During the past few decades, Alzheimer’s disease (AD) has emerged as a leading global health concern (Alzheimer’s Disease International, 2019). Despite constant efforts to develop new drugs, the complexity of pathologies has left no promising treatments for AD (Cummings et al., 2014; Bachurin et al., 2017). Owing to the limitations and uncertainty of current treatments, targeting modifiable risk factors for preventing AD incidence and delaying progression has gained importance in recent years. Management of hypertension with antihypertensive drugs has been recommended for the primary prevention of AD (Crous-Bou et al., 2017; Livingston et al., 2020; Hefner et al., 2021). The experimental results, however, remain controversial, and further analyzes are required to reach a general consensus on the use of antihypertensive drugs (Yasar et al., 2013; Chuang et al., 2014; Walker et al., 2020). Among the antihypertensive agent classes, renin-angiotensin system (RAS) inhibitors, particularly angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin II type 1 receptor blockers (commonly used as angiotensin II receptor blockers, ARBs), have been reported to have potential benefits on reducing the risk of incident AD (Li et al., 2010; Davies et al., 2011; Barthold et al., 2018).

RAS is a hormonal system responsible for regulating blood pressure (BP), fluid balance, electrolyte homeostasis, and vascular resistance. This system is mediated by angiotensin (Ang) ligands that interact with various receptors, including angiotensin II type 1 receptor (AT1R), angiotensin II type 2 receptor (AT2R), angiotensin IV receptor (AT4R), and Mas receptor (MASR; Gebre et al., 2018). In addition to the peripheral RAS, the receptors in the central nervous system (CNS) are involved in oxidative stress, neuroinflammation, and neuronal apoptosis, causing neurodegeneration (Abiodun and Ola, 2020; Xu et al., 2021). Accordingly, blood–brain barrier (BBB)-crossing RAS inhibitors that block neurodegenerative pathways may provide neuroprotective effects in brain disorders including AD.

A few clinical studies have reported that CNS penetration of RAS inhibitors was associated with reduced cognitive decline and a lower conversion rate from mild cognitive impairment (MCI) to AD (Ohrui et al., 2004; Sink et al., 2009; O’Caoimh et al., 2014; Wharton et al., 2015). However, other studies have shown unclear relationship between BBB permeability and increased neuroprotection by RAS inhibitors (Hebert et al., 2013; Fazal et al., 2017). Considering that most studies focused on ACEIs and the results were inconclusive, further studies are essential in clarifying the effect of BBB-crossing RAS inhibitors, independent from blood pressure lowering effect, on the risk of AD incidence.

This study aimed to assess the effects of RAS inhibitors on the risk of incident AD by using a longitudinal national health insurance database in Korea. In addition to comparing the effect of RAS inhibitors on AD based on class and BBB permeability, we further investigated the cumulative duration-response relationship.

As the protective effect of antihypertensive agents on cognitive decline beyond their blood pressure-lowering effects has emerged, the potential effect of reducing the risk of AD via the renin-angiotensin system has been demonstrated in animal and human studies (Li et al., 2010; Davies et al., 2011; Barthold et al., 2018; Abiodun and Ola, 2020). However, the neuroprotective effects of RAS inhibitors reported in previous studies have been conflicting (Ohrui et al., 2004; Sink et al., 2009; Hebert et al., 2013; Hsu et al., 2013; O’Caoimh et al., 2014; Qiu et al., 2014; Wharton et al., 2015). In this nationwide population-based cohort study, patients with IHD who used BBB-crossing ARBs had a lower risk of incident AD than those who did not use RAS inhibitors. Notably, our study showed a significant reduction in the risk of incident AD in patients who used BBB-permeable ARBs at higher cumulative doses. While previous studies have focused on comparing the effects of ARBs and ACEIs (Marcum et al., 2022) or BBB permeability within RAS inhibitors (Hebert et al., 2013; Qiu et al., 2014; Ho et al., 2021), to the best of our knowledge, this is the first study to simultaneously assess risk reduction considering both BBB permeability and cumulative doses. Our results were robust owing to a valid study design with a long-term follow-up based on nationwide study samples with appropriate comparisons and sensitivity analyzes.

