European Journal of Heart Failure (2023) 25, 373–385 RESEARCH ARTICLE doi:10.1002/ejhf.2745 𝛃3 adrenergic agonist treatment in chronic pulmonary hypertension associated with heart failure (SPHERE-HF): a double blind, placebo-controlled, randomized clinical trial Ana García-Álvarez1,2,3,4* , Isabel Blanco2,5,6, Inés García-Lunar3,4,7, Paloma Jordà1, Juan J. Rodriguez-Arias1, Leticia Fernández-Friera3,4,8, Isabel Zegri9, Jorge Nuche3,4,10, Manuel Gomez-Bueno4,11, Susanna Prat1,2, Sandra Pujadas9, Eduard Sole-Gonzalez1, Maria D. Garcia-Cossio4,10, Mercedes Rivas9,11, Estefanía Torrecilla1, Daniel Pereda2,3,4,12, Javier Sanchez3,13, Pablo García-Pavía3,4,11, Javier Segovia-Cubero4,11, Juan F Delgado4,10, Sonia Mirabet4,9, Valentín Fuster3,14, Joan A. Barberá2,5,6, and Borja Ibañez3,4,15, for the SPHERE-HF Investigators† 1 Department of Cardiology, Hospital Clínic Barcelona-IDIBAPS, Barcelona, Spain; 2 Universitat de Barcelona, Barcelona, Spain; 3 Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain; 4 Centro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain; 5 Department of Pulmonary Medicine, Hospital Clínic Barcelona-IDIBAPS, Barcelona, Spain; 6 Centro de Investigación Biomédica en Red de Enfermedades Respiratorias (CIBERES), Madrid, Spain; 7 Cardiology Department, University Hospital La Moraleja, Madrid, Spain; 8 HM Hospitales-Centro Integral de Enfermedades Cardiovasculares HM-CIEC, Madrid, Spain; 9 Cardiology Department, Hospital Santa Creu I Sant Pau, IIb-Sant Pau, Universitat Autonoma de Barcelona, Barcelona, Spain; 10 Department of Cardiology, University Hospital 12 de Octubre, Instituto de Investigación Sanitaria i+12, Universdad Complutense, Madrid, Spain; 11 Department of Cardiology, University Hospital Puerta de Hierro-IDIPHISA, Universidad Autónoma de Madrid (UAM), Madrid, Spain; 12 Department of Cardiovascular Surgery, Hospital Clínic Barcelona-IDIBAPS, Barcelona, Spain; 13 Philips Healthcare Iberia, Madrid, Spain; 14 Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA; and 15 IIS-Fundación Jiménez Diaz University Hospital, Madrid, Spain Received 2 August 2022; revised 15 November 2022; accepted 17 November 2022 ; online publish-ahead-of-print 4 December 2022 Aims Pulmonary hypertension (PH) associated with left heart disease is an increasingly prevalent problem, orphan of targeted therapies, and related to a poor prognosis, particularly when pre- and post-capillary PH combine. The current study aimed to determine whether treatment with the selective β3 adrenoreceptor agonist mirabegron improves outcomes in patients with combined pre- and post-capillary PH (CpcPH). ..................................................................................................................................................................... Methods The β3 Adrenergic Agonist Treatment in Chronic Pulmonary Hypertension Secondary to Heart Failure (SPHERE-HF) and results trial is a multicentre, randomized, parallel, placebo-controlled clinical trial that enrolled stable patients with CpcPH associated with symptomatic heart failure. A total of 80 patients were assigned to receive mirabegron (50 mg daily, titrated till 200 mg daily, n = 39) or placebo (n = 41) for 16 weeks. Of them, 66 patients successfully completed the study protocol and were valid for the main analysis. The primary endpoint was the change in pulmonary vascular resistance (PVR) on right heart catheterization. Secondary outcomes included the change in right ventricular (RV) ejection fraction by cardiac magnetic resonance or computed tomography, other haemodynamic variables, functional class, and quality of life. The trial was negative for the primary outcome (placebo-corrected mean difference of 0.62 Wood units, 95% confidence interval [CI] −0.38, 1.61, p = 0.218). Patients receiving mirabegron presented a .......................................................................................................... *Corresponding author. Department of Cardiology, Hospital Clínic Barcelona-IDIBAPS, Universitat de Barcelona, Villarroel 170, 08036 Barcelona, Spain. Tel: +34 932 279400 (ext 9305), Fax: +34 932 279305, Emal: anagarci@clinic.cat † A list of the SPHERE-HF investigators in provided in online supplementary Appendix S1. © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 374 A. García-Álvarez et al. significant improvement in RV ejection fraction as compared to placebo (placebo-corrected mean difference of 3.0%, 95% CI 0.4, 5.7%, p = 0.026), without significant differences in other pre-specified secondary outcomes. ..................................................................................................................................................................... Conclusions SPHERE-HF is the first clinical trial to assess the potential benefit of β3 adrenergic agonists in PH. The trial was negative since mirabegron did not reduce PVR, the primary endpoint, in patients with CpcPH. On pre-specified secondary outcomes, a significant improvement in RV ejection fraction assessed by advanced cardiac imaging was found, without differences in functional class or quality of life. .......................................................................................................... Graphical Abstract Summary of the study outline, flowchart, and results of the primary and secondary outcomes. The β3 selective agonist mirabegron did not improve pulmonary haemodynamics in adult patients with combined pre- and post-capillary pulmonary hypertension associated with symptomatic heart failure. On secondary outcomes, a significant improvement in right ventricular ejection fraction assessed by advanced cardiac imaging was found, without differences in functional class or quality of life. PP, per protocol population. .......................................................................................................... Keywords Pulmonary hypertension • Heart failure • Mirabegron • β3 adrenoreceptors Introduction vasodilatation9,10 and prevention of cardiac remodelling in exper- .................................................. Pulmonary hypertension (PH) commonly complicates heart fail- imental animal models.11–13 In addition, prior experimental ure (HF) and confers a bad prognosis.1,2 Up to two thirds of research has demonstrated the presence of mRNA expression of patients with HF with both reduced (HFrEF) or preserved ejection hB3AR in human pulmonary arteries and has shown that treatment fraction (HFpEF) develop PH.3–5 Of them, approximately 13% with β3AR agonists produced a beneficial effect on pulmonary progress to combined pre- and post-capillary PH (CpcPH), char- haemodynamics and RV performance, associated with an attenua- acterized by higher pulmonary vascular resistance (PVR), more tion in pulmonary vascular proliferation in a swine model of chronic severe pulmonary vascular remodelling and worse prognosis than post-capillary PH.14 The selective oral β3AR agonist mirabegron isolated post-capillary PH.3 Right ventricular (RV) function is a is currently approved for the treatment of overactive bladder syn- well-recognized prognostic factor in PH associated with left heart drome and has demonstrated a good safety profile in healthy sub- disease (LHD), with an impact even greater than the severity of jects and patients, both suffering from the urinary condition15–17 the increase in PVR.1 Currently there are not specific pharma- or HFrEF.18 The potential efficacy of treatment with β3AR agonists cological therapies approved for patients with CpcPH either to on patients affected by PH has never been evaluated. On the basis improve pulmonary haemodynamics or RV adaptation.6 Beta-3 of these considerations, the β3 Adrenergic Agonist Treatment adrenoreceptors (β3AR) are expressed in the human myocardium7 in Chronic Pulmonary Hypertension Secondary to Heart Failure and vessels8 and have been described to be upregulated in LHD.7 (SPHERE-HF) trial was set up to evaluate the efficacy and safety Their stimulation or overexpression has been shown to cause of the selective β3AR agonist mirabegron in patients with CpcPH. © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License β3 adrenergic