Abstract
PurposeIn this paper we investigated a new method for dose-response analysis of longitudinal data in terms of precision and accuracy using simulations.MethodsThe new method, called Dose-Response Mixed Models for Repeated Measures (DR-MMRM), combines conventional Mixed Models for Repeated Measures (MMRM) and dose-response modeling. Conventional MMRM can be applied for highly variable repeated measure data and is a way to estimate the drug effect at each visit and dose, however without any assumptions regarding the dose-response shape. Dose-response modeling, on the other hand, utilizes information across dose arms and describes the drug effect as a function of dose. Drug development in chronic kidney disease (CKD) is complicated by many factors, primarily by the slow progression of the disease and lack of predictive biomarkers. Recently, new approaches and biomarkers are being explored to improve efficiency in CKD drug development. Proteinuria, i.e. urinary albumin-to-creatinine ratio (UACR) is increasingly used in dose finding trials in patients with CKD. We use proteinuria to illustrate the benefits of DR-MMRM.ResultsThe DR-MMRM had higher precision than conventional MMRM and less bias than a dose-response model on UACR change from baseline to end-of-study (DR-EOS).ConclusionsDR-MMRM is a promising method for dose-response analysis.
Highlights
Traditional dose-response analyses are strongly dependent on the choice of model when the response is highly variable due to unexplained variability
Dose-Response Mixed Models for Repeated Measures (DR-Mixed Models for Repeated Measures (MMRM)) is a promising method for doseresponse analysis
For the dose-response informed methods, the bias increased with increasing ED50 which may be due to that the studied dose range (3–100 mg) covered a smaller part of the true dose-response relationship
Summary
Traditional dose-response analyses are strongly dependent on the choice of model when the response is highly variable due to unexplained variability. The model uncertainty results in large sample size and/or uncertainty in the dose selection for the following study. At each visit a placebo and treatment response is estimated independent of other visits. This approach has proved superior in terms of precision and accuracy to analysis of (co)variance (ANOVA/ANCOVA) with the end-of-study data in cases with dropout, where the traditional alternative is to use last observation carried forward (LOCF) [4,5,6,7,8,9,10,11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
