Comparative studies of radiolabeled choline positron emission tomography, histology of primary tumor and other imaging modalities in prostate cancer: a systematic review and meta-analysis

Abstract

The aim of this systematic review and meta-analysis was to assess the diagnostic performance of 18F/11C-labeled choline positron emission tomography (PET) or PET/computed tomography (CT) in the detection of prostate cancer and its metastases in comparison with histology of primary prostate cancer, other tracers and other imaging modalities. A PubMed and Web of Knowledge search was carried out to select English language articles, published before October 2012, dealing with the diagnostic performance of 18F- and 11C-labeled choline PET in the detection of prostate cancer in both staging and restaging, comparing it with histology of primary tumor, other imaging modalities and other tracers. Articles were included only if absolute numbers of true positive, true negative, false positive and false negative test results were available or derivable from the text. Reviews, clinical reports, and editorial articles were excluded. We re-analyzed all complete studies, performing qualitative and quantitative analyses. For the period 2003 to October 2012, we found 40 complete articles that critically evaluated the role of radiolabeled choline PET in comparison with histology of primary prostate cancer, magnetic resonance imaging (MRI), 18F-fluorodeoxyglucose (18F-FDG) PET/CT or other imaging modalities in prostate cancer patients. A meta-analysis was carried out on eight selected studies, comprising a total of 276 analyzed patients. The meta-analysis gave a pooled sensitivity of 62.6 % (95 % CI: 54–70.6) vs. 59.7 % (95 % CI: 51.1–67.9), and a pooled specificity of 76.3 % (95 % CI: 65.4–85.1) vs 76.1 % (95 % CI: 65.9–84.6) for radiolabeled choline PET/CT as compared to MRI for the detection of the primary lesion; a pooled sensitivity of 72 % (95 % CI: 66.9–76.6), and a pooled specificity of 61.6 % (95 % CI: 55.1–67.7) for the identification of the primary tumor in comparison with step section histopathology, and finally a pooled sensitivity of 65.1 % (95 % CI: 53.8–75.2) for radiolabeled choline PET/CT vs 39.8 (95 % CI: 29.2–51.1) for 18F-FDG PET/CT in the detection of prostate cancer metastases. Heterogeneity ranged between 0.0 and 94.0 %. The use of radiolabeled choline PET for the detection of primary prostate cancer may be unnecessary. On the contrary, radiolabeled choline PET and PET/CT are useful techniques in the detection of loco-regional and distant metastases in prostate cancer patients, being more sensitive than other tracers and other imaging modalities in any given clinical scenario.