Longitudinal brain age in first-episode mania youth treated with lithium or quetiapine

Abstract

It is unclear if lithium and quetiapine have neuroprotective effects, especially in early stages of bipolar and schizoaffective disorders. Here, an age-related multivariate brain structural measure (i.e., brain-PAD) at baseline and changes in response to treatment after a first-episode mania (FEM) were examined. FEM participants were randomized to lithium (n=21) or quetiapine (n=18) monotherapy. T1-weighted scans were acquired at baseline, 3-months (FEM participants only) and 12-months. Brain age predictions for healthy controls (n=29) and young people with bipolar or schizoaffective disorder (15–25 years) were derived using a deep learning model trained on one of the largest datasets (N=53,542) to date. Notably, a higher brain-PAD value (predicted brain age - age) signifies an older-appearing brain. Baseline brain-PAD was higher in young people with FEM compared to controls (+1.70 year, p=0.04; Cohen's d=0.53 [SE=0.25], CI 95% [0.04 to 1.01]). However, no significant effects of time or treatment group, nor an interaction between the two, were observed throughout the course of the study. Baseline brain-PAD did not predict any change in symptomatic, quality of life or functional outcome scores over 12 months. In young individuals with FEM, baseline findings show their brains appeared older than controls. However, brain-PAD remained stable over time across treatment groups and neither baseline values nor treatment predicted 12-month outcomes. A longer follow-up with a larger sample is warranted to determine if treatment effects emerge later in bipolar and schizoaffective disorders.

Trial registration

Australian and New Zealand Clinical Trials Registry – ACTRN12607000639426.