agonist treatment in chronic pulmonary hypertension associated with heart failure 375 Our hypothesis was that treatment with mirabegron in patients Blockrand package (R Foundation). Generation of the randomiza- ........................................................................................................................................................................ with CpcPH would result in a beneficial effect due to: (i) a reduction tion list and allocation concealment were done by statisticians who in PVR, (ii) an increase in RV performance, and (iii) improvement in were independent from researchers and the people involved in the clinical status without an increase in severe adverse events (AEs). implementation of assignments. The randomization list was provided exclusively to the pharmacy department, which was responsible for the preparation of medication kits by identifying them with a unique Methods sequential number for the entire study and provide them to the cen- tres. Researchers assigned medication kits in a sequential order and Study design recorded the kit number provided to each patient in the data collection system. The SPHERE-HF trial was a Phase II double-blind multicentre, with bal- anced randomization (1:1), placebo-controlled, parallel-group clinical trial conducted in Spain (five sites). The study received ethical approval from regional and national health service research ethics committees Procedures and was conducted in accordance with the principles of good clinical Patients underwent the following baseline procedures and assess- practice. All patients provided written informed consent before ments within 4 weeks before random allocation: demographic and randomization. The Hospital Clínic de Barcelona coordinated the trial medical history data collection; physical examination (including blood in collaboration with the Centro Nacional de Investigaciones Car- pressure, heart rate, and pulse oxymetry); NYHA functional class; diovasculares (CNIC). An independent data safety monitoring board blood sample analysis including N-terminal pro-B-type natriuretic pep- reviewed safety data every 6 months throughout the trial. This trial was tide (NT-proBNP); ECG; echocardiography; right heart catheterization registered at ClinicalTrials.gov NCT02775539 and EudraCT: 2016- (RHC); 6-min walking test, and a cardiac magnetic resonance (CMR) or 002949-32. Details of the trial design have been previously reported.19 a cardiac computed tomography (CCT). CMR was used preferably but in patients with severe claustrophobia or cardiac devices, a dedicated CCT examination was performed instead to measure RV volumes Participants and function. The protocol for the RHC was previously written and Patients with PH associated with HF were screened for possible inclu- standardized in all participating hospitals. Briefly, the procedure was sion. Eligible participants were adults aged 18 or over with symp- performed in a supine position with patients breathing mostly room tomatic HF (New York Heart Association [NYHA] functional class air or supplementary oxygen when needed through a venturi mask II–III) and secondary CpcPH who were on optimized evidence-based with a fixed FiO2 . Zero reference level was established at mid tho- pharmacological treatment and stable clinical condition during the racic level, at the intersection of the frontal plane at the mid thoracic 30 days preceding recruitment. Inclusion and exclusion criteria are level, and the transverse plane at the level of fourth anterior inter- presented on Table 1. The exclusion criteria of QTc interval on costal space. Determination of all pressure values was an averaged over electrocardiogram (ECG) >430 ms in men and >450 ms in women, 3–5 respiratory cycles (continuous registry). Systemic blood pres- that had been established as a general safety standard, was modified sure was firstly measured, followed by the measurement of systolic, by an amendment to QTc >480 ms, based on the high percentage of diastolic and mean pulmonary artery pressure (PAP), pulmonary artery patients with CpcPH who presented this exclusion criteria at baseline wedge pressure (PAWP, including ‘v’ wave when existed) and right atrial and the absence of data suggesting that mirabegron at the doses used pressure. When patients were in atrial fibrillation a longer stretch of could significantly increase QTc. the record was analysed. Determination of cardiac output was made by thermodilution at least three times without a difference greater than 10%. Finally, a sample from the pulmonary artery was drawn Manufacturing of study medication to measure PvO2 and SvO2 . A complete description of the exami- Manufacturing of mirabegron and placebo was carried out at the nations, haemodynamic and imaging protocols have been previously Hospital Clínic Pharmacy department. The excipient for placebo cap- published.19 sules (mixture of microcrystalline cellulose and colloidal silica) was Eligible patients were randomized to receive mirabegron (50 mg) received from the manufacturer (Fagron Iberica S.A.U, Terrassa, Spain) or placebo once daily on Visit 0, started study medication on Visit and included in capsules. Similarly, mirabegron pills were bought from 1 (during the 5 days after randomization), and were reviewed 1 week Astellas Pharma (Chuo, Tokyo, Japan), taken out of the blisters, and later in a safety visit (Visit 2), all in a blinded fashion. Thereafter, automatically encapsulated (a single pill into a capsule). Immediately, medication dose was uptitrated every 2 weeks for 8 weeks (up to a capsules containing mirabegron or excipient (exactly the same appear- maximum of 200 mg daily) based on patients’ blood pressure, heart ance and taste) were introduced in polyethylene bottles (30 capsules rate, QTc interval, blood analysis and clinical status assessed at each each), labelled as study medication with the batch number and expi- visit, and then maintained for another 8 weeks. Once finished the ration, and kept in the clinical trials area in conservation conditions 16-week period under treatment, patients underwent the same study (atmosphere temperature <25∘ C). Labelled was done in a blinded man- examinations performed at baseline and stopped the study medication. ner to guarantee masking of patients and physicians. Extra tablets were A final security monitoring visit took place 30 days after the last study given at each clinical visit. Therapeutic compliance was evaluated by medication dose, following clinical trials regulation. Figure 1 shows the counting empty blisters and remaining tablets. scheme of the study conduct and procedures. Randomization and blinding Outcomes Patients were randomly allocated (1:1) to mirabegron or placebo The primary outcome was the change from baseline to week 16 using randomly selected block sizes stratified by center using the in PVR on RHC, calculated in Wood units (WU), as the difference © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 376 A. García-Álvarez et al. Table 1 Inclusion and exclusion criteria Inclusion criteria Exclusion criteria ........................................................................................................................................... Written informed consent Non-coronary cardiac surgery (e.g. valvular surgery) or non-coronary structural percutaneous procedure (e.g. percutaneous mitral repair) within the 12 months preceding recruitment or scheduled ≥18 years old Myocardial infarction or coronary revascularization within the 3 months preceding recruitment HF with reduced, mildly reduced or preserved LVEF, according to the CRT implantation within the 6 months preceding recruitment definition of the European Society of Cardiology guidelines.6 Combined pre- and post-capillary PH determined by RHC showing Sinus tachycardia or uncontrolled atrial fibrillation (HR >100 bpm) the following: • PAWP or LVEDP ≥15 mmHg. • Mean PAP ≥25 mmHg; and: • PVR ≥3 WU and/or diastolic gradient ≥7 mmHg, or • Transpulmonary gradient ≥12 mmHg NYHA functional class II–IV Uncontrolled systemic hypertension (systolic BP >180 mmHg or diastolic BP >110 mmHg) or symptomatic hypotension (systolic BP <90 mmHg) On optimized evidence-based pharmacological treatment. Diagnosis of infiltrative cardiomyopathy Stable clinical condition defined as no changes in therapeutic regimen Pre-menopausal women who have not undergone total hysterectomy for HF or hospitalization in the 30 days preceding recruitment and no current plan for changing therapy Expected survival <1 year due to a disease other than HF Severe renal failure (GFR < 30 ml/min/1.73 m2 ) Severe hepatic impairment (transaminase elevation >3 times ULN) Prolonged cQT interval on the ECG (>430 ms in male or >450 ms in female)a Concomitant use with specific pulmonary vasodilators (sildenafil, bosentan, macitentan, riociguat or other endothelin receptor blockers, phosphodiesterase 5 inhibitors or guanylate cyclase stimulators) Treatment with digoxin, flecainide, propafenone, dabigatran, tricyclic antidepressants or other CYP2D6 inhibitors (other than beta-blockers) Severe COPD (FEV1 /FVC ratio <0.7 together with FEV1 <50% predicted value) Severe restrictive lung disease (TLC <50%) Participation in another clinical trial Known allergy to mirabegron or any of the excipients BP, blood pressure; COPD, chronic obstructive pulmonary disease; CRT, cardiac resynchronization therapy; FEV1 , forced expiratory volume in the first second; FVC, forced vital capacity; GFR, glomerular filtration rate; HR, heart rate; LVEDP, left ventricular end-diastolic pressure; NYHA, New York Heart Association; PAP, pulmonary artery pressure; PAWP, pulmonary artery wedge pressure; PVR, pulmonary vascular resistance; RHC, right heart catheterization; TLC, total lung capacity; ULN, upper limit of normal; WU, Wood units. a This criterion was changed by an amendment to QTc >480 ms. between mean PAP (mmHg) and PAWP (mmHg), divided by the centered in CNIC measured all acquired images (echocardiograms, ..................................... cardiac output (l/min). Pre-specified secondary outcomes were the CMR and CCT) in a completely blinded fashion and reported directly change from baseline in: RV ejection fraction assessed by advanced to the external statistics board. imaging (CMR or CCT), NYHA functional class, 6-min walk distance, dyspnoea Borg scale score, quality of life evaluated by the Kansas City Cardiomyopathy Questionnaire overall summary score, mean Statistical analysis PAP, transpulmonary gradient, diastolic gradient, cardiac output, and NT-proBNP. Safety measures included HF decompensation, death, As described previously, assuming a standard deviation of 3.2 WU urgent heart transplantation, AEs and adverse drug reactions, as (based in a population with CpcPH)20 and a correlation between well as monitoring of heart rate and the QTc interval on ECG. An baseline and 16-week measurement of 0.6, a sample size of 31 sub- independent data safety monitoring board reviewed all major events jects per group was required to detect a placebo-adjusted differ- and determined the potential relationship with study medication, ence of 2.1 WU, with a power of 80% and a 2-sided significance according to standard operational procedures. An imaging core-lab level of 5%. Anticipating a 20% dropout rate, the estimated num- ber of patients needed was 80. Calculations were made using the © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License β3 adrenergic agonist treatment in chronic pulmonary hypertension associated with heart failure 377 Figure 1 Study conduct and procedures. Summary of the study conduct and procedures performed at baseline and after 16 weeks of study medication (mirabegron or placebo). 6MWT, 6-min walking test; CCT, cardiac computed tomography; CMR, cardiac magnetic resonance; ECG, electrocardiogram; RHC, right heart catheterization. power.t.test function available in R software (R Foundation, Vienna, performed descriptively in the safety population. Data are presented .................................................................................... Austria). as mean ± standard deviation or median (first–third quartile) as cor- The following analysis populations were predefined: intention to responding. A p-value < 0.05 was considered as statistically signifi- treat (ITT) population (all randomized patients); per protocol (PP) cant. All statistical analysis were undertaken using R package (Vienna, population (patients from the ITT who have the primary outcome Austria). measured at 16 weeks and who took at least 80% of all medication doses); and the safety population (all patients who took at least one dose of the assigned treatment). As pre-established in the statistical analysis plan (and previously pub- Results lished19 ), due to the nature of the trial (proof-of-concept) and the Patients selected primary event (PVR on RHC), the primary analysis was per- formed in the PP population, with supportive analysis in patients on Of the 151 patients screened from June 2017 to December 2020, the ITT population. For the ITT population, missing values in the final 81 were initially considered eligible for randomization (Figure 2). measure of the primary outcome were replaced by the baseline value One patient was excluded after randomization because a lung (last observation carried forward). The treatment effect on the pri- ventilation–perfusion scan revealed a segmental perfusion defect mary and pre-specified secondary outcomes was analysed using an raising doubts about the potential thromboembolic origin of PH. ANCOVA model that included the baseline value, treatment, and base- Of the 80 patients who were included in the ITT population, line × treatment interaction. Interaction term was removed if the Wald 67 underwent a second RHC at 16 weeks and all of them but test was not statistically significant (p > 0.10). The centre was not one achieved at least 80% of therapeutic compliance, so these 66 included in the model due to small centre sizes.21 Model conditions were investigated in all cases (normality of residuals by Shapiro–Wilk patients composed the PP population (30 allocated to mirabegron test and the presence of observations with influence by graphical meth- and 36 to placebo). Reasons for withdrawal are summarized in ods). In case of deviations from normality or observations with influ- Figure 2. ence, the change from baseline to final measure (delta) was analysed Demographic and clinical characteristics were well balanced by randomization test and bootstrapping (9999 replicates), as imple- between treatment groups at baseline (Table 2). Overall, nearly mented in the boot.t.test function of the R MKinfer package. In case of two thirds of the patients had HFpEF and non-ischaemic aetiol- NYHA functional class, logistic regression instead of ANCOVA was ogy. About half of the patients had marked limitation of physi- used. In addition, patients were analysed by protocol-pre-specified cal activity (NYHA functional class III). Approximately one third subgroups according to left ventricular ejection fraction (<40% vs. (n = 19) had previous non-coronary cardiac surgery (aortic and/or ≥40%) and the maximum tolerated dose (mg/dl). Safety analyses were mitral valve surgery in 18 patients, and septal myectomy in one) © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 378 A. García-Álvarez et al. Figure 2 Patients flow chart. Screening and randomization. *One patient was mistakenly randomized because did not fulfil the inclusion criteria as she suffered from thromboembolic pulmonary hypertension. HF, heart failure; HR, heart rate; PP, per protocol population. and no patient had severe degenerative valve disease at inclusion. Primary and secondary outcomes ....................................................................... In concordance with the study protocol, patients were receiv- The primary outcome, placebo-corrected change from baseline to ing state-of-the-art medical and/or device therapy at baseline. In week 16 in PVR, was not met since there was a median change of HFpEF, optimized evidence-based pharmacological treatment was +0.09 WU (Q1–Q3, −0.49, +1.05 WU) in the mirabegron arm focused to the optimal treatment of comorbidities (hypertension, and a median change of −0.71 WU (Q1–Q3, −1.46, +0.84 WU) diabetes, heart rate control in the case of atrial fibrillation) and in the placebo arm (placebo-corrected difference in PVR of 0.62 hypervolaemia with the use of diuretics. Most patients were treated WU; 95% confidence interval [CI] −0.38, 1.61; p = 0.218). A sim- with beta-blockers, being selective beta-1 blockers the majority ilar result was obtained in the ITT population (placebo-corrected (bisoprolol [n = 32, 48%], carvedilol [n = 10, 15%], atenolol [n = 5; difference in PVR of 0.63 WU; 95% CI −0.22, 1.51; p = 0.152). 