Purpose: To evaluate the use of semi-quantitative dynamic contrast-enhanced MRI (DCE-MRI) parameters for the detection of prostatic carcinoma in correlation to whole-mount histopathology.
Materials and methods: Fifty-three patients with biopsy-proven prostate cancer were examined by DCE-MRI at 1.5-T. Cancerous and benign prostatic tissue regions of interest were delineated based on the histopathology of whole-mount sections and several semi-quantitative parameters were calculated: time to peak (TTP), maximal contrast enhancement (Cpeak), speed of contrast uptake (Wash-in) and clearance rate of the contrast agent (Wash-out). The area under the ROC curve was determined for each parameter.
Results: Within individual patients, a consistently higher Cpeak and faster Wash-in were present in cancerous compared to benign prostatic tissue. Both the TTP and the Wash-out occurred more rapidly in tumour tissue than in normal prostatic tissue. Despite a considerable inter-patient overlap of parameter values between tumour and normal prostatic tissue, area under the ROC curve analysis demonstrated that the Wash-in was a good discriminator for cancer and benign tissue (AUC 0.82). Combination of the Wash-in and the Wash-out proved to be even more accurate (AUC 0.87) to discriminate between cancerous and benign prostatic regions.
Conclusion: The Wash-in is a useful parameter for prostate cancer detection by DCE-MRI.
Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.