PO-50 | Enhanced integration and synergy in EEG- and fNIRS-based networks in migraine patients following three-month galcanezumab treatment: associations with long-term clinical effects

Giulia Paparella,^1,2 Marianna La Rocca,³ Livio Clemente,¹ Emmanuella Ladisa,¹ Chiara Abbatantuono,¹ Elena Ammendola,¹ Marianna Diletta Delussi,⁴ Katia Ricci,¹ Giusy Tancredi,¹ Antonio Lacalamita,³ Alfonso Monaco,³ Nicola Amoroso,⁵ Sebastiano Stramaglia,³ Roberto Bellotti,³ Marina de Tommaso¹ | ¹Neurophysiopathology Unit, Department of Translational Biomedicine and Neuroscience (DiBraiN), University of Bari Aldo Moro, Bari, Italy; ²IRCCS Neuromed, Pozzilli (IS), Italy; ³Interateneo Department of Physics "M. Merlin", University of Bari Aldo Moro, Bari, Italy; ⁴Department of Education, Psychology, Communication (For.Psi.Com.), University of Bari Aldo Moro, Bari, Italy; ⁵Department of Pharmacy–Drug Sciences, University of Bari Aldo Moro, Bari, Italy

Background: Galcanezumab (GCA) is a well-tolerated monoclonal antibody developed for migraine prophylaxis. Although its mechanisms of action are not yet fully understood, they may involve changes in cortical connectivity observed in migraine patients. This study further investigates the central changes induced by GCA by examining the metabolic-electrical brain interaction and network tendencies.

Methods: Twenty migraine patients and a control group of ten healthy subjects were enrolled. A multimodal approach was used, combining electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS), with high-order connectivity analysis obtained through Partial Information Decomposition (PID). Patients underwent clinical and EEG/fNIRS recordings at baseline (B), one hour after the first GCA administration (B’), and after three months of treatment (3M). Clinical outcomes were reassessed after one year (1Y). EEG and fNIRS signals were preprocessed and harmonized, and global network properties (strength, global efficiency, and clustering coefficient) were computed.

Results: Compared to controls, patients showed reduced strength and clustering coefficients across all time points for both EEG and fNIRS (p<0.05), with no significant changes over time. Global efficiency was significantly lower in patients at B and B’ compared to controls and patients at 3M. Synergy at 3M increased significantly in the fNIRS data, though not significantly in the EEG data. However, cluster analysis of synergy matrices over time showed an expansion of high-synergy channel clusters at 3M for both modalities. Synergy at 3M significantly correlated with clinical outcomes at 1Y, including the Migraine Disability Assessment and the monthly number of acute medications (p<0.05).

Conclusion: The results reveal distinct neurophysiological changes following three months of GCA treatment, including increased synergy and the expansion of high-synergy clusters in EEG and fNIRS networks. These findings may reflect a specific anti-nociceptive effect of GCA associated with enhanced functional integration. The association with long-term clinical improvement suggests that the modulation of early network properties alterations due to GCA may contribute to therapeutic outcomes.