7.5%], nebivolol [n = 1, 1.5%], and sotalol [n = 1, 1.5%]). Median Table 3 shows the effect of mirabegron on the primary and therapeutic compliance was 100% of prescribed doses in both pre-specified secondary outcomes. The study resulted positive groups and median achieved dose was 150 mg in the mirabegron for the pre-specified secondary outcome RV ejection fraction by and 125 mg in the placebo group. Of the 66 patients who con- advanced cardiac imaging (placebo-corrected change from baseline stituted the PP cohort (with pre-post evaluation of the primary to week 16 of 3.0%; 95% CI 0.4, 5.7%; p = 0.026). No significant dif- outcome), 58 also had pre-post evaluable advanced cardiac imag- ferences were observed between groups in the rest of pre-specified ing. At baseline, all except two patients underwent CMR (n = 42) secondary outcomes. or CCT (n = 22), whereas all but other two had CMR or CCT performed at 16-week follow-up (n = 37 and n = 27, respectively). Reasons for not performing advanced imaging in these four patients Pre-specified subgroups analysis were technical problems (n = 3) and patient fragility due to recent Pre-specified analysis of subgroups according to baseline left ven- hospital admission (n = 1). In another four cases, images were con- tricular ejection fraction (<40% vs. ≥40%) and maximal tolerated sidered to be of insufficient quality to be evaluable by the Imaging dose were performed for the primary outcome. The interac- Core Lab. tion terms between dichotomized left ventricular ejection fraction © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License β3 adrenergic agonist treatment in chronic pulmonary hypertension associated with heart failure 379 with β3AR agonists, was very low. Mirabegron seemed not to affect ........................................................................................................................................................................ Table 2 Baseline demographic and clinical heart rate, systolic blood pressure or oxygen saturation either at characteristics according to treatment allocation rest or after exercise. The effect of mirabegron on other (not included in the primary and secondary outcomes) systemic and pul- Mirabegron Placebo (n = 30) (n = 36) monary haemodynamics evaluated by right heart catheterization ................................................................ was neutral (online supplementary Table S2). Online supplementary Age (years) 69.7 ± 9.2 64.8 ± 10.6 Table S3 shows cardiac advanced imaging parameters not included Male sex 12 (40.0) 17 (47.2) in the primary or secondary outcomes and measures obtained by Diabetes mellitus 14 (46.7) 18 (50.0) echocardiography. Globally, there were no robust differences in Atrial fibrillation 22 (73.3) 28 (77.8) echocardiographic parameters between groups, except for a sig- Heart disease nificant reduction in left atrial diameter (p = 0.016) and a trend Ischaemic 9 (30.0) 10 (27.8) Non-ischaemic 21 (70) 26 (72.2) towards reduction of RV wall thickness (p = 0.056) in the group HFpEF 21 (70) 26 (72.2) of mirabegron. Regarding the association between imaging tech- Left ventricular ejection 50.7 ± 14.5 51.7 ± 15.6 niques, at baseline and 16-week evaluations, RV ejection fraction by fraction (%) CMR/CT significantly correlated with tricuspid annular plane sys- NYHA functional class tolic excursion, tricuspid S wave velocity, RV fractional area change II 14 (46.7) 23 (63.9) and RV global longitudinal strain (p < 0.002 for Pearson correlation III 16 (53.3) 13 (36.1) in all cases). The highest correlation was found for RV fractional Previous non-coronary 9 (30.0) 10 (27.8) area change (R = 0.75 at baseline and R = 0.66 at follow-up), as cardiac surgery previously observed.22 Changes in RV ejection fraction by CMR/CT Pacemaker 2 (6.7) 4 (11.1) significantly correlated with changes in RV fractional area change Cardiac resynchronization 2 (6.7) 0 (0.0) (R = 0.3, p = 0.023). therapy Defibrillator 7 (23.3) 11 (30.6) Beta-blockers 22 (73.3) 27 (75) ACEI, ARB or ARNI 22 (73.3) 24 (66.6) Safety measures Mineralocorticoid receptor 14 (46.6) 22 (61.1) A total of 238 AEs were reported, 119 in the mirabegron and 119 antagonists in the placebo group, regarding 74 patients (38 in the mirabegron Loop diuretics 28 (93.3) 32 (88.9) and 36 in the placebo group) (Table 5). Severe AEs were reported Thiazide diuretics 3 (10.0) 2 (5.6) in 11 patients in the mirabegron group and 9 in the placebo Dihydropyridine calcium 2 (6.7) 2 (5.6) group. Of them, six events were considered to be related with channel blockers Ivabradine 5 (16.7) 1 (2.8) the study medication by the independent data safety monitoring Amiodarone 1 (3.3) 5 (13.9) board, three in the mirabegron (chest pain in one patient, and Oral anticoagulants 22 (73.3) 29 (80.6) incident atrial fibrillation and aggravated HF in another patient); and three in the placebo group (aggravated HF, incident atrial fibrillation Data are presented as mean ± standard deviation, or n (%). and urinary tract infection). Study medication was discontinued ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; ARNI, angiotensin receptor–neprilysin inhibitor; HFpEF, heart failure because of serious AEs in two patients in the mirabegron group with preserved ejection fraction; NYHA, New York Heart Association. (HF decompensation and evidence of mitral rupture requiring surgery) and two patients in the placebo group (long QT in the context of sepsis, and a gastrointestinal bleeding). There were two (Table 4) and treatment or maximal tolerated dose (as an ordinal deaths during the study, both in patients allocated to mirabegron. variable) and treatment were not significant (p-values of 0.513 and One patient presented a sudden cardiac death despite wearing an 0.340, respectively). Similar results were obtained using the cut-off implantable automatic defibrillator. The second patient had been of 50% (data not shown). The effect of mirabegron after adjust- withdrawn from the study because of the presence of a mitral ment by left ventricular ejection fraction remained not statistically rupture requiring surgery and died in the postoperative period. significant (0.52 WU; 95% CI −0.5, 1.54 WU) and neither after None of them were considered to be related with the study adjustment by maximal tolerated dose (0.62 WU, 95% CI −0.37, drug. As shown in online supplementary Table S4, no clinically 1.6 WU). An exploratory (not pre-specified) analysis did not show relevant changes were seen in vital signs, ECG or laboratory significant differences in treatment effect in patients in sinus rhythm variables. or atrial fibrillation (data not shown). Other variables Discussion Online supplementary Table S1 displays symptoms, physical exam- SPHERE-HF is the first study to assess the effect of a β3AR agonist ination and variables monitored during the 6-min walking test, at in patients with PH secondary to LHD. The primary endpoint, baseline and 16-week follow-up. The percentage of patients pre- placebo-corrected change from baseline to week 16 in PVR, was senting genitourinary symptoms or headache, typically associated not met. Consequently, interpretation of secondary endpoints © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 380 A. García-Álvarez et al. Table 3 Effect of mirabegron on primary and secondary outcomes Mirabegron (n = 30) Placebo (n = 36) Placebo-corrected p-value mean difference (95% CI) ........................................................................................................................................... Primary outcome PVR (WU) Baseline 4.00 (3.42–4.62) 4.66 (3.75–6.23) 0.62 (−0.38, 1.61) 0.218 Week 16 4.09 (3.00–5.88) 4.54 (2.23–6.14) Secondary outcomes RV ejection fraction (%) Baseline 44.9 ± 10.1 45.5 ± 9.3 3.0 (0.4, 5.7) 0.026 Week 16 48.3 ± 8.8 45.0 ± 9.6 Mean PAP (mmHg) Baseline 40.2 ± 9.6 44.0 ± 10.1 −0.5 (−4.2, 3.3) 0.812 Week 16 38.3 ± 10.6 40.9 ± 8.2 TPG (mmHg) Baseline 16.0 (14.0–21.5) 19.0 (15.8, 28.0) 2.1 (−1.0, 5.2) 0.206 Week 16 15.6 (12.0–19.8) 18.0 (13.8–23.2) CO (L/min) Baseline 4.2 ± 1.2 4.4 ± 1.3 −0.2 (−0.6, 0.2) 0.391 Week 16 4.0 ± 1.1 4.3 ± 1.3 Diastolic gradient (mmHg) Baseline 2.0 (0.0–5.8) 5.5 (1.8–8.2) 1.1 (−1.3, 3.3) 0.385 Week 16 3.0 (0.0–5.0) 3.0 (1.0–8.2) NYHA functional class Baseline OR 0.34 (0.09–1.08) 0.077 II 14 (46.7) 21 (58.3) III 16 (53.3) 15 (41.7) Week 16 I 3 (10.0) 4 (11.1) II 15 (50.0) 24 (66.7) III 12 (40.0) 8 (22.2) Quality of life (KCCQ overall summary score) (points) Baseline 58.8 ± 21.3 64.9 ± 18.9 −7.9 (−16.4, 0.6) 0.067 Week 16 60.3 ± 20.2 70.6 ± 20.7 6MWT distance (m) Baseline 357 (270–399) 368 (322–444) 5.3 (−25.1, 35.8) 0.720 Week 16 340 (281–414) 370 (314–434) Borg scale (points) Baseline 3.0 (1.0–5.0) 3.0 (1.2–4.0) −0.4 (−1.7, 0.9) 0.581 Week 16 2.0 (1.0–4.0) 3.0 (1.2–4.8) NT-proBNP (pmol/L) Baseline 1391 (802–3106) 1750 (1140–2485) −92 (−562, 364) 0.703 Week 16 1151 (741–2540) 1743 (1057–2645) Data are presented as mean ± standard deviation, n (%), or median (first–third quartile). 6MWT, 6-min walking test; CI, confidence interval; CO, cardiac output; KCCQ, Kansas City Cardiomyopathy Questionnaire; NT-proBNP, N-terminal pro-B-type natriuretic peptide; OR, odds ratio; PAP, pulmonary arterial pressure; PVR, pulmonary vascular resistance; RV, right ventricular; TPG, transpulmonary gradient; WU, Wood units. requires caution. In relation to these, patients receiving mirabegron Although different pharmacological treatment options are available ........................ presented a significantly better evolution of RV ejection fraction in patients with pulmonary arterial hypertension (PAH), none of during the course of the trial as compared to placebo without them has shown a beneficial effect in PH-LHD.23 Randomized other significant differences in the rest of pre-specified secondary clinical trials evaluating the effect of prostanoids and bosentan in outcomes. Importantly, mirabegron was well tolerated and had a patients with advanced HF and left ventricular systolic dysfunction favourable safety profile (Graphical Abstract). were stopped due to an increase in AEs.24–26 While these initial Treatment of PH in patients with HF remains an unmet clinical clinical trials did not select the population based on the pres- need due to its high prevalence and negative prognostic impact. ence or absence of PH, the subsequent results of the MELODY © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License β3 adrenergic agonist treatment in chronic pulmonary hypertension associated with heart failure 381 Table 4 Pre-specified subgroup analysis Mirabegron (n = 30) Placebo (n = 36) p-value interaction ..................................... ..................................... LVEF ≥40% LVEF <40% LVEF ≥40% LVEF <40% (n = 21) (n = 9) (n = 26) (n = 10) ........................................................................................................................................... PVR 0.513 Baseline 5.05 (2.46) 3.77 (0.90) 5.37 (2.36) 4.66 (1.83) Week 16 4.91 (3.09) 4.98 (3.14) 4.87 (1.99) 4.68 (3.37) LVEF, left ventricular ejection fraction; PVR, pulmonary vascular resistance. Table 5 Adverse events Mirabegron (n = 39) Placebo (n = 41) ................................... ................................... No. of SAEs No. of patients No of SAEs No of patients with AEs (%) with AEs (%) ........................................................................................................................................... Safety endpoint Death 2 0 Hospitalization or need for i.v. furosemide due to heart failure 6 5 (13) 3 3 (7) Investigator-reported adverse event SAEs 17 11 (28) 14 9 (22) SAE possibly related to the study drug 3 2 (5) 3 3 (7) Any AE 119 38 (97) 119 36 (88) Cardiac Heart failure, dyspnoea or oedema 23 16 (41) 17 15 (37) Incident or decompensated atrial flutter or fibrillation 3 3 (8) 3 3 (7) Ventricular arrhythmias 2 2 (5) 1 1 (2) QT interval prolongation 2 2 (5) 4 4 (10) Respiratory Respiratory tract infections 24 15 (38) 18 13 (32) Urinary Urinary symptoms or infection 3 3 (8) 8 7 (17) Nervous system Headache 4 3 (8) 2 2 (5) AE, adverse event; i.v., intravenous; SAE, serious adverse event. clinical trial with macitentan27 focused on patients with CpcPH, improve the primary outcome (mean PAP) both in PH secondary ..................................................... reinforced the message of lack of benefit compared to placebo. to HFrEF34 and HFpEF.35 In this regard, the consensus of the 6th Phosphodiesterase 5 inhibitors have also failed to demonstrate a World Symposium on Pulmonary Hypertension36 and recently beneficial effect in PH-LHD patients. Within this group, in PH-LHD published 2022 ESC/ERS guidelines6 included a strong recom- secondary to HFrEF, there is only one small single-centre study mendation against the use of pulmonary vasodilators in PH-LHD, published by Lewis et al.28 in which sildenafil improved exercise particularly in isolated post-capillary PH-LHD. Regarding riociguat, capacity and quality of life associated with a significant reduc- a very recent trial has reported a significant increase in the cardiac tion in PVR, and some observational studies that have reported output (in this case defined as the main outcome) in patients with haemodynamic improvement prior to heart transplantation.29 In PH associated with HFpEF.37 Of note, most of the aforementioned HFpEF, the randomized study by Guazzi et al.30 using sildenafil trials lacked a comprehensive evaluation of RV function by cardiac showed encouraging results; however, these were not replicated imaging. in the RELAX trial31 or in the randomized trial by Hoendermis Mirabegron is a novel class of drugs that act stimulating β3AR. et al.,32 that failed to show beneficial effects, and neither in the Like other adrenoceptors, β3ARs are coupled to G proteins and multicentre clinical trial SIOVAC.33 Indeed, in this latest trial, the downstream pathway after activation includes nitric oxide which included 200 patients with residual PH after valve surgery, (NO) synthase, NO-activated guanylyl cyclase and cyclic guano- the sildenafil-treated group demonstrated a higher rate of AEs. sine and adenosine monophosphate synthesis.37 Loss of cyclic Finally, riociguat, a soluble guanylate cyclase stimulator, did not guanosine and adenosine monophosphates is a hallmark of PH. © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 382 A. García-Álvarez et al. Within pulmonary circulation, cyclic nucleotides are responsible III–IV HFrEF receiving mirabegron (300 mg) for 7 days presented a ........................................................................................................................................................................ for mediating endothelin-dependent dilatation, thereby maintaining significant increase in cardiac output and reduction of PVR, com- pulmonary vascular homeostasis, but they also have valuable pared to placebo. This discrepancy may relay in differences between actions on pulmonary vascular remodelling, fibrosis and RV func- trials in patients’ characteristics (left ventricular ejection fraction of tion.38 Because of its promising pre-clinical results in a swine model 26 ± 7% vs. 48 ± 17%, NYHA functional class ≥III in all patients vs. of post-capillary PH14 and the safe profile demonstrated in patients 44% in ours), duration (7 days vs. 16 weeks) and dose (300 mg vs. with overactive bladder syndrome and HF,15–18 we considered that median of 150 mg). In contrast, mirabegron was associated with mirabegron deserved to be evaluated in patients with PH sec- a significantly better evolution of RV ejection fraction assessed by ondary to HF. cardiac advanced imaging, in the absence of significant differences SPHERE-HF is the first phase II trial exploring the effect of a in systemic blood pressure or heart rate. Clinical