jpad journal

AND option

OR option

DONANEMAB (LY3002813) PHASE 1B STUDY IN ALZHEIMER’S DISEASE: RAPID AND SUSTAINED REDUCTION OF BRAIN AMYLOID MEASURED BY FLORBETAPIR F18 IMAGING

 

S.L Lowe1, C. Duggan Evans2, S. Shcherbinin2, Y.-J. Cheng2, B.A. Willis2, I. Gueorguieva3, A.C. Lo2, A.S. Fleisher2, J.L. Dage2,4, P. Ardayfio2, G. Aguiar3, M. Ishibai5, G. Takaichi5, L. Chua1, G. Mullins2, J.R. Sims2 on behalf of AACD Investigators

 

1. Eli Lilly and Company, Lilly Singapore, Singapore; 2. Eli Lilly and Company, Indianapolis, Indiana, USA; 3. Eli Lilly and Company, Bracknell, UK; 4. Stark Neurosciences Research Institute, Indiana University School of Medicine, Indianapolis, IN, USA; 5. Eli Lilly Japan, K.K., Kobe, Japan.

Corresponding Author: John R. Sims, Eli Lilly and Company, Lilly Corporate Center DC 1532, Indianapolis, IN, 46285, Telephone: 317-655-2206,e-mail: sims_john_r@lilly.com

J Prev Alz Dis 2021;4(8):414-424
Published online September 21, 2021, http://dx.doi.org/10.14283/jpad.2021.56

 


Abstract

Background: Donanemab (LY3002813) is an IgG1 antibody directed at an N‑terminal pyroglutamate of amyloid beta epitope that is present only in brain amyloid plaques.
Objectives: To assess effects of donanemab on brain amyloid plaque load after single and multiple intravenous doses, as well as pharmacokinetics, safety/tolerability, and immunogenicity.
Design: Phase 1b, investigator- and patient-blind, randomized, placebo-controlled study.
Setting: Patients recruited at clinical research sites in the United States and Japan.
Participants: 61 amyloid plaque-positive patients with mild cognitive impairment due to Alzheimer’s disease and mild-to-moderate Alzheimer’s disease dementia.
Intervention: Six cohorts were dosed with donanemab: single dose 10-, 20- or 40- mg/kg (N = 18), multiple doses of 10-mg/kg every 2 weeks for 24 weeks (N = 10), and 10- or 20-mg/kg every 4 weeks for 72 weeks (N=18) or placebo (N = 15).
Measurements: Brain amyloid plaque load, using florbetapir positron emission tomography, was assessed up to 72 weeks. Safety was evaluated by occurrence of adverse events, magnetic resonance imaging, electrocardiogram, vital signs, laboratory testing, neurological monitoring, and immunogenicity.
Results: Treatment with donanemab resulted in rapid reduction of amyloid, even after a single dose. By 24 weeks, amyloid positron emission tomography mean changes from baseline for single donanemab doses in Centiloids were: -16.5 (standard error 11.22) 10-mg/kg intravenous; 40.0 (standard error 11.23) 20 mg/kg intravenous; and -49.6 (standard error 15.10) 40-mg/kg intravenous. Mean reduction of amyloid plaque in multiple dose cohorts by 24 weeks in Centiloids were: 55.8 (standard error 9.51) 10-mg/kg every 2 weeks; -50.2 (standard error 10.54) 10-mg/kg every 4 weeks; and -58.4 (standard error 9.66) 20-mg/kg every 4 weeks. Amyloid on average remained below baseline levels up to 72 weeks after a single dose of donanemab. Repeated dosing resulted in continued florbetapir positron emission tomography reductions over time compared to single dosing with 6 out of 28 patients attaining complete amyloid clearance within 24 weeks. Within these, 5 out of 10 patients in the 20 mg/kg every 4 weeks cohort attained complete amyloid clearance within 36 weeks. When dosing with donanemab was stopped after 24 weeks of repeat dosing in the 10 mg every 2 weeks cohort, florbetapir positron emission tomography reductions were sustained up to 72 weeks. For the single dose cohorts on day 1, dose proportional increases in donanemab pharmacokinetics were observed from 10 to 40 mg/kg. Dose proportional increases in pharmacokinetics were also observed at steady state with the multiple dose cohorts. Donanemab clearance was comparable across the dose levels. Mean donanemab elimination-half-life following 20 mg/kg single dose was 9.3 days with range of 5.6 to 16.2 days. Greater than 90% of patients had positive treatment-emergent antidrug antibodies with donanemab. However, overall, the treatment-emergent antidrug antibodies did not have a significant impact on pharmacokinetics. Donanemab was generally well tolerated. Amongst the 46 participants treated with donanemab, the following amyloid-related imaging abnormalities, common to the drug class, were observed: 12 vasogenic cerebral edema events (12 [19.7%] patients), 10 cerebral microhemorrhage events (6 [13.0%] patients), and 2 superficial siderosis events (2 [4.3%] patients).
Conclusions: Single and multiple doses of donanemab demonstrated a rapid, robust, and sustained reduction up to 72 weeks in brain amyloid plaque despite treatment-emergent antidrug antibodies detected in most patients. Amyloid-related imaging abnormalities were the most common treatment-emergent event.

Key words: Alzheimer’s disease, amyloid plaque, donanemab, florbetapir PET, immunogenicity.


 

Introduction

The deposition of amyloid-beta peptide (Aβ) is essential to the pathophysiology and progression of Alzheimer’s disease (AD) (1), and thereby has led to the discovery and development of active and passive immunotherapies with mechanisms of action that reduce Aβ accumulation in the brain (2). Some of the initial active immunotherapies targeted at brain amyloid plaques were associated with a high rate of unacceptable adverse events in clinical trials (e.g., meningoencephalitis (3).
Donanemab (LY3002813) is an immunoglobulin IgG1 antibody directed at an N-terminal pyroglutamate Aβ epitope that is present only in brain amyloid plaques. Donanemab was developed to remove existing amyloid plaques through microglial-mediated phagocytosis. Administration of the murine surrogate of donanemab in aged amyloid precursor protein transgenic mice resulted in dose-dependent plaque reduction without microhemorrhage liability (4). In the first-in-human single-dose and multiple-dose, placebo-controlled, dose-escalation Phase 1a study, donanemab 10-mg/kg was associated with 40–50% reductions in amyloid plaque deposits in amyloid-positive patients with mild cognitive impairment (MCI) due to AD or mild to moderate AD dementia (5). Overall, donanemab was generally well tolerated up to 10-mg/kg in this Phase 1a study. The most common treatment-emergent adverse events among 51 donanemab-treated participants were mild-to-moderate infusion reactions (6 of 37 patients with AD who had IV dosing) and asymptomatic cerebral microhemorrhage (2 out of 51 donanemab treated participants). No cases of vasogenic cerebral edema (ARIA-E) were reported. Approximately, 90% of participants developed anti-drug antibodies at 3 months following a single intravenous dose (5).
Based on the positive safety and pharmacodynamic (PD) findings from the Phase 1a study (5), a second Phase 1 study was initiated with donanemab in patients with MCI due to AD or mild to moderate AD. The overall goal of this Phase 1b study was to determine whether different dosing regimens (single-dose, dosing frequency, and chronic dosing for maximal PD effect) could mitigate immunogenicity, potential immune safety issues and produce sustained amyloid reduction. The primary objective was to assess the effect of donanemab on brain plaque load using florbetapir positron emission tomography (PET) imaging. The secondary objectives were to assess the safety, pharmacokinetics (PK), immunogenicity, and cognitive function effects of donanemab following single intravenous (IV) and multiple IV doses.

 

Methods

Study Design and Treatment

This Phase 1b study was conducted between December 22, 2015 and July 08, 2020 at 8 clinical research centers in the United States and Japan among patients with MCI due to AD or mild to moderate AD (6, 7). The study was a 3-part, patient- and investigator-blind, randomized within cohort, placebo-controlled, parallel-group, single- and multiple-dose study. Each of the 6 cohorts was designed to include approximately 6 (single dose) or 9 (multiple dose) patients treated with donanemab and 2 to 3 patients treated with placebo. Patients in Cohorts 1–3 were each administered a single, IV dose of donanemab (Cohort 1: 10-mg/kg, Cohort 2: 20-mg /kg, Cohort 3: 40-mg/kg) or placebo (Supplemental Figure 1). Follow-up was 72 weeks (Cohorts 1 and 2) or 24 weeks (Cohort 3). Patients in Cohort 4 were each administered multiple IV doses of donanemab (10-mg/kg) or placebo every 2 weeks (Q2W) for up to 24 weeks followed by a 48-week follow-up period to obtain amyloid clearance and safety data. Patients in Cohorts 5 and 6 were each administered multiple IV doses of donanemab (Cohort 5: 10-mg/kg ; Cohort 6: 20-mg/kg) or placebo every 4 weeks (Q4W) for up to 72 weeks followed by a 12-week follow-up period.

Study Population

The study enrolled men or nonfertile women ≥50 years of age with evidence of memory impairment on the Free and Cued Selective Reminding Test with Immediate Recall (FCSRT-IR, picture version; <27 for free recall), a Mini–Mental State Examination (MMSE) score of 16 to 30, a Clinical Dementia Rating (CDR) of 0.5 to 2 and memory box score ≥0.5, and a florbetapir PET scan consistent with the presence of amyloid pathology (as determined using visual assessments and composite standardized uptake value ratio [SUVr] cut-points). The florbetapir F 18 interpretation method used for the eligibility decision included quantification as an adjunct to a visual assessment. The PET imaging core lab was responsible for performing both visual and quantitative analysis of the florbetapir F 18 images. All patients had gradual and progressive change in memory function reported by the patients themselves or informants over a period of more than 6 months. Patients with contraindication for magnetic resonance imaging (MRI), presence of more than four microhemorrhages on MRI, or history or evidence on MRI of macrohemorrhage were excluded. In addition to US patients, Japanese patients were included in this study to explore the safety and PK of donanemab in patients of Japanese decent.

Study Evaluations

Florbetapir PET scans were performed at baseline and at 12, 24, 36, 48, and 72 weeks after starting treatment to estimate mean change in amyloid plaques. Calculation of SUVr with cerebellum reference region was as described previously (8). The latter SUVr values were converted to Centiloid units (9). Volumetric measurements were obtained from structural (3D T1-weighted) MR images acquired at screening and at 24, 48, and 72 weeks. Volume and atrophy were assessed in multiple brain regions including whole brain, lateral ventricles, and hippocampus.
Apolipoprotein E (APOE) genotyping was performed at baseline to determine genetic variants that may influence response to treatment.
Emergence of antibodies against donanemab was evaluated to asses the immunogenicity risk. Antidrug antibodies (ADAs) were detected using an affinity capture elution (ACE) Bridge assay validated at BioAgilytix Labs in Durham, North Carolina, USA. The ACE Bridge immunogenicity assay was developed based on published methods (10-13). Serum for determination of ADAs was collected at screening/baseline and then at regular intervals throughout the study period.
Safety in the study was assessed at regular intervals with MRIs, electrocardiograms, safety laboratory tests (clinical chemistry, hematology, and urinalysis), physical/neurological examinations, and by monitoring the occurrence of adverse events, vital signs, and immunogenicity. In addition, the Columbia Suicide Severity Rating Scale (child version) and (if applicable) the Self-Harm Supplement Form were completed prior to dosing and at most study visits.
Cognition was assessed at screening or baseline for all patients using the CDR, the MMSE, the FCSRT-IR, the Alzheimer’s Disease Assessment Scale Cognitive Subscale (ADAS-Cog-14), the Alzheimer’s Disease Cooperative Study-Mild Cognitive Impairment-Activities of Daily Living, 24-item questionnaire (ADCS-MCI-ADL-24), and the Neuropsychological test battery (NTB). Additionally, these assessments were also performed at 24, 48, and 72 weeks after starting treatment or at the end of the study (eg, Week 24 for Cohort 3) or upon early discontinuation.

Bioanalytical Methods

Serum and CSF samples were evaluated for donanemab using a validated enzyme-linked immunosorbent assay method at Covance Laboratories in Chantilly, Virginia, USA. The lower and upper limit of quantification for the serum assay was 200 ng/mL and 5000 ng/mL, respectively. During validation, the inter-assay accuracy (% relative error) ranged from -1.5%–7.0% and -2.9–5.3% and the inter-assay precision (% relative standard deviation) was 4.0–9.7% and 5.2–8.7%.

Pharmacokinetic and Pharmacodynamic Analyses

Serum PK parameter estimates were calculated by standard noncompartmental methods using Phoenix WinNonlin Version 6.3 (Certara L.P., Raleigh, North Carolina, USA). Parameters estimated after IV administration included maximum observed drug concentration (Cmax), area under the concentration versus time curve (AUC) from time 0 to time infinity (AUC(0-∞)), and terminal half-life (t1/2). Mean plasma concentration versus time profiles and summary statistics of PK parameter estimates by treatment group were generated. To evaluate the potential effect of anti-donanemab antibodies on PK, observed trough donanemab concentrations were plotted by dose separately with time-matched anti-donanemab antibody results. A sample collection time window of 168–672 hours (1–4 weeks) and 168–1344 hours (1–8 weeks) from the most recent dose was used to identify trough concentrations for the Q2W and Q4W dosing regimens, respectively. Samples of CSF and serum were collected at baseline and approximately 72 hours following donanemab administration for the single dose cohorts or at baseline and approximately 72 hours following the dose administered at Week 24 for the multiple dose cohorts and assessed for donanemab concentration. These concentrations were compared to calculate a CSF:serum concentration ratio.
Composite SUVr from florbetapir scans were analyzed to estimate change (14) in amyloid burden. Furthermore, those SUVr values were converted to the Centiloid scale, a standardized methodology to quantify amyloid burden from PET scans (9).

Statistical Analysis

This study intended to enroll approximately 72 patients, a sample size that is customary for studies evaluating safety, PK, and/or PD parameters. Based on prior clinical trials conducted by the sponsor, randomizing 6 patients to each donanemab dose was expected to provide approximately 90% power to detect 17% mean florbetapir SUVr reduction of a dose compared to placebo without multiple comparison adjustment.
The demographic variables, other baseline characteristics, and safety parameters were summarized using standard descriptive statistics. Safety analyses were conducted for all enrolled patients, whether or not they completed all protocol requirements.
PD analyses were conducted on the full analysis set, which included all data from all randomized patients receiving at least one dose of the investigational product according to the treatment the patients actually received. The PD measures included florbetapir PET scans in Centiloid units and were analyzed using a mixed model repeated measure (MMRM) with fixed effects of treatment doses, study visit, interaction between treatment and visit, baseline amyloid PET scan (Centiloid unit), and APOE-ε4 status (carrier /non-carrier) as covariate adjustment. An unstructured covariance matrix was used to model the within-subject variance-covariance errors.
Immunogenicity evaluation was based on antibody formation, that was summarized over time. Treatment-emergent ADAs (TE-ADAs) were defined as those with a titer 2-fold (1 dilution) greater than the minimum required dilution if no ADAs were detected at baseline or those with a 4-fold (2 dilutions) increase in titer compared to baseline if ADAs were detected at baseline. The minimum required dilution of the ADA assay was 1:5.
Cognitive outcomes (CDR, MMSE, FSCRT-IR, ADAS-Cog-11, ADCS-MCI-ADL-24, and NTB) were analyzed using a MMRM with baseline cognitive measures as a baseline covariate, fixed-effects of dose, visit, the dose-visit interaction, and appropriate covariance structures for model convergence. Statistical analyses were performed using SAS EG 9.4 software.

 

Results

Demographics and Baseline Characteristics

For patients receiving at least 1 dose of study drug, the demographic and baseline characteristics were generally balanced across the treatment groups (Table 1). A total of 61 patients (donanemab, n = 46; placebo, n = 15) participated in this study. Patients were male (n =27) and female (n = 34) with a mean age of 73.2 years (range: 54 to 90 years). Forty-three (70.5%) patients were non-Japanese and 18 (29.5%) patients were Japanese. At baseline, the mean MMSE total score was 21.1 (Standard Deviation [SD] = 4.04) and the mean florbetapir PET Centiloid units was 104.5 (SD = 32.77). Seventy-seven percent (47 of 61) of patients were APOE-ε4 carriers (11 homozygotes and 36 heterozygotes).

Table 1. Demographic and Baseline Characteristics

*Data from the single dose, Q2W, and Q4W placebo arms were pooled; Abbreviations: APOE = apolipoprotein E; MMSE = Mini–Mental State Examination; N = number of patients; n = number of patients in a subgroup; PET = positron emission tomography; Q2W = every 2 weeks; Q4W = every 4 weeks; SD = standard deviation.

 

Disposition

Among 276 patients screened, 61 patients satisfied entry criteria and were enrolled into the study (7, 7, and 4 patients were randomized to the 10-mg/kg, 20-mg/kg, and 40-mg/kg single dose cohorts respectively; 10 patients were randomized to the 10-mg/kg Q2W for 24 weeks cohort and 8 and 10 patients were randomized to the 10-mg/kg Q4W and 20-mg/kg Q4W cohorts respectively). For simplicity, all patients receiving placebo were pooled into one group. Main reasons for screen failure were not meeting threshold criteria for amyloid PET (40 of 154 patients; 26.0%), cognition (MMSE/FCSRT-IR; 33 of 154 patients; 21.4%), and microhemorrhage greater than 4 on MRI (16 of 154 patients; 10.4%). Of the 61 patients who received at least 1 dose of study treatment, 46 (75.4%) patients completed the study (Supplemental Figure 2). Fifteen patients did not complete the study, which included 6 due to investigator decision (3 in the 10-mg/kg Q4W cohort and 3 in the 20-mg/kg Q4W cohort); 5 due to the patient’s withdrawal of consent (1 in the 10-mg/kg single dose cohort, 1 in the 10-mg/kg Q2W cohort, 1 in the 10-mg/kg Q4W cohort, 1 in the 20-mg/kg Q4W cohort, and 1 placebo); 3 patients discontinued due to adverse events (ARIA-E [20-mg/kg Q4W cohort], hypertensive crisis [20-mg/kg Q4W cohort] and myocardial infarction, considered a serious adverse event, resulting in death [placebo Q4W cohort]); and 1 patient was lost to follow-up (20-mg/kg single dose cohort).

Florbetapir Positron Emission Tomography – Centiloid Scale and Standardized Uptake Value Ratio

Single and multiple doses of donanemab showed a consistent reduction from baseline in cerebral amyloid (Centiloid units) observed by PET from Week 12 through Week 72 (Figure 1). At Week 24, amyloid PET least squares mean Centiloid changes from baseline for single donanemab doses were: -16.5 (standard error [SE] = 11.22) 10-mg/kg IV; -40.0 (SE = 11.23) 20-mg/kg IV; and -49.6 (SE = 15.10) 40-mg/kg IV. In contrast, in the placebo group there was no significant reduction in florbetapir PET at 72 weeks (90.9 Centiloids at 72 weeks compared to 104.4 Centiloids at baseline). Corresponding Centiloid changes for multiple doses at Week 24 included: -55.8 (SE = 9.51) 10-mg/kg Q2W; -50.2 (SE = 10.54) 10-mg/kg Q4W; and -58.4 (SE = 9.66) 20-mg/kg Q4W. Patients in the 20 mg/kg Q4W cohort tended to achieve greater plaque reduction earlier in the study than patients in either of the 10 mg/kg multiple dose cohorts (Figures 1 and 2). After dosing, a sustained reduction of brain amyloid level without significant reaccumulation for up to 72 weeks was observed across all single- and multiple-dose cohorts.

Figure 1. LS mean change of florbetapir PET scans from baseline (Centiloid units) through Week 72 following single and multiple dosing of IV donanemab

Error bars = SE; *Treatment duration of 24 weeks; Abbreviations: IV = intravenous; LS mean = least squares mean; N = number of patients; PET = positron emission tomography; Q2W = every 2 weeks; Q4W = every 4 weeks; SE = standard error.

igure 2. Cerebral amyloid over time as measured by quantitative amyloid PET imaging (florbetapir SUVr). Absolute Centiloid value as calculated from SUVr

*Treatment duration of 24 weeks; Notes: Color indicates APOE-ε4 status and symbol indicates ADA titer of ≥1:5120. The black dashed horizontal line indicates threshold Centiloid value for being amyloid positive; Abbreviations: APOE = apolipoprotein E; LY = LY3002813 (donanemab); PET = positron emission tomography; Q2W = every 2 weeks; Q4W = every 4 weeks; SUVr = standardized uptake value ratio.

