CGRP-provoked headache is associated with a neuroimmune signature in idiopathic intracranial hypertension

Abstract

Background

Idiopathic intracranial hypertension (IIH) is characterised by raised intracranial pressure and disabling headaches, yet its underlying mechanisms remain poorly defined. Calcitonin gene-related peptide (CGRP) is a key neuropeptide implicated in migraine and headache attributed to IIH. CGRP-driven neuroinflammation is implicated in migraine generation but whether CGRP-provoked headache in IIH is accompanied by dynamic neuroimmune activation during the ictal phase remains unknown. We aimed to characterise cytokine alterations during experimentally-induced IIH headache attacks provoked by CGRP.

Methods

In a randomised, double-blind, placebo-controlled, two-way crossover study (IIH Provoke: ISRCTN13251508), 16 women with IIH underwent two inpatient visits receiving intravenous CGRP (1.5 µg/min for 20 min) or placebo (isotonic saline). Blood samples were collected at baseline and either at peak headache or at the end of the visit, if no headache occurred. Plasma cytokines were measured using the Olink^® Target 48 Cytokine Panel. CGRP induced a typical IIH headache attack within the sampling period in 7/16 participants.

Results

Compared with baseline, eight cytokines were significantly elevated during CGRP-provoked headache attacks: C-C motif chemokine 4 (CCL4), CCL8, interleukin-6 (IL-6), IL-17C, IL-27, oncostatin M (OSM), hepatocyte growth factor (HGF), and transforming growth factor-α (TGF-α) (P < 0.05). These changes were not observed after placebo or in participants who received CGRP without developing headache.

Conclusions

CGRP-provoked IIH headache attacks were associated with elevations in pro-nociceptive mediators (IL-6, CCL4, OSM) and immune-related cytokines (IL-17C, IL-27, HGF, TGF-α). These findings suggest that CGRP-provoked headache in IIH is associated with changes in circulating cytokines consistent with a neuroimmune response during the ictal phase. Larger cohorts are required to validate these observations and explore neuroinflammation as a therapeutic target in IIH headache.

Supplementary Information

The online version contains supplementary material available at 10.1186/s10194-026-02385-0.

Introduction

Idiopathic intracranial hypertension (IIH) is defined by raised intracranial pressure (ICP) and optic disc swelling (papilloedema) in the absence of a secondary cause [1–3]. Consensus guidance for diagnosis, investigation and management was published in 2018 [2]. Epidemiological data indicate that both the incidence and prevalence of IIH have risen substantially over the past decade, mirroring the increasing burden of obesity [4–6].

Headache is the predominant symptom, reported in more than 80% of individuals with IIH [7–11]. It is highly disabling, a major driver of poor quality of life and remains challenging to treat [12–14]. Neuroinflammation, including cytokine-driven activation of the trigeminovascular system [15–17] and downstream protein kinase pathways [16, 18], is increasingly implicated in headache pathophysiology and may contribute to both headache generation and chronicity [15, 16, 18–20]. Headache is commonly reported during systemic infections and is also a recognised adverse effect of immune modulating/cytokine-based therapies, supporting a role for inflammatory signalling pathways in headache generation [15]. Furthermore, the efficacy of non-steroidal anti-inflammatory drugs in acute headache treatment is consistent with involvement of inflammatory mediators [21].

The majority of IIH headaches are migraine-like [7, 22]. Calcitonin gene-related peptide (CGRP) is a vasoactive and nociceptive neuropeptide implicated in migraine and is a key modulator of the above-mentioned pathways [23, 24]. CGRP has also been implicated in IIH headache. Animal studies demonstrate that the CGRP antagonist olcegepant alleviates ICP-induced allodynia [25], and elevated CGRP levels have been reported in patients with IIH, particularly those with a migraine-like phenotype [26, 27]. Clinical trials in both primary and secondary headache disorders, including IIH, have demonstrated that CGRP induces headache attacks [23, 28–30].

IIH was historically regarded as a disorder confined to the central nervous system (CNS) and neuro-ophthalmic axis [31]. Growing evidence, however, suggests that IIH has features of a systemic metabolic disease, with distinct hormonal, metabolic and inflammatory profiles that inform disease pathophysiology [32, 33]. Inflammatory dysregulation is increasingly implicated in IIH, with case–control studies demonstrating altered serum and cerebrospinal fluid (CSF) cytokine profiles, supporting the presence of both systemic and central neuroinflammation [33, 34], although findings remain heterogeneous and sometimes conflicting [34–38]. Reported changes include elevations in pro- and anti-inflammatory cytokines, with evidence of association with visual function, CSF opening pressure and relapse risk, suggesting potential mechanistic relevance [34–37]. Overall, these data indicate dysregulated inflammatory signalling in IIH, which may be aetiologically relevant through potential effects on regulating CSF dynamics, venous outflow and neural tissue integrity, although precise mechanisms remain undetermined [34, 37, 39].

