Improved correlation to quantitative DCE-MRI pharmacokinetic parameters using a modified initial area under the uptake curve (mIAUC) approach.

PURPOSE: To propose a modified initial area under the uptake curve (mIAUC) dynamic contrast-enhanced (DCE)-MRI approach to achieve better distinction of underlying physiology.
MATERIALS AND METHODS: The mIAUC is formulated on common characteristics of tissue contrast uptake curves observed over a wide range of physiological conditions. The new metrics, IAUC(Ktrans) and IAUC(ve), are related to the transfer constant (K(trans)) and interstitial volume (v(e)), respectively. Tissue uptake curves were simulated over a range of physiological values and analyzed using the proposed mIAUC, conventional IAUC, and Tofts' pharmacokinetic model.
RESULTS: IAUC(Ktrans) and IAUC(ve) are highly correlated to the true K(trans) and v(e) (rho = 0.97 and 0.95, respectively), approaching the performance of Tofts' model based on a 1.5-s sampled arterial input function (AIF) (rho = 0.98 and 0.98) under noise conditions typical in DCE-MRI experiments. Lower correlations were obtained with conventional IAUC(60) and IAUC(120) (rho = 0.82 and 0.61) and Tofts' parameters fitted using a biexponential AIF (rho = 0.81 and 0.90).
CONCLUSION: The proposed mIAUC approach retains advantages associated with nonmodel based methods (robust to noise and model fit failure, obviates need for an AIF) while providing better distinction of underlying physiological parameters. It can be a valuable alternative to pharmacokinetic modelling in the analysis of DCE-MRI data. (c) 2009 Wiley-Liss, Inc.