Does statin use affect amyloid beta deposition and brain metabolism?

Abstract Background There are contradictory findings regarding the effect of statin drugs on amyloid β (Aβ) deposition as one of the main hallmarks of Alzheimer's disease (AD), along with tau pathology. We aimed to longitudinally investigate the therapeutic and preventive role of statin drugs by examining the brain Aβ deposition and metabolism rate in AD, mild cognitive impairment (MCI), and healthy controls (HC). Methods The data of 828 subjects including 178 HC, 492 MCI, and 158 AD individuals were obtained from ADNI. The baseline and longitudinal [18F] AV45 and 18‐fluorodeoxyglucose (FDG) PET standard uptake value ratio (SUVR) measures were investigated among statin users and non‐users. Results Our results showed that there is no significant difference in baseline Aβ deposition and metabolism rate between statin users and non‐users among HC, MCI, and AD subjects. While there was no significant effect of statin on metabolism rate, there was a significant difference in Aβ deposition change after 4 years (from baseline) between statin users and non‐users within HC subjects (p = 0.011). The change of Aβ deposition at 4 years from baseline was −2.0 ± 6.3% for statin users and 1.4 ± 4.7% for non‐users. There was no significant association between statin duration use with baseline and longitudinal Aβ deposition and metabolism rate. However, statin dosage was significantly associated with Aβ deposition in 2 years (r = −0.412, p = 0.021) in the HC group. Moreover, our analysis showed a significant correlation between total statin exposure (duration×dosage) and Aβ deposition in 2 years visit (r = −0.198, p = 0.037) in HC subjects. Furthermore, we investigated the longitudinal changes within each group of statin users and non‐users separately in linear mixed models. Our findings showed that there are no significant changes in AV45 and FDG SUVR among both groups. Conclusion The present longitudinal analysis revealed that using statins might be beneficial in slowing down or stabilizing the Aβ deposition due to aging in subjects without cognitive impairment. However, once the clinical symptoms of cognitive impairment appear, statins fail to slow down Aβ deposition. Overall, our findings revealed that statin users might have slower Aβ aggregation than non‐users.


| INTRODUC TI ON
Alzheimer's disease (AD) is an aggressive and progressive neurodegenerative disease known as the primary cause of dementia. 1 The first presentation of AD is memory loss in the vast majority of cases with other symptoms such as cognitive dysfunction, psychiatric symptoms, and behavioral disturbances. 1 Although the exact etiology of AD is yet to be known, it is believed that two factors play a crucial role in AD pathology: amyloid β (Aβ) plaques and tau tangles. 1 Moreover, reduced cerebral blood flow (CBF) and neurovascular dysfunction associated with Aβ plaques are major contributions to the progress of AD. 2 Elevated levels of Aβ in the brain are significantly associated with cognitive decline, and specific regions of the brain elevate even before the occurrence of tau pathology, the other pathological hallmark of AD. 3 Besides Aβ plaques and tau neurofibrillary tangles, glucose hypometabolism is also a part of the pathological mechanism of AD and could be used as a diagnostic factor in preclinical stages. 4 Along with Aβ deposition, cerebral hypometabolism assessed by fluorodeoxyglucose PET (FDG-PET) is significantly correlated with AD biomarkers along disease progression. 5 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor (HMG-CoA inhibitor) also known as statins are lipid-lowering agents (LLAs) used in lowering low-density lipoprotein (LDL) cholesterol. 6 Statins are the most widespread LLAs used for primary and secondary prevention of cardiovascular events. 7 For many years, there were reports on the impact of these drugs on neurological and psychiatric disorders. Statins could significantly reduce depression and hospitalization in patients prone to psychiatric disorders. 8 Moreover, based on several studies, use of statin is correlated with less severe symptoms in patients with neurodegenerative diseases such as AD and Parkinson's disease (PD). 9 There have been reports on the probable effect of statins on future AD development. Recently, a meta-analysis conducted by Zhang et al. concluded that statins, by inhibiting the formation of intracerebral amyloid, have a protective role against dementia and AD. 10 However, preventing cognitive decline may not be noticeable in short follow-ups among AD patients. In line with these findings, another study also suggested that statins have no contributions to dementia but may need additional investigations before using them in aggressive lipoprotein therapies in elderly people. 11 Previous studies showed the alternated cholesterol profile in AD patients and suggested that statins could be used as an efficacious therapy in AD. 12 Different epidemiological studies indicated that statin use can be preventive and reduce the risk of AD and regulates the Aβ and possibly tau metabolism. 13 Due to the contradictory findings regarding the role of statins in prevention, treatment, or even worsening AD, there is a strong need for further investigation to examine the effect of statins on AD development not only epidemiologically. 10,14,15 Therefore, we aimed to longitudinally investigate the therapeutic and preventive role of statin drugs by examining the brain Aβ deposition and metabolism rate in AD, mild cognitive impairment (MCI), and healthy controls (HC). without cognitive impairment. However, once the clinical symptoms of cognitive impairment appear, statins fail to slow down Aβ deposition. Overall, our findings revealed that statin users might have slower Aβ aggregation than non-users.

