Modeling, Simulation, and the Critical Path
Daniel Weiner, PhD
Weiner is senior vice president, software products, at Pharsight Corp., Mountain View, Calif.
The benefits of modeling and simulation fall into two categories: Increased certainty in trial designs and faster decisions to kill or advance compounds. The first, more tactical benefit means trials are less likely to fail because they are properly powered and optimally designed to address the hypotheses of interest. In some disease indications, half to two-thirds of all trials either fail or generate findings of limited value. In some cases, inadequacies in the study design fail to show a difference that actually is there. These cases waste money and time, and may expose patients to risks while offering no benefit.
The second, more strategic benefit of modeling and simulation, faster decisions to kill or advance compounds, aligns directly with FDA’s Critical Path goal to “ensure that basic scientific discoveries translate more rapidly into new and better medical treatments.” Results of a proposed trial may be simulated to see whether the information likely to be gained is worthwhile, given the money and time consumed by the trial. There are documented cases where a trial, or even an entire phase II trial, was skipped as a consequence of modeling and simulation. In addition to the saved trial costs, the new therapy may reach thousands of patients years earlier.
Modeling and simulation can demonstrate that no dose will likely outperform existing therapies, often resulting in a decision to kill the compound. The FDA sees this as beneficial, because the kill decision, especially an early kill decision, allows the sponsor to reallocate people and money to other development programs promising more benefit.
Management theory tells us that compelling strategic and tactical benefits are insufficient to create enduring change in the real world of complex organizations like the drug companies and the FDA. It is encouraging, therefore, to see senior FDA and pharmaceutical industry management addressing the challenges of organizational structure, systems, skills, and processes required for real change. Major companies are reorganizing themselves to implement modeling and simulation in separate, high-profile departments. For example, Novartis recently announced the formation of a large modeling and simulation department headed by a former FDA official.
But finding enough people with the skill set needed for modeling and simulation continues to be a challenge because too few people have the necessary math, programming, biology, and pharmacology training. To address this deficiency, Pharsight Corp., Mountain View, Calif., provides free software to thousands of students and teachers in more than 800 pharmacy and medical schools worldwide and invests heavily in the development of high-productivity modeling and simulation tools. But our company is not alone in recognizing the need for expanding the corps of people capable of doing modeling and simulation. For example, Pfizer recently donated $5 million to the University of Buffalo to support expanded training in modeling and simulation.
As increasing emphasis is placed on effective use of modeling and simulation in drug development, we anticipate an increase in university-based training programs to develop scientists with the requisite skills to support these efforts. We expect that FDA will continue to lead the drive to further use modeling and simulation to improve both decision quality and development efficiency.
This article was published in Drug Discovery & Development magazine: Vol. 9, No. 12, December, 2006, p. 54.