Premium
Comparison of three physiologically‐based pharmacokinetic models for the prediction of contrast agent distribution measured by dynamic MR imaging
Author(s) -
Barboriak Daniel P.,
MacFall James R.,
Viglianti Benjamin L.,
Dewhirst DVM Mark W.
Publication year - 2008
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.21344
Subject(s) - physiologically based pharmacokinetic modelling , contrast (vision) , pharmacokinetics , dynamic contrast enhanced mri , nuclear medicine , dynamic contrast , glioblastoma , distribution (mathematics) , medicine , magnetic resonance imaging , biomedical engineering , radiology , computer science , mathematics , artificial intelligence , mathematical analysis , cancer research
Purpose To compare the performance of three physiologically‐based pharmacokinetic (PBPK) models for predicting gadolinium contrast agent concentration‐time curves (Gd‐CTCs) obtained in superior sagittal sinus (SSS), cerebral cortex, and psoas muscle. Materials and Methods Three published whole‐body PBPK models were modified to predict Gd‐CTCs in normal‐appearing tissue. The models differed in the number of organs modeled and total number of compartments, and were designated as the “well‐mixed,” “delay,” and “dispersion” models. The suitability of the three models to predict Gd‐CTC was studied using data from dynamic contrast‐enhanced MR perfusion imaging obtained at 1.5T from 10 patients with glioblastoma multiforme and at 3.0T from five patients with liver metastases. Results The dispersion model produced better fits than the delay model in the SSS ( P < 0.0001) and cerebral cortex ( P < 0.0001), and better fits than the well‐mixed model in psoas muscle ( P < 0.005). No model produced better fits than the dispersion model at any of the three locations. Conclusion In this evaluation, the dispersion model was most robust for prediction of Gd‐CTCs derived from dynamic contrast‐enhanced (DCE)‐MRI. This represents a preliminary step in the development of a PBPK model useful for predicting Gd‐CTCs at a time resolution appropriate for dynamic MRI applications. J. Magn. Reson. Imaging 2008;27:1388–1398. © 2008 Wiley‐Liss, Inc.