Bharati M. Sanjanwala

Introduction: Migraine is a prevalent and disabling neurological disorder with subtypes mainly classified and managed based on monthly headache frequency. This study investigates a novel, data-driven clustering approach that uses comprehensive clinical features in addition to headache frequency to further understand migraine subtyping and management. Methods: This cross-sectional observational study included participants with a diagnosis of migraine based on self-reported surveys recruited from the Stanford Headache Clinic and surrounding community between 2012 and 2020. This study represents a secondary analysis of a previously collected cohort, incorporating newly added episodic migraine (EM) data alongside existing clinical and demographic variables. This approach extends prior work to refine data-driven migraine subtyping. Participants completed comprehensive surveys capturing clinical, demographic, and treatment-related information. A data-driven clustering approach was applied using hierarchical agglomerative clustering. To determine the optimal number of clusters, visual inspection of the dendrogram, scree plot, and principal component analysis was used. Descriptive statistics and logistic regression models were used to explore clinical characteristics and medication outcomes for each cluster. Results: 180 participants were included in the clustering analysis. Two distinct clusters were identified and compared. Cluster 1 ( n = 66; chronic migraine (CM) = 62; EM = 4) had a significantly ( p ≤ 0.001) higher migraine frequency, greater migraine-related disability, higher pain catastrophizing, and more comorbidities (including anxiety, depression, sleep disorders, post-traumatic stress disorder, and somatic symptoms). Cluster 2 ( n = 114; CM = 68; EM = 46) had significantly ( p ≤ 0.001) more years with headache and higher levels of pain self-efficacy. Treatment response differed significantly between clusters for acute medications, with Cluster 2 showing more favorable response rates to triptans and non-steroidal anti-inflammatory drugs ( p ≤ 0.001) and a higher proportion of individuals who indicated that acute medication helped to improve their migraine compared to Cluster 1. Conclusion: This study supports the potential use of a comprehensive phenotype-based clustering approach to determine migraine subtypes that correlate with acute treatment outcome.

Background: Menstrual-related migraine (MRM) is a subtype of migraine that significantly impacts women. Despite its high prevalence, MRM remains under-researched compared to other migraine types. This study investigates the phenotypic characteristics and treatment outcomes of MRM compared to non-menstrual migraine (non-MRM). Methods: This cross-sectional study included women aged 18 years or older with a diagnosis of migraine and an established menstruation pattern. Informed consent was properly obtained from all participants from the Stanford Headache Clinic and surrounding community between 2014 and 2024 prior to participation in the study. Data were collected from self-reported surveys, including demographic information, headache characteristics, physical and mental health status, and treatment outcomes from the Stanford Headache Clinic and surrounding community. Participants were then categorized into MRM and non-MRM groups. An exploratory analysis was performed to investigate phenotypic features, treatment outcomes, and brain volume through magnetic resonance imaging (MRI) study for each group. Statistical analyses included descriptive statistics, analysis of covariance (ANCOVA), and logistic regression. Results: A total of 84 subjects were included, with 38 in the MRM group and 46 in the non-MRM group. The mean age of participants was 34 (SD = 9) years. The MRM group (mean age 38 ± 9) showed significantly higher prevalences of recent change in migraine frequency, migraine attacks occurring in the same site, prodromal signs, speech aura, nausea, and vomiting associated with migraine, and a higher prevalence of reported family history of headache as compared with the non-MRM group (mean age 31 ± 10) (p < 0.05). MRM group also showed significantly higher rate of migraine improvement following the use of triptans and the overall acute and preventive medications (p < 0.05). No significant differences were found in brain volumes between the groups (p > 0.05). Conclusion: The MRM group was associated with significantly more bothersome symptoms associated with migraine, higher family history of headache, and more favorable response to overall acute medication especially triptan as compared with the non-MRM group.

Background: Integrating brain structure and function may help characterize neurobiological heterogeneity in migraine alongside symptom presentation. Aim: To apply a multimodal, exploratory, data-driven approach to identify migraine subgroups using structural and functional MRI, and to describe the clinical characteristics of the resulting subgroups. Methods: Resting-state functional connectivity (FC) across cortical and subcortical regions, along with structural measures including cortical thickness, cortical volume, and subcortical volumes, were extracted from 111 individuals with migraine (75 chronic, 36 episodic) classified according to ICHD-3 criteria. After dimensionality reduction using principal component analysis, hierarchical agglomerative clustering was applied to identify multimodal imaging-derived subgroups. For comparison, secondary unimodal clustering models were constructed using functional-only and structural-only feature sets. The optimal number of clusters was determined using silhouette coefficients, and clustering concordance across models was quantified using the Adjusted Rand Index (ARI). Group differences in clinical characteristics, FC, and cortical and subcortical structure were assessed using covariate-adjusted statistical models with false discovery rate (FDR) correction. Results: Multimodal clustering identified two subgroups with distinct clinical and imaging profiles, Migraine Cluster 1 (M1 _{f + s} ) and Migraine Cluster 2 (M2 _{f + s} ). M2 _{f + s} showed older age, longer disease duration, greater migraine disability, widespread increases in cortical-subcortical FC (including Dorsal Attention, Somatomotor, and Visual networks), and reduced cortical volumes across frontal, parietal, temporal, and insular regions compared with M1 _{f + s} . This subgroup also exhibited increased connectivity relative to controls. In contrast, M1 _{f + s} showed preserved cortical structure and stronger Control-network–subcortical connectivity compared to M2 _{f + s} , and no significant functional or structural deviations from controls. Unimodal analyses revealed that Functional-only clustering aligned moderately with the multimodal cluster solution (ARI = 0.427), showing that FC was a primary determinant of the multimodal cluster structure, whereas structural-only clustering showed negligible overlap (ARI = 0.001), reflecting an orthogonal dimension of heterogeneity captured by structural variation. Conclusion: Data-driven multimodal neuroimaging-based clustering in migraine identified two subgroups with distinct clinical and imaging patterns, highlighting heterogeneity and providing a framework for further investigation of imaging-informed characterization.