Background: Numerous knockout mouse studies have revealed that P-glycoprotein (P-gp) significantly limits drug distribution across the mouse blood-brain barrier (BBB). To determine the importance of P-gp at the human BBB, we developed a state-of-the-art, noninvasive, quantitative imaging technique to measure P-gp activity by use of carbon 11-labeled verapamil as the P-gp substrate and cyclosporine (INN, ciclosporin) as the P-gp inhibitor.
Methods: In brief, 11C-verapamil (approximately 0.2 mCi/kg) was administered to healthy volunteers (n = 12 [6 women and 6 men]) as an intravenous infusion over a period of approximately 1 minute before and after at least a 1-hour infusion of cyclosporine (2.5 mg x kg(-1) x h(-1)). Arterial blood samples and brain positron emission tomography images were obtained at frequent intervals for 45 minutes. Both blood and plasma radioactivity contents were determined in each verapamil sample. The content of verapamil and its metabolites in the 20- and 45-minute plasma samples was determined by a rapid solid-phase extraction method. The brain uptake of 11C-radioactivity (brain area under the curve [AUCbrain ]/blood area under the curve [AUCblood]) was determined in the presence and absence of cyclosporine.
Results: The AUCbrain/AUCblood ratio of 11C-radioactivity was increased by 88% +/- 20% (1.02 +/- 0.18 versus 0.55 +/- 0.10, P < .001) in the presence of cyclosporine (mean blood concentration, 2.8 +/- 0.4 micromol/L) without affecting 11C-verapamil metabolism or plasma protein binding. The corresponding increases for the brain white and gray matter were 84% +/- 13% and 84% +/- 18%, respectively.
Conclusions: This is the first time that P-gp activity at the human BBB has been measured. The modest inhibition of human BBB P-gp by cyclosporine has implications for P-gp-based drug interactions at the human BBB. Our method for imaging P-gp activity can be used to identify multidrug-resistant tumors or to determine the contribution of P-gp polymorphism, inhibition, or induction to interindividual variability in drug response.