 

The change in absolute Centiloid value did not appear to be influenced by APOE-ε4 status with no clear association between presence of the APOE-ε4 allele and florbetapir PET response (Figure 2). TE-ADAs (see below) also appeared not to impact the reduction in amyloid as some participants with high TE-ADA titers (≥1:5120) still had a reduction in amyloid in this study (Figure 2).
Overall, 2 participants in single-dose cohorts (1 in 20-mg /kg and 1 in 40-mg /kg) and 9 participants in the multiple-dose cohorts (2 in 10mg/kg Q2W; 2 in 10-mg /kg Q4W; and 5 in 20-mg /kg Q4W) achieved complete amyloid clearance status based on a threshold 24.1 Centiloid value. Most participants achieving amyloid clearance starting at 12 or 24 weeks remained amyloid negative for the duration of their florbetapir PET measurements.
Reduction in cerebral amyloid (Centiloid units) and SUVr changes from baseline were visually comparable between non-Japanese and Japanese patients (Supplemental Figure 3).

Single- and Multiple-Dose Serum and Cerebrospinal Fluid PK

Dose proportional increases were observed in both Cmax and exposure (AUC) following single and multiple doses. Single doses of 10, 20, and 40 mg/kg had measurable donanemab concentration for at least 56 days post-dose with elimination t1/2 of approximately 10 days. Multiple doses resulted in either no (10 mg/kg Q4W) or very limited exposure accumulation (10 mg/kg Q2W; 20 mg/kg Q4W). PK parameters for single and multiple dose cohorts are summarized in Supplemental Tables 1 and 2, respectively. Single dose PK characteristics were similar between Japanese and non-Japanese participants, albeit based on small sample size (5 patients who are Japanese out of 18 patients given donanemab). Quantifiable concentrations were detected in CSF samples collected from patients treated with single and multiple donanemab doses with CSF to serum concentration ratio of approximately 0.2% across all patients and dose levels.

Table 2. Treatment-Emergent Adverse Events

*Data from the single dose Q2W and Q4W placebo arms were pooled; †One patient reported 2 SAEs; Abbreviations: ARIA = amyloid-related imaging abnormalities; CNS = central nervous system; E = vasogenic cerebral edema; H = cerebral microhemorrhage; N = number of patients; n = number of patients in a group; Q2W = every 2 weeks; Q4W = every 4 weeks; SAE = serious adverse event; TEAE = treatment emergent adverse event.

 

Treatment-emergent Antidrug Antibodies and Effect on Donanemab Serum Concentration

Postbaseline, 46 donanemab-treated participants were evaluable for TE-ADAs. Except for 1 patient in the 10-mg/kg single-dose cohort, all other 45 patients randomized to donanemab developed TE-ADAs. All 6 treatment groups randomized to single or multiple IV administration of donanemab exhibited distinctly higher TE-ADA titers relative to the placebo group. No relationship between dose and TE-ADA was identified in this study. The overall incidence of TE-ADA and titer dynamics were similar for each dose group. The majority of participants exhibited TE-ADAs 3 months after the first dose of donanemab, which returned to or towards baseline after discontinuation of treatment. All 45 donanemab-treated TE-ADA-positive participants were also positive for neutralizing antibody (Nab) to donanemab. Maximum titers for TE-ADA+ participants ranged from 1:10–1:327680 with a median maximum titer of 1:2560. A total of 17 out of 46 of AD patients exposed to donanemab developed high titers (≥1:5120).
To evaluate the potential effect of the kinetics (onset and duration) of TE-ADA on donanemab PK after multiple dosing in the 10 mg/kg Q2W, 10 mg/kg Q4W and 20 mg/kg Q4W cohorts, the observed trough drug concentrations were plotted by dose with ADA results (time-matched with PK) for each visit. Based on graphical analyses, overall there did not appear to be a significant effect of TE-ADAs on the PK of donanemab despite the high incidence of TE-ADAs. Observed trough concentrations among TE-ADA+, NAb+ samples (N=59) appeared similar to those that were TE- ADA- (N=54) across all multiple dose groups (one sample was TE-ADA+, NAb-). Exceptions were observed following 20 mg/kg Q4W beyond Week 48 (Figure 3) where mean trough donanemab concentrations of TE-ADA+, NAb+ samples appeared lower compared with those earlier than Week 48. However, these observations are based on a small number of trough samples, namely Week 48 (5 samples), Week 60 (5 samples), and Week 72 (4 samples). Specific individual participants associated with these lower trough samples were identified to graphically evaluate any effect of titer value on low trough donanemab concentrations. Out of these participants with lower than previous trough samples, there were 2 participants with concentrations below the limit of quantification and low titers, as well as 2 participants with low but quantifiable concentrations and high titers (selected data shown in Supplemental Figure 4).

Figure 3. Serum trough concentrations with available time matched PK and TE ADA evaluable data in the 10 mg/kg Q2W, 10 mg/kg Q4W, and 20 mg/kg Q4W cohorts

Note: Dashed line represents BQL (0.2 µg/mL); Abbreviations: BQL = below the limit of quantification; N = number of patients; NAb = neutralizing antidrug antibody; PK = pharmacokinetics; Q2W = every 2 weeks; Q4W = every 4 weeks; TE-ADA = treatment emergent antidrug antibody.

Figure 4. Least squares mean atrophy on A) whole brain volume, B) average hippocampal volume, and C) lateral ventricle volume (mm3) per study intervention group at 72 weeks

Abbreviations: Q2W = every 2 weeks; Q4W = every 4 weeks; p-values are versus placebo

 

Safety

A total of 7 serious adverse events among 6 patients were reported. Of these, 1 patient (randomized to placebo) discontinued from the study because of a SAE of death due to myocardial infarction (considered not drug-related by the investigator). One of the SAEs, intermittently symptomatic ARIA-E was considered drug-related (20-mg/kg Q4W cohort). The remaining 5 SAEs were considered not drug-related by the investigator.
A total of 223 treatment-emergent adverse events (TEAEs) across all cohorts were reported in this study, regardless of causality (Table 2). Of 61 patients, 55 patients (90.2%) reported least 1 TEAE (generally mild to moderate in severity) and 24 (39.3%) reported at least 1 study drug-related TEAE. The most common TEAE of ARIA-E was experienced by 12 out of 46 donanemab-treated patients with AD and occurred in all donanemab-dosing cohorts except the 10-mg/kg single-dose cohort. The most common study drug-related TEAEs after a single dose of study drug were ARIA-E (n = 4) and cerebral microhemorrhage (n = 4). The most common study drug-related TEAEs in the Q2W- and Q4W-dose cohorts were ARIA-E (n = 2 and n = 6, respectively) and cerebral microhemorrhage (n = 2 and n = 3, respectively).
One infusion-related reaction was reported in 1 patient in the 10-mg/kg Q2W cohort. An additional event of hypertensive crisis had timing consistent with an infusion-related reaction. Three patients discontinued the study prematurely due to an AE: fatal myocardial infarction (placebo Q4W cohort), mild hypertensive crisis (20-mg/kg Q4W cohort), and mild ARIA-E (20-mg/kg Q4W cohort). The patients with hypertensive crisis and ARIA-E both recovered after approximately 20 mins and 8 weeks , respectively. There were no clinically significant changes in other safety assessments, including vital signs, safety laboratories, electrocardiograms, and neurological examinations. Overall, all safety analyses showed no clinically relevant differences between non-Japanese and Japanese patients.

ARIA

Overall, ARIA-E events occurred in 12 of the 46 donanemab-treated patients of whom 2 were symptomatic with mild to moderate symptoms (headache, confusion, hyper-somnolence, and nausea) (Table 2). All patients with ARIA-E were discontinued from study drug as per protocol. All ARIA-E events (including symptoms) resolved following dose discontinuation. All events were considered drug-related.
There were 10 events of cerebral microhemorrhage among 6 of the 46 donanemab-treated patients (Table 2). The majority of cerebral microhemorrhage events (9 of 10) were considered drug-related. Superficial siderosis was reported for 1 patient in the 10-mg/kg Q4W cohort and 1 patient in the 20-mg/kg Q4W cohort. Macrohemorrhage was not observed.

Volumetric Magnetic Resonance Imaging (vMRI)

Overall, administration of donanemab did not result in consistent significant reductions in whole brain volume or hippocampal brain volume nor were there consistent significant increases in lateral ventricular volume when compared to placebo (Figure 4 and Supplemental Figure 5). The changes in whole brain, hippocampal, and ventricular volume were generally numerically greater at 72 weeks with donanemab treatment compared to placebo. However, there was no dose response in the changes, and there were no significant changes in most donanemab treatment cohorts.

Cognition and Function

Across all dose groups, there were no significant changes from baseline in any of the cognitive measures with donanemab treatment (data not shown).

 

Discussion

This Phase 1b study was a randomized, placebo-controlled, single- and multiple-dose study in patients with MCI due to AD or mild to moderate AD (amyloid detected by a positive florbetapir scan). PD, PK, immunogenicity, safety, and tolerability of single and multiple IV doses of donanemab were assessed. The main findings in this study were that:
1) single and multiple doses of donanemab up to 40 mg and 20-mg/kg Q4W, respectively, reduced amyloid plaque deposits in patients with AD; 5 out of 10 patients in the 20 mg/kg Q4W cohort attained complete amyloid clearance within 36 weeks
2) the observed amyloid plaque lowering by donanemab was rapid, robust, and sustained
3) nearly all donanemab-treated patients developed anti-drug antibodies, however, there was no overall significant effect of the antibodies on the PK of donanemab for the duration of the study, given the observed linear PK
4) donanemab was generally well tolerated with manageable ARIA-E events that resolved completely upon treatment discontinuation.

A reduction in cerebral amyloid plaque has also been reported with other anti-amyloid monoclonal antibodies, like gantenerumab, lecanemab, and aducanumab (15-18). The findings of a rapid and dose-dependent reduction in cerebral amyloid plaque after donanemab treatment extend those of a previous ascending dose donanemab study (5), which demonstrated a similar reduction in cerebral amyloid at 10 mg/kg (the highest dose administered in that study). A novel finding in this study is that a significant reduction in cerebral amyloid plaque was observed, even after single doses of donanemab, and the reduction was sustained up to 72 weeks after the single dose. Importantly, the rate of the observed amyloid plaque lowering was rapid, with a greater than 50 Centiloid reduction observed after 24-weeks of multiple-dose donanemab treatment. Furthermore, complete amyloid clearance, as measured by florbetapir PET, was observed for 5 of 10 patients (50.0%) treated with 20-mg/kg Q4W donanemab. This result was sustained through 18 months.
Notably, these robust effects of donanemab on cerebral amyloid were observed in the background of a high incidence of TE-ADAs. Although nearly all donanemab-treated patients developed anti-drug antibodies, there did not appear to be a clinically meaningful effect of the antibodies on the PK of donanemab. However, further analysis are planned where these and other longitudinal data will be analysed via population PK analyses with immunogenicity evaluated as a potential covariate. Despite the background of high TE-ADAs, the PK after single and multiple doses of donanemab were linear from 10- to 40-mg/kg. This result extends the dose range from the earlier Phase 1a study, where the PK of donanemab appeared to be non-linear in nature (5). The reason for this nonlinearity was unclear, and it was speculated that it might be attributed to either donanemab target-mediated disposition and/or anti-drug antibodies impacting PK (5). In this study, the high incidence of anti-drug antibodies was not associated with a high incidence of infusion-related reactions or hypersensitivity reactions (including anaphylaxis).
Donanemab was generally well tolerated with ARIA-E reported as the most common adverse event, which completely resolved upon treatment discontinuation. The incidence of ARIA-E (12 of 46 donanemab-treated patients; 26.1%) was within the range of rates of ARIA-E observed with other amyloid lowering antibodies (19). Several studies with amyloid-lowering therapies have shown a reduction in brain volume and/or an increase in ventricular volume with treatment (20-23). There were no consistent significant changes in vMRI measurements in this study. However, vMRI was an exploratory endpoint in the study and the sample size was small, thus the effect of donanemab on brain volume will need to be more fully addressed in larger clinical studies.
There was no statistically significant effect of donanemab on cognition and function at any dose level or dosing regimen, although this is not unexpected given the small sample size and range of disease stages from MCI to moderate AD dementia enrolled in this study. In contrast, a larger, clinically and pathologically more homogenous Phase 2 trial TRAILBLAZER-ALZ (NCT03367403) met the prespecified primary endpoint of change from baseline to 76 weeks in the Integrated Alzheimer’s Disease Rating Scale with a statistically significant slowing of decline by 32% relative to placebo. Donanemab-treated patients also showed consistent improvements in all prespecified secondary endpoints measuring cognition and function compared to placebo but did not reach nominal statistical significance on every secondary endpoint (24).

 

Conclusion

Single and multiple doses of donanemab demonstrated a rapid and robust reduction in brain amyloid plaque. Single and multiple doses of donanemab yielded sustained amyloid plaque reduction without evidence of significant reaccumulation when measured at 72 weeks. The presence of ADAs were consistent with previous studies, and events of ARIA were manageable. These findings support donanemab dosing up to 1400 mg (approximately 20 mg/kg) Q4W in the TRAILBLAZER-ALZ phase 2 study (NCT03367403), TRAILBLAZER-EXT extension study (NCT04437511), the TRAILBLAZER-ALZ 2 Phase 3 study (NCT04640077) and the planned TRAILBLAZER-ALZ 3 study.

 

Acknowledgments: Data analyses were performed by Eli Lilly and Company. Writing support was provided by Teresa Tartaglione, PharmD (Synchrogenix, a Certara Company, Wilmington, DE, USA) and Paula Hauck, PhD (Eli Lilly and Company).

Funding: This work was supported by Eli Lilly and Company. The sponsors of the study were involved in the design and conduct of the study as well as the collection, analysis, and interpretation of data; in the preparation of the manuscript; and in the review or approval of the manuscript.

Conflict of Interest: JLD – previous employee and minor stockholder of Eli Lilly and Company; currently at Indiana University School of Medicine. All other authors are employees and minor stockholders of Eli Lilly and Company.

Ethical Standards: The study protocol was reviewed and approved by the ethics review board for each of the study sites. The studies were conducted according to Good Clinical Practice, consensus ethics principles derived from international ethics guidelines, including the Declaration of Helsinki and Council for International Organizations of Medical Sciences International Ethical Guidelines, ICH GCP Guideline [E6], and applicable laws and regulations. Patients and/or patients’ legally acceptable representatives provided written informed consent before undergoing study procedures.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

 

SUPPLEMENTARY MATERIAL

 

References

1. Karran E, Mercken M, De Strooper B. The amyloid cascade hypothesis for Alzheimer’s disease: an appraisal for the development of therapeutics. Nat Rev Drug Discov. 2011 Aug 19;10(9):698-712. 10.1038/nrd3505
2. Cummings J, Lee G, Ritter A, Sabbagh M, Zhong K. Alzheimer’s disease drug development pipeline: 2019. Alzheimers Dement (N Y). 2019;5:272-293. 10.1016/j.trci.2019.05.008
3. Gilman S, Koller M, Black RS, et al. Clinical effects of Abeta immunization (AN1792) in patients with AD in an interrupted trial. Neurology. 2005 May 10;64(9):1553-62. 10.1212/01.WNL.0000159740.16984.3C
4. Demattos RB, Lu J, Tang Y, et al. A plaque-specific antibody clears existing beta-amyloid plaques in Alzheimer’s disease mice. Neuron. 2012 Dec 6;76(5):908-20. 10.1016/j.neuron.2012.10.029
5. Lowe SL, Willis BA, Hawdon A, et al. Donanemab (LY3002813) dose-escalation study in Alzheimer’s disease. Alzheimers Dement (N Y). 2021;7(1):e12112. 10.1002/trc2.12112
6. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011 May;7(3):270-9. 10.1016/j.jalz.2011.03.008
7. McKhann GM. Changing concepts of Alzheimer disease. JAMA. 2011 Jun 15;305(23):2458-9. 10.1001/jama.2011.810
8. Barthel H, Gertz HJ, Dresel S, et al. Cerebral amyloid-beta PET with florbetaben (18F) in patients with Alzheimer’s disease and healthy controls: a multicentre phase 2 diagnostic study. Lancet Neurol. 2011 May;10(5):424-35. 10.1016/S1474-4422(11)70077-1
9. Navitsky M, Joshi AD, Kennedy I, et al. Standardization of amyloid quantitation with florbetapir standardized uptake value ratios to the Centiloid scale. Alzheimers Dement. 2018 Dec;14(12):1565-1571. 10.1016/j.jalz.2018.06.1353
10. Bourdage JS, Cook CA, Farrington DL, Chain JS, Konrad RJ. An Affinity Capture Elution (ACE) assay for detection of anti-drug antibody to monoclonal antibody therapeutics in the presence of high levels of drug. J Immunol Methods. 2007 Oct 31;327(1-2):10-7. 10.1016/j.jim.2007.07.004
11. Butterfield AM, Chain JS, Ackermann BL, Konrad RJ. Comparison of assay formats for drug-tolerant immunogenicity testing. Bioanalysis. 2010 Dec;2(12):1961-9. 10.4155/bio.10.136
12. Chen YQ, Pottanat TG, Carter QL, et al. Affinity capture elution bridging assay: A novel immunoassay format for detection of anti-therapeutic protein antibodies. J Immunol Methods. 2016 Apr;431:45-51. 10.1016/j.jim.2016.02.008
13. Sloan JH, Conway RG, Pottanat TG, et al. An innovative and highly drug-tolerant approach for detecting neutralizing antibodies directed to therapeutic antibodies. Bioanalysis. 2016 Oct;8(20):2157-68. 10.4155/bio-2016-0161
14. Clark CM, Schneider JA, Bedell BJ, et al. Use of florbetapir-PET for imaging beta-amyloid pathology. JAMA. 2011 Jan 19;305(3):275-83. 10.1001/jama.2010.2008
15. Bohrmann B, Baumann K, Benz J, et al. Gantenerumab: a novel human anti-Abeta antibody demonstrates sustained cerebral amyloid-beta binding and elicits cell-mediated removal of human amyloid-beta. J Alzheimers Dis. 2012;28(1):49-69. 10.3233/JAD-2011-110977
16. Ostrowitzki S, Deptula D, Thurfjell L, et al. Mechanism of amyloid removal in patients with Alzheimer disease treated with gantenerumab. Arch Neurol. 2012 Feb;69(2):198-207. 10.1001/archneurol.2011.1538
17. Sevigny J, Chiao P, Bussiere T, et al. The antibody aducanumab reduces Abeta plaques in Alzheimer’s disease. Nature. 2016 Sep 1;537(7618):50-6. 10.1038/nature19323
18. Swanson CJ, Zhang Y, Dhadda S, et al. DT-01-07: Treatment of early AD subjects with BAN2401, an anti-Aβ protofibrial monoclonal antibody, significantly clears amyloid plaque and reduces clinical decline. Alzheimer’s & Dementia. 2018;14(7S_Part_31):P1668-P1668. https://doi.org/10.1016/j.jalz.2018.07.009
19. Tolar M, Abushakra S, Hey JA, Porsteinsson A, Sabbagh M. Aducanumab, gantenerumab, BAN2401, and ALZ-801-the first wave of amyloid-targeting drugs for Alzheimer’s disease with potential for near term approval. Alzheimers Res Ther. 2020 Aug 12;12(1):95. 10.1186/s13195-020-00663-w
20. Sur C, Kost J, Scott D, et al. BACE inhibition causes rapid, regional, and non-progressive volume reduction in Alzheimer’s disease brain. Brain. 2020 Dec 1;143(12):3816-3826. 10.1093/brain/awaa332
21. Zimmer JA, Shcherbinin S, Devous MD, Sr., et al. Lanabecestat: Neuroimaging results in early symptomatic Alzheimer’s disease. Alzheimers Dement (N Y). 2021;7(1):e12123. 10.1002/trc2.12123
22. Novak G, Fox N, Clegg S, et al. Changes in Brain Volume with Bapineuzumab in Mild to Moderate Alzheimer’s Disease. J Alzheimers Dis. 2016;49(4):1123-34. 10.3233/JAD-150448
23. Swanson CJ, Zhang Y, Dhadda S, et al. A randomized, double-blind, phase 2b proof-of-concept clinical trial in early Alzheimer’s disease with lecanemab, an anti-Abeta protofibril antibody. Alzheimers Res Ther. 2021 Apr 17;13(1):80. 10.1186/s13195-021-00813-8
24. Mintun MA, Lo AC, Duggan Evans C, et al. Donanemab in Early Alzheimer’s Disease. N Engl J Med. 2021 Mar 13. 10.1056/NEJMoa2100708

TAU PATHOLOGIES MEDIATE THE ASSOCIATIONS OF VASCULAR RISK BURDEN WITH COGNITIVE IMPAIRMENTS IN NONDEMENTED ELDERS: THE CABLE STUDY

 

G.-X. Yu1,#, Y.-N. Ou2,#, Y.-L. Bi3, Y.-H. Ma2, H. Hu2, Z.-T. Wang2, X.-H. Hou2, W. Xu2, L. Tan1,2, J.-T. Yu4

 

1. Department of Neurology, Qingdao Municipal Hospital, Dalian Medical University, Dalian, China; 2. Department of Neurology, Qingdao Municipal Hospital, Qingdao University, Qingdao, China; 3. Department of Anesthesiology, Qingdao Municipal Hospital, Qingdao University, China; 4. Department of Neurology and Institute of Neurology, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai, China; # Contributed equally to this work.

