Measurement of enhancement kinetics using Heuristic parameters in multiparametric MRI as markers of prostate cancer aggressiveness
Session type: Poster / e-Poster / Silent Theatre session
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has been used as a diagnostic modality with known added value in multi-parametric MRI (mp-MRI) characterisation of prostate cancers (PCa). Several methods have been applied to analyse DCE MRI data. The purpose of this study is to assess the ability of mp-MRI for detecting grade of prostate carcinomas in comparison to pathology using heuristic model-free parametric evaluation (wash-in and wash-out slope, etc.) provided by Tissue 4D (Siemens Multi-Modality Work Platform).
21 patients with histologically proven prostate cancer who were scanned at 3T mp-MRI and opted for laparoscopic radical prostatectomy were recruited in this study. Post-prostatectomy pathology reports were considered as reference standard. The pathology specimens were processed by a special method to ensure exact orientation between imaging and slices of section during histopathology. Multi-parametric MRI data including conventional T1 and T2 weighted imaging (T1WI and T2WI), DWI and DCE-MRI was recorded. DCE-MRI was analysed in Tissue 4D using the two different methods: tofts model (Ktrans, kep values and the colour-distribution map for the perfusion parameters); wash-in and –out rates, which were calculated according to the parametric time-concentration curves.
There are 46 lesions detected in pathology of 21 prostatectomy specimen and mp-MRI correctly picked up 33 of them. Smaller lesions (2mm and less) with Gleason grade 3+4 were not identified by DCE-MRI. There was a strong trend towards a positive correlation between grade of cancer and enhancement rate as assessed by toft model.
DCE-MRI with the two difference methods could predict aggressiveness of prostate cancers more than 2mm in size. Smaller lesions and those with Gleason score 7 and less could still be a challenge for multiparamteric MRI detection method.