APOE ε4 Allele Distribution and Association With Scores of Subjective Cognitive Decline Questionnaire 9 in a Large Chinese Memory Clinic Cohort

Background Previous reports on APOE ε4 allele distribution in different populations have been inconclusive. The Subjective Cognitive Decline-Questionnaire 9 (SCD-Q9) was developed to identify those at risk of objective cognitive impairment [OCI; including mild cognitive impairment (MCI) and dementia groups), but its association with APOE ε4 and discriminatory powers for SCDwith subtle cognitive decline (SCDs) and OCI in memory clinics are unclear. Objectives To investigate demographic distribution of APOE ε4, its association with SCD-Q9 scores, and its ability to discriminate SCDs and OCI groups from normal control (NC). Methods A total of 632 participants were recruited (NC = 243, SCDs = 298, OCI = 91). APOE ε4 allele distribution and association with SCD-Q9 scores were calculated and the effects on cognitive impairment were analyzed. Receiver operating characteristic (ROC) analysis was applied to identify discriminatory powers for NC, SCDs, and OCI. Results Total APOE ε4 frequency was 13.1%. This did not vary by demography but was higher in patients with OCI. The SCD-Q9 scores were higher in APOE ε4 carriers than non-carriers in the OCI group. The area under the curve (AUC) for discriminating from OCI using APOE ε4 were 0.587 and 0.575, using SCD-Q9 scores were 0.738 and 0.571 for NC and SCDs groups, respectively. When we combined APOE ε4 and SCD-Q9 scores into the model, the AUC increased to 0.747 for discriminating OCI from NC. However, when OCI group was split into MCI and dementia groups, only total SCD-Q9 score was the independent affecting factor of MCI. Conclusion This study demonstrated that the distribution of APOE ε4 alleles did not vary with different demographic characteristics in a large-scale cohort from a memory clinic. APOE ε4 alleles may be associated with scores of SCD-Q9 reflecting the degree of cognitive complaints but their additional contribution to SCD-Q9 scores is marginal in discriminating between NC, SCDs, and OCI.


