Abstract
P
In 2014 in the U.S., 245,000 patients with hypoglycemia checked into an emergency department, and 168,000 patients were discharged after being treated for DKA. 7 There is an urgent need for additional tools to help manage glycemia in type 1 diabetes, but available treatment options have so far been limited to insulin and pramlintide. The latter modestly reduces glycated hemoglobin and weight, but is associated with an increased risk of severe hypoglycemia and is used only by a few patients with type 1 diabetes. 1,8
Several antihyperglycemic agents (orals or injectables) approved for the treatment of type 2 diabetes have been evaluated as possible adjunctive therapies for type 1 diabetes. After sitagliptin failed to show any clinically meaningful effect on HbA1c levels, no further efforts were made to study dipeptidyl peptidase 4 (DPP4) inhibitors in type 1 diabetes. 8 –10 In ADJUNCT ONE, a randomized, placebo-controlled study involving almost 1400 patients, liraglutide 1.2 and 1.8 mg demonstrated significant placebo-adjusted reductions in HbA1c of 0.15% and 0.20%, respectively, with significant reductions in weight of 3.6 and 4.9 kg and insulin dose by 2% and 5% with the 1.2 and 1.8 mg doses, respectively. However, liraglutide was also associated with significant increases in the rate of symptomatic hypoglycemia as well as hyperglycemia with ketosis. Gastrointestinal adverse events were more frequent with liraglutide and were the most frequently cited reason for discontinuation by 15%, 13%, and 3% of patients in the liraglutide 1.8 and 1.2 mg and placebo groups, respectively. 11 Novo Nordisk, the sponsor of the ADJUNCT Trials, decided not to file for New Drug Application (NDA) to the Food and Drug Administration (FDA) or European Medicines Agency (EMA).
The REMOVAL (REducing with MetfOrmin Vascular Adverse Lesions) study offered a much-anticipated evaluation of the cardiovascular benefits of metformin in type 1 diabetes, which was prompted by metformin's well-established benefits in type 2 diabetes, which include improvements in cardiovascular risk factors as well as HbA1c. The primary endpoint was a 3-year progression of Carotid artery Intima-Media Thickness (CIMT) in 493 adults >40 years of age with type 1 diabetes and cardiovascular risk. Metformin treatment added to insulin therapy failed to significantly reduce CIMT progression during the study period. It was associated with a small but statistically significant decrease in average HbA1c of 0.13% over the 3 years, but this was due mainly to a reduction of 0.24% in the first 3 months that was not sustained afterward. Body weight and low-density lipoprotein cholesterol (LDL-C) decreased with metformin; there was no increase in hypoglycemia, but gastrointestinal adverse events were more frequent and led to discontinuation by 27% of metformin-treated patients vs. 12% of placebo-treated patients. 12 Its off-label use in clinical practice in patients with type 1 diabetes is very limited due to the side-effects profile.
To date, the most promising results have come from studies of the sodium glucose cotransporter (SGLT) inhibitors, which include selective SGLT2 inhibitors and dual inhibitors of SGLT1 and SGLT2. Three selective SGLT2 inhibitors are approved for the treatment of type 2 diabetes. In phase 2 studies in type 1 diabetes, empagliflozin and canagliflozin significantly reduced HbA1c, insulin dose, and weight. 13,14 In the 28-day empagliflozin study involving 75 patients, HbA1c decreased by up to 0.49%, weight by 2 kg, and total daily insulin dose by 0.09 IU/kg. No DKA occurred and the rates of hypoglycemia were similar between treatment groups during the 28-day study period. 13 In an 18-week study with 351 patients, canagliflozin 100 and 300 mg reduced HbA1c by 0.29% and 0.25% relative to placebo from a baseline of 7.9% and 8.0%, respectively.
