Abstract
METHODS
We included 2590 participants from the Rotterdam Study (2009-2015), free of dementia and prevalent cancer at baseline, with gait and cognitive assessment data to define MCRS and mild cognitive impairment (MCI). Cox proportional Hazard models examined the relation of MCRS, MCI, global gait, and individual gait domains with risk of cancer, adjusting for covariates.
CONCLUSIONS
Both individuals with MCRS and individuals with MCI tended to have an increased risk of cancer, supporting the notion of shared mechanisms between neurodegeneration and cancer. The association with gait impairment was most pronounced in the years preceding diagnosis, suggesting subclinical disease may partly underlie this link. These findings highlight motor-cognitive prodromes as informative phenotypes for studying shared vulnerability across neurodegenerative and oncological diseases.
RESULTS
Individuals with MCRS tended to have an increased risk of cancer (HR: 1.26, 95% CI: 0.88-1.80), primarily driven by slow gait. Similarly, individuals with MCI tended to have an increased risk of cancer (HR: 1.29, 95% CI: 0.82-2.04), primarily driven by objective cognitive impairment. Worse global gait conveyed a slightly increased risk of cancer (standardised HR: 1.10, 95% CI: 0.99-1.22), driven by Rhythm, Phases and Variability, with a stronger effect when follow-up was truncated at 3 years (standardised HR: 1.22, 95% CI: 1.03-1.45).
BACKGROUND
Findings on the relation between dementia and cancer have been inconsistent and influenced by methodological biases. Hypothesising that studying preclinical brain dysfunction may better unmask this link, we investigated the association of cancer with motoric cognitive risk syndrome (MCRS), an established dementia prodrome combining slow gait with subjective cognitive complaints.