Crucially, no study has evaluated cytokines across the ictal (headache attack) and interictal (outside of an attack) phases of headache attributed to IIH. The aim of this study was to determine changes in inflammatory cytokines during the ictal phase by provoking typical IIH headaches with CGRP in a randomised, double-blind, placebo-controlled, two-way crossover design. We hypothesised that cytokine levels change during IIH headache attacks and that characterising these alterations would provide insight into the neuroinflammatory signature in IIH headache. Specifically, we aimed to identify cytokine changes unique to the ictal phase of provoked headache, distinct from profiles observed after placebo or in participants receiving CGRP without headache.

Materials and methods

These data were derived from a parent study (IIH Provoke: ISRCTN13251508, registered: 24/10/2022) that was approved by the National Research Ethics Committee of North East - Newcastle & North Tyneside 1 (22/NE/0081). All participants provided written informed consent in accordance with the Declaration of Helsinki and its later amendments. A detailed description of the study and its primary outcomes have been reported elsewhere according to the CONSORT guidelines [30].

Study population

Females aged 18–60 years with a prior diagnosis of IIH by the IIH consensus guideline criteria [2] (papilloedema, elevated lumbar puncture opening pressure ≥ 25 cm H₂O, normal neuroimaging except markers of raised ICP but excluding venous thrombosis, normal CSF constituents and normal neurological examination except sixth nerve cranial palsy) and headache attributed to IIH (International classification of headache disorders 3rd edition (ICHD-3)) [40], were screened at University Hospitals Birmingham National Health Service Foundation Trust, UK. Key exclusions included a history of migraine predating IIH, > 1 day/month non-IIH headache, pregnancy, significant comorbidity or cardio/cerebrovascular disease, medication overuse, use of glucagon-like peptide 1 receptor agonists or other immune-modulating medications.

Participants attended two inpatient visits (> 6 days apart) and received a 20-minute continuous intravenous infusion of either CGRP (1.5 µg/min) or placebo (isotonic saline). Peripheral venous blood samples were taken on both visits. The order and administration of the provocation agent was randomised and double-blinded. Drug preparation/administration was undertaken by independent staff, as both investigators and participants were masked. Participants abstained from caffeine, nicotine, and alcohol for 12 h and withheld ICP-modifying, vasoactive, diuretic, preventive and analgesic medications for at least 5.5 half-lives prior to each visit. Visits were deferred if a spontaneous headache exacerbation occurred in the preceding 72 h from the scheduled study visit.

A priori criteria defined a provoked IIH headache attack as one mimicking the participant’s usual IIH headache as previously published (Supplemental Fig. 1) [30]. Head pain severity was recorded using an 11-point numeric rating scale (NRS): 0–10 where 0 = no pain and 10 = maximum pain.

Blood sample collection procedures

An indwelling intravenous cannula was placed for administration of the provocation agent. Peripheral venous blood was obtained from the cannula whenever feasible; if not, a separate venepuncture in the arm was performed. At both CGRP and placebo visits, the baseline (BL) sample was drawn on arrival, before any food intake at the research facility. Participants attended fasted or after a light breakfast. The second sample was taken at the peak of the provoked headache ictal phase (H+) or, if no attack occurred, at the end of the inpatient visit (H-), approximately 4 h after baseline. Sampling time points are illustrated in Fig. 1. Plasma was collected into 4 mL EDTA (purple-top) VACUETTE^® tubes and centrifuged within 30 min of collection at 2,000 g for 10 min at 4 °C and stored at -80 °C. All samples were processed following one freeze-thaw cycle.

Fig. 1.

Blood sampling time points. BL: Baseline blood sample collected on arrival, H+: peak headache blood sample, H-: blood sample 4 h after the baseline sample. CGRP: calcitonin gene-related peptide

Plasma analyses – Olink^® Target 48 Cytokine Panel

The Olink^® Target 48 Cytokine Panel multiplex Proximity Extension Assay (PEA) with reagents (Olink Proteomics, Uppsala, Sweden) was used to analyse proteins in the plasma samples from participants following the manufacturer’s protocol [41]. PEA uses pairs of antibodies tagged with complementary DNA oligonucleotides [42]. When both antibodies bind their target, the oligonucleotides hybridise and are extended by DNA polymerase, generating a protein-specific DNA sequence that is quantified by quantitative polymerase chain reaction (qPCR). This enables sensitive multiplexed protein measurement from minimal sample volumes (∼1 µL for up to 45 proteins, Olink^® Target 48) [43], with background noise minimised by multiple internal controls [41].

