Purpose: (R)-[(11)C]rolipram and (S)-[(11)C]rolipram have been proposed to investigate phosphodiesterase-4 and, indirectly, cAMP-mediated signaling with PET. This study assessed binding of these tracers to phosphodiesterase-4 in canine myocardium.
Procedures: Seven dogs underwent (R)-[(11)C]rolipram and (S)-[(11)C]rolipram dynamic PET imaging at baseline and with co-injection of saturating doses of (R)-rolipram. Dual-input compartment models were applied to estimate the volumes of distribution (V(T)).
Results: The model comprising one compartment for unmetabolized tracer and one compartment for labeled metabolites provided excellent fits to data acquired with (S)-[(11)C]rolipram at baseline and with both enantiomers during co-injection scans. Use of two compartments for unmetabolized (R)-[(11)C]rolipram at baseline was warranted according to Akaike and Schwarz criteria. V(T) estimates obtained with these models were robust (CV ≤ 8.2%) and reproducible (CV ≤ 15%).
Conclusion: An important fraction (~65%) of the V (T) of (R)-[(11)C]rolipram at baseline reflects specific binding. Thus, the latter may be a useful index of phosphodiesterase-4 levels in canine myocardium.