Objective: Empagliflozin, a potent, selective sodium glucose cotransporter 2 inhibitor, inhibits
renal glucose reabsorption, leading to glucosuria and improved fasting and postprandial
glucose. Due to its renal mechanism of action, we hypothesized that glucose lowering
with empagliflozin is independent of β-cell function and insulin resistance.
Methods: Using pooled data from three 24-week trials that investigated empagliflozin 10 mg
(n = 632) and empagliflozin 25 mg (n = 626) vs. placebo (n = 622) as monotherapy or
add-on to metformin or metformin plus sulfonylurea, we investigated the influence
of baseline HbA1c, body mass index (BMI), HOMA-B and HOMA-IR on change from baseline
in HbA1c using 2-way interaction models.
Results: Empagliflozin significantly reduced HbA1c from baseline vs. placebo at week 24; in
a model without interaction, mean HbA1c reductions vs. placebo were -0.65% with empagliflozin
10 mg and -0.70% with empagliflozin 25 mg. Baseline HbA1c had a significant influence
on the treatment effect (p < 0.001 for interaction). There were no interactions between
treatment and baseline BMI (p = 0.606), HOMA-B (p = 0.384) or HOMA-IR (p = 0.199).
Using a model including baseline HbA1c by treatment interaction, as baseline HbA1c
increased from 7.5% to 10.0%, estimated reductions in HbA1c with empagliflozin 10
mg increased from -0.48% to -1.32% and with empagliflozin 25 mg increased from -0.55%
to -1.33%.
Conclusion: Treatment effects of empagliflozin on HbA1c reductions appear to be driven by baseline
HbA1c and to be independent of baseline β-cell function, insulin sensitivity and BMI.
