(Circulation. 1999;99:198-200.)
© 1999 American Heart Association, Inc.
Editorials |
Good News for Experimental Concept but Bad News for Clinically Effective Therapy
From the Department of Pathology and the Division of Cardiology, Department of Medicine, Duke University Medical Center, Durham, NC.
Correspondence to Robert M. Califf, MD, Duke Clinical Research Institute, PO Box 17969, Durham, NC 27715.
Key Words: Editorials • drugs • trials • myocardial infarction • mibefradil
Although it is widely accepted from experimental studies that timely reperfusion limits myocardial infarct size, reperfusion by itself may not achieve the greatest possible effect. In a clinical setting, reperfusion, whether by thrombolytic therapy or emergency coronary angioplasty, never can be achieved instantaneously. Thus, adjunctive therapy, which could either slow ischemic metabolism and cellular injury pending successful reperfusion or protect myocytes against undesired, potentially lethal effects of reperfusion ("lethal reperfusion injury"), should have added clinical benefit for limiting infarct size. Moreover, it would be possible to treat patients at high risk of infarction prophylactically if a safe and effective cardioprotective agent could be developed.
Several endogenous mechanisms or exogenous interventions are known to both slow the rate of ischemic metabolism and delay the onset of lethal myocyte injury. For example, the speed at which ischemic cell injury occurs is markedly temperature-dependent. Hypothermia substantially slows ischemic metabolism and the onset of lethal myocyte injury.1 This approach is commonly used in cardiac surgical procedures requiring a cessation of myocardial blood flow to slow ischemic injury and thereby prolong the tolerable period of ischemia.
Studies in several animal species have shown that exposing
myocardium to 
1 short (5- to 15-minute) episodes of
ischemia ("ischemic preconditioning") slows energy
metabolism and delays the onset of myocardial necrosis in a
subsequent prolonged period of ischemia.2 3
Although the mechanism is not completely established, there is
considerable evidence that the effect requires activation of 1
signaling pathways, the end effect of which is enhanced opening of
ATP-dependent potassium (KATP)
channels.3
Calcium channel blockers also generally have had cardioprotective, infarct-limiting effects in studies based on an animal model in which infarcts were induced by relatively short periods of ischemia (40 to 90 minutes) followed by reperfusion.4 5 6 Thus, calcium antagonists delay but do not prevent lethal ischemic injury in experimental models.
Calcium channel antagonists are now frequently used in the treatment of clinical myocardial ischemia (angina pectoris). However, assessment of their clinical benefit has been controversial, and the value of calcium antagonists in patients with acute myocardial infarction is even less clear. Earlier clinical trials testing the effect of calcium antagonists, including verapamil, diltiazem, and nifedipine, in the setting of acute myocardial infarction failed to demonstrate a beneficial effect on patient survival.7 8 9 One hypothesized reason for the lack of overall clinical benefit of calcium antagonists has been that the classic agents have substantial negative inotropic effects, a serious drawback especially when myocardial function may already be depressed by the evolving myocardial infarct.10 11 Because of these drawbacks of earlier calcium antagonists, a calcium antagonist with little or no intrinsic negative inotropic activity could be an attractive alternative for clinical application if it showed equivalent infarct-limiting efficacy in experimental models.
Mibefradil (Ro 40-5967) is a calcium channel antagonist that, in experimental animals12 13 as well as clinical studies (involving patients with angina pectoris),14 was found to lack the substantial negative inotropic effect of verapamil. This attribute suggested that mibefradil might offer unique advantages over previous calcium channel blockers in clinical ischemic heart disease. An initial study of short-term ischemia and reperfusion in dogs showed that mibefradil limited myocardial infarct size to a degree similar to that achieved by verapamil in the same study.15 Mibefradil also was reported to limit myocardial infarct size in rats.16
In the present issue of Circulation, Schulz et al17 report the results of a comparison of the effects of 3 calcium antagonists on myocardial infarct size in pigs. These authors observed that mibefradil limited infarct size independently of any hemodynamic effect, whereas amlodipine did not limit infarct size. Verapamil limited infarct size, but its effect was abolished when the associated hemodynamic effects of verapamil were eliminated. In addition, Schulz et al report that the infarct-limiting effect of mibefradil was abolished by blockade of KATP channels, which raises the possibility of a common mechanism between mibefradil and ischemic preconditioning.
