Biomarkers in Migraine
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.315988
Source of Support: None, Conflict of Interest: None
Keywords: Biomarker, cerebral spinal fluid, marker, migraine, saliva, serum
Migraine is the second-leading cause of years lived with disability and affects over 1 billion people globally. For a disease with such prevalence and profound impact on the world's economy, little is known about the objective attributes of the disease to help characterize and differentiate it from others. Although migraine remains a clinical diagnosis based on ICHD-3 criteria, extensive research has been ongoing to identify biomarkers and genetic markers to aid in the diagnosis and treatment of migraine. However, before these biomarkers may be applied to clinical use for precision medicine, they must first be substantiated. This review paper will highlight the purported biomarkers and their role in the diagnosis and management of migraine.
A biomarker can be defined as a physical sign or laboratory measurement that occurs in association with pathological processes and has putative diagnostic or prognostic utility. Here we divide biomarkers into two categories: diagnostic and therapeutic. Diagnostic biomarkers are those which act as an indication of the presence of the disease, whereas therapeutic biomarkers help predict prognosis and guide management with insight into treatment response.
A comprehensive literature search was conducted using PubMed, Google Scholar, and clinicaltrials.gov using keywords: “migraine” OR “headache” combined with “biomarkers” OR “marker.” Other keywords used include “serum,” “cerebral spinal fluid,” “inflammatory,” and “neuroimaging.”
A subset of diagnostic biomarkers that have proven useful as predictors of either the presence of or susceptibility to migraine is genetic in nature. Perhaps some of the most widely studied and elucidated genetic biomarkers for migraine are the ones implicated in familial hemiplegic migraine (FHM). Three causative genes and subtypes have been identified in FHM: CACNA1A on chromosome 19p13 (FHM1), ATP1A2 on 1q23 (FHM2), and SCN1A on 2q24 (FHM3). In FHM1, the most prevalent form of FHM, a mutation in the CACNA1A gene coding for the α1A subunit of the voltage-gated P/Q-type calcium channel (also referred to as CaV2.1), results in a gain-of-function in CaV2.1. This gain-of-function mutation yields increased cortical excitatory transmission and propagation of cortical spreading depression. The ATP1A2 gene associated with FHM2 encodes the α2 subunit of the Na/K+-ATPase expressed in astrocytes, which aids in the clearance of extracellular K+ and establishment of a Na+ gradient used in glutamate reuptake. This loss-of-function mutation results in increased glutamate in the synaptic cleft and resultant hyperexcitability. For FHM3, a mutation in the SCN1A gene encoding the α1 subunit of the voltage-gated sodium channel NaV1.1 results in accelerated channel recovery from fast inactivation. Given the role of NaV1.1 in generating and propagating action potentials, this mutation also leads to hyperexcitability by increasing dendritic response and neuronal firing. In a recent case report, Paucar et al. discovered a novel protein-truncating variant in SLC1A3 that appeared to be associated with acetazolamide-responsive hemiplegic migraine and cerebellar atrophy.
Another channel of interest in migraine is the TWIK-related spinal cord potassium channel (TRESK), also referred to as potassium channel subfamily K member 18 (KCNK18) or K2P18. In addition to its expression in the spinal cord, TRESK channels are located in the dorsal root ganglion as well as trigeminal sensory neurons and are believed to play a role in somatosensory perception and pain. A frameshift mutation in the KCNK18 gene (F139WfsX24) coding for TRESK is linked to a syndrome of familial migraine with aura. This mutation results in the truncation of TRESK from 384 aa to 162 aa, rendering the channel nonfunctional. In addition, the mutation has been found to generate a protein fragment that leads to the inhibition of TREK1 and TREK2, which are members of another family of potassium channels termed TWIK-related K+ channel (TREK). However, not all mutations that inactivate TRESK appear to be implicated in migraine. A recent review conducted by Andres-Bilbe et al. found that missense variants R10G, A34V, C110R, S231P, and A233V were not associated with migraine.
