The concept of personalized medicine has long been promoted to bring more benefits to patients. One aspect of personalized medicine is the right dose. Because of inherent differences between patients the pharmacokinetics (PK) and pharmacodynamics can substantially vary between patients. As a consequence, in some cases, the dose should also be individualized to ensure efficacy and safety for each patient. Methodologically and practically it is a challenging modelling task. Using the MonolixSuite the process of deriving a personalized dose has become much easier thanks to integrated software. On the example of tobramycin, an antibioticum, we demonstrate how to go from a PK data set and the safety and efficacy requirements on exposure to an individualized dose. The figure above shows the expected PK for a single patient depending on the prior information that was used. Using the population PK parameters we get the widest prediction interval; this is because we see the patient like the entire population. Including the patients covariates we can already narrow down the prediction interval. Evidently, when using individual PK observations as priors it allows us to get the most accurate prediction required to have a robust individualized dose prediction. For the individual parameter estimation the first 8 hours PK observations from this patient were used, and for the simulation the standard error of the individual parameter estimates was applied. The fully documented step-by-step modelling procedure for the tobramycin case study can be found on our website.

We are looking forward to see more advanced applications of the MonolixSuite to bring the benefits to patients. Let us know if you have new ideas and would like to have our support to make sure to get your model running.