Patients with migraine were prospectively recruited from two dedicated Neurology units (AOUI and IRCCS Sacro Cuore Don Calabria Hospital, Verona, Italy) from August 2023 to July 2024. Age-matched healthy controls (HC) were also included for comparison.

The study was approved by the local ethic committees (Prog. 4066CESC and Prog. 19CET), and informed consent was obtained from all patients and controls.

Inclusion criteria were (a) either presence (for patients) or absence (for HC) of EM or CM diagnosed according to the ICHD-3 [19], (b) age at sampling between 18 and 50 years old, (c) exclusion of other headache disorders including hemiplegic migraine, cluster headache or secondary headache disorders other than medication-overuse headache, (d) absence of other neurological or psychiatric comorbidities, and (e) absence of systemic conditions possibly influencing biomarkers levels (including kidney dysfunctions).

The fulfillment of inclusion/exclusion criteria for migraine patients was verified by two neurologists specialized in headache (FM and FR). Healthy controls were volunteers recruited by neurologists (SC, RT, and AP).

Demographic and clinical data including age, sex, comorbidities, attack presence at sampling, days from last attack, migraine frequency (monthly migraine headache days), mean severity of attacks during the previous month (collected through visual analogue scale -VAS-), migraine course (years), and presence of aura before attacks were collected in a dedicated database. Treatment strategies including previous preventive treatments, actual preventive treatments, attack medications, and history of medication overuse in the previous 2 years, as defined in previous studies [20] were also analyzed.

Blood samples were obtained and centrifuged for 10 min at 800 g at room temperature. Serum was stored within 2 h at − 80 °C until sample analysis. Investigators blinded to clinical data measured serum NfL and GFAP levels using SIMOA two-plex kit in SR-X immunoassay analyzer, Simoa (Quanterix Corporation, Billerica, Massachusetts, USA), which runs ultrasensitive paramagnetic bead-based enzyme-linked immunosorbent assays. The analysis was performed at the Neuropathology and Neuroimmunology Laboratory, University of Verona, Italy, according to manufacturer’s instructions. Briefly, 25 μL of frozen samples and calibrator was equilibrated to room temperature and diluted with specific sample diluent. Calibrators, samples, detector, and beads were dispensed in each well, and plates were incubated at 30 °C with shaking 800 rpm for 30 min. After washing steps, 100 μL SBG was added to each well, and plates were incubated at 30 °C with shaking 800 rpm for 10 min. After washing steps, beads were resuspended twice at 1000 rpm for 1 min. A final washing step was performed, and plates were dried for 10 min before being transferred to the SR-X reader [21, 22].

Sample size (migraine patients vs. HC) estimation was based on NfL and GFAP levels previously reported, assuming a clinically meaningful difference of 20%, a significance level of 0.05, and a statistical power of 80% [23, 24].

Descriptive statistics were performed using median (interquartile ranges [IQR]) and percentages for categorical variables). Group comparisons (migraine patients vs. HC and EM vs. CM) were assessed using non-parametric tests (χ2 and Mann–Whitney test), as appropriate. Correlation analyses between biomarkers and relevant clinical features were performed using a 2-tailed Spearman analysis with a Bonferroni correction for multiple comparisons. Analyses were performed using IBM SPSS 25; p values < 0.05 were considered statistically significant.