Abstract
OBJECTIVE: To evaluate how model-based parameters of β-cell function change with glucose-lowering treatment and associate with glycemic deterioration in adults with type 2 diabetes (T2D). RESEARCH DESIGN AND METHODS: In the Glycemia Reduction Approaches in Diabetes: A Comparative Effectiveness Study (GRADE), β-cell function parameters derived from mathematical modeling of oral glucose tolerance tests were assessed at baseline (N = 4,712) and 1, 3, and 5 years following randomization to insulin glargine, glimepiride, liraglutide, or sitagliptin, added to baseline metformin. Parameters included insulin secretion rate (ISR), glucose sensitivity (insulin response to glucose), rate sensitivity (early insulin response), and potentiation. Linear mixed-effects models were used to compare changes across treatments. With Cox proportional hazards and Classification And Regression Tree (CART) analyses we evaluated associations between model parameters and glycemic failure (A1C >7.5%; 58.5 mmol/mol). RESULTS: β-Cell function parameters increased variably at year 1 across treatments but subsequently declined for all treatments. Statistically significant changes were noted. Liraglutide led to the greatest increases in ISR, glucose sensitivity and potentiation, remaining above baseline at study end. Sitagliptin improved glucose sensitivity, with modest effects on other parameters. Glimepiride temporarily increased ISR and rate sensitivity but minimally increased glucose sensitivity or potentiation. Rate sensitivity increased most with glargine. Higher β-cell function parameters were protective against glycemic deterioration, but treatment did not alter the relationship between these parameters and glycemic outcomes. CONCLUSIONS: Common glucose-lowering medications impact different physiologic components of β-cell function in T2D. Regardless of treatment modality, lower β-cell function associated with early glycemic failure, and β-cell function progressively declined after initial improvement.