A Physiologically Based Pharmacokinetic Model for the Prediction of Plasma and Bone Tissue Exposure after Prophylactic Administration of Ampicillin/Sulbactam in Patients with Osteonecrosis of the Jaw.

Abstract

Background and objective: The combination of ampicillin (AMP) together with sulbactam (SBC) is a widely used choice for infection prophylaxis in the context of numerous surgical procedures, especially those performed in the field of maxillofacial surgery. Since the pharmacokinetic behavior of these two substances in body tissues is not known in detail owing to sparse tissue data in the literature, the aim of this work was to develop a physiologically based pharmacokinetic (PBPK) model that can predict the concentration versus time courses of AMP and SBC after intravenous administration in plasma, especially bone tissue. Furthermore, the effectiveness of an established prophylaxis regimen based on the developed PBPK model was to be evaluated.

Methods: A PBPK model for middle-aged and elderly populations was created using PK-Sim^® software. A total of nine human clinical studies which included data from plasma, lung, skin, and bone tissue were utilized to verify the model. In addition to the physicochemical properties and ADME (Absorption, Distribution, Metabolism, and Excretion) characteristics of AMP and SBC, the measured drug concentrations from the clinical studies were used for development and validation. The performance of the model was evaluated on the basis of established fold error acceptance criteria for selected pharmacokinetic parameters. Here, the model predictions were compared with the observed values.

Results: The final PBPK model for AMP and SBC could well describe the measured mean concentrations in plasma and in the different body tissues, as these fell within the predicted 5th-95th percentile range for the most part. This applies to 97% of the AMP and 88% of the SBC measurements. Exactly 81% of the fold error values of the pharmacokinetic parameters are within the twofold acceptance criterion. Overall, the average fold errors for the evaluated pharmacokinetic parameters were within the range of 1.01-1.43.

Conclusions: In this work, we present the first PBPK model that simultaneously predicts AMP and SBC in plasma and various tissues. In addition to observed plasma data, the model was also developed and verified with experimentally measured data from the above-mentioned tissues. This allowed a significant limitation of previous PBPK models to be overcome. The effectiveness of established prophylaxis regimes is demonstrated through our model, whereby it must be assumed, owing to measured data for bone tissue, that some individuals do not reach the target values for adequate prophylaxis.