San Francisco, CA- An investigational drug envisioned as a safer alternative to amiodarone for maintaining sinus rhythm in patients with atrial fibrillation (AF) has shown that it may also lower the risk of some clinical outcomes, at least in a population with the arrhythmia described at increased cardiovascular risk, in a large randomized trial [1].

Dr Stefan H Hohnloser

Compared with those who received placebo, patients who received dronedarone (Multaq, Sanofi-Aventis) showed a 24% drop in risk of CV hospitalizations or death over almost two years in the ATHENA trial, reported here at the Heart Rhythm Society (HRS) 2008 Scientific Sessions.

A range of secondary clinical end points also improved with dronedarone, according to Dr Stefan H Hohnloser (JW Goethe University, Frankfurt, Germany). A significant drop in risk of cardiovascular death with the drug was driven by fewer arrhythmic deaths, and a significant decline in CV hospitalizations owed primarily to reductions in admissions for AF and acute coronary syndromes, he said when formally presenting ATHENA at the meeting. Hohnloser chairs the trial's steering committee.

There was no dronedarone effect on all-cause mortality, in either direction, and about as many actively treated patients as controls dropped out due to side effects.

Many are viewing dronedarone's effects on CV hospitalization or death as something of an accomplishment, as few antiarrhythmics used in AF, if any, have shown they make much of a difference on hard clinical end points. And while dronedarone is viewed as less efficacious against the arrhythmia than amiodarone, although that wasn't tested in ATHENA, the drug does seem to carry a low risk of adverse events while the more established Vaughn-Williams class 3 agent is legendary for its toxic side effects.

The two drugs have similar molecular structures and appear to share many biochemical actions, but dronedarone, a benzofuran analog of amiodarone, lacks the iodine component that is largely responsible for the latter's multiple end-organ toxicities, which hit the lungs, thyroid, eyes, and other organs.

Hohnloser described ATHENA, with its >4500 "moderate- to high-risk elderly atrial-fibrillation patients" randomized in 37 countries, as "the largest antiarrhythmic drug trial ever conducted." The trial enrolled hemodynamically stable patients >75 years old or >70 if they had at least one CV risk factor, such as hypertension, diabetes, prior stroke, or transient ischemic attack, an enlarged left atrium, or an LVEF <40%. About equal proportions of the 2301 randomized to dronedarone (400 mg twice daily) and 2327 to placebo were on beta blockers, calcium-channel blockers, ACE inhibitors, or angiotensin-receptor blockers, digoxin, statins, or oral anticoagulation.

Hazard ratios for primary and secondary end points in ATHENA, dronedarone vs placebo, mean follow-up of 21 months

End point	Hazard ratio	p
Time to first CV hospitalization or death from any cause*	0.76	<0.001
All-cause mortality	0.84	NS
CV mortality	0.71	0.034
CV hospitalization	0.75	<0.001
Death from cardiac arrhythmia	0.55	0.01
Cardiac nonarrhythmic death	0.95	NS

As the trial's results were being made public, Sanofi-Aventis launched an aggressive media campaign that emphasized dronedarone's good safety profile and reduction of clinical end points in ATHENA, with far less mention of the drug's level of efficacy at suppressing AF [2]. "Based upon this new clinical data, Sanofi-Aventis plans to submit a registration dossier to the European Medicines Agency and a new drug application to the US Food and Drug Administration during the third quarter of 2008," according to a company press release.

Cautiously anticipating a dronedarone approval, as all experts heartwire contacted appear to be doing, Dr Andrea Natale (Texas Cardiac Arrhythmia Institute at St David's Medical Center, Austin, TX) said clinicians are hoping that the drug's efficacy against AF will at least approach that of amiodarone while being free of significant adverse effects. "We don't have that information right now, because the drug was compared with placebo," he noted, observing, however, that ATHENA's design was typical of drug trials intended to win regulatory approval.

Natale, who isn't connected with ATHENA, said dronedarone had "a good result" in the trial, "considering that even amiodarone, which is one of the most effective drugs we have [for AF], has a long-term efficacy only in the 50% range." He observed that no antiarrhythmic drug is going to be effective against AF in every patient; treatment is highly individualized.

Dr L Brent Mitchell

"So this is just another drug we can use, which is promising because it's proven to be more effective than placebo and in a range that is comparable to what we have available," he said. "It would be another option for our patients."

Dr L Brent Mitchell (Foothills Hospital, Calgary, AB) said to heartwire that he sees the dronedarone effect on ATHENA's primary composite outcome as "moderate to strong" and said its effect on AF "looked better than in the clinical trials preceding it. In ATHENA, for the first time, there's an antiarrhythmic drug for atrial fibrillation that actually seems to reduce [cardiovascular] mortality. That's going to make it very attractive, considering that other drugs when compared with placebo haven't done that—with the possible exception of amiodarone, and that's debatable." Mitchell wasn't involved in ATHENA.

