Shannon Y. Lee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shannon Y. Lee has a total of 69 co-authors. Here are the top co-authors by number of shared publications:
- Breton Michael Asken
Department of Clinical and Health Psychology, University of Florida, Gainesville
2 shared publications Michael Marsiske
Department of Clinical and Health Psychology, University of Florida, Gainesville
2 shared publicationsAdam Staffaroni
Department of Neurology Memory and Aging Center Weill Institute for Neurosciences University of California San Francisco San Francisco California USA
1 shared publication- Alexander P. Lin
Center for Clinical Spectroscopy, Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA
1 shared publication Allison Holgerson
Department of Clinical and Health Psychology, University of Florida, Gainesville
1 shared publicationAnnalise E. Miner
Boston University CTE Center, Boston University Chobanian & Avedisian School of Medicine, MA
1 shared publication
Latest Publications
- 24 March 2026
- Published by Wolters Kluwer Health in Neurology
- Vol. 106 (6) , e214646
Inflammation, Limbic White Matter Microstructure, and Clinical Symptoms in Retired American Football Players With Repetitive Head Impacts
The link between repetitive head impact (RHI) exposure, later-life cognitive decline, and neurobehavioral dysregulation (NBD) is not well understood. Recent work has implicated inflammation and limbic dysfunction as relevant RHI correlates. Our goal was to integrate plasma and CSF inflammatory biomarkers, structural brain imaging, and clinical measures in former elite American football players to better understand reasons for RHI-related cognitive and neurobehavioral changes. Participants were from the Diagnostics, Imaging, and Genetics Network for the Objective Study and Evaluation of Chronic Traumatic Encephalopathy Research Project, which recruited male former college/professional football players with RHI and asymptomatic unexposed (UE) controls with no history of contact sports, military combat, or traumatic brain injury/concussion. Our study focused on plasma/CSF inflammatory biomarkers (interleukin [IL]-6, tumor necrosis factor [TNF]-α, glial fibrillary acidic protein), limbic white matter (WM) microstructure (diffusion tensor imaging: fractional anisotropy [FA], mean diffusivity [MD]), and clinical measures (memory, executive function, NBD). Hierarchical linear regressions assessed change in variance explained (ΔR2) among inflammation, WM, and clinical outcomes in former football players. Post hoc analyses tested whether associations differed by group (football vs UE; group interactions) or were stronger in football players considered at highest risk of CTE. Our sample included 223 men (n = 170 football players: age 57.2 ± 8.1 years, 33% non-Hispanic/Black; n = 53 UE participants: age 59.4 ± 8.6 years, 34% non-Hispanic/Black). In football players, higher inflammation was associated with lower limbic FA (plasma IL-6: ΔR2 = 0.03 [0.001-0.09], p = 0.03; CSF IL-6: ΔR2 = 0.03 [-0.01 to 0.11], p = 0.03; plasma TNF-α: ΔR2 = 0.05 [0.01-0.11], p = 0.003) and higher limbic MD (CSF IL-6: ΔR2 = 0.06 [0.007-0.15], p = 0.01). Inflammation was more strongly related to limbic WM microstructure in football players than in UE participants. Worse WM microstructure was associated with worse memory in football players (FA: ΔR2 = 0.05 [0.003-0.14], p = 0.007; MD: ΔR2 = 0.07, p = 0.003 [0.008-0.16]). Most of the observed associations were stronger in the CTE probable subgroup. There were no direct associations between plasma or CSF markers of inflammation and cognition. In former elite football players, elevated plasma and CSF inflammatory markers were associated with poorer limbic WM microstructure, which in turn related to worse cognition. Given the limbic system's role in cognition and behavior, inflammation may be a modifiable target for RHI-related neurodegeneration. Limitations include the cross-sectional design and limited generalizability to other contact sports, lower levels of play, female athletes, or other RHI sources.
- 24 March 2026
- Published by Wolters Kluwer Health in Neurology
- Vol. 106 (6) , e214646
Inflammation, Limbic White Matter Microstructure, and Clinical Symptoms in Retired American Football Players With Repetitive Head Impacts
The link between repetitive head impact (RHI) exposure, later-life cognitive decline, and neurobehavioral dysregulation (NBD) is not well understood. Recent work has implicated inflammation and limbic dysfunction as relevant RHI correlates. Our goal was to integrate plasma and CSF inflammatory biomarkers, structural brain imaging, and clinical measures in former elite American football players to better understand reasons for RHI-related cognitive and neurobehavioral changes. Participants were from the Diagnostics, Imaging, and Genetics Network for the Objective Study and Evaluation of Chronic Traumatic Encephalopathy Research Project, which recruited male former college/professional football players with RHI and asymptomatic unexposed (UE) controls with no history of contact sports, military combat, or traumatic brain injury/concussion. Our study focused on plasma/CSF inflammatory biomarkers (interleukin [IL]-6, tumor necrosis factor [TNF]-α, glial fibrillary acidic protein), limbic white matter (WM) microstructure (diffusion tensor imaging: fractional anisotropy [FA], mean diffusivity [MD]), and clinical measures (memory, executive function, NBD). Hierarchical linear regressions assessed change in variance explained (ΔR2) among inflammation, WM, and clinical outcomes in former football players. Post hoc analyses tested whether associations differed by group (football vs UE; group interactions) or were stronger in football players considered at highest risk of CTE. Our sample included 223 men (n = 170 football players: age 57.2 ± 8.1 years, 33% non-Hispanic/Black; n = 53 UE participants: age 59.4 ± 8.6 years, 34% non-Hispanic/Black). In football players, higher inflammation was associated with lower limbic FA (plasma IL-6: ΔR2 = 0.03 [0.001-0.09], p = 0.03; CSF IL-6: ΔR2 = 0.03 [-0.01 to 0.11], p = 0.03; plasma TNF-α: ΔR2 = 0.05 [0.01-0.11], p = 0.003) and higher limbic MD (CSF IL-6: ΔR2 = 0.06 [0.007-0.15], p = 0.01). Inflammation was more strongly related to limbic WM microstructure in football players than in UE participants. Worse WM microstructure was associated with worse memory in football players (FA: ΔR2 = 0.05 [0.003-0.14], p = 0.007; MD: ΔR2 = 0.07, p = 0.003 [0.008-0.16]). Most of the observed associations were stronger in the CTE probable subgroup. There were no direct associations between plasma or CSF markers of inflammation and cognition. In former elite football players, elevated plasma and CSF inflammatory markers were associated with poorer limbic WM microstructure, which in turn related to worse cognition. Given the limbic system's role in cognition and behavior, inflammation may be a modifiable target for RHI-related neurodegeneration. Limitations include the cross-sectional design and limited generalizability to other contact sports, lower levels of play, female athletes, or other RHI sources.
