The Anti-Anginal Efficacy of Bosentan in the Coronary Slow Flow Phenomenon

Original Article

Austin J Cardiovasc Dis Atherosclerosis. 2023; 10(1): 1057.

The Anti-Anginal Efficacy of Bosentan in the Coronary Slow Flow Phenomenon

Victoria Kopetz, BSc (Hons), PhD1,3; Rosanna Tavella, BSc (Hons), PhD1-3; Sivabaskari Pasupathy, BSc (Hons), PhD1-3; John Horowitz, BSc, MBBS, PhD1,3; John Beltrame, BSc, BMBS, FRACP, PhD, FESC, FACC, FAHA, FCSANZ1-3*

¹School of Medicine, Faculty of Health Sciences, The University of Adelaide, Adelaide, SA 5000, Australia

²Central Adelaide Local Health Network, Adelaide, SA 5000, Australia

³Basil Hetzel Institute for Translational Health Research, Adelaide, SA 5011, Australia

*Corresponding author: John F Beltrame Faculty of Health Sciences, University of Adelaide Central Adelaide Local Health Network, 28 Woodville Road, Woodville South, Adelaide, SA, Australia. Tel: +61-8-8222 6740; Fax: +61-8-8222 6042 Email: john.beltrame@adelaide.edu.au

Received: July 28, 2023 Accepted: September 07, 2023 Published: September 14, 2023

Abstract

Background: The Coronary Slow Flow Phenomenon (CSFP) is a coronary microvascular disorder that presents with recurrent angina for which there is no available effective therapy. The vasoconstrictor Endothelin-1 (ET-1) has been implicated in the pathophysiology of the CSFP. This study aimed to determine whether the ET-1 antagonist (Bosentan) is effective in ameliorating angina symptoms in CSFP patients.

Methods: Using a randomised, double-blind, placebo-controlled, cross-over trial methodology, the anti-anginal efficacy of Bosentan 125mg bd was assessed in 23 patients with symptomatic CSFP. Study endpoints were measured following 4 weeks of therapy with the primary endpoint being angina frequency. Secondary endpoints included (1) clinical measures of angina and health status, (2) endothelial function parameters, and (3) inflammatory and oxidative stress markers.

Results: There was no statistically significant reduction in angina frequency with Bosentan therapy compared to placebo (median angina episodes/4weeks [25-75% IQ]: Bosentan 14[5,36] vs Placebo 25[9,55]) although a trend for improvement in symptoms during the Bosentan phase was observed. Furthermore, Bosentan did not impact on any of the secondary endpoints compared with placebo.

Conclusion: Bosentan did not have a major impact on angina frequency in patients with the CSFP. A small anti-anginal effect cannot be excluded and requires a larger trial.

Introduction

The Coronary Slow Flow Phenomenon (CSFP) is a coronary microvascular disorder characterised by delayed distal vessel opacification in the absence of obstructive coronary artery vessel disease [1]. Clinically distinct from other microvascular disorders such as cardiac syndrome X [2], the mechanisms responsible for the microvascular dysfunction remain elusive, although a plethora of pathophysiological investigations have reported abnormalities in endothelial function [3-5], inflammatory markers [6,7], oxidative stress markers [5,8], and platelets [9]. However some studies have been unable to replicate these observed abnormalities [10-12] there by making it difficult to identify specific therapeutic targets. One line of pathophysiological investigations in the CSFP with a potential therapeutic target is Endothelin-1 (ET-1). This potent endogenous vasoconstrictor peptide is a combined Et-A and Et-B receptor antagonist and has been implicated in the pathogenesis of the CSFP based upon the following observations: (a) intracoronary ET-1 administration mimics the angiographic appearance of the CSFP in canine [13] and rabbit [14] studies; (b) ET-1 levels have been shown to be elevated in patients with the CSFP [15]; (c) intravenous ET-1 infusion into healthy individuals produces a fall in coronary sinus oxygen saturation [16] mimicking those observed in the CSFP [17]; and (d) a selective hyper-responsiveness to ET-1 in isolated subcutaneous microvessels of patients with the CSFP [18]. To provide confirmatory evidence that ET-1 plays a major role in the CSFP, the influence of a dual endothelin-1 receptor antagonist on clinical and pathophysiological measures should be investigated. Accordingly, the primary objective of this study was to examine if ET-1 blockade with Bosentan 125mg bd would influence the angina frequency in patients with symptomatic CSFP. Supporting secondary objectives examined the effects of Bosentan therapy on other clinical and pathophysiological endpoints including: (i) other clinical markers (prolonged angina episodes, sublingual nitrate consumption, health status measures); (ii) endothelial function assessed by pulse wave analysis and Asymmetric Dimethylarginine (ADMA) plasma levels; (iii) inflammation status as evaluated by High-Densitivity C-Reactive Protein (hsCRP) and Myeloperoxidase (MPO); and (iv) oxidative stress determined by Malondialdehyde (MDA) and homocysteine biomarkers.

