Plasma CGRP Levels in Migraine: A Registry for Migraine Study.

Migraine is a prevalent and disabling neurologic disorder characterized by recurrent attacks of moderate-to-severe headache, often accompanied by photophobia, phonophobia, nausea, and vomiting. 1 Despite extensive research, the precise molecular mechanisms underlying migraine remain incompletely understood, and no circulating biomarker has been validated for clinical use in diagnosis, prognosis, or therapeutic monitoring. 2 Among candidate molecules, calcitonin gene-related peptide (CGRP) has occupied a central role in migraine research for decades, owing to its involvement in nociceptive signaling within the trigeminovascular system. 3 The prominence of CGRP in migraine research stems, in part, from a small 1990 study, which reported elevated plasma CGRP in the external jugular vein during spontaneous migraine attacks. 4 Although the study included only 22 participants with migraine and relied on small historical control groups (12 external jugular and 31 cubital fossa samples), 4 these findings shaped the widely held belief that jugular CGRP levels increase during migraine attacks, supposedly reflecting disease activity. Furthermore, the same authors later demonstrated in 8 participants with ongoing migraine attacks that subcutaneously administered sumatriptan reduced CGRP levels, 5 supporting the idea that CGRP release is implicated in migraine pathophysiology and may be modulated by acute treatment. The subsequent demonstration that IV CGRP infusion could trigger migraine attacks further reinforced the peptide's mechanistic importance and positioned CGRP as both a therapeutic target and a potential biomarker. 6 Building on this foundation, a wave of therapeutic innovations has emerged, including small-molecule receptor antagonists (gepants) and monoclonal antibodies targeting either the ligand or the receptor, thereby expanding the therapeutic arsenal for migraine care. 7 Yet a fundamental question has remained clearly underexamined: do peripheral plasma CGRP concentrations reliably reflect migraine disease activity, and can they serve as a biomarker across the clinical spectrum? As there are several methodological limitations of the 1990 study 4 —including a small sample size, use of an assay with unvalidated human specificity, and comparison to historical controls—it is surprising that many later investigations have been largely unsystematic. Subsequent investigations employing assays specific to human CGRP even failed to replicate the original findings of elevated CGRP during migraine attacks. 8 It has been questioned whether using the jugular vein is necessary to access trigeminovascular CGRP, but the same study found a strong correlation between concentrations when comparing the external jugular and cubital veins, suggesting only minor differences related to sampling site. Nevertheless, the assumption that plasma CGRP rises during migraine attacks and could serve as a clinically meaningful biomarker has persisted, shaping research priorities for decades. This is particularly intriguing as the original findings have been difficult to reproduce, with only modestly sized studies yielding inconsistent findings across populations and methods being published since then. 9-15 To address this critical gap, we performed this rigorously controlled study of unprecedented sample size to compare plasma CGRP concentrations between adults with migraine and matched healthy controls (HCs). We further examined whether plasma CGRP varied across migraine subtypes and disease states. In doing so, this work offers a critical re-evaluation of CGRP's broader role as a biomarker, providing a framework for rethinking long-standing assumptions and guiding the next chapter of migraine research and biomarker discovery.

