Author + information
- Received August 6, 2002
- Revision received October 22, 2002
- Accepted November 19, 2002
- Published online May 7, 2003.
- Christopher M O’Connor, MD, FACC*,
- Wendy A Gattis, PharmD*,* (, )
- Kirkwood F Adams Jr, MD, FACC†,
- Vic Hasselblad, PhD*,
- Bleakley Chandler, MD, FACC‡,
- Aline Frey, PharmD§,
- Isaac Kobrin, MD§,
- Maurizio Rainisio, PhD§,
- Monica R Shah, MD*,
- John Teerlink, MD, FACC∥,
- Mihai Gheorghiade, MD, FACC¶,
- RITZ-4 Investigators
- ↵*Reprint requests and correspondence:
Dr. Wendy A. Gattis, Duke Clinical Research Institute, 2400 N. Pratt Street, 0311 Terrace Level, Durham, North Carolina 27705, USA.
Objectives We sought to determine the effect of tezosentan in patients with acute decompensated heart failure (HF) associated with acute coronary syndrome (ACS).
Background Tezosentan is a dual endothelin receptor antagonist that has been shown to improve cardiac output, decrease pulmonary capillary wedge pressure, and reduce pulmonary and systemic vascular resistance in initial clinical studies in acute decompensated HF.
Methods The Randomized Intravenous TeZosentan (RITZ)-4 study was a multicenter, randomized, double-blinded, placebo-controlled study of tezosentan in patients with acute decompensated HF associated with ACS. A total of 193 patients were randomized to receive tezosentan (25 mg/h for 1 h, then 50 mg/h for 23 to 47 h) or placebo. Patients with evidence of acute decompensated HF and ACS were eligible to participate. The primary end point was the composite of death, worsening HF, recurrent ischemia, and recurrent or new myocardial infarction within 72 h.
Results No significant differences were observed between placebo and 50 mg/h tezosentan in the composite primary end point: 24.2% (95% confidence interval [CI] 16.0% to 34.1%) and 28.9% (95% CI 20.1% to 39.0%), respectively (p = 0.5152). Symptomatic hypotension was more frequent in the treatment group.
Conclusions At the doses studied, tezosentan did not result in a significant improvement in the composite primary clinical end point in the RITZ-4 trial. Tezosentan did not demonstrate pro-ischemic effects in this population. Symptomatic hypotension may have resulted in an increased number of adverse events in the treatment group. Further studies with lower tezosentan doses are warranted.
Acute heart failure (HF) is a major worldwide health care concern, but few data describing this population are available. The American Heart Association 2002 Heart and Stroke Statistical Update reported 978,000 discharges for HF in 1998, as compared with 377,000 in 1979, a 160% increase in less than two decades (1). Despite the increasing prevalence of hospital admissions for acute HF, treatment strategies supported by data from well-designed clinical trials are lacking. Current therapies focus on improving the hemodynamic abnormalities that are present in these patients—namely, volume overload and/or reduced cardiac output. Clinical outcomes have not been improved when standard inotropic therapies, such as milrinone, have been tested in large, randomized trials (2).
Heart failure is a known complication of acute coronary syndrome (ACS). Several recent studies have also shown elevated levels of troponin in patients admitted primarily with HF in the absence of classic ACS (3–7). These data suggest that some degree of myocardial necrosis may be ongoing and contributing to the HF state, even in the absence of frank ACS. Treatment of HF symptoms in the setting of ACS has not been well studied. A rational investigative approach for this population is to test therapies that improve the balance between myocardial oxygen supply and demand and diminish myocardial damage while also improving pulmonary congestion and other HF symptoms. Endothelin promotes ischemia and worsens hemodynamics; thus, blockade of endothelin would be expected to lessen the ischemic burden and reduce HF symptoms (8). The Randomized Intravenous TeZosentan (RITZ-4) study was designed to evaluate the effects of tezosentan, a non-selective endothelin antagonist, in patients with acute HF and ACS.