Methods

To achieve the above objectives, the study employed a randomised, double-blind, placebo-controlled, cross-over study design involving symptomatic CSFP patients. The study was approved by the local Hospital and University Human Research Committees. The active study medication utilised was Bosentan monohydrate 125mg tablet (Actelion Pharmaceuticals Australia Pty Ltd), administered orally twice daily. A corresponding placebo that was indistinguishable in appearance to the active drug was also supplied by the manufacturers.

Study Patients

Both recently diagnosed and patients with established CSFP were recruited if they fulfilled the following inclusion criteria: (1) stable recurrent angina episodes =3 times/week; and (2) angiographic evidence of the CSFP as defined by (a) requiring =3 beats to opacify pre-specified branch points in the distal vasculature of any of the three major epicardial coronary arteries (i.e. equivalent of TIMI-2 flow), and (b) the absence of obstructive CAD (i.e. <50% in any epicardial coronary artery). Exclusion criteria were based upon contra-indications to Bosentan therapy and included: (a) elevated plasma hepatic transaminases twice the upper normal limit; (b) concurrent use of cyclosporine A or oral contraceptives; (c) anaemia (haemoglobin =100g/L); and (d) pregnancy.

Study Protocol

Following an initial visit when clinical history and informed consent were obtained, patients were randomised to either twice-daily Bosentan (125mg) or matching placebo treatment using a computer-generated algorithm, with the sequence known only to the hospital clinical study pharmacist who had no contact with the patients. After two weeks of Bosentan/placebo therapy, patients were assessed for any adverse events and withdrawn from the study if transaminases demonstrated a =5-fold change from baseline. Patients continued with the assigned therapy for a further two weeks and at 4-weeks of therapy were clinically re-evaluated, venesection performed for biomarker determination and endothelial function assessed. Subsequently, patients commenced a one-week washout period and then crossed-over to the alternative Bosentan/placebo therapy with the above study protocol repeated. Throughout the study, the patient’s maintenance medications were maintained constant.

Study Endpoints

The primary endpoint was the frequency of total angina episodes recorded during each 4 weeks of treatment. Secondary endpoints included (1) Clinical measures of angina and health status, (2) Endothelial function parameters, and (3) Inflammatory and oxidative stress markers.

Clinical Measures

These included an angina diary and health status measures. Throughout the study period, patients were asked to maintain an angina diary, documenting the frequency of angina, duration of episodes (prolonged anginal episodes defined as >20 minutes) and sublingual nitrate consumption. During each visit, clinical observations (pulse and blood pressure) were recorded, the angina diary reviewed and any changes in medication or occurrence of adverse events documented. In addition to these clinical measures, in the last week of each treatment phase an electrocardiograph was performed and patients completed both a generic (Short Form- 36; SF-36) and disease-specific (Seattle Angina Questionnaire; SAQ) health status questionnaire. These instruments have been previously validated for patients with angina and provide important insights into the impact on health status [19,20]. Patient medication compliance was also assessed at the end of each phase by pill count.

Endothelial Function Parameters

For the assessment of endothelial function both a physiological functional test and a plasma biomarker were performed. The functional test employed pulse wave analysis using the method described and validated by Wilkinson et al [21]. ADMA is an endogenous inhibitor of nitric oxide and has been used as an indirect measure of endothelial function.

Pulse Wave Analysis for Endothelial Function

This technique involved placing a pressure transducer probe (Sphygmocor Software, Atcor Medical, Sydney, Australia) over the maximal arterial pulsation in the radial artery, so that adequate applanation of the vessel could be achieved to derive the radial pressure waveform. Aortic pressure waveforms were obtained from the peripheral pulse waveform utilising a validated generalised transfer function [22] in the integrated software. Systemic endothelial function was assessed through analysis of changes in aortic pressure waveforms and quantitative measurement of Augmentation Index (AIx) following serial administration of endothelium-independent (sublingual Glyceryl Trinitrate [GTN] 50mcg) and endothelium-dependent (salbutamol inhaler 400mcg) vasodilators. The AIx represented the difference in values between ventricular ejection (first systolic peak, P1) and systolic peak pressure (second systolic peak, P2), expressed as a percentage of pulse pressure and corrected for heart rate.

The following procedure was adopted for the applanation tonometry assessment of endothelial function. On arrival, subjects rested for 30 minutes in a supine position during which serial baseline blood pressure and pulse rate were non-invasively measured from the left brachial artery. Pulse wave analysis of the right radial artery waveforms was also measured during this resting period to ascertain the baseline AIx values. After establishing baseline parameters, 50μg GTN was administered sublingually and recordings of blood pressure, pulse and AIx values occurred at 3, 5, 10, 15 and 20-minute intervals. The protocol was repeated for 400μg salbutamol administration after AIx readings had returned to baseline values. For each time point, three AIx values were obtained and subsequently averaged. The arterial responses to GTN and salbutamol administration were defined as the maximum change in AIx over the total time interval compared to baseline as previously described [21].