Corresponding Author: Prof. Jin-Tai Yu, Department of Neurology, Huashan Hospital, Fudan University, No. 12 Wulumuqi Road, Shanghai, China; Prof. Lan Tan, Department of Neurology, Qingdao Municipal Hospital, Dalian Medical University, No.5 Donghai Middle Road, Qingdao, China, E-mail addresses: jintai_yu@fudan.edu.cn (J.T. Yu); dr.tanlan@163.com (L. Tan), Tel: +86 21 52888160; Fax: +86 21 62483421

J Prev Alz Dis 2021;
Published online September 20, 2021, http://dx.doi.org/10.14283/jpad.2021.55

 


Abstract

BACKGROUND: Studies suggested that vascular dysfunction might increase the risk of developing Alzheimer’s disease (AD), but the underlying mechanisms still remain obscure.
Objective: To evaluate the associations of vascular risk burden with AD core pathologies and investigate the effects of AD core pathologies on relationships between vascular risk burden and cognitive impairments.
Design: The Chinese Alzheimer’s Biomarker and LifestyLE (CABLE) study was principally focusing on aging, as well as the risk factors and biomarkers of AD initiated in 2017.
Setting: The CABLE study was a large cohort study established in Qingdao, China.
Participants: A total of 618 non-demented elders were obtained from CABLE study.
Measurements: The general vascular risk burden was assessed by the Framingham General Cardiovascular Risk Score (FGCRS). Multivariate linear regression analyses were performed to evaluate the associations of FGCRS with cerebrospinal fluid (CSF) AD biomarkers and cognition. Casual mediation analyses were performed to investigate the mediating effects of AD biomarkers on cognition.
Results: Increased FGCRS was related to higher levels of CSF total tau (t-tau, p < 0.001), phosphorylated tau (p-tau, p < 0.001) as well as the ratio of t-tau and amyloid-β 42 (t-tau/Aβ42, p = 0.010), and lower Chinese-Modified Mini-Mental State Examination (CM-MMSE, p = 0.010) score. Stratified analysis indicated that age modified the associations, with FGCRS being significantly associated with tau pathology (p < 0.001 for t-tau and p-tau) in middle-aged group (<65 years old), instead of older group. The influences of FGCRS on cognitive impairments were partially mediated by tau pathologies (a maximum proportion of 20.9%).
Conclusions: Tau pathology might be a pivotal mediator for effects of vascular risk on cognitive decline. Early and comprehensive intervention for vascular risk factors might be a potential approach to delaying or preventing cognitive impairment and AD.

Key words: Alzheimer’s disease, vascular risk burden, biomarkers, cognitive impairment, mediation.


 

Introduction

Alzheimer’s disease (AD) is an age-related neurodegenerative disorder which is characterized by progressive cognitive impairment. The pathologic hallmarks of AD are neuritic plaques composed of aggregated amyloid-β (Aβ), and neurofibrillary tangles (NFT) harboring hyper-phosphorylated tau and diffuse plaques (1-3). AD has traditionally been regarded as a neurodegenerative disorder affecting neurons, and vascular damage has also been implicated in AD as a potentially modifiable factor of cognitive decline (4). Autopsy studies indicated that intracranial vascular damage often co-occurred with the AD core pathologies in sporadic late-onset AD and that vascular impairment might lower the threshold for dementia (5). In addition, considering the brain’s critical dependence on finely regulated blood supply and blood-brain barrier (BBB) exchange, the vascular alterations could play an essential part in neuronal dysfunction which is a mechanism underlying dementia (6). Accumulating evidence demonstrated that vascular risk factors had a key role in the progression of AD (7). Most vascular risk factors were implicated in AD, including hypertension, diabetes mellitus, smoking and hypercholesterolemia (8-11).
It has been a challenge to establish a direct causal relationship of vascular risk burden with human core AD pathologies. The underlying mechanisms by which these risk factors worsen cognition are still unclear. This might be achieved by their direct influences on AD-related neurodegeneration, or by leading to other cerebral damage which in conjunction with ongoing neurodegeneration could result in cognitive decline (12, 13). However, AD cerebrospinal fluid (CSF) biomarkers provide an opportunity to assess the relationships between vascular risk burden and AD core pathologies. According to the 2018 National Institute on Aging-Alzheimer’s Association (NIA-AA) Research Framework (14), the AD core biomarkers included CSF Aβ42, total tau (t-tau) and phosphorylated tau (p-tau). Previous studies showed that CSF Aβ42 or tau levels changed in twenty years before AD onset (15, 16). Moreover, cognitive decline was used to stage the severity of AD according to the NIA-AA Research Framework (14). Therefore, understanding the associations of vascular risk burden with CSF AD biomarkers and cognition is critical to establishing prevention strategies in preclinical AD.
However, there were inconsistent results on the associations of vascular risk burden with AD core pathologies (13, 17-21). Framingham General Cardiovascular Risk Score (FGCRS) was a well-validated, multivariable risk algorithms of vascular risk burden (22). Herein, in a cohort of non-demented Han Chinese elders, the purposes of our research were: 1) to assess whether FGCRS was associated with CSF AD biomarkers and cognition; 2) to investigate the influences of age and sex on the above associations; and 3) to examine whether AD core pathologies mediated the effects of FGCRS on cognitive impairments.

 

Methods

Participants

Non-demented northern Han Chinese participants were recruited from the Chinese Alzheimer’s Biomarker and LifestyLE (CABLE) study. CABLE study is a large cohort principally focusing on aging, as well as the risk factors and biomarkers of AD since 2017. Participants were patients in several departments of Qingdao Municipal Hospital. They signed informed consent at study entry, and agreed to provide cerebrospinal fluid (CSF) and blood samples for further detection, and underwent a series of clinical and neuropsychological assessments to evaluate their cognitive status. All participants were aged from 40 to 90 years. The exclusion criteria include: 1) cranial injury, infections of the central nervous system, epilepsy, multiple sclerosis or other major neurological diseases; 2) major psychological diseases; 3) severe systemic diseases which may have influences on AD biomarkers; and 4) family history of genetic diseases. Approval of CABLE study was obtained from the Institutional Review Board of Qingdao Municipal Hospital. The present study included non-demented participants who provided adequate information to calculate FGCRS and data of core CSF biomarkers. Their cognitive diagnoses were in compliance with the NIA-AA workgroup diagnostic criteria (23). The thresholds of the adapted Chinese-Modified Mini-Mental State Examination (CM-MMSE) to exclude participants with dementia tendency were 17 for illiterate participants, 20 for participants with 1 to 6 years of education, and 24 for participants with 7 or more years of education (24).

Framingham General Cardiovascular Risk Score

FGCRS was calculated based on a weighted summary of age, sex, systolic blood pressure, treatment for hypertension, smoking attitude, total cholesterol, high-density lipoprotein cholesterol and diabetes (22). The score for age ranges from 0 to 12; systolic blood pressure -3 to 7; total cholesterol 0 to 5; high-density lipoprotein cholesterol -2 to 2; smoker 0 to 3; and diabetes 0 to 4 in woman. And in man, the score for age ranges from 0 to 15; systolic blood pressure -2 to 5; total cholesterol 0 to 4; high-density lipoprotein cholesterol -2 to 2; smoker 0 to 4; and diabetes 0 to 3. The total FGCRS ranged from -5 to 33 for woman, and ranged from -4 to 33 for man. FGCRS was a multivariable risk factor algorithm for the prediction of 10-year risk of vascular events (22). When the risk is greater than 20% (FGCRS of 17 points for women and 14 points for men), professional intervention is warranted (22).

CSF AD biomarkers

In CABLE, CSF specimens were collected in 10 ml polypropylene tubes via lumbar puncture and then transported to the laboratory within 2 hours collection. These specimens were centrifuged at 2000×g for 10 minutes and stored in an enzyme-free EP (Eppendorf) tube at -80℃. CSF Aβ42, t-tau, and p-tau levels were measured by the ELISA kit (Innotest β-AMYLOID (1-42), hTAU-Ag, and PHOSPHO-TAU (181p); Fujirebio, Ghent, Belgium) on the microplate reader (Thermo Scientific™ Multiskan™ MK3). The mean intra-batch coefficient of variation (CV) was <5% (4.9% for Aβ42, 4.5% for t-tau, and 2.4% for p-tau). The mean inter-batch CV was <15% (13.3% for Aβ42, 13.8% for t-tau, and 10.9% for p-tau).

APOE and cognitive assessment

DNA was obtained from overnight fasting blood specimens using the QIAamp®DNA Blood Mini Kit (250). APOE ε4 genotyping was conducted using restriction fragment length polymorphism (RFLP) technology on the basis of 2 specific loci associated with APOE ε4 status, rs7412 and rs429358. Participants were finally divided into APOE ε4 carriers and non-carriers. The global cognitive functioning of all the participants was assessed by CM-MMSE score. Total CM-MMSE scores ranged from 0 to 30. The greater the total score was, the better the cognitive performance was.

Statistical Analyses

CSF values situated outside 3 standard deviations (SD) were excluded for further analysis. Participants were further dichotomized into high and vascular risk groups based on a cut-off of a predicted risk of 20%. Demographic factors were compared using Chi-square tests for categorical variables and Kruskal-Wallis test for continuous variables, respectively. The skewed independent or dependent variables were log10-transformed to normalize the distributions.
Multivariate linear regression analyses were used to evaluate the relationships of FGCRS with CSF AD biomarkers, with the score regarded both as continuous and dichotomous. We further calculated t-tau/Aβ42 and p-tau/Aβ42 which were regarded as better predictors of AD and cognitive impairment (25, 26). As age and sex were incorporated into FGCRS calculation, model 1 only included educational level, APOE ε4 status, CM-MMSE score, and history of stroke. We further additionally adjusted age and sex in the model 2. Age and sex were not only associated with AD, but also played an important part in vascular burden. Subgroup analyses stratified by age (mid-life stage and late-life stage based on a cut-off of 65 years old) and sex (female vs male) to investigate the effects of age and sex on the association between FGCRS and AD pathology were conducted.
We further evaluated the association between FGCRS and CM-MMSE score by multivariate linear regression analyses. Then the influences of CSF AD biomarkers on relationships between FGCRS and CM-MMSE score were assessed by a mediation analysis (27). If all the following 4 criteria were satisfied simultaneously, the mediation effects existed: 1) FGCRS was significantly associated with CSF AD biomarkers; 2) FGCRS was significantly associated with CM-MMSE score; 3) CSF AD biomarkers were significantly associated with CM-MMSE score; and 4) the relationship between FGCRS and CM-MMSE score was weakened after additional adjustment for CSF AD biomarkers. We further estimated the attenuation or indirect effect, with the significance determined using 10,000 bootstrapped iterations. Adjusted covariants included educational level, APOE ε4 status, and history of stroke in the above analyses. Statistical analyses were performed with R version 3.6.1 software. And a two-tailed p value < 0.05 was considered significant.

 

Results

Characteristics of participants

The demographic characteristics of the total population included in our analysis were summarized in Table 1. A total of 618 non-demented participants were enrolled with an average age of 61.93±10.21 years, including 253 (40.9%) females and 93 (15.0%) APOE ε4 carriers. The mean FGCRS was 14.14±4.85 and there were 243 (39.3%) individuals in high vascular risk burden group.

Table 1. Basic demographic information of the analytical population

P values of between-group comparisons were obtained using the Chi-square tests for categorical variables and Kruskal-Wallis test for continuous variables; Abbreviations: APOE, apolipoprotein E; CM-MMSE, Chinese-modified mini-mental state examination; Total-C, total cholesterol; HDL-C, high-density lipoprotein cholesterol; SBP, systolic blood pressure; FGCRS, Framingham General Cardiovascular Risk Score; Aβ42, β-amyloid 42; t-tau, total tau; p-tau, phosphorylated tau.

 

Associations between FGCRS and CSF AD biomarkers

We found that increased FGCRS was related to higher levels of CSF t-tau (β = 0.190, p <0.001), p-tau (β = 0.099, p <0.001) and t-tau/Aβ42 (β = 0.121, p = 0.010) after adjustment for educational level, APOE ε4 status, CM-MMSE score, and history of stroke (Model 1 in Table 2). No significant associations between FGCRS and levels of CSF Aβ42 (β = 0.069, p = 0.113) or p-tau/Aβ42 (β = 0.030, p = 0.488) were observed. Moreover, high vascular risk showed closer associations with increased levels of CSF t-tau (β = 0.046, p < 0.001), p-tau (β = 0.020, p = 0.007) and t-tau/Aβ42 (β = 0.049, p = 0.002) than low vascular risk (Figure 1B, 1C and 1D). High vascular risk group had non-significant associations with CSF Aβ42 (β = -0.003, p = 0.850) and p-tau/Aβ42 (β = 0.023, p = 0.111) levels compared with low vascular risk group (Figure 1A, 1E). In the fully adjusted model, increased FGCRS was still related to higher CSF p-tau (β = 0.059, p = 0.044; Model 2 in Table 2). The associations of FGCRS with other AD CSF biomarkers became non-significant.

Table 2. Associations of FGCRS with CSF AD biomarkers and cognition

Model 1 was adjusted for education level, APOE ε4 status, CM-MMSE score, and history of stroke. Model 2 was adjusted for age, sex, educational level, APOE ε4 status, CM-MMSE score, and history of stroke; * Model 1 was adjusted for education level, APOE ε4 status, and history of stroke. Model 2 was adjusted for age, sex, educational level, APOE ε4 status, and history of stroke. Abbreviations: FGCRS, Framingham General Cardiovascular Risk Score; AD, Alzheimer’s disease; CSF, cerebrospinal fluid; Aβ42, β-amyloid 42; t-tau, total tau; p-tau, phosphorylated tau, CM-MMSE, Chinese-modified mini-mental state examination; APOE, apolipoprotein E.

 

Figure 1. Associations between vascular risk burden and CSF AD biomarkers

Compared to low vascular risk group, significant associations of increased FGCRS with higher t-tau (B), p-tau (C) and t-tau/Aβ42 (D) levels were found, but no significant relationships with Aβ42 (A) or p-tau/Aβ42 (E) levels were found. Abbreviations: FGCRS, Framingham General Cardiovascular Risk Score; AD, Alzheimer’s disease; CSF, cerebrospinal fluid; Aβ42, β-amyloid42; t-tau, total tau; p-tau, phosphorylated tau.

 

Subgroup analyses stratified by age and sex

Considering that age and sex not only associate with AD, but also play important roles in vascular risk burden, we conducted subgroup analyses stratified by age (mid-life stage and late-life stage based on a cut-off of 65 years old) and sex (female vs male) to investigate the effects of age and sex on the association between FGCRS and AD pathology. Results indicated that FGCRS was significantly associated with tau pathology (β = 0.159, p < 0.001 for t-tau; β = 0.120, p < 0.001 for p-tau) in middle-aged group, instead of older group (Table 3). However, sex didn’t modify the above associations. Higher FGCRS was associated with tau pathology in both female (β = 0.179, p < 0.001 for t-tau; β = 0.097, p = 0.001 for p-tau) and male (β = 0.275, p < 0.001 for t-tau; β = 0.130, p = 0.001 for p-tau) participants. FGCRS was also associated with t-tau/Aβ42 (β = 0.246, p = 0.002) in the male participants.

Table 3. Associations of FGCRS with CSF AD biomarkers and CM-MMSE score in age and sex subgroups

NOTE: Associations of FGCRS between CSF AD biomarkers and CM-MMSE score were accessed by multiple linear regression models; All models were adjusted for education level, CM-MMSE score, history of stroke and APOE ε4 status; * CM-MMSE was adjusted for education level, history of stroke and APOE ε4 status; Abbreviations: FGCRS, Framingham General Cardiovascular Risk Score; AD, Alzheimer’s disease; CSF, cerebrospinal fluid; Aβ42, β-amyloid42; t-tau, total tau; p-tau, phosphorylated tau; CM-MMSE, China-modified mini-mental state examination; APOE, apolipoprotein E.

 

Causal Mediation Analyses

In all individuals, there was a significant association between FGCRS and CM-MMSE score (β = -0.019, p = 0.010) after adjustment for education level, APOE ε4 status, and history of stroke. According to the above analyses, we found FGCRS was significantly associated with both tau pathologies and cognition. We further explored whether the FGCRS contributed to cognitive decline via mediating tau pathologies (Figure 2). The association of FGCRS and CM-MMSE score was weakened after separately additional adjustment for CSF t-tau, p-tau and t-tau/Aβ42 levels. Thus, tau pathology was identified as a significant mediator for effects of vascular risk burden on cognitive impairments. Moreover, we considered the effects as partial mediation. The mediation proportions were 20.9% for p-tau with a significant indirect effect (p = 0.006), 10.9% for t-tau/Aβ42 with a significant indirect effect (p = 0.017), and 15.9% for t-tau with a marginally significant indirect effect (p = 0.064).

Figure 2. Causal Mediation Analyses

The association between FGCRS and CM-MMSE score was mediated by tau pathologies; Abbreviations: IE, indirect effect; FGCRS, Framingham General Cardiovascular Risk Score; CM-MMSE, China-modified mini-mental state examination; Aβ42, β-amyloid42; t-tau, total tau; p-tau, phosphorylated tau.

 

Discussion

This is a population-based cross-sectional study, which aimed to explore the associations between FGCRS and a series of AD CSF biomarkers in a cohort of non-demented Han Chinese elders. The primary findings of our research were as follows: 1) FGCRS was positively associated with tau pathologies, which was more evident in the middle-aged individuals; 2) FGCRS was negatively associated with cognitive performance; 3) the influence of FGCRS on cognitive impairments was partially mediated by tau pathologies.
Our results are in line with the finding of many previous studies that individual vascular risk burden could lead to increased tau pathologies (18, 28). Long-term exposure to vascular risk factors can lead to cerebral small vessel diseases (CSVDs), including lacunar infarcts, white matter hyperintensities (WMHs), microinfarcts, and BBB disruption. These diseases are important mechanisms underlying cognitive impairment and dementia. Increased vascular risk burden might lead to atherosclerosis, infarction and other vascular damage, therefore lowering cerebral blood flow (CBF) (29) and further resulting in increased tau pathology (30-32). Our results raised the possibility that cerebrovascular dysfunction might be associated with pre-symptomatic AD pathology (33). Nevertheless, we found no clear association between vascular risk burden and Aβ burden, which was consistent with previous studies (18, 21, 28). Our findings supported the possibility that separate amyloid and vascular pathways might both enhance neurodegeneration (34, 35). Furthermore, CSF t-tau/Aβ42 ratio has been proposed to provide more accurate risk assessments for the development of AD (26, 36). The ratio reflects two aspects of AD pathology: amyloid plaques (Aβ42) and neurodegeneration (tau) (37). P-tau is a marker of axonal damage and neuronal degeneration, and it has stronger associations with AD pathophysiology and the formation of neurofibrillary tangles than t-tau (38). However, our study found that FGCRS was related to higher levels of t-tau/Aβ42 rather than p-tau/Aβ42, possibly because a composite of vascular risk factors in the FGCRS may affect the effects of individual risk factors. Using the same analytical population of CABLE database, previous studies have found that blood pressure(39) and dyslipidemia (40) mainly affect tau pathology, whereas blood glucose (41) affects Aβ pathology. Notably, the underlying mechanisms in which vascular risk mediates Aβ or tau pathology warrant further investigation.
Higher FGCRS was associated with cognitive decline, which was in line with several large longitudinal studies (17, 18, 42). Using casual mediation analysis, we further found that tau pathology was a mediator of the effect of FGCRS on cognitive impairment. Imaging studies suggested that the influences of CBF and soluble platelet-derived growth factor receptor beta (sPDGFRβ), two biomarkers of vascular health, on global cognition were partially mediated by tau pathologies (43). Several potential mechanisms in which tau pathologies mediate the association of vascular risk burden with cognitive impairments have been identified. According to a neuropathological study, microvessels obtained from human AD prefrontal cortex with increased tau pathology upregulate genes participating in endothelial senescence and recruitment of leukocytes into the endothelium, contributing to AD-related cerebrovascular damage and decreased CBF (44). Studies also found that reduced CBF was associated with cognitive decline (45, 46). Moreover, previous studies found that vascular risk burden would become more related to CSF neurofilament light (NFL) in the context of greater CSF t-tau or p-tau levels (47). NFL was also supposed to be an AD biomarker and higher CSF NFL levels were posited to reflect axonal injury and cognitive impairment (48, 49). In conjunction with vascular risk burden, tau pathology could aggravate the impairment of nerve function. Furthermore, greater tau levels were related to decreased levels of claudin-5 (CLDN5) and occludin (OCLN) which played an important part in regulating endothelial barrier integrity (50). Decreased levels of CLDN5 and OCLN indicated BBB damage which was found to be associated with human cognitive dysfunction (51).
Although single vascular risk factors could be particularly effective in the neurovascular unit, but a cluster of them could lead to the final vascular derangement (52, 53). FGCRS was a simple and reliable instrument for assessing the general vascular risk burden, in which most indicators were preventable. It has been confirmed that FGCRS is superior to the Cardiovascular Risk Factors, Aging and Dementia (CAIDE) dementia risk score for the application in prevention programs for evaluating cognitive impairments and targeting modifiable factors (54). Moreover, studies have shown that AD has a long asymptomatic period during which there is accumulation and progression of pathologies and brain structural changes. Symptoms appear when compensatory mechanisms have been overcome, initially as mild cognitive impairment (MCI) and ultimately as dementia (2). This present study focused on the non-demented elders might give us a hint that vascular risk factors may be mainly associated with higher CSF biomarkers in the stage without severe cognitive impairment. Till now, there is a lack of effective drugs to prevent or treat AD. Therefore, findings from this work might highlight the importance of early and integrated management of vascular risk burden to protect cognitive health or delay dementia.
There were several limitations that should be mentioned. Firstly, the associations of FGCRS with CSF AD biomarkers and cognition were only evaluated cross-sectionally in the CABLE cohort. Therefore, the temporality for the associations is unclear. Secondly, some data were obtained from self-reports of participants, such as the history of stroke and diabetes, which might lead to reporting bias. Thirdly, the CABLE is a hospital registry-based study with a specific population profile. The conclusions have a limited power to be generalizable to the general population, but the present findings may have important implications for AD prevention and early warning. Moreover, it is still controversial whether to adjust for age and sex when using the FGCRS, because these variables have been controlled for within the FGCRS calculation. Our results were obtained without correction for age and sex. Lastly, ethnic homogeneity of Chinese Northern Han subjects limited the generalizability of our findings to other studies with different ethnicities. It is necessary to further confirm our results in longitudinal cohorts with multiracial participants.
In summary, our study emphasized the close associations of vascular risk burden with tau pathologies and cognitive impairments. Tau pathologies partially mediated the influences of vascular risk burden on cognitive impairments. Our findings suggesting that early and comprehensive intervention for vascular risk factors might be a potential approach to delaying or preventing cognitive impairment and AD.