In this study, we revealed that the use of ARBs, but not ACEIs, was associated with a reduced risk of AD. This result is consistent with some clinical studies that have shown the advantageous effect of ARBs over ACEIs in reducing AD risk (Li et al., 2010; Davies et al., 2011; Barthold et al., 2018; Marcum et al., 2022). Additionally, a number of animal studies have supported this difference in protective effect by suggesting potential underlying mechanisms. ACEIs target the angiotensin-converting enzyme (ACE), which is responsible for converting Ang I to Ang II, thus attenuating AT1R and AT2R activation, whereas ARBs selectively block the Ang II/AT1R axis (Gebre et al., 2018). AT1R activation induces oxidative stress, neuroinflammation, and apoptosis, whereas AT2R counteracts AT1R-mediated neurodegeneration by various mechanisms (Lanz et al., 2010; Faraco et al., 2016; Abiodun and Ola, 2020). Hence, the blockade of AT1R by ARBs may induce indirect activation of the Ang II/AT2R axis to provide neuroprotection (Mogi and Horiuchi, 2013). Enhancement of cognitive function by direct stimulation of AT2R was also demonstrated in the animal study (Jing et al., 2012). Moreover, the benefits of ARBs can be attributed to the conversion of Ang II into Ang IV and Ang (1–7), which are selective for AT4R and MASR, respectively. AT4R has been suggested to have a positive effect on cerebral blood flow, memory, and neuroprotection (Näveri et al., 1994; Royea and Hamel, 2020), whereas enhancement of the Ang (1–7)/MASR axis has been reported to have a potential anti-inflammatory effect and facilitate hippocampal long-term potentiation (Hellner et al., 2005; Wright and Harding, 2019). Several studies have also found that the expression of ACE and ACE2 is related to a decreased amyloid-beta (Aβ) load (Hemming and Selkoe, 2005; Zou et al., 2007; Kehoe et al., 2016). However, further translational investigation is essential to confirm the association between the neuroprotective effects of ARBs and Aβ pathology (Loera-Valencia et al., 2021).

Our analyzes indicated that the use of BBB-crossing ARBs was associated with a reduced risk of AD. The finding was especially significant as CI did not overlap with those of the other three types of RAS inhibitors. Additional benefit of using BBB-crossing RAS inhibitors has been demonstrated in previous studies assessing the effect on cognitive function and MCI to AD conversion (Wharton et al., 2015; Ouk et al., 2021). Another longitudinal study, which investigated the effect of ACEIs depending on central exposure, reported no significant association between BBB-crossing ACEIs and AD and a risk of incident AD with poor BBB-crossing ACEIs (Sink et al., 2009). However, a previous meta-analysis (Ho et al., 2021) assessing the effect of BBB-crossing RAS inhibitors on seven cognitive domains reported that poor BBB-crossing RAS inhibitors demonstrated a better effect in the attention domain compared to that of BBB-crossing RAS inhibitors. As previous studies on cognitive decline and incident AD by BBB permeability of RAS inhibitors largely focused on ACEIs (Ohrui et al., 2004; Sink et al., 2009; Hebert et al., 2013; O’Caoimh et al., 2014; Ouk et al., 2021), our results on BBB-crossing ARBs are noteworthy, but the benefit of using BBB-crossing ARBs as potential drugs for preventing AD should be carefully interpreted. In addition to BBB-crossing effects, some in vitro/vivo and animal studies have reported that some BBB-crossing ARBs, such as telmisartan, showed partial peroxisome proliferator-activated receptors (PPAR) gamma activation effects that have beneficial effects on cognitive functions (Mogi et al., 2008; Pang et al., 2012; Garg et al., 2021). However, the PPAR-gamma activation effect of ARBs is still controversial, with limited evidence for the attenuation of cognitive functions in only some ARBs (Benson et al., 2004; Erbe et al., 2006; Kajiya et al., 2011). Therefore, further comprehensive studies that have considered the PPAR-gamma binding affinities on ARBs, as well as the BBB-crossing characteristics, are needed.

Remarkably, the significantly reduced risk of AD by BBB-crossing ARBs was robust in patients with a larger cumulative dose or longer duration, regardless of the daily equivalent dose. These results implied that the cumulative exposure duration was a more crucial factor in the neuroprotective effect of ARBs than the daily exposure dose. Risk-reducing effect of ARBs on AD with larger cumulative dose and longer exposure were also demonstrated in a previous longitudinal study, supporting our findings (Chiu et al., 2014). Considering that antihypertensive drugs are generally used for an extended period and our results showed a cumulative effect of BBB-crossing ARBs on AD, they could be suggested as promising targets for drug repurposing. The current treatment for AD shows modest effects only on symptoms (Atri, 2019; Cummings et al., 2020), and the efficacy of the newly approved drug, aducanumab, is also controversial (Whitehouse et al., 2022). Moreover, midlife hypertension has been associated with an increased risk of AD, and blood pressure control is a modifiable risk factor for cognitive decline (Lennon et al., 2019). Taken together, BBB-crossing ARBs might be a promising disease-modifying drug option for reducing the risk of AD in patients with cardiovascular diseases, such as hypertension.

