SGLT2 Drug For T1D Gets Thumbs Down From FDA Panel

SILVER SPRING, Md. — An FDA advisory committee on Wednesday declined to recommend approving a low-dose version of empagliflozin (Jardiance) for adults with type 1 diabetes.

In a 2-14 vote, members of the Endocrinologic and Metabolic Drugs Advisory Committee felt the available data on the sodium-glucose cotransporter-2 (SGLT2) inhibitor as an adjunct to insulin suggested that the risks simply didn’t outweigh the benefits of Boehringer Ingelheim’s once-daily oral treatment.

“I don’t feel that the data presented today on empagliflozin 2.5 mg were acceptable from a risk-benefit profile to qualify for approval by the FDA,” stated panel member Thomas Weber, MD, of Duke University Medical Center in Durham, North Carolina.

“I do feel the data is very promising, and would recommend a more robust assessment of efficacy and safety,” he added. “Specifically, at least a 2-year clinical trial, adequately powered, and adequate to establish efficacy in HbA1c, and also [to] gather adequate patient-year exposure to more definitely and acceptably characterize the risk of diabetic ketoacidosis.”

Empagliflozin is already FDA approved in once-daily 10 mg and 25 mg doses for patients with type 2 diabetes as an adjunct to diet and exercise, and to reduce the risk of cardiovascular death in those with cardiovascular disease.

Looking to combine the benefits of the agent’s glucose-lowering ability while reducing the risk for diabetic ketoacidosis (DKA) seen with higher doses, Boehringer Ingelheim is seeking approval of empagliflozin at a 2.5-mg dose for adults with type 1 diabetes. They have also proposed a different trade name for this indication to reduce the likelihood that patients with type 1 diabetes accidentally receive the higher dose.

Many panel members pointed out limitations to the one available trial assessing this 2.5-mg dosage, including its small sample size and short duration. Other concerns raised throughout the day included the risk for potentially fatal DKA.

Although there are four currently approved SGLT2 inhibitors indicated for type 2 diabetes — empagliflozin, ertugliflozin (Steglatro), canagliflozin (Invokana), and dapagliflozin (Farxiga) — no agents in this class are approved for type 1 diabetes. A dual SGLT1/2 inhibitor, sotagliflozin (Zynquista), was turned down by the FDA in March for type 1 diabetes following a split advisory committee vote, mainly due to the risk for DKA.

The only approved, non-insulin therapy option for type 1 diabetes is the amylin analogue pramlintide (Symlin), injected along with mealtime insulin, underscoring the unmet need for more therapeutic options for this population.

Jennifer Green, MD, also of Duke University Medical Center, stated during Boehringer Ingelheim’s morning presentations that more type 1 patients are using non-approved adjuncts off-label — such as other SGLT2 inhibitors — than pramlintide.

Few discussions during Wednesday’s meeting revolved around empagliflozin’s glycemic efficacy. Rather, safety was the big concern, and particularly the risk of DKA. As seen in the phase III EASE-3 trial, the 241 patients randomized to the 2.5-mg dose of once-daily oral empagliflozin had a significant 0.28% (95% CI -0.42% to -0.15%) reduction in HbA1c compared with placebo over 26 weeks. The small study population was cited as a limitation by several panelists and as “borderline insulting” by another.

During data presentations, Boehringer Ingelheim pointed out that these HbA1c reductions are “comparable” to that achieved by pramlintide.

Some panel members raised the concern of whether this HbA1c benefit would be sustained long-term years down the road, beyond just the 26 weeks reported in the clinical trial.

In EASE-3, empagliflozin was not associated with any excess risk of hypoglycemia, but also wasn’t tied to a reduction in hypoglycemia risk. Compared with the placebo group, there was a modest benefit seen in body weight, with a 3.9 lb (1.77 kg) reduction for patients on empagliflozin, as well as 2.01 mm Hg and 0.35 mm Hg reductions in systolic and diastolic blood pressures.

The 52-week EASE-2 trial exclusively assessed 10 mg and 25 mg doses, but the higher rates of DKA at these doses (21 certain events with the 10 mg dose and 16 with the 25 mg dose) led to an FDA recommendation that the lower 2.5-mg dose be included in the subsequent EASE-3 trial.

There were two certain DKA events seen with the 2.5 mg empagliflozin dose versus three in the placebo group in EASE-3. These were labeled as mild. “Certain” events were defined as both acidosis and ketosis being present in the event. In addition, the rate of “potential” cases — where either acidosis or ketosis symptoms were present with symptoms suggestive of DKA — were similar between the low-dose empagliflozin group and placebo.

According to FDA briefing documents, Boehringer Ingelheim proposed a post-marking study to continue to assess the safety profile of the 2.5-mg dose in a real-world clinical practice setting, especially looking at the risk for DKA.

But prior and subsequent to the vote, nearly all panel members said that despite the huge unmet need for additional type 1 diabetes therapies, there are still too many lingering questions revolving around the safety of empagliflozin for type 1 diabetes. Based on the available data, several members explained that although a 0.3% HbA1c reduction is clinically meaningful, it’s still fairly modest.

Overall, nearly all the panel members said they were left uncertain of the risk-benefit profile due to the small sample size on this specific dosage and the short study period. Some suggested that if the 2.5 mg dose would have been included in the EASE-2 trial, there might have been sufficient data to ease their concerns.

Rita Kalyani, MD, MHS, of Johns Hopkins School of Medicine in Baltimore, said the EASE-3 trial “raised uncertainty regarding the adjudication of ketoacidosis and the reliability of those categories,” which is in part why she voted no. She continued, noting her concern with use of this drug in a real-world setting where patients won’t be as closely monitored and might not be checking blood ketones as frequently as the study population.

Also having voted no, Brendan M. Everett, MD, MPH, of Brigham and Women’s Hospital in Boston, stated: “We owe it to patients with type 1 diabetes to do this right and to provide sufficient evidence to know that we’re offering them a benefit and to well-characterize the risks. I think doing a study with 240 patients on active therapy doesn’t quite meet that bar, in terms of respecting the breadth and severity of illness [of] the patient population that we’re hoping to address with this therapy.”

Although the FDA is not required to follow its advisory committees’ recommendations, it typically does.-Med Page Today