A Tale of Two Endothelin Receptor Antagonists in Eisenmenger Syndrome: It was the best of times, it was the worst of times

In 1897, Dr. Victor Eisenmenger described a 32-year-old man with a history of cyanosis beginning as a child. After death from massive hemoptysis, his autopsy showed a large ventricular septal defect. In the 1950s, Paul Wood began to describe the hemodynamic effects of chronic volume and pressure overload leading to the reversal of a central shunt due to irreversible pulmonary vascular obstructive disease.¹ Wood also recognized that Eisenmenger syndrome (ES) develops in patients with large unrestrictive shunt defects, such as a large ventricular septal defect, large patent ductus arteriosus, or atrioventricular septal defect.

In the subsequent decades, conservative medical management has led to improved survival. The prevalence of ES is also decreasing as surveillance is more standard and surgical repair is more available. In a Nordic cohort, the incidence of ES decreased from 2.5/million inhabitants/year in 1977 to 0.2/million inhabitants/year in 2012.² Further, sub-classification of congenital heart defects using the Nice World Symposium of Pulmonary Hypertension criteria has revealed disparate survival between CHD groups. In general patients with ES live longer than those patients with fully repaired shunts and postoperative PAH.³ However, these ES patients remain at risk for progressive disease and death. Treatment with targeted PAH therapy in ES was concerning for lowering systemic vascular resistance relative to pulmonary vascular resistance leading to increased right to left shunting. One such target is the endothelin pathway.

Endothelin is a potent vaso-active peptide discovered in 1988 by Yanagisawa.⁴ The effects of endothelin are mediated by ET_A and ET_B receptors. ET_A receptors are located on vascular smooth muscle cells and mediate vasoconstriction. ET_B receptors are on both endothelial cells and vascular smooth muscle cells. ET_B receptors on endothelial cells may lead to clearance of endothelin and production of nitric oxide whereas ET_B receptors on smooth muscle cells also mediate vasoconstriction. Several targeted therapies have been developed toward the endothelin receptors. In 2001, bosentan, a dual endothelin receptor antagonist (ERA), was approved by the FDA as a treatment for Group 1 Pulmonary Arterial Hypertension (PAH).

The Bosentan Randomized Trial of Endothelin Antagonist Therapy-5 (BREATHE-5) study was the first randomized, double-blind, placebo-controlled, multi-center study completed in patients with ES, with results reported in 2006.⁵ Fifty-four patients with ES were randomized 2:1 to bosentan or placebo for 16 weeks. Systemic pulse oximetry (SPO2) did not worsen with bosentan. Bosentan reduced pulmonary vascular resistance index, mean pulmonary arterial pressure, and improved exercise capacity. Subsequent observational research suggested that targeted PAH therapy improved survival in ES.⁶

A newer generation of ERA, macitentan, has fewer complications, requires less liver function monitoring, and is dosed daily. In this issue of Circulation, Gatzoulis and colleagues evaluated macitentan in patients with ES in the MAESTRO trial (MAcitentan in Eisenmenger Syndrome To RestOre exercise capacity).⁷ MAESTRO is only the second randomized, double-blind, placebo-controlled, multi-center study to be completed in patients with ES. Patients with ES who were over 12 years of age and were in functional class II-IV were randomized to placebo or macitentan for 16 weeks. The primary endpoint was the change from baseline in 6-minute walk distance (6-MWD). Secondary endpoints included the change from baseline at week 16 in functional class. Other exploratory endpoints were included such as NT-proBNP change and hemodynamic evaluation in a subgroup of patients. MAESTRO did not meet its primary endpoint nor many secondary endpoints. The mean change in 6-MWD from baseline was an increase of 18.3 meters (m) in the macitentan group and 19.7 m in the placebo group. The 6-MWD least-squares mean difference at week 16 was −4.7 m between macitentan and placebo (p=0.612, intention-to-treat (ITT)). NT-proBNP improved in the main group and pulmonary vascular resistance index and 6-MWD improved in the hemodynamic substudy.

The MAESTRO trial design was different from the BREATHE-5 study. An expanded and more heterogeneous population of patients was included in MAESTRO: patients with more complex forms of ES, patients with Down syndrome, patients with WHO functional class II-IV and patients with ES treated with type-5 phosphodiesterase inhibitors were included in MAESTRO. In BREATHE-5, only simple forms of ES, those in WHO functional class III, and those without Down syndrome or pre-existing treatment were enrolled. In MAESTRO, 226 patients, randomized 1:1 were enrolled. The majority were functional class II and 27% were receiving background phosphodiesterase type 5 therapy

Comparison between BREATHE-5 and MAESTRO is challenging. The endpoints in these two trials were very different. The BREATHE-5 study first measured pulse oximetry as the primary safety endpoint and pulmonary vascular resistance as the primary efficacy endpoint. MAESTRO evaluated 6-MWD as the primary endpoint and WHO functional class and NT-proBNP as secondary endpoints. It is likely that the inclusion criteria contributed to the different outcome of these trials. In spite of multiple post hoc analyses suggesting that the inclusion of a broader population with ES did not contribute to the results, an expanded population of patients with more complex forms of ES, broader WHO functional class, and use of pre-existing therapy when added together suggests a different patient cohort. Further recent studies suggest the patients with Down syndrome may not respond as well to targeted PAH therapy.⁸

It is impossible to predict whether macitentan treatment with the inclusion criteria in BREATHE-5 would have revealed a positive study. Both agents are dual endothelin receptor antagonists and macitentan may have enhanced tissue specificity. Unfortunately, the differences in design cannot be overcome using post-hoc analyses. A study of macitentan in a more homogeneous group with similar endpoints will be important to determine if macitentan is effective in treating patients with ES. This study does provide interesting insight into clinical trial design. It is known that patients with worse functional class are more likely to improve with PAH targeted therapy. While improvement in 6-MWD has been found to be much more modest in patients on background therapy, stronger evidence for a benefit of combination PAH therapy has subsequently been shown in multiple long-term studies with composite primary endpoints focused on clinical worsening, suggesting further limitation of the use of 6-MWD in patients on multiple background therapies in clinical trials. Some studies have suggested that patients with Down syndrome may not respond as well to targeted PAH therapy as seen in the Sildenafil in Treatment-Naïve Children, Aged 1–17 Years, With Pulmonary Arterial Hypertension (STARTS-1) study.

The MAESTRO trial brings into question the optimal trial design of targeted PAH therapy in patients with long-standing PAH. The exaggerated placebo effect in the trial may be overcome by a longer treatment time. Patients in the placebo cohort of MAESTRO had worsening pulmonary vascular resistance index over the 16 weeks suggesting that over time these ES patients without targeted PAH therapy would have more clinical worsening events.

So, the question remains: two acceptably run randomized studies with very different conclusions on the use of dual ERA drugs in ES. Little is mentioned on the difference in ET_A and ET_B affinity between these two drugs. Could it be that the much greater affinity for ET_A over ET_B receptors of macitentan is revealing our ignorance on the importance of the ratio on the efficacy of pulmonary vasodilation? The MAESTRO study is certainly better powered and employs the 6-MWD which is surely a more widely available test to patients and clinicians. We would certainly not throw the baby out with the bathwater and dismiss macitentan, but we would advocate for a longer trial duration and didactic invasive evaluation of hemodynamics in a larger cohort of patients in any future study involving macitentan in this disease. We also need to evaluate the significance of the marked difference in ET_A and ET_B and its effect on pulmonary vasodilation.

The comparison of the two studies (BREATHE-5 and MAESTRO) strongly suggest that bosentan should continue to be the first choice dual ERA agent in patients with ES.