Aberrant functional connectivity (FC) of the brain regions, evaluated by functional magnetic resonance imaging (fMRI), affects clinical courses in inflammatory arthritis (IA). The static analysis methods would be simplistic to estimate the whole picture of resting-state brain function because blood oxygen level-dependent (BOLD) signals fluctuate over time. The effects of FC dynamics on clinical course are unknown in IA. Therefore, we aimed to evaluate dynamic FC for therapeutic responsiveness to biologics in IA patients. We analyzed resting-state fMRI data of 64 IA patients in 2 cohorts. Dynamic FC was derived as a correlation coefficient of the windowed BOLD signal time series. We determined representative whole-brain dynamic FC patterns by k-means++ cluster analysis, leading to 4 distinct clusters. In the first cohort, occurrence probability of the distinct cluster was associated with favorable therapeutic response in disease activity and patients' global assessment, which was validated by the second cohort. The whole-brain FC of the distinct cluster indicated significantly increased corticocortical connectivity, and probabilistically decreased after therapy in treatment-effective patients compared with -ineffective patients. Taken together, frequent emergence of corticocortical connections was associated with clinical outcomes in IA. The coherence of corticocortical interactions might affect pain modulation, possibly relevant to therapeutic satisfaction.
Keywords: dynamic functional connectivity; neuroimaging; resting-state functional magnetic resonance imaging; rheumatoid arthritis; spondyloarthritis.
© The Author(s) 2023. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.