Sample analysis was conducted by the Birmingham Genomics Service. Each run processed 40 plasma samples using a 48 × 48 integrated fluidic circuit (IFC). Plasma (1 µL) was incubated with antibodies for 18 h. An Applied Biosystems™ Veriti™ 96-well Thermal Cycler was used for extension and pre-amplification of the hybridised DNA tags. The pre-amplified samples and primers were then loaded into a primed IFC. Forty amplification cycles were performed via qPCR using the Olink Signature Q-100 instrument. Quality control and data analysis were performed using Olink’s Normalised Protein eXpression (NPX) Signature software (Data normalisation and standardisation, https://olink.com/knowledge/documents, accessed 20/05/2025).

The Olink^® Target 48 PEA incorporates internal and external controls to ensure assay reliability and minimise technical variation. Each sample contains four internal controls: two incubation controls to monitor the entire assay process, an extension control to assess extension and amplification, and a detection control to evaluate the qPCR step. External controls included inter-plate controls to adjust for run-to-run variation, negative controls to define background signal, and pooled plasma samples analysed in duplicate to assess intra- and inter-plate variability. Plate and sample quality were evaluated using Olink quality control criteria: internal control deviation within ± 0.3 NPX, plate control variation within ± 30%, and an intra-assay coefficient of variation < 30%. Data were normalised using standard Olink procedures to minimise residual inter-plate variation, and runs failing quality thresholds were excluded. An internal standard curve was used to convert normalised protein concentrations to picograms per millilitre (pg/mL). Laboratory analyses were performed blinded to treatment allocation and visit order.

Statistical analysis

Baseline cytokine values were plotted in histograms and a Shapiro–Wilk test revealed that cytokine values were overall normally distributed. Where otherwise specified, data is reported as mean (standard deviation (SD)). Outliers, which deviated significantly from the main data distribution (mean ± > three standard deviations) were inspected individually and removed only when considered likely to reflect technical artefacts rather than biologically plausible extremes.

Paired t-tests were used to compare cytokine levels; at baseline between the first and the second visits, at different timepoints (baseline vs. headache attack (H+) or baseline vs. no headache at approximately 4 h (H-), in both the CGRP and the placebo visit). Pearson’s correlation was used to test for relationships between cytokine levels and clinical parameters. To assess for period or carryover effects, unpaired t-test was used to test for differences between cytokine levels of the participants in the two allocation arms (randomisation for the order of visits of each of the provocation infusions i.e. CGRP and placebo) and to compare baseline cytokine levels on the CGRP visit between participants who developed a CGRP-provoked headache (H+) and those who did not (H-). This was an exploratory, hypothesis generating study and did not aim to delineate diagnostic biomarkers, hence data were not corrected for multiple testing in the primary analysis. Results should be interpreted cautiously, with emphasis on consistency across biologically related cytokines rather than individual p-values. As a post-hoc sensitivity analysis, p-values were adjusted for multiple comparisons using the Benjamini–Hochberg false discovery rate (FDR) method. This was applied to within-participant comparisons between baseline and ictal sampling (H+) during the CGRP visit in participants who developed a headache, across all cytokines passing quality control. However, the study was not powered for multiple comparison-adjusted inference and such correction may be overly conservative in this hypothesis-generating context. A p-value of 0.05 was considered as the level of significance. Analyses were performed with SPSS Statistics (Version 29 Armonk, NY: IBM Corp) and GraphPad Prism 10.1.2 (GraphPad software).

Results

Study sample

A total of 20 women with IIH were enrolled in the study and 16 completed both the CGRP and placebo visits providing blood samples for each visit. Mean (SD) age was 26.2 (6.3) years and BMI 39.2 (8.4) kg/m². Clinical characteristics are summarised in Table 1. During the inpatient monitoring following CGRP, 7 participants developed a typical IIH headache and were sampled at baseline and peak headache (H+), whereas 9 remained headache-free and were sampled at baseline and end-of-visit (~ 4 h) (H-). During the placebo visit, no participant developed a headache, and all samples were obtained at baseline and end-of-visit (~ 4 h) (H-). The mean (SD) time from CGRP administration to headache attack (H+) sample collection was 195 (62) min. The mean time to sample collection among participants who did not experience a headache attack (H-) was 261 (37) min.