Compared with results from the placebo group, weight decreased significantly by 2.8 and 4.4 kg, and total daily insulin dose decreased by 4.1 and 7.6 IU/day with canagliflozin 100 and 300 mg, respectively. Rates of documented hypoglycemia (blood glucose ≤70 mg/dL) and severe hypoglycemia (episodes requiring assistance from another person or resulting in seizure or loss of consciousness) were higher with canagliflozin than placebo, and DKA occurred in 4.3% and 6.0% of patients treated with canagliflozin 100 mg and 300 mg, respectively, compared with none in the placebo group. 15
Results of the phase 3 DEPICT 1 (Dapagliflozin Evaluation in Patients with Inadequately Controlled Type 1 Diabetes) study will be presented at the European Association for the Study of Diabetes (EASD) annual meeting in Lisbon, on September 14, 2017.
The only phase 3 study published to date involved 1402 adult patients with type 1 diabetes. The dual SGLT1 and SGLT2 inhibitor sotagliflozin 400 mg or placebo was added (in double-blind randomized fashion) to patients' existing insulin regimens, and significantly more sotagliflozin-treated (28.6%) than placebo-treated (15.2%) patients achieved a composite primary endpoint of HbA1c <7.0% without severe hypoglycemia or DKA during the 24-week treatment period. Sotagliflozin yielded significant placebo-adjusted decreases in HbA1c (−0.46%); weight (−3.0 kg); and mean daily total (−5.3 IU/day), bolus (−2.8 IU/day), and basal (−2.6 IU/day) insulin doses. Systolic blood pressure (SBP) also decreased significantly in patients with baseline SBP ≥130 mm Hg (−3.5 mm Hg), and SBP and diastolic blood pressure (DBP) also decreased significantly in the entire population—this study was the first to demonstrate significant blood pressure reductions with SGLT inhibition in type 1 diabetes. Severe hypoglycemia rates were similar (3.0% with sotagliflozin vs. 2.4% with placebo), while less documented hypoglycemia occurred in the sotagliflozin group (69.8 vs. 77.9 events per person-year of blood glucose ≤70 mg/dL and 11.8 vs. 15.4 events per person-year of blood glucose ≤55 mg/dL). However, a higher incidence of DKA was observed with sotagliflozin (3.0% vs. 0.6%). 16
SGLT2 inhibition has been associated with DKA in numerous case reports, and the FDA mandates DKA warnings on the prescribing information labels for the class. 14,15,17 –21 A recent review of 2500 cases of SGLT2 inhibitor–associated DKA suggested that 5% of patients with T1D treated off-label with SGLT2 inhibitors develop DKA. 21 Careful monitoring and appropriate discontinuation of treatment during metabolically stressful events (e.g., scheduled surgeries, physically taxing sporting events) is required. 17,20
The rates of DKA were lower with sotagliflozin than canagliflozin in their respective studies. Although no definitive conclusions about relative safety can be drawn, sotagliflozin's dual inhibition of SGLT1 and SGLT2 may offer an advantage. SGLT1 inhibition reduces glucose absorption in the proximal intestine, significantly blunting and delaying postprandial hyperglycemia, whereas SGLT2 inhibition decreases renal glucose reabsorption just like canagliflozin. 22 –24 These complementary effects may have led to more modest reductions in insulin dosages, which may have mitigated DKA risk, as decreases in insulin dose may be a risk factor for DKA. 17,25 Among sotagliflozin-treated patients the total, bolus, and basal daily insulin doses decreased by 9.7%, 12.3%, and 9.9%, respectively, relative to placebo. 16 Larger basal insulin dose reductions of 19% and 22% were reported with canagliflozin 100 and 300 mg, respectively. 15
None of the SGLT inhibitors are approved by the FDA or EMA for patients with type 1 diabetes. To date, studies with SGLT inhibitors have suggested that these agents may satisfy a long-unmet need in type 1 diabetes treatment for adjunctive therapy that can improve HbA1C without increasing the risk of weight gain or severe hypoglycemia. The increased risk of DKA with SGLT inhibition is, however, an important concern that warrants close-monitoring and further evaluations.