 

Acknowledgements: The authors thank all participants of the present study as well as all members of staff of the CABLE study for their role in data collection.

Fundings: This study was supported by grants from the National Natural Science Foundation of China (81971032), the National Key R&D Program of China (2018YFC1314700), Shanghai Municipal Science and Technology Major Project (No.2018SHZDZX01) and ZHANGJIANG LAB, Tianqiao and Chrissy Chen Institute, and the State Key Laboratory of Neurobiology and Frontiers Center for Brain Science of Ministry of Education, Fudan University.

Conflict of interest: The authors declare that they have no competing interests.

Ethical Standards: The CABLE database was conducted in accordance with the Helsinki declaration, and the research program was approved by the Institutional Ethics Committee of Qingdao Municipal Hospital. All subjects or their proxies provided written consents.

 

References

1. Jack CR, Jr., Knopman DS, Jagust WJ, Petersen RC, Weiner MW, Aisen PS, et al. Tracking pathophysiological processes in Alzheimer’s disease: an updated hypothetical model of dynamic biomarkers. The Lancet Neurology. 2013;12(2):207-16. https://doi.org/10.1016/S1474-4422(12)70291-0.
2. Jack CR, Jr., Knopman DS, Jagust WJ, Shaw LM, Aisen PS, Weiner MW, et al. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. The Lancet Neurology. 2010;9(1):119-28. https://doi.org/10.1016/s1474-4422(09)70299-6.
3. Hyman BT, Phelps CH, Beach TG, Bigio EH, Cairns NJ, Carrillo MC, et al. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement. 2012;8(1):1-13. https://doi.org/10.1016/j.jalz.2011.10.007.
4. Iadecola C, Gottesman RF. Cerebrovascular Alterations in Alzheimer Disease. Circulation research. 2018;123(4):406-8. https://doi.org/10.1161/CIRCRESAHA.118.313400.
5. Sweeney MD, Montagne A, Sagare AP, Nation DA, Schneider LS, Chui HC, et al. Vascular dysfunction-The disregarded partner of Alzheimer’s disease. Alzheimer’s & dementia : the journal of the Alzheimer’s Association. 2019;15(1):158-67. https://doi.org/10.1016/j.jalz.2018.07.222.
6. Iadecola C. The Neurovascular Unit Coming of Age: A Journey through Neurovascular Coupling in Health and Disease. Neuron. 2017;96(1):17-42. https://doi.org/10.1016/j.neuron.2017.07.030.
7. Liesz A. The vascular side of Alzheimer’s disease. Science. 2019;365(6450):223-4. https://doi.org/10.1126/science.aay2720.
8. Gabin JM, Tambs K, Saltvedt I, Sund E, Holmen J. Association between blood pressure and Alzheimer disease measured up to 27 years prior to diagnosis: the HUNT Study. Alzheimers Res Ther. 2017;9(1):37. https://doi.org/10.1186/s13195-017-0262-x.
9. Zhang J, Chen C, Hua S, Liao H, Wang M, Xiong Y, et al. An updated meta-analysis of cohort studies: Diabetes and risk of Alzheimer’s disease. Diabetes Res Clin Pract. 2017;124:41-7. https://doi.org/10.1016/j.diabres.2016.10.024.
10. Niu H, Qu Y, Li Z, Wang R, Li L, Li M, et al. Smoking and Risk for Alzheimer Disease: A Meta-Analysis Based on Both Case-Control and Cohort Study. J Nerv Ment Dis. 2018;206(9):680-5. https://doi.org/10.1097/NMD.0000000000000859.
11. Solomon A, Kivipelto M, Wolozin B, Zhou J, Whitmer RA. Midlife serum cholesterol and increased risk of Alzheimer’s and vascular dementia three decades later. Dement Geriatr Cogn Disord. 2009;28(1):75-80. https://doi.org/10.1159/000231980.
12. Vemuri P, Knopman DS, Lesnick TG, Przybelski SA, Mielke MM, Graff-Radford J, et al. Evaluation of Amyloid Protective Factors and Alzheimer Disease Neurodegeneration Protective Factors in Elderly Individuals. JAMA Neurol. 2017;74(6):718-26. https://doi.org/10.1001/jamaneurol.2017.0244.
13. Gottesman RF, Schneider AL, Zhou Y, Coresh J, Green E, Gupta N, et al. Association Between Midlife Vascular Risk Factors and Estimated Brain Amyloid Deposition. Jama. 2017;317(14):1443-50. https://doi.org/10.1001/jama.2017.3090.
14. Jack CR, Jr., Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement. 2018;14(4):535-62. https://doi.org/10.1016/j.jalz.2018.02.018.
15. Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, et al. Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):280-92. https://doi.org/10.1016/j.jalz.2011.03.003.
16. Villemagne VL, Burnham S, Bourgeat P, Brown B, Ellis KA, Salvado O, et al. Amyloid β deposition, neurodegeneration, and cognitive decline in sporadic Alzheimer’s disease: a prospective cohort study. Lancet Neurol. 2013;12(4):357-67. https://doi.org/10.1016/s1474-4422(13)70044-9.
17. Rabin JS, Schultz AP, Hedden T, Viswanathan A, Marshall GA, Kilpatrick E, et al. Interactive Associations of Vascular Risk and beta-Amyloid Burden With Cognitive Decline in Clinically Normal Elderly Individuals: Findings From the Harvard Aging Brain Study. JAMA Neurol. 2018;75(9):1124-31. https://doi.org/10.1001/jamaneurol.2018.1123.
18. Bos I, Vos SJB, Schindler SE, Hassenstab J, Xiong C, Grant E, et al. Vascular risk factors are associated with longitudinal changes in cerebrospinal fluid tau markers and cognition in preclinical Alzheimer’s disease. Alzheimers Dement. 2019;15(9):1149-59. https://doi.org/10.1016/j.jalz.2019.04.015.
19. Rabin JS, Yang HS, Schultz AP, Hanseeuw BJ, Hedden T, Viswanathan A, et al. Vascular Risk and β-Amyloid Are Synergistically Associated with Cortical Tau. Ann Neurol. 2019;85(2):272-9. https://doi.org/10.1002/ana.25399.
20. Reed BR, Marchant NL, Jagust WJ, DeCarli CC, Mack W, Chui HC. Coronary risk correlates with cerebral amyloid deposition. Neurobiol Aging. 2012;33(9):1979-87. https://doi.org/10.1016/j.neurobiolaging.2011.10.002.
21. Lane CA, Barnes J, Nicholas JM, Sudre CH, Cash DM, Malone IB, et al. Associations Between Vascular Risk Across Adulthood and Brain Pathology in Late Life: Evidence From a British Birth Cohort. JAMA Neurol. 2020;77(2):175-83. https://doi.org/10.1001/jamaneurol.2019.3774.
22. D’Agostino RB, Sr., Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743-53. https://doi.org/10.1161/circulationaha.107.699579.
23. McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Jr., Kawas CH, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):263-9. https://doi.org/10.1016/j.jalz.2011.03.005.
24. Cui GH, Yao YH, Xu RF, Tang HD, Jiang GX, Wang Y, et al. Cognitive impairment using education-based cutoff points for CMMSE scores in elderly Chinese people of agricultural and rural Shanghai China. Acta Neurol Scand. 2011;124(6):361-7. https://doi.org/10.1111/j.1600-0404.2010.01484.x.
25. Harari O, Cruchaga C, Kauwe JS, Ainscough BJ, Bales K, Pickering EH, et al. Phosphorylated tau-Aβ42 ratio as a continuous trait for biomarker discovery for early-stage Alzheimer’s disease in multiplex immunoassay panels of cerebrospinal fluid. Biol Psychiatry. 2014;75(9):723-31. https://doi.org/10.1016/j.biopsych.2013.11.032.
26. Mattsson N, Zetterberg H, Hansson O, Andreasen N, Parnetti L, Jonsson M, et al. CSF biomarkers and incipient Alzheimer disease in patients with mild cognitive impairment. Jama. 2009;302(4):385-93. https://doi.org/10.1001/jama.2009.1064.
27. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173-82. https://doi.org/10.1037//0022-3514.51.6.1173.
28. Vemuri P, Lesnick TG, Przybelski SA, Knopman DS, Lowe VJ, Graff-Radford J, et al. Age, vascular health, and Alzheimer disease biomarkers in an elderly sample. Ann Neurol. 2017;82(5):706-18. https://doi.org/10.1002/ana.25071.
29. Suri S, Topiwala A, Chappell MA, Okell TW, Zsoldos E, Singh-Manoux A, et al. Association of Midlife Cardiovascular Risk Profiles With Cerebral Perfusion at Older Ages. JAMA Netw Open. 2019;2(6):e195776. https://doi.org/10.1001/jamanetworkopen.2019.5776.
30. Pluta R, Ułamek-Kozioł M, Januszewski S, Czuczwar SJ. Tau Protein Dysfunction after Brain Ischemia. J Alzheimers Dis. 2018;66(2):429-37. https://doi.org/10.3233/jad-180772.
31. Nedelska Z, Senjem ML, Przybelski SA, Lesnick TG, Lowe VJ, Boeve BF, et al. Regional cortical perfusion on arterial spin labeling MRI in dementia with Lewy bodies: Associations with clinical severity, glucose metabolism and tau PET. Neuroimage Clin. 2018;19:939-47. https://doi.org/10.1016/j.nicl.2018.06.020.
32. Shimada T, Shindo A, Matsuyama H, Yata K, Niwa A, Sasaki R, et al. Chronic cerebral hypoperfusion upregulates leptin receptor expression in astrocytes and tau phosphorylation in tau transgenic mice. Neurosci Lett. 2019;704:133-40. https://doi.org/10.1016/j.neulet.2019.04.009.
33. Iturria-Medina Y, Sotero RC, Toussaint PJ, Mateos-Pérez JM, Evans AC. Early role of vascular dysregulation on late-onset Alzheimer’s disease based on multifactorial data-driven analysis. Nature communications. 2016;7:11934. https://doi.org/10.1038/ncomms11934.
34. Bos I, Verhey FR, Ramakers I, Jacobs HIL, Soininen H, Freund-Levi Y, et al. Cerebrovascular and amyloid pathology in predementia stages: the relationship with neurodegeneration and cognitive decline. Alzheimer’s research & therapy. 2017;9(1):101. https://doi.org/10.1186/s13195-017-0328-9.
35. Vemuri P, Knopman DS. The role of cerebrovascular disease when there is concomitant Alzheimer disease. Biochimica et biophysica acta. 2016;1862(5):952-6. https://doi.org/10.1016/j.bbadis.2015.09.013.
36. Hansson O, Zetterberg H, Buchhave P, Londos E, Blennow K, Minthon L. Association between CSF biomarkers and incipient Alzheimer’s disease in patients with mild cognitive impairment: a follow-up study. Lancet Neurol. 2006;5(3):228-34. https://doi.org/10.1016/s1474-4422(06)70355-6.
37. Ferreira D, Rivero-Santana A, Perestelo-Pérez L, Westman E, Wahlund LO, Sarría A, et al. Improving CSF Biomarkers’ Performance for Predicting Progression from Mild Cognitive Impairment to Alzheimer’s Disease by Considering Different Confounding Factors: A Meta-Analysis. Front Aging Neurosci. 2014;6:287. https://doi.org/10.3389/fnagi.2014.00287.
38. Holtzman DM. CSF biomarkers for Alzheimer’s disease: current utility and potential future use. Neurobiol Aging. 2011;32 Suppl 1(Suppl 1):S4-9. https://doi.org/10.1016/j.neurobiolaging.2011.09.003.
39. Hu H, Meng L, Bi YL, Zhang W, Xu W, Shen XN, et al. Tau pathologies mediate the association of blood pressure with cognitive impairment in adults without dementia: The CABLE study. Alzheimers Dement. 2021. https://doi.org/10.1002/alz.12377.
40. Huang SJ, Ma YH, Bi YL, Shen XN, Hou XH, Cao XP, et al. Metabolically healthy obesity and lipids may be protective factors for pathological changes of alzheimer’s disease in cognitively normal adults. J Neurochem. 2021;157(3):834-45. https://doi.org/10.1111/jnc.15306.
41. Ou YN, Shen XN, Hu HY, Hu H, Wang ZT, Xu W, et al. Fasting blood glucose and cerebrospinal fluid Alzheimer’s biomarkers in non-diabetic cognitively normal elders: the CABLE study. Aging (Albany NY). 2020;12(6):4945-52. https://doi.org/10.18632/aging.102921.
42. Song R, Xu H, Dintica CS, Pan KY, Qi X, Buchman AS, et al. Associations Between Cardiovascular Risk, Structural Brain Changes, and Cognitive Decline. J Am Coll Cardiol. 2020;75(20):2525-34. https://doi.org/10.1016/j.jacc.2020.03.053.
43. Albrecht D, Isenberg AL, Stradford J, Monreal T, Sagare A, Pachicano M, et al. Associations between Vascular Function and Tau PET Are Associated with Global Cognition and Amyloid. J Neurosci. 2020;40(44):8573-86. https://doi.org/10.1523/jneurosci.1230-20.2020.
44. Bryant AG, Hu M, Carlyle BC, Arnold SE, Frosch MP, Das S, et al. Cerebrovascular Senescence Is Associated With Tau Pathology in Alzheimer’s Disease. Front Neurol. 2020;11:575953. https://doi.org/10.3389/fneur.2020.575953.
45. Iadecola C. The pathobiology of vascular dementia. Neuron. 2013;80(4):844-66. https://doi.org/10.1016/j.neuron.2013.10.008.
46. Benedictus MR, Leeuwis AE, Binnewijzend MA, Kuijer JP, Scheltens P, Barkhof F, et al. Lower cerebral blood flow is associated with faster cognitive decline in Alzheimer’s disease. Eur Radiol. 2017;27(3):1169-75. https://doi.org/10.1007/s00330-016-4450-z.
47. Osborn KE, Alverio JM, Dumitrescu L, Pechman KR, Gifford KA, Hohman TJ, et al. Adverse Vascular Risk Relates to Cerebrospinal Fluid Biomarker Evidence of Axonal Injury in the Presence of Alzheimer’s Disease Pathology. J Alzheimers Dis. 2019;71(1):281-90. https://doi.org/10.3233/jad-190077.
48. Olsson B, Lautner R, Andreasson U, Öhrfelt A, Portelius E, Bjerke M, et al. CSF and blood biomarkers for the diagnosis of Alzheimer’s disease: a systematic review and meta-analysis. Lancet Neurol. 2016;15(7):673-84. https://doi.org/10.1016/s1474-4422(16)00070-3.
49. Olsson B, Portelius E, Cullen NC, Sandelius Å, Zetterberg H, Andreasson U, et al. Association of Cerebrospinal Fluid Neurofilament Light Protein Levels With Cognition in Patients With Dementia, Motor Neuron Disease, and Movement Disorders. JAMA Neurol. 2019;76(3):318-25. https://doi.org/10.1001/jamaneurol.2018.3746.
50. Liu CC, Yamazaki Y, Heckman MG, Martens YA, Jia L, Yamazaki A, et al. Tau and apolipoprotein E modulate cerebrovascular tight junction integrity independent of cerebral amyloid angiopathy in Alzheimer’s disease. Alzheimers Dement. 2020;16(10):1372-83. https://doi.org/10.1002/alz.12104.
51. Nation DA, Sweeney MD, Montagne A, Sagare AP, D’Orazio LM, Pachicano M, et al. Blood-brain barrier breakdown is an early biomarker of human cognitive dysfunction. Nat Med. 2019;25(2):270-6. https://doi.org/10.1038/s41591-018-0297-y.
52. Viticchi G, Falsetti L, Buratti L, Boria C, Luzzi S, Bartolini M, et al. Framingham risk score can predict cognitive decline progression in Alzheimer’s disease. Neurobiol Aging. 2015;36(11):2940-5. https://doi.org/10.1016/j.neurobiolaging.2015.07.023.
53. Polidori MC, Pientka L, Mecocci P. A review of the major vascular risk factors related to Alzheimer’s disease. J Alzheimers Dis. 2012;32(3):521-30. https://doi.org/10.3233/jad-2012-120871.
54. Kaffashian S, Dugravot A, Elbaz A, Shipley MJ, Sabia S, Kivimaki M, et al. Predicting cognitive decline: a dementia risk score vs. the Framingham vascular risk scores. Neurology. 2013;80(14):1300-6. https://doi.org/10.1212/WNL.0b013e31828ab370.

