Biomarkers, validation and pharmacokinetic-pharmacodynamic modelling

Abstract

Four elements are crucial to successful pharmacokinetic-pharmacodynamic (PK/PD) modelling and simulation for efficient and effective rational drug development: (i) mechanism-based biomarker selection and correlation to clinical endpoints; (ii) quantification of drug and/or metabolites in biological fluids under good laboratory practices (GLP); (iii) GLP-like biomarker method validation and measurements and; (iv) mechanism-based PK/PD modelling and validation. Biomarkers can provide great predictive value in early drug development if they reflect the mechanism of action for the intervention even if they do not become surrogate endpoints. PK/PD modelling and simulation can play a critical role in this process. Data from genomic and proteomics differentiating healthy versus disease states lead to biomarker discovery and identification. Multiple genes control complex diseases via hosts of gene products in biometabolic pathways and cell/organ signal transduction. Pilot exploratory studies should be conducted to identify pivotal biomarkers to be used for predictive clinical assessment of disease progression and the effect of drug intervention. Most biomarkers are endogenous macromolecules, which could be measured in biological fluids. Many exist in heterogeneous forms with varying activity and immunoreactivity, posting challenges for bioanalysis. Reliable and selective assays could be validated under a GLP-like environment for quantitative methods. While the need for consistent reference standards and quality control monitoring during sample analysis for biomarker assays are similar to that of drug molecules, many biomarkers have special requirements for sample collection that demand a well coordinated team management. Bioanalytical methods should be validated to meet study objectives at various drug development stages, and possess adequate performance to quantify biochemical responses specific to the target disease progression and drug intervention. Protocol design to produce sufficient data for PK/PD modelling would be more complex than that of PK. Knowledge of mechanism from discovery and preclinical studies are helpful for planning clinical study designs in cascade, sequential, crossover or replicate mode. The appropriate combination of biomarker identification and selection, bioanalytical methods development and validation for drugs and biomarkers, and mechanism-based PK/PD models for fitting data and predicting future clinical endpoints/outcomes provide powerful insights and guidance for effective and efficient rational drug development, toward safe and efficacious medicine for individual patients.