Abstract
Aim
To investigate the occupancy at brain 5-hydroxytryptamine (5-HT) 1B receptors in human subjects after administration of the antimigraine drug zolmitriptan.
Methods
Positron emission tomography (PET) studies were undertaken using the radioligand [11C]AZ10419369 in eight control subjects at baseline and after administration of zolmitriptan orodispersible tablets. The subjects were examined after two consecutive administrations of 10 mg zolmitriptan, approximately 1 week apart. Two of the subjects were subsequently examined after administration of 5 mg zolmitriptan. One week after the last administration of zolmitriptan five of the subjects underwent additional PET measurements without drug pretreatment.
Results
After administration of 10 mg zolmitriptan, mean receptor occupancy was 4–5%. No consistent changes in 5-HT1B receptor binding were observed for subjects who received 5 mg zolmitriptan. There was a statistically significant negative relationship between binding potential (BPND) and plasma concentration of zolmitriptan and the active metabolite 183C91, respectively. All of the five subjects who were examined 1 week after dosing with zolmitriptan showed higher BPND post drug administration compared with baseline.
Conclusion
This is the first demonstration of CNS 5-HT1B receptor occupancy of a triptan. The findings are consistent with the low receptor occupancy previously reported in PET studies with agonists at other G protein coupled receptors.
Introduction
Selective 5-hydroxytryptamine (5-HT) 1B/1D receptor agonists (triptans) have revolutionized the acute treatment of migraine, providing patients with an effective migraine-specific therapy. Several mechanisms have been proposed to account for the antimigraine efficacy of triptans, including constriction of intracranial blood vessels, and central and peripheral inhibitory effects on trigeminal nerve activity (1).
Evidence from neuroimaging studies in migraine patients suggests the involvement of central neuronal processes in the pathophysiology of migraine (2). Thus increases in brainstem cerebral blood flow have been consistently shown in positron emission tomography (PET) studies during spontaneous migraine attacks (3,4). In addition, as assessed using functional magnetic resonance imaging, blood oxygenation level-dependent signal changes in the occipital cortex have been found during visual aura (5) and visually triggered headache (6) in patients with migraine.
The emerging evidence for a neurogenic basis of migraine raises a question on the central mechanisms of actions of triptans. The presence of triptan binding sites has been demonstrated in brainstem regions proposed to be involved in the pathophysiology of migraine (7). A chronically low 5-HT state interictally, that facilitates activation of the trigeminovascular nociceptive pathway, and increase in 5-HT release during attacks, has been suggested in migraine, where triptans may act directly on serotonin receptors to modulate brain 5-HT levels (8).
Zolmitriptan is a potent and selective 5-HT1B/1D receptor agonist that has been shown to inhibit activation of the trigeminovascular system both peripherally and centrally (9). Preclinical studies have shown, that zolmitriptan: 1. has central binding sites as demonstrated by ex vivo autoradiography in cats using [3H]zolmitriptan (10); 2. can inhibit activity in the trigeminal nucleus caudalis after intravenous administration (11); 3. has antiaggressive effects, consistent with activation of central 5-HT1B receptors (12). Clinical studies have shown that zolmitriptan modifies the intensity dependence of cortical auditory evoked potentials (13).
PET studies after [11C]zolmitriptan injection and nasal administration of zolmitriptan in human subjects (14) indicate that zolmitriptan enters the brain and at therapeutic dose (a 5 mg nasal spray) reaches brain concentrations of 0.5 ng/ml at 30 min and 1.5 ng/ml at 2 h post-intake. A subsequent autoradiography study with [11C]zolmitriptan showed a heterogeneous binding pattern in the brain of rhesus monkey (15) consistent with the previously reported 5-HT1B receptor distribution in the human brain (16).
Altogether, the available data suggest that the brain serotonin system is involved in the neurobiology of migraine, and that zolmitriptan may enter the central nervous system (CNS) and modulate brain 5-HT function. However, to our knowledge the extent of binding to CNS 5-HT1B receptors at clinical doses of zolmitriptan has not been previously characterized.