 

IMMUNOTHERAPY FOR ALZHEIMER’S DISEASE: CURRENT SCENARIO AND FUTURE PERSPECTIVES

 

M.B. Usman1,*, S. Bhardwaj2, S. Roychoudhury3, D. Kumar4, A. Alexiou5,6, P. Kumar7, R.K. Ambasta7, P. Prasher8, S. Shukla9, V. Upadhye10, F.A. Khan11, R. Awasthi12, M.D. Shastri13, S.K. Singh14, G. Gupta15, D.K. Chellappan16, K. Dua9,17, S.K. Jha18, J. Ruokolainen19, K.K. Kesari19,20, S. Ojha21, N.K. Jha18

 

1. Department of Life Sciences, School of Basic Sciences and Research, Sharda University, Greater Noida, Uttar Pradesh, India; 2. Department of Biotechnology, HIMT, CCS University, Greater Noida, UP, India; 3. Department of Life Science and Bioinformatics, Assam University, Silchar, India; 4. Amity Institute of Molecular Medicine and Stem Cell Research (AIMMSCR), Amity University Uttar Pradesh, Sec 125, Noida, India; 5. Novel Global Community Educational Foundation, Hebersham, 2770 NSW, Australia; 6. AFNP Med Austria, Wien, Austria; 7. Molecular Neuroscience and Functional Genomics Laboratory, Department of Biotechnology, Delhi Technological University (Formerly DCE), Delhi, India; 8. Department of Chemistry, University of Petroleum & Energy Studies, Energy Acres, Dehradun, India; 9. Discipline of Pharmacy, Graduate School of Health, University of Technology Sydney, Ultimo NSW 2007, Australia; 10. Centre of Research for Development (CRD4), Parul Institute of Applied Sciences, Parul University, Vadodara-391760, Gujrat, India; 11. Department of Stem Cell Biology, Institute for Research and Medical Consultations, Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia; 12. Amity Institute of Pharmacy, Amity University Uttar Pradesh, Noida, India; 13. School of Pharmacy and Pharmacology, University of Tasmania, Hobart, Australia; 14. School of Pharmaceutical Sciences, Lovely Professional University, Phagwara 144411, Punjab, India; 15. School of Pharmacy, Suresh Gyan Vihar University, Jagatpura, Mahal Road, Jaipur, India; 16. Department of Life Sciences, School of Pharmacy, International Medical University, Bukit Jalil, Kuala Lumpur, Malaysia; 17. Faculty of Health, Australian Research Centre in Complementary and Integrative Medicine, University of Technology Sydney, Ultimo, 2007 New South Wales, Australia; 18. Department of Biotechnology, School of Engineering & Technology, Sharda University, Greater Noida, Uttar Pradesh, India; 19. Department of Applied Physics, School of Science, Aalto University, Espoo, Finland; 20. Department of Bioproducts and Biosystems, School of Chemical Engineering, Aalto University, Espoo, Finland; 21. Department of Pharmacology and Therapeutics, College of Medicine and Health Sciences, United Arab Emirates University, Al Ain 17666, United Arab Emirates; * These authors contributed equally to this work

Corresponding Author: Dr. Niraj Kumar Jha, Assistant Professor, Department of Biotechnology, School of Engineering & Technology (SET), Sharda University, Knowledge Park III, Greater Noida, Uttar Pradesh-201310, India, Email: nirajkumarjha2011@gmail.com; niraj.jha@sharda.ac.in, Tel: +91-7488019194, ORCID: https://orcid.org/0000-0001-9486-4069; Dr. Shreesh Ojha, Department of Pharmacology and Therapeutics, College of Medicine and Health Sciences, UAE University, PO Box – 17666, Al Ain, UAE, E-mail: shreeshojha@uaeu.ac.ae, Tel: +971-3-7137524, ORCID: https://orcid.org/0000-0001-7801-2966

J Prev Alz Dis 2021;4(8):534-551
Published online September 15, 2021, http://dx.doi.org/10.14283/jpad.2021.52

 


Abstract

Alzheimer’s disease (AD) is a global health concern owing to its complexity, which often poses a great challenge to the development of therapeutic approaches. No single theory has yet accounted for the various risk factors leading to the pathological and clinical manifestations of dementia-type AD. Therefore, treatment options targeting various molecules involved in the pathogenesis of the disease have been unsuccessful. However, the exploration of various immunotherapeutic avenues revitalizes hope after decades of disappointment. The hallmark of a good immunotherapeutic candidate is not only to remove amyloid plaques but also to slow cognitive decline. In line with this, both active and passive immunotherapy have shown success and limitations. Recent approval of aducanumab for the treatment of AD demonstrates how close passive immunotherapy is to being successful. However, several major bottlenecks still need to be resolved. This review outlines recent successes and challenges in the pursuit of an AD vaccine.

Key words: Alzheimer’s disease, amyloid plaque, passive immunotherapy, active immunotherapy, monoclonal antibody.


 

Introduction

Alzheimer’s disease (AD) is a brain disorder characterized by progressive, chronic neurodegenerative symptoms, such as memory loss, cognitive disabilities, and dementia. The global prevalence rates of dementia among people over 85 years and people over 60 years are 20% and 6%, respectively (1). AD is the most common form of dementia among aging individuals in North America and western Europe. It can lead to a decrease in cognitive function, judgment, decision-making, and language abilities among people over 65 years of age (2, 3). Gradual neurodegeneration in the cortex and hippocampus explains the continued loss of memory and dementia observed in patients with AD. This degenerative process can last for up to 25 years after the initial symptoms appear (4).
AD is a significant global health issue associated with a significant economic burden. The global AD prevalence is 24 million, with the United States (US) alone having nearly 5.5 million cases, including 200,000 cases of early-onset AD (2). According to the World Health Organization, an estimated 81.1 million people will have AD by 2040. Unfortunately, the number of people with the disease is projected to multiply in every 20 years (5, 6). AD is the fifth leading cause of death among elderly people globally and sixth in the US (7). While it ranks third in terms of total health care costs in the US after cancer and cardiovascular disease, it is projected to surpass the two diseases in terms of mortality rate and overall financial burden on US health care in the next two decades (8, 9). Approximately $172 billion is spent annually on AD-related healthcare costs (10).
AD, like other neurodegenerative disorders, is a proteinopathy, in that it arises due to protein misfolding or failure of certain peptides to adopt their usual functional and conformational state. Misfolding results in protein accumulation (and/or fibril formation), gain of toxic function, or loss of function. The major causes of protein misfolding include genetic mutations, exposure to external or internal toxins, impairments in the posttranslational modification machinery, and oxidative damage. AD is characterized by the pathological accumulation of two forms of proteinaceous inclusions: the extracellular amyloid beta (Aβ) plaques that develop during the initial disease phase and intracellular neurofibrillary tangles that manifest at later stages (11, 12).
Although we now have a deeper understanding of the pathological features of AD, several questions related to its complex pathways remain unanswered. At present, no single theory has accounted for the various risk factors leading to the pathological and clinical manifestations of dementia-type AD (3, 13). Treatment options targeting various molecules that play essential roles in the development of the disease have been unsuccessful due to multiple drawbacks (14, 15). However, these failures have led to a better understanding of the disease and a shift in focus toward preventive approaches that can avert or delay disease onset (5, 7, 16). One of the major therapeutic avenues being explored currently is immunotherapy, which involves manipulation of the immune system by suppressing, inducing, or enhancing its activity in vivo. Immunotherapy or vaccination against AD-specific peptides inspired considerable optimism in preventing or treating AD through an adaptive immune response (7, 16). Vaccines or immunotherapies for AD utilize the power of the immune system to attack the body’s own proteins or molecules that seem to be dangerous. Despite the practical challenges and decade of disappointments, hopes for Alzheimer’s vaccine are increasing again. Moreover, the enticing allure of being able to curtail the disease through vaccination makes the idea very appealing and worth striving for (16, 17). This review explores the recent successes and challenges in the pursuit of developing an AD vaccine.

 

Pathophysiology and Molecular Concept of AD

AD was first described in 1907 by the German physician, Alois Alzheimer (18). This discovery was ensued by many studies that led to various discoveries and hypothesis on the disease. While the exact cause of AD remains unknown, the most widely acknowledged hypothesis involves abnormal processing of the β-amyloid precursor protein (AβPP), resulting in the overproduction of or reduced clearance of amyloid β-protein (Aβ) in the cortex (19). AβPP, the progenitor molecule of Aβ, is a membrane-bound protein that plays crucial roles in the regulation of neuronal survival, synaptic stabilization and plasticity, cell adhesion, and neuritic outgrowth formation (20, 21). Under normal circumstances, α-secretase cleaves the large AβPP molecule at the middle of the Aβ sequence. In AD, the β-secretase-mediated endoproteolytic cleavage of AβPP generates the primary N-terminal cut, while γ-secretase generates pathogenic Aβ fragments. The length of the deadly Aβ peptide fragments can be determined from the exact site of γ-secretase cleavage. Although β- and γ-secretases are active throughout a person’s lifetime, their undesirable effects on Aβ production are observed in individuals aged ≥60 years (22, 23). The two major forms of Aβ peptides are the 40-residue (Aβ1-40) and 42-residue (Aβ1-42) moieties, which are considered more pathogenic due to their higher aggregation tendency, longer length, and higher quantity in amyloid plaques incases of sporadic and early-onset AD (4, 24). Aβ peptides, like other proteins, have N and C terminals; the N-terminal constitutes the hydrophilic domain with 1-28 residues that are mostly charged, while the C-terminal domain is completely hydrophobic with 29–40 or 29–42 residues. Aβ42, when produced, assumes a beta-pleated structure that clumps to form fibrils that are insoluble in the extracellular space. Over time, amyloid plaques are formed by the deposition of complement protein, microglia, and reactive astrocytes (11, 25). Amyloid plaque formation results in a cascade of neuropathogenic events characterized by neurotoxicity, local inflammation, neuronal apoptosis, complement activation, and disruption of calcium homeostasis, ultimately leading to cognitive decline and AD manifestations. Aβ impairs neuronal function even before its deposition in amyloid plaques. Similarly, Aβ oligomers induce hyperphosphorylation of microtubule-associated protein tau (cytoskeletal protein), leading to the formation of insoluble intracellular neurofibrillary tangles and consequent tauopathy that affects neuronal function (26-28). Hence, the two abnormal protein deposits, amyloid plaques and neurofibrillary tangles, result in the pathophysiological, clinical, and microscopic manifestations of AD (Figure 1).

Figure 1. Neurobiology of Alzheimer’s disease

 

Although the amyloid hypothesis suggests that Aβ deposition and plaque formation are the first steps in the pathogenesis of AD, the relationship between amyloid burden and cognitive symptoms remains unclear. Similarly, the order and timing of amyloidosis and other processes of AD that result in the clinical onset of dementia are not well understood (12). Moreover, the failure of different therapeutic approaches in preventing Aβ aggregation or production raises questions about the hypothesis. So far, there has not been a single successful treatment based on the amyloid hypothesis (3). Recent studies point to the protective and anti-microbial roles of Aβ peptides along with increased formation of tau-positive tangles in AD cell lines, rodent models, and nematodes. In some cases, Aβ is produced in response to bacterial and neurotoxic fungal infection, indicating its neuroprotective role (29). Hence, overproduction of Aβ may be due to downstream immune dysregulation and not the disease process itself.
AD can be divided into two types based on symptom onset: (i) late-onset or sporadic AD is the most common type of AD, in which a majority of patients are diagnosed after 65years of age, and its incidence increases with age (30, 31) and (ii) early-onset AD accounts for 1–2% of AD cases and is characterized by symptom presentation before the age of 65 years (32, 33). Early-onset AD is also referred to as autosomal-dominant AD as it results from mutations in the following genes: amyloid precursor protein (APP) (chromosome 21), presenilin 1/PSEN1(chromosome 14), and presenilin 2/PSEN2 (chromosome 1). Mutations in these genes can lead to abnormal Aβ processing, its excessive accumulation, and consequently, AD with complete penetrance (12). The age of clinical onset of autosomal-dominant AD is influenced by genetic background and is similar among different generations in a family (34, 35). While dominantly inherited mutations have no significant role in sporadic AD, polymorphisms in the apolipoprotein E gene (ε4 allele) increase the risk of developing AD, particularly in females (36, 37). Increasing evidence shows that sporadic and autosomal-dominant AD share pathophysiological features (12, 32).
Modified vaccine formulations use Aβ-specific sequences and epitope-based DNA, while emerging vaccine candidates target other proteins and molecules involved in AD etiology.

 

Vaccines and Immunotherapies for AD

Most immunotherapies and vaccines directly or indirectly target Aβ42 peptides to elicit an appropriate immune response (anti-Aβ antibodies) that will not only clear the Aβ deposits, but also help in improving cognitive and functional abilities (38). Immunotherapy related to AD may be divided into two forms: injection of Aβ42-containing antigens is termed active immunotherapy (vaccination), whereas passive immunotherapy involves administering preformed antibodies against the Aβ42 peptide (such as monoclonal antibodies) (24). Thus, an immunotherapeutic approach involves active injection of Aβ-based immunogens or passive infusion of Aβ-specific antibodies (Figure 2).

Figure 2. Classification of Alzheimer’s disease immunotherapy and hypothetical mechanisms of anti-Aβ antibody action

Active Immunotherapy

In active immunotherapy, patients are injected with a purified form of an antigen, usually coupled with a different protein carrier or adjuvant that helps in the optimization of the immune response. Active AD vaccines are aimed at eliciting an appropriate immune response that clears accumulated proteins. While active immunotherapy has the potential to generate long-term polyclonal antibodies through short-term administration of vaccines at a limited cost, it may cause inconsistent immune responses and long-lasting adverse reactions, especially in older people with low immune competence (25, 39). Most active vaccine trials involve the administration of Aβ42 antigenic peptides. However, more recent studies make use of small Aβ peptides, their DNA sequences, or prime-boost approaches to elicit the anti-Aβ antibody production. This is usually achieved through B-cell activation while avoiding T-cell activation, which may cause autoimmunity (24, 39). As the presence of Aβ plaques is common across different forms of AD, the Aβ peptide is a notable target across immunotherapeutic approaches.

Mechanism of Anti-Aβ Antibodies

Anti-Aβ antibodies are versatile in nature owing to the intrinsic diversity of the human immune system. This versatility is necessary because of the uncertainty about the role of Aβ in physiological conditions and lack of knowledge of the pathogenic forms of Aβ (40). The mechanism by which anti-Aβ antibodies are transported into the central nervous system (CNS) is not well understood. However, it is thought to involve the lymphatic system, passive diffusion through perivascular spaces, and leaky areas in the CNS within the blood-brain barrier (BBB). Consequently, only a small fraction of antibodies in the peripheral circulation is detectable in the CNS (25).
Three hypotheses have been formulated to outline the mechanisms by which anti-Aβ antibodies achieve plaque clearance and reduce AD symptoms (Figure 2). First, anti-Aβ antibodies bind directly to the peptides in the senile plaques, protofibrils, fibrils, or oligomers to destabilize their aggregates and eventually disrupt them (direct action hypothesis). Second, specific antibodies would bind to Aβ plaques and trigger phagocytosis mediated by microglial cells and Fc receptors (41). Third, specific antibodies do not cross the BBB, but bind to and remove the Aβ molecules circulating in the plasma. This generates a concentration gradient that leads to the efflux of Aβ molecules from the brain to the plasma (peripheral sink hypothesis) (14). Most AD vaccine studies prioritize the reduction in senile plaques in the brain by active immunization, which can stimulate the production of anti-Aβ antibodies (38, 42, 43).
Anti-Aβ antibodies are also involved in several other mechanisms that contribute to Aβ reduction or clearance. For instance, antibodies can interact with and alter the transport system of Aβ that includes the receptor for advanced glycation end products (RAGE), the influx channel for Aβ in the CNS, and efflux via the low-density lipoprotein receptor. Theoretically, antibodies that block RAGE could enhance reduction in Aβ levels in the cerebrospinal fluid (CSF) by hindering their transport from the blood (44, 45). While some antibodies may interfere with the interaction between Aβ and other molecules, thereby reducing toxicity, others could act as signals that induce or reduce inflammation by binding to receptors on immune effectors. Further, when antibodies enter the synaptic cleft between neurons or are internalized by neurons, they can alter the cell-to-cell transmission of Aβ and its aggregates (46).

First- and Next-Generation Active Vaccines

Active vaccines aim to stimulate the patient’s immune system to prevent or reduce amyloidosis and restore cognitive and functional abilities. It is commonly believed that immunotherapy must start when the two common features amyloid plaques and neurofibrillary tangles are not obvious. Efforts to develop active AD vaccines have been punctuated by drawbacks, which have led to the evolution of vaccine generations (Table 1).

Table 1. Summary of active vaccines of AD in clinical trial stage

First-generation active vaccines

AD immunotherapy research began with a major breakthrough published by Schenk et al., who demonstrated that active immunization with Aβ42 and an immune-stimulating adjuvant improved cognition in transgenic mice (47). They also showed prevention of or reduction in β-amyloid plaque formation in transgenic mice overexpressing human APP. This discovery led to the rapid development of a first-generation active vaccine called AN-1792.
AN-1792, the first anti-Aβ immunotherapy candidate, consists of aggregated human Aβ42 coupled to a saponin-based adjuvant (QS-21). It elicits an immunological response against the host Aβ42, which can improve cognition and reduce plaque burden (48). The phase 1 trial showed evidence of the tolerability and safety of the vaccine. Moreover, anti-Aβ42 antibodies developed by the recipient patients could recognize the β-amyloid plaque in the extracellular space and the β-amyloid within the blood vessels of the brain. The antibodies were also selective and did not cross-react with native full-length APP or other physiological components (43). Amyloid clearance is facilitated by the solubilization of Aβ42, leading to its exit from the brain through the perivascular pathway. Vaccination also resulted in reduced hippocampal tau pathology mediated by a decrease in tau phosphorylation and inhibition of inflammatory processes that result in neurodegeneration (49-51). Approximately 20% of the vaccinated patients developed antibody titers above the present therapeutic cut-off level (52, 53). However, despite the desirable outcomes, AN-1792 clinical trials were halted in phase 2, owing to adverse inflammatory reactions resulting in subacute meningoencephalitis in nearly 6% of the patients and one death. Subsequent follow-up studies attributed these consequences to the activation of proinflammatory T helper (Th)-1 cell-mediated responses that result in autoimmunity (25, 54). Inflammatory infiltrates in the CNS of the deceased patient were mainly CD8+ cells; to a lesser extent, CD4+, CD3+, and CD5+ cells; and rarely CD7+ cells. In contrast, the patient tested negative for T cytotoxic markers such as CD16 and CD57, turbidimetric immunoassay, granzymes, and B lymphocytes (54). The Aβ42 epitopes are located in the carboxyl-terminal and central region of the Aβ peptide (55). These findings were supported by studies conducted to develop next-generation vaccines containing only B-cell epitopes (primarily located in the N-terminal region of the Aβ peptide). As vaccines that induce only humoral or Th2-mediated responses aim to avoid the undesirable inflammatory effects of Th1 stimulation (Table 1), next-generation vaccines usually contain B-cell epitopes as antigenic determinants coupled to an appropriate adjuvant (56-59).

