Tumour volume delineation in prostate cancer assessed by [11C]choline PET/CT: validation with surgical specimens

Eur J Nucl Med Mol Imaging. 2013 Jun;40(6):824-31. doi: 10.1007/s00259-013-2345-7. Epub 2013 Feb 7.

Abstract

Purpose: PET has been proven to be helpful in the delineation of gross tumour volume (GTV) for external radiation therapy in several tumour entities. The aim of this study was to determine if [(11)C]choline PET could be used to localize the carcinomatous tissue within the prostate in order to specifically target this area for example with high-precision radiation therapy.

Methods: Included in this prospective study were 20 patients with histological proven prostate carcinoma who underwent [(11)C]choline PET/CT before radical prostatectomy. After surgical resection, specimens were fixed and cut into 5-mm step sections. In each section the area of the carcinoma was delineated manually by an experienced pathologist and digitalized, and the histopathological tumour volume was calculated. Shrinkage due to resection and fixation was corrected using in-vivo and ex-vivo CT data of the prostate. Histopathological tumour location and size were compared with the choline PET data. Different segmentation algorithms were applied to the PET data to segment the intraprostatic lesion volume.

Results: A total of 28 carcinomatous lesions were identified on histopathology. Only 13 (46 %) of these lesions had corresponding focal choline uptake. In the remaining lesions, no PET uptake (2 lesions) or diffuse uptake not corresponding to the area of the carcinoma (13 lesions) was found. In the patients with corresponding PET lesions, no suitable SUV threshold (neither absolute nor relative) was found for GTV segmentation to fit the volume to the histological tumour volume.

Conclusion: The choline uptake pattern corresponded to the histological localization of prostate cancer in fewer than 50 % of lesions. Even when corresponding visual choline uptake was found, this uptake was highly variable between patients. Therefore SUV thresholding with standard algorithms did not lead to satisfying results with respect to defining tumour tissue in the prostate.

MeSH terms

  • Algorithms
  • Carbon Radioisotopes / pharmacology
  • Carcinoma / diagnosis
  • Carcinoma / pathology
  • Choline / pharmacology
  • Humans
  • Image Processing, Computer-Assisted
  • Male
  • Multimodal Imaging*
  • Prospective Studies
  • Prostatectomy
  • Prostatic Neoplasms / diagnosis*
  • Prostatic Neoplasms / pathology*
  • Radiopharmaceuticals / pharmacology
  • Reproducibility of Results
  • Tumor Burden*

Substances

  • Carbon Radioisotopes
  • Radiopharmaceuticals
  • Choline