Model selection and averaging of nonlinear mixed-effect models for robust phase III dose selection

Integration of individual preclinical and clinical anti-infective PKPD data to predict clinical study outcomes

The AIDA randomized clinical trial found no significant difference in clinical failure or survival between colistin monotherapy and colistin-meropenem combination therapy in carbapenem-resistant Gram-negative infections. The aim of this reverse translational study was to integrate all individual preclinical and clinical pharmacokinetic-pharmacodynamic (PKPD) data from the AIDA trial in a pharmacometric framework to explore whether individualized predictions of bacterial burden were associated with the trial outcomes. The compiled dataset included for each of the 207 patients was (i) information on the infecting Acinetobacter baumannii isolate (minimum inhibitory concentration, checkerboard assay data, and fitness in a murine model), (ii) colistin plasma concentrations and colistin and meropenem dosing history, and (iii) disease scores and demographics. The individual information was integrated into PKPD models, and the predicted change in bacterial count at 24 h for each patient, as well as patient characteristics, was correlated with clinical outcomes using logistic regression. The in vivo fitness was the most important factor for change in bacterial count. A model-predicted growth at 24 h of ≥2-log10 (164/207) correlated positively with clinical failure (adjusted odds ratio, aOR = 2.01). The aOR for one unit increase of other significant predictors were 1.24 for SOFA score, 1.19 for Charlson comorbidity index, and 1.01 for age. This study exemplifies how preclinical and clinical anti-infective PKPD data can be integrated through pharmacodynamic modeling and identify patient- and pathogen-specific factors related to clinical outcomes - an approach that may improve understanding of study outcomes.

The Role of Model Master Files for Sharing, Acceptance, and Communication with FDA

With the evolving role of Model Integrated Evidence (MIE) in generic drug development and regulatory applications, the need for improving Model Sharing, Acceptance, and Communication with the FDA is warranted. Model Master File (MMF) refers to a quantitative model or a modeling platform that has undergone sufficient model Verification & Validation to be recognized as sharable intellectual property that is acceptable for regulatory purposes. MMF provides a framework for regulatorily acceptable modeling practice, which can be used with confidence to support MIE by both the industry and the U.S. Food and Drug Administration (FDA). In 2022, the FDA and the Center for Research on Complex Generics (CRCG) hosted a virtual public workshop to discuss the best practices for utilizing modeling approaches to support generic product development. This report summarizes the presentations and panel discussions of the workshop symposium entitled "Model Sharing, Acceptance, and Communication with the FDA". The symposium and this report serve as a kick-off discussion for further utilities of MMF and best practices of utilizing MMF in drug development and regulatory submissions. The potential advantages of MMFs have garnered acknowledgment from model developers, industries, and the FDA throughout the workshop. To foster a unified comprehension of MMFs and establish best practices for their application, further dialogue and cooperation among stakeholders are imperative. To this end, a subsequent workshop is scheduled for May 2-3, 2024, in Rockville, Maryland, aiming to delve into the practical facets and best practices of MMFs pertinent to regulatory submissions involving modeling and simulation methodologies.

Statistical recommendations for count, binary, and ordinal data in rare disease cross-over trials

BACKGROUND

Recommendations for statistical methods in rare disease trials are scarce, especially for cross-over designs. As a result various state-of-the-art methodologies were compared as neutrally as possible using an illustrative data set from epidermolysis bullosa research to build recommendations for count, binary, and ordinal outcome variables. For this purpose, parametric (model averaging), semiparametric (generalized estimating equations type [GEE-like]) and nonparametric (generalized pairwise comparisons [GPC] and a marginal model implemented in the R package nparLD) methods were chosen by an international consortium of statisticians.

