2026 American Academy of Neurology Abstract Website

Weige Zhao¹, Alexander Pantelyat², Carmela Tartaglia³, David Coughlin⁴, Marian Dale⁵, Zoltan Mari⁶, Federico Rodriguez-Porcel⁷, Ruth Schneider⁸, Zbigniew Wszolek⁹, Tao Xie¹⁰, James Rowe¹¹, Lawrence Golbe¹², Anne Marie Wills¹
¹Massachussetts General Hospital, ²Johns Hopkins University School of Medicine, ³Toronto Western Hospital, University of Toronto, ⁴University of California San Diego, ⁵Oregon Health and Science University, ⁶Cleveland Clinic Lou Ruro Center for Brain Health, ⁷Medical University of South Carolina, ⁸University of Rochester, ⁹Mayo Clinic- Jacksonville, ¹⁰University of Chicago, ¹¹University of Cambridge, ¹²Rutgers Robert Wood Johnson Medical School

Objective:

To identify sociodemographic, lifestyle, environmental, and medical risk factors associated with diagnosis of incident progressive supranuclear palsy (PSP) in the UK Biobank.

Background:

PSP is a rare neurodegenerative disorder presenting with parkinsonism, eye movement and neurobehavioral abnormalities. Prior studies have suggested risk factors for PSP include lower education, tobacco and well water use, but these studies were limited by selection and recall bias due to their retrospective nature.

Design/Methods:

The UK Biobank recruited >500,000 participants from 2006 to 2010, followed prospectively for close to two decades (census date August 2025). We identified incident PSP cases and used an iterative logistic regression approach to identify premorbid risk factors compared to age- and sex-matched controls.

Results:

240 participants developed incident PSP over 18 years. In multivariate analyses, lower education level, smoking, and >1 cup daily coffee were significant environmental risk factors. Weekly alcohol consumption exhibited a J-shaped risk curve.

Medical history risk factors included BMI ≥25, epilepsy, autonomic dysfunction, protein calorie malnutrition, delirium, depression, and functional GI disorders. Cancer was inversely associated with PSP diagnosis (OR 0.56, 95% CI 0.40-0.79 in fully adjusted model). In lead time analysis, cancer, delirium and depression retained a significant relationship >5 years prior to PSP diagnosis, with depression remaining a risk factor >10 years prior. Pharmacotherapy exposures were not significant factors.

Conclusions:

Our findings include several novel risk factors for PSP diagnosis, however many of the medical history items identified may be early premorbid or premotor symptoms. Interestingly, cancer had a protective effect, similar to that seen in other neurodegenerative disorders. Alcohol intake had a J-shaped relationship, with higher risk in non-drinkers and heavy drinkers, that was not fully explained by the “sick quitter” effect. Unlike Parkinson’s disease, coffee and tobacco appear to be potential risk factors for PSP.