Abstract
INTRODUCTION: To examine the risk of Alzheimer's disease (AD) among cancer survivors in a national database. METHODS: Retrospective cohort of 3,499,378 mostly male US veterans aged ≥65 years were followed between 1996 and 2011. We used Cox models to estimate risk of AD and alternative outcomes (non-AD dementia, osteoarthritis, stroke, and macular degeneration) in veterans with and without a history of cancer. RESULTS: Survivors of a wide variety of cancers had modestly lower AD risk, but increased risk of the alternative outcomes. Survivors of screened cancers, including prostate cancer, had a slightly increased AD risk. Cancer treatment was independently associated with decreased AD risk; those who received chemotherapy had a lower risk than those who did not. DISCUSSION: Survivors of some cancers have a lower risk of AD but not other age-related conditions, arguing that lower AD diagnosis is not simply due to bias. Cancer treatment may be associated with decreased risk of AD.