Purpose: Evaluation of the diagnostic value of the combined endorectal body-phased array technique regarding the staging of prostate cancers, especially in the differentiation between stages T2 and T3.
Materials and methods: Forty-two patients with biopsy-proven or clinically suspected prostate cancer were examined on a 1.5 T scanner (Siemens, Symphony) prior to radical prostatectomy. T (2)-weighted TSE (axial, coronal) and T (2)-weighted FSE (axial) sequences were obtained with and without fat suppression. After application of 0.2 mmol/kg body-weight Gd-DTPA, T (1)-weighted GRE sequences were obtained using dynamic MRI. All images were prospectively interpreted by two observers. The MR images were correlated with the histopathological findings of wide-area sections of prostatectomy specimens.
Results: For the detection of extracapsular growth and seminal vesicle infiltration (T2 versus T3) the accuracy was between 94 % and 97 % (sensitivity 100 %, specificity between 87 % and 93 %, observer 1 and 2). In two cases with a histologically proven stadium pT2b, observer 1 had diagnosed stadium pT3a. The results of observer 2 were marginally better in only one case, which was histologically proven to be pT2b and overstaged as pT3a. MRI did not lead to under-staging of a single tumor with regard to the differentiation between T2 and T3. Overall, the staging of the tumor stages (T1 - T4) was correct in 25 of 33 cases (75 %). The dynamic MRI showed no improvement regarding sensitivity (100 %) and specificity (62 %) and achieved a staging accuracy of only 75 %.
Conclusion: MRI performed with a combination of a pelvic phased-array coil (PPA) and integrated endorectal coil plays a significant role in the preoperative staging of prostate cancer. However, differentiation between capsular infiltration (T2) and penetration (T3) as well as evaluation of the seminal bladder (T3b) seem to be difficult.