Logo for Stony Brook University
News and Media Relations header
  • Search Press Releases
University News

 

New Mathematical Model to Predict Pharmacodynamic Activity May Improve Drug Discovery
Findings by Stony Brook Professor and colleagues published in Nature Chemical Biology
placeholder

Professor Peter Tonge, PhD, left, along with Eleanor K. H. Allen, PhD, Fereidoon Daryaee, a graduate student, and other colleagues (not pictured) have created a mathematical model as a foundation to improve drug discovery. 

Stony Brook, NY, April 21, 2015 – A new mathematical model that uses drug-target kinetics to predict how drugs work in vivo may provide a foundation to improve drug discovery, which is frequently hampered by the inability to predict effective doses of drugs. The discovery by Peter Tonge, a Professor of Chemistry and Radiology, and Director of Infectious Disease Research at the Institute for Chemical Biology and Drug Discovery (ICB & DD) at Stony Brook University, along with collaborators at Stony Brook University and AstraZeneca, is published advanced online in Nature Chemical Biology .

Drugs normally work by binding to proteins in the human body. In the process of discovering new drugs, investigators try to ensure that a drug binds much more tightly to the desired protein target, rather than binding to other off-target proteins, which cause drug toxicity and/or side effects in patient. This difference in binding is thought to control the therapeutic index or window of a drug. The “window” is the difference in concentration of a drug that causes the desired therapeutic effect, compared to the concentration that causes toxicity. An insufficient therapeutic window is often the principal reason for drug failure in clinical trials.

In the paper, titled “Translating slow-binding inhibition kinetics into cellular and in vivo effects,” Dr. Tonge and colleagues describe a new approach that will dramatically improve the therapeutic window of drugs.

“Current methods utilized in drug discovery are unable to account for the fact that in vivo drug levels fluctuate, which is a major factor in the failure to accurately predict drug action in vivo,” explained Dr. Tonge. “We have created a mathematical model that accurately predicts drug activity based not only on how strongly the drug binds to the protein target, but also on how long the drug remains bound to the target.”

The researchers validated their approach by successfully predicting the in vivo activity of an antibacterial agent that targets the human pathogen Pseudomonas aeruginosa.

Dr. Tonge says the approach, which is the first mechanistic pharmacokinetic-pharmacodynamic model that integrates precise measurements of enzyme inhibition (or drug-target kinetics), is applicable not only to other drug targets in the area of infectious diseases but also to targets in all other disease areas.

“We believe the model and entire approach will significantly improve the selection and development of drug candidates,” he added, “and also improve our ability to develop chemical tools for dissecting and understanding cellular biology.” As a next step to realizing this vision, Professor Tonge has recently founded a company on this concept, Chronus Pharmaceuticals, Inc., with his colleague and co-lead author Dr. Stewart Fisher.

Co-authors of the study include: Eleanor K.H. Allen and Fereidoon Daryaee of the ICB & DD at Stony Brook University; Stephen G. Walker, of the Department of Oral Biology and Pathology, Stony Brook University School of Dental Medicine; Grant K. Walkup and Stewart L. Fisher Visiting Professors at Stony Brook University, and the Infection Innovation Medicines Unit, AstraZeneca Research and Development; and Zhiping You, Philip L. Ross, Michael R. Hale, John O’Donnell, David E. Ehmann, Virna J.A. Schuck, Ed T. Buurman, Allison L. Choy, Laurel Hajec, Kerry Murphy-Benenato, Valerie Marone, Sara A. Patey, Lena A. Grosser, and Michele Johnstone of AstraZeneca.

The research is supported in part by the National Institutes of Health.