Evaluation of bolus and infusion methods for quantification of pancreatic VMAT2 binding using [¹⁸F]FP-(+)-DTBZ in humans

Objectives Vesicular monoamine transporter type 2 (VMAT2) imaging with [¹⁸F]FP-(+)-DTBZ is considered a promising biomarker for insulin-secreting pancreatic β cells (Normandin et al, 2012). Previous PET studies with this tracer used a bolus injection protocol. However, modeling analysis indicated the need for a long scan to estimate stable kinetic parameters. The aim of this study was to compare bolus+infusion (B+I), which would facilitate using a shorter scan period and simplify analysis, and bolus (B) study designs in healthy controls (HC) and patients with type 1 diabetes (T1D).

Methods Subjects were scanned on a PET/CT scanner for 4 h with B (2 HC) or B+I (3 HC and 1 T1D, K_bol=360 min) administration of [¹⁸F]FP-(+)-DTBZ (254 ± 33 MBq). The metabolite-corrected input function was measured. Three time-activity curves were generated from regions of interest (ROIs) delineated on pancreas, renal cortex, and spleen. Multilinear analysis (MA1) was applied to estimate distribution volume (V_T). The tissue to plasma ratio method was also applied to the 150-240 min scan data to estimate apparent V_T (V_A).

Results [¹⁸F]FP-(+)-DTBZ V_T values were highest in the pancreas, followed by spleen and renal cortex. Kidney and spleen are not clearly visible in late images (>150 min). Pancreas V_T in T1D was smaller than that of HC (93 vs. 165 ± 34 mL/cm³). Equilibrium with plasma was reached in 90-120 min in the spleen and renal cortex. For pancreas, equilibrium time was after 150 min post-injection, and varied across subjects (e.g., equilibrium time was ~200 min in one HC). For B studies, V_A showed substantial overestimation of V_T ((1.8 ± 0.1) * V_T)) . For B+I studies, V_A values were similar to V_T ((1.1 ± 0.1) * V_T).

Conclusions B+I studies of [¹⁸F]FP-(+)-DTBZ provided very similar V_T estimates to the B method. B+I method with only a late scan may be an acceptable approach if accurate definition of low uptake ROIs can be accomplished.

Research Support JDRF 37-2011-22