Reference region-based pharmacokinetic modeling in quantitative dynamic contract-enhanced MRI allows robust treatment monitoring in a rat liver tumor model despite cardiovascular changes.

In this work, two pharmacokinetic modeling techniques, population arterial input function model, and reference region model, were applied to dynamic contract-enhanced MRI data, to test the influence of a change in heart rate on modeling parameters. A rat population arterial input function was generated by dynamic contrast-enhanced computed tomography measurements using the MR contrast agent gadolinium diethylenetriamine penta-acetic acid. Then, dynamic contract-enhanced MRI was used for treatment monitoring in two groups of hepatocellular carcinoma bearing rats. Whereas group 1 had the same heart rate as animals analyzed for the population arterial input function (263 ± 20 bpm), group 2 had a higher heart rate (369 ± 11 bpm) due to a different anesthesia protocol. The pharmacokinetic modeling parameters volume transfer constant K(trans) and relative extravascular extracellular space v(e) were calculated with both models and statistically compared. For group 1, good correlation and agreement was found between the models showing no difference in K(trans) and v(e) (ΔK(trans): 4 ± 19% and Δv(e): 4 ± 12%, P = 0.2). In contrast, for group 2, a bias in parameter values for the population arterial input function model was detected (ΔK(trans): -45 ± 7% and Δv(e): -31 ± 7%, P ≤ 0.001). The presented work underlines the value of the reference region model in longitudinal treatment monitoring and provides a straightforward approach for the generation of a rat population arterial input function.