Health Sci Rep. 2026 Feb 22;9(2):e71810. doi: 10.1002/hsr2.71810. eCollection 2026 Feb.
ABSTRACT
BACKGROUND AND AIMS: Apolipoprotein E (APOE) and proprotein convertase subtilisin/kexin type 9 (PCSK9) are both lipid proteins and related to immunity/inflammation. We hypothesized that PCSK9 impacts on Alzheimer's disease (AD) risk in an APOE genotype dependent manner.
METHODS: We used the Framingham Heart Study (FHS) Offspring cohort (Gen 2), with data on plasma PCSK9 protein concentration, as the baseline exposure for 1,704 study subjects. Using Cox regression models, the outcomes were incidents of AD or all‐cause dementia. Using another FHS dataset with 3,048 individuals with genetic data, we examined the association between PCSK9 genotypes and the incidence of AD/dementia, stratifying the analysis based on APOE ε4 status. The Alzheimer's Disease Neuroimaging Initiative (ADNI) study was used to validate some of the main findings.
RESULTS: Higher plasma PCSK9 protein levels were associated with a lower risk of AD (HR [95%CI]: 0.74 [0.58, 0.94]; p = 0.01) in APOE ε4 noncarriers; in contrast, PCSK9 levels were not significantly associated with AD risk in APOE ε4 carriers, after adjusting for common confounders, lipid profile, and lipid treatment. Using the three SNPs (rs502576, rs529787, rs676297) of the PCSK9 gene associated with PCSK9 levels in blood, we consistently found that the genotypes, which determine a low concentration of PCSK9, were associated with AD risk only in APOE ε4 noncarriers. These findings were validated by the ADNI study, which showed that the PCSK9 genotypes were associated with AD risk and with the AD biomarker—a low concentration of Aβ42 in CSF, only in APOE ε4 noncarriers.
CONCLUSIONS: Our study suggests that high blood PCSK9 levels are protective against AD risk in APOE ε4 noncarriers, potentially through mechanisms related to lipid metabolism. The findings may highlight the importance of considering APOE genotype when prescribing the drugs targeting PCSK9.
PMID:41737429 | PMC:PMC12927996 | DOI:10.1002/hsr2.71810