Abstract
Background
Pituitary adenylate cyclase-activating polypeptide-38 (PACAP38) and vasoactive intestinal polypeptide can provoke cluster headache attacks in up to half of cluster headache patients in their active phase. At present, it is unknown whether provoked attacks are mediated via calcitonin gene-related peptide or mast cell activation.
Methods
All enrolled patients with cluster headache were randomly allocated to receive a continuous infusion of either PACAP38 (10 pmol/kg/min) or vasoactive intestinal polypeptide (8 pmol/kg/min) over 20 min. We collected clinical data and measured plasma levels of calcitonin gene-related peptide and markers of mast cell activation (tryptase and histamine) at fixed time points: at baseline (T0), at the end of the infusion (T20), 10 min after the infusion (T30), and 70 min after the infusion (T90).
Results
Blood was collected from episodic cluster headache patients in active phase (n = 14), episodic cluster headache patients in remission (n = 15), and chronic cluster headache patients (n = 15). At baseline, plasma levels of calcitonin gene-related peptide, tryptase, and histamine were not different among the three study groups. Plasma levels of calcitonin gene-related peptide (p = 0.7074), tryptase (p = 0.6673), or histamine (p = 0.4792) remained unchanged during provoked attacks compared to attack-free patients.
Conclusion
Cluster headache attacks provoked by either PACAP38 or vasoactive intestinal polypeptide were not accompanied by alterations of plasma calcitonin gene-related peptide, tryptase or histamine. The provoked attacks may not be mediated by calcitonin gene-related peptide or mast cell activation.
Introduction
The hallmarks of cluster headache (CH) include prominent autonomic symptoms ipsilateral to severe headache (1). The anatomical and physiological substrates of CH attacks include the trigeminovascular system (TVS) (2). Upon activation of the TVS, the first division of the trigeminal nerve triggers the release of several parasympathetic neuropeptides, including pituitary adenylate cyclase activating polypeptide-38 (PACAP38) and vasoactive intestinal polypeptide (VIP) (3). In active episodic CH (eCH), PACAP38 and VIP induced attacks in 43% and 36% of patients, respectively, while they induced attacks in 47% of patients with chronic CH (cCH) (4). Mechanisms underlying the initiation of PACAP38- and VIP-induced CH attacks remain unknown.
Potential pathways involved in CH attacks may include the release of calcitonin gene-related peptide (CGRP) from the trigeminal sensory fibres, and mast cell activation. CGRP provokes CH attacks (5), while galcanezumab, a monoclonal antibody neutralising the CGRP peptide, reduces the frequency of attacks in eCH (6). Histamine, a biologically active molecule released by mast cells with inflammatory mediators, provokes CH attacks (7). Mast cells are in a close spatial relationship to nociceptive fibers (8,9) and they release different mediators upon stimulation by neurotransmitters (10,11). Of interest, mast cells have been described more numerous in the painful area of patients with CH, in close proximity to trigeminal nerve endings (12–14). To date, the involvement of CGRP and/or mast cells in CH attacks induced by either PACAP38 or VIP has never been investigated.
We previously described the ability of PACAP38 and VIP to induce CH attacks (4). In the present study, we investigated the contribution of several biochemical variables in experimentally induced CH attacks. We hypothesised that PACAP38- and VIP-induced CH attacks are associated with alterations of plasma levels of CGRP and markers of mast cell activation (tryptase and histamine).
Material and methods
The participants were recruited between December 2017 and August 2019, from the Danish Headache Center (Rigshospitalet, Glostrup). They were aged 18–65 years and were eligible for inclusion with a verified diagnosis of eCH or cCH, according to the International Classification of Headache Disorders, 3rd edition (beta version) (15). eCH patients were defined as being in active phase with the occurrence of typical CH attacks within the last 30 days, or in remission when attack free for at least 30 days. cCH patients had attack-free periods not exceeding 30 days in the last 12 months. Exclusion criteria were any other type of headache (except tension-type headache <5 days per month and/or migraine if >12 months since last attack), pregnancy or nursing women, any serious somatic or psychiatric condition, and drug misuse. All CH preventive medications were allowed with stable dosing, except steroids within 30 days.
