The present study was approved by the Committee on Clinical Investigation at Tohoku University Graduate School of Medicine, Japan, and was performed in accordance with the principles of the Declaration of Helsinki. All experiments were performed at the Cyclotron and Radioisotope Centre, Tohoku University.
Subjects and study design
Eight male Japanese volunteers (mean age ± SD 24.4 ± 3.3 years), who provided written informed consent after receiving a detailed description of the study, were recruited. All subjects were in good health with no clinical history of major physical or mental illness, showed no abnormality in brain magnetic resonance imaging (MRI), and were not receiving any concomitant medication likely to interfere with the study results. Nicotine, caffeine, grapefruit and grapefruit juice were not allowed on the test day, and alcohol was not allowed on the test day or the day before PET measurement.
All subjects underwent PET measurement after single oral administration of bepotastine (10 mg), diphenhydramine (30 mg) or a lactobacteria preparation (6 mg) used as placebo in a three-way crossover study, with minimum wash-out intervals of 7 days between treatments. The lactobacteria preparation has been widely used as placebo in Japan, and its administration has produced no statistical difference between pre- and post-administration in previous cognitive studies [7, 9, 10, 21]. The present PET study was conducted in a single-blinded manner, and after drug administration each subject was asked to remain seated comfortably on a sofa. To determine bepotastine and diphenhydramine plasma concentrations, blood samples were collected from each subject before drug administration and at 0, 60, 120 and 180 min post administration. Subjective sleepiness of each subject was also measured at 0, 60, 120 and 180 min post administration using the Line Analogue Rating Scale (LARS)  and the Stanford Sleepiness Scale (SSS) as used in previous studies [9, 23].
Measurement of plasma concentrations of bepotastine and diphenhydramine
Plasma concentrations of bepotastine and diphenhydramine were measured using liquid chromatography/tandem mass spectrometry (LC/MS/MS) together with an electrospray ionization method . LC was performed on an Agilent 1100 Series LC instrument (Agilent Technologies Inc., Santa Clara, CA, USA) equipped with an analytical column. The MS/MS system was the API 4000 (Applied Biosystems/MDS Sciex, ON, Canada). The Solid Phase Extraction (SPE) cartridge (OASIS HLB 3 ml/60 mg; Waters Corp., Milford, MA, USA) was pretreated with 2 ml of methanol, 2 ml of water and 2 ml of 0.2 m Na2CO3/HCl buffer (pH 11).
For measurement, an internal standard solution (10 μl) and methanol (10 μl) were added to each plasma sample (50 μl). To the resulting solution, 930 ml of 0.2 m Na2CO3/HCl buffer was added for bepotastine measurement, and 1000 μl of 0.1% formic acid containing acetonitrile/methanol (50 : 50, v/v) was added for diphenhydramine measurement. The mixture was applied onto the SPE cartridge after pretreatment as mentioned above. Separations were carried out on a high-performance liquid chromatography (HPLC) column [CAPCELL PAK C18 MG II (3 μm) 2.0 mmφ × 100 mm; Shiseido Co., Ltd, Tokyo, Japan) at a flow rate of 0.2 ml min−1 and at a column temperature of 40 °C. The reconstituted extract (5 μl) was injected onto an HPLC system with mobile phases for bepotastine measurement including 10 mmol l−1 ammonium acetate and acetonitrile of varied concentrations, namely, 32% (0–9 min), 70% (9.5–12.5 min) and 32% (12.6–24 min), and with mobile phases for diphenhydramine measurement including 0.1% heptafluorobutyric acid and acetonitrile of varied concentrations, namely, 40% (0–7 min), 70% (7.5–10.5 min) and 40% (11–21 min).
