Abstract
Nine healthy volunteers aged 18-28 years were recruited into this open, single-centre, two-phase trial. In phase 1, two volunteers received a single dose of 11C-zolmitriptan 2.5 mg administered as a nasal spray and then underwent positron emission tomography (PET) scanning to determine the most appropriate times for scanning in phase 2. In phase 2, six volunteers received two doses and an additional volunteer one dose of 11C-zolmitriptan 2.5 mg intranasally. Volunteers underwent PET scanning over sectors covering one of the nasopharynx, lungs or abdomen, for up to 1.5 h postdose. The brain was also scanned and plasma zolmitriptan levels were measured. Almost 100% of the administered dose was detected in the nasopharynx immediately after dosing. This declined thereafter to about 50% at 20 min and to 35% at 80 min after dosing. Radioactivity appeared slowly in the upper abdomen, with 25% of given radioactivity detected at 20 min and persisting until 80 min after dosing. Minimal radioactivity was detected in the lungs. Radioactivity was detectable within brain tissue suggesting central penetration of zolmitriptan. Zolmitriptan in plasma had approached its maximum concentration by 15 min postdose. The data indicate initial absorption across the nasal mucosa contributing to an early systemic availability. 11C-Zolmitriptan administered intranasally was well tolerated.
Introduction
Zolmitriptan is a 5-HT1B/1D receptor agonist that is highly effective in the acute treatment of migraine when administered by the oral route (1–3). Preclinical studies show that this agent inhibits the trigeminovascular system both peripherally and centrally, suggesting that zolmitriptan is able to cross the blood–brain barrier (4, 5).
As many patients experience gastrointestinal disturbances such as nausea, vomiting and gastric stasis during their migraine attacks, a swallowed oral tablet may not be ideal for all attacks. To meet the needs of such patients and those who have difficulty swallowing a tablet, a nasal spray formulation of zolmitriptan has been developed. Pharmacokinetic studies show that plasma concentrations of zolmitriptan are achieved earlier following intranasal administration than with the oral tablet, indicating that part of the dose is absorbed via the nasal mucosa (6, 7). Thus, zolmitriptan nasal spray may offer earlier relief of migraine symptoms than oral medication and make the benefits of zolmitriptan available to patients unable or unwilling to use oral treatments.
Positron emission tomography (PET) is an imaging and measurement modality which allows the generation of tomographic images, which with high resolution allows the visualization and quantification of radiolabelled drugs used in pharmacokinetic studies (8–10) or labelled tracers used to investigate certain biochemical systems such as receptors, transporter proteins or enzymes (11).
The primary objective of this study was to assess the distribution of zolmitriptan in the nasopharynx, lungs and abdomen using PET methodology after intranasal administration of zolmitriptan to healthy volunteers. The secondary objective was to obtain evidence of any distribution of zolmitriptan into the brain.
Methods
Volunteers
Nine healthy male and female volunteers aged 18–28 years were recruited. Volunteers were required to have normal clinical examination, medical history and resting electrocardiogram. Exclusion criteria were: any abnormality that may affect nasal absorption; acute illness within 2 weeks of the trial; history of clinically relevant disease; participation in a clinical trial within the previous 3 months; previous participation in a PET study; history of respiratory or cardiovascular disease; positive screen for HIV, hepatitis B or hepatitis C; known intolerance of zolmitriptan; concurrent medication except hormone replacement therapy or the oral contraceptive pill; pregnancy or lactation. All volunteers were required to provide written informed consent and the study was conducted according to Good Clinical Practice and the Declaration of Helsinki and following approval from the Research Ethics Committee, Local Radiation Ethics Committee and the Medical Products Agency.
Study design
During this open, two-phase trial performed at the Uppsala University PET Centre, Sweden, volunteers underwent a medical examination during the 14 days before the start of the trial and also a magnetic resonance imaging (MRI) scan to obtain an anatomical picture of the nasopharynx. Volunteers were also familiarized with the PET scanning equipment and the nasal spray device.
In the first phase of the study, two volunteers (one male, one female) received a single intranasal dose of 11C-zolmitriptan 2.5 mg (100 µl solution containing approximately 12 MBq radioactivity). In phase 2, a further six volunteers received two 2.5-mg intranasal doses of 11C-zolmitriptan separated by a washout period of between 3 days and 3 weeks and an additional volunteer received a similar single 2.5-mg dose. Dosing was performed with volunteers sitting up on the couch of the PET scanner. 11C-Zolmitriptan was administered to the least restricted nostril, after which the volunteer immediately lay down ready for scanning. The average administered radioactivity was 12 MBq (range 7–14).