The present study is part of a larger parent protocol (H-17011659) approved from the Regional Health Research Ethics Committee of the Capital Region and conducted in accordance with the Helsinki Declaration of 1964, with later revisions. The parent protocol was registered at clinicaltrials.gov (NCT03814226) and approved by the Danish Data Protection Agency. The current data were collected during the previously reported study (4). In the previous study we reported the ability of CGRP and VIP to induce CH attacks. In the present study, we explored the contribution of several biochemical variables to provoke CH attacks.
Experimental design
The study was conducted as a randomised, double-blind, cross-over trial. All participants were randomly allocated to receive a continuous infusion of either PACAP38 (10 pmol/kg/min) or VIP (8 pmol/kg/min) over 20 min, using a time- and volume-controlled infusion pump (Braun Perfusor, Melsungen, Germany). Infusions were performed on two study days, separated by at least 7 days. Before the first study day, a full medical examination was conducted for all patients. On both study days, a pregnancy test was performed in women of childbearing age. Active eCH or cCH patients were attack-free at least 4 h before the experimental procedures. eCH patients in remission were headache free at least 8 h before the experiments. Participants were placed in the supine position and a venous catheter (Venflon, Braun Melsungen, Germany) was inserted into the antecubital vein on the right or left arm for experimental infusions and drawing blood. A winged infusion set (SAFETY Blood Collection Set, Greiner-Bio, Kremsmünster, Austria) was inserted into the contralateral antecubital vein for an additional blood sample for histamine evaluation (T10), then immediately disposed of. Participants were at rest for 15 min before measuring baseline values. Blood for analysis of CGRP, histamine and tryptase was collected at fixed time points: At baseline (T0), at the end of the infusion (T20), 10 min after the infusion (T30), and 70 min after the infusion (T90).
Randomisation
PACAP38 and VIP were prepared in identical vials and randomised by the Regional Central Pharmacy. During the study, the randomisation code remained in the hospital and was unavailable to medical investigators until the study was completed and data were analysed. Allocation of PACAP38 and VIP was performed to ensure balanced group sizes.
Blood collection and processing
Before the sampling, 5 ml of blood were collected through the venous catheter and discarded. Blood was collected at fixed time points using a 20 ml syringe. The venous catheter was flushed with saline after the blood sampling. The blood was thereafter transferred into standard tubes containing tripotassium ethylenediaminetetraacetic acid (EDTA). Tubes were inverted several times and stored at room temperature for 20 min until centrifugation. The test tube for histamine was place on ice until centrifugation. Tubes were centrifuged at 1600 g for 10 min, then plasma was moved to cryotubes (Thermo Fisher Scientific, Jiangsu, China) and stored at −80°C until analysis.
Plasma CGRP
Plasma CGRP concentrations were determined by radioimmunoassay using antibody AB 4-2905 and α-CGRP as calibrator (16). The tracer was prepared by iodination of [Tyr0] α-CGRP (25–37) amide and purification by high liquid chromatography (HPLC). Samples and calibrators were incubated at 4°C for about 90 h prior to addition of tracer and subsequent incubation for 48 h. Free and antibody-bound tracer were separated by Sac-Cel separation.
Plasma histamine
The concentration of histamine in plasma was measured in technical duplicates using a commercially available enzyme-linked immunosorbent assay (ELISA) for in vitro diagnostics (cat no. IM2562, Immunotech, Beckman Coulter, Prague, Czech Republic) according to the manufacturer’s instructions.
Plasma tryptase
The concentration of tryptase in plasma was quantified using a fluorescence immunoassay (Phadia, Thermo Fisher Diagnostics, Uppsala, Sweden). Limit of detection was 1 ng/ml.