This prospective, single-center, cross-sectional observational study analyzed data from the parental Registry for Migraine (REFORM) study, conducted between September 2020 and June 2022. 16 In this substudy, plasma CGRP was measured at a single timepoint before initiation of erenumab treatment, and associations with clinical characteristics were examined. The REFORM protocol received approval from the Ethics Committee of the Capital Region of Denmark (H-20033264) and was preregistered at ClinicalTrials.gov (Identifiers: NCT04603976 ; NCT04674020 ). The study was conducted in accordance with the Declaration of Helsinki, 17 and written informed consent was obtained from all participants before any procedures. Recruitment of participants with migraine was conducted primarily via referral from the outpatient clinic of a tertiary care center. Eligible participants were adults (≥18 years of age) diagnosed with migraine without aura, migraine with aura, or chronic migraine according to the International Classification of Headache Disorders, third edition (ICHD-3). 18 Participants were required to experience at least 4 monthly migraine days before enrollment. Key exclusions included inability to distinguish migraine from other headache types, migraine onset after 50 years, hemiplegic migraine, cluster headache, or secondary headache disorders, with the exception of medication-overuse headache. Participants were excluded if they had any prior exposure to erenumab or if they had used other anti-CGRP monoclonal antibodies within 3 months before enrollment. Concomitant preventive medications, including onabotulinumtoxinA, were permitted if the dosage had been stable for at least 2 months before screening. Inclusion and exclusion criteria are presented in eTable 1. HCs were recruited through web-based advertising campaigns and were excluded if they reported a personal or family history of primary headache disorders, except infrequent episodic tension-type headache (see eTable 2). Upon enrollment, site investigators conducted a semistructured interview to collect comprehensive information on demographics, clinical features, and both current and medical and treatment history. Moreover, participants completed the 12-item Allodynia Symptom Checklist-12 (ASC-12). 19 At the time of blood sampling, site investigators recorded the presence, characteristics, and associated symptoms of headache, and recent use of an nonsteroidal anti-inflammatory drug (NSAID) or triptan (≤72 hours). Venous blood was drawn by antecubital phlebotomy into 9 mL K 2 EDTA tubes and within 5 minutes centrifuged at 2,200 g for 10 minutes at 4°C. Plasma was then promptly aliquoted into cryotubes and stored at −80°C until analysis, with all samples coded and randomly arranged to ensure blinding. Plasma CGRP was quantified between April 2022 and August 2024 at the Department of Biochemistry, Copenhagen University Hospital – Rigshospitalet, using a validated, high-affinity radioimmunoassay (RIA) specific for human CGRP. 20 This assay uses a rabbit polyclonal antibody directed against the amidated C-terminal of human α-CGRP and therefore does not distinguish between α- and β-CGRP. It demonstrates intra-assay and interassay coefficients of variation of 2% and 7%, respectively, with a detection limit below 1 pmol/L. 20 All samples were analyzed in duplicate by a board-certified laboratory technician blinded to group allocation and clinical information. The tracer was prepared by iodination of [Tyr 0 ]-α-CGRP-amide using the Iodo-Gen method (Pierce, Rockford, IL) and purified by high-performance liquid chromatography (Merck-Hitachi, Merck, Darmstadt, Germany), as previously described. 20 Samples, antibody, and calibrators were incubated at 4°C in a nonequilibrium assay for ∼90 hours, followed by 48 hours with the tracer. Bound and free tracer were separated using a solid-phase system (SAC-CEL; IDS, Boldon, the United Kingdom). Samples exceeding 300 pmol/L were diluted to extend the upper limit of quantitation to 1,200 pmol/L. Previous validation experiments have confirmed the assay's specificity to human CGRP, robust peptide stability at room temperature for up to 6 days without addition of protease inhibitors, and resilience to repeated freeze–thaw cycles. 20 Consistent with this, no correlation was observed between storage time at −80°C and plasma CGRP concentrations (Spearman ρ = −0.032, p = 0.37), supporting peptide stability under long-term storage conditions. Moreover, its performance in vivo has been demonstrated in human studies, showing comparable plasma concentrations across multiple vascular compartments under baseline conditions, 21 and sensitivity to dynamic changes following IV α-CGRP infusion in migraine patients. 