Genome-wide screens have been conducted in an effort to identify novel loci associated with migraine. An association study performed by Carlsson et al. on a multigenerational Swedish family with migraine revealed a significant linkage of chromosome 6p12.2-p21.1 to migraine, both with and without aura. To replicate this novel finding, Oterino et al. in 2011 studied a 134-family cohort and similarly found significant evidence linking migraine to markers located at the 6p12.2-p21.1 locus. Unfortunately, the large size of 6p12.2-p21.1 hinders the sequencing of individual genes for more specific identification of the link between this locus and migraine.
The role of dopamine signaling has also been investigated in migraine. Namely, studies have suggested a link between susceptibility to migraine and dopamine receptors type 2 (D2R) and type 4 (D4R). D2R is encoded by the DRD2 gene located on chromosome 11q22.2–22.3, and its most studied single-nucleotide polymorphisms (SNPs) are rs1799732 and rs6275. DRD2 rs1799732 is a deletion polymorphism that results in decreased D2R expression, whereas DRD2 rs6275 is a synonymous polymorphism. However, results varied among genetic studies, and the relationship has been inconclusive. A meta-analysis performed by Chen et al. revealed that there was no association between the risk of migraine for DRD2 rs1799732 and DRD2 rs6275. D4R, a D2R-like G protein-coupled receptor encoded by DRD4 on chromosome 11 at 11p15.5, however, appears to be associated with a predisposition to migraine. More specifically, it has been linked to migraine without aura and not migraine with aura.
In addition to the dopaminergic pathway, the hypocretin (orexin) system has been investigated in association with migraine. Aside from the system's primary role in regulating arousal, wakefulness, and appetite, the hypocretins have been implicated in the modulation of pain, as well as the autonomic and nociceptive characteristics seen in primary headache disorders. The hypocretin signaling pathway consists of the neuropeptide transmitters hypocretin-1 and hypocretin-2 and their respective G-protein coupled hypocretin receptors, HCRTR1, and HCRTR2. Polymorphisms in the HCRTR1 gene, which encodes HCRTR1, have been posited as risk factors of migraine. In a study by Rainero et al., the non-synonymous polymorphism rs2271933 (G1222A) showed significant association with migraine susceptibility, as carriers of the AA genotype were at twice the risk for migraine in comparison to carriers of the GG genotype. However, polymorphisms rs10914456 and rs4949449 did not appear to be linked to migraine.
Another pathway believed to be related to the pathogenesis of migraine is the serotonergic system. Given the success of 5-HT1B/1D receptor agonists (triptans) in alleviating symptoms of migraine attacks, studies have been conducted to probe whether polymorphisms in serotonin receptor genes contribute to migraine susceptibility. Although the contributions of polymorphisms in 5-HT1A, 5-HT1B, 5-HT2A, and 5-HT2C receptors to migraine pathogenesis remain under investigation, the general consensus is that mutations affecting the 5-HT receptor sequence do not have a significant association with the disease. Polymorphisms in the 5-HT transporter gene have also been studied. Specifically, Yılmaz et al. and Park et al. found that 5-HT transporter gene SLC6A4 variants 2.10 and 2.12 appeared more frequently in persons with migraine., A later meta-analysis conducted by Liu et al. corroborated that the 2.12 variant was associated with migraine. However, studies of insertion/deletion polymorphisms of the 5-HT-transporter-linked polymorphic region (5-HTTLPR) in the regulatory region of SLC6A4 did not prove a relationship. The deletion variant of 5-HTTLPR, also referred to as the short (S) allele, is a polymorphism that causes decreased expression of the 5-HT transporter and consequently decreased reuptake of 5-HT. Another variant, termed the long (L) allele, exists as well. Although the S variant has been shown to be associated with depression, a meta-analysis conducted by Schürks et al. failed to find a statistically significant association between the SLC6A4 5-HTTLPR polymorphism and migraine.