Statistical analysis
All values are presented as mean and range, or median and interquartiles (IQR). To test for distribution of baseline variables we used the Shapiro–Wilk test and group comparisons were subsequently analysed by nonparametric statistics. The primary endpoint were the differences of CGRP, histamine and tryptase between patients who developed an attack and patients who did not in active eCH and cCH groups. For all biochemical variables, we calculated the sum of the differences between the baseline measurement and the measurement at 10 min, 20 min, 30 min, and 90 min after PACAP-38 or VIP infusion. In this way, we obtained a summary score of the changes of CGRP, tryptase and histamine during each experimental day. Then, we compared the scores between patients who developed an attack, and patients who did not, by the Mann–Whitney U-test. The co-primary endpoint was the baseline difference of biochemical variables (CGRP, histamine and tryptase) between groups (active eCH patients, eCH patients in remission and cCH patients). Baseline values were compared using a generalised linear model with repeated measurements, as previously described (17). As post-hoc analysis, baseline values of CGRP from CH groups (active eCH patients, eCH patients in remission and cCH patients) from our study and a previous study (17) were pooled and compared using the generalised linear model. We used GraphPad Prism 9.0.2 and SAS Enterprise for statistical analysis. All p-values were two-sided and considered significant if <0.01.
Results
In total, 41 patients (34 men and seven women) completed the study (Figure 1). Three eCH patients participated in both the active and remission phases, thus resulting in data from 44 participants that were given both infusions. The mean age was 38 years (range, 18–60). Clinical characteristics of participants are shown in Table 1. Results of PACAP38 and VIP infusions on CH attack induction have been reported previously (4). Briefly, PACAP38 infusion induced a CH attack in 13 out of 44 participants, VIP induced a CH attack in 12 out of 44 participants. None of the participants reported attacks during remission. Characteristics of provoked attacks are listed in Table 2.
Flow chart of participant recruitment from Vollesen et al. (4).
Clinical characteristics of study participants.
Note: Age is expressed as years, body mass index (BMI) is expressed as kg/m2.
SD: standard deviation, W/M: women/men.
*Verapamil (between 100 and 800 mg), lithium, and gabapentin.
**Verapamil (120 and 160 mg).
***Verapamil (between 120 and 800 mg), lithium, gabapentin (between 300 and 2700 mg), and melatonin.
Clinical characteristics of provoked attacks in episodic CH patients in active phase (eCHa) and chronic CH (cCH) patients.
Note: Time to onset and duration: Min, starting with experimental infusion. Peak headache intensity was measured from 0 to 10.
Sumatriptan: sumatriptan 6 mg sc. Con: nasal congestion. Ede: eyelid edema. Inj: conjunctival injection. Lac: lacrimation. Mio: miosis. Pto: ptosis. Rhi: rhinorrhea. Swe: forehead and facial sweating.
Experimental days during which plasma samples of CGRP, tryptase and histamine were successfully collected and measured at all time points are described in Table 3. The concentration of CGRP, histamine, and tryptase were within the detection limits in all samples. Due to technical problems (poor/inaccessible peripheral venous accesses and laboratory technical issues), there were missing values in 285 out of 1144 planned samples (24.9%). Baseline CGRP values of a single cCH patient were considered outliers (1304 and 1080 pmol/l) and therefore excluded from statistical analysis.
Experimental days during which plasma samples of CGRP, tryptase and histamine were successfully measured at all time points.
Note: Blood for analysis of CGRP, histamine and tryptase was collected at fixed time points: At baseline (T0), at the end of the infusion (T20), 10 min after the infusion (T30), and 70 min after the infusion (T90). An additional sample was collected for histamine analysis at T10.
CGRP, tryptase and histamine during provoked attacks
We found no differences in plasma CGRP (21 attacks vs. 28 without attacks, p = 0.7074), tryptase (13 attacks vs. 29 without attacks, p = 0.6673), and histamine (15 attacks vs. 15 without attacks, p = 0.4792) between patients who developed CH attacks and those who did not develop attacks (Table 4 and Figure 2). In addition, plasma concentrations of CGRP (p = 0.4258), tryptase (p = 0.8836) and histamine (p = 0.8366) were not different among timepoints (T0, T10, T20, T30 and T90).
Mean percentage changes of CGRP, histamine and tryptase during provoked attacks.
Note: Results at each time point are expressed as mean percentage change (standard error of the mean).
Mean percentage change of plasma concentration of CGRP (a), tryptase (b) and histamine (c) during and after 20-min infusion of either PACAP38 or VIP. The solid black line with black dots represents the average of individuals who developed a CH attack, while the dashed black line with white dots represents the average of individuals who did not develop CH attacks. Error bars are the standard errors of the mean. The red area represents the infusion period of either PACAP38 or VIP.