Despite the promise of molecules such as mibefradil based on animal studies, clinical efficacy must be proven through clinical trials and safety must be validated through postmarketing surveillance. The standard for clinical efficacy is prolongation of life, reduction in symptoms, or prevention of major events in patient groups representing the population of intended use. To evaluate one of these end points without considering the others would be a mistake. Incorrect therapeutic application has occurred in several previous drugs because of insufficient information on all of these aspects from the relevant population of patients.18 19 In the specific case of mibefradil, an unfortunate series of events has led to its withdrawal from the market and consequent elimination of its promising experimental efficacy in limiting infarct size.
In a series of elegant and carefully controlled phase I, II, and III clinical trials, mibefradil was developed as a therapy to reduce angina and treat hypertension. The angina trials recruited patients without complex comorbidities. Using serial treadmill exercise testing, patients given mibefradil were able to exercise longer than patients given placebo. Mibefradil-treated patients also had fewer episodes of angina. The hypertension trials recruited patients with simple essential hypertension and proved that mibefradil can lower blood pressure and that its antihypertensive efficacy was comparable to or better than that of other currently available therapies. "Hard" clinical events such as death, myocardial infarction, or new-onset heart failure occurred at a very low rate in these studies because of the low-risk nature of the population and the short duration of the studies. The drug was noted to cause an abnormality of the QT interval; debate ensued over whether this abnormality represented true QT prolongation or simply a normal QT variant. Comparison with other calcium channel blocking agents showed similar QT effects of diltiazem and verapamil. Only 1 small study was completed in patients with heart failure at the time FDA approval was sought, and a nonsignificant excess of deaths was seen in mibefradil- versus placebo-treated patients.
The company proposing to market mibefradil, to its credit, had commissioned a properly designed mortality trial in heart failure (MACH-1) that was adequately sized to detect a major negative impact of the compound, but at the time of application for marketing, the results of this trial were not available. Pharmacokinetic studies had demonstrated that the P450 enzyme system was involved in its metabolism, and interaction studies had demonstrated a substantial increase in concentrations of both mibefradil and at least 1 statin when both were given to human subjects.
On the basis of the above information, the Cardiorenal Panel of the FDA recommended approval of mibefradil for marketing, and the agent was marketed in June 1997. Clinicians were advised to use caution when combining mibefradil with other compounds metabolized by the P450 system. Within 4 months, >400 000 patients were taking mibefradil for either hypertension or angina.
Shortly after marketing, however, reports began to surface of toxicity, including torsades de pointes, clearly related to interactions between mibefradil and other therapeutic drugs. Apparently, no significant toxicity could be attributed to mibefradil alone; it always occurred in conjunction with another major complicating factor, almost always another potential culprit drug. The MACH-1 trial unfortunately found a nonsignificant trend toward a higher mortality with mibefradil than placebo (27% versus 24.6%). The adverse outcome reports became numerous despite major efforts to better educate physicians about the potential for drug interactions. The MACH-1 trial also failed to eliminate the adverse events. The result was the removal of the agent from the market. It is no longer available for clinical use.
How could a drug with so many positive properties and with so many well-done clinical trials have gotten onto the market and been used by so many patients before it met its terminal fate? We believe that review of the outcome of the development of mibefradil provides convincing evidence that chemical entities (and devices) must be evaluated in trials that assess total health outcomes in patient groups reflecting the population of intended use. Settling for surrogate end points, such as physiological parameters, and small trials of low-risk populations studied for a short duration will expose our patients to substantial risk of unintended adverse consequences.
A systemically administered drug almost always has a combination of desirable effects and undesirable effects, at least some of which cannot be anticipated until the drug has been studied in patients with the diseases being treated who have relevant comorbidities commonly seen with the target diseases. In an increasingly complex clinical environment, in which a greater proportion of patients are older and often have multiple comorbid conditions, "proof-of-principle" studies in small populations provide an excellent foundation for establishing potential benefit, but such studies are not sufficient.
