¹University of Naples "Federico II", Department of Neuroscience, Reproductive and Odontostomatological Sciences, Naples, Italy, ²University of Naples "Federico II", Department of Advanced Biomedical Sciences, Naples, Italy, ³IRCCS, SYNLAB SDN, Naples, Italy

Background: Treatment-resistant schizophrenia (TRS) affects almost 30% of patients with the disorder. This condition is associated with functional impairment and has been suggested to have a putatively different neurobiology compared to treatment-responsive schizophrenia. One striking feature of TRS is the presence of relevant cognitive deficits, which, in many cases, resemble those of Mild Cognitive Impairment (MCI) associated with amyloid-β (Aβ) pathology. Moreover, MCI can overlap at certain points with the trajectory of schizophrenia.

Aims & Objectives: This study investigated whether TRS patients with cognitive deficit showed brain Aβ pathology similar to that observed in subjects with cognitive impairment in the Alzheimer’s Disease spectrum using ¹⁸(F)-Flutemetamol Positron Emission Tomography (PET), examining potential associations between cortical Aβ load and clinical symptoms.

Method: TRS patients (n = 12) were categorized based on cognitive deficits into MCI (n = 5) and Severe Cognitive Impairment (SCI) (n = 7). All subjects underwent a PET scan, with images acquired 90 to 110 min post-injection. Neuroimaging data were analyzed using the Diagnosis ORIented ANalysis (DORIAN) software, which provides a single averaged Integrated score as a proxy of regional cortical Aβ load, based on three semi-quantitative parameters: Standardized Uptake Value region (SUVr), Effectively Linear Binding Affinity (ELBA), and Time Delayed ratio (TDr). Clinical evaluation included the Positive and Negative Syndrome Scale (PANSS) and the Brief Assessment of Cognition in Schizophrenia (BACS). Independent-sample Student t-test, one-way Analysis of Variance (ANOVA), and Chi-square test were used to compare quantitative and categorical data for demographic and clinical features (P value ≤ 0.05). Spearman’s correlation analyses were performed to evaluate the association between the whole brain and regional Integrated score values, PANSS scores, and the BACS composite t-score. The associations were also tested after adjustment for chlorpromazine equivalents and age. A P value < 0.05 was considered significant for group comparison, and a P value < 0.01 for correlation analysis, with post-hoc analysis using Bonferroni correction (P value ≤ 0.005).

Results: The t-test analysis between the two groups of TRS patients (MCI vs. SCI) revealed no significant differences in sex, age, chlorpromazine equivalent, duration of illness, age of onset, or education. The scores of the PANSS Positive and Disorganization factor were statistically significantly higher in the SCI than the MCI group. In contrast, the BACS composite t-score was statistically significantly lower in patients with SCI. PET images qualitative and semi-quantitative analysis reveal no significant Aβ cortical radiotracer “binding” difference between MCI and SCI subgroups. Spearman’s correlation analyses revealed a significant negative association between the Aβ Integrated score values in the Right and Left Posterior Parietal Cortices and the PANSS Positive score (ρs=-0.78, P=0.01 and ρs=-0.84, P=0.005, respectively).

Discussion & Conclusions: This pilot study suggests that although increased cortical Aβ deposition is not evident in TRS patients, a correlation between amyloid pathology and psychotic symptoms may still be inferred. Future longitudinal study designs should assess how amyloid levels and cognitive function may evolve during the TRS patient's disease trajectory.