APOE ε4 Allele Modulates The Differential Effects of Education on Cognition in Alzheimer’s Continuum: The SILCODE Study

Background: Elders with subjective cognitive decline (SCD) in the absence of objective cognitive impairment are at increased risk of developing mild cognitive impairment (MCI) or Alzheimer’s disease (AD). Previous study has reported that higher education was associated with better performance in cognitive performance in elderly people with SCD. However, it is not clear whether there is a differential effect of education on cognition between healthy controls (HC), individuals with SCD and patients with cognitive impairment (CI). Methods: We performed a cross-sectional study that included 186 HC, 279 SCD subjects and 79 patients with CI from the Sino Longitudinal Study on Cognitive Decline (SILCODE) project. A series of neuropsychological tests of memory, executive, language, and general cognitive function were used to assess the subjects cognitive performance. We performed multiple linear regression models to examine the effect of education on neuropsychological test scores in the total sample and then described differences in the effect according to three different groups. In addition, we presented the effect of education on total test score expressed as a global composite z-scores in the total sample and three different groups, respectively. Furthermore, we examined whether apolipoprotein E ε4 allele (APOE ε4) status would regulate the effect of education on the global composite score among three different groups. Results: Education has a positive effect on cognition in the total sample. Stratication for different groups showed that the positive effect of education on cognition is found in HC and SCD group, while education has a negative effect on cognition in the CI group. Furthermore, we found the APOE ε4 allele does not modify the positive effect of education on cognition in the HC group, but the APOE ε4 allele weaken that benecial effect in the SCD group and maintain the negative effect in the CI group. Conclusions: High education may delay the progression from SCD to cognitive impairment and yield differential effects on cognition across the spectrum of AD. Furthermore, that differential effects are subjected to modulation by the APOE ε4 status.


Introduction
Alzheimer's disease (AD) is the leading cause of dementia, which was characterized by a progressive deterioration in memory, language, and logical thinking [1]. Currently, there is no disease-modifying treatment for patients with AD [2]. A consensus has been reached regarding that AD has a long preclinical stage, as its pathology like β-amyloid (Aβ) and tau start to accumulate as early as 15 to 20 years before symptom onset, which indicated that AD can be considered as a continuum [3][4][5]. This asymptomatic stage of AD may offer a golden opportunity for successful prevention and treatment of dementia. What's more, several potentially modi able risk factors for AD and dementia have been identi ed by epidemiological studies [6,7]. Among them, less education is a very important risk factor. It was estimated that the population-attributable risk (PAR) of AD of low educational attainment was the highest around the world (19.1%, 95% CI 12.3-25.6) [8]. A study based on Mendelian randomisation analysis supported that higher educational attainment is associated with a reduced risk of AD [9]. Evidence obtained in recent years suggests that the prevalence and incidence of dementia in some Western countries are stable or declining, which may be interpreted by reductions in absolute inequalities including improvement in education [10].
Subjective cognitive decline (SCD) refers to a condition that cognitively normal individuals reported they have a reduced cognitive function [11,12]. There is growing evidence supports that individuals with SCD are at an increased risk of developing AD or dementia [13][14][15]. In addition, previous studies con rmed that SCD is associated with AD pathology [16,17]. Therefore, individuals with SCD may be a potential target population for AD prevention. A recent study by our group found that higher education was associated with better performance in all cognitive domains in elderly people with SCD and apolipoprotein E ε4 (APOE ε4) status does not modify the bene cial effect of education on cognition [18]. However, it is not clear whether there is a differential in uence of education on cognition between cognitively normal older adults without SCD, individuals with SCD and patients with cognitive impairment.
Here, we used data from the Sino Longitudinal Study on Cognitive Decline (SILCODE) project to examine the effect of education on cognition in cognitively normal older adults without SCD (normal control), subjects with SCD and cognitive impairment, respectively. Furthermore, we explored whether the APOE ε4 status would modify those effects.

Standard protocol approvals, registrations and patients consents
The study protocol was approved by the medical ethics committee of Xuanwu Hospital of Capital Medical University, Beijing, China, and all the participants gave their written informed consent before any study procedures began.