Table 1.

Demographics and clinical baseline characteristics of the study participants

Clinical Characteristics, n = 16	Mean (SD)
Age (years)	26.2 (6.3)
Weight (kg)	107.1 (23.0)
BMI (kg/m2)	39.2 (8.4)
Time from diagnosis (months)	17.4 (19.6)
Papilloedema measured by OCT RNFL (worst eye) (µm)	145.5 (58.5)
Monthly headache days	24.9 (8.3)
Non-migrainous headache severity	5.9 (2.9)
Monthly migraine-like days	16.3 (9.4)
Monthly migraine-like days severity	9.0 (1.5)
	n (%)
Ethnicity
• White - British	13 (81%)
• Mixed - White & Black Caribbean	3 (19%)
Headache phenotype
• Migraine-like	15 (94%)
◦ Episodic Migraine-like	2 (13%)
◦ Chronic Migraine-like	13 (81%)
• Tension-type headache-like	1 (6%)
Family history of migraine	4 (25%)
Acetazolamide use	3 (19%)
Migraine preventative use	3 (19%)
Triptan use	5 (31%)
Simple analgesic use	12 (75%)

BMI: Body mass index, IIH: Idiopathic intracranial hypertension, OCT: Optical coherence tomography, RNFL: peripapillary retinal nerve fibre layer

Baseline cytokine measurements

Out of the 45 proteins analysed, 7 proteins (16%) (granulocyte-macrophage colony-stimulating factor (CSF2), Interleukins IL-17A, IL-17F, IL-1β, IL-2, IL-33, and thymic stromal lymphopoietin (TSLP)) had levels below the limit of quantification and therefore failed quality control and were excluded from the final analysis, leaving 38 proteins for analysis. Cytokines were compared between baseline visits of CGRP and placebo administration to assess for period or carryover effects. There were no significant differences in the baseline cytokine levels between participants in either of the randomised allocation arms (order of the visits for provocation agents CGRP or placebo, Supplementary Table 1), nor in baseline cytokine levels on the CGRP visit between participants who developed CGRP-provoked headache (H+) and those who did not (H-) (Supplementary Table 2). There were also no significant correlations between monthly headache days, monthly migraine-like days and baseline cytokine levels.

Ictal cytokine measurements

We compared cytokine levels at baseline and during the ictal phase of an experimentally induced IIH headache attack after administration of CGRP revealing a number of changes (Fig. 2). In comparison to baseline levels, a significant decrease was noted in Fms-related tyrosine kinase 3 ligand (FLT3LG) (p = 0.022) (Fig. 2A). A significant increase was noted in C-C motif chemokine 4 (CCL4) (p = 0.038), C-C motif chemokine 8 (CCL8) (p = 0.049), hepatocyte growth factor (HGF) (p = 0.007), interleukin-17C (IL-17C) (p = 0.044), interleukin-27 (IL-27) (p < 0.001), interleukin-6 (IL-6) (p = 0.014), oncostatin M (OSM) (p = 0.011) and transforming growth factor alpha (TGF-α) (p = 0.042) (Fig. 2B - I). FDR-adjusted p-values for these primary cytokine comparisons are provided in Supplementary Table 3.

Fig. 2.

Cytokine levels in participants at the ictal phase during an experimentally-induced headache attack after receiving calcitonin gene-related peptide (CGRP). Mean change (SD) in cytokine levels in participants with idiopathic intracranial hypertension (n = 7) during an experimentally-induced headache attack after CGRP. Changes are compared from baseline (BL) to the peak headache timepoint (H+), after CGRP. A Fms-related tyrosine kinase 3 ligand (FLT3LG). B C-C motif chemokine 4, (CCL4). C C-C motif chemokine 8 (CCL8). D Hepatocyte growth factor (HGF). E Interleukin-17C (IL-17C). F Interleukin-27 (IL-27). G Interleukin-6 (IL-6). H Oncostatin M (OSM). I Transforming growth factor alpha (TGF-α). * p < 0.05 between BL and H+ after CGRP, ** p < 0.01 between BL and H+ after CGRP, *** p < 0.001 between BL and H+ after CGRP

In participants who did not experience headache attack following CGRP (H-), there were no significant changes in cytokine levels from baseline, including no significant changes in FLT3LG, CCL4, CCL8, HGF, IL-17C, IL-27, IL-6, OSM, and TGF-α (Fig. 3).

Fig. 3.

