Comparing the Efficacy of Various Insulin Types: Pharmacokinetic and Pharmacodynamic Modeling of Glucose Clamp Effects in Healthy Volunteers

This study compares the pharmacokinetics and efficacy of various subcutaneously (SC) dosed insulin analogs, including rapid-acting, intermediate-acting, long-acting, and regular human insulin, using mechanistic pharmacokinetic (PK) and pharmacodynamic (PD) models. These models were applied to data from euglycemic clamp studies in healthy volunteers, where insulin pharmacokinetics and its effects on glucose utilization were monitored. Data from published studies were digitized and modeled using MONOLIX (Version 2024). The PK model described insulin absorption via sequential first-order processes and linear elimination. The PD effects were captured using a model combination of biophase, indirect, and receptor down-regulation components. While PK parameters-especially absorption rates-varied between insulin types, a common set of nonlinear PD parameters was sought to account for dose-related differences in glucose utilization. The maximum glucose stimulation ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:msub><mml:mi>S</mml:mi> <mml:mi>max</mml:mi></mml:msub> <mml:annotation>${{{\mathrm{S}}}_{{\mathrm{max}}}}$</mml:annotation></mml:semantics> </mml:math> ) was 163, and the insulin concentration for a half-maximal effect ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mi>S</mml:mi> <mml:msub><mml:mi>C</mml:mi> <mml:mn>50</mml:mn></mml:msub> </mml:mrow> <mml:annotation>${\mathrm{S}}{{{\mathrm{C}}}_{50}}$</mml:annotation></mml:semantics> </mml:math> ) were 1156 pmol/L for insulin lispro, regular human insulin, neutral protamine hagedorn (NPH) insulin, and insulin glargine; 674 pmol/L for insulin aspart; and 5335 pmol/L for insulin detemir. Insulin detemir showed similar overt effects as the other insulin types but with smaller clearances and lower potency. This mechanism-based glucose-insulin model demonstrated that most insulin analogs exhibit similar receptor- and transporter-related parameters. The model, with specific PK but unified PD parameters, may enable clinical optimization of insulin therapy by highlighting differences in pharmacokinetics and operating common intrinsic glucose utilization parameters.