22 The primary outcome was the comparison of plasma CGRP concentrations between participants with migraine and HCs. Secondary outcomes included comparisons of plasma CGRP concentrations across clinically relevant nonmutually exclusive migraine subgroups based on ICHD-3 18 : Migraine with aura vs migraine without aura. Episodic migraine vs chronic migraine. Ictal (participants who experienced headache fulfilling definite or probable migraine criteria at the time of sampling) vs interictal (headache-free at the time of sampling). Ongoing use of 1 or more preventive migraine medication(s) vs no current use of preventive medication use. Migraine with aura vs migraine without aura. Episodic migraine vs chronic migraine. Ictal (participants who experienced headache fulfilling definite or probable migraine criteria at the time of sampling) vs interictal (headache-free at the time of sampling). Ongoing use of 1 or more preventive migraine medication(s) vs no current use of preventive medication use. Each migraine subgroup was also compared against HCs to contextualize findings across the broader clinical spectrum. In the subgroup analysis of ictal vs interictal participants, we furthermore stratified by recent triptan intake (≤72 hours). Finally, we explored associations between plasma CGRP concentrations and key clinical variables. No formal sample size calculations were performed. The sample size was determined by participant enrollment in the parental REFORM study, with an expected sample of about 600 participants with migraine and 150 HCs. To reduce confounding and ensure balance, each HC was matched to 4 participants with migraine based on age, sex, body mass index (BMI), and time from blood collection to analysis, using optimal propensity score matching. Categorical variables were summarized as counts and percentages. Continuous variables were reported as means with ±SDs or medians with interquartile ranges (IQRs), depending on data distribution assessed by histograms and quantile-quantile plots. Given skewed distribution, CGRP concentrations were reported as medians (IQRs), with means ± SDs provided in tables. CGRP levels were compared between participants with migraine and HCs in the matched sample using the Mann-Whitney U test. As a sensitivity analysis, group comparisons were repeated in the subgroup of participants who were headache-free at the time of blood sampling. In addition, group comparisons were repeated with inclusion of an extreme outlier CGRP value that was excluded from the primary analysis. Linear regression was used for exploratory analyses on associations between plasma CGRP concentrations and key clinical variables. Variables with p < 0.20 in univariate analyses were included in a multivariable model. The final model also incorporated predefined covariates, including chronic migraine, migraine with aura, ictal status, preventive medication use, and NSAID or triptan use within 72 hours. Plasma CGRP concentrations were analyzed on the untransformed scale, as model fit was adequate and estimates are directly interpretable; log-transformation was evaluated but did not materially improve model diagnostics and was therefore not retained. Model assumptions were evaluated using standard diagnostic procedures, and multicollinearity was assessed using variance inflation factors, with values below 5 considered acceptable. Spearman ρ was calculated to examine the correlation between the number of months from sample collection to laboratory analysis and measured plasma CGRP concentrations. Significance was set at a 2-sided p value of 0.05. The Bonferroni method was applied to correct for p value multiple pairwise comparisons within each subgroup analysis for the primary and secondary outcome analyses, while exploratory analyses were corrected using the Benjamini-Hochberg procedure. For the main analyses, complete case analysis was performed, as missing data were below 5%. For exploratory regression models, where missingness ranged from 5% to 10% for certain covariates, multiple imputation of 20 data sets was conducted using a random forest algorithm. The influence of imputing missing data was evaluated by comparing results from imputed data sets with those derived from complete case analysis. All statistical analyses were conducted using R (version 4.5.1). Data and study materials that support the analyses presented in this manuscript can be obtained from the corresponding author upon reasonable request, subject to applicable ethical and legal restrictions.

From September 2020 to June 2022, 751 participants with migraine were enrolled in the REFORM study. Of these, 740 underwent blood sampling, and plasma CGRP concentrations were quantified in 632 samples. One plasma CGRP measurement was far outside the range of all other observations (26,704 pmol/L; ∼25 SDs above the mean) and exceeded the assay's upper limit of quantification. Given the unphysiological level and high likelihood of an analytical deviation, it was excluded from the primary analysis and included in a sensitivity analysis, yielding an analytical dataset of 631 participants with migraine. For comparative analyses, 588 participants with migraine were matched in a 4:1 ratio to 147 HCs, ensuring balance across age, sex, BMI, and time from sample collection to laboratory analysis. Demographic and clinical characteristics of the study population are summarized in Table 1 , with missing data shown in eTable 3. Among 588 participants with migraine in the matched sample, the mean (±SD) age was 43.1 ± 11.9 years, and 88.8% (n = 522) were female, with a mean BMI of 24.9 ± 4.8 kg/m 2 . Chronic migraine was present in 64.6% (n = 380), and 29.3% (n = 172) had