Baseline values of plasma CGRP, tryptase and histamine
Group comparison revealed no differences in plasma CGRP between active eCH patients (n = 27) (96.0 pmol/l, IQR 80.0–151.0), eCH patients in remission (n = 29) (99.0 pmol/l, IQR 62.5–138.5) and cCH patients (n = 25) (104.0 pmol/l, IQR 83.5–133.5) (p > 0.050) (Table 5). When we pooled our CGRP data with a previous study (17), plasma CGRP levels showed no difference between CH groups (p > 0.050). In the post-hoc analysis, we removed data from patients who participated in both trials. We found no difference in plasma tryptase between active eCH patients (n = 20) (3.02 μg/l, IQR 2.32–3.92), eCH patients in remission (n = 25) (3.65 μg/l, IQR 2.67–4.42) and cCH patients (n = 22) (4.47 μg/l, IQR 3.74–6.44) (p > 0.050) (Table 3). Plasma histamine did not differ between active eCH patients (n = 22) (4.75 nmol/100 μl, IQR 2.90–7.10), eCH patients in remission (n = 18) (7.25 nmol/100 μl, IQR 3.30–21.45) and cCH patients (n = 16) (7.00 nmol/100 μl, IQR 3.90–23.98) (p > 0.050) (Table 3). The baseline values are displayed in Figure 3.
Baseline levels of plasma CGRP, histamine and tryptase presented as median (interquartiles).
Baseline plasma concentration of CGRP (a), tryptase (b) and histamine (c).
Discussion
The major finding of our study is that PACAP38- and VIP-induced CH attacks are not associated with alterations of plasma CGRP, tryptase or histamine. These data suggest that CH attacks provoked by PACAP38 and VIP are unlikely to be mediated by CGRP or mast cell activation.
CGRP
The first study on plasma concentrations of CGRP was performed in eCH patients (n = 9) during spontaneous attacks (18). The authors reported that treatment with either oxygen or sumatriptan resulted in a reduction of plasma CGRP collected from the external jugular vein (18). This study did not investigate CH patients during interictal periods. Two other studies found that plasma CGRP collected from the external jugular vein increased in eCH patients (n = 12 and n = 11) during glyceryl trinitrate-provoked attacks (19,20). Furthermore, both spontaneous and sumatriptan-induced remissions reversed the plasma CGRP elevation (19,20). In the present study, we found no alterations in plasma CGRP levels during PACAP38 or VIP provoked attacks in eCH and cCH patients. Similar findings were observed in patients with migraine, in which infusion of PACAP38 did not alter plasma CGRP (21). Data derived from animal studies concerning the relationship between PACAP38 and CGRP are inconclusive, suggesting that other factors are likely to play a role in CGRP increase. In rats, PACAP38 caused the release of CGRP from the trigeminal nucleus caudalis, but not from the trigeminal ganglion or the dura mater (22). In the rat ear, PACAP38 inhibited both chemically and electrically induced release of CGRP from capsaicin-sensitive fibres in a concentration-dependent manner (23). Whether PACAP38- and VIP-induced attacks and CGRP-induced attacks are mutually independent remains to be seen. It would be interesting to investigate if monoclonal antibodies targeting the CGRP pathway prevent PACAP38- or VIP-induced CH attacks.
Mast cells
Plasma tryptase is a marker of mast cell activation, released by human mast cells together with histamine (24). Agonists of the proteinase-activated receptor 2 (PAR2), including tryptase, activate trigeminal nociceptors and dural mast cells, causing hyperalgesia and neurogenic inflammation (25,26). Tryptase has never been investigated in CH. In a prior study, histamine was investigated in the whole blood of 20 CH patients who experienced 22 CH attacks (27). Of these, six attacks occurred spontaneously, five were induced by the ingestion of alcohol, and 11 by the sublingual administration of nitroglycerin. Blood histamine was higher during attacks than the pre-headache period, with a mean increase around 20% (27). Another study reported an increased platelet-rich plasma histamine in five CH patients during spontaneous attacks, compared to free periods (28). In the current study, we did not find alterations of plasma tryptase or histamine during provoked CH attacks. Clinical trials demonstrated that the histamine H1 and H2 receptor antagonists, alone and/or in combination, were not efficacious for the prevention of CH (29–31). Interestingly, the concentration of serum tryptase did not change during PACAP38-induced migraine attacks (32). Collectively, these data suggest that is unlikely that mast cells play a major role in PACAP38- and VIP-induced CH. The implementation of a novel monoclonal antibody targeting the PAR2 receptor would reveal the involvement of mast cells in CH (33).
