The Impact of Arterial Input Function Determination Variations on Prostate Dynamic Contrast-Enhanced Magnetic Resonance Imaging Pharmacokinetic Modeling: A Multicenter Data Analysis Challenge

Pharmacokinetic analysis of dynamic contrast-enhanced (DCE) MRI data allows estimation of quantitative imaging biomarkers such as K(trans) (rate constant for plasma/interstitium contrast reagent (CR) transfer) and v(e) (extravascular and extracellular volume fraction). However, the use of quantitative DCE-MRI in clinical practice is limited with uncertainty in arterial input function (AIF) determination being one of the primary reasons. In this multicenter study to assess the effects of AIF variations on pharmacokinetic parameter estimation, DCE-MRI data acquired at one center from 11 prostate cancer patients were shared among nine centers. Individual AIF from each data set was determined by each center and submitted to the managing center. These AIFs, along with a literature population averaged AIF, and their reference-tissue-adjusted variants were used by the managing center to perform pharmacokinetic data analysis using the Tofts model (TM). All other variables, including tumor region of interest (ROI) definition and pre-contrast T(1), were kept constant to evaluate parameter variations caused solely by AIF discrepancies. Considerable parameter variations were observed with the within-subject coefficient of variation (wCV) of K(trans) obtained with unadjusted AIFs being as high as 0.74. AIF-caused variations were larger in K(trans) than v(e) and both were reduced when reference-tissue-adjusted AIFs were used. These variations were largely systematic, resulting in nearly unchanged parametric map patterns. The intravasation rate constant, k(ep) (= K(trans)/v(e)), was less sensitive to AIF variation than K(trans) (wCV for unadjusted AIFs: 0.45 vs. 0.74), suggesting that it might be a more robust imaging biomarker of prostate microvasculature than K(trans).