experience and β3AR in patients with CpcPH secondary to HFrEF or HFpEF. We previous studies have shown that PVR and RV ejection fraction may chose this population because we share the opinion20 that it is respond to PH therapies in a different manner. Thus, in the study the added pre-capillary component which has the greatest impact by van der Veerdonk et al.,42 PAH patients receiving PAH-targeted on prognosis in group 2 PH. Although we recognize that patho- therapies presented a significantly reduction in PVR and a signifi- physiology of HFrEF and HFpEF differs, we considered that CpcPH cantly increase in cardiac output but not differences in RV ejection selects a cohort of patients who share a common phenotype with fraction. Several mechanisms have been proposed to play a role severe pulmonary remodelling and RV dysfunction,20 and there- in RV ejection fraction besides afterload.42,44 Regarding the clin- fore most likely to benefit from a targeted therapy. The design ical relevance that a 3% increase of RV ejection fraction could of the trial coincided with the publication of the 2015 ESC/ERS have, a recent meta-analysis published by Alabed et al.45 demon- guidelines.23 The reason for using PVR, diastolic pressure and strated that a 1% reduction in RV ejection fraction in patients transpulmonary gradient in the discrimination of CpcPH was that with PAH was associated with a 4.9% increase in the risk of clin- transpulmonary gradient was widely used in clinical practice at that ical worsening and a 2.1% increase in the risk of death. To our moment and diastolic pressure had been recently incorporated in knowledge there are no similar quantitative data for CpcPH, other ESC/ERC guidelines. However, all patients fulfill the current defi- than the demonstrated independent and additive prognostic value nition of CpcPH based on the recently published 2022 ESC/ERS of RV ejection fraction.1 The indistinct use of CMR or CCT for guidelines6 and only six patients do not fulfill definition of CpcPH the measurement of RV ejection fraction is justified by previous according to the 2018 World Symposium27 (three randomized to literature that demonstrates an extraordinarily high concordance mirabegron and three to placebo). At inclusion, PVR severity was between both techniques, with a correlation coefficient of 0.98 lower than in the trial by Guazzi et al.30 in HFpEF (significant reduc- and a pooled standard mean difference of 0.65 (95% CI −2.60, tion of mean PAP using sildenafil), and slightly higher compared 1.29) in a recent meta-analysis.46 However, beyond a reduction to the SIOVAC trial33 in PH associated with corrected valve dis- in left atrial diameter and a trend towards a reduction in RV wall ease (negative results using sildenafil) and the recently published thickness, no robust changes were observed in the echocardio- DYNAMIC trial in HFpEF36 (significant improvement of cardiac graphic parameters. In this regard, it is known that the statisti- output with vericiguat). We decided to use the change in PVR as cal power to assess changes in cardiac remodelling and function the primary outcome as the most robust haemodynamic parame- using CMR is significantly higher than using echocardiography47 and ter in this population.39,40 Selecting endpoints for phase II clinical therefore it could be plausible that an eventual beneficial effect of trials in PH is not straightforward and have relevant implications, mirabegron was noted by CMR or CCT and not by echocardiog- for example, as discussed in the editorial comment by Bauersachs raphy. In fact, changes in RV ejection fraction by CMR significantly and Olsson,41 the clinical relevance of the positive result of the correlated with changes in RV fractional area change. If real, this DYNAMIC trial using cardiac output as the endpoint is unclear. beneficial effect would be independent of pressure afterload relief However, we acknowledged that RV systolic function has even a and could be associated with the cardioprotective effect observed higher prognostic impact than haemodynamic severity in group 2 in previous studies of ischemia–reperfusion12,48,49 and HF11,50,51 PH1 . Moreover, it is not uncommon that both variables vary in that involved an improvement in myocardial perfusion, antioxidant a non-coordinated way. Thereby, PAH patients may show a pro- and antifibrotic effects. In this sense, Bundgaard et al.50 demon- gressive decline in RV function despite a therapeutic response to strated that the β3AR stimulation produced an increased cardiac vasodilators.42 contractility in sheep after induction of HF that did not occur In our study, compared to patients assigned to placebo, patients before induction of the model (healthy myocardium). Protection receiving mirabegron failed to achieve a significant reduction in of the alpha1 subunit of the Na+ /K+ ATPase from inactivation by PVR after 16 weeks of treatment. In addition, in the pre-specified oxidative S-glutathionylation and subsequent reduction of intracel- subgroup analysis we found no evidence of a differential effect lular Na+ content has been considered critical for the improve- of mirabegron on PVR in the HFrEF or HFpEF populations. In ment of inotropism in the context of HF.50 In addition, β3AR may fact, pulmonary haemodynamics, including systolic and mean PAP, regulate cardiac function through paracrine effects from NO- and PAWP and cardiac output, as assessed by RHC, remained vir- endothelium-derived hyperpolarizing factor-dependent vasorelax- tually steady between the baseline and 16-week evaluations in ation of coronary arteries.8 Finally, demonstrated coupling of both groups. These results differ from those recently reported by β3AR to cGMP and downstream activation of PKG-dependent sig- Bundgaard et al.43 in which patients with NYHA functional class nalling may result in inhibition of hypertrophy.52 A direct inotropic © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License β3 adrenergic agonist treatment in chronic pulmonary hypertension associated with heart failure 383 effect attributed to off-target stimulation is unlikely due to the Supplementary Information ........................................................................................................................................................................ mirabegron’s high selectivity towards β3AR, the high percentage of patients under beta-blockers and the absence of significant changes Additional supporting information may be found online in the in heart rate or systemic blood pressure. Nevertheless, this even- Supporting Information section at the end of the article. tual beneficial effect observed in RV ejection fraction should be evaluated with caution as it was a secondary outcome, available Acknowledgements for the 88% of the cohort, and was not associated with a signif- The authors would like to thank all SPHERE-HF participants for icant improvement in functional class or quality of life. In addi- their invaluable participation in this trial. The authors also acknowl- tion, although there were no differences in the number and sever- edge the work of Celia Sánchez and Iris Dueñas, nurses with excel- ity of AEs, a higher number of patients in the mirabegron group lent dedication in the coordination of the trial at Hospital Clínic. withdrew study medication and there were two patients who died Also, we would like to thank Dr. Eulalia Roig for her valuable sup- (though considered not attributable to study medication by the port at the beginning of the project. safety external committee). The present study was designed and developed with a robust methodology (multicentre, randomized, double-blind clinical trial), Funding however we can point out some limitations. It included a hetero- The SPHERE-HF trial is an investigator-initiated noncommercial trial inde- geneous cohort of patients with both HFrEF and HFpEF. Despite pendent of the pharmaceutical industry. This work was funded by a grant being aware of the differences in the pathophysiology of both from Fundació La Marató de TV3 (20151730-31-32) to A.G.