Methylenetetrahydrofolate reductase (MTHFR), the rate-limiting enzyme in the methyl cycle encoded by the MTHFR gene also appears to play a role in migraine. MTHFR is involved in the methylation of homocysteine into methionine. An excess of homocysteine has been shown to cause vascular tissue remodeling, and some studies suggest that migraine initiation and maintenance may be the result of endothelial dysfunction precipitated by hyperhomocysteinemia. The most common polymorphism of MTHFR is rs1801133 (C677T), resulting in an alanine-to-valine substitution at position 222 in the catalytic domain that alters the quaternary structure and reduces enzymatic activity approximately by 50%. This reduced enzymatic activity consequently leads to the accumulation of homocysteine in the blood, which is a risk factor for migraine and a number of cardiovascular diseases. In meta-analyses conducted by Rubino et al. and Schürks et al., the authors found that the TT genotype was associated specifically with an increased risk for migraine with aura. However, a later genetic association study conducted by Essmeister et al. in 2016 failed to replicate previous findings. Instead, their data suggested that MTHFR C677T had no influence on migraine susceptibility.
Another enzyme implicated in migraine and often closely studied with MTHFR is the angiotensin-converting enzyme (ACE). ACE is a key enzyme involved in the renin-angiotensin system (RAS), converting angiotensin I to the vasoconstrictor angiotensin II. Aside from its role in modulating blood pressure via regulation of fluid volume in the body, ACE has also been implicated in migraine. Numerous studies have been dedicated to studying the association of ACE deletion/insertion (D/I) polymorphism (rs1799752) with migraine, but results have largely been contradictory. Some studies have also suggested an interaction between MTHFR C677T and ACE D/I polymorphisms, but more recent data from Essmeister et al. suggested that ACE polymorphisms do not increase susceptibility to migraine, either alone or in combination with MTHFR polymorphisms.
The casein kinase 1 (CK1) family of serine/threonine kinases has also been studied in relation to migraine. These kinases are important regulators of multiple signal transduction pathways involved in cellular differentiation, proliferation, chromosome segregation, and circadian rhythms. Studies regarding mutations in CK1 isoform delta (CK1δ), encoded by CSNK1D on chromosome 17 (17q25.3) and largely responsible for the regulation of the circadian clock protein PER2, have suggested a link to migraine with aura. Specifically, the missense mutations T44A and H46R in CSNK1D appear to be co-segregated with migraine with aura in two independent families with familial advanced sleep phase syndrome (FASPS). Both mutations occur in the catalytic domain and result in decreased enzymatic activity.
Attention has also been turned to syntaxins as genetic biomarkers of migraine. Syntaxins are important mediators in the vesicular docking, fusion, and exocytic process observed in neurotransmitter release. Studies have generally revolved around one particular isoform of this protein: Syntaxin 1A. Encoded by the STX1A gene on chromosome 7q11.23, Syntaxin 1A has been shown to be involved in the regulation of GABA by inhibiting its reuptake as well as 5-HT by decreasing 5-HT transporter expression. Literature investigating STX1A polymorphism rs941298 and rs2293489 have shown a significant association to migraine without aura. Interestingly, the STX1A rs6951030 G allele contributed to migraine susceptibility whereas the T allele appeared to have a protective effect for migraine both with and without aura. However, polymorphisms rs4363087 and rs3793243 did not appear to be linked to migraine.
Given that migraine is considered to be a multifactorial disease with contributions from both endogenous and exogenous factors, some studies have investigated the potential role that hepatic enzymes may play in its pathophysiology. Glutathione S-transferases (GSTs) are expressed in the liver and purported to be found in the brain as well. GSTs are responsible for the detoxification of electrophilic compounds via conjugation with glutathione, thus providing cellular protection against oxidative stress. One of the GST classes, GST Mu 1 (GSTM1), is prone to acquiring a homozygous deletion (null) genotype that reduces its metabolic activity. Mattsson et al. in 2000 were the first to explore this null polymorphism in relation to migraine but concluded that there was no association between GSTM1 and migraine in a Swedish population. Building on the work of Mattsson et al., Kusumi et al. conducted the study in a Japanese population and reported that the GSTM1 null genotype appeared to be associated with migraine but not migraine with aura.