migraine with aura. Preventive migraine medications were used by 49.7% (n = 292), of which 76 (12.9%) received onabotulinumtoxinA, and 32.8% (n = 193) had a history of multiple preventive treatment failures. Of the 522 females in the matched sample, 153 (29.6%) reported being postmenopausal and 212 (40.6%) reporting using hormonal treatments in the form of oral contraceptives, hormonal intrauterine devices, or hormonal replacement therapy. Demographics and Clinical Characteristics Abbreviations: — = not applicable; ASC-12 = allodynia symptom checklist-12; BMI = body mass index; IQR = interquartile range; NSAID = nonsteroidal anti-inflammatory drug. Participants with migraine matched 4:1 with healthy controls on age, sex, BMI, and time from sample collection to analysis. Unpaired t test. Pearson's χ 2 test. Mann-Whitney U test. <5% missing data. 5%–10% missing data. ASC-12 score ≥6. At the time of blood sampling, 45.3% (n = 265) reported ongoing migraine headache, 18.8% (n = 110) reported nonmigraine headache, and 35.9% (n = 210) were headache-free. Within the preceding 72 hours, 28.3% (n = 152) had used an NSAID, and 34.9% (n = 188) had used a triptan. Of the 265 participants reporting ongoing migraine headache, 89 (33.6%) had used a triptan within 72 hours, 167 (63.0%) had not, and data on recent acute medication use lacked in 9 participants (3.4%). Gepants were not used by any participants, as these were not available at study baseline. Participants with migraine and HCs were well-matched on age, sex distribution, BMI, and the proportion of current smokers ( Table 1 ). The median time from sample collection to laboratory analysis was also comparable between groups (30.0 months [IQR 26.9–33.7] vs 32.8 months [IQR 22.1–33.7]; p = 0.23). Plasma CGRP concentrations are presented in Figure 1 and Table 2 . Among the matched sample, participants with migraine had significantly lower median (IQR) CGRP levels compared with HCs (125 [68–173] vs 151 [118–199] pmol/L; p 0.999). CGRP levels were also similar between participants with migraine with aura and those without aura (125 [65–180] vs 124 [72–171] pmol/L; p > 0.999). Furthermore, ictal and interictal participants had comparable CGRP levels (116 [64–163] vs 133 [81–177] pmol/L; p = 0.092). Participants receiving 1 or more preventive migraine medications also showed no difference compared with those not using preventives (125 [75–175] vs 124 [62–172] pmol/L; p > 0.999). Notably, all migraine subgroups exhibited significantly lower CGRP levels compared with HCs (all p ≤ 0.001). Plasma calcitonin gene-related peptide (CGRP) concentrations are shown for (panel A) participants with migraine vs healthy controls (HCs); (panel B) migraine with aura (MA), migraine without aura (MO), and HCs; (panel C) chronic migraine (CM), episodic migraine (EM), and HCs; and (panel D) ictal vs interictal participants compared with HCs. Box plots indicate median values (horizontal bar), interquartile ranges (hinges), and 1.5 × interquartile ranges (whiskers). Diamonds represent means, while individual measurements are displayed as dots. Statistically significant differences after Bonferroni correction were assessed using the Mann-Whitney U test; ** p < 0.01 and *** p < 0.001. Plasma CGRP Levels (Matched Samples) Abbreviations: CGRP = calcitonin gene-related peptide; CM = chronic migraine; EM = episodic migraine; HC = healthy control; IQR = interquartile range; MA = migraine with aura; MO = migraine without aura. Mann-Whitney U tests, with p value corrected for multiple testing using the Bonferroni procedure. Participants with migraine matched with healthy controls on age, sex, body mass index, and time from sample collection to analysis. Data on recent triptan intake missing in 9 participants with migraine headache and 28 participants who were headache-free. Sensitivity analyses restricted to headache-free participants produced results largely consistent with the main analysis ( Figure 2 ). Median plasma CGRP concentrations in HCs did not differ significantly from those in participants with migraine with aura (151 [118–199] vs 143 [86–179] pmol/L; p = 0.126) or chronic migraine (151 [118–199] vs 142 [90–178] pmol/L; p = 0.090). Sensitivity analysis including the extreme CGRP outlier that was excluded from the main analysis yielded conclusions consistent with the primary findings. Mean-based summaries were, however, distorted with SDs inflated beyond ±1,000 pmol/L, whereas median (IQR) and nonparametric inferential statistics remained unchanged. For further details, see eTables 4 and 5. Plasma calcitonin gene-related peptide (CGRP) concentrations are shown for headache-free participants in the matched sample, comparing (panel A) participants with migraine vs healthy controls (HCs); (panel B) migraine with aura (MA), migraine without aura (MO), and HCs; and (panel C) chronic migraine (CM), episodic migraine (EM), and HCs. Box plots indicate median values (horizontal bar), interquartile ranges (hinges), and 1.5 × interquartile ranges (whiskers). Diamonds represent means, and individual measurements are displayed as dots. Statistically significant differences after Bonferroni correction were assessed using the Mann-Whitney U