Minimally invasive quantification of cerebral P2X7R occupancy using dynamic [18F]JNJ-64413739 PET and MRA-driven image derived input function
Nathalie Mertens, Mark E. Schmidt, Anja Hijzen, Donatienne Van Weehaeghe, Paulien Ravenstijn, Marleen Depré, Jan de Hoon, Koen Van Laere, Michel Koole
Abstract
Abstract [ 18 F]JNJ-64413739 has been evaluated as PET-ligand for in vivo quantification of purinergic receptor subtype 7 receptor (P2X7R) using Logan graphical analysis with a metabolite-corrected arterial plasma input function. In the context of a P2X7R PET dose occupancy study, we evaluated a minimally invasive approach by limiting arterial sampling to baseline conditions. Meanwhile, post dose distribution volumes (V T ) under blocking conditions were estimated by combining baseline blood to plasma ratios and metabolite fractions with an MR angiography driven image derived input function (IDIF). Regional postdose V T,IDIF values were compared with corresponding V T,AIF estimates using a arterial input function (AIF), in terms of absolute values, test–retest reliability and receptor occupancy. Compared to an invasive AIF approach, postdose V T,IDIF values and corresponding receptor occupancies showed only limited bias (Bland–Altman analysis: 0.06 ± 0.27 and 3.1% ± 6.4%) while demonstrating a high correlation (Spearman ρ = 0.78 and ρ = 0.98 respectively). In terms of test–retest reliability, regional intraclass correlation coefficients were 0.98 ± 0.02 for V T,IDIF compared to 0.97 ± 0.01 for V T,AIF. These results confirmed that a postdose IDIF, guided by MR angiography and using baseline blood and metabolite data, can be considered for accurate [ 18 F]JNJ-64413739 PET quantification in a repeated PET study design, thus avoiding multiple invasive arterial sampling and increasing dosing flexibility.