552. Can locus coeruleus structural and functional changes serve as a biomarker for Alzheimer's disease: a systematic review and meta-analysis

Abstract

Abstract Background The locus coeruleus (LC), the brain’s principal noradrenergic nucleus, is implicated early in Alzheimer’s disease (AD) pathophysiology. Growing evidence from animal models, human post-mortem studies, and in vivo neuroimaging suggests that LC structural and functional alterations may reflect AD-related neurodegeneration and could serve as a biomarker across the AD continuum. We aimed to systematically evaluate LC-related evidence in AD and to quantify group differences using meta-analysis restricted to LC neuroimaging outcomes. Aims & Objectives This systematic review aims to evaluate whether LC structural and functional changes can serve as a biomarker for AD, including diagnostic differentiation among AD, mild cognitive impairment (MCI), and healthy controls (HC). Method We systematically reviewed the literature in accordance with PRISMA guidelines. Eligible studies included: (1) animal studies evaluating LC-related AD pathology; (2) human post-mortem studies assessing LC degeneration and AD pathology; and (3) human neuroimaging studies examining LC structural and/or functional alterations in AD, MCI, and HC. Observational designs (cross-sectional, cohort, longitudinal) and randomized controlled trials involving LC-related interventions were eligible. Meta-analysis was conducted only for neuroimaging studies reporting quantitative LC measures (neuromelanin-sensitive MRI signal/contrast, LC volume, diffusion metrics, resting-state functional connectivity, PET-related indices), pooling standardized effect sizes under a random-effects model. Two reviewers independently screened records, extracted data, and resolved discrepancies by consensus. Risk of bias was assessed using appropriate tools based on study design. Results A total of 239 studies were included (animal: 88; post-mortem: 83; neuroimaging: 68). The neuroimaging meta-analysis comprised 24 studies. Across animal and post-mortem studies, convergent evidence supports LC vulnerability in AD, including LC-related pathological burden and degeneration with accompanying noradrenergic alterations. In neuroimaging meta-analysis, LC measures were substantially reduced in AD compared with HC (pooled SMD = −1.21; high heterogeneity), with the largest and most consistent effects observed for neuromelanin-sensitive MRI contrast/signal indices. In contrast, the pooled difference between MCI and HC was not significant overall (SMD = −0.08; very high heterogeneity), although some structural metrics (e.g., LC length and, to a lesser extent, volume) suggested potential early-stage change in subset analyses. Publication bias tests indicated possible small-study effects in AD–HC comparisons, while bias signals were not evident for MCI–HC. Discussion & Conclusions LC integrity measured by neuroimaging, particularly neuromelanin-sensitive MRI, shows strong promise as a biomarker for AD, supported by complementary animal and post-mortem evidence. However, substantial methodological heterogeneity and potential small-study effects underscore the need for standardized LC acquisition/segmentation and larger, longitudinal cohorts to establish robust diagnostic and prognostic utility.