The Effect of Weight Loss on Brain Age in Schizophrenia

662 

Abstract Submission 

Vittal Korann¹, Nicolette Stogios², Bjørn Ebdrup³, Margaret Hahn², Mahavir Agarwal²

¹University of Toronto, Toronto, ontario, ²University of Toronto, Toronto, Ontario, ³University of Copenhagen, Copenhagen, Copenhagen

Vittal Korann  
University of Toronto
Toronto, ontario

Nicolette Stogios  
University of Toronto
Toronto, Ontario

Bjørn Ebdrup  
University of Copenhagen
Copenhagen, Copenhagen

Margaret Hahn  
University of Toronto
Toronto, Ontario

Mahavir Agarwal  
University of Toronto
Toronto, Ontario

Individuals with schizophrenia (SCZ) often have metabolic comorbidities, such as type 2 diabetes, and experience a reduced life expectancy due to cardiovascular diseases. Obesity, a common comorbidity in SCZ, can negatively affect brain health. However, there is limited understanding of how metabolic disorders impact brain structure in individuals with SCZ, and the effects of weight changes following pharmacological interventions have not been explored. In this study, we will investigate changes in brain morphology, specifically brain-age, in overweight or obese individuals with or without diabetes who have been diagnosed with SCZ. Our primary objective will be to assess these changes before and after a 12-week period of pharmacological treatment targeting metabolic dysfunction. We hypothesized that 1) a change in BMI will be positively associated with the change in brain age between baseline and endpoint; 2) there will be no significant difference in the strength of the correlation between the medication and placebo groups. As exploratory analyses, we also looked at the association between brain age and cognition and metabolic parameters.

This analysis includes 48 participants, aged 18 to 65, from three double-blind studies investigating interventions for antipsychotic-induced metabolic dysfunction: TAO study (NCT01794429, 9 on medication and 8 on placebo), Metformin for prediabetes/diabetes study (NCT02167620, 11 on medication and 8 on placebo), and Topiramate in clozapine study (NCT02808533, 12 on medication). In brief, inclusion criteria include patients with a diagnosis of schizophrenia, schizoaffective disorder, or bipolar disorder, metabolic comorbidity of prediabetes or type 2 diabetes, and BMI of above 25 kg/m2. We collected brain structural MRI, metabolic measures, cognition data, and body mass index at baseline and week 12. We utilized a convolution neural network-based classifier that was trained and tested to estimate the brain age of each participant using high-quality brain anatomical T1 image. The study aims to examine the changes in BMI and estimated brain age scores using baseline and endpoint data.

The BMI alteration demonstrated statistical significance within the whole sample (p < 0.001), as well as in both the medication (p = 0.005) and placebo groups (p = 0.008). Likewise, significant changes were found only in total and HDL (high-density lipoprotein) cholesterol levels across all three groups in terms of metabolic parameters. However, none of the groups exhibited any substantial changes in psychopathological scores or cognitive data between the baseline and endpoint assessments. Multiple regression analysis revealed a positive correlation between BMI change and alterations in brain age for the whole sample (beta = 0.263; t = 1.85; p = 0.05) and the medicated group (beta = 0.372; t = 2.12; p = 0.04), but not in the placebo group (beta = -0.106; t = -0.40; p = 0.69). However, no significant difference in the correlation strength was observed between the medication and placebo groups (p = 0.12). Furthermore, there was no significant association between changes in brain age and metabolic indicators such as total and HDL cholesterol. Lastly, no significant correlation was found between brain age and cognition.

In conclusion, our study showed a link between brain health (as assessed by delta brain age) and significant weight loss by anti-diabetic medication in patients with SCZ and comorbid obesity. These findings imply that large and extended weight loss, together with general improvements in cardiometabolic alterations, can prevent obesity-related abnormalities in brain health.

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

Aging

Physiology, Metabolism and Neurotransmission Other ²

Cognition

Machine Learning

Schizophrenia

STRUCTURAL MRI

Other - body mass index