Commentary
ROCKVILLE, MD. – In a 12-11 vote, a Food and Drug Administration advisory panel just barely came down in favor of the agency adding a new labeling entry to the already-approved diabetes drug empagliflozin (Jardiance) that would say the drug reduces cardiovascular mortality.
While several members of the panel wished the FDA’s staff good luck in weighing both the evidence and the advisory committee’s closely split endorsement when deciding whether to grant this unprecedented labeling to a diabetes drug, the fact that a majority of panelists favored this course marked a watershed moment in the development of new agents for treating hyperglycemia.
Mitchel L. Zoler/Frontline Medical News
Dr. Marvin A. Konstam
“It’s the first time we have evidence that a diabetes drug can reduce cardiovascular risk. That’s never been seen before, and it’s huge,” said Marvin A. Konstam, MD, a temporary member of the Endocrinologic and Metabolic Drugs Advisory Committee (EMDAC) and chief physician executive of the cardiovascular center at Tufts Medical Center in Boston. “The question is whether or not this effect is real, and the vote was 50-50, but I think there is a good chance it’s real, and, if so, it’s a game changer,” Dr. Konstam said in an interview. He voted in favor of the new labeling, and, like many of his colleagues on the panel, he admitted to agonizing over the decision during the postvote comment period.
What he and the other committee members struggled with was a remarkably strong effect by empagliflozin on reducing cardiovascular mortality by a relative 38%, compared with placebo, in more than 7,000 patients with type 2 diabetes selected for their high cardiovascular disease risk. The major sticking point was that the study enrolled patients into a randomized, placebo-controlled trial that was primarily designed to test the drug’s cardiovascular safety and not its efficacy, and where cardiovascular death was not even a prespecified secondary endpoint.
“It’s very hard to go from safety to superiority in one study,” said Peter W.F. Wilson, MD, who voted against the added indication. Like many panel members who voted no, Dr. Wilson said that any claim to preventing cardiovascular mortality with empagliflozin should meet the standard FDA requirement to have consistent results from at least two studies. “This is the first drug in its class [the sodium-glucose cotransporter 2 inhibitors], and we should have a high bar for the quality of the evidence,” said Dr. Wilson, professor of medicine and public health at Emory University in Atlanta.
“There is substantial evidence [to support the mortality claim], but not yet to the extent to put it on the label,” said another voter on the no side, Judith Fradkin, MD, also a temporary committee member and director of the division of diabetes, endocrinology and metabolic diseases at the National Institutes of Health. The data collected so far in favor of the mortality claim “are very compelling, but what I couldn’t get past is my long-standing belief that a positive result to a study’s secondary outcome is hypothesis generating. We need a second study to put this on the label,” Dr. Fradkin said.
That dramatic and highly meaningful clinical effect of empagliflozin on cardiovascular mortality jumped out at the investigators who ran the EMPA-REG OUTCOME trial as well as to many others from the moment the results had their unveiling less than a year ago, at the annual meeting of the European Association for the Study of Diabetes in Stockholm, and in a concurrently published article (N Engl J Med. 2015 Nov 26;373[22]:2117-28).
Mitchel L. Zoler/Frontline Medical News
Dr. Stuart Pocock
The primary efficacy endpoint placed into the EMPA-REG OUTCOME safety trial as the study developed following the 2008 FDA mandate for cardiovascular safety trials for all new hypoglycemic drugs was a three-part, combined-outcome endpoint of cardiovascular death, nonfatal MI, and nonfatal stroke. Although this primary, combined endpoint had a statistically significant but much more modest benefit with a 14% relative risk reduction, compared with placebo, “the benefit was all driven by the reduction in cardiovascular death that had an astonishing P value of less than .0001 with no suggestion of benefit or risk for MI or stroke,” said Stuart Pocock, PhD, professor of medical statistics at the London School of Hygiene and Tropical Medicine who appeared before the committee as a consultant brought in by the applicant, Boehringer Ingelheim.
The total 309 cardiovascular deaths seen during the study that led to this finding provided more data than most cardiovascular trials, and while in some respects, the cardiovascular benefit seemed “too good to be true,” it also turned out that “the data were so strong that they overwhelm skepticism,” Dr. Pocock told the panel while presenting some advanced statistical test results to prove this assertion. The trial results showed “overwhelming evidence of benefit, beyond a reasonable doubt,” and while cardiovascular death was just one part of the efficacy endpoint, “mortality merits special attention,” he said. The statistical analyses also showed an equally robust 32% relative risk reduction in all-cause mortality, and both the cardiovascular and all-cause death benefits seen in EMPA-REG OUTCOME were consistent across both dosages of empagliflozin tested in the study (10 mg and 25 mg daily) and across the sensitivity analyses applied by the investigators.