INTRODUCTION
Previous studies have reported that mutation of Apolipoprotein E (APOE) and regulation of its expression have an important connection with Alzheimer's disease (AD) dementia (Kim and Tsai, 2009) because of the pivotal role of APOE in lipoprotein metabolism in the brain. Among the three alleles of APOE (ε2, ε3, and ε4), presence of ε4 can increase the risk of AD by approximately 3-(single allele) to 15-fold (double alleles) (Saunders et al., 1993;Kim and Tsai, 2009;Koffie et al., 2012).
An earlier meta-analysis and review of the distributions of APOE alleles showed that APOE ε4 alleles were the second most common allele (besides APOE ε3) (Farrer et al., 1997;Abondio et al., 2019), but distributions vary with age, gender, and ethnicity (Eisenberg et al., 2010;Kern et al., 2015;Le Couteur et al., 2020). Among cognitively normal subjects aged 21-97 years, APOE ε4 carriers were younger than non-carriers, but no significant differences were found in education level and gender (Caselli et al., 2009). Farrer et al. (1997) demonstrated that APOE ε4 was associated with the risk of AD from 40 to 90 years, but the relationship diminished after 70 years of age. They also reported that the frequency of the APOE ε4 allele in males was lower than that in females among the elderly without dementia (Lehmann et al., 2006). In addition, previous studies investigating different ethnic groups revealed that Asian populations might have lower APOE ε4 frequencies than Oceania in the natural population (Singh et al., 2006;Eisenberg et al., 2010), and Chinese cohorts presented relatively low frequencies of APOE ε4 alleles (Katzman et al., 1997), but the conclusions need to be confirmed by more research.
More recently, researchers have also focused on understanding the relevance of APOE ε4 prevalence and the early stages of cognitive impairment, such as subjective cognitive decline (SCD), which is an intermediate stage between mild cognitive impairment (MCI) and normal cognition. A systematic review (Ali et al., 2018) (n = 36 articles) showed that the frequency of APOE ε4 was significantly lower in healthy controls than in groups with objective cognitive impairment (OCI), including MCI and AD dementia, but showed no difference in the SCD group, suggesting that APOE ε4 may not be directly related to the development of SCD. However, another earlier meta-analysis included a total of 28 studies that indicated a weak positive correlation between APOE ε4 and SCD (Zhang et al., 2017). This inconsistency may be due to different study designs (e.g., age) and the smaller sample size, especially in memory clinics, most of which evaluated less than 100 subjects (Striepens et al., 2010;Fortea et al., 2011). However, memory clinics, as the primary setting for individuals with memory complaints seeking care, are crucial for the early identification of those at risk of cognitive decline.
Additionally, since the standardization of SCD with cognitive complaints but unimpaired cognition (Jessen et al., 2014) in 2014, there is growing evidence demonstrating that subtle cognitive decline is already present in populations with SCD (Kielb et al., 2017;Jessen et al., 2018;Hao et al., 2019). To improve the likelihood of preclinical AD diagnosis, a more sensitive and reliable neuropsychological standard for subtle cognitive decline (Jak/Bondi) was developed in 2015 (Edmonds et al., 2015). However, the relationship between SCD with subtle cognitive decline (SCDs) diagnosis based on the two updated criteria and APOE ε4 within a large cohort has not been reported.
Finally, in recent years the SCD-Questionnaire 9 (SCD-Q9), developed as an easy and quick screening tool, was used to identify patients with SCD at risk of OCI at an early stage using advanced statistical methods (Gifford et al., 2015). Since its development, several studies (Alber et al., 2018;Bott et al., 2018;Kumar et al., 2018;Hao et al., 2020) have applied it to assess changes in memory complaints and defined SCD. Previous evidence has also demonstrated the links between the subjective observations included in SCD-Q9 and objective pathological alterations (Amariglio et al., 2012). APOE ε4 as an objective biomarker of AD has been universally accepted. However, the association between SCD-Q9 scores and APOE ε4 and their predictive powers for OCI with SCD complaints are not clear. Elucidation of this association would help us better understand the biomarkers of AD reflected by subjective cognitive complaints. The combination of the two methods may help us to more quickly and accurately identify early AD patients and could also reduce the economic burden of the society and families. Therefore, our current investigation studies larger cohorts in memory clinic settings, mainly aimed to (1) investigate the distribution characteristics of APOE ε4 alleles in different demography and cognitive impairment groups diagnosed based on updated criteria (SCD-I) (Jessen et al., 2014) and Jak-Bondi (Edmonds et al., 2015); (2) analyze the association of APOE ε4 and scores of SCD-Q9 reflecting cognitive impairment; and (3) assess the discriminatory powers of APOE ε4 alleles themselves and their combination with scores of SCD-Q9 for diagnosing cognitive impairment.

Subject Recruitment
Six hundred thirty-two individuals participated our study, SCDs, MCI, and AD dementia patients were recruited at first routine visits to the memory clinic of the Neurology Department, Xuanwu Hospital, Capital Medical University and normal control (NC) subjects were recruited from communities in Beijing, China from March, 2017 to January, 2020. The details of the study, including its purpose, procedure and contact information, was advertised in the memory clinic and via broadcasting at large-scale gatherings in the communities. People were asked for their consent to join the study.

Inclusion Criteria for All Subjects
Jak/Bondi criteria as follows (Bondi et al., 2014;Edmonds et al., 2015) were used for the diagnosis of MCI and SCDs.
Mild cognitive impairment was assigned when (1) the answer needed to be "yes" to the question "Do you have a problem with your memory?"; (2) scores of two measures in the same cognitive domain were >1.0 standard deviation (SD) below the normative mean; or (3) scores of at least one measure in each of the three cognitive domains (Memory, Execution, and Language) were >1.0 SD below the normative mean; (4) failure to meet the criteria of dementia; and (5) ADL had to be normal.
For the diagnosis of SCDs, the following requirement was to be met: (1) the answers needed to be "yes" to both of the questions "Do you have a problem with your memory?" and "Are you concerned about your memory?"; (2) subtle cognitive decline was observed in the neuropsychological examination, indicated by the decreased score of two measures in different cognitive domains (>1.0 SD below the normative mean); (3) failure to meet the criteria of MCI; and (4) ADL was normal.
The diagnosis of mild AD dementia fulfilled standardized diagnostic criteria (Mckhann et al., 1984;American Psychiatric Association, 1994;Dubois et al., 2007): (1) met the diagnostic criteria of dementia; (2) gradual and progressive decline in memory function over more than 6 months; (3) impaired episodic memory revealed by the objective testing listed above; (4) impaired basic and elementary functioning for ADL; (5) CDR = 1; and (6) hippocampal atrophy confirmed by structural magnetic resonance imaging (MRI).