The RITZ-4 trial was a pilot, multicenter, randomized, double-blinded, placebo-controlled trial of tezosentan or placebo in patients with acute HF in the setting of ACS (9). Tezosentan is a dual endothelin (ET) antagonist with a high binding affinity for both ETaand ETbreceptors (10). The study was approved by local ethics committees (institutional review boards) in accordance with the Declaration of Helsinki. Patients were eligible for the study if they met the criteria listed in Table 1. Qualifying patients were randomized in a 1:1 fashion to receive tezosentan or placebo for at least 24 and up to 48 h. Background oral therapy was continued and adjusted as clinically indicated by the treating physician. Parenteral inotropic, sympathomimetic, and vasodilator therapy started at least 2 h (4 h for milrinone) before randomization was allowed to continue, but the dose or infusion rate was kept constant during treatment with the study drug up to 24 h. Increases in the dose or initiation of new inotropic, sympathomimetic, or vasodilatory therapy during the initial 24 h after randomization was done only if necessary, based on the patient’s clinical condition, and was counted as worsening HF. To ensure no pharmacologic interference, intravenous (IV) diuretics were not given within 2 h before randomization or during the first 6 h after randomization, unless clinically necessary based on the condition of the patient. Furosemide was the only IV diuretic allowed per protocol (replaced by ethacrynic acid in patients allergic to furosemide).
After obtaining written, informed consent, patients were randomized to tezosentan or matching placebo. Patients randomized to the tezosentan group received 25 mg/h IV for 1 h followed by 50 mg/h for 23 h and up to 48 h of treatment. The infusion rate could be decreased by half two times if systolic blood pressure (SBP) decreased to <95 mm Hg or symptomatic hypotension occurred, or if the patient experienced study drug-related adverse events. During study infusion, patients were monitored by continuous telemetry. A 12-lead electrocardiogram and cardiac enzymes and/or troponin levels were obtained at baseline, 6 h (only cardiac enzymes), 12 h (only cardiac enzymes), 24 h, 48 h, 72 h, treatment discontinuation, and, in case of ischemic symptoms, within the first 72 h of study drug treatment.
The primary efficacy end point was the composite incidence of death, worsening HF, recurrent ischemia, and recurrent or new myocardial infarction (MI) within the first 72 h of study drug treatment. Secondary efficacy end points included: 1) change in HF score at 24, 48, and 72 h after treatment initiation; 2) death or worsening HF during 72 h after treatment initiation; 3) recurrent ischemia or recurrent/new MI during 72 h after treatment initiation; 4) total days alive out of the hospital within 30 days of treatment initiation; 5) all-cause mortality and rehospitalization 30 days after treatment initiation; 6) length of initial hospital stay; and 7) total dose of IV diuretics administered over 72 h after treatment initiation.
End point definitions
“Worsening HF” was defined as initiation or increase of IV treatment for HF with inotropic, vasodilator, or sympathomimetic treatments; mechanical ventilatory or circulatory support during 72 h after treatment initiation; or IV administration of furosemide during the first 6 h of treatment.
“Recurrent ischemia” was defined as the following: 1) ischemic symptoms (chest pain or its equivalent) associated with new ST-segment changes ≥1 mm or definite new T-wave inversion in two or more contiguous leads; or 2) ischemic symptoms (chest pain or its equivalent of ≥5 min duration) relieved within 10 min of administration of an anti-ischemic drug.
“Recurrent or new MI” was defined as ischemic symptoms (chest pain or equivalent) associated with a new ST-segment shift (≥1 mm), definite T-wave inversion in two or more contiguous leads, or a new Q-wave, as well as an increase in creatine kinase-MB fraction or troponin T or I during 72 h after treatment initiation that was at least two times higher than the previous level if it was above the upper limit of normal or to a level above the limits of normal if the previous level was within normal limits.
The sample size estimate was based on a projected event rate for the primary end point (death, worsening HF, recurrent ischemia, recurrent or new MI) in the placebo and tezosentan groups of 40% and 20%, respectively. At an alpha level of 0.05, 91 patients in each arm were required to have 80% power to detect a significant difference between treatment arms. The analysis was based on the intention-to-treat principle, and the Fisher exact test was used to test the primary end point. The p values provided for secondary end points are given for exploratory purposes only.