Volunteers were required to fast from midnight before each study day, but were allowed free access to water until 2 h before dosing. Alcohol was not permitted from 72 h before each study day until 4 h after dosing and smoking was not permitted from midnight prior to the study day until 4 h after dosing.
Tracer synthesis
Zolmitriptan (S)-4-[[3-[2-(dimethylamino)ethyl]-1H-indol-5-yl]methyl]-2-oxazolidinone, was labelled with 11C according to a standard manufacturing procedure developed at Uppsala University PET Centre. Basically, 11C-carbon dioxide was produced by irradiation of nitrogen gas by 17 MeV protons from a Scanditronix MC-17 cyclotron. The 11C-carbon dioxide was converted to 11C-methyl iodide to be used in a methylation reaction of N-desmethyl-zolmitriptan to yield the 11C-labelled zolmitriptan. After preparative high-performance liquid chromatography (HPLC) and evaporation of the solvents, 5–10 ml of a solution containing zolmitriptan 25 mg/ml was added to the labelled zolmitriptan and the solution sterile filtered. To the sterile vial of the spray device, a well-defined volume of the formulated labelled drug was added. The spray device was assembled and transported to the PET-camera room. The total radioactivity of the spray device was measured before and after application, and administered radioactivity was calculated with due corrections for radioactive decay.
PET scans
The scanning was performed using a GE-4096 PET camera (General Electric Medical Systems, Milwaukee, WI, USA) with an axial field of view of 10 cm and a simultaneous acquisition of 15 tomographic slices (12). The image resolution was approximately 7 mm.
On each study day, before drug administration, a transmission scan was performed with an external rotating 68Ge-source to allow attenuation correction of measured data, necessary for the generation of quantitative data. The transmission scan lasted 10 min over each region.
In phase 1 of the study, the distribution of the 11C-radioactivity in the pharynx, lungs, abdomen and brain was determined by PET scanning for 1.5 h postdose. The imaging sequence for each volunteer was 5 + 5 frames of 1 min alternating between two sectors over the nasal passages and oral cavity, then two 5-min scans over the mid-lung, one 5-min scan over two sectors over the abdomen and two 5-min scans over the brain. This sequence was repeated over all sectors. The data collected were used to guide the timings of the PET scans during phase 2 of the study.
In phase 2, following each dose of 11C-zolmitriptan, each volunteer underwent PET scanning for 1.5 h postdose focusing on one of three regions (pharynx, lungs or abdomen) during each scanning period. For each volunteer, two regions were scanned (one on each study day), with the region to be scanned determined by a randomization scheme produced by the study centre. Each scan comprised 3 + 3 frames of 1 min (0–6 min postdose) and 2 + 2 frames of 5 min (7–27 min postdose). The pharynx and abdomen each consisted of two fields, with scanning alternating between nasopharynx/throat and the upper/lower abdomen, respectively. The lung consisted of a single field, with all scans performed over the central lung region. In addition, the distribution of 11C-radioactivity in the brain was assessed by two frames of 5 min conducted between 28 and 37 min postdose. Finally, six frames of 10 min were obtained from the same selected regions (nasopharynx, abdomen or lung) between 37 and 97 min postdose. Short times for bed movement were included in the total times.
Reconstruction of PET images was performed after collection of the raw data and correcting for attenuation, scattered radiation, randoms and deadtime. A 6-mm Hanning filter was used.
Pharmacokinetic assessments
Blood samples (7.5 ml) were taken for pharmacokinetic analysis via an intravenous cannula prior to and at 15, 30, 45, 60, 75, 90, 105 and 120 min after dosing. Samples were collected into lithium heparin tubes and centrifuged at 1000 r.p.m. for 10 min. Plasma samples were stored at − 20°C until analysis. Plasma concentrations of zolmitriptan and its active N-desmethyl metabolite, 183C91, were analysed using HPLC with tandem mass spectroscopy (LCMS/MS). Briefly, zolmitriptan and 183C91 were isolated using solid-phase extraction and the solid-phase extraction eluate was analysed by cation exchange liquid chromatography coupled with tandem mass spectroscopy with positive ion atmospheric pressure ionization using a turboionspray interface and multiple reaction monitoring. Duplicate quality control samples at three concentrations were included with each analytical run. The limit of quantification was 0.1 ng/ml for zolmitriptan and 0.2 ng/ml for 183C91 and the method was linear up to 15 ng/ml.
Radioactivity in whole blood was measured in a calibrated scintillation well counter for comparison with activity detected in the brain.
Tolerability
Adverse events and their relationship with study medication, as assessed by the investigator, were recorded throughout the study. Vital signs, blood pressure, 12-lead ECG recordings and routine laboratory safety assessments were performed during pre- and post-trial medical examinations.
