Brain network localization of treatment resistance in schizophrenia

Background: Treatment-resistant schizophrenia (TRS) is associated with poorer clinical and functional outcomes than its non-TRS counterpart. A more detailed characterization of shared and distinct neuropathological signatures between TRS and non-TRS is urgently needed to guide prevention and intervention strategies for TRS. The objective of this study was to examine brain network localization of TRS and non-TRS. Methods: We first conducted a systematic review of published neuroimaging literature to identify brain regions exhibiting structural and functional differences across TRS, non-TRS, and healthy controls (HC). By integrating these affected regions with large-scale discovery (N = 1113) and validation (N = 1093) resting-state functional MRI datasets, we then applied functional connectivity network mapping to construct 3 specific brain networks. Results: The TRS vs. HC and non-TRS vs. HC networks commonly involved the canonical default and frontoparietal networks, suggesting their deficits as a shared feature inherent to schizophrenia regardless of treatment response status. Concurrently, the TRS vs. HC network additionally implicated the subcortical and limbic networks, while the non-TRS vs. HC network uniquely affected the visual network. Remarkably, the TRS vs. non-TRS network most prominently involved the ventral attention network, highlighting the critical importance of its dysfunction in treatment resistance in schizophrenia. Conclusion: Our findings facilitate a better understanding of the commonalities and differences in neuropathology between TRS and non-TRS, potentially informing the early identification and effective treatment of TRS.