PURPOSE: To determine whether cancers of the prostate transition zone (TZ) possess a unique metabolic pattern by which they may be identified at proton magnetic resonance (MR) spectroscopic imaging. MATERIALS AND METHODS: Findings in 40 patients who underwent combined endorectal MR imaging and hydrogen 1 MR spectroscopic imaging before radical prostatectomy and who had TZ tumor identified subsequently at step-section pathologic analysis were retrospectively reviewed. Within this population, a subset of 16 patients whose TZ tumor had a largest diameter of 1 cm or greater and was included in the MR spectroscopic imaging excitation volume was identified. In these 16 patients, the ratios of choline-containing compounds (Cho) and creatine/phosphocreatine (Cr) to citrate (Cit) (ie, [Cho + Cr]/Cit), Cho/Cr, and Cho/Cit were compared in tumor and control tissues. The presence of only Cho and the absence of all metabolites were also assessed. RESULTS: The mean values of (Cho + Cr)/Cit, Cho/Cr, and Cho/Cit were different between TZ cancer and control tissues (P =.001, P =.003, and P =.001, respectively; Wilcoxon signed rank test). Nine (56%) of 16 patients had at least one tumor voxel in which Cho comprised the only detectable peak, while no control voxels showed only Cho (P =.008, McNemar test). The percentage of voxels in which no metabolites were detected did not differ between tumor and control tissues (P =.134, McNemar test). CONCLUSION: TZ cancer has a metabolic profile that is different from that of benign TZ tissue; however, the broad range of metabolite ratios observed in TZ cancer precludes the use of a single ratio to differentiate TZ cancer from benign TZ tissue. Copyright RSNA, 2003
PURPOSE: To determine whether cancers of the prostate transition zone (TZ) possess a unique metabolic pattern by which they may be identified at proton magnetic resonance (MR) spectroscopic imaging. MATERIALS AND METHODS: Findings in 40 patients who underwent combined endorectal MR imaging and hydrogen 1 MR spectroscopic imaging before radical prostatectomy and who had TZtumor identified subsequently at step-section pathologic analysis were retrospectively reviewed. Within this population, a subset of 16 patients whose TZtumor had a largest diameter of 1 cm or greater and was included in the MR spectroscopic imaging excitation volume was identified. In these 16 patients, the ratios of choline-containing compounds (Cho) and creatine/phosphocreatine (Cr) to citrate (Cit) (ie, [Cho + Cr]/Cit), Cho/Cr, and Cho/Cit were compared in tumor and control tissues. The presence of only Cho and the absence of all metabolites were also assessed. RESULTS: The mean values of (Cho + Cr)/Cit, Cho/Cr, and Cho/Cit were different between TZcancer and control tissues (P =.001, P =.003, and P =.001, respectively; Wilcoxon signed rank test). Nine (56%) of 16 patients had at least one tumor voxel in which Cho comprised the only detectable peak, while no control voxels showed only Cho (P =.008, McNemar test). The percentage of voxels in which no metabolites were detected did not differ between tumor and control tissues (P =.134, McNemar test). CONCLUSION:TZcancer has a metabolic profile that is different from that of benign TZ tissue; however, the broad range of metabolite ratios observed in TZcancer precludes the use of a single ratio to differentiate TZcancer from benign TZ tissue. Copyright RSNA, 2003
Authors: Terence P F Gade; Jason A Koutcher; William M Spees; Bradley J Beattie; Vladimir Ponomarev; Michael Doubrovin; Ian M Buchanan; Tatiana Beresten; Kristen L Zakian; H Carl Le; William P Tong; Philipp Mayer-Kuckuk; Ronald G Blasberg; Juri G Gelovani Journal: Cancer Res Date: 2008-04-15 Impact factor: 12.701