This page was translated automatically from Turkish. Read the original

This page is an AI-assisted Turkish summary of the study presented at the 26th World Congress of Psychiatry. The study belongs to the presenting authors; please refer to the original source when citing it.

EV0916

Occipital and sensorimotor hypoconnectivity and structure–function decoupling as network biomarkers in treatment-resistant schizophrenia

OCCIPITAL–SENSORIMOTOR HYPOCONNECTIVITY AND FUNCTIONAL–STRUCTURAL DECOUPLING AS NETWORK BIOMARKERS OF TREATMENT-RESISTANT SCHIZOPHRENIA

Treatment-resistant schizophrenia is characterized by robust and reproducible within-network hypoconnectivity in occipital and right sensorimotor networks, together with a loss of structure–function coupling in the superior occipital network.

The aim was to identify MRI-based structural and functional biomarkers and their clinical correlates in treatment-resistant schizophrenia (TRS) using hypothesis-free whole-brain analysis. Four groups were compared: healthy controls (HC, n=43), drug-naïve first-episode schizophrenia (FESdn, n=54), first-episode schizophrenia in remission with oral risperidone (FESrm, n=21), and TRS (n=54). Resting-state fMRI (303 volumes) and high-resolution structural imaging were acquired using 3T MRI; the data were decomposed into 30 independent component networks (ICNs), and dual regression, permutation testing, structure–function coupling, and correlations with PANSS scores were examined. Consistent within-network hypoconnectivity was identified in occipital (superior, inferior, medial, and temporo-occipital) and right sensorimotor networks in TRS; significance was maintained after covariate adjustment, the use of alternative template ICNs, and validation in an external TRS sample (n=10). Effect sizes were largest in the bilateral temporo-occipital junction, with Cohen’s d=2.23 [1.76, 2.68] for TRS versus HC, followed by the inferior posterior occipital network, d=1.97 [1.52, 2.41], and the right sensorimotor network, d=1.06 [0.70, 1.42] (all p<0.001); effects were smaller in comparisons with FESdn and FESrm (d=1.45 and d=1.30, respectively, for the bilateral temporo-occipital junction). In the TRS group, higher resting-state functional connectivity in the inferior posterior occipital (r=-0.286, p=0.036) and bilateral temporo-occipital junction (r=-0.316, p=0.020) networks was associated with lower PANSS positive symptom scores. Structure–function coupling in the superior occipital network differed significantly between HC and TRS (z=2.14, p<0.05); significant positive coupling was observed in the right sensorimotor network in HC (r=0.44) and in the medial occipital (r=0.30) and superior occipital (r=0.28) networks in FESdn, whereas none was detected in any network in TRS. No significant differences in between-network resting-state functional connectivity or structural morphometry were found between TRS and the other groups; the most prominent contribution to regional hypoactivity outside the networks came from the bilateral cerebellar cortices. The findings offer a reproducible candidate network biomarker for identification and stratification, indicating disruption of posterior perceptual–predictive and sensory–cognitive integration networks in TRS, but require replication in larger cohorts.

All WCP 2026 summaries