Asymmetric Dimethylarginine Assay: Plasma ADMA, symmetric dimethylarginine (SDMA) and arginine concentrations were determined using a modified version of a previously described protocol [23]. Briefly, 150μL of plasma was combined with internal standard N- monomethylarginine [NMMA (5μg/mL)] and extracted on pre-conditioned Bond Elut Strong Cation Exchange (Varian) cartridges loaded on an automated solid phase extraction system (Gilson Aspec GX-274). Arginine and the olimethylargines were eluted and prepared for HPLC as per the published protocol. ADMA and SDMA were quantified by HPLC as described previously, from a 20μL injection volume. Arginine was quantified by injecting 2.5μL of mixture onto the HPLC.

Inflammatory and Oxidative Stress Markers

The inflammatory biomarkers assessed included hsCRP and MPO, while oxidative stress markers included MDA and homocysteine. Fasting plasma samples were obtained for these assays with homocysteine and hsCRP concentrations assessed at the time of sampling and the remaining plasma was stored at -80°C for subsequent batch analysis of ADMA, MPO, and MDA.

Homocysteine and high-sensitivity CRP assays. Plasma homocysteine was assessed using a well-established and validated chemiluminescence immunoassay [24]. Concentrations of plasma hsCRP were determined using a validated immuno-turbidimetric assay (Olympus OSR6199 Analyzer, Australia) [25].

Myeloperoxidase assay. MPO concentration was determined using the Zen™ Myeloperoxidase 96-well ELISA Kit (Invitrogen, Australia) as per the specified protocol. All antibody solutions, standards and patient samples were diluted with phosphate buffered saline/0.1%BSA. Briefly, individual ELISA wells: 1) were hydrated with PBS + 0.001% Tween (PBST) and incubated with 500ng/mL (100μL) solutions of mouse anti-MPO primary capture antibody; 2) underwent sequential washing steps with PBST; 3) incubated with standards, samples (100μL) and 1.0μg/mL of secondary rabbit anti-MPO capture antibody; 4) washed and incubated with 100μL goat anti-rabbit HRP conjugate; 5) and incubated with 100μL of Amplex UltraRed reagent mix (30 min) followed by addition of Amplex® stop solution. Fluorescence was measured using filters at λexcitation = 530nm and λemission = 590nm.

Malondialdehyde assay. Plasma MDA levels were determined based upon a modified protocol that involved fluorescent detection of an extracted MDA-TBA acid adduct [26]. Briefly, plasma samples were thawed and underwent protein precipitation (2.3M perchloric acid) and lipid removal (200μL chloroform) protocols, respectively. For the adduct reaction to occur, 200μL of supernatant was aliquoted into conical glass tubes containing 0.15M H3PO4 (750μL), 42mM TBA (250μL) and made up to 1.5mL with MilliQ water. Samples were boiled for 60 minutes at 100°C and cooled on ice thereafter. Extraction of the MDA adduct was performed by addition of 70:30 chloroform:methanol (vol/vol) and the top aqueous layer was collected for detection of MDA fluorescence at λexcitation = 530nm, λemission = 547nm.

Data Analysis

The study was undertaken using double-blind methodology and all analyses were conducted blinded to the study treatment. The clinical angina parameters were analysed using cross-over trial methodology with comparison between patients and respect to treatment order [27]. Large differences in angina frequency between phases for individual patients were observed and therefore the data was log transformed and non-parametric (Mann Whitney U) tests used for comparison. The continuous, normally distributed health status questionnaires, endothelial function and biomarker assay endpoints were expressed as mean ± SEM and analysed by unpaired Student t-tests. Statistical values with a=0.05 were considered as significant.

The study sample size was calculated based upon the primary endpoint of total angina frequency. From a previous study examining the anti-anginal efficacy of mibefradil in the CSFP [28], the total angina frequency during placebo therapy was 28±31 episodes/month with 22 less episodes on active therapy. Thus, to detect a difference of 22 episodes or more with Bosentan therapy, a minimum of 22 patients would be required for 90% power at a=0.05 level.

Results

Over a thirty-month period, 26 patients were recruited into the study with 3 patients subsequently withdrawing, before completion of the study protocol. Two of these patients were withdrawn because of concerns with the study drug (mild elevations in hepatic transaminases and potential drug interactions with medications). A further patient was withdrawn due to non-compliance with study visits. Amongst the patients completing the study protocol, there were no major adverse effects and medication compliance was over 95% on pill count.

Patient Characteristics

The recruited patients were more often male, with over half having at least 2 atherosclerotic risk factors (Table 1). All of the patients were on conventional anti-anginal therapy, including long-acting nitrates (68%), calcium-channel blockers (68%) and/or beta-blockers (14%); yet their mean angina frequency at baseline was 15±3 episodes/week.