Vernakalant Hydrochloride for Rapid Conversion of Atrial Fibrillation

Currently available antiarrhythmic agents have modest efficacy in converting atrial fibrillation (AF) to sinus rhythm, and the risk of proarrhythmia or hypotension is of concern.^1–3 Time to conversion with these drugs often is unpredictable and may be long, especially with oral therapies.⁴ Although electric cardioversion is more effective than drug administration, it is associated with adverse effects such as skin burns, heart block, ventricular proarrhythmia, and pacemaker or internal defibrillator malfunction.^2,4,5 Electrical cardioversion requires general anesthesia or conscious sedation from which patients must recover and may prolong hospitalization. In addition, the procedure generally is delayed for at least 6 hours after meals.⁶ A rapidly acting, efficacious, and safe drug that targets the fibrillating atria would be a valuable alternative to current treatments for patients with this common arrhythmia. Prompt pharmacological conversion of AF may prove to be a cost-saving strategy.

Vernakalant hydrochloride (RSD1235), an investigational compound, is a relatively atrium-selective, early-activating K⁺ and frequency-dependent Na⁺ channel blocker with a half-life of 2 to 3 hours.^7,8 In animal models of AF and in a recent clinical study, vernakalant selectively prolonged the atrial refractory period without affecting ventricular refractoriness.^7–9 Vernakalant effectively converted acute-onset AF (AF lasting 3 to 72 hours) in a placebo-controlled phase 2 trial (n=56).¹⁰

Methods

Study Design

This prospective, randomized, double-blind, placebo-controlled trial was conducted in 44 centers in Canada, the United States, and Scandinavia. Patients were stratified by duration of AF: 3 hours to 7 days (short duration) and 8 to 45 days (long duration). Patients within the 2 strata were then randomized 2:1 to vernakalant or placebo using block randomization and a computer-generated centralized list via an Interactive Voice Response System. This 2:1 randomization scheme was used to gather extra safety data for vernakalant. The study protocol was developed by the Steering Committee and the sponsor. Study oversight was provided by the Steering Committee and an unblinded independent Data Safety Monitoring Committee. The Data Safety Monitoring Committee held monthly teleconferences, with 2 planned interim safety evaluations once 90 and 180 patients, respectively, were enrolled. The protocol was approved by the institutional or regional review board at each site, and patients gave written informed consent before starting study procedures. The Steering Committee approved the text of this manuscript before publication.

Selection and Description of Participants

To be eligible, patients had to have sustained AF for 3 hours to 45 days, be ≥18 years of age, have a body weight of 45 to 136 kg, be receiving adequate anticoagulation, and have a systolic blood pressure >90 mm Hg and <160 mm Hg and a diastolic blood pressure <95 mm Hg. Women could not be pregnant or nursing and, if premenopausal, had to use an effective form of birth control. Patients were excluded if they had sick-sinus syndrome or QRS >0.14 seconds without a pacemaker; ventricular rate of <50 bpm; uncorrected QT >0.440 seconds; typical atrial flutter; New York Heart Association class IV heart failure; acute coronary syndrome, myocardial infarction, or cardiac surgery within 30 days before enrollment; an investigational drug within 30 days before enrollment; a reversible cause of AF; end-stage disease; previously failed electric conversion; uncorrected electrolyte imbalance; or digoxin toxicity. Treatment initiation with β-blockers, calcium antagonists, and digoxin for control of ventricular rate was permitted for up to 2 hours before study drug infusion. Treatment with intravenous class I and III antiarrhythmics was allowed up to 24 hours before study drug infusion, and background therapy of oral antiarrhythmics was allowed.

Treatment Plan

Patients received a 10-minute infusion of vernakalant (3.0 mg/kg) or placebo, followed by a 15-minute observation period. If the patient did not convert to sinus rhythm, an additional dose of vernakalant (2.0 mg/kg) or placebo was administered. The infusion was to be discontinued if the uncorrected QT interval increased to >0.550 seconds or by >25%; heart rate decreased between 40 and 50 bpm with symptoms or to <40 bpm; systolic blood pressure increased to >190 mm Hg or decreased to <85 mm Hg; new bundle-branch block developed or QRS increased ≥50%; or polymorphic ventricular tachycardia, a sinus pause of ≥5 seconds, or intolerable side effects occurred. The use of other antiarrhythmic medications and electrical cardioversion was not recommended until at least 2 hours after study drug infusion.

