New Mathematical Model Helps Predicts Optimal Use of Antibiotics
Researchers at the Yale School of Public Health have developed a new mathematical model to help predict the optimal dosing of antibiotics.
Although antibiotics were first introduced more than 70 years ago, substantial uncertainty remains about how the drugs should be used by patients to ensure recovery, while minimizing toxic side effects and the risk of developing antibiotic resistance.
“To date, antibiotic treatment recommendations have been arrived at by trial-and-error or with complex models of drug distribution and action, and it is not possible in most cases to know whether the recommended dose, frequency and length of most treatment regimens are appropriate,” said Ted Cohen, M.D., M.P.H, D.Ph., associate professor at Yale School of Public Health and senior author of the study in the journal Science Translational Medicine.
While current mathematical models predict the effect of antibiotics on bacteria in specific settings, these models fail to reliably predict treatment efficacy across a range of biologically relevant scenarios.
In their study, Pia Abel zur Wiesch, postdoctoral fellow at Yale School of Public Health and lead author, Cohen and a team of international colleagues describe a model that uses information about how antibiotics bind to bacterial target molecules to predict how these drugs will affect individual bacterial cells and populations of bacteria. The model suggests that the apparent complexity of antibiotic action may be explained by the simple chemical kinetics of antibiotics binding to bacterial targets. Accordingly, the new model offers an alternative explanation for the mechanisms that are responsible for complex antibiotic actions.
Abel zur Wiesch said that the complexity of antibiotic action has traditionally been ascribed to various biological phenomena. For example, it is not well understood why very long treatment courses are needed to eradicate some bacterial infections. One common explanation has been that some bacteria are “asleep” and not susceptible to antibiotics, but can then “wake up” after therapy is discontinued and cause patients to suffer a relapse.
The authors hope the model may eventually help inform the design of more effective antibiotic dosing regimens based on chemical kinetic properties of antibiotics alone. Furthermore, the model may be useful for speeding the development of new antibiotics by identifying favorable chemical kinetic characteristics of new drug compounds.
In the future, the authors plan to extend this model to understand how chemical kinetics of antibiotics influence the risk of drug resistance, particularly for diseases of global health importance such as tuberculosis. “While we are excited about the promise of this modeling framework for explaining antibiotic effects, additional work is needed to evaluate whether this simple modeling framework can be used to inform the design of regimens for diseases requiring complex multidrug treatments,” said Abel zur Wiesch.