The aim of the present investigation was to study the occupancy at brain 5-HT1B receptors in human control subjects after administration of therapeutic doses of zolmitriptan. Binding to brain 5-HT1B receptors was investigated using PET and the radioligand [11C]AZ10419369, at baseline and after administration of single doses of zolmitriptan. In addition, the relationship between plasma concentration of zolmitriptan and the active metabolite 183C91, respectively, and binding to 5-HT1B receptors was analyzed.
Materials and methods
Radiochemistry
[11C]AZ10419369 was prepared at Karolinska Institutet by N-methylation of the desmethyl precursor (8-(1-piperazinyl)-5-methylchrom-2-en-4-one-2-(4-morpholinophenyl) carboxamide, obtained from AstraZeneca R&D, Wilmington, DE, USA), using carbon-11 methyl triflate, as described previously (17).
Subjects and study design
The study was approved by the Research Ethics Committee in Uppsala, Sweden, Radiation Safety Committee at Karolinska University Hospital, Stockholm and the Medical Products Agency of Sweden, and was performed in accordance with the current amendment of the Declaration of Helsinki and International Conference on Harmonization/Good Clinical Practice guidelines. Written informed consent was obtained from all subjects prior to the initiation of the study. The study was conducted as a research collaboration between AstraZeneca, Wilmington, DE, USA and Södertälje, Sweden and the Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden.
The investigation was an open, non-randomized, repeated dose, single-centre, exploratory study conducted between May 2010 and October 2010 in male control subjects. Enrollment procedures were carried out at the Quintiles Clinical Pharmacology Unit, Uppsala, Sweden, and subsequent magnetic resonance imaging and PET examinations were performed at Karolinska University Hospital, Solna, Sweden.
Eight men aged 20–29 years were enrolled as recruited for the main part of the study from a Quintiles (Uppsala Clinical Pharmacology Unit) database of subjects who expressed an interest in participating in clinical studies. The subjects were healthy according to medical history, including psychiatric symptoms and physical examination, ECG, routine blood tests, negative cotinine test and urine drug screen. The use of nicotine products was forbidden and caffeine intake was restricted to a maximum of five cups per day. No anatomical brain abnormalities were detected by magnetic resonance imaging.
The subjects underwent [11C]AZ10419369 PET measurements at baseline and after administration of single doses of zolmitriptan orodispersible tablets (Zomig® Rapimelt). PET measurements were conducted 2–3 h after drug administration to target the time of maximum plasma concentration for zolmitriptan. All subjects underwent PET measurements after two consecutive administrations of 10 mg zolmitriptan, approximately 1 week apart. Two of the subjects were subsequently examined after administration of a 5 mg zolmitriptan dose. One week after the last single dose administration of zolmitriptan, five of the subjects underwent a second PET measurement without drug pretreatment.
Magnetic resonance imaging
Brain magnetic resonance imaging was performed in a 3 Tesla General Electric Discovery MR750 (GE, Milwaukee, WI, USA) system. T2-weighted images were acquired for clinical evaluation regarding pathology, and T1-weighted images were acquired for co-registration with PET and delineation of anatomical brain regions. The T1 sequence was a 3D FSPGR BRAVO protocol in the axial plane with the following settings: TR 8.132 ms, TE 3.18 ms, flip angle 12°, FOV 256 mm, matrix 256 × 256×176, slice thickness 1.0 mm, NEX = 1.
Positron emission tomography procedures
Individualized plaster helmets were made for each subject and used with a head fixation system to minimize movement artifacts and to reproduce the head positioning in each examination (18). The subject was placed recumbent with his head in the PET system. A cannula was inserted into the left or right cubital vein. A sterile physiologic phosphate buffer (pH 7.4) solution containing [11C]AZ10419369 was injected as a bolus during 3 s into the cubital vein. The cannula was then immediately flushed with 10 ml saline.