One hundred ninety-three patients were enrolled in RITZ-4: 96 were randomized to placebo and 97 to tezosentan. One patient (in the placebo group) did not start the study drug and is therefore not included in this analysis. This patient was determined to be ineligible for the study after randomization but before initiation of the study drug. Given the fact the study was double-blinded, it was determined that excluding this patient was appropriate. The baseline characteristics of the study population are presented in Table 2. The median age was 70 years in the placebo group and 74 years in the tezosentan group—an older population than typically represented in HF trials. The median time (25th, 75th percentiles) from ACS onset to initiation of randomized treatment was 28 h (18, 39) and 29 h (19, 39) for the placebo and tezosentan groups, respectively.
At the dose studied, there was no significant difference in the incidence of the primary end point between the treatment groups. The incidence of death, worsening HF, recurrent ischemia, or recurrent/new MI within 72 h was 24.2% in the placebo group and 28.9% in the tezosentan group (p = 0.52) (Fig. 1). When examining each component of the composite separately, no excess risk of death, recurrent ischemia, or recurrent/new MI was observed (Table 3). The incidence of worsening HF was higher in the tezosentan group than in the placebo group (19.6% vs. 11.6%, p = 0.16). There were no clinically relevant differences in the secondary end points of the study. The HF score was similar between the groups at 24, 48, or 72 h (Table 4).
The most common adverse effects in RITZ-4 were related to the vasodilatory properties of tezosentan (Table 5). Overall, 81.1% of the placebo group and 91.8% of the tezosentan group experienced an adverse event. In the placebo group, 13.7% of the patients experienced a serious adverse event, compared with 14.4% of patients in the tezosentan group. The adverse events occurring with a frequency of >3% in the tezosentan group are displayed in Table 5. Adverse events requiring discontinuation of the study drug are displayed in Table 6. There were no clinically meaningful differences in the incidence of liver enzyme or other laboratory abnormalities between the groups. Substantial reductions in blood pressure were observed in the study population, despite eligibility criteria that excluded patients with SBP <105 mm Hg (Table 7). Hypotension requiring a decrease or discontinuation of the study drug occurred in seven placebo-treated patients and 17 tezosentan-treated patients. A non-significant trend toward a higher incidence of worsening HF was observed in the tezosentan group, as described earlier. No changes in heart rate were observed for either treatment group. The median change (25th, 75th) in heart rate at 48 h was 2 beats/min (−4, 11) and 2 beats/min (−6, 10) in the placebo and tezosentan groups, respectively. Despite the degree of hypotension observed in the study, there was no evidence of a pro-ischemic effect. The incidence of renal failure was higher in the tezosentan group (7.2%) than in the placebo group (2.1%). Renal failure was reported at the discretion of the investigator. No specific criteria or definitions for renal failure were provided to the investigators. The exact mechanism for the development of renal failure is unknown. It may be related to the incidence of hypotension, as there is no published or unpublished direct evidence from clinical or experimental studies to suggest that tezosentan has a direct nephrotoxic effect. It is also possible that excess vasodilation of the efferent arterioles may have occurred, resulting in decreased filtration pressure and development of renal failure. It is also possible that direct nephrotoxic effects exist that have not been discovered in clinical trials conducted to date. Other significant adverse effects include nausea and headache, both of which were higher in the tezosentan arm (Table 5). Survival was similar between the groups at six months.
Although no significant differences were detected in the primary end point, RITZ-4 produced meaningful data on patients with ACS and acute HF. The increased risk of hypotension, headache, and nausea in this study suggests that the dosing strategy chosen may have been too high. The higher incidence of both worsening HF and renal failure may have been related to the high incidence of hypotension. The fact that no pro-ischemic effects were observed is important for future studies with tezosentan. Dose-ranging studies with tezosentan are ongoing to evaluate whether favorable hemodynamic effects are present at lower doses with less potential to cause hypotension and other reported adverse effects. Thus, if safety can be confirmed at lower doses while maintaining a pharmacologic effect, further evaluation of tezosentan’s efficacy in this population can be examined.
