-transformed PVR pre–post percent. Model covariates included randomised treatment, stratification factors per informative response technology (pulmonary hypertension-specific therapies and chronic thromboembolic pulmonary hypertension population) and the loge-transformed baseline PVR value.

Following the decision to terminate the study, the pre-planned testing hierarchy was no longer applicable, and all efficacy analyses other than the primary end-point were exploratory. No treatment differences were observed for any of the secondary end-points (supplementary table S5). Exploratory subgroup analyses of treatment effect on PVR by PH-specific therapy, CTEPH population, sex, WHO FC, dose group and NT-proBNP did not identify any differences between selexipag and placebo, except for race and geographical location, both of which showed a trend towards favouring placebo over selexipag for patients who were Asian and those from Asia/Latin America (supplementary figure S2).

Post-hoc analyses of PVR ratio of week 20 to baseline suggested a treatment effect favouring selexipag over placebo in several subgroups; however, none of these results were statistically significant. First, in patients with inoperable CTEPH excluding for Asia or Latin America, a treatment effect of 0.77 (95% CI 0.63–0.94) was observed (n=19). In patients receiving treatment with PH-specific sGC stimulators at baseline, the treatment effect was 0.83 (95% CI 0.71–0.98) compared with 1.02 (95% CI 0.85–1.24) for patients not receiving sGC stimulators at baseline. Of the patients receiving riociguat concomitantly, one (5.6%) and three (14.3%) were from Asia/Latin America in the selexipag and placebo groups, respectively; for those not receiving riociguat, nine (31.0%) and 14 (60.9%) were from Asia/Latin America, respectively. Finally, most patients with the highest PVR response (≥30% decrease from baseline) underwent PEA and/or BPA prior to baseline (selexipag n=8/11 (72.7%); placebo n=5/6 (83.3%)).

Safety

The mean±sd treatment duration and individual maximum tolerated dose were 11.8±8.0 months and 1001.6±534.7 µg twice daily for selexipag and 11.4±7.7 months and 1271.9±524.2 µg twice daily for placebo, respectively (SAF). The most commonly reached individual maximum tolerated dose was high (1200, 1400 or 1600 µg twice daily) in both the selexipag (45.3% (n=29)) and placebo (71.9% (n=46)) groups.

The safety results are presented in table 2. The proportion of patients reporting ≥1 adverse events (AEs) was higher in the selexipag group (n=63 (98.4%)) compared with placebo (n=53 (82.8%)). The most frequently reported AEs were prostacyclin-associated AEs (selexipag n=57 (89.1%); placebo n=29 (45.3%)). The incidence of AEs leading to discontinuation was 12.5% (n=8) and 6.3% (n=4) in the selexipag and placebo groups, respectively, and there were more patients with treatment-related AEs in the selexipag group (n=55 (85.9%)) compared with placebo (n=31 (48.4%)). In contrast, the incidence of serious AEs was lower in the selexipag group (n=9 (14.1%)) compared with placebo (n=16 (25.0%)). Two deaths were reported from each group while on study treatment during the double-blind period, both of which were treatment emergent (occurring within 3 days after the last dose of study treatment). The causes of death were cited as right ventricular failure and pneumonia (selexipag) and acute respiratory failure and death (placebo). Another death was reported in the selexipag group, 11 days after the last dose of selexipag in the double-blind period, due to multiple organ dysfunction syndrome.

TABLE 2

Summary of safety results during the double-blind treatment period of the study (safety analysis set)

Discussion

SELECT was discontinued for futility after the planned interim analysis of the primary end-point (percent of baseline PVR at week 20) did not show a statistically significant difference between selexipag and placebo. The mean±sd change in PVR from baseline to week 20 was −92.4±163.4 dyn·s·cm−5 in the selexipag group. However, an unexplainable placebo effect was observed, with a mean±sd change in PVR from baseline to week 20 of −82.8±179.0 dyn·s·cm−5 for placebo. Analyses of the secondary and exploratory end-points (6MWD, time to clinical worsening (TTCW) and other haemodynamic parameters) were consistent with the primary end-point, with no differences observed between groups. The most frequently reported AEs were those associated with prostacyclin receptor agonist treatment, CTEPH progression and other comorbidities. Treatment-emergent AEs with a fatal outcome were reported in two patients in each treatment group, with another death being reported 11 days after the last dose of selexipag; this death was not considered to be treatment emergent. The reported causes of death were consistent with the underlying CTEPH disease and associated comorbidities.

