Noninvasive quantification of [18F]UCB-H binding using microPET and population-based input function.

Introduction: [18F]UCB-H is a validated radiotracer with a high affinity for the synaptic vesicle glycoprotein 2A (SV2A), known as the binding site of the antiepileptic drug levetiracetam [1, 2]. Due to the absence of reference region, the major drawback of SV2A microPET imaging in the living rat brain is the invasiveness and the complexity of the arterial input function measurement needed for a full quantification. We provide here evaluation of a population-based input function (PBIF) to estimate input function of [18F]UCB-H.
Methods: Standard arterial input functions were measured with an arteriovenous shunt and a β-microprobe system from eight anesthetized Sprague-Dawley (SD) rats, as previously described [2]. The distribution volume (Vt) for [18F]UCB-H was calculated with Logan graphic analysis. After normalization, all individual AIFs were averaged to provide the PBIF, and the Logan graphical analysis was computed on each individual rat using the PBIF instead of individual AIF. Correlations analyses were computed between Vt obtained with each methods (individual AIF vs PBIF). Finally, another cohort of five SD rats was scanned with [18F]UCB-H, and Vt were computed using the PBIF and Logan analysis. Single blood samples were harvested at 15 min after radiotracer injection, just to ensure the consistency of the metabolic parent fraction.
Results: The Vt computed with individual AIFs were higly consistent with previously reported values, so are the Vt computed with the PBIF [2]. Individual AIFs Vt and PBIF Vt are highly correlated through all brain areas for the height subjects (r2 =0.9). Coefficients of variance are slightly higher with the PBIF method compare to the individual AIF method (14 % and 9 % respectively for the whole brain). Finally, Vt measurement in the second cohort were consistent with previously reported values, and the metabolization profile matched the parent fraction described by Warnock and coll. [2].
Conclusions: The present study described a method for the noninvasive estimation of the AIF using a PBIF, carrying a potential that might substitute for conventional invasive, indivi- dual AIF measurement. We propose that this method can provide a reasonable solution for longitudinal quantitative [18F]UCB-H microPET studies.