FDA Panel Rejects Empagliflozin for Use in Type 1 Diabetes

Alicia Ault

November 14, 2019

SILVER SPRING, Maryland — A US Food and Drug Administration (FDA) advisory committee has recommended against approval of empagliflozin (Boehringer Ingelheim, Jardiance), a sodium-glucose cotransporter type 2 (SGLT2) inhibitor, as an adjunct to insulin therapy to improve glycemic control in adults with type 1 diabetes mellitus.

If the FDA follows the committee's advice, it would be the third drug in the class to be rejected for the treatment of type 1 diabetes.

The FDA turned down approval of another SGLT2 inhibitor, dapagliflozin (Forxiga, AstraZeneca), in July, as well as the dual SGLT1/SGLT2 inhibitor sotagliflozin (Zynquista, Sanofi/Lexicon) in March, as reported by Medscape Medical News.

The European Union, however, has approved both dapagliflozin and sotagliflozin as an adjunct to insulin for the treatment of type 1 diabetes.

Physician Panelists Cite Insufficient Data

The FDA's Endocrinologic and Metabolic Drugs Advisory Committee voted 14 to 2 that empagliflozin should not be approved for type 1 diabetes. The drug has been on the market for type 2 diabetes since 2014.

Most of the panelists said there were too few data to evaluate either the efficacy or the safety of empagliflozin — especially given that the SGLT2 inhibitor class is associated with an elevated risk for diabetic ketoacidosis (DKA) in the setting of type 1 diabetes.

The panel said that Boehringer needed to conduct longer-term studies with a much larger group of patients. The company submitted data from a single phase 3 trial — the Empagliflozin as Adjunctive to Insulin Therapy (EASE-3) — which included just 241 patients taking empagliflozin. Primary endpoints were collected at 26 weeks.

"I think we have insufficient data on the benefits and the risks of this drug," said panelist Jack A. Yanovski, MD, PhD, a pediatric endocrinologist at the Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland.

James De Lemos, MD, agreed: "I'm just completely unconvinced by the data provided because it's too small and too short an exposure period."

De Lemos, a professor of medicine and chair in cardiology at UT Southwestern Medical Center, Dallas, Texas, said he believes empagliflozin holds promise in this indication, but for now, "Giving a drug that can cause DKA to patients with type 1 diabetes makes me nervous intrinsically."

Brendan Everett, MD, MPH, FACC, FAHA, said he too thought empagliflozin could be an effective therapy.

But he said, "We owe this to patients with type 1 diabetes to do this right."

A 241-patient study "doesn't quite meet that bar in terms of respecting the breadth and severity of illness in the patient population that we're hoping to address with this therapy," said Everett, director of general cardiology inpatient service at Brigham and Women's Hospital, Boston, Massachusetts.

"It's out of respect for them that I voted no," he emphasized.

But Some Patients Call for Approval

But one of the patient advocates on the panel voted to recommend approval of empagliflozin, and some advocates who spoke at the public hearing also urged the panel to give the drug the thumbs up, noting that SGLT2 inhibitors are being used off label in type 1 diabetes anyway.

Bruce A. Perkins, MD, MPH, who helped coordinate the empagliflozin trials for Boehringer and who himself has type 1 diabetes, said that despite technologic advances, glucose is still not adequately controlled for people with type 1 diabetes.

"We need to acknowledge there's an epidemic of high A1C in the type 1 population." The ability of empagliflozin to reduce A1C "compares similarly to some of the incredibly complex interventions that we apply in type 1 diabetes," said Perkins, professor of medicine in the Division of Endocrinology and Metabolism, University of Toronto, Ontario, Canada.

And given the ongoing off-label use, the drug should be approved, said Anna McCollister-Slipp.

"Do we want people to take this class of medications without knowing what's happening?" she asked fellow panelists. "I would rather have that happen in a regulated environment" where manufacturers have greater responsibility to collect data, she stressed.

Kelly Close, cofounder of the nonprofit diaTribe Foundation, also argued during the public hearing that a "more conservative approval" could help physicians decide which patients with type 1 diabetes would be good candidates for an SGLT2 inhibitor.

It also would give physicians and patients more concrete guidance on side effects, Close said.

A diaTribe survey of some 90 patients with type 1 diabetes who were taking the drugs off label found that most were doing it to lower their A1C level, to have fewer instances of hyperglycemia, and to spend more time in their target range, she noted.

And McCollister-Slipp, who is also a type 1 diabetes patient, stressed that patients "live with risk" every day. "You hear all the time about patients dying not because of this study drug, or other study drugs, but because of the inadequacy of treatments," she stressed.

Fellow panelist and advocate Carling Lellock, who also has type 1, said she was concerned about a heightened possibility of DKA with empagliflozin, but she noted, "It's a risk we're already taking."