Groupings
The population was divided into three groups according to these listed diagnostic criteria: (1) NC was assigned when participants did not have SCD complaints (the answers needed to be "no" to both of the questions "Do you have a problem with your memory?" and "Are you concerned about your memory?") and mild AD dementia, MCI, or SCDs, and had normal ADL scores; (2) SCDs group; and (3) the OCI group included people who were diagnosed with MCI and mild AD dementia.

Exclusion Criteria for All Subjects
(a) A history of stroke; (b) severe depression (HAMD >30), and other psychiatric disorders or current psychotropic drugs treatment; (c) other central nervous system diseases that could cause cognitive decline (e.g., brain tumors, Parkinson's disease, encephalitis, or epilepsy); (d) other systemic diseases which could cause cognitive decline (e.g., alcoholism, thyroid dysfunction, severe anemia, syphilis, HIV, or vitamin B 12 abnormalities); (e) a history of psychosis or congenital mental growth retardation; (f) cognitive decline caused by traumatic brain injury; (g) use of anti-dementia agents in SCDs, MCI, and; or (h) those who could not complete neuropsychological tests or with contraindication to MRI.

APOE Genotyping
DNA sequences for each subject were extracted for SNPs rs7412 and rs429358 from the APOE ε2/ε3/ε4 haplotype. 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 amplified using the following conditions: 1 cycle of 98 • C for 10 s, 35 cycles of 72 • C for 5 s, and 1 cycle of 72 • C for 5 min. PCR was performed in a final 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 the GC Buffer (Takara Bio).

Statistical Analysis
We conducted all analyses using the Statistical Package for the Social Sciences version 17.0 (SPSS Inc., Chicago, IL, United States). Descriptive statistics (APOE alleles and scores of SCD-Q9) were calculated by percentages or mean ± SD (x ± S) or median (percentile 25, 75). The x 2 or T-test or Mann-Whitney test was used to assess group differences, and p < 0.05 was considered to be statistically significant. For three groups comparison, p < 0.05 was considered to be statistically significant and corrected p' value (p < 0.017) was used in the partitions of Pearson's Chi-square statistics. To examine the potential affecting factors of SCDs and OCI, we performed univariate and binary logistic regression analysis (Supplementary Materials 1). More specifically, we used APOE ε4 alleles and scores of SCD-Q9 that significantly differed between two groups as the independent variables, and "diagnosis" as the dependent variable. Besides, odds ratios (ORs) were calculated for the two variables. p < 0.05 was required for variables to be in the model. Finally, we obtained the receiver operating characteristic (ROC) curves and calculated area under the curves (AUCs) for the factors.

Distribution of APOE ε4 Alleles and Genotypes in the Total Population
In total, 632 individuals were recruited in our study, including 218 (34.5%) males and 414 (65.5%) females, and the mean age and education years were 65.4 ± 6.76 and 12.4 ± 3.21 years, respectively. The proportions of APOE genotypes were listed in Table 1.
The proportions of APOE ε3/4 and ε2/4 were 20.3% and 2.8%, respectively. The proportion of homozygous ε4 was <2.0%. For the frequencies of the APOE alleles ε4 was 13.1% (see Table 1). The Demography of APOE ε4 Carriers and Non-carriers For age, gender, and years of education, we did not find significant differences between APOE ε4 carriers and non-carriers (p > 0.05) (see Table 2).