Patients were observed in the hospital for a minimum of 8 hours after study drug infusion. ECGs were recorded and vital signs were measured at screening; at baseline; every 5 minutes from the start of infusion to 50 minutes after infusion; at 90 minutes and 2, 4, 8, and 24 hours; and at 1 week after dose. Lead II or V₅ tracings were used to measure ECG intervals by a central ECG laboratory. Generally, the intervals were evaluated from 3 (consecutive if possible) complexes. QT intervals were corrected for heart rates with Bazett’s (QTcB) and Fridericia’s (QTcF) formulas. A Holter continuously monitored the cardiac rhythm from screening to 24 hours after dose.

Study End Points

The primary efficacy end point was the proportion of patients in the short-duration AF (3 hours to 7 days) group who had conversion to sinus rhythm for at least 1 minute within 90 minutes of drug initiation. Secondary and exploratory efficacy end points in the short-duration AF group included the time to conversion from first exposure to study drug and the proportion of patients who remained in sinus rhythm at 24 hours, respectively. Other secondary efficacy end points included the proportion of patients with AF duration of 3 hours to 45 days and 8 to 45 days who had termination of AF (defined as the absence of AF or atrial flutter, which included conversion to sinus rhythm and a paced rhythm).

Conversion to sinus rhythm and termination of AF were adjudicated by a Clinical Events Committee blinded to treatment assignment. The Clinical Events Committee also reviewed all episodes of suspected torsade de pointes. All 12-lead ECGs and 24-hour Holter recordings were reviewed by a cardiologist at the central ECG laboratory who was blinded to treatment assignment. Ventricular tachycardia was defined as ≥3 wide complex beats with a rate of ≥100 bpm.

A serious adverse event was defined as any adverse event occurring from the start of the infusion through the 30 days after study treatment that, at any dose of study drug, was fatal or life threatening, required or prolonged hospitalization, was significantly incapacitating, or required medical or surgical intervention.

Statistical Considerations

All randomized patients who received any amount of study drug were included in the efficacy and safety analyses (prespecified modified intention-to-treat population). Baseline characteristics were compared between groups by use of a 1-way ANOVA with a fixed effect for treatment for continuous variables and a χ² test or Fisher’s exact test for categorical variables. The primary end point was analyzed with the Cochran-Mantel-Haenszel test stratified by center; this test also was used in the analysis of the secondary end point in the group of patients with AF duration of 3 hours to 45 days. Fisher’s exact test was used in the analysis of the secondary end point in the group with AF duration of 8 to 45 days. The Kaplan-Meier method was used to summarize time to conversion, with the log-rank test used to compare the distributions. The sample sizes were based on assumed conversion rates of 25% and 50% for the placebo and active groups, respectively, in the short-duration AF group and 5% and 30% in the long-duration AF group. Based on a 2-sided χ² test (significance level, P=0.05), 240 patients in the short-duration AF group and 120 patients in the long-duration AF group, each allocated in a 2:1 ratio of vernakalant to placebo, provided >90% power to detect a 25% difference between treatments. Except for AF duration, data are given as mean±SD.

The authors had full access to and take responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.

Results

Patient Disposition

A total of 356 patients were randomized to either vernakalant or placebo. Twenty patients did not receive study drug and were withdrawn: 14 spontaneously converted to sinus rhythm; 2 violated inclusion or exclusion criteria; 2 were diagnosed with myocardial infarction; 1 could not obtain the study drug; and 1 discontinued for an unspecified reason. Thus, the efficacy and safety evaluable populations included 220 patients in the short-duration AF group and 116 patients in the long-duration AF group (Figure 1).