Even so, she voted against approval because she wanted to see more data.

One physician panel member who voted to approve the drug said that though it would be nice to see more data, he did not expect different outcomes.

Given what's already known about the class, "the data were what I'd expected them to be," said Kashif Munir, MD, vice chief of diabetes and nutrition in the Division of Endocrinology, Diabetes and Nutrition at the University of Maryland School of Medicine, Baltimore.

"I didn't feel like this trial was a surprise or anything unexpected or anything I wouldn't have predicted," he said.

DKA Risk Not Fully Characterized, Some Cases Overlooked

In EASE-3, Boehringer came up with a handful of classifications of DKA and DKA-related events that were referred to a blinded safety committee for adjudication. The FDA reviewers were critical, however, of that classification system.

The company said that 2 of 241 patients taking the pivotal 2.5-mg dose of empagliflozin experienced "certain DKA," compared to three patients taking placebo. Four patients in each arm had "certain or potential DKA," while seven patients taking empagliflozin (and two taking placebo) had "unlikely DKA but ketosis" and zero patients (one taking placebo) had "unlikely DKA."

But FDA reviewer Mahtab Niyyati, MD, said the agency had concerns about the reliability of the adjudication and disagreed with what seemed to be a glossing over of the seriousness of some of the events by virtue of how they were classified.

Some cases that were classified as "unlikely DKA but ketosis" were clinically significant, serious events that necessitated hospitalization and prompt intervention, said Niyyati.

She also said the "unlikely" and the "potential" categories were "an artificial categorization of DKA risk."

Key is the fact that the FDA and panelists both were concerned that some cases may have been overlooked.

"I personally think some of the cases were missed or not appropriately categorized," said panel chairman Kenneth Burman, MD, chief of the Endocrine Section at Medstar Washington Hospital Center, Washington, DC.

He and other panel members also said that once the drug was widely available, DKA incidents could greatly increase.

Head of drug safety, metabolism, at Boehringer Ingelheim, Ona Kinduryte Schorling, MD, said: "We understand that the risk of DKA might be higher in a postmarketing setting."

As such, the company proposed a risk mitigation strategy with a host of measures designed to ensure that prescribers and patients are made aware of the risks, the signs and symptoms, and what to do in an emergency.

But many on the panel said the company had not submitted any evidence that such a strategy would work.

Effectiveness of 2.5-mg Dose of Empagliflozin Debated

Empagliflozin was initially approved in 2014 at doses of 10 mg and 25 mg as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes. For type 1 diabetes, Boehringer studied 2.5-mg, 10-mg, and 25-mg doses.

EASE-2, a double-blind, randomized, controlled phase 3 study, only investigated the 10- and 25-mg doses, which the company realized presented too high a risk for DKA. EASE-3 was the only pivotal study that included the 2.5-mg dose. The company also submitted a simulation model that was meant to serve as a substitute for a second phase 3 trial.

Results from the whole EASE program were first presented at the EASD meeting in 2018, as reported by Medscape Medical News.

Both Boehringer and the FDA reviewers agreed that at 26 weeks, the 2.5-mg dose achieved a significant reduction of A1C from baseline of 0.26%. But the FDA noted that the figure was a result of a 0.20% increase in the placebo group and a 0.5% reduction in the empagliflozin group.

Most panelists said they could not draw conclusions on whether the A1C reduction was significant or meaningful.

"I feel that this database on the dose of 2.5 mg is too small and too short of a time period to even know what the efficacy in terms of glycemic control may be," said Connie Newman, MD.

"The purported reduction of 0.26% may not actually be clinically meaningful," said panelist Newman, adjunct professor of medicine at NYU Langone School of Medicine, New York City.

De Lemos agreed: "The clinical meaningfulness of this small reduction is completely unclear."

He and Newman, along with others on the panel, also raised questions about how long the reduction might last.

And the FDA said that secondary measures, such as a 1.77-kg reduction in body weight when compared with placebo and a 2-mmHg drop in systolic blood pressure compared to placebo were too small to be significant, especially in patients whose measures at baseline were normal for both.

The data also showed no difference from placebo regarding hypoglycemia risk; there was no excess hypoglycemia, but also no reduction compared to placebo, said the agency.

Newman, among others on the panel, agreed that the need for additional safe and effective treatments for type 1 patients is great. She and others called for longer studies to verify the duration of treatment effect and to better characterize the safety of empagliflozin.

"There should be at least one additional clinical randomized controlled trial of at least 1 year duration with a consideration to including microvascular outcomes," said Newman, who added that any such research should perhaps be extended another year through an open-label study.

For more diabetes and endocrinology news, follow us on Twitter and Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.