Shahid Beheshti Medical University
Our aim was to detect the performance characteristics of multiparametric magnetic resonance imaging (mp-MRI) in patients with clinical suspicion, or previous diagnosis, of prostate cancer. The study sought to prospectively evaluate which technique among T2-weighted images, dynamic contrast-enhanced (DCE), magnetic resonance spectroscopy (MRS), diffusion-weighted (DW) MRI, or a combination of the 2, is best suited for prostate cancer detection and local staging.
Patients & Methods:
Thirty-seven consecutive patients with biopsy result of the prostate underwent MRI on a 1.5T scanner with combination of surface phased-array coil and endorectal coil (ERC) prior biopsy. Combined anatomical and functional imaging was performed with the use of T2-weighted sequences, DCE MRI, MRS and DW MRI. We compared the imaging results based on PI-RADS schem with Gleason score.
For the multiparametric approach, significantly higher sensitivity values, that is, 63% (95% confidence interval [CI]) were obtained as compared with each modality alone or any combination of the 3 modalities (P < .05). The specificity for this multiparametric approach, being 90.3% (95% CI) was not significantly higher (P < .05) as compared with the values of the combination of T2DCE MRI, T2 DW MRI, or T2MRS. Among the 3 techniques, T2DW MRI had the best performance for tumour detection in both the peripheral and the transition zone. High negative predictive value rates (>86%) were obtained for both tumor detection and local staging.
The combination of T2-weighted sequences, DCE MRI, MRS and DW MRI yields higher diagnostic performance for tumor detection and local staging than can any of these techniques alone or even any combination of them. These results suggest that mp-MRI can be used as a triage test among different patient populations, to select patients that can avoid biopsy and those that need re-biopsy before entering an active surveillance program. Time and cost can be saved by using only certain MRI sequences in patients.