TY - JOUR T1 - Quantitation of glutamate mGluR5 receptor with 18F-FPEB PET in humans JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 215 LP - 215 VL - 51 IS - supplement 2 AU - Olivier Barret AU - Gilles Tamagnan AU - Jeff Batis AU - Danna Jennings AU - George Zubal AU - David Russell AU - Kennth Marek AU - John Seibyl Y1 - 2010/05/01 UR - http://jnm.snmjournals.org/content/51/supplement_2/215.abstract N2 - 215 Objectives Altered glutaminergic function is implicated in several neuropsychiatric disorders such as schizophrenia, neurodegenerative diseases, and autism and medications with potency at the mGluR5 receptor are now in clinical trials. The aim of this study is to characterize 18F-FPEB, a novel mGluR5 tracer in humans, and evaluate simple tissue ratio against different non invasive kinetic models using the cerebellum as reference region (very low mGluR5 levels in humans). Methods 12 subjects (4 Healthy Control, 6 Parkinson’s Disease and 2 Huntington’s Disease) were imaged with 18F-FPEB PET over 4.5h. Images were realigned and spatially normalized using an in-house FPEB template. VOI templates were overlaid to get regional brain time activity curves (TAC). Binding potential BPnd, proportional to mGluR5 receptor density, was evaluated assuming the cerebellum as reference region using simple tissue ratio over the last 2h of acquisition (TACs at quasi-equilibrium), multilinear (MRTM) and linear (Logan) as well as non linear compartmental (FRTM, SRTM and Watabe) regression analyses where Watabe uses two compartments for the cerebellum and FRTM and SRTM only one. Results Both FRTM and SRTM failed to describe the TACs whereas Watabe gave a very good fit in all the regions. Logan and MRTM with t*=30min also described adequately the TACs in all regions. A paired samples T-test on BPnd revealed no significant difference between MRTM and Logan (p>0.05) and Watabe to be higher by 3% compared to MRTM and Logan (p<0.001). The ratio method gave an apparent BPnd significantly higher (70%) than other methods, with however a strong correlation to Watabe, MRTM and Logan BPnd (r=0.9, p<0.001). Conclusions These preliminary results suggest that quantitation of mGluR5 receptor with 18F-FPEB with non-invasive modeling using cerebellum as reference region may be feasible. The slightly higher Watabe BPnd may be due to low specific binding in the cerebellum. The strong correlation of the ratio method apparent BPnd with other estimates makes it a valuable outcome measure ER -