Cytokine levels in participants after receiving calcitonin gene-related peptide (CGRP) in the absence of an inpatient headache attack. Mean change (SD) in cytokine levels in participants with idiopathic intracranial hypertension (n = 9) after receiving CGRP in the absence of an inpatient headache attack. Changes are compared from baseline (BL) to the second sampling timepoint (H-) after CGRP. A Fms-related tyrosine kinase 3 ligand (FLT3LG). B C-C motif chemokine 4, (CCL4). C C-C motif chemokine 8 (CCL8). D Hepatocyte growth factor (HGF). E Interleukin-17C (IL-17C). F Interleukin-27 (IL-27). G Interleukin-6 (IL-6). H Oncostatin M (OSM). I Transforming growth factor alpha (TGF-α). ns: no significant difference between BL and H- after CGRP

Cytokine changes were subsequently evaluated following placebo administration to determine whether the responses were specific to CGRP-induced headache (Fig. 4). Compared to baseline, a significant decrease was noted in FLT3LG (p = 0.038) (similar to CGRP-induced headache attack, H+) and a significant decrease in CCL8 (p = 0.04) (as opposed to an increase during CGRP-induced headache attack, H+) (Fig. 4). There were no significant correlations between changes in cytokine levels and changes in headache severity, in participants who experienced a CGRP-induced headache attack, although this analysis was likely underpowered to draw meaningful inferences.

Fig. 4.

Cytokine levels in participants after receiving placebo. Mean change (SD) in cytokine levels in participants with idiopathic intracranial hypertension (n = 16) after receiving placebo in the absence of an inpatient headache attack. Changes are compared from baseline (BL) to the second sampling timepoint (H-) after placebo. A Fms-related tyrosine kinase 3 ligand (FLT3LG). B C-C motif chemokine 4, (CCL4). C C-C motif chemokine 8 (CCL8). D Hepatocyte growth factor (HGF). E Interleukin-17C (IL-17C). F Interleukin-27 (IL-27). G Interleukin-6 (IL-6). H Oncostatin M (OSM). I Transforming growth factor alpha (TGF-α). * p < 0.05 between BL and H- after placebo, ns: no significant difference between BL and H- after placebo

Discussion

This study uniquely characterises ictal cytokine alterations during CGRP-provoked IIH headache attacks. Eight cytokines, including CCL4, CCL8, IL-6, IL-17C, IL-27, OSM, HGF, and TGF-α, were significantly elevated during CGRP-provoked headache attacks, but not after placebo infusion or in participants who received CGRP without developing headache. These mediators include pro-inflammatory and pro-nociceptive cytokines, pleiotropic immunoregulatory factors, and signalling molecules implicated in glial activation, vascular modulation, and nociception. Although the provoked attacks frequently fulfilled migraine phenotypic criteria, these are best interpreted as IIH-attributed, migraine-like exacerbations rather than de novo primary migraine, consistent with ICHD-3 classification [30, 44, 45]. Taken together, these findings suggest that CGRP-provoked headache in IIH is accompanied by an immune response that is temporally linked to the ictal phase.

Cytokine profile of CGRP-induced headache in IIH

Immune recruitment, glial activation and BBB-associated inflammation

Elevations in the chemokines CCL4 and CCL8 suggest activation of glial–immune and perivascular inflammatory pathways during CGRP-induced headache in IIH. CCL4 (MIP-1β), a CCR5 ligand involved in microglial activation and immune-cell recruitment [46–48], is not consistently elevated in spontaneous migraine [49], but its upregulation in IIH headache may reflect mechanisms related to raised ICP and glial–immune interactions in IIH [49, 50]. CCL8 (MCP-2), a chemokine involved in recruitment of monocytes and lymphocytes and associated with leukocyte trafficking across the blood–brain barrier (BBB) [51, 52], has been implicated in chronic migraine, cluster headache and neuropathic pain [53–55]. Its elevation in IIH aligns with emerging evidence of BBB vulnerability and neurovascular interface dysfunction in this condition [50, 56]. Together, these findings suggest that CGRP exposure in IIH headaches may be associated with immune recruitment and nociceptor sensitisation in the context of an already susceptible BBB [50, 56–58].