RESULTS

It was found that there is no uniformly best method for the aforementioned types of outcome variables, but in particular situations, there are methods that perform better than others. Especially if maximizing power is the primary goal, the prioritized unmatched GPC method was able to achieve particularly good results, besides being appropriate for prioritizing clinically relevant time points. Model averaging led to favorable results in some scenarios especially within the binary outcome setting and, like the GEE-like semiparametric method, also allows for considering period and carry-over effects properly. Inference based on the nonparametric marginal model was able to achieve high power, especially in the ordinal outcome scenario, despite small sample sizes due to separate testing of treatment periods, and is suitable when longitudinal and interaction effects have to be considered.

CONCLUSION

Overall, a balance has to be found between achieving high power, accounting for cross-over, period, or carry-over effects, and prioritizing clinically relevant time points.

How to analyze continuous and discrete repeated measures in small-sample cross-over trials?

To optimize the use of data from a small number of subjects in rare disease trials, an at first sight advantageous design is the repeated measures cross-over design. However, it is unclear how these within-treatment period and within-subject clustered data are best analyzed in small-sample trials. In a real-data simulation study based upon a recent epidermolysis bullosa simplex trial using this design, we compare non-parametric marginal models, generalized pairwise comparison models, GEE-type models and parametric model averaging for both repeated binary and count data. The recommendation of which methodology to use in rare disease trials with a repeated measures cross-over design depends on the type of outcome and the number of time points the treatment has an effect on. The non-parametric marginal model testing the treatment-time-interaction effect is suitable for detecting between group differences in the shapes of the longitudinal profiles. For binary outcomes with the treatment effect on a single time point, the parametric model averaging method is recommended, while in the other cases the unmatched generalized pairwise comparison methodology is recommended. Both provide an easily interpretable effect size measure, and do not require exclusion of periods or subjects due to incompleteness.

Treatment of evolving cancers will require dynamic decision support

Cancer research has traditionally focused on developing new agents, but an underexplored question is that of the dose and frequency of existing drugs. Based on the modus operandi established in the early days of chemotherapies, most drugs are administered according to predetermined schedules that seek to deliver the maximum tolerated dose and are only adjusted for toxicity. However, we believe that the complex, evolving nature of cancer requires a more dynamic and personalized approach. Chronicling the milestones of the field, we show that the impact of schedule choice crucially depends on processes driving treatment response and failure. As such, cancer heterogeneity and evolution dictate that a one-size-fits-all solution is unlikely-instead, each patient should be mapped to the strategy that best matches their current disease characteristics and treatment objectives (i.e. their 'tumorscape'). To achieve this level of personalization, we need mathematical modeling. In this perspective, we propose a five-step 'Adaptive Dosing Adjusted for Personalized Tumorscapes (ADAPT)' paradigm to integrate data and understanding across scales and derive dynamic and personalized schedules. We conclude with promising examples of model-guided schedule personalization and a call to action to address key outstanding challenges surrounding data collection, model development, and integration.

Robust and Adaptive Two-stage Designs in Nonlinear Mixed Effect Models

To get informative studies for nonlinear mixed effect models (NLMEM), design optimization can be performed based on Fisher Information Matrix (FIM) using the D-criterion. Its computation requires knowledge about models and parameters, which are often prior guesses. Thus, adaptive designs composed of several stages may be used. Robust approach can also be used to account for various candidate models. In the estimation step of a given stage, model selection (MS) or model averaging (MA) can be performed. In this work we propose a new two-stage adaptive design strategy, based on the robust expected FIM and MA over several candidate models. The methodology is applied to a clinical trial simulation in ophthalmology to optimize doses and time measurements. A set of dose-response candidate models is defined, and one-stage designs are compared to two-stage 50/50 designs (i.e., each stage performed with half of the available subjects), using either local optimal design or robust design, and performing analysis with one model, MS or MA. Performing a two-stage design with MS at the interim analysis can correct the choice of a wrong model for designing the first stage. Overall, starting from a robust design (1- or 2-stage) is valuable and leads to reasonable bias and precision. The proposed robust adaptive design strategy is a new tool to design longitudinal studies that could be used in different therapeutic areas.

