PET quantification of brain O-GlcNAcase with [(18)F]LSN3316612 in healthy human volunteers

BACKGROUND: Previous studies found that [(18)F]LSN3316612 was a promising positron emission tomography (PET) radioligand for imaging O-GlcNAcase in nonhuman primates and human volunteers. This study sought to further evaluate the suitability of [(18)F]LSN3316612 for human clinical research. METHODS: Kinetic evaluation of [(18)F]LSN3316612 was conducted in a combined set of baseline brain scans from 17 healthy human volunteers and test-retest imaging was conducted in 10 of these volunteers; another 6 volunteers had whole-body scans to measure radiation exposure to body organs. Total distribution volume (V(T)) estimates were compared for the one- and two-tissue compartment models with the arterial input function. Test-retest variability and reliability were evaluated via mean difference and intraclass correlation coefficient (ICC). The time stability of V(T) was assessed down to a 30-min scan time. An alternative quantification method for [(18)F]LSN3316612 binding without blood was also investigated to assess the possibility of eliminating arterial sampling. RESULTS: Brain uptake was generally high and could be quantified as V(T) with excellent identifiability using the two-tissue compartment model. [(18)F]LSN3316612 exhibited good absolute test-retest variability (12.5%), but the arithmetic test-retest variability was far from 0 (11.3%), reflecting a near-uniform increase of V(T) on the retest scan in nine of 10 volunteers. V(T) values were stable after 110 min in all brain regions, suggesting that no radiometabolites accumulated in the brain. Measurements obtained using only brain activity (i.e., area under the curve (AUC) from 150–180 min) correlated strongly with regional V(T) values during test-retest conditions (R(2) = 0.84), exhibiting similar reliability to V(T) (ICC = 0.68 vs. 0.64). Estimated radiation exposure for [(18)F]LSN3316612 PET was 20.5 ± 2.1 μSv/MBq, comparable to other (18)F-labeled radioligands for brain imaging. CONCLUSIONS: [(18)F]LSN3316612 is an excellent PET radioligand for imaging O-GlcNAcase in the human brain. Alternative quantification without blood is possible, at least for within-subject repeat studies. However, the unexplained increase of V(T) under retest conditions requires further investigation.