In addition to proof-of-principle studies, we must have clinical outcome studies that follow patients for more than several months and represent the patients and clinical environment in which the therapy will be used. Patients with common comorbidities and with other commonly coadministered therapies should not be excluded. Sample sizes should be large enough to rule out substantial negative effects on morbidity and mortality. Considerable debate has arisen in the clinical community about the appropriate definition of the term "substantial" in this context; however, we cannot afford to allow the complexity of this issue to divert us from the clear knowledge that the effects of therapies for chronic disease often lead to unanticipated outcomes, both positive, as with carvedilol,20 and negative, as with mibefradil. It is often stated that an approach of determining the outcomes associated with therapies for chronic diseases would be too expensive. A serious effort to streamline clinical trials to reduce the cost while maintaining the sample sizes required to find clinically important differences is urgently needed.
Must the baby be thrown out with the bathwater? In the case of mibefradil, it is difficult to imagine that the compound can be resuscitated specifically for the purpose of limiting infarct size. The clinical setting of myocardial infarction is even more complex than the clinical settings of angina and hypertension, in which the adverse outcomes occurred and led to the withdrawal of mibefradil from the market. Nevertheless, the offsetting negatives of this particular compound should not dampen the enthusiasm for other compounds that block T-type calcium channels, which may hold promise for reducing the size of myocardial infarcts and thereby improving the plight of patients with coronary heart disease. Despite decades of effort to test dozens of pharmacological entities, reperfusion with fibrinolysis or direct mechanical intervention remains the only clinically proven method of limiting infarct size. Even with the use of these therapies, however, >25% of patients have postinfarction heart failure, and the mortality of these patients is >8% per year. We must continue to focus therapeutic development on solutions to this major public health problem; perhaps other T-type calcium channel blocking agents without the drug interaction problems of mibefradil will lead to reduction in infarct size and improvement in outcomes for our patients.
Footnotes
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
References
-
Schwartz LM, Verbinski SG, Vander Heide RS, Reimer
KA. Epicardial temperature is a major predictor of myocardial infarct
size in dogs. J Mol Cell Cardiol. 1997;29:1577–1583.[Medline]
[Order article via Infotrieve]
-
Murry CE, Jennings RB, Reimer KA. Preconditioning with
ischemia: a delay of lethal cell injury in ischemic
myocardium. Circulation. 1986;74:1124–1136.
[Abstract/Free Full Text]
-
Przyklenk K, Kloner RA. Ischemic
preconditioning: exploring the paradox. Prog Cardiovasc Dis. 1998;40:517–547.[Medline]
[Order article via Infotrieve]
-
Reimer KA, Jennings RB. Verapamil in two
reperfusion models of myocardial infarction: temporary protection of
severely ischemic myocardium without limitation of
ultimate infarct size. Lab Invest. 1984;51:655–666.[Medline]
[Order article via Infotrieve]
-
Higginson L, Tang A, Knoll G, Calvin J. Effect of
intracoronary diltiazem on infarct size and regional myocardial
function in the ischemic reperfused canine heart. J
Am Coll Cardiol. 1991;18:867–874.
-
Lo H-M, Kloner RA, Braunwald E. Effect of
intracoronary verapamil on infarct size in the
ischemic, reperfused canine heart: critical importance of the
timing of treatment. Am J Cardiol. 1985;56:672–677.[Medline]
[Order article via Infotrieve]
-
Roberts R. Review of calcium antagonist
trials in acute myocardial infarction. Clin Cardiol.
1989;12(suppl III):4–47.
-
Roberts R. Thrombolysis and its sequelae:
calcium antagonists as potential adjunctive therapy.
Circulation. 1989;80(suppl IV):IV-93–IV-101.
-
Yusuf S. An overview of the clinical trials of agents
(other than beta-blockers) that potentially limit myocardial infarct
size. J Cardiovasc Pharmacol. 1988;12(suppl
I):548–555.
-
Goldstein RE, Boccuzzi ST, Cruess D, Nattel S.
Diltiazem increases late-onset heart failure in postinfarction patients
with early reduction in ejection fraction. Circulation. 1991;83:52–60.
