The trouble with triples: Examining the impact of measurement error in mediation analysis.

Mediation analysis is used in genetic mapping studies to identify candidate gene mediators of quantitative trait loci (QTL). We consider genetic mediation analysis of triplets-sets of three variables consisting of a target trait, the genotype at a QTL for the target trait, and a candidate mediator that is the abundance of a transcript or protein whose coding gene co-locates with the QTL. We show that, in the presence of measurement error, mediation analysis can infer partial mediation even in the absence of a causal relationship between the candidate mediator and the target. We describe a measurement error model and a corresponding latent variable model with estimable parameters that are combinations of the causal effects and measurement errors across all three variables. The relative magnitudes of the latent variable correlations determine whether or not mediation analysis will tend to infer the correct causal relationship in large samples. We examine case studies that illustrate the common failure modes of genetic mediation analysis and demonstrate how to evaluate the effects of measurement error. While genetic mediation analysis is a powerful tool for identifying candidate genes, we recommend caution when interpreting mediation analysis findings.

A Bayesian model selection approach to mediation analysis.

Genetic studies often seek to establish a causal chain of events originating from genetic variation through to molecular and clinical phenotypes. When multiple phenotypes share a common genetic association, one phenotype may act as an intermediate for the genetic effects on the other. Alternatively, the phenotypes may be causally unrelated but share genetic loci. Mediation analysis represents a class of causal inference approaches used to determine which of these scenarios is most plausible. We have developed a general approach to mediation analysis based on Bayesian model selection and have implemented it in an R package, bmediatR. Bayesian model selection provides a flexible framework that can be tailored to different analyses. Our approach can incorporate prior information about the likelihood of models and the strength of causal effects. It can also accommodate multiple genetic variants or multi-state haplotypes. Our approach reports posterior probabilities that can be useful in interpreting uncertainty among competing models. We compared bmediatR with other popular methods, including the Sobel test, Mendelian randomization, and Bayesian network analysis using simulated data. We found that bmediatR performed as well or better than these alternatives in most scenarios. We applied bmediatR to proteome data from Diversity Outbred (DO) mice, a multi-parent population, and demonstrate the power of mediation with multi-state haplotypes. We also applied bmediatR to data from human cell lines to identify transcripts that are mediated through or are expressed independently from local chromatin accessibility. We demonstrate that Bayesian model selection provides a powerful and versatile approach to identify causal relationships in genetic studies using model organism or human data.

Quantification of the Impact of Partition Coefficient Prediction Methods on Physiologically Based Pharmacokinetic Model Output Using a Standardized Tissue Composition.

Tissue:plasma partition coefficients are key parameters in physiologically based pharmacokinetic (PBPK) models, yet the coefficients are challenging to measure in vivo. Several mechanistic-based equations have been developed to predict partition coefficients using tissue composition information and the compound's physicochemical properties, but it is not clear which, if any, of the methods is most appropriate under given circumstances. Complicating the evaluation, each prediction method was developed, and is typically employed, using a different set of tissue composition information, thereby making a controlled comparison impossible. This study proposed a standardized tissue composition for humans that can be used as a common input for each of the five frequently used prediction methods. These methods were implemented in R and were used to predict partition coefficients for 11 drugs, classified as strong bases, weak bases, acids, neutrals, and zwitterions. PBPK models developed in R (mrgsolve) for each drug and each set of partition coefficient predictions were compared with respective observed plasma concentration data. Percent root mean square error and half-life percent error were used to evaluate the accuracy of the PBPK model predictions using each partition coefficient method as summarized by strong bases, weak bases, acids, neutrals, and zwitterions characterization. The analysis indicated that no partition coefficient method consistently yielded the most accurate PBPK model predictions. As such, PBPK model predictions using all partition coefficient methods should be considered during drug development. SIGNIFICANCE STATEMENT: Several mechanistic-based methods exist to predict tissue:plasma partition coefficients critical to PBPK modeling. Controlled comparisons are confounded by the use of different tissue composition values for each method; a standardized tissue composition was proposed. Resulting assessments indicated that no method was consistently superior; therefore, sensitivity of PBPK predictions to each method may be warranted prior to model optimization.