SELECT was designed to account for the evolving landscape of CTEPH and diversity of multimodal management. The chances of success for the primary (PVR) and two key secondary end-points (6MWD and TTCW) were maximised with a predefined testing hierarchy and stratification based on baseline therapies and CTEPH subpopulation. In addition, a group-sequential strategy and multiple analysis time-points allowed for early examination of futility and efficacy. It is important to consider the complexity of multimodal management of CTEPH where disease-specific pharmacological interventions are not widely available; thus, the end-points employed in CTEPH trials should include disease progression and response to treatment.

SELECT was the largest global trial in CTEPH and comprised a heterogeneous population who were receiving standard of care in the multimodal treatment era. As such, SELECT was the first global study of a pharmacological intervention in CTEPH that included patients with a history of BPA. The standard of care differs globally due to variation in the level of experience with PEA and/or BPA at different centres, whether these centres are accessible and operability status, which can be partly related to centre experience with the procedures. As a result, the study population was diverse in terms of baseline exercise capacity, haemodynamics and background PH medical therapies (n=70 (76.9%) were receiving PH therapy). Based on the post-hoc analyses, the differences between treatment effect of selexipag versus placebo might reflect the differences in standard of care by geographical location of patients and centres.

The change in PVR with selexipag in SELECT was consistent with that observed in two randomised controlled trials of selexipag in Japanese patients with inoperable/non-operated or persistent/recurrent CTEPH [10, 16]. In the phase 3 Japanese study, a significant reduction in PVR from baseline to week 16 was observed for selexipag versus placebo (treatment effect −93.5 (95% CI −156.8– −30.3) dyn·s·cm−5; p=0.006) [10]. However, in our study, we observed a placebo effect. The imbalance between the groups of patients receiving treatment with sGC stimulators (selexipag 38.3%; placebo 47.7%) could be partly responsible, as riociguat can have long-term effects on vascular disease function and 6MWD [17]. Most patients with the highest PVR response in the placebo group underwent PEA and/or BPA prior to baseline, and thus the effect of these procedures may have persisted beyond the 90-day restriction outlined in the eligibility criteria. The literature indicates that no significant improvements are seen after 6 months post-intervention [18, 19], although does not exclude the possibility of an effect between 3 and 6 months post-intervention, as speculated here. Another potential explanation for this placebo effect is the geographical location and race of patients. As reported in the exploratory analyses (supplementary figure S2), treatment effect favoured placebo (not significant) among patients in Asia and Latin America. However, the sample size of these subgroups was small [10].

In TRITON, the addition of selexipag to ERA+PDE5 inhibitor combination therapy in patients with pulmonary arterial hypertension (PAH) also had no effect on PVR at week 26, but had a potential positive effect on disease progression and potentially clinically meaningful long-term benefits [20], which is also reflected in the label for selexipag [11]. While the two diseases are different and there is a thrombotic component involved in CTEPH, there are known similarities in the pathophysiology and clinical features between PAH and CTEPH, with both diseases being subtypes of PH as per official nomenclature [1, 2]. While PVR was selected as the futility end-point in SELECT, long-term benefits of the drug could not be formally tested due to futility as per the clinical protocol being met, and thus there was no remaining α to allow for further formal testing nor the possibility to continue the study long-term.

Overall, the safety findings of SELECT were consistent with the established safety profile of selexipag [21], and the natural progression of CTEPH and its associated comorbidities [10]. Also, no safety concerns were reported, consistent with the Japanese studies of selexipag [10, 16].

Since the approval of riociguat in 2013 [22], selexipag (approved in Japan only) [11] and treprostinil (approved in the European Union) [12] are the only other medical therapies to be approved in the CTEPH field. Treprostinil subcutaneous is indicated for the treatment of adults with WHO FC III/IV and inoperable or persistent/recurrent CTEPH to improve exercise capacity, based on the pivotal CTREPH trial, which reported a treatment effect (high dose (target: 30 ng·kg−1·min−1) versus low dose (3 ng·kg−1·min−1)) for change in 6MWD at 24 weeks (primary end-point) of +41 (95% CI 16–66) m (p=0.016) [23]. There were several differences between SELECT and CTREPH, including the primary end-point and study population eligibility criteria and characteristics. Unlike SELECT, CTREPH only enrolled patients who had undergone PEA (not BPA) and only allowed ERAs and PDE5 inhibitors as baseline therapies (not sGC stimulators) [23]. A greater percentage of SELECT versus CTREPH participants were in WHO FC II (48% versus 6%) and receiving PH-specific therapies (78% versus 30%) at baseline [23]. The cohort of patients in SELECT being more heavily pre-treated may have contributed towards the lack of treatment effect observed.