Pro-nociceptive and glial–immune signalling

The elevation of IL-6 and oncostatin M (OSM) might be consistent with involvement of pro-nociceptive and central sensitisation pathways during CGRP-induced headache. IL-6 is a well-established mediator of nociceptor excitability and central sensitisation via signal transducer and activator of transcription 3 (STAT3) and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signalling, and is elevated during the ictal phase of migraine [15, 16, 18, 49, 59–61]. Chronically raised ICP in IIH may further prime glial and meningeal networks, amplifying CGRP-induced IL-6 release and pain sensitisation [62–66]. OSM, a member of the IL-6 cytokine family [67], activates JAK/STAT, mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase/protein kinase B (PI3K/AKT) pathways [67–70] and has been reported to show altered expression in cluster headache across both central and peripheral compartments [17, 53]. Experimental pre-clinical models link OSM to pain chronification and hypersensitivity [70–72], suggesting a role for glial–immune interactions in driving susceptibility to headache in IIH.

Vascular, barrier and tissue-remodelling pathways

HGF and TGF-α elevations suggest involvement of vascular, barrier-associated or tissue-remodelling responses during CGRP-induced headache. HGF acts via the c-Met receptor and exerts angiogenic, neurotrophic and immunomodulatory effects [73–77], with evidence linking it to nociceptive neuron differentiation and CGRP release [78, 79]. Elevated HGF has been reported in cluster headache [53] and linked to migraine risk in Mendelian randomisation analyses of inflammatory cytokines [78]. Elevated HGF in this study may reflect amplified neurogenic inflammation or compensatory vascular or glial repair responses in IIH headache [80]. TGF-α, an EGF receptor (EGFR) ligand involved in proliferation and tissue repair [81, 82], has been reported in chronic cluster headache [17], although its role in headache remains poorly characterised. Elevated TGF-α levels during CGRP induced IIH headache may reflect a compensatory response to CGRP-driven trigeminovascular inflammation [83].

Immune modulation and counter-regulatory responses

IL-17C and IL-27 elevations may be consistent with activation of barrier-associated and counter-regulatory immune pathways. IL-17C, produced by epithelial and barrier-associated cells, promotes neutrophil recruitment and secondary pro-inflammatory signalling [84–88]. IL-17C has not been previously reported in headache disorders where IL-17A has been implicated [85–87, 89]. IL-27, a pleiotropic cytokine of the IL-6/IL-12 family [90], exerts both pro- and anti-inflammatory effects, including suppression of Th17 differentiation and induction of IL-10 [90–93], and has demonstrated antinociceptive and neuroprotective effects in experimental models [94–96]. Their upregulation in IIH headache may reflect compensatory immune mechanisms in response to CGRP-induced headache and neuroinflammatory activation. Importantly, not all cytokine changes were headache-specific: FLT3LG levels declined following both CGRP and placebo administration, indicating a non-specific temporal or procedural effect rather than a CGRP-mediated response. In contrast, the cytokines elevated during CGRP-provoked headache showed a coordinated pattern that may reflect neuroimmune modulation.

Overall, these findings may support a model in which CGRP exposure in susceptible individuals with IIH triggers a multifaceted neuroimmune response encompassing immune recruitment, glial activation, BBB modulation and nociceptive sensitisation. This integrated inflammatory profile provides mechanistic insight into headache in IIH. The results are hypothesis-generating given the exploratory nature of the study. Taken together, these cytokine-specific findings support a broader neuroimmune framework for CGRP-provoked headache in IIH.

Proposed neuroimmune mechanisms in IIH headache attacks

Pro-nociceptive cytokines such as IL-6, CCL4, OSM, and IL-17C are established drivers of trigeminal sensitisation and central pain amplification [15, 16, 18, 46–49, 59–61, 67–72, 84–88, 97]. In contrast, while HGF and TGF-α have been reported in cluster headache, upregulation of IL-17C and IL-27 has not previously been described, highlighting immune pathways that may warrant further investigation in IIH. However, the current study design does not permit determination of whether these cytokine changes contribute causally to headache generation or instead represent downstream responses to CGRP-induced neuronal activation. The signature of elevated cytokines observed here may be consistent with amplified neuroinflammatory signalling, increased vascular permeability, and immune cell trafficking alongside nociceptor sensitisation in susceptible individuals (Fig. 5). Within this cohort of patients with IIH who developed a CGRP-induced headache attack, this pattern may represent a primed neuroimmune state, in which chronically raised ICP and BBB vulnerability may lower the threshold for CGRP-triggered immune and glial responses. Rather than reflecting a uniform or classical inflammatory state [98], the observed cytokine changes likely represent dynamic, state-dependent signalling at the neurovascular and glial interfaces during the ictal phase of headache.

Fig. 5.