Next-generation active vaccines

Next-generation active vaccines target the N-terminal regions of Aβ peptides (B-cell epitope) to stimulate humoral immune responses.
ACC-001: ACC-001 (VanutideCridificar) contains1-7 amino acid-long N-terminal Aβ peptide fragments connected to a carrier protein (CRM197) via a surface-active saponin adjuvant (QS-21). The CRM197 carrier protein is a nontoxic Diphtheria toxin mutant (60, 61). ACC-001 elicits an Aβ-specific B-cell response without the adverse T-cell response recorded following AN-1792 administration (62). A phase 1, single ascending dose trial of ACC-001 showed safety and tolerability, which paved the way for phase 2, multiple ascending dose studies (61) conducted in Europe (ClinicalTrials.gov Identifier: NCT00479557), US (ClinicalTrials.gov Identifier: NCT00498602), and Japan. These trials involved administration of different doses of the vaccine (3, 10, and 30μg) with or without the adjuvant. The patients who received doses of ACC-001+QS-21 adjuvant showed sustained anti-Aβ IgG titers and consistently higher peaks. While no case of meningoencephalitis was reported, few patients showed side effects such as insignificant microhemorrhage, treatment-related vasogenic edema, local injection reaction, and headache (61,62). Phase 2a extension studies carried out in these countries showed that long-term exposure to ACC-001+ QS-21 was well-tolerated and gave the highest anti-Aβ IgG titer compared to other regimens (63). However, the phase 2 trial of this vaccine was aborted in 2014 owing to adverse effects linked to autoimmune responses, lack of efficacy, and case of treatment-related angina pectoris recorded in a patient who received ACC-001 (30μg) + QS-21 (62, 64).
AD01, AD02, AD03: While AD01 and AD02 contain Aβ1-6 (B-cell epitope) peptides that mimic the N-terminal region of Aβ42 coupled with an Alum adjuvant, AD03 consists of N-terminal-truncated and pyroglutamated Aβ conjugated with an Aluma djuvant (58). Phase 1 trials of AD01/ AD02 have been announced to be completed by AFFiRiS (Wien, Austria). So far, the trial has demonstrated safety of AD02 and its ability to stabilize cognitive parameters based on a potential correlation between cognitive function and post-vaccination antibody levels; however, these data have not yet been published (58). Phase 2 trials of AffitopeAD02 have been performed in patients with early-onset AD; however, these trials were terminated due limited efficacy and adverse side effects (64). AFFiRiS also conducted a phase 1 trial usingAD03 (58). However, the follow-up study was aborted due to organizational reasons (65, 66).
ACI-24: ACI-24 is based on tetra-palmitoylated amyloid 1–15 peptide in β conformation coupled with liposomes containing monophosphorylated lipid A as an adjuvant. ACI-24 aims to induce antibodies specific to the beta-sheet conformation, thereby targeting Aβ1-15 (67). It is similar to the liposomal vaccine against Aβ1-15, which showed the ability to restore memory defects and reduced plaques in mice (67, 68). Having achieved the desired outcomes in the preclinical trial, a combined phase1/2a clinical trial was initiated (67, 69). The trial compared vaccine doses of 10,100, 300, and 1000 µg/ml to placebo; the dose was administered subcutaneously for the first year, followed by an additional 1 or 2 years. The primary outcomes included tolerability, safety, and serum titers of anti-Aβ42 IgG antibodies. The secondary outcomes included biomarker measures such as T-cell activation measures; magnetic resonance imaging (MRI)-based volumetry; and tau, phospho-tau, and Aβ levels in the CSF. ACI-24 was the first anti-Aβ vaccine to be examined for the treatment of AD patients with Down’s syndrome. The study involved subcutaneous injection of ACI-24 in 24 patients (age: 35–55 years). The study ended in June 2020 and reported positive outcomes and no serious adverse effects. The AC Immune registered additional phase 2 trial in the same syndrome by May 2020, it was set to commence in October 2020 and designed to enroll 72 patients aged 40–50 years who had only brain amyloid deposition without dementia. The primary outcome measures include safety parameters and incidence of adverse events such as suicidal ideation, heart rate, and changes in blood pressure studied for up to 2 years. The secondary outcome measures include changes in cognitive and behavioral measures, levels of amyloid and tau in the blood, neurodegeneration, blood Aβ antibody titers, and levels of amyloid and tau in the brain as determined by positron emission tomography (PET). The trial is projected to end in October 2024 (70).
CAD-106 (Novartis): Novartis’s CAD-106 is composed of multiple copies of B-cell epitope (Aβ1-6) fragments as the immunogenic sequence, attached to a carrier with 180 copies of bacteriophage QB protein coat as an adjuvant (57, 69). The formulation stimulates Aβ-specific antibodies unique to the N-terminus, while avoiding T-cell autoimmune responses (57). As the vaccine could reduce Aβ plaques in APP transgenic mice in a preclinical trial, a phase 1 trial was conducted among patients with mild AD. The trial showed reasonable antibody response and evidence of safety, with no meningoencephalitis, autoimmunity, or other adverse reactions (71). Although phase 2 trials showed adequate antibody production in 75% of the patients without the adverse effects observed in the AN-1792 trials, there was no significant difference between the control and treated groups (71). Phase 2a randomized control trials and two open extension studies showed effective antibody response in approximately 64% of the treated patients. There were sustained anti-Aβ IgG titers in extension versus core studies. Although there was no evidence of Aβ-specific T-cell response or vasogenic edema, a few patients showed intracerebral hemorrhage and imaging abnormalities corresponding to amyloid-related microhemorrhage (57). The phase 2/3 clinical trial (GENERATION 1) sponsored by Novartis Pharmaceuticals was initiated in 2015. It aimed to investigate whether CAD-106 and CNP520, an inhibitor of aspartyl protease beta-secretase or beta-site APP cleaving enzyme, can stall the onset and progression of clinical symptoms in cognitively unimpaired individuals with two APOE4 genes. The clinical trial consists of 1340 enrolled patients and is set to end in 2024. While half of the participants will receive CAD-106 injections four times a year, the other half will receive 50 mg CNP520 once daily; the outcomes in both groups will be compared to that of an age-matched placebo group (72). An additional phase 2/3 prevention study (GENERATION 2) was initiated in August 2017, which enrolled 2000 heterozygous carriers with evidence of brain amyloid protein (age 65–70 years) or homozygous ApoE4 carriers. Patients were randomized to one of three groups: while groups 1 and 2 are given one capsule of CNP520 (group 1: 15 mg; group 2: 50 mg) daily for 60–84 months, group 3 is given one capsule of placebo daily. The GENERATION 1 and 2 trials of CNP520 were both prematurely terminated by the sponsors in July 2019, owing to worsening of cognitive abilities in the treatment groups (73). A phase 3 trial is expected to show whether CAD-106 is more effective than placebo in delaying AD symptoms among individuals with genetic susceptibility to AD. Therefore, CAD-106 remains the only vaccine to advance to phase 3 trials and was selected for an AD prevention initiative (API) in theAPOEε4 homozygote study (39).
Lu AF20513: Lu AF20513 consists of three B-cell epitopes (Aβ1-12) attached to two Th epitopes obtained from tetanus toxoid P2 and P30 (74). The formulation is designed to activate memory Th cells present in majority of the population immunized with the conventional tetanus vaccine, thereby enhancing response against Aβ1-12 in elderly people. The phase 1 study aimed to determine the tolerability and safety of multiple immunizations of the drug. The trial enrolled 24 patients with a recent MRI consistent with an AD diagnosis and Aβ antibodies in the CSF. Multiple shots of either low-, medium-, or high-dose Lu AF20513 were administered to the participants. Although the study aimed to evaluate the safety, tolerability, and antibody titers for around 2 years, the study was terminated on account of new efficacy data from another study (59).
UB-311: UB-311 contains synthetic Aβ1-14 (B-cell epitope) coupled with CpG/Alum as an adjuvant (58). A novel form of the vaccine contains two synthetic Aβ targeting peptides, each of which is conjugated with different Th epitopes and designed in a Th2-based delivery system (56). A successful phase 1 trial led to the advancement to phase 2. The recruitment for this trial is now complete, and the outcomes show early evidence of safety and immunogenicity (59).
V-950: V-950 is a multivalent vaccine containing Aβ1-15 coupled with Alum/ISCOMATRIX as an adjuvant. Although a phase 1 study was initiated to determine its safety, tolerability, and immunogenicity, the study was suspended for unknown reasons (69).
Anti-tau Vaccines: Given the failure of vaccine candidates that target Aβ to provide the desired results in clinical trials, recent efforts seek to include tau protein as another target antigen in preventing or controlling AD.
ACI-35: ACI-35 is a liposomal vaccine based on a synthetic human tau protein sequence phosphorylated at S396 and S404 (75); phase 1 trials to study ACI-35 are ongoing (64, 76).
AADvac1: AADvac1 contains synthetic peptides that mimic the naturally occurring truncated and misfolded tau protein, conjugated with keyhole limpet hemocyanin and aluminum hydroxide as adjuvants (77). AADvac1 is formulated to elicit antibodies against the pathological tau protein, prevent the aggregation or progression of the tau protein aggregates, and thereby hinder the spread of the pathology and the disease. A phase 1 trial was conducted in patients with mild-to-moderate AD. A 24-month, randomized, placebo-controlled, parallel group, double-blind, multi-center, phase 2 study aimed at assessing the safety and efficacy of AADvac1 in patients with mild AD (ADAMANT) is ongoing. Patients with pathological tau protein and/or hippocampal atrophy and CSF amyloid were enrolled in the phase 2 trial, in which they would be given 11 vaccinations within a period of 11 months. Although the study was set to conclude in summer 2019 (77), the results are yet to be published.
Given the failures and practical uncertainties associated with several peptide vaccines in clinical trials, new formulations that do not require adjuvant-like peptides such as DNA vaccines, epitope/protein-based vaccines, and the prime-boost approach have been developed (Table 2).

Table 2. Summary of some active vaccines at preclinical stage

 

DNA Vaccines (genetic vaccines): These are considered as third-generation vaccines; they are constructed by inserting a gene of interest or target gene (Aβ) into an expression vector. The construct is then introduced into a host, which expresses the protein of interest that elicits an immune response in the recipient host (78). DNA vaccines have been found to elicit both humoral and cellular immune responses characterized by Th2 cell stimulation and IgG1 antibody generation in animals (79). The vaccine formulations employ the concept of fusion with immune-modulatory sequences, such as the pan-human leucocyte antigen DR-binding peptide (PADRE) sequence, a non-self Th-cell epitope being used together with other modulators or by itself (7, 80, 81). The vaccine formulation demonstrated evidence of induction of an Aβ-specific immune response without the undesired cytotoxic response.
Some epitope vaccines are obtained from the fusion of Aβ with immunomodulatory sequences such as PADRE, which are either attached to adjuvants or incorporated into chimeric vaccines, such as virus-like particles. The formulation shows good immunogenicity, induction of humoral immune response, and Th2 modulation (58, 82, 83). Vaccines based on recombinant viruses encode an Aβ-specific epitope. However, they are costly and may have adverse effects due to the generation of antibodies with altered epitope specificities (84).
The prime-boost approach seeks to enhance the immune response by administering priming doses (like synthetic peptides) followed by booster doses (like DNA vaccines). This delivery approach facilitates the expansion and selection of B cells with a high degree of affinity for the target gene. Further, the initial boost stimulates T-cell generation, while the second boost activates regulatory T cells that help in the Aβ-specific T-cell-mediated prevention of autoimmune reactions (85, 86).

 

Passive Immunotherapy

Passive immunotherapy involves the administration of preformed antibodies to stimulate the immune system. These antibodies are either derived from humanized murine monoclonal antibodies (mAbs) or naturally occurring polyclonal antibodies obtained from various young healthy donors (intravenous immunoglobulin [IVIG]). Humanized mAbs are derived from non-human sources and have their protein sequences modified to increase similarity with naturally produced human antibodies, whereas fully human mAbs are obtained using phage display or transgenic mice to avoid the side effects of human antibodies (93). Unlike active immunotherapy, passive immunotherapy ensures consistent antibody titer volumes (through infusion of known amount of antibody) and rapid antibody clearance. Drawbacks of the therapy include repeated infusion of antibodies, high cost of production, BBB penetration, proper selection of antigen targets, and generation of an immune response to the injected antibodies (67,94). The antibodies injected into human subjects have different modes of action based on their antigenic targets (Table 3).

Table 3. Summary of monoclonal antibodies (mAbs) with their targets and current statuses

Mechanisms of Action of Anti-Aβ Monoclonal Antibodies

Monoclonal antibodies (mAbs) originate from a single clone of a unique parent cell and bind to a single epitope given their monovalent affinity. For the treatment of AD, various mAbs have been designed to target various epitopes of Aβ species (95) and are administered either subcutaneously or through intravenous infusions.
Monoclonal antibody action begins with binding to a specific antigenic epitope, which triggers an effector function mediated by the Fc portion of the mAb (96). While one hypothesis suggests that mAb binding to amyloid initiates a cascade of processes resulting in complement activation and macrophage-mediated phagocytosis, another suggests that the peripheral sink leads to the efflux of Aβ from the CNS (see mechanism of anti-Aβ antibody and Figure 2 and 3). However, the first hypothesis is based on the assumption that mAbs enter the CNS in sufficient amounts and enhance the phagocytic action of resident microglia or infiltrating monocytes (97). This hypothesis is not widely acknowledged because only 0.1% of the mAbs cross the BBB; the failures of these agents can be linked to poor CNS penetration (67). A novel approach targets receptors on the BBB to induce active transport of the antibodies into the CNS or deliver the gene encoding the antibodies (98).
A recent approach for mAbs is targeting pyroglutamate-3 Aβ, which may be considered as a seed of Aβ aggregation owing to its neurotoxicity and resistance to degradation (93). A preclinical study showed that passive immunization with mAbs reduces plaque deposits while minimizing vaccination side effects (99,100). Another approach involves targeting the N-terminus of Aβ, which could be the most effective way of removing aggregated Aβ (98).

Figure 3. Alzheimer’s disease immunotherapy or anti-Aβ vaccines associated adverse effects

Monoclonal Antibodies in Clinical Trials

Bapineuzumab

Bapineuzumab was the first mAb developed for passive immunotherapy in AD; it entered testing after the failure of the AN-1792 trial. It is a humanized mAb (IgG1) targeting the Aβ N-terminus (Aβ1-5), which binds to and clears fibrillar Aβ42 as well as amyloid plaques. A 12-month, phase 1, single ascending dose trial of 0.5, 1.5, or 5 mg/kg of bapineuzumab showed safety and tolerability in patients with mild-to-moderate AD (101). The phase 2 study involved intravenous administration of either 0.15, 0.5, 0.1, or 2 mg/kg of bapineuzumab in 124 patients with the same form of AD; vasogenic edema was recorded, especially among APOEε4 carriers (102). APOEε4 is one of the alleles of polymorphic apolipoprotein E involved in cholesterol metabolism. It is associated with an increased risk of late-onset AD and Aβ production (103). Given differences in the incidence of vasogenic edema between APOEε4 carriers and non-carriers, phase 3 trials included separate protocols for the two. The mAb was intravenously administered to 2452 patients with mild-to-moderate symptoms in two 18-month phase 3 trials. The results of the two large, multi-center, randomized, double-blind, placebo-controlled, parallel group phase 3 studies did not match the expected outcomes, which were largely negative. Although there was a small reduction in the CSF tau level, there were no significant differences between the bapineuzumab-treated groups and placebo-treated control group (104). The adverse effects included significant vasogenic edema and intracerebral microhemorrhages, referred to as amyloid-related imaging abnormalities with parenchymal edema (ARIA-E) and hemorrhage (ARIA-H), respectively. These conditions could be detected by MRI even when lower doses were administered to APOEε4 carriers (105). Other adverse effects include neuropsychiatric and gastrointestinal symptoms, headache, and confusion. The bapineuzumab trial was terminated because of these side effects and lack of clinical efficacy (94, 106).

Solanezumab

Solanezumab (Eli Lilly) is another humanized IgG1mAb that binds to monomeric, soluble, and toxic Aβ species at the mid-region of the peptide (Aβ16-26) (107). After its apparent success in improving cognitive deficits in transgenic mice, a phase 1 trial of solanezumab at doses of 0.5, 1.5, 4.0, or 10.0 mg/kg in 19 patients with mild-to-moderate AD and healthy volunteers showed good tolerability without any MRI evidence of microhemorrhage, vasogenic edema, or inflammation (108, 109). No adverse events were observed in a multiple-dose study involving 33 patients with mild-to-moderate AD taking 400 mg/month intravenous solanezumab. However, pharmacodynamic biomarker studies showed changes in plasma and CSF levels of Aβ40 and Aβ42. The phase 2 study involved administering 100–1600 mg/month of solanezumab to patients with mild-to-moderate AD. The drug showed a good safety profile and adequate tolerability even at high doses; although the dose-dependent increases in Aβ (Aβ40 and Aβ42) levels in the plasma and CSF indicate mobilization of Aβ from the senile plaques in the brain, there was no change in cognitive function (110). Double-blind, placebo-controlled phase 3 studies (EXPEDITION-1 and EXPEDITION-2) were conducted with over 2000 patients with mild-to-moderate AD who were administered a drug dose of 400 mg/month. Subgroup analysis in the EXPEDITION-1 trial revealed a 34% reduction in cognitive decline in patients with mild AD. The incidences of ARIA-E and ARIA-H in the treatment and placebo groups across the two studies were not significantly different (107). Consequently, Lilly launched the phase 3 (EXPEDITION-3) trial on 2100 patients with brain amyloid burden and mild AD. Although the secondary outcome of this trial slightly favored the drug, solanezumab had no effect on Aβ and tau PET biomarkers; therefore, its development was discontinued (111). Nonetheless, given the drug’s good safety profile and encouraging performance in mild AD cases, it was considered as a candidate in two secondary prevention studies. One prevention study was conducted by the Dominantly Inherited Alzheimer’s Network Trials Unit (DIAN-TU) in 2012. The study was targeted at 210 asymptomatic and very mildly symptomatic carriers of APP, PSEN1, and PSEN2 mutations. The study began as a two-year, phase 2 biomarker study and later proceeded to phase 3 registration with endpoint measurement of cognition after 4 years of treatment. The dose was 400 mg/month initially, which was increased to 1600 mg/month halfway through the trial. However, as the trial did not meet its primary endpoint and there was no reasonable treatment-related change on the DIAN multivariate cognitive endpoint, it was considered to have failed (112).
The other prevention study was initiated by the Alzheimer’s Disease Cooperative as a three-year trial in February 2014. The study recruited 1150 very mildly symptomatic or asymptomatic patients (age: 65 years or more) and investigated biomarker-based evidence of brain amyloid deposition. Solanezumab or placebo was administered intravenously once every four weeks and the drug dose was increased from 400 to 1600 mg/month in June 2017. This trial is expected to continue until mid-2020 (113). Therefore, solanezumab is being evaluated for treatment in patients with mild AD and for prevention in cognitively normal individuals at risk of AD (NCT02008357) and those with familial AD mutations (NCT01760005) (114, 115).

Gantenerumab

Gantenerumab (Hoffman-La, Roche/Ganentech) is the first fully human anti-Aβ mAb(IgG1) that binds specifically to the fibrillar form of Aβ (116). Gantenerumab is a conformational protein that binds to epitopes expressed on Aβ fibrils at the N-terminal (3-12) and central (18-27) amino acids of Aβ. Therefore, the antibody shows a higher affinity for Aβ oligomers and fibrils than for Aβ monomers (116). Gantenerumab significantly reduced Aβ plaques in transgenic mice by mobilizing microglia and hindering the formation of new plaques without altering plasma Aβ levels (116).
Four phase 1 trials in 308 patients conducted internationally showed safety, tolerability, and reduction in brain amyloid plaques in a dosage-dependent manner; however, ARIA remains a major concern. A phase 2 trial was started by Roche in 2010, consisting of 360 participants receiving subcutaneous gantenerumab injections (105 or 225 mg). The study was later expanded into a multinational, 159-center, phase 2/3 registration trial called SCarlet RoAD and recruited 799 participants. Double-blind, placebo-controlled, phase 2/3 studies conducted on Aβ-PET-positive patients with prodromal AD were terminated due to lack of efficacy and incidence of ARIA that increased in a dose-dependent and APOEε4 genotype-dependent manner; this was subsequently converted to an open extension study (117). Participants of SCarlet RoAD who became part of the open-label extension study were administered up to 1200 mg subcutaneous gantenerumab, and slow titration resulted in less ARIA-E (118). Although the open-label trial was set to continue until July 2020, it was later reported to have failed futility analysis (119). GRADUATE-1 and GRADUATE-2 are two new phase 3, double-blind, placebo-controlled studies initiated in 2018, each with the goal of recruiting 760 patients with Aβ pathology and prodromal-to-mild AD; the target enrollment was later raised to 1016 in 2020 (14). The participants will be administered up to 1020 mg subcutaneous gantenerumab or placebo for 2 years. The trial might be completed in 2023. Gantenerumab together with solanezumab are being tested by DIAN-TU for prevention of AD in a phase 2/3 trial for 210 individuals at risk of AD due to autosomal-dominant APP, PSEN1, and PSEN2 mutations (114, 115). The researchers increased the dosage of the two drugs and began a two-year, phase 2 biomarker study. The trial failed to meet its primary endpoint as gantenerumab did not provide reasonable treatment-related changes on the DIAN multivariate cognitive endpoint (114).

Crenezumab

Crenezumab (Genetech/Hoffman-La Roche) was obtained from a mouse antibody and modified to a novel human IgG4mAb that binds to pentameric oligomeric forms of Aβ oligomers, plaques, and fibrils. It also promotes disaggregation while hindering aggregation (120). Phase 1 studies in patients with mild-to-moderate AD reported no case of ARIA-E in either single dose or multiple ascending doses (121). A phase 2, double-blind, placebo-controlled study of patients with mild-to-moderate AD did not report sufficient efficacy (122). A smaller phase 2 imaging study (BLAZE) also failed to show cognitive or clinical benefits of the drug, while a double-blind, placebo-controlled phase 1b trial reported adverse effects like ARIA-H. Currently, a phase 3, double-blind, placebo-controlled study (NCT02670083) is being conducted on Aβ-PET-positive patients with prodromal-to-mild AD to evaluate higher doses of crenezumab (39). As part of the API, crenezumab has also been tested in secondary prevention trials in cognitively normal PSEN1 mutation carriers from the world’s largest early-onset AD kindred in Columbia (NCT01998841) (123).

Ponezumab

Ponezumab (Pfizer Inc.) is a humanized IgG2mAb designed to recognize the C-terminus of Aβ40 (Aβ30-40) (124). It elicits lower immune effector functions than IgG1. Although phase 1 trials showed a good safety profile without evidence of ARIA, CSF antibody levels were poor. The development of ponezumab was halted when two consecutive phase 2 studies revealed no clinical efficacy (125).