This study has several limitations. Statistical comparisons between groups were not conducted for all end-points due to early termination. The combination of great diversity in a small sample size may have contributed to the unexplained placebo effect and may have meant the study was underpowered. Protocol deviations may have contributed towards differences in numbers between the randomisation tool (informative response technology) where stratification factors were recorded and the subpopulation entered in the database (electronic case report forms).

Conclusions

SELECT was the first globally conducted randomised controlled trial in patients treated with standard of care for CTEPH in the multimodal treatment era, employing a unique study design tailored to the evolving landscape of CTEPH. SELECT investigated selexipag as an add-on to standard of care in patients with inoperable CTEPH or persistent/recurrent CTEPH after PEA and/or BPA. SELECT was discontinued for futility, as no treatment effect on the primary end-point (PVR) was observed because of the unexpected placebo effect. The safety data were consistent with the established safety profile of selexipag, with no safety signals identified.

Supplementary materialSupplementary Material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material ERJ-00193-2024.Supplement

Acknowledgements

We wish to thank all the patients, family members and staff from all the units that participated in the study, and the independent data monitoring committee, central adjudication committee, country-specific adjudication committees (supplementary table S6), and the institutional review boards and ethics committees. Medical writing support was provided by Shaun Hall, Clare Lowe and Izabel James of Ashfield MedComms, an Inizio company (https://ashfieldmedcomms.com), and funded by Actelion Pharmaceuticals Ltd, a Johnson & Johnson Company.

Footnotes

This article has an editorial commentary: https://doi.org/10.1183/13993003.01560-2024

This clinical trial was prospectively registered at ClinicalTrials.gov with identifier NCT03689244.

Ethics approval: The trial protocol was reviewed and approved by the institutional review board or ethics committee at each participating centre. SELECT was conducted in accordance with the Declaration of Helsinki, and all patients provided written informed consent before participation.

Conflict of interest: N.H. Kim has consulted for, received research grants from or spoken for Janssen, Bayer, Merck, United Therapeutics, Enzyvant, Gossamer Bio, Polarean and Pulnovo. R. Channick has consulted for, received research grants from or spoken for Janssen, Bayer, United Therapeutics, Respira, Third Pole, Merck, Aria CV and Gossamer. M. Delcroix reports research grants from Janssen, speaker and consultant fees from Altavant, Acceleron, AOP, Bayer, Ferrer, Gossamer, INARI, Janssen, United Therapeutics and MSD, outside the submitted work, and all paid to her institution; and is holder of a Janssen Chair for Pulmonary Hypertension at KU Leuven. M. Madani has served as a consultant and has received fees from Wexler Surgical, Janssen, Bayer and MSD. J. Pepke-Zaba reports research grants from MSD and consultant fees from Ferrer, Gossamer, Janssen, United Therapeutics and MSD. V. Easton, S. Gesang and D. Richard are employees of Johnson & Johnson and own shares in the company. J.I. Borissoff is an employee of Johnson & Johnson. H-A. Ghofrani reports consultancy fees from Gossamer Bio, Inc., Aerovate, Altavant, Bayer AG, Attgeno, Janssen/Actelion, MSD/Acceleron, Pfizer, Liquidia, Morphic and Keros, payment or honoraria for lectures, presentations, manuscript writing or educational events from Bayer AG, Janssen/Actelion, Gossamer Bio, Keros and MSD/Acceleron, participation on a data safety monitoring board or advisory board with Aerovate, Altavant, Bayer AG, Attgeno, Janssen/Actelion, Insmed, MSD/Acceleron and Pfizer; H-A. Ghofrani's spouse is an employee of Liquidia.

Support statement: SELECT was funded by Actelion Pharmaceuticals Ltd, a Johnson & Johnson Company. Funding information for this article has been deposited with the Crossref Funder Registry.

Received February 6, 2024.Accepted July 25, 2024.Copyright ©The authors 2024.http://creativecommons.org/licenses/by-nc/4.0/

This version is distributed under the terms of the Creative Commons Attribution Non-Commercial Licence 4.0. For commercial reproduction rights and permissions contact permissionsersnet.org