Morphine Dose Optimization in Critically Ill Pediatric Patients With Acute Respiratory Failure: A Population Pharmacokinetic-Pharmacogenomic Study.

OBJECTIVE: To develop a pharmacokinetic-pharmacogenomic population model of morphine in critically ill children with acute respiratory failure. DESIGN: Prospective pharmacokinetic-pharmacogenomic observational study. SETTING: Thirteen PICUs across the United States. PATIENTS: Pediatric subjects (n = 66) mechanically ventilated for acute respiratory failure, weight greater than or equal to 7 kg, receiving morphine and/or midazolam continuous infusions. INTERVENTIONS: Serial blood sampling for drug quantification and a single blood collection for genomic evaluation. MEASUREMENTS AND MAIN RESULTS: Concentrations of morphine, the two main metabolites, morphine-3-glucuronide and morphine-6-glucuronide, were quantified by high-performance liquid chromatography tandem mass spectrometry/mass spectroscopy. Subjects were genotyped using the Illumina HumanOmniExpress genome-wide single nucleotide polymorphism chip. Nonlinear mixed-effects modeling was performed to develop the pharmacokinetic-pharmacogenomic model. A two-compartment model with linear elimination and two individual compartments for metabolites best describe morphine disposition in this population. Our analysis demonstrates that body weight and postmenstrual age are relevant predictors of pharmacokinetic parameters of morphine and its metabolites. Furthermore, our research shows that a duration of mechanical ventilation greater than or equal to 10 days reduces metabolite formation and elimination upwards of 30%. However, due to the small sample size and relative heterogeneity of the population, no heritable factors associated with uridine diphosphate glucuronyl transferase 2B7 metabolism of morphine were identified. CONCLUSIONS: The results provide a better understanding of the disposition of morphine and its metabolites in critically ill children with acute respiratory failure requiring mechanical ventilation due to nonheritable factors. It also provides the groundwork for developing additional studies to investigate the role of heritable factors.

Midazolam Dose Optimization in Critically Ill Pediatric Patients With Acute Respiratory Failure: A Population Pharmacokinetic-Pharmacogenomic Study.

OBJECTIVES: To develop a pharmacokinetic-pharmacogenomic population model of midazolam in critically ill children with primary respiratory failure. DESIGN: Prospective pharmacokinetic-pharmacogenomic observational study. SETTING: Thirteen PICUs across the United States. PATIENTS: Pediatric subjects mechanically ventilated for acute respiratory failure, weight greater than or equal to 7 kg, receiving morphine and/or midazolam continuous infusions. INTERVENTIONS: Serial blood sampling for drug quantification and a single blood collection for genomic evaluation. MEASUREMENTS AND MAIN RESULTS: Concentrations of midazolam, the 1' (1`-hydroxymidazolam metabolite) and 4' (4`-hydroxymidazolam metabolite) hydroxyl, and the 1' and 4' glucuronide metabolites were measured. Subjects were genotyped using the Illumina HumanOmniExpress genome-wide single nucleotide polymorphism chip. Nonlinear mixed effects modeling was performed to develop the pharmacokinetic-pharmacogenomic model. Body weight, age, hepatic and renal functions, and the UGT2B7 rs62298861 polymorphism are relevant predictors of midazolam pharmacokinetic variables. The estimated midazolam clearance was 0.61 L/min/70kg. Time to reach 50% complete mature midazolam and 1`-hydroxymidazolam metabolite/4`-hydroxymidazolam metabolite clearances was 1.0 and 0.97 years postmenstrual age. The final model suggested a decrease in midazolam clearance with increase in alanine transaminase and a lower clearance of the glucuronide metabolites with a renal dysfunction. In the pharmacogenomic analysis, rs62298861 and rs28365062 in the UGT2B7 gene were in high linkage disequilibrium. Minor alleles were associated with a higher 1`-hydroxymidazolam metabolite clearance in Caucasians. In the pharmacokinetic-pharmacogenomic model, clearance was expected to increase by 10% in heterozygous and 20% in homozygous for the minor allele with respect to homozygous for the major allele. CONCLUSIONS: This work leveraged available knowledge on nonheritable and heritable factors affecting midazolam pharmacokinetic in pediatric subjects with primary respiratory failure requiring mechanical ventilation, providing the basis for a future implementation of an individual-based approach to sedation.