Optimized method for normal range estimation of standardized uptake values (SUVmax, SUVmean) in liver SPECT/CT images with somatostatin analog [99mTc]-HYNIC-TOC (Tektrotyd)

BACKGROUND: 99mTc-hydrazinonicotinyl-Tyr3-octreotide ([99mTc]-HYNIC-TOC [Tektrotyd]) is a radiopharmaceutical used for the diagnosis of lesions with overexpression of somatostatin receptors. The purpose of this study was to optimize the method and estimate normal ranges for standardized uptake values of Tektrotyd in healthy livers. MATERIAL AND METHODS: An analysis of standardized uptake value (SUVs) normal ranges was performed for images acquired in a selected "healthy group" of 42 patients evaluated for neuroendocrin tumors. The "pathological group" comprised 20 patients with liver lesions detected by scintigraphic imaging. Normal ranges for radiopharmaceutical uptake values were estimated based on the quantitative analysis of images acquired with a GE Healthcare NM/CT 850 gamma camera. RESULTS: The method for healthy liver segmentation in single photon emission computed tomography/computed tomography (SPECT/CT) was optimized. The normal range of SUVs for the liver was: standardized uptake value body weight (SUVbw) max [5.2-14.0] g/mL and standardized uptake value lean body mass (SUVlbm) [3.5-9.5] g/mL. The relative standard error (relative SE) of activity concentration estimated in the phantom study for the largest hot spheres was: ϕ = 37 mm - 5.9%, ϕ = 28 mm - 7.1%, ϕ = 22 mm - 11.4%, and ϕ = 17 mm - 22%. CONCLUSIONS: Segmentation in the mid-coronal computed tomography (CT) image, at one-fourth of the height of the liver measured from the top, with a medium-sized volume of interest (VOI) outlined on a given transverse SPECT slice was regarded as the optimal method for estimating normal ranges for standardized uptake values. It is necessary to standardize quantification methods in the SPECT/CT studies. Our work is a step forward in obtaining standardization of SPECT/CT SUV calculation methods. Calculations for radiopharmaceutical uptake in tumors with volumes smaller than 5 mL are biased with a significant measurement error.