External Evaluation of Population Pharmacokinetic Models for Precision Dosing: Current State and Knowledge Gaps

Predicting drug exposures using population pharmacokinetic models through Bayesian forecasting software can improve individual pharmacokinetic/pharmacodynamic target attainment. However, selecting the most adapted model to be used is challenging due to the lack of guidance on how to design and interpret external evaluation studies. The confusion around the choice of statistical metrics and acceptability criteria emphasises the need for further research to fill this methodological gap as there is an urgent need for the development of standards and guidelines for external evaluation studies. Herein we discuss the scientific challenges faced by pharmacometric researchers and opportunities for future research with a focus on antibiotics.

Comparison of Seven Non-Linear Mixed Effect Model-Based Approaches to Test for Treatment Effect

Analyses of longitudinal data with non-linear mixed-effects models (NLMEM) are typically associated with high power, but sometimes at the cost of inflated type I error. Approaches to overcome this problem were published recently, such as model-averaging across drug models (MAD), individual model-averaging (IMA), and combined Likelihood Ratio Test (cLRT). This work aimed to assess seven NLMEM approaches in the same framework: treatment effect assessment in balanced two-armed designs using real natural history data with or without the addition of simulated treatment effect. The approaches are MAD, IMA, cLRT, standard model selection (STDs), structural similarity selection (SSs), randomized cLRT (rcLRT), and model-averaging across placebo and drug models (MAPD). The assessment included type I error, using Alzheimer's Disease Assessment Scale-cognitive (ADAS-cog) scores from 817 untreated patients and power and accuracy in the treatment effect estimates after the addition of simulated treatment effects. The model selection and averaging among a set of pre-selected candidate models were driven by the Akaike information criteria (AIC). The type I error rate was controlled only for IMA and rcLRT; the inflation observed otherwise was explained by the placebo model misspecification and selection bias. Both IMA and rcLRT had reasonable power and accuracy except under a low typical treatment effect.

Optimization of trial duration to predict long-term HbA1c change with therapy: A pharmacometrics simulation-based evaluation

Glycated hemoglobin (HbA1c) is the main biomarker of diabetes drug development. However, because of its delayed turnover, trial duration is rarely shorter than 12 weeks, and being able to predict long-term HbA1c with precision using data from shorter studies would be beneficial. The feasibility of reducing study duration was therefore investigated in this study, assuming a model-based analysis. The aim was to investigate the predictive performance of 24- and 52-week extrapolations using data from up to 4, 6, 8 or 12 weeks, with six previously published pharmacometric models of HbA1c. Predictive performance was assessed through simulation-based dose-response predictions and model averaging (MA) with two hypothetical drugs. Results were consistent across the methods of assessment, with MA supporting the results derived from the model-based framework. The models using mean plasma glucose (MPG) or nonlinear fasting plasma glucose (FPG) effect, driving the HbA1c formation, showed good predictive performance despite a reduced study duration. The models, using the linear effect of FPG to drive the HbA1c formation, were sensitive to the limited amount of data in the shorter studies. The MA with bootstrap demonstrated strongly that a 4-week study duration is insufficient for precise predictions of all models. Our findings suggest that if data are analyzed with a pharmacometric model with MPG or FPG with a nonlinear effect to drive HbA1c formation, a study duration of 8 weeks is sufficient with maintained accuracy and precision of dose-response predictions.

Improved Confidence in a Confirmatory Stage by Application of Item-Based Pharmacometrics Model: Illustration with a Phase III Active Comparator-Controlled Trial in COPD Patients

PURPOSE

The current study aimed to illustrate how a non-linear mixed effect (NLME) model-based analysis may improve confidence in a Phase III trial through more precise estimates of the drug effect.

METHODS

The FULFIL clinical trial was a Phase III study that compared 24 weeks of once daily inhaled triple therapy with twice daily inhaled dual therapy in patients with chronic obstructive pulmonary disease (COPD). Patient reported outcome data, obtained by using The Evaluating Respiratory Symptoms in COPD (E-RS:COPD) questionnaire, from the FULFIL study were analyzed using an NLME item-based response theory model (IRT). The change from baseline (CFB) in E-RS:COPD total score over 4-week intervals for each treatment arm was obtained using the IRT and compared with published results obtained with a mixed model repeated measures (MMRM) analysis.