Cytokines profile observed during a calcitonin gene-related peptide (CGRP) provoked idiopathic intracranial hypertension (IIH) headache attack. BBB: blood-brain barrier, CCL4: C-C motif chemokine 4, CCL8: C-C motif chemokine 8, HGF: Hepatocyte growth factor, IL-6: Interleukin-6, IL-17C: Interleukin-17C, IL-27: Interleukin-27, OSM: Oncostatin M, TGF-α: Transforming growth factor alpha

These findings suggest that CGRP-provoked headache in IIH is associated with a coordinated neuroimmune response during the ictal phase, consistent with the possibility of a primed neuroimmune state in individuals who developed a headache following CGRP infusion [99–105]. Whether these cytokine changes contribute to headache susceptibility or represent secondary responses to trigeminovascular activation requires confirmation in larger mechanistic studies.

Limitations

While this study provides important mechanistic insight, certain limitations should be acknowledged. Headache attacks were experimentally provoked and may not fully replicate the pathophysiology of spontaneous IIH headaches, potentially limiting generalisability to naturally occurring attacks. However, this controlled experimental design enabled precise timing of blood sampling in relation to headache onset, which would be difficult to achieve during spontaneous attacks and was intended to minimise biological variability and enhance reproducibility of ictal measurements. We cannot completely exclude direct pharmacological effects of CGRP on inflammatory markers [104], although this was mitigated by including a control group that received CGRP infusion but did not develop headache. We also acknowledge that the small sample size may have reduced statistical power and inevitably limited the ability to detect associations with clinical measures. Nevertheless, this highly controlled experimental medicine approach, with sampling immediately prior to infusion and at peak headache, enabled within-person comparisons, which is a key strength and enhances the internal validity of the results. Seven cytokines were below the lower limit of quantification, restricting comprehensive evaluation of the inflammatory profile; however, these are unlikely to represent meaningful markers of the ictal phase given their very low concentrations. Finally, cytokines were measured in peripheral blood rather than cerebrospinal fluid and therefore do not directly reflect central neuroimmune processes. Nonetheless, blood-based cytokines provide insight into systemic inflammatory responses and offer greater practicality and translational potential as accessible biomarkers [106, 107].

Conclusion

We demonstrate that CGRP-provoked headache attacks in IIH were associated with elevations in CCL4, CCL8, IL-6, IL-17C, IL-27, OSM, HGF, and TGF-α, indicating activation of both pro-nociceptive and immune pathways. The higher levels of IL-6, CCL4, IL-17C and OSM are consistent with glial activation and trigeminal sensitisation, while the upregulation of IL-17C, IL-27, HGF, and TGF-α may highlight immune pathways of relevance in IIH that warrant further investigation. These exploratory data are consistent with the concept that IIH headache may reflect immune and glial priming in the setting of raised ICP, with CGRP acting as a central trigger. Further studies in larger cohorts are needed to determine the functional relevance of these cytokines and to clarify whether they contribute to headache generation or represent secondary responses within the landscape of IIH. Understanding neuroimmune pathways in IIH headache could reveal future therapeutic targets which modulate neuroinflammation.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

BBB

Blood brain barrier

BL

Baseline

BMI

Body mass index

CCL11

C-C motif chemokine 11 or Eotaxin

CCL13

C-C motif chemokine 13

CCL19

C-C motif chemokine 19

CCL2

C-C motif chemokine 2

CCL3

C-C motif chemokine 3

CCL4

C-C motif chemokine 4

CCL7

C-C motif chemokine 7

CCL8

C-C motif chemokine 8

CCR5

C-C chemokine receptor type 5

CGRP

Calcitonin gene-related peptide

CNS

Central nervous system

CSF

Cerebrospinal fluid

CSF1

Macrophage colony-stimulating factor (M-CSF)

CSF2

Granulocyte–macrophage colony-stimulating factor (GM-CSF)

CSF3

Granulocyte colony-stimulating factor (G-CSF)

