The ‘theory of everything’ for migraine pathophysiology will never exist: Insights from glial cells

Over the years, an emerging interest has grown in studying the role of neuroinflammation in migraine pathophysiology, attracting the interest of scientists worldwide. Preclinical evidence showed the contributions of glial cells in sustaining neuroinflammation, supporting their role in sensitization of the pain processing system and initiation of migraine attacks (1). Several human neuroimaging studies showed signs of cortical and meningeal inflammation in migraine patients, particularly in patients with aura (2,3) with features of neuroimmune activation in different nociceptive processing areas (4). Additionally, an elegant and recent combined 7-Tesla magnetic resonance imaging/positron emission tomography study showed a link between altered trigeminal nerve microstructure, brainstem activity and neuroinflammation (5), supporting a linkage between neuroinflammation and peripheral and central factors.

However, despite great interest and increasing efforts, the role of neuroinflammation in migraine pathogenesis is not yet completely understood. Glial fibrillary acidic protein (GFAP) and neurofilament light chain (NfL) are potential biomarkers that reflect underlying glial and neuronal alterations, potentially unravelling neuroinflammatory processes (6). NfL is a neuronal cytoplasmic protein highly expressed in large myelinated axons and is a sensitive indicator of axonal injury. On the other hand, GFAP is predominantly expressed in astrocytes and reflects glial activation and potential blood–brain barrier compromise. Although they were found to be elevated in several neurological disorders (6), few studies on migraine patients showed negative results (7,8).

A recent cross-sectional study by Gozalov et al. (9), published in the last issue of Cephalalgia, contradicted some of these results. This study represents the most extensive study of GFAP and NfL in migraine patients. Gozalov et al. (9) investigated a mixed cohort of 603 migraine patients, including chronic and episodic patients with and without aura, and 154 healthy controls (HC). No differences emerged in the NfL concentrations between migraine patients and HC, confirming that migraine does not involve axonal damage. By contrast, they found a modest but statistically significant increase in the serum GFAP in the migraine cohort, which had higher (6.8%) serum GFAP concentrations. This modest increase was hypothesized to reflect persistent, low-grade glial activation at multiple sites of nociceptive messaging, which may exacerbate migraine pain by modulating the excitability of peripheral and central pain-processing neurons. However, following this intriguing hypothesis, one would have expected higher levels of GFAP in chronic than in episodic migraine patients because the nociceptive signaling should be more activated in patients with high-frequency attacks. By contrast, neither biomarker concentration was influenced by any clinical characteristics related to the degree of pain system activation, including the headache frequency, disease duration or headache status at the time of blood sampling.

At this point, a careful reader could point out that most evidence in this editorial suggests a link between neuroinflammation and migraine with aura, leading to an expectation of elevated GFAP levels in such patients. However, no differences were found between patients with and without aura. Gozalov et al. (9) propose that the slight GFAP increase may represent a stable migraine trait unaffected by disease subtype and progression.

Taking all these factors into account, would it be unreasonable to consider this modest GFAP increase as a secondary effect of the migraine process? Indeed, although it is easier to envision how neuroinflammation can play a role in migraine chronification (1), it is difficult to explain how neuroinflammation can adapt to the dynamic nature of the episodic forms and explain the cyclical brain changes that start days before the headache phase (10).

Although thought-provoking, we do not have answers to our questions. The desire to find a unique theory is part of human nature and the main aim of research. However, in this complex interplay between central and peripheral actors, in the harmony of this intricate imbalance, we may start considering that a ‘theory of everything' is no longer possible to explain one of the most dynamic and fascinating disorders in human pathology. Longitudinal studies evaluating intra-individual variations in these biomarkers across different migraine phases may elucidate whether glial cells are contributors or observers in migraine pathophysiology. This understanding could provide valuable insights into migraine mechanisms and potentially lead to novel therapeutic approaches.