Data analysis
In phase 2, a total of seven volunteers (three male, four female) were required to provide four full datasets for each of the three regions of interest. This had previously been shown to be sufficient to describe the distribution of inhaled 11C-nicotine in a study performed at the same centre (8).
The primary endpoint was the relative distribution of radioactivity to each of the three regions of interest, and changes to the distribution pattern during the 1.5-h monitoring period postdose. The secondary endpoint was the qualitative distribution of radioactivity in the brain over the same time period.
In the PET images, regions of interest were generated to represent the areas of interest based on the visualization of radioactivity distributions, utilizing the anatomical delineation as shown in the transmission scans. For the definition of total amount of radioactivity in the nasal cavity, throat, oesophagus stomach and intestines, the regions were drawn to safely include all visual activity by interactive changes of the window setting. For the local determinations including brain, lung and heart, the regions were made smaller than the anatomy to ensure representative values. Regions in adjacent slices were added together to calculate volumes of interest. The in-built routine in the image display analysis program calculated total radioactivity for the volumes of interest by outlining total radioactivity in a sector and applying average radioactivity concentration in local anatomical regions, with due correction for radioactive decay. Data on total radioactivity and local radioactivity concentration were collected as time–activity curves. For the nasopharynx, abdomen and lungs, data were expressed as a percentage of the dose administered, while for the brain and blood, data were expressed as standardized uptake values (SUV) by dividing with average administered radioactivity per body weight. The values for studies performed over the same sector were averaged over individuals.
Plasma concentrations of zolmitriptan and 183C91 were summarized as geometric means with coefficient of variation (CV).
Due to the small sample size, no formal statistical analysis of data was performed.
Results
A total of nine volunteers were recruited to this trial, received study medication and were included in subsequent analyses. Of these nine volunteers, five were female and four male. Their mean age was 23.4 years (range 18–28), mean height 174.2 cm (range 153–192) and mean weight 71.7 kg (range 53–101); all were caucasian.
There were no protocol violations leading to exclusion from PET scan evaluations. However, in one volunteer, cannulation problems prevented blood samples being taken during scanning of the abdomen. This deviation was felt to justify the exclusion of her data from the summaries and plots for that region of interest. An additional volunteer was recruited and her abdomen scanned after a single 2.5-mg intranasal dose of 11C-zolmitriptan so that four datasets were available for each region of interest.
11 C-zolmitriptan distribution
PET scanning data from the first two volunteers (phase 1) were purely explorative and are not presented. Assessment of distribution is based on the PET scans from phase 2.
The images obtained over the nasopharynx showed high amounts of radioactivity within the nasal cavity (Fig. 1). The anatomical location of this high concentration is readily observed in the images where PET has been superimposed on MRI images. In sectors over the lung and heart, radioactivity could be seen only in the oesophagus, while in lower sectors radioactivity was observed in the stomach and intestines (data not shown).
Axial (a) and coronal (b) images with radioactivity distribution in the nasopharynx as seen with positron emission tomography, overlapped on magnetic resonance images.
For all volunteers, close to 100% of the administered dose was detected in the nasopharynx immediately after intranasal administration (Fig. 2a). The concentration of 11C-zolmitriptan in the nasopharynx subsequently decreased with time as it was swallowed and entered the abdomen (Fig. 2a,b). Thus, 50% of the initial dose remained in the nasopharynx at 20 min postdose, declining to 35% of the initial dose by 80 min postdose. Considerable intersubject variability existed with respect to the rate of clearance from the nasopharynx.
Mean percentage of dose recovered in the nasopharynx (a) and the upper abdomen (b) following intranasal administration of 11C-zolmitriptan 2.5 mg. Error bars denote standard deviation, n = 4.
Radioactivity first appeared in the abdomen during the first 20–40 min postdose and total radioactivity concentrations in the upper abdomen at 20, 30 and 60 min postdose were, respectively, approximately 20%, 25% and 35% of the initial dose. Total radioactivity in the lower abdomen was generally lower than in the upper abdomen, with 15% and 5% of the initial dose observed at 25 and 80 min postdose, respectively. In three individuals, no radioactivity could be detected before 20 min.
Radioactivity concentrations in the lungs were very low in all four volunteers, with only 0.2% and 0.3% of the initial dose observed in the lungs at 20 and 80 min postdose, respectively.
Absolute radioactivity levels in the brain were low but were detectable for all volunteers on both study days with SUV ranging from 0.006 to 0.07 (recalculated to drug concentrations 0.16–1.77 ng/g). The mean brain concentration of radioactivity for all volunteers over both study days was SUV 0.02 (0.54 ng/g). The brain SUV and blood SUV were correlated with a correlation coefficient of 0.95 and brain values were 0.2 times the blood values.