Participants
We included 544 participants in the present study. The sample was selected from the SILCODE project, an observational longitudinal study carried out in Beijing, which was registered at ClinicalTrials.gov (number NCT03370744). Recruitment of SILCODE started in April 2017 and, at time of data extraction for the present study (August 2020), was still ongoing. The goal of SILCODE is to collect longitudinal data including clinical and imaging data from older adults with SCD and then to develop a model for the ultraearly diagnosis of AD. In addition, SILCODE also recruit healthy controls (HC) and patients with mild cognitive impairment (MCI) and AD dementia. The detailed protocol for the SILCODE project has been published elsewhere [19]. All participants underwent a standardized clinical evaluation at baseline visit including a medical history interview, physical and neurological examinations, neuropsychological testing, laboratory tests, and brain magnetic resonance imaging (MRI) scanning and as well as optional [ 18 F] orbetapir (AV-45) positron emission tomography (Aβ-PET) or [ 18 F] uorodeoxyglucose (FDG) positron emission tomography (FDG-PET). The same examinations will be performed by each participant during the 15-month follow-up. The HC group had to achieve unimpaired cognitive performance according to the results of standard neuropsychological assessments and no self-reported persistent decline in cognition. The SCD group was de ned by the presence of self-perceived continuous cognitive decline compared to a previous normal status and unrelated to other diseases or conditions that would lead to cognitive decline, and failure to meet the following criteria for MCI and AD. The CI group was consisted of both patients with MCI and AD. The diagnostic criteria for MCI are de ned by an actuarial neuropsychological method proposed by Jak and Bondi [20] and the clinical diagnostic criteria for AD according to diagnostic guidelines established by National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) [21]. All the participants were required to be aged between 60 and 80 years old and had at least 6 years of education. Further requirements were Mandarin-speaking and right-handed. Here, we included n=186 HC, n=279 SCD subjects and n=79 CI patients for the present study. For details see gure 1 as a owchart.
Clinical and cognitive function assessments A paper-printed case report form (CRF) was used to record demographic features (name, birthdate, sex, education, occupation, ethnic origin, residential address and contact information etc.), medical history  [24,25]. And Montreal Cognitive Assessment-Basic (MoCA-B) scores were used as an index of global cognitive condition [26]. To obtain a total test score of cognitive performance, all raw neuropsychological test scores were rst converted into Z scores calculated by subtracting the score from the mean test score and dividing it by the standard deviation of initial test scores, and then the total test score was expressed as a global composite z-scores created by averaging all neuropsychological test z-scores.

APOE genotyping
A fasting blood sample was drawn from each participant in the department of laboratory, Xuanwu Hospital of Capital Medical University at the baseline visit. A part of the blood sample was used for analysis of the level of blood glucose, blood lipids, anti-syphilis, homocysteine, folic acid, vitamin B12, thyroid hormone, hemoglobin, blood coagulation and the other part was used to determine the apolipoprotein E (APOE) gene polymorphism status. APOE was genotyped using the standard Sanger sequencing method (Sangon, Shanghai, China) with the following primers: 5′-ACGCGGGCACGGCTGTCCAAGG-3′ (forward) and 5′-GGCGCTCGCGGATGGCGCTGA-3′ (reverse). APOE was ampli ed using the following conditions: 1 cycle at 98 °C for 10 s, 35 cycles at 72 °C for 5 s, and 1 cycle at 72 °C for 5 min. Polymerase chain reaction (PCR) was performed in a nal volume of 30 μl containing 10 pmol of forward and reverse primers, and 50 ng of genomic DNA template using PrimeSTAR HS DNA polymerase with GC Buffer (Takara Bio, Kusatsu Shiga, Japan). In the present study, APOE genotype was dichotomized into individuals with 1 or 2 copies of the ε4 allele (APOE ε4 carriers) and those without any copies of the ε4 allele (APOE ε4 non-carriers).

Statistical analyses
First, Shapiro-Wilk test was used to test continuous variables for normality. Means and standard deviations were reported for normally distributed variables, while non-normally distributed variables were described as median (range). Categorical variables were expressed as number (%). Statistical differences between the three groups were analysed using one-way analysis of variance (ANOVA), Kruskal-Wallis test or Chi-square test. Second, the method of simple linear correlation analysis was employed to measure the association between education and raw score of all neuropsychological tests including AVLT-H-N5, AVLT-H-N7, STT-A, STT-B, AFT, BNT and MoCA-B. Third, we used multiple linear regression model, adjusted for age, sex, the presence of hypertension, diabetes, dyslipidemia, and APOE ε4 status, to examine the effect of education on cognition in the total sample with the raw score of neuropsychological tests as the dependent variable and education as independent variable. Next, we examined whether the effect of education on cognition differed according to different groups, by performing previous multiple linear regression analysis in HC group (n=186), SCD group (n=279), and CI group (n=79). Then, we presented the effect of education on the global composite score of cognitive performance in the total sample and three different groups, respectively. Furthermore, we explored whether the APOE ε4 status would modify the effect of education on the global composite score among three different groups. Regression diagnostics were performed to ensure the assumptions for linear regression were met. Residuals were normally distributed. Durbin-Watson test statistics indicated independence of observations and heteroscedasticity was in conformance with test assumptions (results not shown). The "PerformanceAnalytics", "tableone", "forestplot" and "effects" packages in R 3.6.3 software (https://www.r-project.org) were used to perform all the analyses. The signi cant level was set at p< 0.05 (2-sided).

