• Top
• Meeting abstract

Meeting abstract

The Working Group 'Markers of Pharmacological and Toxicological Action' has been established as part of the COST-B15 European Concerted Action 'Modelling in Drug Development'. Its objective is to develop a theoretical framework for, and to disseminate a European view on, the development, evaluation and validation of biomarkers.

In 1999 an 'NIH/FDA Working Group on Definitions' proposed definitions for 'Biomarkers' as well as 'Surrogate Endpoints'. According to these definitions a biomarker is a characteristic that is measured and evaluated as an indicator of normal biological processes, pathogenic processes or pharmacological responses. A surrogate endpoint is defined as a biomarker, which may serve as a substitute for a clinical endpoint (= an indicator of 'how a patient feels, functions or survives'). These definitions focus exclusively on the use of biomarkers as the basis for regulatory decisions (i.e. the approval of new drug products). So far, no proposals have been presented on experimental approaches that might be applied for the validation of biomarkers as surrogate endpoints.

According to the view of the COST-B15 Working Group, biomarkers have applicability that reaches beyond the regulatory decision-making process. Biomarkers are often used successfully for decision-making in early drug development (i.e. drug candidate selection, lead optimisation), well before information is available on their 'surrogacy' with respect to the clinical effects. At the other end of the spectrum, biomarkers may be used as the basis for drug selection and dose optimisation (individualisation) in clinical practice. These various applications of biomarkers require different types of validation.

On the basis of a recent expert meeting, the COST-B15 Working Group proposes a new classification for biomarkers, which is based on the mechanism of action and the 'directness' to the clinical response. According to this classification 8 different kinds of biomarkers can be distinguished ranging from type 0 (=genotype/phenotype) to type 7 (=clinical scales). In addition it is also proposed to use mechanism-based pharmacokinetic/ pharmacodynamic (PK/PD) modelling as the scientific basis for the selection, evaluation and validation of biomarkers in early, pre-clinical animal studies. Briefly, mechanism-based PK/PD models utilise principles from receptor theory and dynamic systems analysis. This allows the evaluation of in vivo/in vitro correlations as the basis for the validation of biomarkers. The evaluation and validation of biomarkers for the application in clinical practice is the subject of a new expert meeting in the autumn of 2001.

Have something to say? Post a comment on this article!