Another hepatic enzyme that has been investigated is Cytochrome P450 1A2 (CYP1A2). CYP1A2 is a member of the cytochrome P450 superfamily of enzymes that are responsible for the metabolism of xenobiotics in the body, such as the selective serotonin receptor agonist zolmitriptan used in the treatment of migraine attacks. In 2010, Gentile et al. explored whether an association existed between chronic migraine (CM) and allelic variants of CYP1A2. Specifically, Gentile et al. looked into the CYP1A2 -164A > C polymorphism (CYP1A2*1F; rs762551) and –3860G > A polymorphism (CYP1A2*1C; rs2069514). Although the link between CYP1A2*1C and migraine did not prove to be significant, CYP1A2*1F polymorphism appeared to be associated with a liability to migraine. However, the authors acknowledged the multifactorial nature of migraine and emphasized the need for further, more comprehensive studies to fully elucidate the relationship between the CYP1A2*1F variant and migraine.
Obesity has also been shown to be associated with greater susceptibility to migraine, with various studies estimating the risk to be increased by 40-80%. Recent translational human research has investigated this relationship and acquired data that suggest a role of adipokines in the pathogenesis of migraine. Adiponectin is produced largely by adipocytes for the purpose of energy homeostasis and inflammation, and is one such adipokine that has been widely investigated. Adiponectin is readily detectable in the serum, and also is present in cerebrospinal fluid, albeit at lower levels. In the brain, adiponectin acts on receptors found in the cortex, brainstem, and hypothalamus, as well as endothelial cells where they signal through various pathways including AMP-activated protein kinase (AMPK), mitogen-activated protein kinase (MAPK), nuclear factor kappa beta (NFkβ), and peroxisome proliferator-activated receptor alpha (PPARα). According to a systematic review conducted by Peterlin et al. in 2016 looking at interictal studies evaluating adiponectin levels in persons with migraine, the data suggest that migraine may be associated with higher levels of adiponectin. However, the authors acknowledged the wide variation in limitations and methodologies used by the studies, urging for more carefully designed studies with appropriate controls in the future before drawing definitive conclusions on a relationship. More recently in 2018, a cross-sectional study conducted by Dominguez et al. found that adiponectin was increased in persons with migraine.
Inflammatory and neuromodulatory biomarkers
Other diagnostic biomarkers that have shown promise in the detection of migraine are inflammatory and neuromodulatory markers. Namely, these markers include neuropeptides, cytokines, as well as products of biotransformation. Calcitonin gene-related peptide (CGRP) and pituitary adenylate cyclase-activating peptide-38 (PACAP-38) are endogenous neuropeptides that are known to be associated with migraine. CGRP resides in perivascular trigeminal sensory afferents and many CGRP-containing fibers can be found in the cerebral arterial walls of the circle of Willis. PACAP-38 is found in perivascular trigeminal nerve fibers and ganglia, the sphenopalatine ganglion (SPG), and the trigeminal nucleus caudalis (TNC). PACAP-38 belongs to the VIP/secretin/glucagon neuropeptide superfamily and acts at one specific receptor PAC1 and two non-specific receptors (VPAC1 and VPAC2) which are shared with VIP. The presence of mRNA or protein for VPAC1/2 and PAC1 receptors in human/rat middle meningeal arteries, trigeminal ganglia, TNC, SPG, mast cells, and macrophages suggest PACAP signaling mechanisms are involved in mediating cranial autonomic symptoms, but also in mediating dural neuro-inflammatory mechanisms that contribute to dural trigeminovascular activation. Once released, both CGRP and PACAP-38 produce a potent vasodilatory effect that is believed to be linked temporally to the induction of migraine attacks. Furthermore, a review conducted by Kaiser et al. found that multiple studies have reported elevated levels of CGRP and PACAP-38 during a spontaneous attack. Similar to CGRP and PACAP-38, neurokinin A (NKA), a member of the tachykinin family, is another neuropeptide that appears to be associated with migraine. Tachykinins are multi-functional and contribute to a wide variety of processes including nociception, inflammation, and itch. NKA is widely distributed in both the central and peripheral nervous systems, and it is believed to increase vascular permeability in response to trigeminal nerve activation. In a meta-analysis conducted in 2020, Frederiksen et al. found that blood levels of NKA were significantly elevated in migraine patients in the ictal phase compared to the interictal phase across multiple studies.