-A. as the coor- entities, we prioritized the fact that all patients had CpcPH, as we dinator. A.G.A. was granted by Instituto de Salud Carlos III INT19/00022, hypothesized that the main effect exerted by mirabegron would be PI17/00995 and PI20/00742 during the study period. B.I. is funded by the the reduction of PVR. Considering that the beneficial effect might European Commission (ERC-Consolidator Grant agreement No. 819775), be greater at the myocardial level, it would possibly have been the Spanish Ministry of Science and Innovation (PID2019-110369RB-I00), more appropriate to focus on a single HF phenotype. Measurement and the Comunidad de Madrid (S2017/BMD-3867 RENIM-CM). The CNIC of cardiac output by RHC, needed for PVR calculation, was done is supported by the Ministerio de Ciencia, Innovación y Universidades and by thermodilution, which may be inaccurate in patients with severe the Pro CNIC Foundation, and is a Severo Ochoa Center of Excellence (CEX2020-001041-S). CNIC imaging facilities are part of the ICTS ReDIB. tricuspid regurgitation (present in 10 patients at baseline and 11 IDIBAPS belongs to the CERCA Programme and receives partial funding at 16-week follow-up). This fact may explain, at least partially, the from the Generalitat de Catalunya. The study funders were not involved in discordance observed between the increase in RV ejection fraction the study design; the collection, analysis, or interpretation of data; the writ- and the absence of a significant increase in cardiac output. Indeed, ing of the manuscript; or the decision to submit the paper to publication. in the subpopulation with available advanced imaging, cardiac Conflict of interest: CNIC and Fundació Clínic per a la recerca biomèdica output by CMR did increase in patients receiving mirabegron hold a patent for the use of beta-3 agonists for the treatment of pulmonary (though differences did not reach significance as compared to hypertension (B.I., A.G.A., and V.F. are co-inventors). All other authors have placebo). Two thirds of patients had atrial fibrillation at baseline, nothing to disclose. and we cannot exclude a lower benefit in these patients compared to those in sinus rhythm, as has been observed with the use of beta-blockers in HFrEF,53 although the results of the exploratory References analysis do not point in this direction. Also, approximately one 1. Ghio S, Gavazzi A, Campana C, Inserra C, Klersy C, Sebastiani R, et al. Independent and additive prognostic value of right ventricular systolic function third had PH associated with corrected valve disease, subgroup and pulmonary artery pressure in patients with chronic heart failure. J Am Coll probably representing long evolution chronic PH, especially refrac- Cardiol. 2001;37:183–8. tory to treatment. Another limitation is associated with the 2. Bursi F, McNallan SM, Redfield MM, Nkomo VT, Lam CSP, Weston SA, et al. Pulmonary pressures and death in heart failure: a community study. J Am Coll baseline observation carried forward approach used to deal with Cardiol. 2012;59:222–31. missing data in the ITT population, which assumes no changes 3. Gerges M, Gerges C, Pistritto AM, Lang MB, Trip P, Jakowitsch J, et al. Pulmonary in patients receiving placebo or mirabegron. Nevertheless, no hypertension in heart failure. Epidemiology, right ventricular function, and sur- vival. Am J Respir Crit Care Med. 2015;192:1234–46. differences were observed in the results obtained in the PP or the 4. Lam CSP, Roger VL, Rodeheffer RJ, Borlaug BA, Enders FT, Redfield MM. ITT population. Pulmonary hypertension in heart failure with preserved ejection fraction. J Am Coll Cardiol. 2009;53:1119–26. 5. Miller WL, Grill DE, Borlaug BA. Clinical features, hemodynamics, and outcomes of pulmonary hypertension due to chronic heart failure with reduced ejection Conclusions fraction. JACC Heart Fail. 2013;1:290–9. 6. Humbert M, Kovacs G, Hoeper MM, Badagliacca R, Berger RMF, Brida M, The SPHERE-HF trial, the first to evaluate the safety and efficacy of et al. 2022 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary mirabegron in CpcPH, did not achieve the primary outcome, since hypertension. Eur Heart J. 2022;26:3618–731. 7. Moniotte S, Kobzik L, Feron O, Trochu JN, Gauthier C, Balligand JL. Upregulation the placebo-corrected change from baseline to week 16 in PVR of β3 -adrenoceptors and altered contractile response to inotropic amines in was not met. On pre-specified secondary outcomes, a significant human failing myocardium. Circulation. 2001;103:1649–55. improvement in RV ejection fraction assessed by advanced cardiac 8. Dessy C, Moniotte S, Ghisdal P, Havaux X, Noirhomme P, Balligand JL. Endothelial β3-adrenoceptors mediate vasorelaxation of human coronary microarteries imaging was found without differences in functional class or quality through nitric oxide and endothelium-dependent hyperpolarization. Circulation. of life. 2004;110:948–54. © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 384 A. García-Álvarez et al. 9. Dumas M, Dumas JP, Bardou M, Rochette L, Advenier C, Giudicelli JF. 31. Redfield MM, Chen HH, Borlaug BA, Semigran MJ, Lee KL, Lewis G, et al. ........................................................................................................................................................................ Influence of β-adrenoceptor agonists on the pulmonary circulation. Effects RELAX TrialEffect of phosphodiesterase-5 inhibition on exercise capacity and of a β3-adrenoceptor antagonist, SR 59230A. Eur J Pharmacol. 1998;348: clinical status in heart failure with preserved ejection fraction. JAMA. 2013;309: 223–8. 1268–77. 10. Tagaya E, Tamaoki J, Takemura H, Isono K, Nagai A. Atypical 32. Hoendermis ES, Liu LCY, Hummel YM, van der Meer P, de Boer RA, Berger RMF, adrenoceptor-mediated relaxation of canine pulmonary artery through et al. Effects of sildenafil on invasive haemodynamics and exercise capacity in heart a cyclic adenosine monophosphate–dependent pathway. Lung. 1999;177: failure patients with preserved ejection fraction and pulmonary hypertension: a 321–32. randomized controlled trial. Eur Heart J. 2015;36:2565–73. 11. Niu X, Watts VL, Cingolani OH, Sivakumaran V, Leyton-Mange JS, Ellis CL, et al. 33. Bermejo J, Yotti R, García-Orta R, Sánchez-Fernández PL, Castaño M, Cardioprotective effect of beta-3 adrenergic receptor agonism. J Am Coll Cardiol. Segovia-Cubero J, et al. Sildenafil for Improving Outcomes after VAlvular 2012;59:1979–87. Correction (SIOVAC) InvestigatorsSildenafil for improving outcomes in patients 12. Niu X, Zhao L, Li X, Xue Y, Wang B, Lv Z, et al. β3-adrenoreceptor stimulation with corrected valvular heart disease and persistent pulmonary hypertension: protects against myocardial infarction injury via eNOS and nNOS activation. PLoS a multicenter, double-blind, randomized clinical trial. Eur Heart J. 2018;39: One. 2014;9:e98713. 1255–64. 13. Fry N, Liu CC, Garcia A, Hamilton EJ, Galougahi KK, Kim YJ, et al. Targeting 34. Bonderman D, Ghio S, Felix SB, Ghofrani HA, Michelakis E, Mitrovic V, et al. cardiac myocyte Na+ -K+ pump function with β3 adrenergic agonist in rabbit Left Ventricular Systolic Dysfunction Associated With Pulmonary Hypertension model of severe congestive heart failure. Circ Heart Fail. 2020;13:e006753. Riociguat Trial (LEPHT) Study GroupRiociguat for patients with pulmonary hyper- 14. García-Álvarez A, Pereda D, García-Lunar I, Sanz-Rosa D, Fernández-Jiménez R, tension caused by systolic left ventricular dysfunction: a pahse IIb double-blind, García-Prieto J, et al. Beta-3 adrenergic agonists reduce resistance and improve randomized, placebo-controlled, dose-ranging hemodynamic study. Circulation. right ventricular performance in a porcine model of chronic pulmonary hyper- 2013;128:502–11. tension. Basic Res Cardiol. 2016;111:49. 35. Bonderman D, Pretsch I, Steringer-Mascherbauer R, Jansa P, Rosenkranz S, 15. Malik M, van Gelderen EM, Lee JH, Kowalski DL, Yen M, Goldwater R, et al. Tufaro C, et al. Acute hemodynamic effects of riociguat in patients with pul- Proarrhythmic safety of repeat doses of mirabegron in healthy subjects: a monary hypertension associated with diastolic heart failure (DILATE-1): a ran- randomized, double-blind, placebo-, and active-controlled thorough QT study. domized, double-blind, placebo-controlled, single-dose study. Chest. 