Lecanemab

BAN2401 (Biogen/Eisai) is a humanized IgG1mAb that selectively binds, clears, or neutralizes the large soluble Aβ protofibrils. A multi-center phase 1 trial comprised a randomized, double-blind, placebo-controlled study to assess the safety, tolerability, pharmacokinetics, immunogenicity, and pharmacodynamic response to repeated intravenous infusions of BAN2401 (up to 10 mg/kg every 2 weeks for 4 months) in 80 subjects with mild AD and mild cognitive impairment due to AD. While the tolerability of BAN2401 at all the tested doses was good, dosage-dependent increases in ARIA-H and ARIA-E were observed in the treatment and placebo groups. Although the serum elimination half-life was short (7 days) and there was no clear effect on CSF biomarkers, the antibodies entered the CSF and showed dose-dependent exposure (126). A phase 1/2a study of the drug showed adequate tolerability with no cases of ARIA-E (127). Consequently, an 18-month phase 2b trial recruited 856 participants with prodromal-to-mild AD to evaluate the safety, tolerability, and efficacy of BAN2401 at five different intravenous dosages. The study revealed a 47% reduction in cognitive decline and a 93% reduction in brain amyloid with the highest antibody dose (10 mg/kg) administered twice monthly. MRI reports in the highest dose group revealed ARIA in only 10% of the participants and in less than 15% of those with ApoE4 (128). Eisai began a phase 3 trial known as Clarity AD in March 2019, and enrolled 1566 patients with early symptomatic AD across 250 sites in the world. Participants will receive 10 mg/kg drug or placebo every 2 weeks for a period of 18 months, followed by a two-year open-label extension. Changes in Clinical Dementia Rating Scale Sum of Boxes (CDR-SB) at 18 months and the brain amyloid subscale constitute the primary and secondary outcomes, respectively. The trial will continue till 2024. The Alzheimer’s Clinical Trial Consortium began a large BAN2401 phase 3 study, co-funded by Eisai and National Institute of Health (NIH) (AHEAD 3-45). The trial was expected to start in July 2020 and recruit 1400 people who would be divided into two sub-studies. A3 will consist of 400 participants with sub-threshold amyloid levels, and BAN2401 (5 mg/kg titrated to 10 mg/kg) or placebo will be administered every month for 216 weeks; changes in brain amyloid PET at week 216 will constitute the primary outcome. A45 will comprise 1000 participants with amyloid-positive PET scans; BAN2401 (titrated to 10 mg/kg) will be administered at two-week intervals for 96 weeks, followed by a dose of 10 mg/kg every 4 weeks for 216 weeks. A change from baseline in the Preclinical Alzheimer Cognitive Composite 5 score at week 216 constitutes the primary outcome, while changes in brain amyloid PET and cognitive function constitute the secondary outcomes (128).

Aducanumab

Aduhelm (Neurimmune/Biogen) is another fully human IgG1mAb that selectively binds to soluble Aβ aggregates and insoluble fibrils (129).The drug was developed by screening libraries of B-memory cells from healthy elderly individuals for reactivity against aggregated Aβ. The analog of aducanumab has been shown to cross the BBB in transgenic mice; dose-dependent reductions in soluble and insoluble Aβ have also been observed in mice (129). A 12-month phase 1b trial conducted on patients with Aβ-PET-positive prodromal-to-mild AD showed evidence of a dose- and time-dependent reduction in brain fibrillar Aβ. However, the ARIA-E incidence among APOEε4 carriers was high (129). Two identical 18-month phase 3 studies were launched based on the success of the phase 1b trial. These trials sought to evaluate the efficacy of monthly doses of aducanumab in improving cognitive and functional abilities. Although only the data related to doses of 1, 3, and 10 mg/kg were reported, the drug appeared to reduce decline in a dose-dependent manner. Exploratory analyses showed that instances of ARIA-E increased with ApoE4 carriage and dosage (55% inApoE4 homozygotes at 10 mg/kg); these instances occurred in the initial phase of the trial and were later resolved (130).
The development study on patients with mild-to-moderate AD began in Japan in May 2015, with a phase 1 trial of increasing doses up to 6 mg/kg. Later, a phase 3 trial with two efficacy trials was initiated:221AD301ENGAGE and 221AD302EMERGE. 221AD301 ENGAGE enrolled 1350 patients with mild AD or mild cognitive impairment due to AD, as determined by a positive amyloid PET scan. The study, set to continue until 2022, aimed at comparing placebo with monthly infusions of one of the three doses of aducanumab over a period of 18 months. 221AD302 EMERGE, identical to ENGAGE, was conducted at 131 sites in North America with 1350 additional patients. In 2016, Biogen published and presented PRIME data, indicating that a dose titration schedule mitigated ARIA-E and announced its usage in phase 3 (131). However, in March 2019, Biogen and Eisai announced a plan for termination of all aducanumab trials based on an interim analysis that suggested that ENGAGE and EMERGE would miss their primary endpoints; the drug was subsequently removed from the pipeline (132). Interestingly, in October 2019, Biogen faulted the futility analysis and subsequent analysis showed that EMERGE achieved its primary endpoint. Although ENGAGE did not meet the primary endpoint, some exploratory analysis suggested a slow decline in the subgroup that received 10 or more doses of 10 mg/kg.
Following some interactive sessions with the Food and Drug Administration (FDA), Biogen announced plans to apply for regulatory approval of aducanumab in the US and to re-engage eligible patients from the EMERGE, ENGAGE, and PRIME trials with renewed dosing and observations (133).
In January 2020, Biogen launched a phase 3b open-label study called EMBARK, targeting 2400 previous aducanumab trial participants who will receive monthly injections of 10 mg/kg for 2 years. EMBARK has the same endpoints for efficacy as EMERGE and ENGAGE, while biomarker endpoints consist of tauPET, amyloidPET, volumetric MRI, and CSF in a subset of participants. The study is expected to end in 2023.
Biogen submitted the license application in July 2020, demanding priority review (134), and later applied for approval in Japan and the European Union. In November 2020, the FDA advisory committee cited weaknesses in efficacy and voted against approval, while recommending a confirmatory trial. In April 2021, the committee renewed its argument against approval with complaints from public citizens (135). Ultimately, the FDA approved aducanumab in June 2021 under its accelerated approval pathway that requires reasonable likelihood of a meaningful clinical benefit, substantial evidence of effect on an intermediate marker, and phase 4 evidence for such a benefit to be gathered in a subsequent trial after the marketing license has been granted (136).

 

Intravenous Immunoglobulin (IVIG)

IVIG is closely related to passive immunotherapy. It involves the intravenous administration of naturally occurring polyclonal antibodies obtained from the plasma of thousands of healthy young donors. IVIG has already been used as replacement therapy in various clinical conditions, such as certain forms of cancers, immunodeficiency syndromes, and hematological and autoimmune disorders. IVIG primarily contains IgG antibodies, only about 0.5% of which bind to Aβ. The use of IVIG as a potential treatment for AD began in 2002 when human pooled antibodies were shown to have strong affinity for Aβ fibrils and neurotoxic oligomers, while weakly interacting with its monomeric form (16, 137). Moreover, IVIG has some immunomodulatory effects pertinent to the treatment of AD. Early trials of IVIG revealed some benefits in reducing cognitive decline, paving the way for further studies. A phase 2 open-label IVIG trial revealed symptomatic benefits, and a futility study of Gammagard IVIG (Baxter) conducted on patients with mild-to-moderate AD showed positive cognitive scores (25). Baxter and the US funded the phase 3 trial of Gammagard IVIG to determine its efficacy and safety among patients with mild-to-moderate AD. Baxter announced that the primary endpoint for this study was not achieved; the trial has now been discontinued (138).
Grifols conducted a pilot study that involved plasma removal and replacement with Albutein in seven patients with mild-to-moderate AD (7). This procedure was performed twice weekly with a follow-up period of 6 months. Grifols concluded that this is a feasible approach for AD treatment. In 2017, Grifols conducted a phase 2 trial using the same approach and measured similar parameters as the pilot, involving 20 sham-treated and 19 actively treated patients with mild-to-moderate AD. A sawtooth pattern for plasma Aβ40/Aβ42 was seen in the treatment group, while both groups showed similar incidence of adverse events. Grifols’ recent Alzheimer Management by Albumin Replacement (AMBAR) study was a multi-center, randomized, double-blind, placebo-controlled study involving 496 patients with mild-to-moderate AD treated for 14 months. This approach is under phase 3 trial in Europe, while a phase 2 trial in the US is investigating the effect of plasmapheresis with albumin replacement and IVIG. The treatment groups were divided into a sham-treated control group and three treatment groups: plasmapheresis with albumin replacement, plasmapheresis with low dose albumin and IVIG, and plasmapheresis with high-dose albumin and IVIG (139). Changes in the Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) and Alzheimer’s Disease Cooperative Study-Activities of Daily Living (ADCS-ADL) scores between baseline and the endpoint constitute the primary outcome measures. The secondary measures include changes in functional, cognitive, and behavioral tests; measures of disease progression; changes in CSF total tau, p-tau, Aβ40, and Aβ42 levels; changes in plasma Aβ40 and Aβ42 levels; and changes in brain structure and brain glucose metabolism. Subjects in the treatment groups showed 50–75% less worsening of ADAS-Cog scores and 42–70% less worsening of ADCS-ADL scores than control subjects. In addition, pooled data from treated subjects showed that the average decline in ADAS-Cog and ADCS-ADL scores in the treatment group were 66% and 52% lower, respectively, then in the control group. Although some patients with mild AD showed slower disease progression, sham-treated patients with mild AD unexpectedly showed a similar pattern. Grifols reported significant differences in memory, processing speed, quality of life, and language between the control and high albumin/high IVIG treatment groups. Moreover, actively treated patients with moderate AD demonstrated better memory and quality of life than their sham-treated counterparts. Similarly, actively treated patients with mild AD showed better outcomes in language and processing speed tasks than their control counterparts. However, some instances of mild adverse events were noted during high-volume plasma exchange. While the outcomes of AMBAR are promising, some important gaps need to be addressed: mechanism(s) leading to reduction in disease progression; effectiveness of the approach in mild AD as in moderate AD; necessity of including IVIG in the protocol; and how ApoE genotype, age, and sex influence the treatment response (140).

 

Adverse effects of anti-Aβ vaccines

Importantly, both passive and active Aβ immunization elicit CNS inflammation, and can also induce cerebral microhaemorrhage and vasogenic oedema in the already inflamed milieu (141). With the administration of vaccine/antibodies against Aβ, many factors have led to compromised efficacy of immunotherapy in clinical trials. These adverse effects include brain cerebral amyloid angiopathy (CAA), microhemorrhage and meningoencephalitis, which have led to the suspension of clinical trials (Figure 3). Furthermore, patients with AD have well-established neurotic plaques, which are obstacles for a successful vaccine-mediated immune response. Aβ peptide production leads to the activation of the innate immune response marked by activated microglia and elevated levels of complement protein, together they are known to release chemokines and proinflammatory cytokines. Moreover, endogenous sugars can modify Aβ fibrils to advanced glycation end products (AGEs), resulting in proinflammatory signal transduction pathways pertaining to the overproduction of reactive oxygen species and upregulation of AGE receptors. These pathological events constitute a secondary inflammatory response to the early aggregation of Aβ peptides. When the vaccine is administered, the Aβ–antibody complex activates the complement system and microglia, eliciting inflammation in the CNS. Furthermore, activation of T-lymphocytes triggers an adaptive immune response. T-lymphocytes insinuate the brain parenchyma and damage the neural tissue, which is the primary cause of aseptic meningoencephalitis reported in many immunotherapy clinical trials. Moreover, mobilization of Aβ plaques may be an additional concern. As Aβ species cross the BBB, there is a potential risk of neurotoxicity from the brain to the periphery (141). Aβ monomers readily aggregate into oligomers and then into fibrils with β-pleated sheet structures. Aβ oligomers are reported to be more neurotoxic than other Aβ species. Aβ toxicity can be reduced by targeting Aβ oligomers in the early stages rather than plaques. Additionally, current clinical trials based on Aβ-based immunotherapies target Aβ aggregates and do not affect the amount of soluble Aβ. An AN1972 active immunization study reported that increased concentrations of detergent-soluble and water-soluble forms of Aβ in the brain are linked to reduced Aβ plaque load. A series of events such as this aids the formation of Aβ oligomers, which may cause damage to neurons during Aβ clearance (141). This effect of immunotherapy is a significant safety concern and must be investigated.

 

Conclusion and Future Perspectives

After a decade of disappointment in AD prevention through vaccination against Aβ, some vaccine candidates have entered phase 3 clinical trials, while other approaches are in preclinical trials. One of the challenges related to vaccination is its timing. It is now clear that vaccination must start early because plaque removal at later stages does not curtail the progression of the disease, possibly due to progressive tau aggregation. Thus, amyloid removal at pre-symptomatic stages can avert the clinical onset of AD. However, determining the appropriate time for commencement of vaccination is quite challenging. Another point of concern in vaccination is anti-Aβ specificity and antibody titer volumes. Therefore, future studies should evaluate the appropriate timing for vaccination, pathogenic Aβ specificity, and optimization of the titer for antibody response.
Currently, passive immunotherapy appears more promising than active vaccination. The recent approval of aducanumab by the FDA, albeit with some controversies, demonstrates the potential of passive immunotherapy. One of the advantages of passive immunotherapy is that mAbs are amenable to dose and specificity modulation. However, the challenges of short-term antibody effects, low improvement in cognition, and instances of ARIA constitute bottlenecks that need to be addressed.
Given the complex pathophysiology of AD, it is necessary to re-strategize future research in both active and passive immunotherapy. Combination therapy may help in targeting tau protein and Aβ protein, while specific formulations may be beneficial in individuals with specific APOE genotypes, immune phenotypes, and/or Aβ strains. Thus, considering inter-individual differences could improve the prospects of immunotherapeutic prevention of AD.

 

Author’s contributions: NKJ and SO conceptualized the study and hypotheses. MBU, SB, SR, DK, AA and SKS performed literature search. NKJ draw the schemes and drafted the artwork. GG, DKC, KD, JR and KKK drafted the tables. NKJ, and other authors contributed significantly in editing the manuscript. PK, RKA, PP, SS, VU, FAK, RA, SKJ and MDS significantly contributed during revision. All authors read, edited and approved the manuscript.

Acknowledgements: The authors would like to express their gratitude to the unknown referees for carefully reading the paper and giving valuable suggestions.

Conflict of Interest: The authors declare that they have no conflict of interest.

Consent for publication: All authors have read the final version of the manuscript and have given their consent for publication.

 

