FDA panel rejects vernakalant bid for AFib cardioversion indication


A Food and Drug Administration advisory committee has voted 11 to 2 against a recommendation that the agency approve a long-studied antiarrhythmic agent for cardioversion of recent-onset atrial fibrillation (AFib).

It was the second time before an FDA advisory panel for vernakalant (Brinavess, Correvio International Sàrl), which the agency had declined to approve in 2008 due to safety concerns. That time, however, its advisors had given the agency a decidedly positive recommendation.

Since then, registry data collected for the drug’s resubmission seemed only to raise further safety issues, especially evidence that a single infusion may cause severe hypotension and suppress left ventricular function.

Some members of the Cardiovascular and Renal Drugs Advisory Committee (CRDAC), including a number who voted against approval, expressed hopes for further research aimed at identifying specific AFib patient groups who might safely benefit from vernakalant.

Of note, the drug has long been available for AFib cardioversion in Europe, where there are a number of other pharmacologic options, and was recently approved in Canada.

“We do recognize there’s a significant clinical need here,” observed Paul M. Ridker, MD, MPH, of Harvard Medical School and Brigham and Women’s Hospital Boston, a CRDAC panelist.

The results of the safety study that Correvio presented to the panel were “pretty marginal,” Dr. Ridker said. Given the negative safety signals and the available cardioversion alternatives, he questioned whether vernakalant represented a “substantial advance versus just another option. Right now, I’m not convinced it’s a substantial advance.”

FDA representatives were skeptical about vernakalant when they walked into the meeting room, as noted in briefing documents they had circulated beforehand. The drug’s safety experience under consideration included one case of ventricular arrhythmia and cardiogenic shock in a treated patient without apparent structural heart disease, who subsequently died. That case was much discussed throughout the meeting.

In its resubmission of vernakalant to regulators, Correvio also pointed to a significant unmet need for AFib cardioversion options in the United States, given the few alternatives.

For example, ibutilide is FDA-approved for recent-onset AFib or atrial flutter; but as the company and panelists noted, the drug isn’t often used for that indication. Patients with recent-onset AFib are often put on rate-control meds without cardioversion. Or clinicians may resort to electrical cardioversion, which can be logistically cumbersome and require anesthesia and generally a hospital stay.

Oral or intravenous amiodarone and oral dofetilide, flecainide, and propafenone are guideline-recommended but not actually FDA-approved for recent-onset AFib, the company noted.

Correvio made its “pre-infusion checklist” a core feature of its case. It was designed to guide selection of patients for vernakalant cardioversion based on contraindications such as a systolic blood pressure under 100 mm Hg, severe heart failure, aortic stenosis, severe bradycardia or heart block, or a prolonged QT interval.

In his presentation to the panel, FDA medical officer Preston Dunnmon, MD, said the safety results from the SPECTRUM registry, another main pillar of support for the vernakalant resubmission, “are not reassuring.”

As reasons, Dr. Dunnmon cited likely patient-selection bias and its high proportion of patients who were not prospectively enrolled; 21% were retrospectively entered from records.

Moreover, “the proposed preinfusion checklist will not reliably predict which subjects will experience serious cardiovascular adverse events with vernakalant,” he said.

“Vernakalant has induced harm that cannot be reliably predicted, prevented, or in some cases, treated. In contrast to vernakalant, electrical cardioversion and ibutilide pharmacologic cardioversion can cause adverse events, but these are transient and treatable,” he said.

Many on the panel agreed. “I thought the totality of evidence supported the hypothesis that this drug has a potential for a fatal side effect in a disease that you can live with, potentially, and that there are other treatments for,” said Julia B. Lewis, MD, Vanderbilt Medical Center, Nashville, Tenn., who chaired the CRDAC panel.

“The drug clearly converts atrial fibrillation, although it’s only transient,” observed John H. Alexander, MD, MHSc, Duke University, Durham, N.C., one of the two panelists who voted to recommend approval of vernakalant.

“And, there clearly is a serious safety signal in some populations of patients,” he agreed. “However, I was more reassured by the SPECTRUM data.” There is likely to be a low-risk group of patients for whom vernakalant could represent an important option that “outweighs the relatively low risk of serious complications,” Dr. Alexander said.

“So more work needs to be done to clarify who are the low risk patients where it would be favorable.”

Panelist Matthew Needleman, MD, Walter Reed National Military Medical Center, Bethesda, Md., also voted in favor of approval.

“We’ve all known patients with normal ejection fractions who keep coming in with symptomatic atrial fib, want to get out of it quickly, and get back to their lives. So having an option like this I think would be good for a very select group of patients,” Dr. Needleman said.

But the preinfusion checklist and other potential ways to select low-risk patients for vernakalant could potentially backfire, warned John M. Mandrola, MD, Baptist Medical Associates, Louisville, Ky., from the panel.

The FDA representatives had presented evidence that the drug can seriously depress ventricular function, and that the lower cardiac output is what leads to hypotension, he elaborated in an interview after the meeting.

If the checklist is used to exclude hemodynamically unstable patients from receiving vernakalant, he said, “Then you’re really giving this drug to relatively healthy patients for convenience, to decrease hospitalization or the hospital stay.”

The signal for substantial harm, Dr. Mandrola said, has to be balanced against that modest benefit.

Moreover, those in whom the drug doesn’t work may be left in a worse situation, he proposed. Only about half of patients are successfully converted on vernakalant, the company and FDA data suggested. The other half of patients who don’t achieve sinus rhythm on the drug still must face the significant hazards of depressed ejection fraction and hypotension, a high price to pay for an unsuccessful treatment.

Dr. Mandrola is Chief Cardiology Correspondent for theheart.org | Medscape Cardiology; his disclosure statement states no relevant financial relationships.

This article first appeared on Medscape.com.

Recommended Reading

AF risk is elevated after early-stage breast cancer diagnosis
MDedge Cardiology
Regular drinking a greater AFib risk than binge drinking
MDedge Cardiology
Native Americans appear to be at increased risk for AFib
MDedge Cardiology
Patients frequently drive too soon after ICD implantation
MDedge Cardiology
Getting high heightens stroke, arrhythmia risks
MDedge Cardiology
ATTEST: AFib ablation slows progression to persistence
MDedge Cardiology
Evidence builds for AFib ablation’s efficacy in heart failure
MDedge Cardiology
Cardiac arrests peak with pollution in Japan
MDedge Cardiology
Uncertain generalizability limits AFib ablation in HFrEF
MDedge Cardiology
Genetic test stratified AFib patients with low CHA2DS2-VASc scores
MDedge Cardiology