Plasma concentrations of zolmitriptan and 183C91
At 15 min postdose, mean plasma concentrations of zolmitriptan were close to 2 ng/ml, at which time the majority of the administered dose was still within the nasopharynx. Plasma concentrations of zolmitriptan were maintained at values below 3 ng/ml throughout the remainder of the scanning period. Plasma concentrations of 183C91 were initially detected from 15 to 30 min onwards and reached a maximum of 0.7 ng/ml at 2 h.
Tolerability
Seven volunteers reported 11 adverse events, all of which were mild in intensity and did not necessitate study withdrawal. The most frequently reported adverse events were paraesthesia (n = 3) and pharyngitis (n = 3). No serious adverse events occurred and there were no clinically relevant changes in laboratory parameters.
Discussion
In the present study we investigated a spray formulation of zolmitriptan developed for intranasal administration. PET was utilized as a means of recording the site of deposition, disposition rate and transport along the oesophagus into the intestines. In this study, 100% of the radioactivity was recovered in the nasal cavity directly after administration, while only minimal radioactivity (max. 0.3% of dose) was detected in the lung scans. This eliminates the possibility that nasally administered zolmitriptan reaches the lungs and thereafter the systemic circulation. The nasal radioactivity started to decrease rapidly within the first 6 min for three of the volunteers, while for the fourth the decrease started after 20 min. This indicates a rapid disposition, but also large interindividual variations. Since the volunteer group is very small, it is not possible to estimate the variability accurately. The appearance of radioactivity in the abdomen was delayed until 20 or more minutes after dosing. Coupled with the fact that plasma drug concentration had reached close to maximum values by 15 min after administration, this suggests that absorption from the intestines cannot explain the rapid appearance in plasma of zolmitriptan. The most plausible explanation is that a substantial amount of zolmitriptan is absorbed directly through the nasal mucosa. However, due to the high variability and the fact that assessment of disappearance of zolmitriptan from the nasal mucosa and appearance in the gastrointestinal tract could not be made in the same individual, it is difficult to make a good estimate of the proportion of zolmitriptan that would enter the systemic circulation via the nasal mucosa.
Our findings are consistent with previous studies, in which zolmitriptan was shown to appear in the plasma as early as 5 min after intranasal administration, which is far more rapid than after oral administration (6, 7). The rapid appearance of zolmitriptan in plasma while the majority of the administered intranasal dose is still in the nasopharynx suggests direct absorption across the nasal mucosa.
Compared with oral dosing, intranasal administration of zolmitriptan has previously been shown to result in the delayed appearance of the active metabolite of zolmitriptan, 183C91 (6, 7), and this was also observed in the present study. Thus, 183C91 was not detected in the plasma until 30 min postdose. This lag time has been attributed to the initial avoidance of first-pass metabolism by absorption across the nasal mucosa (6, 7) until detectable concentrations of the metabolite are formed in the liver from the remaining fraction of the dose that was eventually swallowed.
Previous studies have demonstrated a central effect of zolmitriptan both in animals (4) and in man (13, 14), suggesting that zolmitriptan is able to cross the blood–brain barrier. Studies with sumatriptan have shown that sumatriptan does not cross the blood–brain barrier (15, 16). In the present study, drug-related radioactivity was indeed measurable in the brain, with SUV values of about 0.02 and brain : blood ratios of about 0.2. Since the cerebral blood volume is of the order of 3% of total brain volume, vascular contribution can only explain a minor fraction of this brain radioactivity, and it is fair to assume that drug-related material does indeed enter the brain parenchyma. However, caution must be taken since it is likely that, particularly at longer times after dosing, some of the radioactivity in plasma is attributable to metabolites of zolmitriptan and it cannot be excluded that some of these cross the blood–brain barrier. A separate PET study using the labelled metabolite could help in clarifying this.
PET scan results in the present study suggest that, although only low levels of radioactivity are detectable, zolmitriptan is absorbed and does indeed cross the blood–brain barrier following intranasal administration. Nevertheless, previous studies have shown that this does not result in an increase in the incidence of centrally related adverse events compared with sumatriptan (17, 18). Indeed, in agreement with previous observations, intranasal administration of zolmitriptan was well tolerated.
In summary, following intranasal administration, zolmitriptan was gradually cleared from the nasopharynx by swallowing, with only minimal quantities detected in the lungs. The rapid increase in plasma concentrations of zolmitriptan immediately after intranasal administration indicates direct absorption across the nasal mucosa. PET scans demonstrated penetration of 11C-zolmitriptan-derived radioactivity into the brain. Finally, zolmitriptan was well tolerated following intranasal administration.