• Download figure
• Download PowerPoint

Baseline Characteristics

There were no statistical differences in patient demographics between the 2 treatment arms in the short-duration AF, long-duration AF, or overall groups (Table 1). The median AF duration at baseline in the placebo and vernakalant groups was 28.4 and 28.2 hours, respectively, in the short-duration AF group, and 465.2 and 613.0 hours, respectively, in the long-duration group.

Efficacy End Points

In the primary efficacy analysis, 75 of the 145 vernakalant patients (51.7%) in the short-duration AF (3 hours to 7 days) group converted to sinus rhythm within 90 minutes compared with 3 of the 75 placebo patients (4.0%; P<0.001; Figure 2). Figure 3 shows the cumulative success of conversion relative to time after the start of the infusion. Patients with AF lasting 3 to 48 hours given vernakalant demonstrated the highest conversion rate (62.1% versus 4.9% with placebo; P<0.001; Table 2). Of the 75 patients who demonstrated conversion, 57 (76.0%) did so with a single dose. The median time to conversion to sinus rhythm for the 75 patients receiving vernakalant who converted was 11 minutes. Only 1 of the 75 vernakalant-treated patients who converted to sinus rhythm relapsed to AF at 24 hours.

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

Six of the 76 vernakalant patients (7.9%) in the long-duration AF group had termination of AF (5 demonstrated conversion to sinus rhythm, 1 had paced rhythm) compared with 0 of the 40 placebo patients (P=0.09; Figure 2). In the overall population, 83 of the 221 vernakalant patients (37.6%) experienced termination of AF compared with 3 of the 115 placebo patients (2.6%; P<0.001). An additional 2 of 4 vernakalant patients with pacemakers in the short-duration AF group converted to a paced rhythm but were considered treatment failures in the prespecified primary analysis of conversion to sinus rhythm.

Nineteen vernakalant patients (8.6%) displayed atrial flutter in the first 90 minutes, 5 of whom subsequently converted to sinus rhythm. Each of the 19 patients had AF confirmed at baseline. None of these episodes of atrial flutter were associated with 1:1 atrioventricular conduction.

Safety Results

Deaths were assessed up to 30 days after study drug infusion. Three patients died; all 3 had received vernakalant. None of these deaths were considered to be related to study drug. A 68-year-old woman received vernakalant and died 28 hours later during a gastroscopy procedure; an autopsy revealed rupture of a dissecting aneurysm of the ascending aorta. A 90-year-old woman died of pulmonary edema and congestive heart failure 26 days after receiving vernakalant. A 67-year-old man with lung cancer died of pneumonia and respiratory arrest 8 days after successful conversion with vernakalant.

Serious Adverse Events

In the overall study population, investigator-reported serious adverse events were recorded from the time of study drug infusion through the 30-day follow-up assessment. There were 21 placebo patients (18.3%) and 29 vernakalant patients (13.1%) with serious adverse events. Most events were of cardiac origin, with the most common being recurrent AF requiring hospitalization (placebo, 12.2%; vernakalant, 5.9%). Four serious adverse events occurring in 3 patients were considered to be possibly or probably related to vernakalant. One patient developed hypotension 14 minutes after receiving a partial second infusion (mean baseline blood pressure, 120/81 mm Hg; lowest recorded blood pressure, 88/53 mm Hg), which responded to a saline intravenous bolus infusion. A second patient had hypotension (mean baseline blood pressure, 113/83 mm Hg; lowest recorded blood pressure, 82/68 mm Hg) after the first infusion, which responded to a saline intravenous bolus infusion, and he did not receive the second infusion; he underwent electrical cardioversion a few hours later and then developed cardiogenic shock. He was successfully treated and later diagnosed as having a tachyarrhythmia-induced cardiomyopathy. The previously mentioned 90-year-old woman had complete heart block after electrical cardioversion, which was performed ≈2.5 hours after the second vernakalant infusion.

