Estimation of serotonin transporter parameters with 11C-DASB in healthy humans: reproducibility and comparison of methods.

The aim of the present study was to define the optimal analytic method to derive accurate and reliable serotonin transporter (SERT) receptor parameters with (11)C-3-amino-4-(2-(dimethylamino)methyl]phenylthio)benzonitrile ((11)C-DASB). METHODS: Nine healthy subjects (5 females, 4 males) underwent two (11)C-DASB PET scans on the same day. Five analytic methods were used to estimate binding parameters in 10 brain regions: compartmental modeling with 1- and 2-tissue compartment models (1TC and 2TC), data-driven estimation of parametric images based on compartmental theory (DEPICT) analysis, graphical analysis, and the simplified reference tissue model (SRTM). Two variations in the fitting procedure of the SRTM method were evaluated: nonlinear optimization and basis function approach. The test/retest variability (VAR) and intraclass correlation coefficient (ICC or reliability) were assessed for 3 outcome measures: distribution volume (V(T)), binding potential (BP), and specific to nonspecific equilibrium partition coefficient (V(3)'). RESULTS: All methods gave similar values across all regions. The variability of V(T) was excellent (< or =10%) in all regions, for the 1TC, 2TC, DEPICT, and graphical approaches. The variability of BP and V(3)' was good in regions of high SERT density and poorer in regions of moderate and lower densities. The ICC of all 3 outcome measures was excellent in all regions. The basis function implementation of SRTM demonstrated improved reliability compared with nonlinear optimization, particularly in moderate and low-binding regions. CONCLUSION: The results of this study indicate that (11)C-DASB can be used to measure SERT parameters with high reliability and low variability in receptor-rich regions of the brain, with somewhat less reliability and increased variability in regions of moderate SERT density and poor reproducibility in low-density regions.