Introduction: The corpus callosum (CC) is essential for interhemispheric communication, and its abnormal integration is central to the neurobiology of schizophrenia (SCZ). SCZ patients have a 10-fold higher risk of cannabis use disorder (CUD) and about 20-35% show a lack or poor response to antipsychotics and are defined as treatment-resistant schizophrenia (TRS) Until now, no study has analyzed the morphology of the CC in TRS compared to healthy controls (HC) and non-TRS patients with and without CUD. Objectives: The aim of the study is to assess whether the diagnosis of psychosis, the response to antipsychotic treatment, and CUD can influence the volume of the CC. To achieve this, we used an innovative artificial intelligence program applied to MRI, which provides structural information on a single subject. Methods: We included 20 HC and 48 SCZ patients, of whom 14 were affected by TRS and 34 were non-TRS. Among the non-TRS group, 20 had CUD comorbidity (non-TRS-CUD+) and 14 did not have CUD (non-TRS-CUD-). All were assessed cross-sectionally through the Neurological Evaluation Scale, the Brief Assessment of Cognition in Schizophrenia, the Positive And Negative Syndrome Scale. We assessed them cross-sectionally using psychometric tools, cognitive tests. All patients underwent a brain MRI 1.5 T, for white matter volume group analysis, and MRI applied to Artificial Intelligence (MRI-AI-Pixyl.Neuro) for single-subjects analysis. Results: TRS was associated with higher PANSS total score (fig.1) and neurological soft signs (fig.2) and lower negative symptoms (trend) than non-TRS groups. The TRS group performs worse in the Tower of London task compared to non- TRS and HC groups. Only the condition of TRS is associated with a significantly smaller CC volume (64.28%) compared to HCs and non-TRS patients (Fig.3). Only one patient from the non-TRS-CUD- group showed a reduction in the volume of the CC like TRS patients.
Corpus Callosum Volume in Treatment-Resistant compared to Treatment Responsive Schizophrenia patients with and without cannabis use disorder: a Novel Artificial Intelligence Method Applied to Single-Subject Magnetic Resonance Imaging / Matrone, Marta; Romano, Andrea; Kotzalidis, Giorgio; Perrini, Filippo; Ciccarelli, Mariateresa; Vellucci, Licia; Barone, Annarita; Iasevoli, Felice; De Persis, Simone; de Bartolomeis, Andrea; Bozzao, Alessandro. - (2024). (Intervento presentato al convegno EPA 2025 – 33rd European Congress of Psychiatry tenutosi a Lisbona).
Corpus Callosum Volume in Treatment-Resistant compared to Treatment Responsive Schizophrenia patients with and without cannabis use disorder: a Novel Artificial Intelligence Method Applied to Single-Subject Magnetic Resonance Imaging
Marta Matrone;Giorgio Kotzalidis;Alessandro Bozzao.
2024
Abstract
Introduction: The corpus callosum (CC) is essential for interhemispheric communication, and its abnormal integration is central to the neurobiology of schizophrenia (SCZ). SCZ patients have a 10-fold higher risk of cannabis use disorder (CUD) and about 20-35% show a lack or poor response to antipsychotics and are defined as treatment-resistant schizophrenia (TRS) Until now, no study has analyzed the morphology of the CC in TRS compared to healthy controls (HC) and non-TRS patients with and without CUD. Objectives: The aim of the study is to assess whether the diagnosis of psychosis, the response to antipsychotic treatment, and CUD can influence the volume of the CC. To achieve this, we used an innovative artificial intelligence program applied to MRI, which provides structural information on a single subject. Methods: We included 20 HC and 48 SCZ patients, of whom 14 were affected by TRS and 34 were non-TRS. Among the non-TRS group, 20 had CUD comorbidity (non-TRS-CUD+) and 14 did not have CUD (non-TRS-CUD-). All were assessed cross-sectionally through the Neurological Evaluation Scale, the Brief Assessment of Cognition in Schizophrenia, the Positive And Negative Syndrome Scale. We assessed them cross-sectionally using psychometric tools, cognitive tests. All patients underwent a brain MRI 1.5 T, for white matter volume group analysis, and MRI applied to Artificial Intelligence (MRI-AI-Pixyl.Neuro) for single-subjects analysis. Results: TRS was associated with higher PANSS total score (fig.1) and neurological soft signs (fig.2) and lower negative symptoms (trend) than non-TRS groups. The TRS group performs worse in the Tower of London task compared to non- TRS and HC groups. Only the condition of TRS is associated with a significantly smaller CC volume (64.28%) compared to HCs and non-TRS patients (Fig.3). Only one patient from the non-TRS-CUD- group showed a reduction in the volume of the CC like TRS patients.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.