Ventricular Arrhythmia During the First 24 Hours

A full characterization of possible ventricular arrhythmia events during the 24-hour period after the administration of vernakalant included investigator-reported adverse events (such as syncope), 12-lead ECGs, and Holter recordings. The incidence of ventricular arrhythmia from all sources was 17.4% for placebo and 9.0% for vernakalant. Confirmed nonsustained ventricular tachycardia was reported in 14.8% of the placebo patients and 6.3% of the vernakalant patients. There were no reports of sustained ventricular tachycardia during this 24-hour interval.

Torsade de Pointes

There were no reports of torsade de pointes or ventricular fibrillation during the first 24 hours after infusion. The 90-year-old woman referenced above experienced an episode of torsade de pointes 32 hours after vernakalant administration. Another patient had 3 episodes of torsade de pointes, 2 of which occurred 16 days after treatment with vernakalant (2 days after cardiac surgery). The third episode occurred 17 days after treatment and resulted in cardiac arrest, for which the patient received an internal defibrillator.

Other Adverse Events

The most common treatment-emergent adverse events reported during the first 24 hours in patients given vernakalant were dysgeusia (29.9% vernakalant, 0.9% placebo), sneezing (16.3% vernakalant, 0% placebo), paresthesia (10.9% vernakalant, 0% placebo), nausea (9.0% vernakalant, 0.9% placebo), and hypotension (6.3% vernakalant, 3.5% placebo). The median duration of related adverse events in vernakalant patients was 12 minutes for dysgeusia, 3 minutes for sneezing, 7 minutes for paresthesia, 12.5 minutes for nausea, and 15 minutes for hypotension.

A total of 5 patients, 4 vernakalant and 1 placebo, had study drug discontinued because of adverse events. Adverse events that led to study drug discontinuation in 3 of the 4 vernakalant patients were bradycardia (n=1), hypotension (n=1), and prolonged uncorrected QT (>25%, as per protocol; n=1); 1 of the 4 patients discontinued as a result of ventricular bigeminy and multiple other minor complaints. The placebo patient discontinued study drug prematurely because of prolonged QT.

Effects of Vernakalant on ECG

ECG effects of vernakalant were compared with those of placebo in patients who remained in AF. Except for a transient 4.5-bpm increase at 10 minutes, the effect of vernakalant on heart rate was unremarkable (Figure 4A). Patients given vernakalant who demonstrated termination of AF had higher mean heart rates at baseline (110±26 versus 91±22 bpm in those who remained in AF). There was a correlation between AF duration and baseline heart rate (log [AF duration], P<0.001) so that patients with shorter AF durations had higher baseline heart rates. Vernakalant prolonged QRS intervals from 100.1±16.6 ms (mean±SD) at baseline to a peak of 106.9±21.4 ms at the end of the first infusion, whereas QRS intervals remained unchanged in placebo patients (Figure 4B). Maximum mean increases in QT, QTcB, and QTcF intervals in vernakalant patients who remained in AF were observed near the end of each infusion (Figure 4C and 4D).

• Download figure
• Download PowerPoint

Similar changes in QT, QTcB, and QTcF intervals were observed in the overall population (converters to sinus rhythm and those who remained in AF; Table 3). At the end of the first infusion, 24% of all patients given vernakalant had QTcB >500 ms compared with 15% of all patients given placebo. If patients with baseline QTcB >500 ms are excluded, these percentages fall to 22% with vernakalant and 12% with placebo. For QTcF, 3% of all vernakalant patients and 1% of all placebo patients had values >500 ms at the end of the first infusion, all of whom had baseline values <500 ms. Within 90 minutes, the percentages of patients with QTcB or QTcF >500 ms in the 2 treatment groups were essentially the same (QTcB, 11% vernakalant versus 14% placebo; QTcF, 1% vernakalant versus 2% placebo at 90 minutes).

In patients given vernakalant who demonstrated conversion to sinus rhythm, QRS intervals increased from 95.1±11.9 ms at baseline to 99.1±12.3 ms at the end of the first infusion, and QTcF intervals were 408.6±24.1 ms at baseline compared with 422.4±22.1 ms after the first infusion.

