Baseline amyloid burden and global cognitive change during a home-based tDCS protocol in late-life depression: an exploratory single-arm study

Abstract

Background Late-life depression often co-occurs with cognitive difficulty, but whether amyloid burden is associated with mood and cognitive change during transcranial direct current stimulation (tDCS) in older adults is unknown. Methods In this single-arm pre–post study, 41 older adults were enrolled; 3 withdrew before receiving the device, and 38 received home-based tDCS and completed baseline (V1), post-stimulation (V2), and follow-up (V3) assessments. Depression, insomnia, subjective cognition, MoCA, and a detailed cognitive battery were examined, and 31 change outcomes were regressed on Centiloid and 21 regional amyloid indicators with covariate adjustment. Results No participant met the ≥50% GDS response criterion, whereas 60.5% showed a ≥2-point MoCA gain; the absent depression response reflects a floor effect expected from the low baseline severity rather than inefficacy. No association survived Benjamini–Hochberg correction across the complete exploratory family. In a biologically motivated Centiloid-only analysis, however, greater Centiloid burden was associated with smaller MoCA gain, and this association survived correction in both the full-analysis and per-protocol sets. Phonemic-fluency findings were less stable, and no comparable FDR-supported association was detected for depression, insomnia, or subjective cognition. Conclusions The findings do not support a broad amyloid-related cognitive pattern, but they identify a focused, hypothesis-generating Centiloid–MoCA association during the protocol. Because the study was single-arm, neither the observed changes nor the amyloid association can be attributed to stimulation itself. Controlled studies with prespecified outcomes are required.