2014;146: Clin Pharmacol Ther. 2012;92:696–706. 1274–85. 16. Chapple CR, Kaplan SA, Mitcheson D, Klecka J, Cummings J, Drogendijk T, et al. 36. Dachs TM, Duca F, Rettl R, Binder-Rodriguez C, Dalos D, Ligios LC, et al. Randomized double-blind, active-controlled phase 3 study to assess 12-month Riociguat in pulmonary hypertension and heart failure with preserved ejection safety and efficacy of mirabegron, a β3-adrenoceptor agonist, in overactive fraction: the haemoDYNAMIC trial. Eur Heart J. 2022;21:3402–13. bladder. Eur Urol. 2013;63:296–305. 37. Rozec B, Gauthier C. β3-adrenoceptors in the cardiovascular system: putative 17. Herschorn S, Barkin J, Castro-Diaz D, Frankel JM, Espuna-Pons M, Gousse AE, roles in human pathologies. Pharmacol Ther. 2006;111:652–73. et al. A phase III, randomized, double-blind, parallel-group, placebo-controlled, 38. Baliga RS, MacAllister RJ, Hobbs AJ. New perspectives for the treatment of multicentre study to assess the efficacy and safety of the β3 adrenoceptor pulmonary hypertension. Br J Pharmacol. 2011;163:125–40. agonist, mirabegron, in patients with symptoms of overactive bladder. Urology. 39. Chang PP, Longenecker JC, Wang NY, Baughman KL, Conte JV, Hare JM, et al. 2013;82:313–20. Mild vs severe pulmonary hypertension before heart transplantation: different 18. Bundgaard H, Axelsson A, Hartvig Thomsen J, Sørgaard M, Kofoed KF, Hassel- effects on posttransplantation pulmonary hypertension and mortality. J Heart balch R, et al. The first-in-man randomized trial of a beta3 adrenoceptor agonist Lung Transplant. 2005;24:998–1007. in chronic heart failure: the BEAT-HF trial. Eur J Heart Fail. 2017;19:566–75. 40. Vonk Noordegraaf A, Westerhof BE, Westerhof N. The relationship between 19. Garcia-Lunar I, Blanco I, Fernández-Friera L, Prat-Gonzàlez S, Jordà P, the right ventricle and its load in pulmonary hypertension. J Am Coll Cardiol. Sánchez J, et al. Design of the β3-Adrenergic Agonist Treatment in Chronic 2017;69:236–43. Pulmonary Hypertension Secondary to Heart Failure trial. JACC Basic Transl Sci. 41. Bauersachs J, Olsson KM. Targeting pulmonary hypertension in patients with 2020;5:317–27. heart failure and preserved ejection fraction: rather static than DYNAMIC 20. Gerges C, Gerges M, Lang MB, Zhang Y, Jakowitsch J, Probst P, et al. development? Eur Heart J. 2022;21:3414–6. Diastolic pulmonary vascular pressure gradient: a predictor of prognosis in 42. van der Veerdonk MC, Kind T, Marcus JT, Mauritz GJ, Heymans MW, Bogaard HJ, “out-of-proportion” pulmonary hypertension. Chest. 2013;143:758–66. et al. Progressive right ventricular dysfunction in patients with pulmonary arterial 21. Pickering RM, Weatherall M. The analysis of continuous outcomes in multi-centre hypertension responding to therapy. J Am Coll Cardiol. 2011;58:2511–9. trials with small centre sizes. Stat Med. 2007;26:5445–56. 43. Bundgaard H, Axelsson Raja A, Iversen K, Valeur N, Tønder N, Schou M, et al. 22. Garcia-Alvarez A, Fernandez-Friera L, Lau JF, Sawit ST, Mirelis JG, Castillo JG, Hemodynamic effects of cyclic guanosine monophosphate-dependent signaling et al. Evaluation of right ventricular function and post-operative findings using through β3 adrenoceptor stimulation in patients with advanced heart failure: a cardiac computed tomography in patients with left ventricular assist devices. randomized invasive clinical trial. Circ Heart Fail. 2022;15:e009120. J Heart Lung Transplant. 2011;30:896–903. 44. Kawut SM, Al-Namnani N, Agerstrand C, Rosenzweig EB, Rowan C, Barst RJ, 23. Galiè N, Humbert M, Vachiery JL, Gibbs S, Lang I, Torbicki A, et al. 2015 et al. Determinants of right ventricular ejection fraction in pulmonary arterial ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension. hypertension. Chest. 2009;135:752–9. Eur Heart J. 2016;37:67–119. 45. Alabed S, Shahin Y, Garg P, Alandejani F, Johns CS, Lewis RA, et al. Cardiac-MRI 24. Califf RM, Adams KF, McKenna WJ, Gheorghiade M, Uretsky BF, McNulty SE, predicts clinical worsening and mortality in pulmonary arterial hyperten- et al. A randomized controlled trial of epoprostenol therapy for severe congestive sion: a systematic review and meta-analysis. JACC Cardiovasc Imaging. 2021;14: heart failure: the Flolan International Randomized Survival Trial (FIRST). Am 931–42. Heart J. 1997;134:44–54. 46. Fu H, Wang X, Diao K, Huang S, Liu H, Gao Y, et al. CT compared to MRI for 25. Kalra PR, Moon JCC, Coats AJS. Do results of the ENABLE (Endothelin Antago- functional evaluation of the right ventricle: a systematic review and meta-analysis. nist Bosentan for Lowering Cardiac Events in Heart Failure) study spell the end for Eur Radiol. 2019;29:6816–28. non-selective endothelin antagonism in heart failure? Int J Cardiol. 2002;85:195–7. 47. Bellenger N, Davies LC, Francis J, Coats A, Pennell D. Reduction in sample size 26. Packer M, McMurray J, Massie BM, Caspi A, Charlon V, Cohen-Solal A, et al. for studies of remodeling in heart failure by the use of cardiovascular magnetic Clinical effects of endothelin receptor antagonism with bosentan in patients with resonance. J Cardiovasc Magn Reson. 2000;2:271–8. severe chronic heart failure: results of a pilot study. J Card Fail. 2005;11:12–20. 48. Aragón JP, Condit ME, Bhushan S, Predmore BL, Patel SS, Grinsfelder DB, et al. 27. Vachiéry JL, Tedford RJ, Rosenkranz S, Palazzini M, Lang I, Guazzi M, et al. Beta3-adrenoreceptor stimulation ameliorates myocardial ischemia-reperfusion Pulmonary hypertension due to left heart disease. Eur Respir J. 2019;53:1801897. injury via endothelial nitric oxide synthase and neuronal nitric oxide synthase 28. Lewis GD, Lachmann J, Camuso J, Lepore JJ, Shin J, Martinovic ME, et al. Sildenafil activation. J Am Coll Cardiol. 2011;58:2683–91. improves exercise hemodynamics and oxygen uptake in patients with systolic 49. García-Prieto J, García-Ruiz JM, Sanz-Rosa D, Pun A, García-Alvarez A, heart failure. Circulation. 2007;115:59–66. Davidson SM, et al. β3 adrenergic receptor selective stimulation during 29. Jabbour A, Keogh A, Hayward C, Macdonald P. Chronic sildenafil lowers transpul- ischemia/reperfusion improves cardiac function in translational models through monary gradient and improves cardiac output allowing successful heart transplan- inhibition of mPTP opening in cardiomyocytes. Basic Res Cardiol. 2014; tation. Eur J Heart Fail. 2007;9:674–7. 109:422. 30. Guazzi M, Vicenzi M, Arena R, Guazzi MD. PDE5 inhibition with sildenafil 50. Bundgaard H, Liu CC, Garcia A, Hamilton EJ, Huang Y, Chia KKM, et al. β3 improves left ventricular diastolic function, cardiac geometry, and clinical status adrenergic stimulation of the cardiac Na+ -K+ pump by reversal of an inhibitory in patients with stable systolic heart failure. Circ Heart Fail. 2011;4:8–17. oxidative modification. Circulation. 2010;122:2699–708. © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology. 18790844, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ejhf.2745 by Readcube (Labtiva Inc.), Wiley Online Library on [18/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License β3 adrenergic agonist treatment in chronic pulmonary hypertension associated with heart failure 385 51. Kamiya M, Asai K, Maejima Y, Shirakabe A, Murai K, Noma S, et al. neurohormone-induced hypertrophic remodeling through nitric oxide synthase. ............ β3 -adrenergic receptor agonist prevents diastolic dysfunction in an angiotensin Circulation. 2014;129:451–62. II–induced cardiomyopathy mouse model. J Pharmacol Exp Ther. 2021;376: 53. Kotecha D, Holmes J, Krum H, Altman DG, Manzano L, Cleland JG, et al.; 473–81. Beta-Blockers in Heart Failure Collaborative Group. Efficacy of β blockers 52. Belge C, Hammond J, Dubois-Deruy E, Manoury B, Hamelet J, Beauloye C, in patients with heart failure plus atrial fibrillation: an individual-patient data et al. Enhanced expression of β3-adrenoceptors in cardiac myocytes attenuates meta-analysis. Lancet. 2014;384:2235–43. © 2022 The Authors. European Journal of Heart Failure published by John Wiley & Sons Ltd on behalf of European Society of Cardiology.