RESULTS

The IRT included a graded response model characterizing item parameters and a Weibull function combined with an offset function to describe the COPD symptoms-time course in patients receiving either triple therapy (n = 907) or dual therapy (n = 894). The IRT improved precision of the estimated drug effect compared to MMRM, resulting in a sample size of at least 3.64 times larger for the MMRM analysis to achieve the IRT precision in the CFB estimate.

CONCLUSION

This study shows the advantage of IRT over MMRM with a direct comparison of the same primary endpoint for the two analyses using the same observed clinical trial data, resulting in an increased confidence in Phase III.

Application of Machine Learning Classification to Improve the Performance of Vancomycin Therapeutic Drug Monitoring

Bayesian therapeutic drug monitoring (TDM) software uses a reported pharmacokinetic (PK) model as prior information. Since its estimation is based on the Bayesian method, the estimation performance of TDM software can be improved using a PK model with characteristics similar to those of a patient. Therefore, we aimed to develop a classifier using machine learning (ML) to select a more suitable vancomycin PK model for TDM in a patient. In our study, nine vancomycin PK studies were selected, and a classifier was created to choose suitable models among them for patients. The classifier was trained using 900,000 virtual patients, and its performance was evaluated using 9000 and 4000 virtual patients for internal and external validation, respectively. The accuracy of the classifier ranged from 20.8% to 71.6% in the simulation scenarios. TDM using the ML classifier showed stable results compared with that using single models without the ML classifier. Based on these results, we have discussed further development of TDM using ML. In conclusion, we developed and evaluated a new method for selecting a PK model for TDM using ML. With more information, such as on additional PK model reporting and ML model improvement, this method can be further enhanced.

OUP accepted manuscript

Background

The antiviral efficacy of remdesivir in COVID-19 hospitalized patients remains controversial.

Objectives

To estimate the effect of remdesivir in blocking viral replication.

Methods

We analysed nasopharyngeal normalized viral loads from 665 hospitalized patients included in the DisCoVeRy trial (NCT 04315948; EudraCT 2020-000936-23), randomized to either standard of care (SoC) or SoC + remdesivir. We used a mathematical model to reconstruct viral kinetic profiles and estimate the antiviral efficacy of remdesivir in blocking viral replication. Additional analyses were conducted stratified on time of treatment initiation (≤7 or >7 days since symptom onset) or viral load at randomization (< or ≥3.5 log₁₀ copies/10⁴ cells).

Results

In our model, remdesivir reduced viral production by infected cells by 2-fold on average (95% CI: 1.5–3.2-fold). Model-based simulations predict that remdesivir reduced time to viral clearance by 0.7 days compared with SoC, with large inter-individual variabilities (IQR: 0.0–1.3 days). Remdesivir had a larger impact in patients with high viral load at randomization, reducing viral production by 5-fold on average (95% CI: 2.8–25-fold) and the median time to viral clearance by 2.4 days (IQR: 0.9–4.5 days).

Conclusions

Remdesivir halved viral production, leading to a median reduction of 0.7 days in the time to viral clearance compared with SoC. The efficacy was larger in patients with high viral load at randomization.

Pharmacometrics meets statistics-A synergy for modern drug development

Modern drug development problems are very complex and require integration of various scientific fields. Traditionally, statistical methods have been the primary tool for design and analysis of clinical trials. Increasingly, pharmacometric approaches using physiology-based drug and disease models are applied in this context. In this paper, we show that statistics and pharmacometrics have more in common than what keeps them apart, and collectively, the synergy from these two quantitative disciplines can provide greater advances in clinical research and development, resulting in novel and more effective medicines to patients with medical need.

