Big world challenges for pharmacometrics


From the outset, the aim of pharmacometrics has been to improve patient therapy through model based methods. The focus of these efforts have shifted over time. Initially, in the 70’s and 80’s, targeting the data collected during routine use of drugs in the target patient population to provide better dosing recommendations. In the 90’ and 00’s the focus shifted towards improved clinical drug development, whereas in the present decade the additonal focus has been towards translation between different stages throughout drug development as well as allowing improved societal use of drugs. Thus, today the overall aim remains the same, but the scope of activities has widened considerably. Specific aims and challenges for activities of different maturity need to be identified and addressed. The performance in core activities needs to improve while also exploring and implementing the utilization of pharmacometric methods and models in new areas of application where the benefit/cost is high or promising.


For core development activities, assure efficient utilization of resources and appropriate impact on decision making through increased communication and integration with other disciplines. The sibling disciplines statistics and systems pharmacology that together with pharmacometrics constitute the quantitative aspects throughout drug development require particular attention. Methodologies that bridge gaps need to be developed. For example, to better bridge towards systems pharmacology, methods to utilize both prior information and new data are required and to bridge towards statistics in late stage clinical development, more methods for pre-specified use of pharmacometric models are desired. For core clinical use activities, there is an opportunity to develop the utilization of models in areas beyond therapeutic drug monitoring. The increased access to, and possibility to develop, models for clinical endpoints can provide not presently available decision support in management of patient care. Such models may become particularly important with increasing access to of patient specific information through new technologies. Models may act as notification and decision support when data amounts prohibit systematic in cerebro synthesis. Increased access to or generation of large databases from different populations: the general population,patients with specific diseases and/or treatments comes from registries, health care providers or through multi-institutional collaborations. Such data can provide information on (patho) physiological and functional characteristics from more patients than would ever be possible to study in clinical trials. However, efficient utilization of these databases will involve collaboration with new disciplines, e.g. epidemiologists and health economists.


In its core application area of clinical drug development, pharmacometrics is becoming a mature discipline. However, its implementation still displays a large heterogeneity between and within organisations. Community consensus, including regulatory acceptance, is only slowly being generated. Data sharing, common in many other disciplines has made little or no progress to the detriment of better based models, in particular in rare, orphan and slowly progressing diseases. Good practises documents, model sharing mechanisms, standards development and other community led efforts have only recently started to emerge. Short comings of existing estimation methodologies and software are impediments in developing otherwise desirable models, especially mechanistic models and models for large databases. Educated and trained pharmacometricians remain a scarce resource. PhD programs and training on the job remains the main influx of new pharmacometricians, while undergraduate programs intentionally set up to suit the need of the disciplines are rare.


To consolidate the use of pharmacometrics in core areas and expand its use to new opportunities there is a need for development in a large number of areas: education, methodology, software, interdisciplinary collaborations, intradisciplinary infrastructure to name a few.