Cerebrospinal fluid biomarkers in primary headache disorders

The object of this review is to examine the published literature for cerebrospinal fluid laboratory measures of primary headache disorders to identify biomarkers and provide recommendations for future biomarker discovery.

Biomarkers may distinguish deviation from a normal state, provide insight into mechanisms of pathophysiology, quantify the degree of change, discriminate what may be clinically overlapping disorders, and allow monitoring and/or selection of specific treatment. High-throughput, discovery technologies fuel the ability to reveal more biomarkers than past hypothesis-driven studies. DESIGN OR METHODS: Publications were identified in PubMed, ISI web of knowledge (both Web of Science and BIOSYS), and SciFinder, using the key words for cerebrospinal fluid (CSF) and migraine, headache, or biomarkers. Additional references were sought from the papers identified in these searches. Data were assessed relating to all primary headache types for clinical and scientific methods and results.

Fifty-five out of 82 biomarkers were found from 55 publications, though none have been validated for clinical utility. Data for site (ventricular, cervical, lumbar) and timing of CSF collection, headache state, and diagnostic description were patchy, and controls were often poorly defined. Most routinely performed CSF measurements were within normal limits. Most levels of pain-related molecules were reduced, and concentrations of most neurotransmitters, neuropeptides, proteins, and small molecules were increased. Though few studies assessed the specificity of biomarkers for primary headaches, it is clear that there are differences in CSF biomarkers between migraine, cluster headache, tension-type headache, and trigeminal neuralgia.

The high proportion (67%) of biomarkers identified from laboratory measures tested thus far predicts that many more biomarkers will be identified for primary headaches when more candidates are evaluated. In order to discover and evaluate more biomarkers, especially those that may have clinical application for headache management, 3 recommendations are encouraged: prospective design of care-independent studies; evaluation of more clinical variables; and evaluation of substantially more candidates by using discovery-based research methods. Outlines of approaches to pursue these aims are proposed.