This week, Entelos announced that the U.S. Food and Drug Administration will be using its Cardiovascular PhysioLab simulation platform to help agency scientists assess the cardiovascular risks of drugs with the help of simulated “virtual patients.”
The agreement is the second collaboration that Entelos has signed with the FDA for its PhysioLab platform. In 2007, it signed a two-year Cooperative Research and Development Agreement with the agency to build a version of the system to predict drug-induced liver injury [BioInform 08-10-07].
In a statement submitted to BioInform via e-mail, an FDA spokesperson said that the cardiovascular project “is an extension of that interest,” though the two projects are not directly related.
The spokesperson noted that the project is in line with the agency’s Critical Path initiative, which it launched in 2004 in an effort to improve drug safety and efficacy and speed the regulatory process. FDA included modeling and simulation among several high-priority areas that it planned to explore under the program.
In the collaboration with Entelos, FDA will apply Cardiovascular PhysioLab “along with simulation to see if we can anticipate relatively rare and serious adverse events prior to either large scale clinical trials or market availability,” the spokesperson said.
In a follow-on step, FDA plans to “compare the simulated outcome for efficacy and safety in specific patient types to those the sponsor reports when the trials are completed.”
Based on the outcome of these studies, “we may choose to expand our evaluation of systems biology linked with simulation to potentially anticipate efficacy and safety in specific subpopulations,” FDA said, adding that ““This is the only evaluation of this nature in [FDA’s Center for Drug Evaluation and Research] today.”
While various simulation methods have played a role in the drug approval process and post-marketing evaluation at the agency, “simulation from systems biology models have not been used previously at CDER to our knowledge,” FDA said.
FDA will study three drugs under the project. “Two of the three are currently being studied for a cardiovascular condition,” FDA said. Additional details of the drugs were not provided.
Fujisaki said that the partnership is not structured as a formal subscription-based model “but rather a close collaboration.”
The contract for Cardiovascular PhysioLab is not connected to the drug-induced liver injury project, which Entelos is conducting with a different FDA research group, Fujisaki said.
The project also differs slightly in its scope. While Entelos is developing a new version of PhysioLab for the DILI project, the company’s Cardiovascular PhysioLab platform is already in use at some of the firm’s pharmaceutical customers.
The platform is a large-scale computer simulation of cholesterol regulation, atherogenesis, and cardiovascular risk. The company said that its customers have used it to predict the effects of drugs in patients and patient populations, evaluate novel drug targets, test combination therapies, identify and interpret biomarker patterns, and predict a drug’s long-term biological effects.
A Tool for Hard Times
Entelos is not the only firm developing tools to simulate patient response to drugs. Companies such as SimCyp and Optimata, and several academic groups, are also taking a similar approach.
Fujisaki said that Entelos distinguishes itself from these other groups by “the variety of data used to build all our disease platforms: from pathway/molecular, to medical histories, to environmental [factors] such as diet, exercise, other exposures, to large-scale clinical trials and patient data.”
“This is the only evaluation of this nature in [FDA’s Center for Drug Evaluation and Research] today.”
The company’s researchers do not “simply ‘dump’ these data into a database; but rather, use the data to test, deduce, and validate the biological mechanisms regulating health and disease processes,” she said.
Another difference lies in the mechanistic modeling engine behind the simulation, which is not built using statistical models, but is rather based on “engineering principles more closely related to those used to build bridges, airplanes, or complex communication systems,” she said.
The platform also has the ability to represent virtual patient populations that span the wide range of real patients seen in the clinic, she said. “We can literally simulate and test thousands of virtual patients, drugs, drug combinations, and conditions to customize applications to customers' specific questions about optimal dosing, best responders, or best combination therapies to pursue.”
Fujisaki said she believes simulation holds particular value in more challenging economic times for the drug discovery industry, where pharmas are “seeing it as a smarter way to use their precious R&D dollars in tough times. Why not simulate before spending hundreds of millions of dollars in clinical trials, and help de-risk your riskiest projects?”
CDER scientists use other tools for mechanistic and empirical drug-disease modeling in the clinical trial area to assist scientists in planning and decisions, the FDA spokesperson said.
For example, in a paper published in the Journal of Clinical Pharmacology earlier this year, CDER scientists gave an overview of their experiences with the use of quantitative knowledge to guide drug development.
The FDA spokesperson said that the agency has also used empiric and mechanistic modeling and clinical trial simulation to compare Phase 2b and Phase 3 trial designs and patient entry criteria prior to trial initiation.
In addition, “modeling pharmacokinetics and dynamics across clinical trials have been used to evaluate dose-response and patient subtypes during the [new drug approval] review process. This information has been added to labeling at times.”
Separately, in post-approval studies, FDA said, “industry scientists have utilized these approaches and systems biology simulation modeling to help plan and make decisions,” though these types of simulations have not been used to date within CDER.
"Although the modern controlled clinical trial is still the international gold standard for evaluating safety and efficacy of new therapies, rare and serious adverse events may only appear after a drug has been administered to a large heterogeneous population, long after it has been approved," Janet Woodcock, director of CDER, said in a statement.
“Having information that may be predictive of likely adverse events or that can help to explain the biological mechanisms leading to adverse events in certain patient types could be extremely valuable,” she added. “This project will thus test the predictive value of using a dynamic, mechanistic computer model of cardiovascular disease processes and a large virtual patient population for detecting rare adverse events.”