Proinflammatory cytokines such as interleukin (IL)-1β, IL-6, and tumor necrosis factor-alpha (TNF-α) as well as anti-inflammatory cytokines like IL-10 have also been purported to play a role in migraine. Proinflammatory cytokines, which are produced by activated macrophages, are responsible for upregulating inflammatory reactions and serve as important mediators of pain. IL-1β is expressed in nociceptive dorsal root ganglion neurons and has been found to increase neuronal and glial cell production of substance P and prostaglandin E2 (PGE2), both of which play a role in hyperalgesia. IL-6 has been shown to regulate neuronal neuropeptide expression and mediate neuropathic pain behavior following peripheral nerve injuries., Similarly, TNF-α is responsible for mediating inflammatory and neuropathic hyperalgesia. Anti-inflammatory cytokines, on the other hand, serve to control the pro-inflammatory cytokine response. IL-10 is a potent cytokine that represses inflammatory cytokine expression by activated macrophages and downregulates proinflammatory cytokine receptors. IL-10 has also been shown to reduce neuropathic pain and can act in the ventrolateral orbital cortex to attenuate allodynia.
Given the roles that these proinflammatory and anti-inflammatory cytokines play in the regulation of pain, many have attempted to elucidate their potential relationship with migraine. TNF-α levels are elevated in persons with migraine, with ictal levels being even higher than the baseline., In 2019, a study conducted by Han found that persons with migraine had increased levels of IL-6, IL-1β, and TNF-α in comparison to non-migraine controls. Likewise, adults with migraine have increased levels of the anti-inflammatory cytokine IL-10 during attacks.,
Products of biotransformation such as nitric oxide (NO) metabolites have also been implicated in migraine. NO produces potent vasodilatory effects mediated by the activation of guanylate cyclase, resulting in the synthesis of cyclic guanosine monophosphate (cGMP) and consequent decrease of intracellular Ca2 + levels. Synthesis of NO has been shown to be stimulated and potentiated by other substances such as serotonin and CGRP, both of which are mediators of pain in migraine attacks. Once the vasodilatory effect has been achieved, NO is rapidly metabolized into nitrites and nitrates, which are stable products that are easier to detect as potential markers. Sarchielli et al., Fidan et al., and Yilmaz et al. found that nitrate and nitrite levels were elevated in systemic blood during an attack, suggesting the utility of NO metabolites in predicting increased vulnerability to migraine.,, However, findings during the interictal period have been more conflicting. Although D'amico et al. reported increased NO metabolites during the interictal period, Guldiken et al. saw no significant difference, thus no conclusion regarding dysfunction of baseline NO metabolism in persons with migraine can be made.
Serum biomarkers have expanded the understanding of a range of neurological pathologies in recent years. Cerebrospinal fluid as compared to serum is believed to be more demonstrative of the biochemical environment of the brain. Therefore, CSF is a primary substance of interest in studying the pathophysiology of migraine. The excitatory neurotransmitter glutamate has been linked to neuronal hyperexcitability in migraine. Glutamate has been implicated in the onset and generation of cortical spreading depression (CSD), the proposed mechanism for migraine aura. A meta-analysis by van Dongen et al. showed glutamate concentrations are consistently increased in cerebrospinal fluid (CSF) of persons with migraine during ictus compared to controls. By contrast, interictal blood levels of glutamate from persons with episodic migraine (EM) were not statistically significantly different from controls. One limitation of this meta-analysis is that there are currently no studies of glutamate concentrations in the blood of persons with CM nor in interictal CSF in persons with EM for comparison.
Glycine and taurine were studied by Rothrock et al. The levels of taurine, glycine, and glutamine were significantly higher in 38 migraine patients than 10 headache-free controls (P < 0.0001 for taurine and glycine; P < 0.0009 for glutamine). Interestingly, in this same study, seven patients were treated with divalproex sodium. Afterward, CSF taurine levels decreased significantly from pretreatment baseline values. Neuropeptide Y-like peptide (NPY-LI) has been hypothesized to play a role in the pathogenesis of migraine. NPY-LI was studied in the CSF via suboccipital puncture, and in plasma from persons with migraine, and compared to other neuropsychiatric disorders. In this study, Vecsei et al. showed that NPY-LI did not change during the attack period in the CSF or plasma. However, this same paper indicated the possible role of somatostatin in the pathogenesis of migraine. The inter-ictal somatostatin-like immunoreactivity (SLI) concentration was decreased in those with migraine compared to concentrations in mixed neuropsychiatric groups. During the migraine attack, the level of SLI decreased further. NPY-LI levels did not change.