Subject characteristics
In Table 1, we showed the demographic and clinical characteristics of the total sample and according to three different groups. The sample consisted predominantly of women (64.5%). The average age was 66 years. The mean years of education was 12.3. The proportion of APOE ε4 carriers was 24.4%. There were no differences in the prevalence of hypertension, diabetes mellitus and dyslipidemia between the three groups. However, signi cant differences in sex, age, education and all of neuropsychological measures were observed between groups. Speci cally, post-hoc tests showed that the proportion of women in the SCD group was higher than the other two groups, while the years of education in the CI group were lower than the other two groups. Regarding age, the CI group was older than the SCD group, but there was no difference between HC and SCD group, nor between HC and CI group. Overall, the CI group performed worse than other two groups on all neuropsychological tests. In addition, there was a relatively poor performance in AVLT-H-N5, BNT and MoCA-B in the SCD group when compared with the NC group. Multiple regression analyses with adjustment for age, sex, the presence of hypertension, diabetes, dyslipidemia, and APOE ε4 status revealed a positive effect of on all neuropsychological test scores for total subjects (all p<0.001; Figure 2). Then, we strati ed the sample according to three different groups and performed multiple regression analyses with adjustment for age, gender, the presence of hypertension, diabetes, dyslipidemia in three subgroups. Subgroup analyses revealed that education showed a positive effect on all neuropsychological test scores in HC and SCD group (p<0.05; Figure 3), except for the effect of education on STT-A in the HC group (p=0.069, Figure 3). And there was no signi cant association between education and all neuropsychological test scores in the CI group (p>0.05, Figure 3), except for there was a positive effect of education on STT-A (p=0.013, Figure 3).
Next, we presented the effect of education on the total test score expressed as a global composite zscores in the total sample and according to three different groups, respectively. As years of educational attainment increased, the total test score increased in the total samples (Figure 4a), HC and SCD group (Figure 4b). However, as years of educational attainment increased, the total test score decreased in the CI group (Figure 4b). Finally, we explored whether the APOE ε4 status would modify those effects of education on cognition among three different groups. Figure 5 shows that there was a similar effect of education on cognition in the HC group regardless of the APOE ε4 status. However, the APOE ε4 allele does attenuate the positive effect of education on cognition in APOE ε4 carriers with SCD group in comparison with the APOE ε4 non-carriers with SCD group. And the negative effect of education on cognition was maintained in APOE ε4 carriers with CI group when compared with the APOE ε4 noncarriers with CI group.