CXCL10

C-X-C motif chemokine 10

CXCL11

C-X-C motif chemokine 11

CXCL12

C-X-C motif chemokine 12 or Stromal cell-derived factor 1

CXCL8

C-X-C motif chemokine 8 or Interleukin-8

CXCL9

C-X-C motif chemokine 9

CXCR

C-X-C motif chemokine receptor

DNA

Deoxyribonucleic acid

EDTA

Ethylenediaminetetraacetic acid

EGF

Epidermal growth factor

EGFR

Epidermal growth factor receptor

FLT3LG

Fms-related tyrosine kinase 3 ligand

H-

No inpatient IIH headache attack

H+

Inpatient IIH headache attack

HGF

Hepatocyte growth factor

HIT-6

Headache impact test-6

ICAM-1

Intercellular adhesion molecule 1

ICHD-3B

International classification of headache disorders 3rd edition beta

ICP

Intracranial pressure

IFC

Integrated fluidic circuit

IFN-γ

Interferon gamma

IIH

Idiopathic intracranial hypertension

IL-10

Interleukin-10

IL-13

Interleukin-13

IL-15

Interleukin-15

IL-17A

Interleukin-17A

IL-17C

Interleukin-17C

IL-17F

Interleukin-17F

IL-18

Interleukin-18

IL-1β

Interleukin-1β

IL-2

Interleukin-2

IL-27

Interleukin-27

IL-33

Interleukin-33

IL-4

Interleukin-4

IL-6

Interleukin-6

IL-7

Interleukin-7

JAK

Janus kinase

JNK

c-Jun N-terminal kinase

KATP

ATP-sensitive potassium channels

LTA

Lymphotoxin-alpha

MAPK

Mitogen-activated protein kinase

MMP1

Matrix metalloproteinase-1 or interstitial collagenase

MMP12

Matrix metalloproteinase-12 or macrophage metalloelastase

mRNA

Messenger ribonucleic acid

NPX

Normalised protein expression

NRS

Numeric rating scale

OCT

Optical coherence tomography

OLR1

Oxidized low-density lipoprotein receptor 1

OSM

Oncostatin M

PCR

Polymerase chain reaction

PEA

Proximity extension assay

PI3K/AKT

Phosphoinositide 3-kinase/ protein kinase B

qPCR

Quantitative polymerase chain reaction

RNFL

Peripapillary retinal nerve fibre layer

SD

Standard deviation

STAT3

Signal transducer and activator of transcription 3

TGF-α

Transforming growth factor alpha

TGFB1

Transforming growth factor beta 1

TNFSF10

Tumor necrosis factor ligand superfamily member 10

TNFSF12

Tumor necrosis factor ligand superfamily member 12

TNF-α

Tumor necrosis factor alpha

TSLP

Thymic stromal lymphopoietin

UK

United Kingdom

VEGFA

Vascular endothelial growth factor A

Author contributions

AY, LJH and AJS conceptualised the study. AY, TPD, SPM, SJEL, MA, CNT, LJH and AJS designed the study. AY acquired and analysed the data with the support of CWM, CNT, OG, LJH. AY drafted the manuscript. LJH and AJS edited the manuscript.All authors contributed to the interpretation of data for the work, reviewed the manuscript critically for intellectual content, have read and approved the final version of the manuscript.

Funding

AY is supported by a Clinical Research Training Fellowship funded by the Association of British Neurologists and the Guarantors of Brain for the duration of this study. MA is supported by institutional grants from Danish National Research Foundation, Lundbeck Foundation and Novo Nordisk Foundation. AJS is funded by the National Institute for Health Research (NIHR) Health Technologies Assessment programme (NIHR131211) and Sir Jules Thorn Award for Biomedical Research.

Data availability

Data will be made available upon reasonable request from 12 months to 3 years after publication of this article, subject to approval by the original study investigators. Proposals should be directed to the corresponding author, and requesters will be required to sign a data access agreement.

Declarations

Ethics approval and consent to participate

This study (IIH Provoke: ISRCTN13251508, https://www.isrctn.com/ISRCTN13251508 registered: 24/10/2022) was approved by the National Research Ethics Committee of North East - Newcastle & North Tyneside 1 (22/NE/0081). All participants provided written informed consent to participate in the study, in accordance with the Declaration of Helsinki.

Competing interests

AY reports speaker fees from Teva UK. SPM reports consultancy fees (Invex Therapeutics); advisory board fees (Ocular Therapeutix, Dompé); Speaker fees (Teva, AbbVie); Travel (Abbvie and European Alliance of Associations for Rheumatology); Receipt of equipment (Heidelberg Engineering). SPM is a section editor at Eye Journal, Eye Open and Journal of Neuro-ophthalmology. OG reports consulting fees from Invex Therapeutics unrelated to this research. MA reports receiving personal fees from AbbVie, Astra Zeneca, Eli Lilly, GlaxoSmithKline, Incyte, Lundbeck, Novartis, Pfizer and Teva Pharmaceuticals outside of the submitted work. MA reports serving as associate editor of The Journal of Headache and Pain and associate editor of Brain. AJS reports she has consulted for Novartis, AbbVie, Vertex and Orion Pharma and speaker fees from Teva Uk and Novartis on unrelated topics.