test; ** p < 0.01 and *** p < 0.001. In univariate linear regression, participants with unilateral cranial autonomic symptoms had higher CGRP levels than those without such symptoms (19 pmol/L higher; 95% CI 3–32; p = 0.018). Similarly, participants with moderate-to-severe ictal cutaneous allodynia (ASC-12 score ≥6) showed higher CGRP levels than those with mild or no allodynia (20 pmol/L higher; 95% CI 6–35; p = 0.005). Moreover, participants with a diagnosis of anxiety also showed higher CGRP levels compared with those without (22 pmol/L higher; 95% CI 1–44 pmol/L; p = 0.044). None of the other assessed clinical variables, including recent use of a triptan or ongoing treatment with onabotulinumtoxinA, demonstrated a significant association with CGRP levels (all p > 0.05). In the multivariable model, associations were no longer statistically significant for unilateral cranial autonomic symptoms, moderate-to-severe ictal cutaneous allodynia, or anxiety (all p > 0.05) ( Figure 3 ). These findings were consistent across analyses using both multiple imputation and complete case data sets available in eTables 6 and 7. Associations between clinical variables and plasma calcitonin gene-related peptide (CGRP) concentrations are shown for the full analytical dataset (n = 631). Estimates are derived from multivariable linear regression and presented as β-coefficients (pmol/L) with 95% CIs. p Values were adjusted for multiple comparisons using the Benjamini–Hochberg procedure.

In this large cross-sectional investigation, we found that plasma CGRP was significantly lower in individuals with migraine compared with matched HCs. Importantly, this pattern was evident across all examined migraine subtypes, with no meaningful differences observed between episodic vs chronic migraine, migraine with vs without aura, and between individuals sampled during ictal vs interictal phases. Furthermore, preventive treatment use, including onabotulinumtoxinA specifically, was unrelated to plasma CGRP concentrations, and sensitivity analyses restricted to headache-free participants confirmed robustness of the findings. Exploratory analyses indicated a possible association of unilateral cranial autonomic symptoms, ictal cutaneous allodynia, and anxiety with elevated CGRP levels in univariate models, though this did not remain after multivariable adjustment. Importantly, plasma storage duration was unrelated to measured CGRP values, strengthening the reliability of the biochemical analyses. Together, these findings challenge longstanding assumptions that migraine is associated with elevated circulating CGRP and cast doubt on the utility of peripheral plasma CGRP as a biomarker for migraine disease activity or treatment monitoring. CGRP has occupied a central role in migraine research for more than 3 decades. Much of this prominence stems from the 1990 study, which reported elevated CGRP concentrations in the external jugular vein during spontaneous migraine attacks, 4 a finding later reinforced with the finding that sumatriptan reduced CGRP levels during migraine attacks. 5 Although these studies included only a small number of participants, 4,5 they catalyzed the enduring hypothesis that migraine is characterized by elevated circulating CGRP and shaped biomarker discovery for decades. However, important methodological limitations of this early work were not fully appreciated at the time. The RIA used was based on a polyclonal rabbit antiserum raised against rat CGRP, 23 which is highly homologous but not identical to human CGRP, 24 and its specificity for human samples has never been validated. Moreover, the use of historical rather than contemporaneous controls further increased bias. 4 Subsequent investigations have produced inconsistent results. Some studies reported ictal increases in CGRP in cubital or jugular venous blood, 10-12,15 while others failed to replicate such elevations even under tightly controlled conditions. 8,9,14 In a pivotal follow-up investigation, 8 both the original 1990 RIA and a more specific human RIA, the latter also applied in the present study, were used to examine plasma CGRP dynamics across venous compartments. The analyses demonstrated no significant ictal increase in external jugular or antecubital plasma CGRP among individuals with migraine, but instead revealed strong correlation in plasma CGRP concentrations across compartments. These results suggested that jugular venous sampling does not necessarily provide a more proximal signal of meningeal peptide release than antecubital sampling. This finding likely reflects the fact that only a small proportion of external jugular venous drainage arises from intracranial circulation, 25,26 where the meningeal contribution remains uncertain. Additional work confirmed cross-compartment consistency of CGRP levels in healthy adults, with comparable concentrations observed in the internal and external jugular veins, cubital vein, and radial artery. 21 Robust, time-locked increases following IV infusion of human α-CGRP, reaching a mean plasma concentration of 428 pmol/L, further validated the RIA assay used in this study and confirmed its sensitivity to dynamic changes. 22 Outside of attacks, the results have been equally contradictory. Some reports described elevated CGRP levels in migraine, particularly chronic migraine, 12,15,27-30 while others found no difference or even numerically lower concentrations compared with controls. 13,15,31 Methodological differences likely contribute to this heterogeneity: early work primarily relied on in-house RIAs of varying specificity, whereas later studies increasingly applied commercial ELISAs. 32 Yet a widely used commercial ELISA kit, applied in numerous studies, was recently shown to fail to detect mature α- and β-CGRP. 33 This finding casts doubt on results generated with ELISA platforms whose analytical specificity and sensitivity are not well-established. Taken together, these inconsistencies underscore the challenges inherent to CGRP quantification and highlight the importance of rigorous assay validation. Beyond methodological factors, clinical heterogeneity and comorbidities may also contribute to inconsistent results. Rosacea, for instance, has been associated with elevated circulating CGRP and is comorbid in up to half of individuals with migraine, most of whom remain undiagnosed. 34 In a large study of 491 individuals, CGRP was elevated in those with rosacea irrespective of migraine status, raising the possibility of confounding effects in smaller cohorts. 35 In this study, only 10 participants had an established diagnosis of rosacea, limiting the ability to explore its potential influence on circulating CGRP levels. Other conditions, including endometriosis, have also been linked to altered CGRP levels. 36 Preclinical evidence indicates that estrogen and progesterone can modulate CGRP signaling in the trigeminovascular system, including CGRP and receptor expression in trigeminal ganglion neurons. 37 Hormonal states and exogenous hormones may therefore contribute to variability in circulating CGRP. In older, small studies, higher plasma CGRP-like immunoreactivity was reported in females using oral contraceptives, 38 with increasing levels during pregnancy followed by postpartum normalization, 39 and higher levels postmenopausally. 40 In this study, we observed no differences in plasma CGRP by contraceptive use or postmenopausal status; however, menstrual cycle phase was not recorded, and thus, residual confounding related to cycle-related variation cannot be excluded. Of note, while hormonal therapy use was common among female participants, its use was comparable to the Danish background population. 41 A further mechanistic hypothesis is that repeated or sustained trigeminal activation could deplete CGRP from perivascular afferents, leading to lower circulating levels. In persistent post-traumatic headache and cluster headache, lower plasma CGRP concentrations have been reported compared with controls, potentially reflecting such depletion. 42,43 Experimental animal models provide supporting evidence: capsaicin injection induces local CGRP release followed by depletion of peptide stores. 44,45 Moreover, early physiologic studies showed that electrical stimulation of the trigeminal ganglion in cats evoked CGRP release into the cranial circulation. 46 In humans, direct ganglion stimulation during thermocoagulation for trigeminal neuralgia caused facial flushing accompanied by increased jugular venous CGRP, raising the theoretical possibility that trigeminovascular activation can release CGRP and, with sustained activation, potentially deplete its stores. 46 Similarly, decreased plasma pituitary adenylate-cyclase-activating polypeptide-38 levels have also been reported interictally compared with controls 47 ; however, the results have been inconsistent. 30 Although our study showed significantly lower CGRP in migraine, we did not observe associations with ictal status comparing participants sampled during attacks with those sampled interictally nor with headache frequency, chronicity, or preventive medication use, raising further questions about the clinical correlates of this reduction. Nevertheless, the wide variability in CGRP concentrations within the migraine group compared with HCs indicates substantial biological heterogeneity, which could increase the risk of inconsistent or false-positive findings in smaller cohorts and leaves open the possibility that CGRP levels may still reflect disease activity in a subset of patients. The present investigation builds on the available literature by applying a validated, high-affinity, human-specific RIA with stringent preanalytical protocols and large-scale matched analyses. 20 With almost 600 participants with migraine and nearly 150 matched controls, this is the largest and most rigorously controlled study of plasma CGRP to date. By demonstrating that circulating plasma CGRP is not elevated—but instead modestly lower—in migraine, the study challenges longstanding assumptions, reframes interpretation of past work, and underscores the need for biomarker strategies that align more closely with migraine pathophysiology. The present findings carry several implications for both migraine research and clinical practice. First, while CGRP remains a key therapeutic target, these findings caution against using peripheral plasma CGRP concentrations as a biomarker for migraine diagnosis, disease activity, or therapeutic monitoring using currently available methods. The consistent absence of ictal elevations, subgroup differences, or associations with preventive use suggests that circulating CGRP lacks the sensitivity and specificity required for clinical translation; thus, plasma CGRP should be interpreted cautiously outside controlled experimental settings. Second, the results emphasize the importance of anatomical and mechanistic context in biomarker discovery. CGRP is primarily released from perivascular trigeminal sensory nerve terminals, where it induces vasodilation and promotes nociceptive messaging. 