TY - JOUR T1 - Quantification of <sup>18</sup>F-Fluorocholine Kinetics in Patients with Prostate Cancer JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 365 LP - 371 DO - 10.2967/jnumed.114.148007 VL - 56 IS - 3 AU - Eline E. Verwer AU - Daniela E. Oprea-Lager AU - Alfons J.M. van den Eertwegh AU - Reindert J.A. van Moorselaar AU - Albert D. Windhorst AU - Lothar A. Schwarte AU - N. Harry Hendrikse AU - Robert C. Schuit AU - Otto S. Hoekstra AU - Adriaan A. Lammertsma AU - Ronald Boellaard Y1 - 2015/03/01 UR - http://jnm.snmjournals.org/content/56/3/365.abstract N2 - Choline kinase is upregulated in prostate cancer, resulting in increased 18F-fluoromethylcholine uptake. This study used pharmacokinetic modeling to validate the use of simplified methods for quantification of 18F-fluoromethylcholine uptake in a routine clinical setting. Methods: Forty-minute dynamic PET/CT scans were acquired after injection of 204 ± 9 MBq of 18F-fluoromethylcholine, from 8 patients with histologically proven metastasized prostate cancer. Plasma input functions were obtained using continuous arterial blood-sampling as well as using image-derived methods. Manual arterial blood samples were used for calibration and correction for plasma-to-blood ratio and metabolites. Time–activity curves were derived from volumes of interest in all visually detectable lymph node metastases. 18F-fluoromethylcholine kinetics were studied by nonlinear regression fitting of several single- and 2-tissue plasma input models to the time–activity curves. Model selection was based on the Akaike information criterion and measures of robustness. In addition, the performance of several simplified methods, such as standardized uptake value (SUV), was assessed. Results: Best fits were obtained using an irreversible compartment model with blood volume parameter. Parent fractions were 0.12 ± 0.4 after 20 min, necessitating individual metabolite corrections. Correspondence between venous and arterial parent fractions was low as determined by the intraclass correlation coefficient (0.61). Results for image-derived input functions that were obtained from volumes of interest in blood-pool structures distant from tissues of high 18F-fluoromethylcholine uptake yielded good correlation to those for the blood-sampling input functions (R2 = 0.83). SUV showed poor correlation to parameters derived from full quantitative kinetic analysis (R2 &lt; 0.34). In contrast, lesion activity concentration normalized to the integral of the blood activity concentration over time (SUVAUC) showed good correlation (R2 = 0.92 for metabolite-corrected plasma; 0.65 for whole-blood activity concentrations). Conclusion: SUV cannot be used to quantify 18F-fluoromethylcholine uptake. A clinical compromise could be SUVAUC derived from 2 consecutive static PET scans, one centered on a large blood-pool structure during 0–30 min after injection to obtain the blood activity concentrations and the other a whole-body scan at 30 min after injection to obtain lymph node activity concentrations. ER -