Simulating Science
by Sarah Stone Wunder
October 2008
Through clinical trial simulation, researchers can generate a fuller understanding of their drug and its potential effects, leading to better designed—and possibly less expensive—trials.
In an effort to refine clinical trials, many drug developers have turned to clinical trial simulations to augment their current programs. Although simulations can never replace patient-driven clinical trials, they can help developers target, refine and focus their trial process.
At the Center for Drug Development Science at the University of California, San Francisco (UCSF), scientists work with pharmaceutical companies to help them create mechanism-based drug development models. Dr. Howard Lee, director of the center and an associate adjunct professor at UCSF, says many of these companies come to the center with limited knowledge of clinical trial simulations.
“Most of them do not know what they should do,” Lee says. “In more than 99 percent of the cases, the first step we took was to teach the sponsor what options they have for mechanism-based drug development through modeling and simulations.”
Essentially, clinical trial simulation involves setting up computer models to mimic how a trial would progress. The assumptions for the simulation’s models are based on knowledge gained from previous studies; knowledge about the drug’s mechanism of action; knowledge gained from public databases, as well as other sources. All of this information is gathered to create a computer model, which simulates the action of the drug and the outcome that would likely result. In the end, researchers discover the potential ways in which a drug can act and the possible outcomes and side effects that might occur.
“In real life, I might be constrained to give a trial drug to 500 or 1,000 patients,” says Badhri Srinivasan, vice president and head of the enterprise transformation unit at Quintiles Transnational, in Durham, N.C., USA. “In the computer world, I can run 10,000, 100,000, 1 million, 10 million simulations. So I’m actually simulating it as if there are 10 million interventions of the drug, 10 million patients out there taking the drug.”
Because researchers can use such a large database, they can better refine and target their subsequent patient-driven clinical trials. According to Lee, clinical trial simulations help researchers answer essential questions, such as: What is the dose-response relationship? Are there differences between populations taking the drug? “Clinical trial simulation is a way to help drug developers understand their drug better,” he says.
A Slow Growth
Although drug developers have been slow to use clinical trial simulations, Srinivasan says he expects that to change. “Companies are exploring possibilities of clinical trials, but I don’t think there’s an active commitment to doing this,” he says. “But it is something that is now getting on everyone’s radar screen, where everybody is saying, ‘Should I be looking at simulation instead?’ So the awareness is increasing.”
Srinivasan says uptake has been slow for three reasons: The exercise can be fairly complex; data used for modeling isn’t available for certain therapeutic areas and indications; and regulatory acceptance is still an uncertainty.
“From a regulatory perspective, can [companies] then go with clinical trial simulation data in lieu of, or as a proxy for, Phase II data and then go into Phase III directly?” he says. “Those questions are yet to be answered because from a regulatory perspective, it is still a bit of an unknown.”
As it stands currently, it is highly unlikely that clinical trial simulation will result in elimination of Phase II studies as there is sufficient complexity and lack of knowledge in drugs, mechanisms of actions and metabolic pathways, Srinivasan says. “While we may use clinical trial simulation to augment our knowledge and more quantitatively understand risk and benefit for a compound, under the current clinical trial paradigm, regulators will be highly reluctant to consider simulation as a proxy for Phase II studies.”
Timing Simulations
Because simulations help researchers learn more about their therapies, the exercise can be used throughout the drug development process. “It can be done at any phase,” Srinivasan says. “Clinical trial simulation is being done at the Phase I stage, where you look at things like T-Max and C-Max, and you can do simulations for that. It also is a great tool to use at the Phase II stage in order to understand what the outcomes might be. So it will allow you to examine potential doses, populations, dosing regimens, etc., to construct a Phase II trial in a more informed way, and even understand it completely before going on to the confirmatory phase.”
Lee says simulations also are useful during Phase III. “If a Phase III clinical trial study failed, many companies have the mentality, ‘Let’s repeat it until we show that the drug is effective,’” Lee says. “But modeling and simulations can be very effective in sorting out those failures, questions and issues from the learning perspective.”
Send us Your Comments
Web Exclusives
- Statistical Advantage
-
Gary Shorter, director of biostatistics for Quintiles, talks to Envisage editor Adam Istas about the growing importance of incorporating biostatistical analysis into all stages of drug development.
- Measure for Measure
-
A proposed Health Care Comparative Effectiveness Research Institute in the United States may force the biopharmaceutical industry to emphasize patient outcomes and quality of life issues when developing a new medicinal product.
Is the same way to construct realistically “Simulating Science” programmes for clinical problems? I tried to build such projects years ago on predicting of the future treatment for peptic ulcer disease (before computer era)?
— Machowski Z.A. · Oct 19, 05:02 AM · #