Detection of bepotastine was based on fragmentation of the precursor ion (m/z = 389 to product ion m/z = 202 with collision energy of 29 eV for bepotastine, and m/z = 256 to product ion m/z = 167 with collision energy of 19 eV for diphenhydramine), and that of the internal standard was based on fragmentation of the precursor ion (m/z = 389 to product ion m/z = 201 with collision energy of 29 eV for bepotastine, and m/z = 270 to product ion m/z = 181 with collision energy of 17 eV for diphenhydramine) in positive multiple reaction monitoring (MRM) mode. Positive ions were detected using an API 4000 system at 550 °C nebulizer gas temperature, with 5000 V ion spray voltage, 68.9 kPa (nitrogen) curtain gas and Level 4 collision gas for bepotastine, and 206.8 kPa curtain gas and Level 4 collision gas for diphenhydramine. Chromatographic data for positive MRM were collected using Analyst software (ver. 1.2, Applied Biosystems/MDS Sciex) with cycle times of 1.010 s per cycle for bepotastine and 0.5200 s per cycle for diphenhydramine. The lowest detectable concentration was around 1 ng ml−1 for both antihistamines, and some values slightly under the threshold (only for diphenhydramine) were extrapolated. As for validation, the following items were checked for bepotastine and diphenhydramine, respectively: accuracy (100.7% and 102.2%), correlation coefficients to standard solutions (r > 0.99 for both), and coefficients of variation (CVs) of three different concentrations (n = 5) (1.7–2.3%, 1.8–3.8%).
For examination of the relationship between estimated binding potential ratio (BPR) of 11C-doxepin and plasma concentration of each antihistamine, the areas under the curves (AUCs) of bepotastine and diphenhydramine were calculated for 0–180 min (AUC0−3 h) post administration (Table 1).
Table 1. Plasma concentrations of bepotastine and diphenhydramine (n = 8)
|Bepotastine time (min)||Mean (ng ml−1)||SEM||CV, %||Diphenhydramine time (min)||Mean (ng ml−1)||SEM||CV, %|
|AUC0−3 h||196.1||24.3||35.0||AUC0−3 h||32.3||4.0||35.3|
PET tracer and image acquisition
Doxepin is one of the tricyclic antidepressants that has binding affinity to other receptors such as muscarinic receptors to some extent. However, its affinity to histamine H1Rs is much higher than to other receptors and is very high compared with other antidepressants . Thus, doxepin's affinity to other receptor systems is negligible in this imaging study, as also confirmed by experiments using histamine H1R knock-out mice, where doxepin binding in the brain was nearly zero . Thereafter, 11C-doxepin has been used to evaluate the distribution of histamine H1Rs. 11C-doxepin kinetics in plasma and the brain are not affected by the sedative antihistamine d-chlorpheniramine using arterial sampling data combined with metabolite analysis .
In the present study, 11C-doxepin was prepared by 11C-methylation of desmethyl doxepin with 11C-methyl triflate as described previously [10, 27]. 11C-doxepin radiochemical purity was >99%, and its specific radioactivity at the time of injection was 120.9 ± 80.55 GBq μmol−1 (3268 ± 2177 mCi μmol−1). 11C-doxepin-containing saline solution was intravenously injected into each subject at 90 min post administration, a time close to Tmax of both antihistamines (1.2 h for bepotastine and 2–3 h for diphenhydramine). The injected dose and cold mass of 11C-doxepin were 135.4 ± 19.83 MBq (3.660 ± 0.536 mCi) and 1.587 ± 0.895 nmol, respectively, and the radiological dose was calculated based on a previous study on radiological exposure .
Approximately 60 min after 11C-doxepin injection, the subjects were positioned on the coach of the PET scanner (SET2400W; Shimadzu Co., Kyoto, Japan) for transmission scan (6 min) and emission scan in the three-dimensional (3D) mode lasting for 15 min (70–85 min post injection of 11C-doxepin) in a similar fashion to our previous work [10, 29]. PET brain images from a 15-min-long emission scan were corrected for scattering based on a previous study  and for tissue attenuation using post-injection transmission scan data according to previous work . Brain images were reconstructed with a filtered back-projection algorithm, with the aid of a supercomputer SX-7 at the Information Synergy Centre, Tohoku University, Sendai, Japan. The brain images were then normalized by plasma radioactivity at 10 min post injection to yield images reflecting distribution volume (DV) based on our static scan protocol reported previously [10, 32]. Validation using venous sampling instead of arterial sampling was carried out by another group, giving no difference between venous and arterial sampling at 10 min post injection (M. Senda, personal communication, 23 August 2007).