Improved Decision-Making Confidence Using Item-Based Pharmacometric Model: Illustration with a Phase II Placebo-Controlled Trial

This study aimed to illustrate how a new methodology to assess clinical trial outcome measures using a longitudinal item response theory–based model (IRM) could serve as an alternative to mixed model repeated measures (MMRM). Data from the EXACT (Exacerbation of chronic pulmonary disease tool) which is used to capture frequency, severity, and duration of exacerbations in COPD were analyzed using an IRM. The IRM included a graded response model characterizing item parameters and functions describing symptom-time course. Total scores were simulated (month 12) using uncertainty in parameter estimates. The 50th (2.5th, 97.5th) percentiles of the resulting simulated differences in average total score (drug minus placebo) represented the estimated drug effect (95%CI), which was compared with published MMRM results. Furthermore, differences in sample size, sensitivity, specificity, and type I and II errors between approaches were explored. Patients received either oral danirixin 75 mg twice daily (n = 45) or placebo (n = 48) on top of standard of care over 52 weeks. A step function best described the COPD symptoms-time course in both trial arms. The IRM improved precision of the estimated drug effect compared to MMRM, resulting in a sample size of 2.5 times larger for the MMRM analysis to achieve the IRM precision. The IRM showed a higher probability of a positive predictive value (34%) than MMRM (22%). An item model–based analysis data gave more precise estimates of drug effect than MMRM analysis for the same endpoint in this one case study.

Assessing Treatment Effects with Pharmacometric Models: A New Method that Addresses Problems with Standard Assessments

Longitudinal pharmacometric models offer many advantages in the analysis of clinical trial data, but potentially inflated type I error and biased drug effect estimates, as a consequence of model misspecifications and multiple testing, are main drawbacks. In this work, we used real data to compare these aspects for a standard approach (STD) and a new one using mixture models, called individual model averaging (IMA). Placebo arm data sets were obtained from three clinical studies assessing ADAS-Cog scores, Likert pain scores, and seizure frequency. By randomly (1:1) assigning patients in the above data sets to “treatment” or “placebo,” we created data sets where any significant drug effect was known to be a false positive. Repeating the process of random assignment and analysis for significant drug effect many times (N = 1000) for each of the 40 to 66 placebo-drug model combinations, statistics of the type I error and drug effect bias were obtained. Across all models and the three data types, the type I error was (5th, 25th, 50th, 75th, 95th percentiles) 4.1, 11.4, 40.6, 100.0, 100.0 for STD, and 1.6, 3.5, 4.3, 5.0, 6.0 for IMA. IMA showed no bias in the drug effect estimates, whereas in STD bias was frequently present. In conclusion, STD is associated with inflated type I error and risk of biased drug effect estimates. IMA demonstrated controlled type I error and no bias.

A Model Averaging/Selection Approach Improves the Predictive Performance of Model-Informed Precision Dosing: Vancomycin as a Case Study

Many important drugs exhibit substantial variability in pharmacokinetics and pharmacodynamics leading to a loss of the desired clinical outcomes or significant adverse effects. Forecasting drug exposures using pharmacometric models can improve individual target attainment when compared with conventional therapeutic drug monitoring (TDM). However, selecting the "correct" model for this model-informed precision dosing (MIPD) is challenging. We derived and evaluated a model selection algorithm (MSA) and a model averaging algorithm (MAA), which automates model selection and finds the best model or combination of models for each patient using vancomycin as a case study, and implemented both algorithms in the MIPD software "TDMx." The predictive performance (based on accuracy and precision) of the two algorithms was assessed in (i) a simulation study of six distinct populations and (ii) a clinical dataset of 180 patients undergoing TDM during vancomycin treatment and compared with the performance obtained using a single model. Throughout the six virtual populations the MSA and MAA (imprecision: 9.9-24.2%, inaccuracy: less than ± 8.2%) displayed more accurate predictions than the single models (imprecision: 8.9-51.1%; inaccuracy: up to 28.9%). In the clinical dataset, the predictive performance of the single models applying at least one plasma concentration varied substantially (imprecision: 28-62%, inaccuracy: -16 to 25%), whereas the MSA or MAA utilizing these models simultaneously resulted in unbiased and precise predictions (imprecision: 29% and 30%, inaccuracy: -5% and 0%, respectively). MSA and MAA approaches implemented in TDMx might thereby lower the burden of fit-for-purpose validation of individual models and streamline MIPD.

