Prince Henrys Institute of Medical Research Clayton, Victoria, Australia 3168
Address all correspondence and requests for reprints to: John W. Funder, M.D., Ph.D., Prince Henrys Institute of Medical Research, P.O. Box 5152, Clayton, Victoria, Australia 3168. Email: John.Funder{at}med monash.edu.au.
As noted elsewhere in this issue of JCEM, aldosterone was isolated and characterized in 1953 (1). For the first decade of its life, aldosterone was the focus of considerable interest in the then much less differentiated endocrine/renal/circulatory community, with the mineralocorticoid antagonist spironolactone first seeing the light of day in the early 1960s. The next 30 yr represented, for aldosterone, the Dark Ages, with about 40 people worldwidecave dwellers, keepers of the flamemeeting very quietly each year for a day and a half before the annual Endocrine Society meetings, a meeting supported by Searle, Monsanto, Pharmacia, and now (hopefully) Pfizer.
Over the past decade things have changed. The first stirrings of a Mr. Hyde face of aldosterone came from the studies of Karl Weber, a cardiologist, who showed that inappropriate aldosterone-for-salt status in rats produced cardiac hypertrophy and global cardiac fibrosis (2). These and subsequent studies from other laboratories prompted G. D. Searle \|[amp ]\| Co. to launch the Randomized Aldactone Evaluation Study (RALES), which was prematurely terminated in August 1998 and published in September 1999 (3). Briefly, RALES showed that the addition of a very modest dose of spironolactone (Aldactone) to current standard of care treatment for severe (New York Heart Association Class III) heart failure provided a 30% improvement in survival, and a 35% improvement in morbidity/hospitalization.
Apart from morbidity and mortality, only two other indices separated the groups. First, despite the low dose used (average, 26 mg/d), the patients on spironolactone consistently had plasma [K+] 0.20.3 meq/liter higher than the group receiving placebo [plus angiotensin-converting enzyme (ACE) inhibitors, diuretics, etc.]. Second, I suspect to the relief of the sponsors, the spironolactone group showed an incidence of gynecomastia of 10%, compared with 1% in the control group.
In the intervening 5 yr since RALES was terminated there has been a healthy spate of studies, basic and clinical, using a much more selective mineralocorticoid receptor blocker (eplerenone) to explore this new biology of aldosterone. The methodology is quintessentially endocrineablate the hormone, not at the level of secretion as was classically the case, but at the level of action. The basic studies have increasingly focused on the vascular inflammatory response to aldosterone and its prevention by eplerenone. It has been known for many years that the vascular wall, and in particular the vascular smooth muscle cell, was an epithelial-type aldosterone target tissue, expressing not only mineralocorticoid receptors but also the aldosterone selectivity-conferring enzyme 11ß-hydroxysteroid dehydrogenase (4). Recently, in vitro (5) and in vivo (6) studies on human blood vessels have confirmed their status as aldosterone target tissues, perhaps not surprising in the context of an integrated renin-angiotensin-aldosterone response to decreased blood volume.
In rat studies from various laboratories, most notably by Rocha et al. (7, 8), aldosterone administration to animals drinking 0.9% saline has been shown to provoke a spectacular vascular and perivascular inflammatory response. This inflammatory response is marked by progressive increases in a range of cytokines and markers over 14 wk; is seen in other high salt/aldosterone models (angiotensin II-infused rats or stroke-prone spontaneously hypertensive rats), on saline to drink; affects cerebral, renal, and coronary vessels and their surrounds; and is blocked by eplerenone. Importantly, the inflammatory response can be blocked by eplerenone in rats given carbenoxolone to block 11ßHSD2 (9), or in pigs undergoing balloon angioplasty (10). In both these situations aldosterone levels are normal or low, suggesting that the mineralocorticoid receptors in the vessel wall may be inappropriately activated by glucocorticoids in the context of enzyme blockade or tissue damage. This, in turn, suggests that the culprit in a mechanistic sense is inappropriate activation of the mineralocorticoid receptor, rather than aldosterone per se.
