Inverse relationship between cancer and Alzheimer’s disease.

Cancer and Alzheimer's disease are common diseases in aging populations. Prior research has reported a lower incidence of Alzheimer's disease-type (amnestic) dementia among individuals with a diagnosis of cancer. Both cancer and amnestic dementia are prevalent and potentially lethal clinical syndromes. The current study (Brain, awac035, https://doi.org/10.1093/brain/awac035 ) was conducted to investigate the association of cancer diagnosis with neuropathological and cognitive features of dementia.

Data were analyzed from longitudinally evaluated participants in a community-based cohort study of brain aging who came to autopsy at the University of Kentucky Alzheimer's Disease Research Center. These data were linked to the Kentucky Cancer Registry, a population-based state cancer surveillance system, to obtain cancer-related data. The investigators examined the relationship between cancer diagnosis, clinical dementia diagnosis, Mini-Mental State Examination scores, and neuropathological features using inverse probability weighting to address bias due to confounding and missing data. To address bias due to inclusion of participants with dementia at cohort baseline, they repeated all analyses restricted to the participants who were cognitively normal at baseline. Included participants (n = 785) had a mean ± SD age of death of 83.8 ± 8.6 years; 60.1% were female. Cancer diagnosis was determined in 190 (24.2%) participants, and a diagnosis of mild cognitive impairment or dementia (MCI/Dementia) was determined in 539 (68.7%). APOE ε4 allele dosage was lower among participants with cancer diagnosis compared to cancer-free participants overall (P = 0.0072); however, this association was not observed among those who were cognitively normal at baseline. Participants with cancer diagnosis had lower odds of MCI/Dementia, and higher cognitive test scores (e.g., MMSE scores evaluated 6 and ≤ 2 years antemortem, P <0.001 for both comparisons). Cancer diagnosis also associated with lower odds of higher Braak neurofibrillary tangle stages (III/IV) or (V/VI), moderate/frequent neuritic plaques, moderate/frequent diffuse plaques, and moderate/severe cerebral amyloid angiopathy (all P < 0.05). By contrast, TDP-43, α-synuclein, and cerebrovascular pathologies were not associated with cancer diagnosis. Cancer diagnosis was associated with a lower burden of Alzheimer's disease pathology and less cognitive impairment. These findings from a community-based cohort with neuropathologic confirmation of substrates support the hypothesis that there is an inverse relationship between cancer and Alzheimer's disease