Additional Value of Dynamic Contrast-enhanced Sequences in Multiparametric Prostate Magnetic Resonance Imaging: Data from the PROMIS Study

Abstract

Background: Multiparametric magnetic resonance imaging (MP-MRI) is established in the diagnosis of prostate cancer, but the need for enhanced sequences has recently been questioned.

Objective: To assess whether dynamic contrast-enhanced imaging (DCE) improves accuracy over T2 and diffusion sequences.

Design, setting, and participants: PROMIS was a multicentre, multireader trial, with, in this part, 497 biopsy-naïve men undergoing standardised 1.5T MP-MRI using T2, diffusion, and DCE, followed by a detailed transperineal prostate mapping (TPM) biopsy at 5 mm intervals. Likert scores of 1-5 for the presence of a significant tumour were assigned in strict sequence, for (1) T2 + diffusion and then (2) T2 + diffusion + dynamic contrast-enhanced images.

Outcome measurements and statistical analysis: For the primary analysis, the primary PROMIS outcome measure (Gleason score ≥4 + 3 or ≥6 mm maximum cancer length) on TPM was used, and an MRI score of ≥3 was considered positive.

Results and limitations: Sensitivity without and with DCE was 94% and 95%, specificity 37% and 38%, positive predictive value 51% and 51%, and negative predictive value 90% and 91%, respectively (p > 0.05 in each case). The number of patients avoiding biopsy (scoring 1-2) was similar (123/497 vs 121/497, p = 0.8). The number of equivocal scores (3/5) was slightly higher without DCE (32% vs 28% p = 0.031). The proportion of MRI equivocal (3/5) and positive (4-5) cases showing significant tumours were similar (23% and 71% vs 20% and 69%). No cases of dominant Gleason 4 or higher were missed with DCE, compared with a single case with T2 + diffusion-weighted imaging. No attempt was made to correlate lesion location on MRI and histology, which may be considered a limitation. Radiologists were aware of the patient's prostate-specific antigen.

Conclusions: Contrast adds little when MP-MRI is used to exclude significant prostate cancer.

Patient summary: An intravenous injection of contrast may not be necessary when magnetic resonance imaging is used as a test to rule out significant tumours in the prostate.

Keywords: Biparametric magnetic resonance imaging; Multiparametric magnetic resonance imaging; Prostate cancer; Prostate magnetic resonance imaging.