1A-04 - Comparative Effectiveness of Current Glucose-Lowering Medications in Type 2 Diabetes Mellitus: Emulation of a modified GRADE trial

Background: The GRADE (Glycemia Reduction Approaches in Diabetes: A Comparative Effectiveness Study) randomized clinical trial evaluated the comparative effectiveness of four glucose-lowering medication classes in reducing hemoglobin A1c (HbA1c) levels in individuals with type 2 diabetes mellitus (T2DM) who had moderately elevated HbA1c and were on metformin monotherapy. However, GRADE did not include a sodium-glucose cotransporter-2 inhibitor (SGLT-2i) treatment arm, despite SGLT-2i currently being a widely used and important group of antidiabetic medications due to their established cardiovascular and kidney benefits, as they had not yet received regulatory approval at the time the trial was designed.

Objectives: To replicate and extend GRADE findings by emulating a modified trial including an SGLT-2i group.

Methods: In a US commercial claims database (Jan 2014–Aug 2023), we used propensity score weighting to balance 87 baseline covariates in eligible patients (T2DM, age ≥30 years, on metformin monotherapy, baseline HbA1c 6-9%); 5,489 initiated SGLT-2i, 3,617 glucagon-like peptide-1 receptor agonists (GLP-1RA), 5,069 dipeptidyl peptidase 4 inhibitors (DPP-4i), and 11,148 sulfonylureas (SU). The primary outcome was the first occurrence of HbA1c ≥7.0%. Secondary outcomes included major adverse cardiovascular events (MACE), hospitalization for heart failure (HHF), and hypoglycemia requiring inpatient or emergency care. Hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated using Cox regression models.

Results: We identified 8,092 individuals who were included in the weighted cohort and achieved balance in observed pre-exposure covariates. GLP-1RAs showed a lower risk of HbA1c ≥7.0% versus SUs (HR 0.73; 95%CI 0.68-0.78) and DPP-4is (HR 0.65; 95%CI 0.60-0.70). SGLT-2is had comparable risk to DPP-4is (HR 1.04; 95%CI 0.97-1.10) but higher risk than GLP-1RAs (HR 1.60; 95%CI 1.48-1.72) and SUs (HR 1.17; 95%CI 1.11-1.23). Secondary cardiovascular outcomes did not show differences except that SGLT2i was associated with a decrease in the risk of hospitalization for heart failure. Hypoglycemic events were infrequent across all groups, with the highest incidence rate among initiators of SU. Results were consistent across subgroups and sensitivity analyses.

Conclusions: GLP-1RAs were more effective at achieving glycemic control compared to SUs and DPP-4is, as observed in the GRADE trial, and to SGLT-2is when combined with metformin in patients with T2DM. These findings can guide glucose-lowering medication selection in combination with metformin for T2DM management.