Distribution of APOE ε4 and Subjective
Cognitive Decline-Questionnaire 9 Scores of Carriers and Non-carriers in Normal Control, SCDs, and Objective Cognitive Impairment Groups The results showed that the difference in the proportion of APOE ε4 carriers among the three groups (NC, SCDs, and OCI) was significant (p = 0.004). Further pairwise comparison showed that the OCI group had a higher proportion of APOE ε4 carriers than the NC (p = 0.001) and SCDs groups (p = 0.005) at corrected test level p , but no significant difference was found between the other two groups (p = 0.487) (see Table 3). When the total population was divided into male and female subgroups, the latter showed consistent results with the total population (see Table 4), but we did not find any significant differences among NC, SCDs, and OCI groups in the male subgroup (see Table 5).
For SCD-Q9 scores, we also found that APOE ε4 carriers scored higher than non-carriers in the OCI group (p = 0.022). However, there were no significant differences in SCD-Q9 scores in the total population (p = 0.057), NC (p = 0.766), and SCDs groups (p = 0.703) between ε4 carriers and noncarriers (see Table 3).Then, we divided the total population into female and male subgroups. For the female subgroup, the results were consistent with the total population (see Table 4). However, a significant difference was found in SCD-Q9 scores between ε4 carrier and non-carriers in the males (p = 0.032), which was different from the female subgroup and the total population. Also, we did not find any significant difference in SCD-Q9 scores between ε4 carriers and non-carriers in different cognitive groups, including NC (p = 0.220), SCDs (p = 0.709), and OCI groups (p = 0.275) (see Table 5).

Logistic Regression Models for Normal Control, SCDs, and Objective Cognitive Impairment Groups
The results of the binary logistic regression analysis for NC and OCI indicate carrier status of APOE ε4 and total SCD-Q9 score were independent risk factors of OCI [OR: 2.050, 95% CI (confidential interval): 1.161-3.620, p = 0.013, and OR: 1.444, 95% CI: 1.285-1.622, p < 0.001, respectively]. When we split OCI group into MCI and AD dementia groups, only total SCD-Q9 score was the independent risk factor of MCI [OR: 1.390, 95% CI: 1.232-1.568, p < 0.001], whereas carrier status of APOE ε4 did not show any relationship (p = 0.191).
Our results also showed that carrying the APOE ε4 allele (OR: 1.960, 95% CI: 1.184-3.243, p = 0.009) was a risk factor for OCI compared with SCDs, whereas scores of SCD-Q9 did not show any relationship (p = 0.153).When we split OCI group into MCI and AD dementia groups, carrying the APOE e4 allele and total SCD-Q9 score were not affecting factors for MCI compared with SCD (p = 0.265 and p = 0.792, respectively).
Then, we included scores of SCD-Q9 and APOE ε4 carrier status together in the model and calculated AUCs for group discrimination. The AUCs increased to 0.747 (95% CI: 0.685-0.808, p < 0.001) for the NC and OCI groups, and 0.593 TABLE 3 | Distribution of APOE ε4 and SCD-Q9 scores of carriers and non-carriers in NC, SCDs, and OCI groups.

