Strong mortality data lifted Boehringer Ingelheim GMBH and Eli Lilly & Co. to a positive FDA advisory committee review of a cardiovascular benefit claim for Jardiance (empagliflozin), but the panel's closely split vote likely means the question is far from settled for FDA.

At a June 28 meeting, the Endocrinologic and Metabolic Drugs Advisory Committee voted 12-11 that results from the EMPA-REG OUTCOME trial provide substantial evidence to establish that the SGLT2 inhibitor reduces CV mortality in the population studied.

If FDA agrees with the panel's recommendation by the early September user fee goal date, Jardiance and Synjardy (empagliflozin/metformin) would become the first type 2 diabetes drugs to be labeled for a CV benefit.

However, the split vote provides some cover for the agency if it decides to deny the claim. "Good luck figuring out what to do," panelist and cardiologist Marvin Konstam, Tufts University, said in remarks directed to FDA staff after the vote.

CV Death Benefit Persuasive

The majority of panelists were swayed by a 38% reduced risk of CV death with empagliflozin, a finding that was coupled with a 32% reduced risk of all-cause mortality. The mortality data were strong enough to overcome the more marginal results from the primary endpoint of major adverse cardiovascular events (MACE), the robustness of which were questioned due to study design issues.

"I feel pretty confident that there's some benefit on cardiovascular mortality," biostatistician Michael Proschan, National Institute of Allergy and Infectious Diseases, said. "I think it's likely that that 38% is an overstatement. ... But I think even if it's 20%, that's still very good."

"I was impressed with the magnitude of reduction of CV and all-cause mortality, which I think really represents the hardest of hard outcomes," said Susan Yanovski, co-director of the Office of Obesity Research at the National Institute of Diabetes and Digestive and Kidney Diseases.

Panelists who voted against approval of a new claim called the CV mortality data "intriguing" but said that given methodological and study design issues with EMPA-REG, the benefit should be confirmed in another trial.

Since empagliflozin would be first diabetes drug to gain a CV benefit claim, "I think we should have a higher bar for the quality of the evidence," said Peter Wilson, an epidemiologist at Emory University.

EMPA-REG was originally designed to satisfy FDA's December 2008 guidance on demonstrating CV safety of new type 2 diabetes drugs, but the study was subsequently expanded to also demonstrate CV superiority.

The committee voted 23-0 that the results satisfied the guidance's recommendations for demonstrating that the drug would not result in an unacceptable increase in CV risk.

Narrow Indication Gets Narrower

The panel's blessing of a CV mortality claim, albeit by the slimmest of margins, would appear to confirm the wisdom of the product sponsors' decision to seek a narrow claim rather than a broad indication for reduced risk of CV events based on EMPA-REG's primary endpoint ([A#14160517002]).

Boehringer, the sNDA sponsor, originally sought a claim for all-cause-mortality, CV death and heart failure-related hospitalization. However, in early May the company amended its proposed indication, opting to focus solely on CV death (see box).

FDA Division of Metabolism and Endocrinology Products Director Jean-Marc Guettier described the new proposal as "the current working version."

Original Proposed Indication

In adult patients with type 2 diabetes mellitus and high CV risk to reduce the risk of all-cause mortality by reducing the incidence of CV death and to reduce the risk of CV death or hospitalization for heart failure.

Current Proposed Indication

In adult patients with type 2 diabetes mellitus and established CV disease, Jardiance is indicated to reduce the incidence of CV death.

In an interview after the meeting, Thomas Seck, Boehringer VP of clinical development and medical affairs-metabolism, said the company decided, after consulting experts, to narrow the claim and focus on the most compelling benefit in EMPA-REG.

However, advisory committee briefing documents indicated that FDA reviewers took a dim view of the heart failure claim ([A#14160624001]). The agency also expressed reluctance about an all-cause mortality risk reduction claim, with the cardio-renal review division noting it rarely gives such indications for CV outcomes trials because this endpoint is typically driven by CV death.

Panelists said the reduction in heart failure-related hospitalization in EMPA-REG was a potentially very important clinical benefit but needed to be confirmed in a separate trial.

"I would not have voted yes on the original claim they made," Proschan said.

Boehringer and Lilly have announced plans for two new outcomes studies of empagliflozin and treatment of chronic heart failure in people with and without type 2 diabetes. Seck said one study would enroll patients with reduced ejection fraction, while the second trial would enroll patients with preserved ejection fraction.

'Ignoring' MACE Results

The proposed indication for empagliflozin is unusual in that it reflects only one component of the three-point composite MACE endpoint in the EMPA-REG trial.

Empagliflozin demonstrated a 14% reduced risk of MACE, which includes CV death, nonfatal myocardial infarction and nonfatal stroke. The results were driven by the CV death benefit. There was no benefit on nonfatal MI, and there was an unfavorable stroke imbalance that was not statistically significant.

Much of the panel's discussion and angst focused on the persuasiveness of the mortality findings given concerns about the robustness of the MACE results. The primary endpoint results narrowly met the statistical criteria for superiority. Further, the MACE results lost their superiority under various sensitivity analyses that involved inclusion of "silent" MIs and exclusion of non-assessable deaths.

Some panelists questioned the company's decision to forego the MACE results in favor of a claim for CV death.

"I don't understand how we're going from MACE, and the totality of the cardiovascular events, to a component of the MACE that seems selective to maximize the indicated benefit of the drug," cardiologist William Hiatt, University of Colorado, said. "There was no hierarchal testing on components of the primary endpoint or other secondary endpoints."

While acknowledging that the mortality benefit is "incredibly important," Hiatt nevertheless questioned how the panel could just "ignore" the other aspects of the primary endpoint because "legitimately the trial wasn't designed to test components of the primary the way you're doing it."

"We describe in our current indications statement the most important finding of trial, which is this large finding of cardiovascular mortality … being reduced remarkably," said Hans-Juergen Woerle, VP-metabolism therapeutic area for Boehringer. "We are of the belief that our data in the indication is best reflected in a descriptive statement that while we have reduced significantly three-point MACE, the most relevant finding to be described is the finding of cardiovascular mortality."

Boehringer consultant Stuart Pocock, a statistician at London School of Hygiene and Tropical Medicine, said that in general it's important to look at the driver for a composite endpoint.

"People can nitpick about type 1 errors ... but cardiovascular death is there. I think it's the most staggering result we've seen in cardiovascular outcome diabetes [trials] probably ever," he said. "Therefore to stay within the strict hierarchy of three-point MACE … I think is missing the point about the strength of evidence regarding cardiovascular death. … And obviously deaths are more important than nonfatal events."

Hiatt ultimately overcame his discomfort and voted for approval of a CV mortality benefit claim, but this hurdle was too high for some committee members to surmount.

"What I just couldn't get past was sort of my longstanding belief that a positive outcome to a secondary outcome is hypothesis generating, and I wasn't convinced that the primary outcome was still positive given some of the methodologic issues in the study," said Judith Fradkin, director of diabetes, endocrinology and metabolic diseases at NIDDK.

Jardiance Mortality Claim Narrowly Endorsed By FDA Advisory Committee

More from United States

More from North America