Tracing the Footprints of Alzheimer’s in the Locus Coeruleus: A Longitudinal Investigation of LC-NE System Integrity and Cognitive Performance Across the AD Spectrum Using a Custom LC NM-MRI Processing Method

Abstract

Alzheimer’s disease (AD) is increasingly recognized as involving early degeneration of the locus coeruleus (LC), the brain’s primary source of norepinephrine and a key regulator of cognitive functions such as attention and memory. Despite its importance, the progression and consequences of LC deterioration have remained elusive due to challenges in reliably imaging this small brainstem nucleus in vivo. This thesis addresses that gap by employing neuromelanin-sensitive MRI (NM-MRI) and introducing a novel automated LC segmentation technique, the funnel-tip (FT) method, to quantify LC integrity across aging and AD stages. In the first set of experiments, we validated the FT approach using a cross-sectional cohort (n = 190) across the Alzheimer’s spectrum. The FT method demonstrated high agreement with manual tracing (ICC = 0.91), strong association with AD diagnosis, Braak stage, and cognition, and robust test–retest reliability (ICC = 0.82). Building on this, we conducted a longitudinal analysis in 266 older adults, including cognitively normal individuals, patients with mild cognitive impairment (MCI), and those with AD, followed over four years with annual NM-MRI, PET imaging for amyloid and tau, and neuropsychological testing. This study’s main findings were that the LC signal declined significantly in early AD (Aβ+, Braak stages 1-3), but remained stable in late stages of AD and healthy aging. Notably, Aβ-negative individuals at baseline exhibited an unexpected LC signal increase linked to early tau accumulation, suggesting a transient biological response before neuronal loss. LC signal was also positively correlated with cognitive performance, particularly in memory and object recognition tasks among AD patients. In conclusion, this work identifies LC degeneration as a dynamic, early feature of AD and introduces a scalable method for its in vivo assessment. The FT segmentation technique, combined with multimodal biomarkers, enhances our ability to track disease progression and opens new avenues for early diagnosis and intervention in AD.