First clinical assessment of [ 18 F]MC225, a novel fluorine-18 labelled PET tracer for measuring functional P-glycoprotein at the blood–brain barrier

Objective 5-(1-(2-[ 18 F]fluoroethoxy))-[3-(6,7-dimethoxy-3,4-dihydro-1H-isoquinolin-2-yl)-propyl]-5,6,7,8-tetrahydronaphthalen ([ 18 F]MC225) is a selective substrate for P-glycoprotein (P-gp), possessing suitable properties for measuring overexpression of P-gp in the brain. This is the first-in-human study to examine safety, radiation dosimetry and P-gp function at the blood–brain barrier (BBB) of [ 18 F]MC225 in healthy subjects. Methods [ 18 F]MC225 biodistribution and dosimetry were determined in 3 healthy male subjects, using serial 2 h and intermittent 4 and 6 h whole-body PET scans acquired after [ 18 F]MC225 injection. Dynamic [ 18 F]MC225 brain PET (90 min) was obtained in 5 healthy male subjects. Arterial blood was sampled at various time intervals during scanning and the fraction of unchanged [ 18 F]MC225 in plasma was determined. T1-weighted MRI was performed for anatomical coregistration. Total distribution volume ( V T ) was estimated using 1- and 2-tissue-compartment models (1-TCM and 2-TCM, respectively). V T was also estimated using the Logan graphical method (Logan plot) ( t * = 20 min). Surrogate parameters without blood sampling (area-under the curve [AUC] of regional time–activity curves [TACs] and negative slope of calculated TACs) were compared with the V T values. Results No serious adverse events occurred throughout the study period. Although biodistribution implied hepatobiliary excretion, secretion of radioactivity from liver to small intestine through the gallbladder was very slow. Total renal excreted radioactivity recovered during 6 h after injection was < 2%ID. Absorbed dose was the highest in the pancreas (mean ± SD, 203 ± 45 μGy/MBq) followed by the liver (83 ± 11 μGy/MBq). Mean effective dose with and without urination was 17 ± 1 μSv/MBq. [ 18 F]MC225 readily entered the brain, distributing homogeneously in grey matter regions. 2-TCM provided lower Akaike information criterion scores than did 1-TCM. V T estimated by Logan plot was well correlated with that of 2-TCM ( r 2 > 0.9). AUCs of TACs were positively correlated with V T (2-TCM) values ( r 2 : AUC 0-60 min = 0.61, AUC 0-30 min = 0.62, AUC 30-60 min = 0.59, p < 0.0001). Negative slope of SUV TACs was negatively correlated with V T (2-TCM) values ( r 2 = 0.53, p < 0.0001). Conclusions This initial evaluation indicated that [ 18 F]MC225 is a suitable and safe PET tracer for measuring P-gp function at the BBB.