K E Y W O R D S
Alzheimer's disease, amyloid beta, cognitive impairment, metabolism, statins

| Classification of statin exposure
Information on statin use was obtained from the ADNI concurrent medication file, which is a record of the longitudinal use of medications among participants. The subjects with at least 2 years of statin use with a minimum of 10 mg per day were categorized as statin users. The duration of statin use was estimated by subtracting the date of the baseline visit date and the start date.

| PET imaging biomarkers
Aβ deposition was visualized with PET tracer [ 18 F]AV45. 16 Measures of regional AV45 standard uptake value ratios (SUVR) which were calculated by ADNI core (

| Cognitive assessments
All participants underwent Mini-Mental State Examination (MMSE), which included 30 questions to measure the cognitive status at baseline. The MMSE score of patients was downloaded from ADNI.

| APOE ε 4 genotyping
The data of APOE ε4 genotyping of the participants were extracted from the ADNI dataset. The participants were divided into APOE ε4 positive and negative.

| Participant's characteristics
Overall, 30.4% of our subjects were statin users. The mean duration of statin use was 6.5 ± 5.7, and 67% of statin users received simvastatin followed by pravastatin (13%). The demographic and clinical characteristics of the subjects are detailed in Table 2.

| Effect of statin on PET findings
We initially investigated the association between statin use and baseline AV45 and FDG SUVR using ANCOVA models. Our results showed that there is no significant difference in baseline Aβ deposition and metabolism rate between statin users and non-users among HC, MCI, and AD subjects ( Table 3).  There was no difference in FDG SUVR change at 2 years between statin users (−9.6 ± 7.4%) and non-users (−7.0 ± 5.6%) in AD patients (p = 0.799). Furthermore, in the AD group, we found that there was no significant difference in change of AV45 SUVR at 2 and 4 years from baseline and at 4 years from 2 years visits between statin users and non-users. The AV45 SUVR change is represented in Table 3. Note: P value as defined using ANCOVA models adjusted for age, sex, APOE e4, MMSE score, cardiovascular conditions, and cholinesterase drugs use.
Abbreviations: AD, Alzheimer's disease; HC, healthy controls; MCI, mild cognitive impairment; SUVR, standard uptake value ratios. models ( Table 4). Our findings showed that there are no significant changes in AV45 and FDG SUVR among both groups.