These basic studies were accompanied by a variety of clinical trials of eplerenone in hypertension, some of which have been published, others are in press, and still others have been presented at meetings. In brief, eplerenone is an effective blood pressure-lowering agent in essential hypertension across a range of populations. It is not unexpectedly better than ACE inhibitors in African-Americans (11) and as potent as calcium channel blockade in isolated systolic hypertension (12). With regard to end-organ protection, eplerenone seems comparable with ACE inhibition in terms of left ventricular hypertrophy (13) and marginally better in terms of proteinuria (14); in both cases, the combination of eplerenone and enalapril adds little to the fall in blood pressure, but significantly to end-organ protection, compared with either agent alone. On the basis of the safety and efficacy studies, the use of eplerenone for essential hypertension was approved in late September 2002, with prescribing guidelines and a package insert that has been characterized as unwarrantedly restrictive (15), and to date the compound has not been released for sale.
Back to heart failure. If RALES was the road to Damascus, the recent publication by Pitt et al. (16) is the letter from Ephesus. The Eplerenone Post-AMI Heart Failure Efficacy and Survival Study (EPHESUS) randomized 6642 patients in 37 countries who had evidence of heart failure 314 d after myocardial infarction. As in RALES, the mineralocorticoid receptor blocker was in addition to existing therapy, for the first 4 wk at 25 mg and subsequently at 50 mg as indicated, in both instances as a single daily dose. The final average dose achieved was 43 mg/d, very much in line with the 26 mg/d spironolactone in RALES; in vivo eplerenone seems to have 5070% the antialdosterone potency of spironolactone, but two orders or greater of magnitude selectivity for mineralocorticoid receptors over androgen receptors and more than one order of magnitude over progesterone receptors, compared with spironolactone. Patients were enrolled December 27, 1999 to December 31, 2001, and the closing date was August 30, 2002 when 1032 deaths had been recorded. The co-primary end points for the study were total mortality and cardiovascular (CV) mortality/CV hospitalization, and major secondary end points CV mortality and total mortality/total hospitalization. On December 23, 2002, a brief press release announced that the trial had shown eplerenone to have been positive in terms of the two primary end points, and that a more detailed analysis would be presented at the American College of Cardiology meeting in late March 2003, and published simultaneously (16).
In terms of primary end points, patients receiving eplerenone showed a relative risk (RR) of 0.85 [95% confidence interval (CI), 0.75, 0.96; P = 0.008] of death compared with controls and a RR of 0.87 (95% CI, 0.790.95; P = 0.002) of CV mortality/hospitalization. The secondary end point of CV mortality was also positive (RR, 0.83; 95% CI, 0.72, 0.94; P = 0.005), as was total mortality/hospitalization (RR, 0.92; 95% CI, 0.860.98; P = 0.02). Within CV mortality the only significant difference between causes was for sudden cardiac death (RR, 0.79; 95% CI, 0.64, 0.91; P = 0.03), with acute myocardial infarct (RR, 0.82) and heart failure (RR, 0.80) not significantly different between groups, reflecting the lower numbers per group. The only significant difference between eplerenone and placebo in terms of CV hospitalization was for heart failure (RR, 0.85; 95% CI, 0.74, 0.99; P = 0.03). Eplerenone-treated patients showed a lower incidence of respiratory, metabolic, and nutritional disorders, and predictably of hypokalemia; equally predictably, they showed a higher incidence of serious hyperkalemia ([K+], >6 meq/liter: 5.5 vs. 3.9%, P = 0.002). No numerical or statistical difference was seen between groups for gynecomastia, impotence, or female disorders (sic).
So, what does it all mean? The improvement in mortality and morbidity in EPHESUS was substantial, but only half that in RALES: why? There are several differences between the two studies that may account, in large part or in whole, for the differences. First, the RALES patients were sicker, with an average left ventricular ejection fraction of 25%, vs. 33% for EPHESUS; the probability is that this figure would rise in the latter group with revascularization and emergence from the state of so-called cardiac stunning. Second, the baseline treatment regimen in the two studies was markedly different; in EPHESUS 75% of the patients were on a ß-blocker compared with 11% in RALES, for instance; in EPHESUS 88% were also on aspirin, 87% on an ACE inhibitor, 60% on diuretics, and 47% on statins. Finally, in RALES the average age was 69 yr compared with 64 yr in EPHESUS. This combination of age, standard of care, and ejection fraction seems reasonable to explain the apparent difference in outcome between the two studies. EPHESUS provided a very respectable extent of improvement in the patient cohort in which it was tested, and there is no reason to believe it would be inferior to spironolactone in older, sicker patients with a lesser standard of care.
