Error quantification in pharmacokinetic parameters derived from DCE-MRI data using a novel anthropomorphic dynamic prostate phantom

Abstract

Pharmacokinetic (PK) modelling of dynamic contrast enhanced (DCE) MRI data has shown considerable promise in the detection of many cancers, including prostate. However, the clinical implementation of DCE-MRI acquisition protocols suitable for PK modelling has been greatly hindered by the lack of consistency with published PK parameter values, such as the volume-transfer constant ( K trans ), with widely varying acquisition protocols and PK modelling approaches adding to the confusion. A novel anthropomorphic dynamic prostate phantom has been developed to address this shortcoming. The device is capable of simultaneously producing two distinct contrast time-intensity curves (CTCs), representative of those observed in DCE-MRI data from healthy and tumorous prostate tissue, in an environment which closely mimics the male pelvic region. CTCs were measured using a custom-built optical scanner, and ‘ground truth’ PK parameter values derived from the data. DCE-MRI data were acquired using a 3T scanner (Achieva, Philips, Netherlands) and a 16-channel phased array detector coil (3D-SPGR, TR/TE = 3.6/1.75 ms, flip = 10°, voxel size = 1.1 × 1.1 × 4 mm 3 , FOV = 224 × 224 × 72 mm 3 , slices = 18). The parallel imaging factor and number of signal averages were varied to give temporal resolutions from 2.3 s to 20.3 s. K trans values derived from MR data from the phantom were compared with the ground truth values and were found to differ by −8.1% to −44.6%, with the lowest variance from ground truth values achieved at a temporal resolution of 6.8 s. The phantom provides a model system on which the quantitative validation of new prostate DCE-MRI protocols can be performed, and could contribute to the standardisation of clinical prostate DCE-MRI acquisition protocols.