Author(s): TINNEY, E., BURTRAND, C., JASON, K., WARREN, A., MORRIS, T., SUTTON, B., KANG, C., HUANG, H., WAN, L., GROVE, G., ROGERS, R., JAKICIC, J., MARSLAND, A., BURNS, J., VIDONI, E., KARIKARI, T., MORRIS, J., MCAULEY, E., KRAMER, A., ERICKSON, K., HILLMAN, C., Institution: NORTHEASTERN UNIVERSITY, Country: UNITED STATES, Abstract-ID: 882

INTRODUCTION:
White matter microstructure is critical for cognitive function and vulnerable to both aging and Alzheimer's disease (AD) pathology, yet its responsiveness to exercise interventions remains poorly understood.
METHODS:
We report white matter findings from the Investigating Gains in Neurocognition in an Intervention Trial of Exercise (IGNITE; NCT02875301), a Phase 3, parallel arm, randomized clinical trial testing moderate-intensity aerobic exercise (AEx) in older adults. As a secondary aim of IGNITE, we examined whether pre-intervention AD-related pathology moderated intervention effects on white matter, addressing whether neuroprotective responses vary by underlying disease risk. Cognitively unimpaired, low-active adults (65-80 years) were randomized to one of three 12-month conditions: 150 min/week of light intensity stretching and toning (control), 150 min/week of moderate-intensity AEx, or 225 min/week of moderate-intensity AEx. Elevated AD pathology was defined using plasma phosphorylated tau-217 (pTau217; cutoff: 0.46 pg/ml). A total of 636 participants with all variables of interest (mean age=69.8; 71% female) were randomized. Intent-to-treat analyses included pre-specified covariates of age, sex, and site, and adjusted for COVID-19 exposure (binary indicator of any pandemic-year intervention weeks and continuous measure of weeks affected). White matter microstructure was assessed using fixel-based analysis, quantifying fiber density in six a priori tracts: right frontal premotor, right premotor, right precentral, corpus callosum body, right striatal, and right superior cerebellar peduncle.
RESULTS:
Across all groups, white matter fiber density declined over 12 months, but there were no significant differences in decline between groups. Baseline pTau217 levels significantly moderated the intervention effect on premotor, frontoparietal, precentral, superior thalamic radiation, and superior cerebellar peduncle tract fiber density. Among individuals with lower pTau217 (<0.46 pg/ml), both interventions resulted in similar modest declines in FD, with a slight numerical advantage for aerobic exercise. However, among those with higher pTau217 (≥0.46 pg/ml), a crossover pattern emerged where stretch-and-tone participants showed preservation or slight improvement in FD, while aerobic exercise participants continued to show decline.
CONCLUSION:
. These findings reveal an unexpected pattern: among individuals with elevated baseline AD pathology, stretching and toning was associated with white matter preservation while aerobic exercise showed continued decline. This interaction challenges assumptions about exercise intensity and suggests that the relationship between intervention type and white matter outcomes may depend critically on underlying neurodegeneration status. These results highlight the complexity of exercise responses in aging populations and highlight the need to consider individual AD biomarker profiles when designing and interpreting exercise intervention studies.