A trade-off between efficacy and safety

"Any antiarrhythmic drug that's to be used for atrial fibrillation has to be compared with placebo," according to Mitchell. "The comparison of two active agents obscures any demonstration of harm." Pointing to the low rate of adverse side effects in the study, he said, "When you consider that atrial fibrillation is for most people a relatively benign arrhythmia, the safety profile of the drug is imperative. [Dronedarone] seemed to have a very good safety profile."

In his presentation, Hohnloser said the adverse event rates for dronedarone and placebo were 72% and 69%, respectively, a nonsignificant difference driven by a slight excess of gastrointestinal symptoms. Rates for adverse events that were respiratory in nature or skin- or thyroid-related, he said, "were absolutely identical between the two groups." Adverse events considered serious were seen in 20% and 21%, respectively; 13% and 8% of patients went off randomized therapy permanently due to side effects.

A lead investigator for two previous dronedarone trials [3], EURIDIS and ADONIS, and an advisor to the ATHENA trialists, Dr Bramah N Singh (University of California, Los Angeles), told heartwire that "in terms of safety, it is the best drug we have for atrial fibrillation. Efficacy? It has not been compared with any drug at this point."

According to Dr Fred Morady (University of Michigan, Ann Arbor), not an ATHENA investigator, the major advantage of dronedarone over amiodarone "is that it won't have long-term pulmonary toxicity. And I think that by itself will be enough motivation for most clinicians to switch over to dronedarone."

Morady continued, "My hunch is that in a head-to-head comparison, dronedarone will not be as effective as amiodarone for atrial fibrillation. It's a modest effect."

Also, he said, there will always be some patients for whom efficacy will be more important than the adverse-effect risk. "If it's someone with frequent shocks whose quality of life is very disturbed and they haven't responded well to other drugs—certainly with amiodarone, perhaps then you'd [accept] the increased risk of organ toxicity."

A mixed record in prior studies

As previously reported by heartwire, in the EURIDIS and ADONIS trials—both much smaller than ATHENA but differing only in geography and so combined for analysis—dronedarone's AF recurrence rate over 12 months was about 64%, modestly but significantly smaller than the 75% for placebo (p<0.001). A secondary end point composite of hospitalization or death was also reduced (23% vs 31%, respectively, p<0.01). Adverse events were generally the same in both groups.

When the trials were published in the September 6, 2007 issue of the New England Journal of Medicine, an accompanying editorial by Dr Michael D Ezekowitz (Lankenau Institute for Medical Research and Main Line Hospitals, Wynnewood, PA) said that head-to-head trials against amiodarone with longer follow-up would be needed to confirm the drugs' relative safety profiles [4].

However, Ezekowitz wrote, "The critical comparison should have been between dronedarone and amiodarone, both because amiodarone is highly efficacious in this setting and because the apparent goal in the development of dronedarone was to produce an agent that would be as effective as amiodarone but have fewer side effects." He also referred to a randomized dronedarone trial called ANDROMEDA that was halted early due to an apparent mortality increase associated with the drug in the trial's "high-risk" patients with systolic heart failure—as reported by heartwire in January 2003.

Safety in heart failure still a question

Dr Andrea Natale

According to Natale, "The hope was that we'd have this drug for heart-failure patients." But because of ANDROMEDA, he said, dronedarone "probably won't get approval immediately for heart-failure patients until they do a properly designed study for that group, those with atrial fibrillation."

At his presentation of ATHENA, Hohnloser observed that the trial had excluded patients with recently decompensated or NYHA class 4 heart failure, although there were some in NYHA class 2 or 3.

"We need to look in more detail, obviously, into the data from these heart-failure patients, but from all we can see, everything is very consistent in terms of hospitalization for heart failure," Hohnloser said. He pointed out that the rates of cardiac nonarrhythmic death, "most likely due to heart failure," weren't significantly different in the dronedarone and placebo groups, suggesting that in patients like those in ATHENA, "dronedarone appears to be very safe."

The next logical step, Hohnloser observed, is to compare dronedarone directly with amiodarone for maintenance of sinus rhythm in a randomized AF trial. Just such a trial, with the name DIONYSUS in keeping with the classical theme of prior dronedarone trials, is currently ongoing.

That DIONYSUS was chosen for the name might also represent wishful thinking on the part of Sanofi-Aventis and the trial's investigators, in that a favorable dronedarone performance against the more toxic drug, another step toward possible FDA approval, just might be a good reason to throw a party.

ATHENA was sponsored by Sanofi-Aventis. Hohnloser reports serving as a consultant, investigator, and/or advisory board member for Sanofi-Aventis, St Jude Medical, ARYx, Cardiome, Bristol-Myers Squibb, and Proctor & Gamble. ATHENA steering committee member Dr Richard L Page (University of Washington, Seattle) is an HRS vice president and chaired the society's 2008 Scientific Sessions Program Committee.