DIFFERENTIAL PATTERNS OF BRAIN METABOLISM IN STABILIZED, ACTIVELY SYMPTOMATIC TREATMENT RESISTANT VS. TREATMENT RESPONDER SCHIZOPHRENIA PATIENTS: A STUDY BY 18F-FDG-POSITRON EMISSION TOMOGRAPHY AND STRUCTURAL 3T MRI

Abstract Background Schizophrenia is a disabling psychiatric disease with a lifetime prevalence of 1%, which usually starts in late adolescence or initial adulthood and has a rare possibility of recovery. According to the extent of antipsychotic response, schizophrenia individuals can be divided into treatment resistant schizophrenia (TRS) and Treatment responders (non-TRS). Currently, it is unclear whether these categories belong to different biological and prognostic groups. Therefore, in order to deepen the knowledge of these conditions, we analyzed the cerebral metabolism of a sample of stabilized but still actively symptomatic TRS and non-TRS patients with substantial cognitive deficits. Methods We recruited 35 schizophrenia patients (18 TRS and 17 non-TRS) among approximately 80 consecutive non-affective psychotic patients, that were referred to our academic outpatient unit for supposed resistance to antipsychotics. The diagnosis of TRS was made by a structured diagnostic flowchart, according to published guidelines. Patients underwent a wide set of clinical and cognitive evaluations by trained raters. All patients who exhibited substantial cognitive deficits (at least two BACS/MATRICS items < than 1 - adjusted scores) and met inclusion/not met exclusion criteria were included in the neuroimaging study by 18F-fluorodeoxyglucose PET scans and structural 3T Magnetic Resonance Imaging. There were no statistically significant differences in age, chlorpromazine equivalents, and duration of illness between the TRS/non-TRS groups. Patients groups were matched with a control group. PET images were normalized on SPM template and then examined in visual and voxel-based analyses. Results Globally, the visual analysis of the images showed a diffuse relative metabolic reduction in the cerebral cortex with prevalent frontal involvement in the majority of the patients of the TRS group and to a lesser extent in the non-TRS group. The voxel-per-voxel analysis revealed an extensive cluster of significant relative metabolic reduction in TRS patients compared to controls and to non-TRS patients (13653 voxels) localized bilaterally in the frontal cortex with a mild left prevalence. In particular, the metabolic reduction peaks involved the upper right and left medial frontal girder (BA6 and BA8, respectively), left opercular frontal (BA44), middle frontal girth (BA10) bilaterally, left lower frontal girth (BA47). Notably, at least three patients from the TRS group were excluded from the initial pool due to the observation of previously unrecognized structural brain anomalies. Discussion In this study, we observed that TRS patients had a relative hypometabolic state in dorsal and medial prefrontal cortex, including frontobasal areas, compared to controls and to non-TRS patients, while the striatum and the occipital cortex appeared relatively preserved. In non-TRS patients, brain distribution of 18F-FDG was more homogeneous and only slightly reduced in the cerebral cortex, as compared to controls. These patterns of brain metabolism cannot be associated with more severe cognitive dysfunctions in TRS patients since all patients included in the study had substantial cognitive impairments, irrespective of being TRS or non-TRS. Although it cannot be excluded that these differential patterns may stem from the lack of response to antipsychotics, our findings suggest distinct neurobiological alterations in TRS vs. non-TRS patients, which may include more relevant brain disconnections in non-responder ones.