But…my projects are special.
Scientists in the field of pharmaceutical research and development face a most daunting challenge. Our understanding of the diseases we work to treat grows more complex and perplexing with each new published study. Take, for example, the gusher of information coming out on Alzheimer’s disease. How can research findings from genetics, neurology, nutrition, protein chemistry, pharmacology, and epidemiology (just to name a few) be tracked, sorted, and used?
Interdisciplinary R&D needs a conceptual synthesis to bridge the disciplines and allow research results from one discipline to be applied to the questions of other disciplines. Disease-drug models can provide this conceptual synthesis. The predictive power of disease-drug models is likely to become the key determinant, and measure, of interdisciplinary research effectiveness.
Currently, creating a model is mostly an ad hoc process that’s dependent on special interests of project team members, previous experience of pharmacometricians, and current understanding of the disease process and drug pharmacology. Models are managed informally, with no established process of maintenance and storage. The result is that crucial information-bearing relationships represented in the model may go unnoticed. This hampers information flow across disciplinary groups and delays recognition of the need for critical studies that would inform the model and improve predictability. Consequently, harvesting the efficiencies that can come from recognizing and exploiting these commonalities are lost to the furious and time-consuming activities of specialty-focused research projects.
A systematic process for envisioning, developing, and maintaining disease-drug models has to be developed before these models can take their place at the center of drug development. This process will become more important and difficult as the models become more mechanistic. The transition to mechanistic models is necessary if the models are going to provide much needed insight into the determinants of drug efficacy and safety—a prerequisite for improving the chances of success in clinical trials. Only then will disease-drug models become a conceptual framework that can be used to inform experimentation strategies and to render research results.
Developing a process for establishing and maintaining disease-drug models will likely take a substantial investment in time, money, patience, and persistence. As a preliminary effort, I have been spending much of my time recently considering the nature of communications among pharmacometricians at the technical level and between pharmacometricians and other members of a development team at the strategic and tactical level. I look for common patterns that emerge in these communications because I believe the patterns hold the key to more efficient and relevant communications across stakeholders and to a strategy on how discipline-specific research results can be more properly incorporated into disease-drug models. Ultimately, the goal is to improve productivity of pharmaceutical research and development programs, a goal that has proved more elusive than understanding the implications of variability in the universal constant. (links to a dead page) LOL.