Poststroke aphasia significantly impacts the quality of life in older adults, yet the underlying neural mechanisms linking macro-scale network hierarchy and micro-scale molecular architecture remain unclear. This study investigated alterations of the principal functional connectivity gradient and their transcriptomic underpinnings in older adults with poststroke aphasia. We recruited 27 patients with aphasia and 29 age-matched healthy controls. Resting-state fMRI data were analyzed using diffusion map embedding to characterize the principal functional connectivity gradient. Patients exhibited a compressed gradient range, characterized by diminished differentiation in unimodal networks (visual and somatomotor) and disordered integration in multimodal networks, including the ventral attention network and the default mode network. These gradient alterations were significantly correlated with language deficits. Furthermore, partial least squares regression revealed that the spatial pattern of gradient changes was associated with normative gene expression profiles related to synaptic transmission, trans-synaptic signaling, and calcium ion binding. Machine learning models incorporating gradient features and lesion volume successfully predicted individual differences in language performance. These findings suggest that poststroke aphasia involves a disruption of the cortical functional hierarchy that is constrained by specific molecular architectures, providing novel insights into the neurobiological mechanisms of language recovery and potential targets for precision rehabilitation in aging populations.
Wang et al. (Sun,) studied this question.