After Metformin, What Comes Next? Empagliflozin?

Gregory A. Nichols, PhD


December 22, 2015

Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes

Zinman B, Wanner C, Lachin JM, et al; EMPA-REG OUTCOME Investigators
N Engl J Med. 2015;373:2117-2128

The EMPA-REG OUTCOME trial is part of the new wave of cardiovascular (CV) outcome trials required by the US Food and Drug Administration (FDA) for all new diabetes drugs. Empagliflozin is in the relatively new class of antihyperglycemics known as sodium-glucose cotransporter 2 (SGLT2) inhibitors, which reduce renal glucose reabsorption and thus increase glucose excretion through the urine.

Risk or Benefit?

EMPA-REG recruited participants from 42 countries. The investigators randomly assigned 7020 patients with established CV disease to receive 10 mg or 25 mg of empagliflozin or placebo once daily against a background of standard diabetes care. Background glucose-lowering therapy was to remain unchanged for the first 12 weeks after randomization; investigators were then encouraged to adjust glucose-lowering therapy to achieve glycemic control according to local guidelines.

The primary composite outcome was death from CV causes, nonfatal myocardial infarction, or nonfatal stroke, as analyzed in the pooled empagliflozin group vs the placebo group. Secondary outcomes included the primary outcome plus hospitalization for unstable angina, hospitalization for heart failure, and all-cause mortality. Effects on glycemia and other cardiometabolic risk factors were also assessed.

A Reduction in Risk

Compared with placebo, the pooled empagliflozin groups had a 38% lower risk for death from CV causes, a 32% lower risk for all-cause mortality, and a 35% lower risk for hospitalization for heart failure (all statistically significant). For independent analyses of the 10-mg dose and the 25-mg dose vs placebo, the risk reduction was virtually identical but did not reach significance because of smaller numbers of events in each dosage group. In addition, the glycated hemoglobin (A1c) level was significantly lower in each empagliflozin dosage group compared with placebo, ranging from -0.60 to 0.24 percentage points at weeks 12 and 206, respectively. Of interest, the dose response of empagliflozin was negligible.


Early Glycaemic Control in Metformin Users Receiving Their First Add-On Therapy: A Population-Based Study of 4,734 People with Type 2 Diabetes

Thomsen RW, Baggesen LM, Sogaard M, et al
Diabetologia. 2015;58:2247-2253

This observational cohort study used electronic databases in Northern Denmark from 2000 to 2012 to assess glycemic control of people who initiated metformin monotherapy and then added another glucose-lowering agent within 3 years. Specifically, the investigators examined attainment of A1c goals (<7.0% and <6.5% among "healthy" patients) within 2 to 6 months following the add-on therapy. They then compared these levels among patients who added one of five categories of second-line agents: sulfonylurea; dipeptidyl peptidase-4 (DPP-4) inhibitors; glucagon-like peptide-1 (GLP-1) receptor agonists; insulin; or other antihyperglycemic drugs (undefined in the study). Analyses used sulfonylurea as the reference category and adjusted for age, sex, preintensification A1c, diabetes duration, complications, and Charlson Comorbidity Index.

Do Second-Line Agents Differ in Effectiveness?

Of the 4734 patients in the analysis, 52% added sulfonylurea, 27% added a DPP-4 inhibitor, 7% added a GLP-1 receptor agonist, 8% added insulin, and 6% added something else. Of note, and as one might expect, that distribution was decidedly different depending on when the add-on occurred (eg, between 2000 and 2003 vs from 2010 to 2012).

In any case, the proportion reaching an A1c of <7% was generally similar across therapies (59% for sulfonylurea, 59% for DPP-4, 62% for GLP-1) although somewhat lower for insulin (42%) and "other" antihyperglycemics (50%). Absolute reductions in A1c were generally similar for each of the drug classes, ranging from -0.8 to -1.3 percentage points, except for insulin (-2.4 percentage points). However, the statistically adjusted relative risk of attaining A1c <7.0% was 6% lower for DPP-4, 10% higher for GLP-1, 12% lower for insulin, and 14% lower for other antihyperglycemics compared with sulfonylureas.

An important confounder was that the proportion of all metformin users who attained A1c targets of <7% after add-on therapy increased from 46% in 2000-2003 to 59% in 2010-2012, a time trend that was associated with decreasing preintensification A1c values over time.



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.