Three brain images of each subject, following oral administration of bepotastine, diphenhydramine and placebo, were coregistered to an identical stereotaxic brain coordinate system using an MRI-T1 image of each subject, with the aid of Statistical Parametric Mapping (SPM2, Wellcome Department, UK) software package . MRI images were obtained with a 1.5-T magnetic resonance (MR) scanner (HiSpeed, Ver. 9.1; General Electric Inc., WI, USA) at Sendai Seiryo Clinic (miyagi, Japan). T1-weighted images (Vascular TOF SPGR: TR/TE 50/2.4 ms, FA 45°, number of excitations 1, matrix size 256 × 256, spatial resolution: x, y, z = 0.86, 0.86, 20.0 mm, respectively) were collected from all subjects.
Regions of interest (ROIs) were first placed on the following brain regions on the T1 images that had precise anatomical information, i.e. anterior and posterior cingulate gyri (ACG and PCG, respectively), prefrontal cortices (PFC), orbitofrontal cortex (OFC), insular cortex (IC), temporal cortex (TC), parietal cortex (PC), occipital cortex (OC), primary sensorimotor cortex (SMC), thalamus, striatum, midbrain, pons, and cerebellum. ROI was defined for each cortical region by two to five circles with a diameter of 7.6 mm for each hemisphere in four to five consecutive brain transaxial slices, as indicated in Figure 2A. For the thalamus, striatum, pons and midbrain, the margin of each region was traced in MRI T1 images. An averaged value from all ROIs was used as a representative value of each region. Information on ROI location was automatically transferred to the coregistered three PET images reflecting DV, and the binding potential ratio (BPR) was calculated for each region using the following equation: BPR = [(DV of each region − DV of cerebellum)/DV of cerebellum][8, 9]. Finally, H1ROs of bepotastine and diphenhydramine were calculated for each cortical region using the following equation: H1RO = [(BPR of placebo − BPR of antihistamine)/BPR of placebo] × 100. BPR brain images were also created by applying the same equation to each DV brain image [8–10, 34, 35]. BPR brain images were analysed statistically on a voxel-by-voxel basis using SPM2 , following spatial normalization and smoothing using the same method as in our previous work. Differences in parameter values between bepotastine, diphenhydramine and placebo (control) were statistically examined, and regional maxima of statistical significance (P < 0.001) were projected onto surface-rendered MRI-T1 standard brain images. Precise locations of statistically significant regions were identified with the Co-Planar Stereotaxic Atlas .
Figure 2. Binding potential ratio (BPR) images of 11C-doxepin in the human brain (A) and results of voxel-by-voxel comparison (B). BPR of 11C-doxepin was calculated in healthy male subjects (n = 8) by positron emission tomography following oral administrations of placebo (left), bepotastine (10 mg, middle) or diphenhydramine (30 mg, right), and their magnetic resonance imaging-T1 images (far right), demonstrated in the transaxial (top), coronal (middle) and sagittal (bottom) sections for each treatment condition, were compared (A). White circles in the transaxial images indicate the regions of interest (ROIs). The brain image of each subject was transformed to fit stereotaxic brain space (spatial normalization) and was averaged across each drug condition to generate the mean images displayed (A). The images demonstrate that diphenhydramine treatment results in BPR significantly lower than those of other drug conditions (B). Height threshold of voxel values was set at P < 0.001 and extent threshold was set at 10 voxel minimum. Results were not corrected for multiple comparisons. There are no areas with significantly lower BPR after bepotastine treatment compared with those after placebo treatment (‘bepotastine 10 mg < placebo’ in the left columns). In contrast, red colour shows areas with significantly lower BPR after diphenhydramine treatment compared with those after placebo treatment (‘diphenhydramine 30 mg < placebo’ in the right columns). In both columns, significant areas are demonstrated in four aspects, namely, left and right medial (L. MED and R. MED) and right and left lateral (R. LAT and L. LAT) aspects (P < 0.001, uncorrected, using SPM2) (B)
Download figure to PowerPoint