DISCUSSION
In the current study, we first reported the distribution characteristics of APOE ε4 alleles in a Chinese memory clinic with a larger cohort. To the best of our knowledge, this is also the first study to reveal the associations of APOE ε4 and SCD-Q9 scores in subjective and OCI diagnosed based on the combination of SCD-I (Jessen et al., 2014) and the Jak/Bondi standards (Edmonds et al., 2015). These results will help us better understand the variation in the unfavorable effect of APOE ε4 on the disease progression of AD and identify individuals with higher risk of cognitive decline in order to intervene earlier and treat more effectively.
Our findings showed that the frequency of APOE ε4 was 13.1%, following the APOE ε3 (78.5%), as the second most common allele. This is similar to the previous report based on the worldwide distribution (Farrer et al., 1997), showing APOE ε4 and APOE ε3 with a frequency of 13.7 and 77.9%, respectively. However, compared with a study in China by Li et al. (2002), we reported a higher frequency of APOE ε4 (13.1 vs. 9.7%), which may be due to the different recruitment protocol and more patients with OCI participating in our study. Participants with more memory impairment were included in our investigation, which may have resulted in a higher carrier rate of APOE ε4. Recently, a higher frequency of APOE ε4 alleles (19.6%) was reported in a population with cognitive impairment from a Chinese memory clinic, which supports our assumption (Wang et al., 2014). Moreover, compared with Oceania, South Africa, and Europe, such as Australia (26.0%) (Kamboh et al., 1991), Khoi San (37.0%) (Sandholzer et al., 1995), and NE England (15.4%), respectively (Mastana et al., 1998), a numeric lower frequency of APOE ε4 alleles was found in our study, which was also consistent with the previous conclusions that the frequency of APOE ε4 varied among different ethnicities and that Asian populations have a relatively lower ε4 frequency (Singh et al., 2006;Eisenberg et al., 2010). This also may be one of the reasons that the worldwide distribution of APOE ε4 appeared at a relatively low level.
Second, we analyzed the distribution characteristics of the APOE ε4 alleles in different demographics. The results showed that their frequencies did not vary with age, which was in agreement with the previous studies (Nakayama and Kuzuhara, 1999;Thelma et al., 2001). However, results from Alzheimer's Disease Neuroimaging Initiative (ADNI) cohorts and Uniform Data Set of the Alzheimer's Disease (UDS) Centers and Australian Imaging, Biomarkers and Lifestyle Flagship Study of Ageing (AIBL), both demonstrated significant correlations between APOE ε4 and aging (Heffernan et al., 2016). This contradiction may be attributed to uncertainties inherent in the study design. For instance, several confounding factors of AD in the UDS and ADNI datasets, such as severe heart disease and diabetes, were excluded from our study. Studies have also reported that the frequencies of APOE ε4 decreased with aging (Jian-Gang et al., 1998;Bonham et al., 2016;Liu and Caselli, 2018;Bellou et al., 2020;Le Couteur et al., 2020) and the risk mainly declined after 75 years of age (Bickeboller et al., 1997;Jian-Gang et al., 1998;Bonham et al., 2016;Liu and Caselli, 2018). In the current study, only 9.3% of individuals older than 75 years agreed to participate in our investigation which may have contributed to this discrepancy. At present, knowledge of education and APOE ε4 alleles is limited. To the best of our knowledge, only one study previously investigated this association. Their results indicated that the relationship was not correlated (Caselli et al., 2009) which is consistent with findings from the current study. Lastly, we did not find a significant difference between gender and APOE ε4 allele distribution, which is consistent with previous findings (Combarros et al., 1998;Vaisi-Raygani et al., 2007;Tsolaki et al., 2018).
Finally, our study reported the distribution of APOE ε4 alleles in different cognitive groups and their associations with the SCD-Q9 scores. The most robust findings have demonstrated that the presence of the APOE ε4 allele imparts a genetic risk for the development of cognitive impairment, specifically that related to AD and vascular dementia (Allan and Ebmeier, 2011;Pink et al., 2015;Chen et al., 2016a;Jiang et al., 2016;Liu et al., 2016). Ali et al. (2018) suggested that the frequencies of carrying the APOE ε4 allele were comparable between healthy controls and SCD samples but were significantly higher in objectively impaired samples (i.e., MCI and AD dementia). In our study, the frequencies of APOE ε4 allele in the NC and SCDs groups were lower than that of the OCI groups in the total population and the female subgroup, but we did not find a difference between the NC and SCDs groups, which was in line with Ali et al.'s (2018) conclusion. In addition, a previous study found that individuals with subjective memory complaints (SMC) with no objective memory impairment did not differ from the NC group in terms of the frequency of APOE ε4 alleles (Lautenschlager et al., 2005). The results of our study did not provide supportive evidence for a positive association between APOE ε4 and SCD. However, a study has previously reported higher APOE ε4 frequency in the SCD group than in normal controls (Jessen et al., 2018), and this inconsistency may be due to different study populations (i.e., gender and ethnicity) and diagnostic criteria. The German Center for Neurodegenerative Diseases (DZNE)-Longitudinal Cognitive Impairment and Dementia Study (DELCODE) enrolled subjects who speak fluent German and defined SCD based on the SCD-I diagnostic frame, which differs from ours. Also, the relatively higher prevalence of SCD but lower conversion rate to OCI in China due to low level of education and income (Wang et al., 2000;Rohr et al., 2020;Si et al., 2020) may result in different conclusions.
