PAGE. Abstracts of the Annual Meeting of the Population Approach Group in Europe.
PAGE 24 (2015) Abstr 3563 [www.page-meeting.org/?abstract=3563]
Poster: Drug/Disease modeling - Absorption & PBPK
M. W. Sadiq (1,2), E. I. Nielsen (1), M. O. Karlsson (1), L. E. Friberg (1)
(1) Uppsala University, Sweden; (2) CVMD iMed DMPK, AstraZeneca R&D, 431 83, M÷lndal, Sweden
Objectives: Aim of this study was to develop a WBPBPK model for ciprofloxacin to predict the tissue concentration time profiles in patients with only plasma concentrations data available. WBPBPK model was further combine to a PKPD model to illustrate the time-course of bacterial killing for infections with E. coli strains with different levels of resistance.
Methods: Based on 102 adult ICU patient’s plasma concentration data, a WBPBPK model for ciprofloxacin was developed . NONMEM was used to apply population approach for data analysis. Tissue to plasma distribution coefficients (Kp) for ciprofloxacin in 10 different tissues including lung, muscle, kidney and adipose were taken from clinical studies available in literature. These literature Kp values were used as informative priors while estimating the individual tissue Kp values. Time-course of the bacterial killing for E. coli in different tissues were quantitatively predicted by coupling the final WBPBPK model to a pharmacokinetic-pharmacodynamic (PKPD) model .
Results: The developed WBPBPK model successfully characterized both the typical trends and variability of the available ciprofloxacin data, as demonstrated by visual predictive checks. Stable PK estimates including clearance and tissue Kp values were generated by model, comparable to previously reported literature values. By connecting the predicted PK profile of unbound ciprofloxacin with the PKPD model the rate and extent of take-over of mutant bacteria in different tissues could be predicted. A series of simulation scenarios of different dosing regimens, mixtures of bacterial population with different levels of resistance and immune response were performed to illustrate the concept and the impact of different PK-profiles.
Conclusions: For prediction of time course of bacterial killing in different tissues a novel method of combining the concentration time profile from WBPBPK with PKPD model was successfully implemented.
Acknowledgements: This work was supported by the DDMoRe (www.ddmore.eu) project and Swedish Foundation for Strategic Research.