[Abstract/Free Full Text]
-
Hansen JF. Calcium antagonists and
myocardial infarction. Cardiovasc Drugs Ther. 1991;5:665–670.[Medline]
[Order article via Infotrieve]
-
Clozel J-P, Veniant M, Osterrieder W. The structurally
novel Ca2+ channel blocker RO 40-5967, which
binds to the
[3H]desmethoxyverapamil receptor is
devoid of the negative inotropic effects of verapamil in
normal and failing rat hearts. Cardiovasc Drugs Ther. 1990;4:731–736.[Medline]
[Order article via Infotrieve]
-
Billman GE. RO 40-5967, a novel calcium channel
antagonist, protects against ventricular
fibrillation. Eur J Pharmacol. 1992;229:179–187.[Medline]
[Order article via Infotrieve]
-
Portegies MCM, Schmitt R, Kraaij CJ, Braat SHJG,
Gassner A, Hagemeijer F, Ponzenel H, Prager G, Viersma JW, Vander Wall
EE, Kleinbloesem CH, Lie KI. Lack of negative inotropic effects of the
new calcium antagonist RO 40-5967 in patients with stable
angina pectoris. J Cardiovasc Pharmacol. 1991;18:746–751.[Medline]
[Order article via Infotrieve]
-
Vander Heide RS, Schwartz LM, Reimer KA. The novel
calcium antagonist RO 40-5967 limits myocardial infarct
size in the dog. Cardiovasc Res. 1994;28:1526–1532.[Medline]
[Order article via Infotrieve]
-
Richard V, Tron C, Blanc T, Thuillez C. Infarct
size-limiting properties of RO 40-5967, a novel
nondihydropyridine calcium channel, in
anesthetized rats: comparison with verapamil.
J Cardiovasc Pharmacol. 1995;25:552–557.[Medline]
[Order article via Infotrieve]
-
Schulz R, Post H, Jalowy A, Backenköhler U,
Dörge H, Vahlhaus C, Heusch G. Unique cardioprotective action of
the new calcium antagonist mibefradil.
Circulation. 1999;99:305–311.
[Abstract/Free Full Text]
-
Epstein AE, Hallstrom AP, Rogers WJ, Liebson PR, Seals
AA, Anderson JL, Cohen JD, Capone RJ, Wyse DG, for the CAST
Investigators. Mortality following ventricular
arrhythmia suppression by encainide, flecainide, and moricizine
after myocardial infarction. JAMA. 1993;270:2451–2455.[Abstract]
-
Packer M, Carver JR, Rodeheffer RJ, Ivanhoe RJ,
DiBianco R, Zeldis SM, Hendrix GH, Bommer WJ, Elkayam U, Kukin ML,
Mallis GI, Sollano JA, Shannon J, Tandon PK, DeMets DL, for the PROMISE
Study Research Group. Effect of oral milrinone on mortality in severe
chronic heart failure. N Engl J Med. 1991;325:1468–1475.[Abstract]
-
Packer M, Bristow MR, Cohn JN, Colucci WS, Fowler MB,
Gilbert EM, Shusterman NH, US Carvedilol Heart Failure Study Group. The
effect of carvedilol on morbidity and mortality in patients with
chronic heart failure. N Engl J Med. 1996;334:1349–1355.
[Abstract/Free Full Text]
This article has been cited by other articles:
 |
|
 |
|
  R. M. Califf and D. L. DeMets Principles From Clinical Trials Relevant to Clinical Practice: Part II Circulation, August 27, 2002; 106(9): 1172 - 1175. [Full Text] [PDF]
|
|
|
|
|
|
R. M. Califf and D. L. DeMets Principles From Clinical Trials Relevant to Clinical Practice: Part I Circulation, August 20, 2002; 106(8): 1015 - 1021. [Full Text] [PDF]
|
|
|
|
 |
|
 G. Tognoni The EMIP Project Eur. Heart J., September 2, 2000; 21(18): 1498 - 1499. [PDF]
|
|
|
|
 |
|
 A. A. Grace and A.J. Camm Voltage-gated calcium-channels and antiarrhythmic drug action Cardiovasc Res, January 1, 2000; 45(1): 43 - 51. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||