46,48 These effects are spatially restricted and local, making them unlikely to generate sustained systemic changes detectable in antecubital or external jugular venous samples. Future studies might focus on compartments more proximal to trigeminovascular activation, including internal jugular venous blood. Third, these findings highlight the critical distinction between therapeutic target validity and biomarker validity. CGRP-targeting therapies are effective, 7 yet this this does not imply that peripheral plasma CGRP is a clinically useful biomarker. This broader principle is relevant to migraine biomarker discovery, emphasizing the need to separate mechanistic mediators from circulating indicators. In addition, the relationship between circulating neuropeptides and biological activity may be more complex than a simple concentration-effect coupling. It may be speculated that variants enhancing receptor affinity or downstream signaling could confer strong biological effects even at lower circulating concentrations. A recent genome-wide association study identified risk loci at CALCA and CALCB , which encode α- and β-CGRP, but none were found in in the genes encoding CGRP receptor components. 49 The biological relevance of these findings remains to be clarified. Fourth, future research could pursue longitudinal, intraindividual sampling designs across migraine phases to capture dynamic fluctuations in CGRP within individuals, minimizing interindividual variability and providing a clearer picture of phase-specific pathophysiology. Other neuropeptides, including substance P and pituitary adenylate cyclase-activating polypeptide might provide complementary or synergistic insights. Indeed, it seems plausible that multimodal biomarkers are likely needed to disentangle the complex biology of migraine and to advance toward precision medicine. Several limitations should be considered in interpreting the present findings. The cross-sectional design precludes longitudinal assessment of within-individual fluctuations in plasma CGRP across migraine phases. Even so, inclusion of both ictal and interictal participants revealed no differences in concentrations, indicating that dynamic variability did not obscure the overall case-control signal. Blood was sampled from the antecubital vein, which reflects predominantly extracranial rather than intracranial drainage. Importantly, previous validation studies demonstrated comparable CGRP concentrations across jugular, cubital, and radial compartments, 8,21 supporting the reliability of antecubital sampling for systemic quantification. The applied RIA cannot distinguish α- from β-CGRP, 20 although α-CGRP predominates in trigeminal afferents, 50 and the assay demonstrated high sensitivity in α-CGRP infusion paradigms. 22 Although single time point sampling cannot exclude short-term variability, analyses of multiple large subgroups—including episodic vs chronic migraine and aura vs no aura—yielded findings consistent with the principal analysis. These subgroup cohorts exceeded the entire sample sizes of many earlier reports, 4,8-13,27,29 substantially reducing the likelihood of overlooked associations. Some subgroup comparisons could still have lacked statistical power, yet directional consistency with the main findings further diminished this concern. Despite matching and multivariable adjustment for relevant covariates, residual confounding from unrecognized comorbidities, hormonal influences, or medication use cannot be completely excluded. Moreover, most participants were recruited from specialized care, and the cohort was predominantly female with a high proportion of chronic migraine, which could limit generalizability to community and primary care populations. Finally, preanalytical variation cannot be entirely dismissed, but biospecimens were uniformly processed, stored at −80°C, analyzed in duplicate, and demonstrated no correlation between storage duration and measured plasma CGRP, supporting biochemical stability. Peripheral plasma CGRP concentrations were significantly lower in participants with migraine compared with matched HCs, with no differences across migraine subtypes or clinical states. Although CGRP signaling remains central to migraine pathophysiology and a validated therapeutic target, these findings challenge the long-standing assumption that migraine is characterized by elevated circulating CGRP. Future biomarker research should prioritize mechanistically relevant compartments and integrative approaches to advance precision medicine in migraine.