Sylvie Retout

Prediction of Shrinkage of Individual Parameters Using the Bayesian Information Matrix in Non-Linear Mixed Effect Models with Evaluation in Pharmacokinetics

ABSTRACTPurposeWhen information is sparse, individual parameters derived from a non-linear mixed effects model analysis can shrink to the mean. The objective of this work was to predict individual parameter shrinkage from the Bayesian information matrix (MBF). We 1) Propose and evaluate an approximation of MBF by First-Order linearization (FO), 2) Explore by simulations the relationship between shrinkage and precision of estimates and 3) Evaluate prediction of shrinkage and individual parameter precision.MethodsWe approximated MBF using FO. From the shrinkage formula in linear mixed effects models, we derived the predicted shrinkage from MBF. Shrinkage values were generated for parameters of two pharmacokinetic models by varying the structure and the magnitude of the random effect and residual error models as well as the design. We then evaluated the approximation of MBF FO and compared it to Monte-Carlo (MC) simulations. We finally compared expected and observed shrinkage as well as the predicted and estimated Standard Errors (SE) of individual parameters.ResultsMBF FO was similar to MBF MC. Predicted and observed shrinkages were close . Predicted and estimated SE were similar.ConclusionsMBF FO enables prediction of shrinkage and SE of individual parameters. It can be used for design optimization.

Powers of the Likelihood Ratio Test and the Correlation Test Using Empirical Bayes Estimates for Various Shrinkages in Population Pharmacokinetics

We compared the powers of the likelihood ratio test (LRT) and the Pearson correlation test (CT) from empirical Bayes estimates (EBEs) for various designs and shrinkages in the context of nonlinear mixed‐effect modeling. Clinical trial simulation was performed with a simple pharmacokinetic model with various weight (WT) effects on volume (V). Data sets were analyzed with NONMEM 7.2 using first‐order conditional estimation with interaction and stochastic approximation expectation maximization algorithms. The powers of LRT and CT in detecting the link between individual WT and V or clearance were computed to explore hidden or induced correlations, respectively. Although the different designs and variabilities could be related to the large shrinkage of the EBEs, type 1 errors and powers were similar in LRT and CT in all cases. Power was mostly influenced by covariate effect size and, to a lesser extent, by the informativeness of the design. Further studies with more models are needed.

Correction to: A phase III randomized trial of gantenerumab in prodromal Alzheimer’s disease

Following publication of the original article [1], the author reported errors in the formatting of the table. The details of the errors are as follows:

Comparison of Model Averaging and Model Selection in Dose Finding Trials Analyzed by Nonlinear Mixed Effect Models

In drug development, pharmacometric approaches consist in identifying via a model selection (MS) process the model structure that best describes the data. However, making predictions using a selected model ignores model structure uncertainty, which could impair predictive performance. To overcome this drawback, model averaging (MA) takes into account the uncertainty across a set of candidate models by weighting them as a function of an information criterion. Our primary objective was to use clinical trial simulations (CTSs) to compare model selection (MS) with model averaging (MA) in dose finding clinical trials, based on the AIC information criterion. A secondary aim of this analysis was to challenge the use of AIC by comparing MA and MS using five different information criteria. CTSs were based on a nonlinear mixed effect model characterizing the time course of visual acuity in wet age-related macular degeneration patients. Predictive performances of the modeling approaches were evaluated using three performance criteria focused on the main objectives of a phase II clinical trial. In this framework, MA adequately described the data and showed better predictive performance than MS, increasing the likelihood of accurately characterizing the dose-response relationship and defining the minimum effective dose. Moreover, regardless of the modeling approach, AIC was associated with the best predictive performances.