Brain age is associated with cognitive improvement following repetitive transcranial magnetic stimulation in...

Late-life depression (LLD) is commonly accompanied by cognitive impairment and poor response to first-line antidepressants. Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for treatment-resistant LLD, yet neurobiological predictors of cognitive and clinical outcomes remain unclear. Brain aging, quantified via magnetic resonance imaging (MRI)-derived centile scores from normative brain charts, may capture individual variability in neurostructural aging relevant to rTMS treatment response.

We examined relationships between centile scores, cognition, and antidepressant outcomes in older adults with LLD (n=112) and healthy controls (HC; n=41) who participated in a noninferiority clinical trial of bilateral dorsolateral prefrontal cortex (DLPFC) rTMS versus theta burst stimulation. Structural MRI was obtained before and after treatment. Centile scores leveraging cortical and subcortical tissue volumes were generated using BrainChart.io. Cognitive outcomes were assessed using standardized neuropsychological measures, including verbal learning and executive function.

Baseline centile scores did not differ between the LLD cohort and HCs and were unrelated to depression severity or antidepressant response. However, participants in a cognitively-diminished LLD subtype exhibited significantly lower centile scores driven by reduced white-matter volume. Across LLD participants, higher baseline centile scores predicted greater improvement in verbal learning following rTMS, but not global cognition or executive function, independent of antidepressant response. Centile scores remained stable post-treatment.

Brain age measures reflect cognitive, but not antidepressant, outcomes following rTMS in LLD. Structural-MRI based centile scores may serve as biomarkers of cognitive function and predictors of rTMS-related cognitive improvement in individuals with LLD.