In our study, a significant difference in SCD-Q9 scores was found between APOE ε4 allele carriers and non-carriers in OCI in the total population and female subgroup. These results indicate that APOE ε4 alleles may be partially reflected in SCD-Q9 scores in patients with OCI in total population and females. Meanwhile, APOE ε4 allele carriers also presented a higher score of SCD-Q9 than non-carriers in the total population and SCDs group, but the differences were not significant. This is the first attempt to explore the relationship between APOE ε4 and SCDs diagnosed based on the combination standards of SCD-I and Jak/Bondi, which needs to be further verified by follow-up studies. Further attention should also be paid to other ethnicities and cohorts to verify this association in the future. Finally, the results of logistic regression analysis showed that APOE ε4 allele was risk factor for the OCI group (MCI and AD dementia) but not for the NC and SCDs groups, although the predictive powers were smaller. However, we found AUCs of SCD-Q9 alone were 0.727 for the NC and SCDs groups, 0.738 for the NC and OCI groups, and 0.571 for the SCDs and OCI groups, respectively. When APOE ε4 carrier status and SCD-Q9 scores together were added to the model, the AUC increased to 0.747 for NC and OCI groups, and 0.593 for SCDs and OCI groups, suggesting that the predictive power of APOE ε4 is limited, especially when OCI group was split into MCI and dementia groups, their discriminating powers for MCI and NC were marginal, but could be increased by combining with scores of SCD-Q9. Previous studies reported that APOE ε4 status appeared to be a more predictive risk factor for progression from MCI to AD dementia than family history, age, gender, or education (Fleisher et al., 2007), but it was only a useful predictor of progression from 70 to 85 years of age while controlling for education, memory scores, and gender (Devanand et al., 2005). Another study reported that ε4 carriers with SMC showed altered AD-related cerebrospinal fluid and fluorodeoxyglucose-positron emission tomography (PET) measures (Mosconi et al., 2008); In addition, they demonstrated that aging, APOE ε4, and SMC were associated with high Aβ burden, indicating that selection based on the presence of SMC and APOE ε4 may help identify healthy elderly participants with high Aβ burden eligible for secondary prevention trials (Zwan et al., 2016). As encouraging as these results may be, the exact role played by APOE ε4 in the development of AD dementia or other OCI continues to be unclear due to a lack of convergent evidence and considerable sample heterogeneity (Ali et al., 2018). Consequently, further investigation is warranted before APOE ε4 genetic testing can be recommended for widescale clinical adoption as a viable diagnostic tool for pathological cognitive decline.
It should be noted that there were obvious limitations associated with this study. (1) Our study is a cross-sectional survey, and follow-up studies should be performed to further confirm the conclusions; (2) the diagnosis of subjective and OCI was not validated by other tests. For instance, it lacks the completeness of Aβ-PET, cerebrospinal fluid tau, or Aβ examinations, given that only parts of the included population underwent Aβ-PET; (3) finally, this study focused on APOE ε4 alleles; thus, no evidence was provided for other related biomarkers and imaging approaches; and (4) the small sample size of OCI group in current study restricts us to further confirm the relationship between APOE ε4 and SCD-Q9 after control the amount of cognitive impairment, and a larger cohort with MCI and mild AD dementia patients was needed to verify our conclusion.

CONCLUSION
In summary, we reported the distribution characteristics of APOE ε4 alleles in different demographics and levels of cognition, and their associations with scores of SCD-Q9 with a larger cohort from a Chinese memory clinic. The findings of this study indicate that clinicians should be attentive to the distributed variation of APOE ε4 alleles and their unfavorable effects on OCI with SCD complaints, but their additional contribution to SCD-Q9 scores is marginal in discriminating individuals with cognitive impairment from normal controls.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Xuanwu Hospital Capital Medical University. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
LH: conceiving, implementing, statistics, and writing the manuscript. JJ: revising the manuscript. YX and YH: conceiving and revising the manuscript. YH: providing funding support. All authors contributed to the article and approved the submitted version.

FUNDING
This study was supported by the National Natural Science Foundation of China (Grant 61633018).