Kinetic evaluation of [¹⁸F]MOZAT PET imaging in humans.

Objectives: The enzyme 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) catalyzes the conversion of cortisone into active cortisol, a neuroendocrine stress hormone. Levels of cortisol are altered in diseases such as Posttraumatic Stress Disorder (PTSD), Major Depressive Disorder (MDD), and obesity^1,2. In vivo imaging of 11β-HSD1 provides a potential method for probing the brain glucocorticoid system in such disorders. We have previously developed [¹¹C]AS2471907 as a novel PET radiotracer for imaging 11β-HSD1 in humans³. This study aims to evaluate its ¹⁸F-labeled counterpart, [¹⁸F]AS2471907, in humans to identify the optimal imaging and kinetic modeling methods, and to determine the test-retest variability of [¹⁸F]AS2471907 kinetic parameters in the brain. Methods Four trauma-exposed, otherwise healthy individuals participated in test and retest [¹⁸F]AS2471907 PET scans within 2 weeks of each other. PET imaging was acquired with a high-resolution research tomograph (HRRT) for 240 min, with a 30-min break at 120-150 min during which the participant exited the camera. An [¹⁸F]AS2471907 bolus injection of (average ± s.d., n=8) 99 ± 3 MBq with high specific activity (54 ± 26 MBq/nmol) was administered between 12 and 1 pm to control for possible diurnal variation in 11β-HSD1 levels due to circadian fluctuation in plasma cortisol concentration. Arterial blood sampling was conducted in all individuals for metabolite analysis and measurement of the plasma input function. The following AAL regions of interest (ROIs) were defined on co-registered participant MRI images: amygdala, caudate, cerebellum, hippocampus, hypothalamus, thalamus, putamen, and anterior and posterior cingulate, frontal, temporal, parietal and occipital cortices. Volumes of distribution (V_T) were estimated with 1-tissue (1T) and 2-tissue (2T) compartment models, and the multilinear analysis (MA1) method with fitting start time (t[asterisk]) of 30 min. Results We observed high uptake of [¹⁸F]AS2471907 throughout the brain (Figure 1A). The unmetabolized fraction of [¹⁸F]AS2471907 was 69 ± 11% at 240 min, with individual parent fraction curves (Fig. 1B) and arterial [¹⁸F]AS2471907 concentrations (Fig. 1C). SUV from 210-240 min varied from 0.47 in caudate to 1.66 in occipital cortex (Fig. 1D). The 1T model did not fit the data well in all regions, thus 2T was selected as the preferred model. K₁ values for the 2T model ranged from 0.018 ± 0.007 in hypothalamus to 0.043 ± 0.005 mL/cm³/min in cerebellum. V_T estimates ranged from 3.3 ± 0.9 in hypothalamus to 13.5 ± 5.6 mL/cm³ in occipital cortex. The MA1 model produced V_T estimates ranging from 3.2 ± 1.5 in hypothalamus to 13.4 ± 5.3 mL/cm³ in occipital cortex, which correlated extremely well with those from the 2T model (r=0.994). Absolute test-retest variability for V_T was 16.3 ± 5.5 for 2T model and 15.8 ± 4.0 for MA1. A high degree of proportionality between SUV and V_T was observed, with r-values ranging from 0.822-0.997 across scans. Conclusions These preliminary data show appropriate properties of [¹⁸F]AS2471907 for PET imaging of the 11β-HSD1 enzyme. Efforts are ongoing to optimize quantitation of [¹⁸F]AS2471907 and assess possible relationships with other variables such as BMI, age, gender, and cortisol levels. [¹⁸F]AS2471907 imaging in individuals with PTSD may yield further insight into differences in the distribution and function of 11β-HSD1 in the brain. <!--EndFragment-->