Hemodynamic Effects of Vernakalant

There were no significant changes from baseline in mean systolic or diastolic blood pressure in vernakalant patients compared with those given placebo. In the 24 hours from the start of drug infusion, hypotension was reported as an adverse event in 14 vernakalant patients (6.3%) and 4 placebo patients (3.5%). Hypotensive episodes occurred within 15 minutes of the end of either infusion of vernakalant, were transient, and resolved without pharmacological intervention. Two previously described hypotensive events were reported as serious; both cases responded to saline, and 1 led to discontinuation of treatment.

Discussion

The identification of an intravenous antiarrhythmic drug that can provide reliable, safe, and prompt pharmacological cardioversion is highly desirable. Vernakalant was effective in converting AF to sinus rhythm. Vernakalant was especially effective in treating short-duration AF. Fifty-two percent of vernakalant patients in the short-duration AF group converted to sinus rhythm compared with only 4% of placebo patients, and patients with the shortest duration of AF demonstrated the highest rate of conversion. The majority (76%) converted after receiving only the first infusion. Additionally, conversion to sinus rhythm was rapid; the median time to conversion was 11 minutes. These findings are consistent with results from the phase 2 study, which provided the first clinical evidence of effective intravenous cardioversion with vernakalant.¹⁰ Vernakalant was associated with a minimal risk of early recurrence of AF.

Efficacy comparisons with other drugs used for acute cardioversion of AF must be made with caution because the drugs are used in different patient populations. The most relevant comparison to vernakalant is ibutilide, a class III agent and the only intravenous drug approved by the Food and Drug Administration for the conversion of AF.⁵ Results from 3 large, randomized, placebo-controlled ibutilide trials indicate placebo-subtracted conversion rates of 28% to 31% for AF.^11–13 Studies comparing ibutilide with other agents have reported AF conversion rates of 32% to 51%.^14–16 In this study, vernakalant displayed an AF conversion rate that exceeded that reported for ibutilide.

Other antiarrhythmic drugs are used for pharmacological cardioversion of AF. Patterns of usage vary in different countries.^17–19 Amiodarone generally is considered of limited value for the acute cardioversion of AF because prolonged infusions generally are required for efficacy.^17,18,20,21 Intravenous flecainide and propafenone reportedly successfully terminate recent-onset AF in >50% to 60% of patients.^22–28 These formulations are not available in North America,^4,19 however, and they have significant proarrhythmic and negative inotropic potential.^22,23,25,26 Procainamide is used despite a risk-to-benefit profile that is inferior to ibutilide or flecainide.^15,28 Studies of intravenous dofetilide have shown modest results.^29–31 Tedisamil, an investigational class III drug, successfully converted 57% of patients with recent-onset AF to sinus rhythm in a preliminary trial; however, an increased risk of proarrhythmia was seen.³²

Safety Considerations of Vernakalant

Vernakalant was well tolerated in most patients. A transient alteration in taste, sneezing, paresthesia, and nausea were the most common adverse reactions with vernakalant. Hypotension may occur; however, most episodes of hypotension in this study were transient. The exception was the patient with tachyarrhythmic cardiomyopathy previously described. Atrial flutter may occur with vernakalant, and atrioventricular nodal-blocking drugs for rate control may be needed.

Vernakalant-treated patients who remained in AF showed statistically significant increases in QRS duration, QT, QTcB, and QTcF. In those who remained in AF, vernakalant had little effect on heart rate. Patients given vernakalant who were successfully treated had a higher baseline heart rate, which was likely related to a shorter duration of AF. Two patients had torsade de pointes: 1 patient 32 hours after treatment with vernakalant and the second patient 16 to 17 days after treatment. These episodes were not considered to be due to vernakalant because the half-life of vernakalant is 2 to 3 hours. There were 3 deaths during the study, none of which were considered to be related to vernakalant because they occurred 28 hours and 8 and 26 days after study drug infusion.

Conclusion

Vernakalant demonstrated rapid conversion of short-duration AF and was well tolerated.

Footnotes