More recently, in a case-control study by Meyer et al., a novel radiological approach was taken to elucidate biomarkers in the CSF of persons with migraine. This was the first controlled study to use a technique called sodium MRI to evaluate CSF. Persons with migraine had significantly higher sodium concentrations in cerebrospinal fluid compared to non-migraine controls. One of the key propositions of this paper is that MRI may become a non-invasive imaging tool for therapeutics that affect sodium levels. Hence, if sodium fluctuates in the CSF during a migraine attack, therapies directed to the molecular level of the cause of migraine could be elicited, thereby leading to more targeted treatments.
Elevated TNF-α has been shown in the cerebrospinal fluid of persons with CM and new daily persistent headache (NDPH). A study from Rozen T et al. indicates the role of inflammation and endothelial dysfunction in the progression of migraine as marked by TNF-α. Most patients with CM and NDPH who had elevated CSF TNF-α levels had normal serum levels. Interestingly, the patients in this study who had elevation of CNS TNF-α showed minimal to no improvement during aggressive inpatient treatment. The authors conclude that persistent elevation of CSF TNF-α levels may be one of the causes of treatment-refractory headache disorders. It has also been postulated that reduced CSF levels of glial cell line derived neurotrophic factor (GDNF) and somatostatin in CM can contribute to sustained central sensitization underlying chronic head pain.
The dopamine metabolite 3,4-dihydroxyphenylacetic acid (DOPAC) increases in CSF during migraine attacks. The increase in this metabolite correlates positively with migraine pain intensity, suggesting that persons with migraine have central dopaminergic hyperfunction. This may be related to a coexisting central noradrenergic neurotransmitter dysfunction. Phosphatidylcholine-specific phospholipase C (PC-PLC) fluctuations in the CSF may also reflect deranged neurotransmission during migraine. Fonteh et al. compared the CSF of persons with migraine to that of non-migraine controls, and also to blood samples taken during their attacks. Plasma PC-PLC activity was 250-300 times lower than in CSF and did not increase in migraine, implicating the brain as the source of these enzymatic changes.
Imaging modalities have suggested different areas of the brain as having possible roles in migraine pathology. Positron emission tomography (PET) in the premonitory phase of triggered migraine shows changes in brain activity that correlate with migraine symptomatology. During the premonitory phase of migraine, the midbrain tegmentum, periaqueductal gray (PAG), dorsal pons, and cortical areas display altered function. Iron deposition in the PAG has been posited as a possible biomarker for CM. However, the correlation of iron deposition in this region has been studied infrequently and is heavily debate. Among many functions, the PAG plays a critical role in behavioral and autonomic responses to threat, and it is a primary control center for pain modulation and suppression. Increasing iron deposition in this area has been linked to a shift from EM to CM. Patients with CM display significantly increased iron deposition in the PAG by MRI evaluation compared with patients with EM. Further, patients with EM display increased iron accumulation in the PAG compared to controls. However, the nature of the relationship between iron accumulation in the PAG and migraine remains unclear with respect to cause, effect, or correlation. There are multiple theories for this finding, including free radical damage to structures involved in pain regulation. It is postulated that atypical iron homeostasis in the PAG may lead to irregular nociceptive regulation. This phenomenon remains ripe for further hypothesis and research.
Functional magnetic resonance imaging (fMRI) data suggest that alterations in mood, energy level, and appetite in the premonitory phase are linked to changes in hypothalamic function, with increased blood flow both before and during attacks. Additionally, fMRI has shown relevant changes in brain metabolic activity and blood flow during migraine aura. The thalamus is implicated as playing a key role in sensory sensitivity during the migraine phase as shown by increased thalamic responses to sensory stimulation based on fMRI findings.
