(CSPOT51) Central vs. Non-central Angiotensin Converting Enzyme-inhibitors and the risk of Alzheimer's disease: A cohort study

Background: Cerebral angiotensin converting enzyme (ACE) has been shown to break down amyloid-beta; a protein that is a hallmark feature of Alzheimer’s disease (AD). Evidence from mendelian randomisation has shown that inhibition of ACE with ACE-inhibitor drugs may increase the risk of AD. However, there is heterogeneity within the ACE-inhibitor drug class and not all ACE-inhibitors can penetrate the blood-brain barrier (BBB) to elicit cerebral ACE inhibition. Therefore, a hypothesis testing cohort study was conducted to investigate if there is a differential risk of AD in patients initiated on central-acting (CA) ACE-inhibitors compared to non-central-acting (NC) ACE-inhibitors.

Objectives: To evaluate the association between CA-ACE-inhibitors compared to NC-acting ACE-inhibitors and the risk of incident AD.

Methods: Primary care electronic health records from IQVIA Medical Research Database UK, previously known as THIN, were used to conduct a new user, active-comparator cohort study in adults aged ≥40 years between 2000-2021. Patients newly prescribed CA-ACE-inhibitors were compared with patients newly prescribed NC-acting-ACE-inhibitors. Diagnostic read-codes were used to measure the primary outcome of incident AD. Potential confounders including, comorbidities, co-prescriptions, laboratory results, socioeconomic status, smoking, alcohol and body mass index were adjusted for using multivariable Cox proportional hazard regression to estimate adjusted hazard ratios (HR) with 95% confidence intervals (CI). To address AD latency and minimise reverse causation, sensitivity analysis was conducted with ACE-inhibitor initiation lagged (delayed) by 1, 3 and 5 years.

Results: The cohort included 621,522 new ACE-inhibitor initiators, with a mean age of 64 (SD 12) years and 48% female sex. There were 8,755 individuals newly diagnosed with AD during a median follow up of 6.6 (IQR 3.3-10.7) years. The crude incidence rate (IR) for initiators of CA-ACE-inhibitors was 19.2 per 10,000 person-years at risk (PYAR) (95%CI 18.8-19.7) and 21.9 per 10,000 PYAR (95%CI 20.2-23.8) in NC-ACE-inhibitor initiators. The adjusted HR of incident AD in CA-ACE-inhibitors compared to NC-ACE-inhibitors was 0.92 (95%CI, 0.81-1.05). The results from sensitivity analysis were consistent with the primary result.

Conclusions: In this large population-based cohort study, there was no evidence to support the hypothesis that CA-ACE-inhibitors are associated with a differential risk of AD compared to NC-ACE-inhibitors. Our findings question the significant contribution of ACE mediated amyloid-beta degradation on AD risk.