Methods for Predicting Diabetes Phase III Efficacy Outcome From Early Data: Superior Performance Obtained Using Longitudinal Approaches

Abstract

The link between glucose and HbA1c at steady state has previously been described using steady-state or longitudinal relationships. We evaluated five published methods for prediction of HbA1c after 26/28 weeks using data from four clinical trials. Methods (1) and (2): steady-state regression of HbA1c on fasting plasma glucose and mean plasma glucose, respectively, (3) an indirect response model of fasting plasma glucose effects on HbA1c, (4) model of glycosylation of red blood cells, and (5) coupled indirect response model for mean plasma glucose and HbA1c. Absolute mean prediction errors were 0.61, 0.38, 0.55, 0.37, and 0.15% points, respectively, for Methods 1 through 5. This indicates that predictions improved by using mean plasma glucose instead of fasting plasma glucose, by inclusion of longitudinal glucose data and further by inclusion of longitudinal HbA1c data until 12 weeks. For prediction of trial outcome, the longitudinal models based on mean plasma glucose (Methods 4 and 5) had substantially better performance compared with the other methods.

Improving Efficacy of Inhaled Technosphere Insulin (Afrezza) by Postmeal Dosing: In-silico Clinical Trial with the University of Virginia/Padova Type 1 Diabetes Simulator

“…Expected MPG 24h differences are then translated into estimated HbA1c differences, assuming a linear correlation between average glucose over 24 h and HbA1c with a slope of 28.7 mg/dL per 1% HbA1c. 16 , 17 The assessment of relative hyporisk versus insulin lispro is derived using the LBGI. 14 , 15 It was calculated from the simulated glucose profiles across 100 virtual patients, each receiving an individual dose after virtual uptitration following the given titration rule.…”

“…If the MPG 6h results from the meal test simulations apply to the three main meals in a day and average glucose concentrations remain at the level of baseline for the remaining 6 h, the average 24-h glucose concentration can be calculated and the expected difference in HbA1c for different treatment arms applying different dosing regimens/titration rules can be estimated using an established correlation. 16 , 17 When this algorithm is applied to compare the previously used TI dosing regimen against insulin lispro, the HbA1c levels of TI patients would be about 0.26% above that of insulin lispro patients ( Fig. 7B ).…”

Comparison of Power, Prognosis, and Extrapolation Properties of Four Population Pharmacodynamic Models of HbA1c for Type 2 Diabetes

“…This allows less flexibility than the IGRH model, which could explain the slightly lower power and slightly higher inaccuracy in 12‐week prognosis observed compared to the IGRH model. The predictive performance of the IGRH and ADOPT models has previously been investigated head‐to‐head . In that study, the IGRH model was used with fixed parameters, results slightly favoring the ADOPT model.…”

“…The purpose of a pharmacometric analysis varies greatly: selection of optimal doses and dosing regimens to maximize the HbA1c‐lowering effect in phase II, quantitative support for the decision of dose reduction in subsequent phase II trials, mechanistic understanding of drug effects on hemoglobin (Hb) and its impact on HbA1c, assessment of demographics and disease progression on HbA1c to inform patient inclusion criteria, evaluation of the impact of genetic pharmacokinetic differences on HbA1c, investigation of differences in HbA1c effects between once‐daily and twice‐daily dosing, and all prospective predictions of HbA1c in phase II using phase I short‐term glucose, phase III using phase II study data, and patients using data from healthy subjects . In most of these examples, pharmacometric analysis was also used to quantify uncertainty.…”

A Longitudinal HbA1c Model Elucidates Genes Linked to Disease Progression on Metformin

“…[17][18][19] To date, current mathematical models that capture the timecourse of HbA1c in relation to metformin therapy have been limited by small sample sizes and sparse measurements. 16,20,21 Furthermore, a comprehensive genetic analysis linking genetic variants to long-term HbA1c trajectories has not yet been performed and, consequently, there is no current knowledge regarding the influence of genetics on long-term HbA1c dynamics.…”