Prospective comparison of computed tomography, diffusion-weighted magnetic resonance imaging and [11C]choline positron emission tomography/computed tomography for preoperative lymph node staging in prostate cancer patients

Abstract

The aim of this study was to prospectively compare diffusion-weighted magnetic resonance imaging (DWI) and [(11)C]choline positron emission tomography/computed tomography (PET/CT) with computed tomography (CT) for preoperative lymph node (LN) staging in prostate cancer (PCa) patients. Between June 2010 and May 2012, CT, DWI and [(11)C]choline PET/CT were performed preoperatively in 33 intermediate- and high-risk PCa patients undergoing radical prostatectomy (RP) and extended pelvic lymph node dissection (ePLND) including obturator fossa and internal, external and common iliac fields. Patient- and field-based performance characteristics for all three imaging techniques based on histopathological results are reported. Imaging techniques were compared by means of the area under the curve (AUC). LN metastases were detected in 92 of 1,012 (9%) LNs from 14 of 33 (42%) patients. On patient-based analysis, sensitivity, specificity and accuracy for CT were 57, 68 and 64%, respectively, for DWI were 57, 79 and 70%, respectively, and for [(11)C]choline PET/CT were 57, 90 and 76%, respectively. On field-based analysis, these numbers for CT were 47, 94 and 88%, respectively, for DWI were 56, 97 and 92%, respectively, and for [(11)C]choline PET/CT were 62, 96 and 92%, respectively. Neither DWI nor [(11)C]choline PET/CT performed significantly better than CT on pairwise comparison of patient- and field-based results. All three imaging techniques exhibit a rather low sensitivity with less than two thirds of LN metastases being detected on patient- and field-based analysis. Overall diagnostic efficacy did not differ significantly between imaging techniques, whereas distinct performance characteristics, esp. patient-based specificity, were best for [(11)C]choline PET/CT followed by DWI and CT.