Ranking the risk factors for Alzheimer's disease; findings from the UK Biobank study

Abstract

Background: The cause of the most common form of dementia, sporadic Alzheimer's disease (AD), remains unknown. This may reflect insufficiently powered studies to date for this multi-factorial disorder. The UK Biobank dataset presents a unique opportunity to rank known risk factors and determine novel variables.

Methods: A custom machine learning approach for high dimensionality data was applied to explore prospectively associations between AD in a sub-cohort of 156,209 UK Biobank participants aged 60-70 including more than 2,090 who were subsequently diagnosed with AD.

Results: After the possession of the APOE4 allele, the next highest ranked risk factors were other genetic variants within the TOMM40-APOE-APOC1 locus. When stratified by their apolipoprotein $\mathrm{epsilon}$ 4 (APOE4) carrier status, the most prominent risk factors in carriers were AST:ALT ratio, the "number of treatments/ medications" taken as well as "time spent in hospital" while protection was conferred by "Sleeplessness/Insomnia". In non-APOE carriers, lower socioeconomic status and fewer years of education were highly ranked but effect sizes were small relative to APOE4 carriers.

Conclusions: Possession of the APOE4 allele was confirmed as the most important risk factor in AD. Other TOMM40-APOE-APOC1 locus variants further moderate the risk of AD in APOE4 carriers. Liver pathology is a novel risk factor in APOE4 carriers while "Sleeplessness/Insomnia" is protective in AD irrespective of APOE4 status. Other factors such as "Number of treatments/ medications" suggest that multimorbidity is an important risk factor for AD. Future treatments aimed at co-morbidities, including liver disease, may concomitantly lower the risk of sporadic AD.

Keywords: Alzheimer’s disease; Apolipoprotein E; Liver; Machine learning; Multimorbidity; UK Biobank.

Fig. 1

UK Biobank timeline and case/control inclusion/exclusion criteria. A schematic demonstrates our selection criteria for hypothetical UKB participants. In this case, Participant 1 would be selected in our analysis as they attended the assessment centre and did not meet any of the exclusion criteria, nor were they ever diagnosed with AD. Participant 2 is included as a case, with a past disease coded as an independent variable as it was diagnosed prior to their attendance at the assessment centre, and they were diagnosed with AD over 2 years after their attendance. Participant 3 is excluded as they received an AD diagnosis prior to 2 years after attending the baseline assessment centre.

Fig. 2

SHAP scores and AD risk. Bar plots showing mean SHAP scores for the entire cohort (A), APOE4 carriers (B) and non-APOE4 carriers (C). The colour of the bar represents features which positively impacted on AD risk in the UKB model (red) and those with a negative or protective association (blue). Black bars indicate confidence intervals. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)