Background: Cyclooxygenase-2 (COX-2) is present in healthy brain at low densities but can be markedly upregulated by excitatory input and by inflammogens. This study evaluated the sensitivity of the positron emission tomography (PET) radioligand [11C]MC1 to detect COX-2 density in healthy human brain. 
Methods: The specificity of [11C]MC1 was confirmed using lipopolysaccharide (LPS)-injected rats and transgenic mice expressing human COX-2 gene, with 120-minute baseline and blocked scans using COX-1 and COX-2 selective agents. Twenty-seven healthy participants were injected with [11C]MC1. Ten of these participants received two PET scans: a baseline study followed by blockade with celecoxib (600mg PO), a preferential COX-2 inhibitor. Seventeen participants underwent test-retest imaging. All scans included concurrent arterial sampling. The tissue to plasma ratio at equilibrium (i.e., total distribution volume VT) was determined using a two-tissue compartment model.
Results: In transgenic hCOX-2 mice, 70-90% of [11C]MC1 brain uptake was blocked by nonradioactive MC1 and celecoxib (a COX-2 selective inhibitor) but not by PS13 (a COX-1 selective inhibitor), thereby confirming specific binding to human COX-2. Radioactivity in human brain peaked at a concentration of about 4.0 standardized uptake value (SUV), indicating good passage through the blood-brain barrier. VT values achieved stability after 80 minutes, indicating no radiometabolite contamination. Celecoxib reduced radioligand binding in neocortical areas by 25% but had little or no effect in subcortical regions and the cerebellum, which correlated with COX-2 mRNA expression levels. Binding site occupancy by celecoxib was virtually complete by the Lassen plots. Test-retest reliability was moderate (intraclass correlation coefficient (ICC)=0.65) but had relatively large variability (retest variability (absolute)=20%). Reference tissue methods yielded comparable results to the two-tissue compartment model but reduced the retest variability by up to 75% and reduced intersubject variability (coefficient of variation) by about half. Thus, compared to the two-tissue compartment model that requires arterial blood, the reference tissue method is expected to significantly reduce the sample sizes required to detect statistically significant differences between groups.
Conclusion: [11C]MC1 has adequate sensitivity to measure the low density of COX-2 in healthy human brain, suggesting it can also quantify COX-2 elevations expected in human disorders associated with neuroinflammation.