Kinetic Modeling of the New σ1 Receptor Ligand (-)-[18F]Fluspidine in the Human Brain

Objectives:

The σ1 receptor, a transmembrane protein located at the endoplasmatic reticulum is involved in a variety of neuropsychiatric diseases, e.g. depression, schizophrenia and drug addiction. The newly developed PET tracer (-)-[18F]Fluspidine was successfully applied to quantify σ1 receptors in the porcine brain [1]. Here we present the first PET quantification of σ1 receptors with (-)-[18F] Fluspidine in humans.
Methods:
After intravenous administration of 269.6±13.3 MBq (-)-[18F]Fluspidine PET brain imaging was performed in 10 healthy subjects (age 36.6±14.8 years; gender 5F/5M) using an ECAT EXACT HR+ system in 3D-acquisition mode. 26 frames were acquired from 0-210 min post injection and motion corrected with SPM2. Kinetic modeling using 1- and 2-tissue compartment models (1TCM, 2TCM) with metabolite corrected arterial input-function was applied to the volume of interest (VOI) based tissue time-activity curves (TACs) in 43 brain regions (anatomically defined via MRI co-registration). Time ranges from 0 to 90 and 0 to 210 min were investigated. Model-based receptor parameter was the total distribution volume VT (ml/cm-3), a linear function of receptor density.
Results:
TACs of all 43 regions could be described with the 1- and 2TCM. VT in all cortical regions could be reliably estimated from 90 min PET data already. In white matter longer measurements can be necessary. The distribution volume was highest in the cerebellar cortex (31.4±6.1), low in the centrum semiovale (17.7±7.1) and ranged in cortical structures between 20.9±3.9 in the orbitofrontal and 24.9±5.7 in the posterior cingulate cortex (pcc) (2TCM, 90 min). The distribution volumes computed from 210 min data were comparable to 90 min results, e.g. in pcc 25.7±5.9 (2TCM) and 25.7±6.0 (1TCM).
Conclusions:
σ1 receptor parameters in cortical structures can be estimated with a 1- or 2TCM from 90 min (-)-[18F]Fluspidine TACs. If a model derived receptor parameter is used in a classification problem, e.g., distinguishing patients with depression from healthy controls, the final model decision should be made on the basis of the PET data of both groups.
References:
P. Brust, ..., O. Sabri, Journal of Nuclear Medicine 2014, 55, 1730-1736