PET measurements were conducted using the High Resolution Research Tomograph (Siemens Molecular Imaging, Knoxville, TN, USA). Listmode data were reconstructed using the ordinary Poisson-3D-ordered subset expectation maximization algorithm, with 10 iterations and 16 subsets including modeling of the point spread function. The corresponding in-plane resolution with ordinary Poisson-3D-ordered subset expectation maximization point spread function was 1.5 mm in the center of the field of view and 2.4 mm at 10-cm off-center directions (19). Attenuation correction was acquired with a 6 min transmission measurement using a single 137Cs source. Brain radioactivity was measured in a consecutive series of time frames for up to 63 min. The frame sequence consisted of eight 10 s frames, five 20 s frames, four 30 s frames, four 1 min frames, four 3 min frames and seven 6 min frames.
The first two PET examinations (baseline and first measurement after administration of 10 mg zolmitriptan) were undertaken on the same day, approximately 5 h apart. The subsequent PET examinations were separated by 4–16 days (longer than an estimated period of five half lives of zolmitriptan). The radioactivity injected ranged from 255 to 334 MBq. The specific radioactivities of the radioligand injected varied between 253 and 787 GBq/µmol, corresponding to an injected mass of 0.18–0.57 µg. As assessed by repeated measurements ANOVA, the injected mass of AZ10419369 did not vary at a statistically significant level between the study occasions.
Blood sampling and bioanalytical determination of zolmitriptan and 183C91
Venous blood samples (4 ml) for determination of the concentration of zolmitriptan and 183C91 in plasma were drawn at the start, middle and end of the PET examination. Samples for the measurements of concentrations of zolmitriptan and 183C91 were analyzed by PRA International Early Development Services, Assen, the Netherlands. The concentrations of zolmitriptan and 183C91 in human plasma were determined by solid phase extraction and liquid chromatography followed by tandem mass spectrometric detection (LC-MS/MS). The area under plasma concentration–time curve during the time period of the PET measurement (AUCPET) was calculated using the linear up/log down method, and the average plasma concentration during the time period of the PET measurement (Cavg) was calculated as AUCPET divided by the time period of the PET measurement.
Data analysis
The high resolution T1-weighted MR images were re-oriented according to the line defined by the anterior and posterior commissures, re-sampled and cropped to generate a 220 × 220 × 170 matrix with 1 mm3 voxels. Subsequently, MR images were co-registered to the PET summation images and segmented into gray and white matter using the SPM5 software (Wellcome Department of Cognitive Neurology, London, UK). The delineations of anatomical brain regions of interest (ROIs) were made manually on the MR images using Karolinska University Hospital PET centre in-house image analysis software, Human Brain Atlas. The ROI for the cerebral cortex was delineated on horizontal projections to include the entire cortical volume. The ROI for the occipital cortex was delineated in six sagittal slices in each hemisphere and the cerebellum ROI was delineated on horizontal projections of the six sections in the center of this region. The cortical ROIs were delineated by an initial manually defined crude region that included the gray matter as well as white matter and cerebrospinal fluid surrounding the ROI. The cortical gray matter was subsequently segmented by intersection of the initial cortical regions with the gray matter mask.
The ROIs were displayed on the corresponding PET images and the average radioactivity concentration for the whole volume of anatomical structure was obtained by pooling data from a series of sections. Further, the radioactivity concentration in each ROI, calculated for each sequential time frame and corrected for 11C decay was plotted versus time.
The BPND was estimated using the simplified reference tissue model (20) and the cerebellum as reference region. This method of analysis has been validated for the quantitative analysis of [11C]AZ10419369 binding (21).
Receptor occupancy was calculated from BPND values obtained at baseline (BPND, baseline) and after zolmitriptan administration (BPND, drug) according to the equation
For visualization of the regional binding to 5-HT1B receptors at baseline and after administration of zolmitriptan, parametric maps of [11C]AZ10419369 binding were created using the wavelet approach as described by Cselényi et al. (22).
