Towards standardization of ^sup 18^F-FET PET imaging: What is the best method for background activity assessment?

Objectives: PET with O-(2-18F-fluoroethyl)-L-tyrosine (18F-FET)has reached increasing clinical significance for patients with brain neoplasms. Regarding assessment of PET-derived parameters such as the maximal tumor-to-background-ratio, there is a need of a background-activity reference using a region-of-interest (ROI) or volume-of-interest (VOI) in unaffected brain tissue. In clinical routine, however, there is no standardized approach for the procedure and shape of such a background reference. Therefore, we intended to evaluate the intra- and inter-reader variability of commonly applied approaches for clinical 18F-FET PET imaging. Methods: The background activity of 20 randomly selected 18F-FET PET scans were consecutively evaluated by 6 readers using I) a simple 2D-ROI, II) a spherical VOI with 3.0 cm diameter and III) a VOI consisting of 6 crescent-shaped ROIs. Each method, derived from the current literature, is applied in the contralateral, non-affected hemisphere including white and grey matter in line with the DGN- and EANM-guidelines. In a second step, 10 further scans were read using a guided and standardized background VOI. To evaluate intra-reader variability, each scan was evaluated ten times by each reader. Results are displayed as 'coefficient of variation' (CoV), the standardized measure of dispersion as defined as ratio of the standard deviation to the mean. Results: In each of the six readers, the use of a crescent-shaped VOI revealed the most stable background activity values with the significantly lowest median intra-reader CoV (p