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 ultra-early 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 [18F] florbetapir (AV-45) positron emission tomography (Aβ-PET) or [18F] fluorodeoxyglucose (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 defined 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 defined 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 figure 1 as a flowchart.
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 (particularly including history of hypertension, diabetes, dyslipidemia, etc.), biochemical examination, results of a battery of neuropsychological tests and clinical diagnosis for each participant at the baseline and at different visits.
A standardized neuropsychological test battery was used to assess performance in 3 cognitive domains: episodic memory (Auditory Verbal Learning Test-Huashan version long-delayed free recall [AVLT-H-N5] and recognition [AVLT-H-N7]) [22], speed/executive function (Shape Trailing Test A [STT-A] and B [STT-B]) [23], language (Animal Fluency Test [AFT]; Boston Naming Test [BNT]) [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 first 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 amplified 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 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 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 significant level was set at p< 0.05 (2-sided).