Dynamic MRI and CAD vs. choline MRS vs. choline PET/CT: Where is the detection level for a lesion characterization in prostate cancer?

Objectives To evaluate prospectively the role of pre-interventional fused high resolution T2 MR images with parametrically analyzed dynamic contrast enhanced T1 MR images (DCE-MRI) and 1H MR spectroscopy (MRS) for a precise biopsy for the detection of prostate cancer and for radiation treatment planning of IMRT.

Methods Inclusion criteria: 55 pts with pathological PSA and/or previously negative TRUS-biopsy. Standardized biopsy of the prostate divided into 20 regions. Image fusion of colored parametric maps derived from DCE-MRI and MRS (single voxel spectroscopy, SVS; chemical shift imaging, CSI) with T2 images. Correlation with histology and cytokeratin (CK34βE1.2) positive areas in prostatectomy species and with choline PET/CT.

Results DCE-MRI: Sensitivity 82%, specificity 89%, accuracy 88%, PPV 61%, NPV 96%. SVS: Sensitivity 55%, specificity 62%. CSI: Sensitivity 68%, specificity 67%. False positive findings due to prostatitis, adenomatous hyperplasia, false negative findings due to low signal (PIN, cut-off level for DCE-MRI: lesions smaller 3mm and less than 30% cancer cells, for SVS: lesions smaller 8mm and less than 50% cancer cells, for CSI: lesions smaller 4mm and less than 40% cancer cells).

Conclusions Choline PET/CT is superior to choline MRS. Similar accuracy of DCE-MRI and choline PET/CT. DCE-MRI and MRS are helpful for a precise biopsy of the prostate. The ESTRO-guidelines 2006 for radiation treatment planning of the prostate have to be revised, if the standardized biopsy will be replaced by a lesion orientated biopsy. Till now it is unclear, if the parametric maps of DCE-MRI and MRS can be used for IMRT of the prostate.