Discussion
In the present study, we investigated whether the education shows a different effect on cognitive performance in HC, SCD and CI group. Furthermore, this study examined whether the APOE ε4 allele regulates those effects among three different groups. The main ndings of our study are (1) education has a positive effect on cognition in the total sample, (2) the positive effect of education on cognition is found in HC and SCD group, while education has a negative effect on cognition in CI group that as years of educational attainment increased, the cognitive performance deteriorated, (3) the APOE ε4 allele has a differential role in regulating the effect of education on cognition in three different groups. Speci cally, the APOE ε4 allele does not modify the positive effect of education on cognition in the HC group, but the APOE ε4 allele weaken that bene cial effect in the SCD group and maintain the negative effect in the CI group.
In a recently published work, we found that high education has a bene cial impact on cognition in subjects with SCD [18]. We extend on that nding by demonstrating that high education has a differential effect on cognition across the spectrum of AD, as high education has a positive effect on cognition in HC, SCD group and a negative effect on cognition in the CI group. Such difference in the effect of education on cognitive symptoms during different disease stage of AD can be explained by the theory of cognitive reserve (CR). The concept of CR grew out of a marked disjunction between an individual's clinical manifestation and the degree of brain pathology or damage [27]. Foe example, among individuals matched for the amount of AD neuropathologic process, some people develop cognitive impairment while others do not, which can be attributed to differences in CR [28]. The concept of CR refers to an ability that uses alternate paradigms or functional brain process more e ciently to maintain function in response to brain changes or damage [27,29]. As CR is a theoretical construct, it is not measured directly but rather the more general sociobehavioral indices like years of education, IQ, occupational attainment, The p value was calculated using one-way ANOVA.
leisure activities, and other protective factors used as a surrogate [29]. According to the theory of CR in AD [30], individuals with high CR can tolerate more AD pathology in the absence of clinical symptoms and therefore, the point at which cognitive performance begin to decline will be delayed. However, the rate of decline in cognition is more rapid in individuals with high CR than those with low CR after they present with the clinical presentation of AD. Indeed, longitudinal studies have con rmed that greater CR attenuated clinical progression at an early stage of AD, while accelerated cognitive decline at an advanced stage AD [31,32]. In particular, a 7-year follow-up study found that high CR delayed progression to MCI about 9 years in SCD subjects in comparison to SCD subjects with low CR, while high CR was a risk factor for progression to AD dementia in MCI patients with APOE ε4 [32]. Therefore, our ndings that education has a dual role at different stages of AD are in line with the theory of CR.
We further found that the differential effect of education on cognition is subjected to regulation by the APOE ε4 allele across the spectrum of AD. The APOE ε4 allele imposed no regulatory action in HC group, while the bene cial effect of education is reduced in SCD group with APOE ε4 and the negative effect of education is retained in CI group with APOE ε4. Genetically, APOE ε4 allele represents a strongest risk factor for AD [33,34]. Previous studies reported that individuals with APOE ε4 heterozygosity were at 3-4fold increased risk of AD, and with APOE ε4 homozygosity was at 9-15-fold increased risk [35,36]. The mechanism that link APOE ε4 status with AD risk is might be mediated by Aβ-dependent and Aβindependent pathways including tau pathology, in ammatory responses and breakdown of the bloodbrain barrier [37][38][39]. Therefore, one explanation for our ndings modulated by the APOE ε4 status is that subjects with APOE ε4 allele who may harbour severe pathological lesions is more sensitive to the depletion of CR than APOE ε4 non-carriers. Previous study has identi ed that the bene cial effect of education on cognitive function can not be modi ed by the APOE ε4 status in healthy older adults with more than 4-year period follow-up [40]. In individuals with SCD, the APOE ε4 allele signi cantly increased the risk of conversion to AD dementia [41], while at the symptomatic stage, the APOE ε4 allele accelerate progression from MCI to AD dementia only in APOE ε4 carriers [32]. However, further studies will be required to understand the mechanisms underlying the role of APOE ε4 allele in regulating the effect of education on cognition across the spectrum of AD.

Limitation
The strength of this study was a relatively large sample size and comprehensive neuropsychological assessment. However, the study also has its limitations. First, current study was a cross-sectional design, and therefore, longitudinal data is needed to further con rm whether high education has a differential effect on cognition in Alzheimer's continuum and that could be regulated by the APOE ε4 allele. Second, the present study is lack of biomarkers of AD. It is a very important issue to clarify if different AD pathology pro les are differentially related to education and cognitive change or clinical outcomes. Third, the participants in our study were relatively well-educated and females were relatively over represented. Therefore, such differences need to be taken into account when those results are extrapolated to other populations.

Conclusions
Our data indicates that high education may delay the progression from SCD to cognitive impairment, while it could accelerate cognitive deterioration in patients with cognitive impairment, which are in line with the model of CR. Furthermore, the differential effect of education on cognition in Alzheimer's continuum is regulated by the APOE ε4 status. Therefore, no effective treatment available for AD makes it urgent to improve the popularization of education, which may be a potentially promising approaches to preventing AD.

Declarations
Ethics approval and consent to participate The protocol of the study was approved by the medical ethics committee of Xuanwu Hospital of Capital Medical University. Beijing, China, and all participants gave their written informed consent before any study procedures began.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This article was supported by the National Natural Science Foundation of China (Grant 61633018).
Authors' contributions GC collected, analyzed, interpreted the data and drafted this manuscript. KY revised this manuscript. YH designed this study and revised this manuscript. All authors read and approved the nal manuscript.
41. Mazzeo S, Padiglioni S, Bagnoli S, Carraro M, Piaceri I, Bracco L, et al. Assessing the effectiveness of subjective cognitive decline plus criteria in predicting the progression to Alzheimer's disease: an 11- year follow-up study. Eur J Neurol. 2020;27:894-9. Figure 1 Flowchart of the sample selection. HC healthy controls, SCD subjective cognitive decline, CI cognitive impairment.

Figure 2
Effect sizes of education on neuropsychological tests in the total sample. β partial regression coe cients, CI con dence interval, AVLT-H-N5 Auditory Verbal Learning Test-Huashan Version for long-  Predicted global composite z-scores. a For the total sample, b For the three subgroups. HC healthy controls, SCD subjective cognitive decline, CI cognitive impairment.

Figure 5
Predicted global composite z-scores among the three subgroups according to the APOE ε4 status. a APOE ε4 non-carriers, b APOE ε4 carriers. HC healthy controls, SCD subjective cognitive decline, CI cognitive impairment.