- Open Access
- 5 September 2025
- Published by Wiley in Alzheimer's & Dementia
- Vol. 21 (9) , e70626
Moderating effects of plasma glial fibrillary acidic protein along the Alzheimer's disease continuum
INTRODUCTION: Glial fibrillary acidic protein (GFAP) may contribute to Alzheimer's pathology at early disease stages. GFAP moderation of Alzheimer's disease (AD)‐related neurodegeneration and cognition is unclear. METHODS: We examined plasma GFAP moderation of AD biomarkers (amyloid beta [Aβ]‐positron emission tomography [PET][A]; plasma phosphorylated tau‐181 [p‐tau181][T 1 ]), neurodegeneration (plasma NfL[N plasma ]; structural magnetic resonance imaging [MRI][N MRI ]), and cognition (Cog memory ; Cog executive ) in two cohorts: University of California San Francisco (UCSF) ( N = 212, 91.0% non‐Hispanic/Latino White [NHLW], age = 74.7 [7.6] years, 75.9% cognitively unimpaired [CU]) and 1Florida Alzheimer's Disease Research Centers (1FLADRC; N = 582, 32.8% NHLW, age = 70.7 [8.5] years, 28.9% CU). RESULTS: Plasma GFAP consistently moderated A–T 1 (UCSF: β = 0.46, p = 0.012; 1FLADRC: β = 0.12, p = 0.029). The association between elevated Aβ‐PET and increased (p‐tau) was strengthened at higher GFAP concentrations. In 1FLADRC, GFAP moderated T 1 –N plasma/MRI. In UCSF, GFAP moderated T 1 –Cog memory/executive and N MRI –Cog memory/executive . Higher GFAP consistently related to worse neurodegeneration and cognition (main effects). DISCUSSION: Across demographically and clinically heterogeneous cohorts, plasma GFAP is a key moderator of AD and may help identify individuals at greatest risk of AD‐related neurodegeneration and cognitive decline. Highlights: AD biomarkers were measured in two demographically and clinically distinct cohorts. Plasma GFAP moderated Aβ‐PET to p‐tau associations in both UCSF and 1FLADRC. Cohort‐dependent, GFAP moderated p‐tau to neurodegeneration and cognition associations. All moderations revealed strengthened disease associations with higher plasma GFAP. Plasma GFAP may help identify individuals at greatest risk of AD‐related decline.
- 5 September 2025
- Published by Wiley in Alzheimer's & Dementia
- Vol. 21 (9) , e70626
Open AccessModerating effects of plasma glial fibrillary acidic protein along the Alzheimer's disease continuum
INTRODUCTION: Glial fibrillary acidic protein (GFAP) may contribute to Alzheimer's pathology at early disease stages. GFAP moderation of Alzheimer's disease (AD)‐related neurodegeneration and cognition is unclear. METHODS: We examined plasma GFAP moderation of AD biomarkers (amyloid beta [Aβ]‐positron emission tomography [PET][A]; plasma phosphorylated tau‐181 [p‐tau181][T 1 ]), neurodegeneration (plasma NfL[N plasma ]; structural magnetic resonance imaging [MRI][N MRI ]), and cognition (Cog memory ; Cog executive ) in two cohorts: University of California San Francisco (UCSF) ( N = 212, 91.0% non‐Hispanic/Latino White [NHLW], age = 74.7 [7.6] years, 75.9% cognitively unimpaired [CU]) and 1Florida Alzheimer's Disease Research Centers (1FLADRC; N = 582, 32.8% NHLW, age = 70.7 [8.5] years, 28.9% CU). RESULTS: Plasma GFAP consistently moderated A–T 1 (UCSF: β = 0.46, p = 0.012; 1FLADRC: β = 0.12, p = 0.029). The association between elevated Aβ‐PET and increased (p‐tau) was strengthened at higher GFAP concentrations. In 1FLADRC, GFAP moderated T 1 –N plasma/MRI. In UCSF, GFAP moderated T 1 –Cog memory/executive and N MRI –Cog memory/executive . Higher GFAP consistently related to worse neurodegeneration and cognition (main effects). DISCUSSION: Across demographically and clinically heterogeneous cohorts, plasma GFAP is a key moderator of AD and may help identify individuals at greatest risk of AD‐related neurodegeneration and cognitive decline. Highlights: AD biomarkers were measured in two demographically and clinically distinct cohorts. Plasma GFAP moderated Aβ‐PET to p‐tau associations in both UCSF and 1FLADRC. Cohort‐dependent, GFAP moderated p‐tau to neurodegeneration and cognition associations. All moderations revealed strengthened disease associations with higher plasma GFAP. Plasma GFAP may help identify individuals at greatest risk of AD‐related decline.