Baseline comparison
This is the first study investigating differences in plasma levels of histamine and tryptase in CH. Two previous studies reported higher plasma CGRP in active eCH patients compared to eCH patients in remission (19,20). We previously reported higher CGRP levels in eCH patients in remission than in cCH patients (17), which were not reproduced in the present study. Our studies were powered to investigate attack induction and not changes in plasma CGRP. Given the similar methodology used in conducting plasma evaluations and the same statistical approach, we pooled data from both studies into a single, merged study population (active eCH patients = 22, eCH patients in remission = 23, cCH patients = 25). The post-hoc analysis revealed no baseline difference of plasma CGRP between CH groups. There were no obvious disparities in the use of prophylactic medications, with verapamil being the most common used in both studies. While acknowledging the limitations associated with the post-hoc analysis, we believe that our study was statistically powered to detect a 25% difference among the three study populations (34).
Study limitations
This is an exploratory study and our results should be interpreted with caution. Previous studies found a more than doubled value of plasma CGRP during CH attacks, compared to the postictal state (18–20). However, we observed no changes in plasma CGRP even with a larger sample size and the presence of a control group. Different methodologies, including intra-assay differences and the location of blood sampling may explain discrepancies. We collected blood from the antecubital vein and not from cranial veins, such as the external jugular vein. However, only a small fraction of CGRP from the dura mater reaches the jugular blood, the total amount of plasma CGRP is largely derived from perivascular nerve terminals and not from the brain (35,36)f. Interpretation of our findings was also made difficult due to the presence of non-normally distributed data, as evident by the standard deviation larger than the mean for each biochemical variable. In addition, the presence of data missing completely at random may result in a loss of statistical efficiency. We discarded all observations with missing data and performed a so-called “complete case analysis”. We chose this approach, rather than imputing these values, because the impact of missing laboratory measurements on sample size was only modest and their probability was essentially due to random issues. To date, there are no accepted robust methods for handling missing data (37). The choice of the most appropriate method for handling missing data is ideally informed by an understanding of the mechanism(s) that led to the missing observations. We therefore consider the current statistical approach a coherent method of dealing with our missing values. It cannot be assumed that increased methodological complexity leads to less bias, there are situations in which multiple imputation methods produce identical bias levels as in a complete case analysis (38). Studies with pre-specified outcomes are warranted in the future to elucidate whether CGRP may be a reliable marker of CH attacks.
Conclusions
The present study demonstrates no evident changes of plasma CGRP, tryptase or histamine collected from the antecubital vein during experimentally provoked CH attacks. Whether CH attacks provoked by CGRP and PACAP38/VIP are mediated by distinct signaling pathways will be worth investigating in forthcoming studies. Targeting the pathophysiology of CH with the VIP/PACAP38 blockade might be beneficial in patients who are not responding to anti-CGRP therapies.
Key findings
PACAP38- and VIP-induced cluster headache attacks are not associated with alterations in plasma CGRP, histamine and tryptase. Cluster headache attacks provoked by PACAP38/VIP might be mediated by CGRP independent pathways.
Footnotes
Acknowledgments
Not applicable.
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: KB is employed by RefLab ApS, where the histamine measurements were conducted. PSS is acting as scientific advisor for RefLab ApS and EP Medical. Messoud Ashina (MA) reports personal fees from AbbVie, Allergan, Amgen, Eli Lilly, Lundbeck, Novartis and Teva. MA has no ownership interest and does not own stocks of any pharmaceutical company. MA serves as associate editor of Cephalalgia, associate editor of The Journal of Headache and Pain, and associate editor of Brain. LP, BA, ALHV, AHS and RHJ declare no actual or potential conflict of interest.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by grants from Lundbeckfonden (grant R252-2017-1317) and Research Foundation of Rigshospitalet.