Where to from now? The remainder of this year should see presentation (and hopefully publication) of a number of the EPHESUS-derived substudies, of which perhaps three promise to be of particular interest. The first of these is on the coagulation and fibrinolytic status of eplerenone-treated patients vs. controls; aldosterone-induced changes in plasminogen activator inhibitor-I levels, for example, have been claimed to contribute to the vascular phenotype seen (17), although its role is currently disputed (18). Second, given the recent studies on the CV implications of elevated C-reactive protein, levels of the inflammatory marker in the two groups will be of particular interest. Third, a full 32% of the patients in EPHESUS are diabetic, and analysis of their response, between treatment and within treatment, is likely to be instructive in terms both of patient care and prescribing guidelines. Regulatory agencies may take an additional level of comfort from a large outcome trial such as EPHESUS, and be able to refine the existing prescribing information with an additional degree of confidence.
So, why should a practicing endocrinologist read about a heart failure trial? The answer, in brief, is that aldosterone is a hormone and endocrinologists need to know about what hormones are doing. With the advent of the steroid receptor subfamily of nuclear transcription factors, reproductive endocrinologists or thyroidologists had to acknowledge the mineralocorticoid receptor as a shadowy, somehow conflicted relative; their most common thought, to the extent that they thought about aldosterone, was to give thanks that they were not involved. For most cardiologists, aldosterone was terra incognita, commonly if unhelpfully dismissed as one of a heterogeneous group of neurohumoral factors. The keepers of the flame during the Dark Ages of aldosterone have been endocrinologists, clinical and basic, to an overwhelming degree; it is no accident that Dr. Bertram Pitt, an eminent cardiologist, chairs the steering committee for EPHESUS, but Dr. Gordon Williams the scientific advisory board.
In contrast to heart failure, for which cardiologists are clearly the key opinion leaders, hypertension draws on a much more diverse range of expertisecardiology, nephrology, clinical pharmacology, and endocrinology. Although the two big outcome trials on mineralocorticoid receptor blockade have been in heart failure, for the majority of heart failure patients in most countries hypertension is a major antecedent cause; in EPHESUS, over 60% of patients had a history of established hypertension. Whereas mineralocorticoid receptor blockade can halt the progression of heart failure both experimentally (19) and in a proportion of patients (20), it can largely prevent the effects on vascular inflammation in animal models of hypertension and improve (rather than just halt the progression of) indices of cardiac and renal status in hypertensive patients. Logically, then, the prime candidate group for eplerenone therapy would seem to be young and middle-aged men and women with elevated blood pressure, to lower blood pressurewhich a number of current antihypertensives can do equally wellbut most importantly to protect the vasculature and prevent end-organ damage. The primary use of eplerenone should be for primary aldosteronism, which is now recognized to contribute 10% or greater of essential hypertension (21), and in the remaining essential hypertensive population, particularly African-Americans, patients with isolated systolic hypertension and in so-called resistant hypertensives, those whose blood pressure is not controlled on three or more other agents.
All this may sound a bit evangelical, in the footsteps of St. Paul, the author of the Letter to the Ephesians. One bright spot of the late Adlai Stevensons doomed presidential campaign, again almost 50 yr ago, was his response to a question about the latter day evangelist Norman Vincent Peale. "As evangelists go," Stevenson replied, "I find Paul appealing, and Peale appalling." Nobody that smart could be expected to survive in politics; but many of us in this high-salt, high-stress world will survive our hypertension better with some degree of specific mineralocorticoid receptor blockade. Here endeth the first lesson, text the Letter from EPHESUS.
Footnotes
Abbreviations: ACE, Angiotensin-converting enzyme; CI, confidence interval; CV, cardiovascular; EPHESUS, Eplerenone Post-AMI Heart Failure Efficacy and Survival Study; RALES, Randomized Aldactone Evaluation Study; RR, relative risk.
Received February 21, 2003.
Accepted February 28, 2003.
References
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |