An International Workshop on the Development of Good Modelling Practice for PBPK Models
Mediterranean Agronomic Institute of Chania, Crete, Greece
Thursday 26 April 2007 - Saturday 28 April 2007

Summary
The use of Physiologically Based Pharmacokinetic Modelling techniques as a tool in the risk assessment process is well established. The power of the technique coupled with the implications of REACH suggest that there will be a rapid increase in usage of PBPK over the next few years. To ensure the transparency, validity and applicability of PBPK modelling, several organisations (EPA, Health Canada) have commissioned reports on the principles of good modelling practice. The information contained in these documents agree on many points but there are a number of areas where expert opinion is divided. The purpose of this workshop is to allow regulators and modellers to debate these disparate positions and progress towards a standard practice for the development, evaluation and interpretation of these models. From this, the workshop aims to establish a 'best practice' that improves transparency and facilitates the understanding and sharing of national and international risk assessment reports.

Introduction
PBPK modelling is recognised as having great potential in the development of a more scientific and quantitative risk assessment of chemicals. The International Programme on Chemical Safety (IPCS) steering group of the World Health Organisation (WHO) identified PBPK modelling as an important component of chemical risk assessment which merits international harmonisation. The ability to review a PBPK model according to accepted criteria would greatly facilitate widespread acceptance, in particular, amongst regulators. A six-step process for the evaluation of PBPK models has been proposed (Clark, et al. (2004) Risk Anal 24, 1697-717) which, lists the obvious and important characteristics one should check. However, there are some ingrained practices that some PBPK modellers consider contentious. Therefore, the six-step process requires further refinement and may not, in any case, be appropriate for regulators and risk assessors who would like to see increased use of PBPK models in the risk assessment process. Identification of key structural components of a model that can be checked by a risk assessor or regulator would be of value. Discussion and agreement amongst PBPK practitioners is paramount for the development of good modelling practice (GMP) guidelines. However, these guidelines should also be acceptable to regulators and risk assessors. Development of GMP guidelines is best achieved through a cross-disciplinary exchange of experience and ideas among laboratory scientists, PBPK modellers, regulators and risk assessors.

Goals
To achieve its goal, this workshop will:

• Review and assess the technical aspects of good modelling pracice, including the outputs of the sister workshop on Uncertainty and Variability held in the US - http://www.epa.gov/comptox/uvpkm/.
• Define methods and tools that determine whether a model is fit for purpose.
• Establish the model predictions that best address the needs of the regulator and how this information should be communicated.
• Make availiable the findings of the workshop through a workshop proceeding and a series of publications in the peer-reviewed literature.

Reference:
Clark, L. H., Setzer, R. W., and Barton, H. A. (2004). Framework for evaluation of physiologically-based pharmacokinetic models for use in safety or risk assessment. Risk Anal 24, 1697-717.

  back to top