Population-based input function is an accurate alternative to arterial sampling in [11C](R)-rolipram PET brain studies

Objectives We explored the use of a population-based input function (PBIF) for [11C](R)-rolipram as an alternative to arterial blood sampling (AIF).

Methods The PBIF was generated using [11C](R)-rolipram parent time-activity curves from 12 healthy volunteers and a “leave-one-out” procedure (a PBIF was generated for each of the 12 subjects by averaging the measured AIF of the remaining 11 subjects). Different scaling methods were tested. The first method was scaling using parent plasma values. To determine the optimal sampling time(s) we performed a Pearson’s linear correlation analysis. We also evaluated non-invasive scaling methods (body weight, body surface area, and body lean mass, calculated with different formulas). The scaling method that gave the best estimation of the area under the curve (AUC) and Logan-VT was prospectively applied to another population of 23 healthy subjects,and to a population of 32 patients with depression.

Results Blood-scaled PBIF gave the best estimate of AIF(Table). Excellent results were obtained when the blood-scaled PBIF was applied to the other healthy subjects (AUC ratio 1.01±0.07; VT ratio 1.02±0.07) and to patients (AUC ratio 1.02±0.06; VT ratio 1.03±0.04).

Conclusions Blood-scaled PBIF accurately estimates AIF in [11C](R)-rolipram studies. A PBIF derived from healthy volunteers can be successfully applied to patients

• Download figure
• Open in new tab
• Download powerpoint

PBIF results according to the scaling method