Statistical analysis
The relationship between BPND values in the pooled cortical region (BPND, CX) and occipital cortex (BPND, OC), respectively, at baseline and after administration of 10 mg zolmitriptan, and plasma concentration of zolmitriptan and 183C91, respectively, was analyzed using a repeated measurement linear mixed effects model using compound symmetry covariance structure. Differences in BPND at baseline and 1 week post zolmitriptan administration were analyzed using the student’s t-test for dependent samples. A p value <0.05 was considered as statistically significant.
Results
After administration of 10 mg zolmitriptan there was a reduction in cortical BPND values consistent with low binding to brain 5-HT1B receptors (Table 1, Figure 1). At the first administration of 10 mg zolmitriptan the calculated mean receptor occupancies were 4.5% and 5.0% for the pooled cortical region and occipital cortex, respectively. The corresponding values for the second 10 mg zolmitriptan administration were 4.8% and 3.6%, respectively (Figure 2). For the two subjects who received 5 mg zolmitriptan, mean receptor occupancy varied between −1.8% in the cortical region and 1.5% in the occipital cortex.
Fused MRI and PET parametric images representing [11C]AZ10419369 binding potential (BPND) at baseline and after the first 10 mg zolmitriptan administration for one control subject (subject 7). Images were generated using the method described by Cselényi et al. (22). 5-HT1B receptor occupancy in the pooled region for the cerebral cortex and occipital cortex at two separate occasions after administration of 10 mg zolmitriptan. Horizontal lines correspond to mean values for each occasion. Binding potential (BPND) values for PET measurements at baseline, after administration of 10 mg zolmitriptan, and ca 1 week (4–11 days) after the last zolmitriptan administration.

Table 2 shows the plasma concentrations of zolmitriptan and the active metabolite 183C91. For both of the regions studied linear mixed effect model analysis for repeated measurements indicated a statistically significant negative relationship (p < 0.05) between BPND and plasma concentration of zolmitriptan and 183C91, respectively (Figure 3).
Relationship between binding potential (BPND) values and mean plasma concentration of zolmitriptan (left) and its active metabolite 183C91 (right) at time of PET measurement. Data are presented for subjects who underwent PET measurements at baseline and after two separate administrations of 10 mg zolmitriptan. Lines describe the relationship between BPND and plasma concentration of zolmitriptan and 183C91, respectively, estimated using a repeated measurement linear mixed effects model. The relationship between BPND for the cerebral cortex (BPND, CX) and occipital cortex (BPND, OC), respectively, and plasma concentration (Cavg) could be described by the following equations (parameter estimates with 95% confidence intervals in parenthesis): Mean plasma concentration of zolmitriptan and the active metabolite, 183C91, at time of PET measurement.
All of the five subjects who underwent a PET examination 1 week after dosing with zolmitriptan showed higher BPND at the post-drug administration measurement when compared with baseline (Table 1). The differences in BPND post-drug administration relative to baseline ranged from 2% to 32% for the cortical region and 0.8% to 15% for the occipital cortex, but did not reach statistical significance (p = 0.07 for both regions).
Zolmitriptan was well tolerated, adverse events judged as associated with intake of zolmitriptan were dizziness and paraesthesia. They were mild to moderate in intensity and resolved spontaneously.
Discussion
This study evaluated the occupancy at brain 5-HT1B receptors after oral administration of therapeutic doses of zolmitriptan using PET and the radioligand [11C]AZ10419369. To our knowledge this is the first examination of 5-HT1B receptor occupancy after administration of a triptan.
Occupancy at 5-HT1B receptors in human control subjects after administration of zolmitriptan at the therapeutically relevant 10 mg dose was 4–5%. A low receptor occupancy has been previously reported in PET studies with agonists at other G protein coupled receptors, including agonists for the 5-HT1A receptor (23,24) and the opioid receptor agonist methadone (25,26), and for allosteric modulators at the benzodiazepine site of the GABAA receptor (27,28). For agonist drugs the degree of occupancy required for therapeutic efficacy is dependent on the intrinsic activity of the ligand and receptor reserve at the target site with low occupancy required for high efficacy agonists (29). As zolmitriptan has been shown to display high intrinsic activity at human 5-HT1B receptors (30), low receptor occupancy may be sufficient to induce therapeutic effects.
