Functional P-glycoprotein in the blood-brain barrier of the post status epilepticus rat: Measurements using (R)-[C-11]verapamil and PET

Background: About one third of epilepsy patients cannot be treated satisfactorily with standard medication, illustrating that pharmacoresistance is a major problem in the treatment of epilepsy. In vitro studies have shown that the multidrug transporter P-glycoprotein (P-gp) is overexpressed in the endothelial cells of the blood-brain barrier (BBB). It has been hypothesized that this overexpression of P-gp could be an important mechanism contributing to pharmacoresistance in epilepsy patients. As many anti-epileptic drugs are substrates for P-gp, overexpression of P-gp could result in brain concentrations of these drugs that are too low to be effective. Aim: The purpose of the present study was to investigate in vivo whether functional P-gp in the BBB was increased in an animal model of epilepsy. Methods: Spontaneous epilepsy was achieved in three Sprague Dawley rats by induction of Status Epilepticus (SE) using i.v. injection of kainic acid. Three months after SE induction, presence of spontaneous epileptic activity was confirmed by EEG measurements. (R)-[C-11]verapamil was synthesized according to a standard procedure as described previously. Specific activity of (R)-[C-11]verapamil ranged from 10 to 40 GBq/μmol, radiochemical purity was > 98%. After a 6 min transmission scan, anesthetized (5% isoflurane) rats were injected with 20 MBq (R)-[C-11]verapamil via the tail vein. A 60 min list mode PET scan was acquired and histogrammed into 20 frames of increasing duration. Data were corrected for randoms, scatter and dead time and reconstructed using attenuation and normalisation weighted OSEM. Arterial blood samples were taken at set times to estimate plasma to whole blood ratios and metabolite fractions, used to convert the left ventricle time-activity curve into a plasma input function. Whole brain Vd over a time interval of 30 to 60 minutes was obtained using Logan analysis. Results: Whole brain Vd 30-60 minutes was in the SE rats was 0.41 ± 0.07 and in control group 0.60 ± 0.11 (p =0.034, two tailed Mann-Whitney U test). Metabolite fractions of (R)-[C-11]verapamil in plasma were different in the SE rats compared with the control group. However, these differences in plasma concentration of the tracer were taken into account in the analysis and therefore the differences in Vd could not be explained by the differences in plasma kinetics. Conclusion: A significant decrease in Vd of (R)-[C-11]verapamil of around 30% was measured in the SE rat relative to control. This indicates increased P-gp function at the BBB, which in turn might explain increased brain elimination of anti-epileptic drugs. Using similar studies in patients, this hypothesis now needs to be tested in patients with and without pharmacoresistant epilepsy.