Visual cortical processing has also been widely correlated with migraine in multiple imaging studies., Network-based dysfunction may be responsible for migraine rather than dysfunction of any single brain structure, as illustrated in an fMRI study aimed at identifying network abnormalities in persons with migraine without aura. Through machine learning, the researchers collected fMRI resting state functional connectivity data from persons with migraine without aura and non-migraine controls and found differences in network-based connections between these groups. Of particular interest was an abnormal connection between the sensorimotor and occipital cortices found only in those with migraine. There was also disruption in the cingulo-opercular networks in these individuals. Considering these results, the authors propose that there may be derangements in sensory integration in the brains of people with migraine.
Although we look for objective markers to assist in the diagnosis of migraine, the hope when exploring therapeutic biomarkers is to help advance practices of precision medicine in migraine by using knowledge of a patient's genetic makeup to individualize clinical treatment plans. Often it is seen that CM is undertreated due to poor treatment response or adverse effects of the treatment. Selecting the safest and effective treatment strategy for those who require preventive medication may help ensure higher adherence rates to treatment and subsequently prevent the progression of migraine from EM to CM. CGRP is one such molecule that has shown promise not only as a diagnostic marker as discussed above but also as a therapeutic biomarker. In a study by Cady RK et al., persons with episodic migraine with and without aura had increasing levels of CGRP in saliva as the headache progressed in intensity. The level of salivary CGRP then returned to near-baseline in those responsive to rizatriptan, a 5-HT1B/1D agonist, clinically. Those who did not respond to rizatriptan had no significant change in salivary CGRP levels at any phase of the attack or in response to rizatriptan, suggesting that elevated saliva CGRP levels correlate with a better response to 5-HT1B/1D agonist.
As discussed earlier, another biomarker that is elevated during an acute migraine attack is NKA. Sarchielli et al. found that both CGRP and NKA were elevated in venous blood samples from the external jugular vein of those with migraine. As compared to rizatriptan non-responders, responders had decreased levels of both biomarkers from external jugular vein samples one hour after administration which correlated with reported symptomatic relief of pain. Non-responders had varied levels detected after administration.
Parallel to CGRP levels in saliva, prostaglandins including PGD2, PGD4, and thromboxane A2 have also been detected during the acute phase of menstrual migraine attacks in a pilot study by Durham et al. The prostaglandin levels also correlated with the intensity of reported migraine pain and were elevated in placebo-treated participants compared to triptan-treated participants, who showed no increase at two and four hours after symptom onset. In a much earlier study in 1989, prostaglandin PGE2 levels were found to be lower than in those with migraine compared to those without but were significantly increased during an attack. Prostaglandin inhibitors such as mefenamic acid have also been tested as treatment options for those with menstrual migraine, resulting in significantly lower pain compared to those treated with placebo.
More recently, the role of microRNAs as key regulators of gene expression at the post-transcriptional level has added a source of potential regulation of drug response via controlled over-expression of drug-metabolizing enzymes, drug transporters, and drug targets. These miRNAs are expressed at aberrant levels in those with neurogenic pain in complex regional pain syndrome. This may account for variation between individuals with respect to pain sensitivity and response to treatment. Fewer studies have been completed looking at the role of miRNAs in migraine. In 2015, Tafuri et al. described the upregulation of miR-27b in patients with migraine without aura, whereas Anderson et al. identified high levels of miR-34a-5p and miR-382-5p as serum miRNA signatures during migraine attacks., Using this knowledge to guide further research can help to realize the full potential that miRNAs carry for migraine research and practice.
PACAP-38 has become another therapeutic biomarker of interest in migraine since the discovery that infusion of PACAP-38 caused headache and vasodilation in participants with or without migraine. Additionally, those with moderate or severe pain had elevated PACAP in the external jugular vein during headache; this reduced one hour after treatment with sumatriptan 6 mg and further upon attack resolution.
Migraine is a complex disease process for which there are several promising markers of identification and treatment. However further study is needed to corroborate and build on these findings before they may be broadly clinically applicable. In an era of precision medicine, it would behoove the medical community to untangle the complexity of migraine pathophysiology so that treatment can be better targeted to suit individuals and improve the prognosis of migraine.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.