J. Mandema,1 S. Terra,2 K. Sweeney,3 V. Sahasrabudhe3; 1Quantitative Solutions, Inc., Menlo Park, CA, 2Pfizer, Inc., Andover, MA, 3Pfizer, Inc., Groton, CT

BACKGROUND: Sodium-glucose cotransporter-2 inhibitors (SGLT2i) have emerged as a new class of drugs for the treatment of type 2 diabetes mellitus (T2DM). In the current crowded anti-diabetic drug market, the use of objective tools to understand the predictors of efficacy and guide dose selection, go-no go decisions, and clinical trial strategy has become critically important.
METHODS: A model-based meta-analysis (MBMA) was undertaken to quantify the time course of dose vs. HbA1c response of SGLT2i and evaluate the impact of baseline HbA1c, eGFR (estimated glomerular filtration rate) and background diabetes treatment on HbA1c-lowering effect. A systematic literature review yielded 73 randomized controlled trials representing >30,000 T2DM patients and 8 SGLT2i (dapa-, cana-, empa-, ipra-, luseo-, tofo-, remogliflozin and LX-4211).
RESULTS: For SGLT2i, there was a 9.2% [5.8 to 12.5%; 95% confidence interval] decrease in HbA1c efficacy for every 10 mL/min/1.73m2 drop in eGFR. Patient populations with a mean baseline HbA1c of 7% and 9% had a 28% [25 to 31%] smaller and 33% [28 to 38%] greater response, respectively, than patient populations with a mean baseline of 8%. The HbA1c efficacy (adjusted for eGFR) was 26% [17 to 37%] greater on a diet background and 14% [5.6 to 22%] lower on an insulin background relative to a background of metformin (or other oral anti-diabetic agents).
CONCLUSION: This analysis provides a quantitative framework to understand the key predictors of the HbA1c-lowering effect of SGLT2i and allows an indirect comparison of SGLT2i across trials.