Refining Risk for Alzheimer's Disease Among Heterozygous APOE epsilon 4 Carriers

In a large population-based cohort, we show not all heterozygous APOE epsilon 4 carriers are at increased risk for Alzheimer's disease (AD); a significantly higher AD proportion was only found for epsilon 3/epsilon 4, not epsilon 2/epsilon 4. Among epsilon 3/epsilon 4 carriers (24% in the cohort), the AD proportion differed considerably by polygenic risk score (PRS). In particular, the AD proportion was lower than the entire cohort for subjects in the bottom 20-percentile PRS and was higher than that of homozygous epsilon 4 carriers for subjects at the top 5th-percentile PRS. Family history was no longer a significant predictor of AD risk after adjusting APOE and PRS.

Automated summary

In a large population-based cohort, not all heterozygous APOE epsilon 4 carriers are at increased risk for Alzheimer's disease (AD). Only epsilon 3/epsilon 4 carriers showed a significantly higher proportion of AD, not epsilon 2/epsilon 4 carriers. Among epsilon 3/epsilon 4 carriers (24% in the cohort), the AD proportion differed significantly based on polygenic risk score (PRS). Particularly, subjects in the bottom 20-percentile PRS had a lower AD proportion compared to the entire cohort, while subjects at the top 5th-percentile PRS had a higher AD proportion than homozygous epsilon 4 carriers. Family history was no longer a significant predictor of AD risk after adjusting APOE and PRS.