To investigate the additional value of (11)C-choline positron emission tomography (PET)-computed tomography (CT) to T2-weighted (T2w) magnetic resonance imaging (MRI) for localization of intraprostatic tumor nodules. Forty-nine prostate cancer patients underwent T2w MRI and (11)C-choline PET-CT before radical prostatectomy and extended lymphadenectomy. Tumor regions were outlined on the whole-mount histopathology sections and on the T2w MR images. Tumor localization was recorded in the basal, middle, and apical part of the prostate by means of an octant grid. To analyze (11)C-choline PET-CT images, the same grid was used to calculate the standardized uptake values (SUV) per octant, after rigid registration with the T2w MR images for anatomic reference. In total, 1,176 octants were analyzed. Sensitivity, specificity, and accuracy of T2w MRI were 33.5%, 94.6%, and 70.2%, respectively. For (11)C-choline PET-CT, the mean SUV(max) of malignant octants was significantly higher than the mean SUV(max) of benign octants (3.69 ± 1.29 vs. 3.06 ± 0.97, p < 0.0001) which was also true for mean SUV(mean) values (2.39 ± 0.77 vs. 1.94 ± 0.61, p < 0.0001). A positive correlation was observed between SUV(mean) and absolute tumor volume (Spearman r = 0.3003, p = 0.0362). No correlation was found between SUVs and prostate-specific antigen, T-stage or Gleason score. The highest accuracy (61.1%) was obtained with a SUV(max) cutoff of 2.70, resulting in a sensitivity of 77.4% and a specificity of 44.9%. When both modalities were combined (PET-CT or MRI positive), sensitivity levels increased as a function of SUV(max) but at the cost of specificity. When only considering suspect octants on (11)C-choline PET-CT (SUV(max) ≥ 2.70) and T2w MRI, 84.7% of these segments were in agreement with the gold standard, compared with 80.5% for T2w MRI alone. The additional value of (11)C-choline PET-CT next to T2w MRI in detecting tumor nodules within the prostate is limited.
Read full abstract