Added value of diffusion-weighted images and dynamic contrast enhancement in multiparametric magnetic resonance imaging for the detection of clinically significant prostate cancer in the PICTURE trial

Objective To determine the additional diagnostic value of diffusion‐weighted imaging (DWI) and dynamic contrast‐enhanced (DCE) imaging in men requiring a repeat biopsy within the PICTURE study.

Patients and Methods PICTURE was a paired‐cohort confirmatory study in which 249 men who required further risk stratification following a previous non‐MRI guided TRUS biopsy underwent a 3‐Tesla mpMRI consisting of T2W, DWI and DCE followed by transperineal template prostate mapping (TPM) biopsy. Each mpMRI was reported using a LIKERT score in a sequential blinded manner to generate scores for T2W, T2W+DWI and T2W+DWI+DCE. Area under the receiver operating characteristic (AUROC) fanalysis was performed to compare the diagnostic accuracy of each combination. The threshold for a positive mpMRI was set as a LIKERT score >/=3. Clinically significant prostate cancer was analysed across a range of definitions including UCL/Ahmed Definition 1 (primary definition), UCL/Ahmed Definition 2, any Gleason >/=3+4 and any Gleason >/=4+3.

Results Of 249, sequential MRI reporting was available for 246. There was a higher rate of equivocal lesions (44.6%) using T2W alone compared to the addition of DWI (23.9%) and DCE (19.8%). Using the primary definition of clinically significant disease, there was no significant difference in the overall accuracy between T2W at AUROC 0.74 (95% CI 0.68‐0.80), T2W+DWI at 0.76 (95% CI 0.71‐0.82) and T2W+DWI+DCE at 0.77 (95% CI 0.71‐0.82) (p=0.55). The AUROCs remained comparable using other definitions of clinically significant disease including UCL/Ahmed 2 (p=0.79), Gleason >/=3+4 (p=0.53) and Gleason >/=4+3 (p=0.53).

Conclusions Using a 3T MRI, a high level of diagnostic accuracy can be achieved using T2W as a single parameter in men with a prior biopsy. However, such a strategy can lead to a higher rate of equivocal lesions.