In the subgroup analysis, no difference in the protective effect of BBB-crossing ARBs was identified between men and women. A study by Barthold et al. reported that ARBs were superior to ACEIs in risk of AD incidence for white men and women, but no association was observed for the black and Hispanic populations (Barthold et al., 2018). Estrogen lowers AT1R expression, prevents the production and action of angiotensin II, and decreases NADPH-oxidase activity and expression of neuroinflammatory markers (De Silva and Faraci, 2012; O’Hagan et al., 2012; Rodriguez-Perez et al., 2015). Aging men with aromatization of androgens to estrogens have a higher estrogen level than that of aging women with dramatic ovarian loss of 17β-estradiol (Rosario et al., 2011). A study on the pathophysiology of sex differences in the protective effect of ARB owing to race is needed. Moreover, further studies in other Asian countries are needed to confirm the sex difference in the protective effect of ARBs in Asians.

This study has several limitations. First, this study used a secondary claims database; therefore, we could not verify detailed clinical information, including symptoms, body weight, blood pressure, smoking, alcohol intake, education level, and genetic factors, such as APOE ε4. In addition, the limitation related to the accuracy of incident AD needs to be considered because the outcome variable was identified based on ICD-10 diagnostic codes. Given that the diagnosis of AD is based on the patient’s symptoms (Atri, 2019) and the protective effects of RAS inhibitors have been reported to vary according to cognitive symptoms (Ho et al., 2021), our results should be carefully interpreted. However, we attempted to use the medication prescription claims along with the diagnostic codes for enhanced accuracy of outcome definitions. Moreover, we adopted various outcome definitions in sensitivity analyzes to confirm the robustness of the study results. Second, the possibility of a selection bias cannot be neglected, as our study population was selected based on a very short identification period of 6 months. Moreover, we could not consider active comparators and make a direct comparison of the effect of drugs with different mechanisms, such as beta-blockers, CCBs, or thiazides, because antihypertensive agents are usually used in combination. To minimize this selection bias, we balanced RAS inhibitor users and non-users by PS matching and adjusted for various confounders using rigorous definitions. Moreover, our sensitivity analysis by shifting the index date provided comparable results. Third, it is difficult to generalize the study results to the entire population, as our study population included patients with IHD, who have a high cardiovascular profile. RAS inhibitors, possessing strong vascular effects, has been used for treating various cardiovascular diseases, and conflicting results have been reported on the neuroprotective effect of ARBs, depending on the study population. Further research on the effects of RAS inhibitors on AD in patients with various cardiovascular diseases is required. Finally, the duration or cumulative doses of concomitant medications could not be considered in our study. Instead of considering the variability of confounder status by using time-varying Cox regression, this study used the precise definition of comorbidities and concurrent medications that appeared at least once every year during the follow-up period.

To the best of our knowledge, this is the first longitudinal study to demonstrate the effect of BBB-crossing ARBs on the incidence of AD with cumulative dose and duration subgroups using a population-based cohort. In this study, we highlighted the neuroprotective effect of ARBs, particularly BBB-crossing ARBs, on AD. Additionally, we present a novel finding of the protective effects against AD conferred by long-term use of BBB-crossing ARBs. In addition to existing evidence, these results are expected to provide valuable insights for AD-targeted drug development.

The datasets presented in this article are not readily available because the primary data analyzed in this study are handled and stored by the Health Insurance Review and Assessment Service. Requests to access the datasets should be directed to Health Insurance Review and Assessment Service, https://www.hira.or.kr.

The studies involving human participants were reviewed and approved by Institutional Review Board (IRB) of Yonsei University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

HWL and SK contributed for study design, data analysis, data interpretation, and writing of the manuscript. YMY contributed for study conceptualization, data interpretation, critical revision of the manuscript, and supervision of the study. YJ and YK contributed to data analysis and manuscript revision. BSY contributed to clinical interpretation of the data and critical revision of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Ministry of Science and ICT (MSIT) of the Korea government (No. 2020R1G1A110120513).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnagi.2023.1137197/full#supplementary-material