Zolmitriptan has been shown to bind with high affinity to the human 5-HT1B and 5-HT1D receptors and could thus be expected to bind to both of these sites in brain (30). However, given the 10 - to 50-fold selectivity of the radioligand [11C]AZ10419369 for 5-HT1B vs. 5-HT1D receptors (31; AZ; data on file) and the low density of 5-HT1D receptors relative to that of 5-HT1B receptors in the human cerebral cortex (16), contribution of the latter subtype to the observed occupancy may be assumed to be negligible.
A consistent increase in BPND between baseline and at 1 week post zolmitriptan administration was an unexpected finding of the study. The reason for this observation is not known and several explanations are plausible. The post drug administration measurement was undertaken after administration of two doses of zolmitriptan, and approximately 1 week after the last dose. Thus, a drug effect on 5-HT1B receptor availability cannot be excluded. Zolmitriptan is a 5-HT1B receptor agonist, and as such could be expected to modulate receptor trafficking (29). Alternatively, this finding could possibly arise from methodological factors related to the repeated measurements design. In this respect, the observed differences in binding potential are unlikely to result from variation in specific radioactivity and injected mass of AZ10419369, as the mass administered was low (<0.6 µg) in all measurements, and did not differ at a statistically significant level between the experimental sessions. Another possible explanation could be the variation in endogenously released serotonin between test occasions, given that the radioligand [11C]AZ10419369 has been shown to be sensitive to the endogenous levels of serotonin (32). This interpretation is, however, not supported by recent test–retest studies with [11C]AZ10419369 showing no evidence for consistent changes in BPND after repeated measurements in human subjects (Nord et al., manuscript in preparation). Given the small number of subjects who underwent PET measurement at 1 week post zolmitriptan administration, future studies in larger groups of subjects are required to confirm this finding.
Accumulating evidence suggests that the brain serotonergic system may be involved in the pathophysiology of migraine (8). Observations of increased availability of brainstem serotonin transporter (33), and increased intensity-dependence in the amplitude of evoked potentials (34) are indicative of low levels of brain 5-HT in migraine patients interictally. In migraine patients, brain 5-HT synthetic rate has been shown to be higher during attacks and to decrease after administration of sumatriptan (35), consistent with an agonist activity at brain 5-HT1B receptors. The present finding that zolmitriptan occupies brain 5-HT1B receptors at therapeutic doses in human subjects, provides additional support for a central binding site of the triptans.
Conclusions
This is the first demonstration of CNS 5-HT1B receptor occupancy of a triptan. The findings are consistent with the low receptor occupancy previously reported in PET studies with agonists at other G protein coupled receptors. Based on the observed trend for increased 5-HT1B receptor availability observed at 1 week post drug administration, potential zolmitriptan-mediated effects on 5-HT1B receptor trafficking cannot be excluded.
Article highlights
PET studies were undertaken to investigate the occupancy at brain 5-HT1B receptors in human subjects after administration of the antimigraine drug zolmitriptan. After administration of 10 mg zolmitriptan mean receptor occupancy was 4–5%.A statistically significant negative relationship was found between binding to 5-HT1B receptors and plasma concentration of zolmitriptan and the active metabolite 183C91, respectively. This study provides the first demonstration of CNS 5-HT1B receptor occupancy of a triptan.
Footnotes
Acknowledgements
The authors thank all members of the PET group at Karolinska Institutet for the excellent technical assistance during the study, and Magnus Backheden, Department of Learning, Informatics, Management and Ethics at Karolinska Institutet for excellent support with statistical analyses.
Funding
The work was supported by AstraZeneca Pharmaceuticals.
Conflict of interest
Aurelija Jučaitė, Dennis J. McCarthy, Lars Farde and Stephen Kanes are employees of AstraZeneca Pharmaceuticals. Parts of the data have been presented at the 41st Annual Meeting of the Society for Neuroscience, Washington, DC, USA, 12–16 November 2011.
