Migraine genetics: New opportunities,new challenges

As migraine runs in families and has a strong genetic component, many investigations have aimed to identify the genetic factors that confer susceptibility to this disease. Initial studies of rare monogenic migraine disorders led to the identification of the first migraine genes (1). Specific mutations in the genes CACNA1A, ATP1A2 and SCN1A, which all encode subunits of ion transporters that play a part in neurotransmission, have been identified for familial hemiplegic migraine, an autosomal dominant subtype of migraine with aura characterized by a transient hemiparesis during the aura. Genes with an apparent vascular function were identified by studying monogenic diseases in which migraine is prominent, such as NOTCH3 in cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy and CSNK1D in familial advanced sleep phase syndrome.

As a result of a recent technical revolution in the way DNA analyses are performed, which allows cost-effective massive genotyping in large cohorts of many thousands of patients, it is now feasible to identify causal genetic variants for common polygenic disorders. This technical breakthrough has led to what is often referred to as the genome-wide association study (GWAS) era. As a consequence, in the last decade, numerous GWASs have been performed, which have led to the discovery of thousands of genes and gene variants for hundreds of traits, including various common neurological diseases (2). An essential factor for a GWAS to be successful in migraine was the fact that clinicians and geneticists with an interest in migraine organized themselves in the International Headache Genetics Consortium (IHGC) (http://www.headachegenetics.org/). The IHGC connects no less than 34 research groups from 12 countries from Europe, the USA and Australia and brings together an ever-increasing wealth of well-characterized genetic information and high-quality diagnostic data, as well as analysis expertise, to discover the basis for future treatments of migraine and other headache disorders. The success of IHGC’s gene discovery efforts is evident from four GWASs that investigated over 23,000 patients and hundreds of thousands of control subjects. Migraine-associated single nucleotide polymorphisms (SNPs) were identified in 13 loci and the assigned genes suggest the involvement of vascular (C7orf10, PHACTR1, TGFBR2) and neuronal (ASTN2, FHL5, LRP1, MEF2D, MMP16, MTDH, PHACTR1, PRDM16) pathways, as well as metalloproteinases (AJAP1, MMP16, TSPAN2) and a pain nociceptor (TRPM8) in migraine pathophysiology (3–6). As the identified gene variants, without exception, have small effect sizes (odds ratio < 1.2), a legitimate question is whether they will really further our understanding of migraine pathophysiology and have practical clinical applications.

This special issue of Cephalalgia contains reviews that present recent findings and viewpoints on how genetic data can be used in research and the clinic. In addition, it contains several studies with original data, most of which were initiated by the IHGC, which address issues related to, for example, the genetic architecture of migraine subtypes, gene variants in relation to drug response and disease pathophysiology.

The review by Chasman et al. (7) discusses the importance of the appropriate ascertainment of the migraine status of individual patients enrolled in genetic studies. Challenges and solutions in diagnosing patients in very large population-based cohorts and how this is different from clinic-based cohorts, where clinical information is, in general, more detailed, are extensively discussed. Because an increase in sample size of at least 10-fold (of the current set of 23,000 patients) is predicted to be necessary to identify gene variants with smaller effect sizes than the already identified variants, the authors make a plea that future efforts should be directed towards reliable questionnaire-based ascertainment methods in much larger initiatives at the national level in multiple countries.

Esserlind et al. (8) evaluated the migraine-associated SNPs in a large independent Danish clinic-based migraine cohort. Although all the SNPs showed the same direction of effect as in the original studies, only about half provided robust statistical replication, which, according to these researchers, points to clinical differences between patients in population- and clinic-based cohorts. With respect to possible clinical use, an increased risk with an increased number of risk alleles was observed for relevant clinical endpoints, such as the frequency of migraine attacks.

Using various statistical approaches, Zhao et al. (9) identified gene-based overlap (pleiotrophy) between the two main migraine types – migraine with and without aura – by mining the available genetic data, beyond the inspection of only the genome-wide significant loci, of over 4000 patients with migraine with aura, a similar number of patients with migraine without aura, and almost 75,000 control subjects. Their findings suggest that the overall genetic architecture of these two migraine types is similar.

Christensen et al. (10) combined genetic information about the migraine-associated SNPs into a genetic score and provided the first evidence that specific SNPs and an increased genetic load influence the patient’s response to migraine drugs. This provides a promising first step towards individualized medicine for migraine patients.

Eising et al. (11) combined the functional information of genes – that is, curated sets of synaptic genes and sets of genes belonging to various types of glial cells – with the IHGC’s genotype information of 5000 migraine patients and over 13,000 control subjects and showed an association of gene sets containing astrocyte- and oligodendrocyte-related genes with migraine. Their findings suggest that, at the gene level, both of the main migraine types, at least in part, seem to be different.

De Vries et al. (12) used IHGC’s genotype information to systematically re-evaluate the most relevant findings from published candidate gene association studies and other genetic studies, but in a much larger data set. Without exception, there was no statistical evidence for the involvement of the 27 candidate migraine genes in migraine, which not only casts doubt on the usefulness of selecting genes based on prior knowledge, but also shows that the independent replication of findings, as is common practice in GWASs, should be advised to avoid false-positive results.

Yang et al. (13) reviewed the available data from genetic studies investigating the bidirectional co-morbid association of migraine and depression. Although a number of claims have been made, it has to be concluded that, at present, no specific genetic variant with an unequivocal association with risk for both migraine and depression has been identified.

The review by Malik et al. (14) discusses the interesting link between migraine and vascular disorders. Evidence for shared mechanisms mainly come from studies that investigated monogenic vascular disorders in which migraine is prominent, such as cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy. Cross-phenotype studies have revealed a genetic overlap between migraine and ischaemic stroke and between migraine and coronary artery disease.

Gerring et al. (15) provide an interesting viewpoint on how gene expression profiles from blood can be used for the development of biomarkers of migraine. The original data obtained from 38 patients and 38 control subjects showed a clear potential as the gene expression data differed significantly between migraine patients and controls.

Chen et al. (16) review research that used transgenic mouse models, primarily models in which a human pathogenic gene mutation for one of the monogenic migraine disorders was introduced. They discuss new insights into migraine-relevant neurobiological mechanisms from gene discoveries made in the last two decades.

This special issue highlights new possibilities for genetic research that have become feasible in the GWAS era. Although much progress has been made, it is clear that we are only at the beginning of understanding what these new gene discoveries really mean and how they can be used in research and clinical practice. Future research that effectively mines the existing genetic information and combines this with the wealth of genetic information that will undoubtedly be obtained has all the elements to fundamentally change our insight into the pathophysiology of migraine and other headache disorders. Whether this genetic research will lead to practical use in predicting who will become a migraine patient and when a migraine will strike, or to novel treatment options, remains a great unknown.