Inhibition of the endothelin system is a rational approach in the treatment of acute HF. Endothelin is a potent vasoconstrictor and contributes to poor outcomes in both ischemic heart disease and HF (8,11–21). Initial experimental and clinical studies with tezosentan appeared promising (22–24). RITZ-4 was the third study within the RITZ program. RITZ-2 evaluated the effect of tezosentan versus placebo on the primary end point of a change in the cardiac index at 6 h (25). Tezosentan was associated with a significant improvement in the cardiac index at 6 h, as compared with placebo, and a reduction in pulmonary capillary wedge pressure (25). RITZ-1 compared a change in the dyspnea score for tezosentan versus placebo (26). Hemodynamic end points were not included in the analysis. There was no difference in the primary end point between the two treatment groups (26). The lack of benefit in RITZ-1 may have been associated with the inherent difficulty in showing differences in subjective symptomatic end points when the comparator is an active control agent and when patients may have received standard therapies during the time between hospital admission and randomization. Such factors decrease the likelihood that a significant difference in the dyspnea score could be observed. The incidences of hypotension in all of the RITZ studies were similar, which could have limited the potential beneficial effects of the drug. Initiating therapy with low initial doses may be a safer approach than dosing to achieve maximal inhibition of the endothelin system.
It is apparent that the dose used in RITZ-4 was too high. Additional work is ongoing to identify the optimal tezosentan dose in the HF population. RITZ-4 was a pilot study and was underpowered to determine the efficacy of tezosentan in patients with acute HF and ACS. The placebo event rate was assumed to be 40%; the actual placebo event rate was 24.2%. Thus, the actual power to detect a 50% reduction in the composite end point was 61%. To detect a smaller difference of 20% to 25%, the power was further compromised at <20%.
The population of patients with acute HF and ACS has not been well studied or previously characterized in terms of clinical outcomes. From RITZ-4, important data have been learned in terms of patient characteristics, treatment patterns, and the rate and type of clinical events. This information, particularly regarding the event rate, was previously unknown. RITZ-4 provides important data for the planning of future trials with tezosentan and other potential therapies in this population. Although caution must be exercised when developing future studies with tezosentan, given the results of RITZ-4, it is possible that different results may be observed with lower doses of tezosentan. Work is ongoing to identify optimal dosing strategies for tezosentan in the HF population. Once these data are known, further evaluation of tezosentan as a potential therapy for decompensated HF may be warranted. Based on the RITZ-4 findings, future trials must include assessments of clinical outcomes.
☆ The RITZ-4 study was supported by Actelion, Ltd., Basel, Switzerland. Drs. Frey, Kobrin, and Rainisio are employees of Actelion, Ltd. Drs. O’Connor, Gattis, Adams, and Chandler received research grants to support the RITZ-4 study from Actelion, Ltd.
- acute coronary syndrome
- heart failure
- myocardial infarction
- Randomized Intravenous TeZosentan program
- systolic blood pressure
- Received August 6, 2002.
- Revision received October 22, 2002.
- Accepted November 19, 2002.
- American College of Cardiology Foundation
- American Heart Association
- Chen Y.N.,
- Wei J.R.,
- Zeng L.J.,
- Wu M.Y.
- Mossov E.,
- Calzolari C.,
- Pau B.
- Clozel M.,
- Ramuz H.,
- Clozel J.P.,
- et al.
- Love M.P.,
- McMurray J.J.V.
- Pousset F.,
- Isnard R.,
- Lechat P.,
- et al.
- Pernow J.,
- Wang Q.D.
- Omland T.,
- Lie R.T.,
- Aakvaag A.,
- et al.
- Stewart D.J.,
- Kubac G.,
- Costello K.B.,
- et al.
- Ray S.G.,
- McMurray J.J.,
- Mortaon J.J.,
- et al.
- Wei C.M.,
- Lerman A.,
- Rodeheffer R.J.,
- et al.
- Ponicke K.,
- Vogelsang M.,
- Heinroth M.,
- et al.
- Cotter G.,
- Kiowski W.,
- Kaluski E.,
- et al.
- Schalcher C.,
- Mandinov L.,
- Schindler R.,
- et al.
- Louis A.,
- Cleland J.G.,
- Crabbe S.,
- et al.