Linked Pharmacometric-Pharmacoeconomic Modeling and Simulation in Clinical Drug Development

Market access and pricing of pharmaceuticals are increasingly contingent on the ability to demonstrate comparative effectiveness and cost-effectiveness. As such, it is widely recognized that predictions of the economic potential of drug candidates in development could inform decisions across the product life cycle. This may be challenging when safety and efficacy profiles in terms of the relevant clinical outcomes are unknown or highly uncertain early in product development. Linking pharmacometrics and pharmacoeconomics, such that outputs from pharmacometric models serve as inputs to pharmacoeconomic models, may provide a framework for extrapolating from early-phase studies to predict economic outcomes and characterize decision uncertainty. This article reviews the published studies that have implemented this methodology and used simulation to inform drug development decisions and/or to optimize the use of drug treatments. Some of the key practical issues involved in linking pharmacometrics and pharmacoeconomics, including the choice of final outcome measures, methods of incorporating evidence on comparator treatments, approaches to handling multiple intermediate end points, approaches to quantifying uncertainty, and issues of model validation are also discussed. Finally, we have considered the potential barriers that may have limited the adoption of this methodology and suggest that closer alignment between the disciplines of clinical pharmacology, pharmacometrics, and pharmacoeconomics, may help to realize the potential benefits associated with linked pharmacometric-pharmacoeconomic modeling and simulation.

Dose-Response Mixed Models for Repeated Measures – a New Method for Assessment of Dose-Response

Purpose

In this paper we investigated a new method for dose-response analysis of longitudinal data in terms of precision and accuracy using simulations.

Methods

The 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.

Results

The 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).

Conclusions

DR-MMRM is a promising method for dose-response analysis.

Electronic supplementary material

The online version of this article (10.1007/s11095-020-02882-0) contains supplementary material, which is available to authorized users.

Model Averaging in Viral Dynamic Models

The paucity of experimental data makes both inference and prediction particularly challenging in viral dynamic models. In the presence of several candidate models, a common strategy is model selection (MS), in which models are fitted to the data but only results obtained with the "best model" are presented. However, this approach ignores model uncertainty, which may lead to inaccurate predictions. When several models provide a good fit to the data, another approach is model averaging (MA) that weights the predictions of each model according to its consistency to the data. Here, we evaluated by simulations in a nonlinear mixed-effect model framework the performances of MS and MA in two realistic cases of acute viral infection, i.e., (1) inference in the presence of poorly identifiable parameters, namely, initial viral inoculum and eclipse phase duration, (2) uncertainty on the mechanisms of action of the immune response. MS was associated in some scenarios with a large rate of false selection. This led to a coverage rate lower than the nominal coverage rate of 0.95 in the majority of cases and below 0.50 in some scenarios. In contrast, MA provided better estimation of parameter uncertainty, with coverage rates ranging from 0.72 to 0.98 and mostly comprised within the nominal coverage rate. Finally, MA provided similar predictions than those obtained with MS. In conclusion, parameter estimates obtained with MS should be taken with caution, especially when several models well describe the data. In this situation, MA has better performances and could be performed to account for model uncertainty.

Optimal designs for frequentist model averaging

We consider the problem of designing experiments for estimating a target parameter in regression analysis when there is uncertainty about the parametric form of the regression function. A new optimality criterion is proposed that chooses the experimental design to minimize the asymptotic mean squared error of the frequentist model averaging estimate. Necessary conditions for the optimal solution of a locally and Bayesian optimal design problem are established. The results are illustrated in several examples, and it is demonstrated that Bayesian optimal designs can yield a reduction of the mean squared error of the model averaging estimator by up to 45%.