| DISCUSS ION
Our longitudinal analysis indicates that using statins might be beneficial in not increasing Aβ deposition in HC subjects. In the first 2 years of our follow-up, no significant difference in Aβ deposition Similarly, free cholesterol accelerates Aβ aggregation, as the number and size of Aβ aggregates generated in the presence of free cholesterol are larger than those formed in the absence of free cholesterol. 22 As with other proteins, the amount of Aβ depends on the balance between its production and clearance. Blood vessels of the brain play a crucial role in controlling Aβ clearance. 23 LRP1, an apoE receptor, may be involved in the clearance of Aβ via blood vessels. 24 Statins promote clearance of Aβ by up-regulating the amount of LRP1 in the vessels. 25 ApoE, one of the brain's primary cholesterol transporters, is the most potent genetic risk factor associated with AD. 26 In both AD patients and cognitively normal individuals, the APOE-ε4 allele promotes the deposition of senile plaques. 27 Furthermore, improved vascular health and blood flow are linked to lower cholesterol levels. Pathologically damaged vessels have been observed within the brain of AD patients. 23 Several studies have demonstrated that statins reduce plasma cholesterol levels and prevent cerebrovascular and cardiovascular events. 28 Thus, in combination with antihypertensive drugs that protect vessels, statins can reduce the risk of vascular dementia. 29 Statins have also been demonstrated to affect brain cells independent of cholesterol metabolism, affecting neurotransmitter levels, synaptic neurotransmitter receptors, cellular survival, neuronal dendritic arborization, and myelination. 30 For instance, a study of animal design found that statins altered gene expression in the brain cortex of mice and that the impacted genetic pathways were primarily connected to apoptotic mechanisms. 31  In recent years, PET imaging, particularly with FDG and AV45 tracers, has become an essential tool in the early detection of AD biomarkers, including Aβ deposition and hypometabolism. 35 FDG PET scan measures glucose consumption in brain regions and is an early biomarker of brain hypoactivity and hypometabolism in neurodegenerative diseases. 36 On the other hand, the AV-45 PET scan technique is an amyloid scan that visualizes amyloid aggregation and plaque formation in AD with high sensitivity.Thus, amyloid PET scan SUVR is accepted as a highly sensitive early-phase predictor for mild stages of AD. 37 It is widely accepted that the therapeutic and preventive effects of statins on AD should be studied separately. 13 Our results demonstrated no significant difference in Aβ deposition between statin users and non-users among AD patients. Similarly, most of the previous studies focusing on statins as a medical solution to improve cognition in AD patients have failed to determine any significant effects. 38 However, a meta-analysis of studies on AD patients suggests that statin consumption may slightly slow cognitive impairment in mild-to-moderate dementia. 12 Based on our findings, using statins seems to slow down and stabilize the Aβ pathology due to the aging process in healthy individuals compared to non-users. However, the within-group analysis showed that there was no increase or decrease in Aβ deposition due to the statin use. Furthermore, several studies have determined the protective effects of statins in preventing AD in healthy individuals. 39 In line with our study, a meta-analysis has revealed that only long-term use of statins can reduce the incidence of AD. 39 Furthermore, the protective effects of statins were more prominent in younger subjects, implying that statin use should begin early in midlife to slow down or reduce the risk of AD. 13

| Limitations and strengths
The limitations of our study are as follows: first, this is an observational study rather than a randomized clinical trial, which is re-

| CON CLUS ION
The present longitudinal analysis revealed that using statins might be beneficial in slowing down or stabilizing the Aβ deposition due to aging in subjects without cognitive impairment. However, once the clinical symptoms of cognitive impairment appear, statins fail to slow down Aβ deposition and make any difference. Moreover, the withingroup analysis showed no significant changes in Aβ among statin users and non-users. In other words, statins seem to be preventive rather than a therapeutic agent. In contrast, statins failed to affect cerebral metabolism during our follow-up. Overall, our findings revealed that statin users might have slower Aβ aggregation compared to non-users. However, further studies are required to confirm our findings.

ACK N OWLED G M ENTS
Data collection and sharing for this project were provided by the

CO N FLI C T O F I NTE R E S T S TATE M E NT
The author declares no conflict of interest regarding the publication of this paper.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets analyzed during the current study are available upon request with no restriction.

CO N S E NT FO R PU B LI C ATI O N
This manuscript has been approved for publication by all authors.