Daniel Olivier Slosman MD, Presenter: Nothing to Disclose

Malika Quinodoz, Abstract Co-Author: Nothing to Disclose

Eric Fréneaux, Abstract Co-Author: Nothing to Disclose

Jerome Lang, Abstract Co-Author: Nothing to Disclose

Christophe Iselin, Abstract Co-Author: Nothing to Disclose

F-18 FluoroCholine (F-CHO) PET is applied in clinical oncology, mainly for prostate cancer imaging. Similarly to FDG PET, it remains a heterogeneous process that justifies recent recommendations for their use in National Cancer Institute clinical trials on evaluation of therapeutic response (Shankar, J Nucl Med, 2006). Our goal was to develop a quality control method feasible in clinical trials. For this purpose, we identified normal tissues of reference, calculated precision of measurement (%CV) of standardized uptake values (SUV) of F-CHO PET and characterized the quality control parameters.

F-CHO PET scan was performed in 50 patients with prostate cancer using PET/CT Biograph 16. Careful standardized protocols were applied. Measures of SUVmean/SUVmax were obtained for normal tissues of reference (lung, liver, and trabecular bone of L4 vertebral body) and pathological tissues (prostate cancer, metastatic lymph node and bone metastasis). Blind analyses were performed by 2 observers. %CV and least significant changes for a 95% level of confidence (LSC-95%) were calculated. Quality assessment was determined by calculating lower/upper limits of acceptance (mean SUV ± 2SD) and compared to LSC-95%.

Means +/- SD of SUVmean were 0.74+/-0.20, 8.28+/-1.59 and 2.88+/-0.67 for lung, liver and bone tissue respectively. For comparison, SUVmean were 5.03+/-1.09 for prostate cancer, 4.17+/-1.81 for metastatic lymph node and 5.71+/-1.38 for bone metastasis. %CV of normal tissues was better for SUVmean than SUVmax and the opposite for pathological tissues. SUVmean %CV were 2.5%, 1.5% and 8.8% for lung, liver and bone normal tissue while LSC-95% were 0.05, 0.42 and 0.62. Based on the inter-individual variability, normal SUVmean lower/upper limits could be characterized as 0.69/0.80 for lung tissue, 7.86/8.69 for liver tissue and 0.56-1.61 for bone tissue. Values of LSMC-95% were within these defined limits supporting the adequacy on these quality control parameters.

Standardized analysis of SUVmean of normal tissues of reference, with smaller precision inter-observer than inter-individual, enables appropriate quality control.

QA of F-CHO PET is feasible in clinical setting.

Slosman, D, Quinodoz, M, Fréneaux, E, Lang, J, Iselin, C, Quality Assessment of F-18 Fluoro-Choline PET Imaging in Clinical Trials: Definition of Standard Indicators and Validation in Patients with Prostate Cancer.  Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL. http://archive.rsna.org/2007/5006523.html