Dose Finding in the Clinical Development of 60 US Food and Drug Administration-Approved Drugs Compared With Learning vs. Confirming Recommendations

This review characterizes clinical development that supported the label dose in 60 drug indications recently approved by the US Food and Drug Administration. With Lewis B. Sheiner's Learning vs. Confirming clinical drug development paradigm as a reference point, the clinical development paths, the design of dose‐ranging trials, and the dose–exposure–response characterization were examined using US Food and Drug Administration approval packages. It was found that 89% of clinical development programs included several doses in the first‐in‐patient trial, 43% proceeded directly to confirmatory trials after the first‐in‐patient trial, and 52% included multiple doses in confirmatory development. A low number of doses and narrow dose ranges were generally included in dose‐ranging trials, with only 20% including at least four doses over an at least 10‐fold dose range. In a third of approval packages, no dose–response or exposure–response evaluation was identified, and model‐based dose–exposure–response characterization was rarely alluded to, as only 2 of 60 approval packages mentioned the use of a model‐based approach. The findings suggest that confirmatory development may often be guided more toward learning than confirming, and furthermore that dose exposure response is robustly assessed in only a minority of clinical drug development programs, indicating that there may be room left for optimizing the benefit/risk profile of confirmatory/marketed dose(s). Significant deviation from Learning vs. Confirming may exist in clinical development practice on several levels, and the reasons for why this may be the case are discussed in light of contemporary literature.

Robust designs accounting for model uncertainty in longitudinal studies with binary outcomes

To optimize designs for longitudinal studies analyzed by mixed-effect models with binary outcomes, the Fisher information matrix can be used. Optimal design approaches, however, require a priori knowledge of the model. We aim to propose, for the first time, a robust design approach accounting for model uncertainty in longitudinal trials with two treatment groups, assuming mixed-effect logistic models. To optimize designs given one model, we compute several optimality criteria based on Fisher information matrix evaluated by the new approach based on Monte-Carlo/Hamiltonian Monte-Carlo. We propose to use the DDS-optimality criterion, as it ensures a compromise between the precision of estimation of the parameters, and hence the Wald test power, and the overall precision of parameter estimation. To account for model uncertainty, we assume candidate models with their respective weights. We compute robust design across these models using compound DDS-optimality. Using the Fisher information matrix, we propose to predict the average power over these models. Evaluating this approach by clinical trial simulations, we show that the robust design is efficient across all models, allowing one to achieve good power of test. The proposed design strategy is a new and relevant approach to design longitudinal studies with binary outcomes, accounting for model uncertainty.

Feasibility of Exposure-Response Analyses for Clinical Dose-Ranging Studies of Drug Combinations

The exposure-response relationship of combinatory drug effects can be quantitatively described using pharmacodynamic interaction models, which can be used for the selection of optimal dose combinations. The aim of this simulation study was to evaluate the reliability of parameter estimates and the probability for accurate dose identification for various underlying exposure-response profiles, under a number of different phase II designs. An efficacy variable driven by the combined exposure of two theoretical compounds was simulated and model parameters were estimated using two different models, one estimating all parameters and one assuming that adequate previous knowledge for one drug is readily available. Estimation of all pharmacodynamic parameters under a realistic, in terms of sample size and study design, phase II trial, proved to be challenging. Inaccurate estimates were found in all exposure-response scenarios, except for situations where no pharmacodynamic interaction was present, with the drug potency and interaction parameters being the hardest to estimate. When previous knowledge of the exposure-response relationship of one of the monocomponents is available, such information should be utilized, as it enabled relevant improvements in parameter estimation and in correct dose identification. No general trends for classification of the performance of the tested study designs across different scenarios could be identified. This study shows that pharmacodynamic interactions models can be used for the exposure-response analysis of clinical endpoints especially when accompanied by appropriate dose selection in regard to the expected drug potencies and appropriate trial size and if information regarding the exposure-response profile of one monocomponent is available.