References

1. LoGiudice D, & Watson R. Dementia in older people: an update. Internal Medicine Journal 2014; 44(11):1066–1073. https://doi.org/10.1111/imj.12572
2. Mayeux R, & Stern Y. Epidemiology of Alzheimer disease. Cold Spring Harbor Perspectives in Medicine 2012; 2(8):a006239. https://doi.org/10.1101/cshperspect.a006239
3. Jha NK, Jha SK, Kar R, Nand P, Swati K & Goswami VK. Nuclear Factor-Kappa β as a therapeutic target for Alzheimer’s disease. J Neurochemistry 2019; 150(2):113-137. doi:10.1111/jnc.14687
4. Uversky VN. Intrinsic Disorder in Proteins Associated with Neurodegenerative Diseases. In: Ovádi J., Orosz F. (eds) Protein Folding and Misfolding: Neurodegenerative Diseases. Focus on Structural Biology, vol 2009; 7. Springer, Dordrecht. https://doi.org/10.1007/978-1-4020-9434-7_2
5. Imtiaz B, Tolppanen AM, Kivipelto M, Soininen H. Future directions in Alzheimer’s disease from risk factors to prevention. Biochemical Pharmacology 2014; 88(4):661–670. https://doi.org/10.1016/j.bcp.2014.01.003
6. Sosa-Ortiz AL, Acosta-Castillo I,Prince MJ. Epidemiology of dementias and Alzheimer’s disease. Archives of Medical Research 2012; 43(8):600–608. https://doi.org/10.1016/j.arcmed.2012.11.003
7. Alves RP, Yang MJ, Batista MT, Ferreira LC. Alzheimer’s disease: is a vaccine possible?. Brazilian JMedical Biological Research 2014; 47(6):438–444. https://doi.org/10.1590/1414-431×20143434
8. Alzheimer’s Association. 2012 Alzheimer’s disease facts and figures. Alzheimer’s &Dementia: The Journal of the Alzheimer’s Association (2012; 8(2):131–168. https://doi.org/10.1016/j.jalz.2012.02.001
9. Hallock P, Thomas MA. Integrating the Alzheimer’s disease proteome and transcriptome: a comprehensive network model of a complex disease. Omics: AJournal of Integrative Biology 2012;16(1-2):37–49. https://doi.org/10.1089/omi.2011.0054
10. Reitz C, Mayeux R. Alzheimer disease: epidemiology, diagnostic criteria, risk factors and biomarkers. Biochemical Pharmacology 2014; 88(4): 640–651. https://doi.org/10.1016/j.bcp.2013.12.024
11. Henderson VW. Alzheimer’s disease: Review of hormone therapy trials and implications for treatment and prevention after menopause. J Steroid Biochem and Mol Biol 2014; 142:99–106. https://doi.org/10.1016/j.jsbmb.2013.05.010
12. Bateman RJ, Xiong C, Benzinger TL, Fagan AM, Goate A, et al. Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. The New England Journal of Medicine 2012; 367(9):795–804. https://doi.org/10.1056/NEJMoa1202753
13. Jack CR, Jr KnopmanDS, Jagust WJ, ShawLM, Aisen PS, Weiner MW, Petersen RC &Trojanowski JQ. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. The Lancet Neurology 2010; 9(1):119–128. https://doi.org/10.1016/S1474-4422(09)70299-6
14. Panza F, Lozupone M, Seripa D, Imbimbo BP. Amyloid-β immunotherapy for alzheimer disease: Is it now a long shot?. Annals of Neurology 2019; 85(3):303–315. https://doi.org/10.1002/ana.25410
15. Haass C. New hope for Alzheimer disease vaccine. Nature medicine 2002; 8(11):1195–1196. https://doi.org/10.1038/nm1102-1195
16. Schnabel J. Vaccines:Chasing the dream. Nature 2011; 475(7355):S18–S19. doi:10.1038/475s18a
17. Sterner RM, Takahashi PY & Yu Ballard AC. Active Vaccines for Alzheimer Disease Treatment. Journal of the American Medical Directors Association 2016; 17(9):862.e11–862.e15. doi:10.1016/j.jamda.2016.06.009
18. Alzheimer A, Stelzmann RA, Schnitzlein HN Murtagh FR. An English translation of Alzheimer’s 1907 paper, “Uber eineeigenartigeErkankung der Hirnrinde”. Clinical Anatomy (New York, N.Y.) 1995; 8(6):429–431. https://doi.org/10.1002/ca.980080612
19. Hardy J,Selkoe DJ. The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science (New York, N.Y.) 2002; 297(5580):353–356. https://doi.org/10.1126/science.1072994
20. Mattson MP. Pathways towards and away from Alzheimer’s disease. Nature 2004; 430(7000):631–639. https://doi.org/10.1038/nature02621
21. Van Gassen G, Annaert W, Van Broeckhoven C. Binding partners of Alzheimer’sdisease proteins: are they physiologically relevant? Neurobiol Dis 2000; 7(3):135– 151.
22. Price JL, Morris JC. Tangles and plaques in nondemented aging and “preclinical” Alzheimer’s disease. Annals of Neurology 1999; 45(3):358–368. https://doi.org/10.1002/1531-8249(199903)45:3<358::aid-ana12>3.0.co;2-x
23. Berg L, McKeel DW, Jr Miller JP, Storandt M, Rubin EH, Morris JC, Baty J, Coats M, Norton J, Goate AM, Price JL, Gearing M, Mirra SS &Saunders AM. Clinicopathologic studies in cognitively healthy aging and Alzheimer’s disease: relation of histologic markers to dementia severity, age, sex, and apolipoprotein E genotype. Archives of Neurology 1998; 55(3): 326–335. https://doi.org/10.1001/archneur.55.3.326
24. Lambracht-Washington D & Rosenberg RN. Advances in the development of vaccines for Alzheimer’s disease. Discov Med 2013; 15(84):319-326.
25. Lannfelt L, Relkin NR & Siemers ER. Amyloid-ß-directed immunotherapy for Alzheimer’s disease. Journal of Internal Medicine 2014; 275(3):284–295. doi:10.1111/joim.12168
26. Jin M, Shepardson N, Yang T, Chen G, Walsh D & Selkoe DJ. Soluble amyloid beta-protein dimers isolated from Alzheimer cortex directly induce Tau hyperphosphorylation and neuritic degeneration. Proceedings of the National Academy of Sciences, U.S.A 2011; 108(14):5819–5824.
27. Oddo S, Caccamo A, Tran L, Lambert MP, Glabe CG, Klein WL & La Ferla FM. Temporal profile of amyloid-beta (Abeta) oligomerization in an in vivo model of Alzheimer disease. A link between Abeta and tau pathology. The Journal of Biological Chemistry 2006; 281(3):1599–1604. https://doi.org/10.1074/jbc.M507892200
28. Walsh DM, Klyubin I, Fadeeva JV, Cullen WK, Anwyl R, Wolfe MS, Rowan MJ &Selkoe DJ. Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal long-term potentiation in vivo. Nature 2002; 416(6880):535–539. https://doi.org/10.1038/416535a
29. Kumar DK, Choi SH, Washicosky KJ, Eimer WA, Tucker S, Ghofrani J, Lefkowitz A, McColl G, Goldstein LE, Tanzi RE & Moir RD. Amyloid-β peptide protects against microbial infection in mouse and worm models of Alzheimer’s disease. Science Translational Medicine 2016; 8(340):340ra72. https://doi.org/10.1126/scitranslmed.aaf1059
30. Kukull WA, Higdon R, Bowen JD, McCormick WC, Teri L, Schellenberg GD, van Belle G, Jolley L & Larson EB. Dementia and Alzheimer disease incidence: a prospective cohort study. Archives of Neurology 2002; 59(11):1737–1746. https://doi.org/10.1001/archneur.59.11.1737
31. Launer LJ, Andersen K, Dewey ME, Letenneur L, Ott A, et al. Rates and risk factors for dementia and Alzheimer’s disease: results from EURODEM pooled analyses. EURODEM Incidence Research Group and Work Groups. European Studies of Dementia. Neurology 1999; 52(1):78–84. https://doi.org/10.1212/wnl.52.1.78
32. Cruchaga C, Haller G, Chakraverty S, Mayo K, Vallania FL, Mitra RD, et al. Rare variants in APP, PSEN1 and PSEN2 increase risk for AD in late-onset Alzheimer’s disease families. PloS one 2012; 7(2):e31039. https://doi.org/10.1371/journal.pone.0031039
33. Campion D, Dumanchin C, Hannequin D, Dubois B, Belliard S, Puel M, Thomas-Anterion C, et al. Early-onset autosomal dominant Alzheimer disease: prevalence, genetic heterogeneity, and mutation spectrum. American Journal of Human Genetics 1999; 65(3):664–670. https://doi.org/10.1086/302553
34. Wijsman EM, Daw EW, Yu X, Steinbart EJ, Nochlin D, Bird TD & Schellenberg GD. APOE and other loci affect age-at-onset in Alzheimer’s disease families with PS2 mutation. American Journal of Medical Genetics. Part B, Neuropsychiatric Genetics: The official publication of the International Society of Psychiatric Genetics 2005; 132B(1):14–20. https://doi.org/10.1002/ajmg.b.30087
35. Lopera F, Ardilla A, Martínez A, Madrigal L, Arango-Viana JC, Lemere CA, Arango-Lasprilla JC et al. Clinical features of early-onset Alzheimer disease in a large kindred with an E280A presenilin-1 mutation. JAMA 1997; 277(10):793–799.
36. Bretsky PM, Buckwalter JG, Seeman TE, Miller CA, Poirier J, Schellenberg GD, Finch CE & Henderson VW. Evidence for an interaction between apolipoprotein E genotype, gender, and Alzheimer disease. Alzheimer Disease and Associated Disorders 1999; 13(4):216–221. https://doi.org/10.1097/00002093-199910000-00007
37. Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, Myers RH, Pericak-Vance MA, Risch N, van Duijn CM. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA 1997; 278(16):1349–1356.
38. Evans CF, Davtyan H, Petrushina I, Hovakimyan A, Davtyan A, Hannaman D, Cribbs DH, Agadjanyan MG & Ghochikyan A. Epitope-based DNA vaccine for Alzheimer’s disease: translational study in macaques. Alzheimer’s &Dementia: The Journal of the Alzheimer’s Association 2014; 10(3):284–295. https://doi.org/10.1016/j.jalz.2013.04.505
39. van Dyck CH. Anti-Amyloid-β Monoclonal Antibodies for Alzheimer’s Disease: Pitfalls and Promise. Biological Psychiatry 2018; 83(4):311–319. https://doi.org/10.1016/j.biopsych.2017.08.010
40. Abramov E, Dolev I, Fogel H, Ciccotosto GD, Ruff E & Slutsky I. Amyloid-beta as a positive endogenous regulator of release probability at hippocampal synapses. Nature Neuroscience 2009; 12(12):1567–1576. https://doi.org/10.1038/nn.2433
41. Wilcock DM, Munireddy SK, Rosenthal A, Ugen KE, Gordon MN, Morgan D. Microglial activation facilitates Abeta plaque removal following intracranial anti-Abeta antibody administration. Neurobiology of Disease 2004; 15(1):11–20. https://doi.org/10.1016/j.nbd.2003.09.015
42. Weiner HL, Frenkel D. Immunology and immunotherapy of Alzheimer’s disease. Nature Reviews Immunology 2006; 6(5):404–416. doi:10.1038/nri1843
43. Hock C, Konietzko U, Papassotiropoulos A, Wollmer A, Streffer J, von Rotz RC, Davey G, Moritz E, Nitsch RM. Generation of antibodies specific for beta-amyloid by vaccination of patients with Alzheimer disease. Nature Medicine 2002; 8(11):1270–1275. https://doi.org/10.1038/nm783
44. Zlokovic BV. Clearing amyloid through the blood-brain barrier. Journal of Neurochemistry 2004; 89(4):807–811. https://doi.org/10.1111/j.1471-4159.2004.02385.x
45. Deane R, Du Yan S, SubmamaryanRK, LaRue B, Jovanovic S, Hogg E, Welch D, et al. RAGE mediates amyloid-beta peptide transport across the blood-brain barrier and accumulation in brain. Nature Medicine 2003; 9(7):907–913. https://doi.org/10.1038/nm890
46. Arbel M, Yacoby I, Solomon B. Inhibition of amyloid precursor protein processing by beta-secretase through site-directed antibodies. Proceedings of the National Academy of Sciences U.S.A 2005; 102(21):7718–7723. https://doi.org/10.1073/pnas.0502427102
47. Schenk D, Barbour R, Dunn W, Gordon G, Grajeda H, Guido T, Hu K, Huang J, et al. Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature 1999; 400(6740):173–177. https://doi.org/10.1038/22124
48. Bayer AJ, Bullock R, Jones RW, Wilkinson D, Paterson KR, Jenkins L, Millais SB, Donoghue S. Evaluation of the safety and immunogenicity of synthetic Abeta42 (AN1792) in patients with AD. Neurology 2005; 64(1):94–101. https://doi.org/10.1212/01.WNL.0000148604.77591.67
49. Zotova E, Bharambe V, Cheaveau M, Morgan W, Holmes C, Harris S, Neal JW, Love S, Nicoll JA &Boche D. Inflammatory components in human Alzheimer’s disease and after active amyloid-β42 immunization. Brain: AJournal of Neurology 2013; 136(Pt 9):2677–2696. https://doi.org/10.1093/brain/awt210
50. Serrano-Pozo A, William CM, Ferrer I, Uro-Coste E, Delisle MB, Maurage CA, Hock C, Nitsch RM, Masliah E, Growdon JH, Frosch MP & Hyman BT. Beneficial effect of human anti-amyloid-beta active immunization on neurite morphology and tau pathology. Brain: A Journal of Neurology 2010; 133(Pt 5):1312–1327. https://doi.org/10.1093/brain/awq056
51. Boche D, Zotova E, Weller RO, Love S, Neal JW, Pickering RM, Wilkinson D, Holmes C & Nicoll JA. Consequence of Abeta immunization on the vasculature of human Alzheimer’s disease brain. Brain: AJournal of Neurology 2008; 131 (Pt12):3299–3310. https://doi.org/10.1093/brain/awn261
52. Gilman S, Koller M, Black RS, Jenkins L, Griffith SG, Fox NC, Eisner L, Kirby L, Rovira MB, et al. Clinical effects of Abeta immunization (AN1792) in patients with AD in an interruptedtrial. Neurology 2005; 64(9):1553–1562. https://doi.org/10.1212/01.WNL.0000159740.16984.3C
53. Orgogozo JM, Gilman S, Dartigues JF, Laurent B, Puel M, Kirby LC, Jouanny P, Dubois B, Eisner L, et al. Subacute meningoencephalitis in a subset of patients with AD after Abeta42 immunization. Neurology 2003; 61(1):46–54. https://doi.org/10.1212/01.wnl.0000073623.84147.a8
54. Ferrer I, Rovira MB, Guerra MLS, Rey MJ & Costa-Jussá F. Neuropathology and Pathogenesis of Encephalitis following Amyloid β Immunization in Alzheimer’s Disease. Brain Pathology 2004; 14(1):11–20. doi:10.1111/j.1750-3639.2004.tb00493.x
55. Cribbs DH, Ghochikyan A, Vasilevko V, Tran M, Petrushina I, Sadzikava N, Babikyan D, Kesslak P, Kieber-Emmons T, Cotman CW &Agadjanyan MG. Adjuvant-dependent modulation of Th1 and Th2 responses to immunization with beta-amyloid. International Immunology 2003; 15(4):505–514. https://doi.org/10.1093/intimm/dxg049
56. Wang CY, Wang PN, Chiu MJ, Finstad CL, Lin F, Lynn S, Tai YH, De Fang X, et al. UB-311, a novel UBITh® amyloid β peptide vaccine for mild Alzheimer’s disease. Alzheimer’s &Dementia (New York, N. Y.) 2017; 3(2):262–272. https://doi.org/10.1016/j.trci.2017.03.005
57. Farlow MR, Andreasen N, Riviere ME, Vostiar I, Vitaliti A, SovagoJ, Caputo A, Winblad B & Graf A. Long-term treatment with active Aβ immunotherapy with CAD106 in mild Alzheimer’s disease. Alzheimer’s Research & Therapy 2015; 7(1):23. https://doi.org/10.1186/s13195-015-0108-3
58. Davtyan H, Bacon A, Petrushina I, Zagorski K, Cribbs DH, Ghochikyan A &Agadjanyan MG. Immunogenicity of DNA- and recombinant protein-based Alzheimer Disease epitope vaccines.Human Vaccines & Immunotherapeutics 2014; 10(5):1248–1255. doi:10.4161/hv.27882.
59. Mantile, F., & Prisco, A. Vaccination against β-Amyloid as a Strategy for the Prevention of Alzheimer’s Disease. Biology, 2020; 9(12), 425. doi:10.3390/biology9120425
60. Mamikonyan G, Necula M, Mkrtichyan M, Ghochikyan A, PetrushinaI, Movsesyan N, Mina E, Kiyatkin A, Glabe CG, Cribbs DH &Agadjanyan MG. Anti-A beta 1-11 antibody binds to different beta-amyloid species, inhibits fibril formation, and disaggregates preformed fibrils but not the most toxic oligomers. The Journal of Biological Chemistry 2007; 282(31):2 2376–22386. https://doi.org/10.1074/jbc.M700088200
61. Arai H, Suzuki H, Yoshiyama T. Vanutidecridificar and the QS-21 adjuvant in Japanese subjects with mild to moderate Alzheimer’s disease: results from two phase 2 studies. Curr Alzheimer Res 2015; 12(3):242-254. doi:10.2174/1567205012666150302154121. PMID: 25731629.
62. Pasquier F, Sadowsky C, Holstein A, Leterme G, Peng Y, Jackson N, Fox NC, et al. Two Phase 2 Multiple Ascending-Dose Studies of VanutideCridificar (ACC-001) and QS-21 Adjuvant in Mild-to-Moderate Alzheimer’s Disease. Journal of Alzheimer’s Disease: JAD 2016; 51(4):1131–1143. https://doi.org/10.3233/JAD-150376
63. Hull M, Sadowsky C, Arai H, Le Prince Leterme G, Holstein A, Booth K, Peng Y, Yoshiyama T, Suzuki H, Ketter N, Liu E & Ryan JM. Long-Term Extensions of Randomized Vaccination Trials of ACC-001 and QS-21 in Mild to Moderate Alzheimer’s Disease. Current Alzheimer research 2017; 14(7):696–708. https://doi.org/10.2174/1567205014666170117101537
64. Godyń J, Jończyk J, Panek D &Malawska B. Therapeutic strategies for Alzheimer’s disease in clinical trials. Pharmacological Reports: PR 2016; 68(1):127–138. https://doi.org/10.1016/j.pharep.2015.07.006
65. Kwan, P., Konno, H., Chan, K. Y., & Baum, L. Rationale for the development of an Alzheimer’s disease vaccine. Human Vaccines & Immunotherapeutics. 2019. doi:10.1080/21645515.2019.1665453
66. Schneeberger A, Hendrix S, Mandler M, Ellison N, Bürger V, Brunner M, Frölich L, et al. Results from a Phase II Study to Assess the Clinical and Immunological Activity of AFFITOPE® AD02 in Patients with Early Alzheimer’s Disease. The Journal of Prevention of Alzheimer’s Disease 2015; 2(2):103–114. https://doi.org/10.14283/jpad.2015.63
67. Lemere CA. Immunotherapy for Alzheimer’s disease: hoops and hurdles. Molecular Neurodegeneration 2013; 8(1):36. doi:10.1186/1750-1326-8-36
68. Muhs A, Hickman DT, Pihlgren M, Chuard N, Giriens V, MeerschmanC, van der Auwera I, van Leuven F, et al. Liposomal vaccines with conformation-specific amyloid peptide antigens define immune response and efficacy in APP transgenic mice. Proceedings of the National Academy of Sciences (U.S.A.) 2007; 104(23): 9810–9815. https://doi.org/10.1073/pnas.0703137104
69. Yu YZ & Xu Q. Prophylactic immunotherapy of Alzheimer’s disease using recombinant amyloid-β B-cell epitope chimeric protein as subunit vaccine. Human Vaccines &Immunotherapeutics 2016; 12(11):2801–2804. https://doi.org/10.1080/21645515.2016.1197456
70. https://www.alzforum.org/therapeutics/aci-24. Last updated 08 October 2020
71. Winblad B, Andreasen N, Minthon L, Floesser A, Imbert G, Dumortier T, Maguire RP, et al. Safety, tolerability, and antibody response of active Aβ immunotherapy with CAD106 in patients with Alzheimer’s disease: randomised, double-blind, placebo-controlled, first-in-human study. The Lancet Neurology 2012; 11(7):597–604. https://doi.org/10.1016/S1474-4422(12)70140-0
72. Lopez C, Tariot PN., Caputo A, Langbaum JB, Liu F, et al. The Alzheimer’s Prevention Initiative Generation Program: Study design of two randomized controlled trials for individuals at risk for clinical onset of Alzheimer’s disease. Alzheimer’s & dementia (New York, N. Y.), 2019; 5, 216–227. https://doi.org/10.1016/j.trci.2019.02.005
73. https://www.alzforum.org/news/conference-coverage/picking-through-rubble-field-tries-salvage-bace-inhibitors 20 Dec 2019
74. Davtyan H, Ghochikyan A, Hovakimyan A, Petrushina I, Yu J, Flyer D, Madsen PJ, Pedersen LO, Cribbs DH &Agadjanyan MG. Immunostimulant patches containing Escherichia coli LT enhance immune responses to DNA- and recombinant protein-based Alzheimer’s disease vaccines. Journal of Neuroimmunology 2014; 268(1-2):50–57. https://doi.org/10.1016/j.jneuroim.2014.01.002
75. Theunis C, Crespo-Biel N, Gafner V, Pihlgren M, López-Deber MP, Reis P, Hickman DT, et al. Efficacy and safety of a liposome-based vaccine against protein Tau, assessed in tau.P301L mice that model tauopathy. PloS one 2013; 8(8):e72301. https://doi.org/10.1371/journal.pone.0072301
76. Grüninger F. Invited review: Drug development for tauopathies. Neuropathology and Applied Neurobiology 2015; 41(1):81–96. https://doi.org/10.1111/nan.12192
77. Novak P, Zilka N, Zilkova M. et al. AADvac1, an Active Immunotherapy for Alzheimer’s Disease and Non Alzheimer Tauopathies: An Overview of Preclinical and Clinical Development. J Prev Alzheimers Dis 2019;6:63–69. https://doi.org/10.14283/jpad.2018.45
78. RobinsonHL & Pertmer TM. DNA vaccines for viral infections: basic studies and applications. Advances in Virus Research 2000; 55:1–74. https://doi.org/10.1016/s0065-3527(00)55001-5
79. Martins YA, Tsuchida CJ, Antoniassi P &Demarchi IG. Efficacy and Safety of the Immunization with DNA for Alzheimer’s Disease in Animal Models: A Systematic Review from Literature. Journal of Alzheimer’s Disease Reports 2017; 1(1):195–217. doi:10.3233/adr-170025
80. Ghochikyan A, Davtyan H, Petrushina I, Hovakimyan A, Movsesyan N, Davtyan A, Kiyatkin A, Cribbs DH & Agadjanyan MG. Refinement of a DNA based Alzheimer’s disease epitope vaccine in rabbits. Human Vaccines & Immunotherapeutics 2013; 9(5):1002–1010. https://doi.org/10.4161/hv.23875
81. Movsesyan N, Ghochikyan A, Mkrtichyan M, Petrushina I, Davtyan H, Olkhanud PB, et al. Reducing AD-Like Pathology in 3xTg-AD Mouse Model by DNA Epitope Vaccine — A Novel Immunotherapeutic Strategy. PLoS ONE 2008; 3(5):e2124. doi:10.1371/journal.pone.0002124
82. Bach P, Tschäpe JA, Kopietz F, Braun G, Baade JK, Wiederhold KH, Staufenbiel M, Prinz M, Deller T, Kalinke U, Buchholz CJ & Müller UC. Vaccination with Abeta-displaying virus-like particles reduces soluble and insoluble cerebral Abeta and lowers plaque burden in APP transgenic mice. Journal of Immunology (Baltimore, Md.:1950) 2009; 182(12):7613–7624. https://doi.org/10.4049/jimmunol.0803366
83. Petrushina, I, Ghochikyan A, Mktrichyan M, Mamikonyan G, Movsesyan N, Davtyan H, Patel A, Head E, Cribbs DH &Agadjanyan MG. Alzheimer’s disease peptide epitope vaccine reduces insoluble but not soluble/oligomeric Abeta species in amyloid precursor protein transgenic mice. The Journal of Neuroscience :The official Journal of the Society for Neuroscience 2007; 27(46):12721–12731. https://doi.org/10.1523/JNEUROSCI.3201-07.2007
84. Zou J, Yao Z, Zhang G, Wang H, Xu J, Yew DT & Forster EL. Vaccination of Alzheimer’s model mice with adenovirus vector containing quadrivalent foldable Abeta(1-15) reduces Abeta burden and behavioral impairment without Abeta-specific T cell response. Journal of the Neurological Sciences 2008; 272(1-2):87–98. https://doi.org/10.1016/j.jns.2008.05.003
85. Lambracht-Washington D, Qu BX, Fu M, Anderson LD, Jr Eagar TN, Stüve O & Rosenberg RN. A peptide prime-DNA boost immunization protocol provides significant benefits as a new generation Aβ42 DNA vaccine for Alzheimer disease. Journal of Neuroimmunology 2013; 254(1-2):63–68. https://doi.org/10.1016/j.jneuroim.2012.09.008
86. Kim H-D, Jin J-J, Maxwell JA & Fukuchi K. Enhancing Th2 immune responses against amyloid protein by a DNA prime-adenovirus boost regimen for Alzheimer’s disease. Immunology Letters 2007; 112(1):30–38. doi:10.1016/j.imlet.2007.06.006
87. Rosenberg RN, Fu M &Lambracht-Washington D. Active full-length DNA Aβ42 immunization in 3xTg-AD mice reduces not only amyloid deposition but also tau pathology. Alzheimer’s Research & Therapy 2018; 10(1). doi:10.1186/s13195-018-0441-4
88. Davtyan H, Chen WW, Zagorski K, Davis J, Petrushina I, Kazarian K, Cribbs DH, Agadjanyan MG, Blurton-Jones M &Ghochikyan A. MultiTEP platform-based DNA epitope vaccine targeting N-terminus of tau induces strong immune responses and reduces tau pathology in THY-Tau22 mice. Vaccine 2017; 35(16):2015–2024. https://doi.org/10.1016/j.vaccine.2017.03.020
89. Matsumoto Y, Niimi N & Kohyama K. Development of a New DNA Vaccine for Alzheimer Disease Targeting a Wide Range of Aβ Species and Amyloidogenic Peptides. PLoS ONE 2013; 8(9):e75203. doi:10.1371/journal.pone.0075203
90. Xing XN, Zhang WG, Sha S, Li Y, Guo R, Wang C & Cao YP. Amyloid β 3-10 DNA vaccination suggests a potential new treatment for Alzheimer’s disease in BALB/c mice. Chinese Medical Journal 2011; 124(17):2636–2641.
91. Olkhanud PB, Mughal M, Ayukawa K, Malchinkhuu E, Bodogai M, Feldman N, Rothman S, Lee JH, Chigurupati S, Okun E, Nagashima K, Mattson MP & Biragyn A. DNA immunization with HBsAg-based particles expressing a B cell epitope of amyloid β-peptide attenuates disease progression and prolongs survival in a mouse model of Alzheimer’s disease. Vaccine 2012; 30(9):1650–1658. https://doi.org/10.1016/j.vaccine.2011.12.136
92. Qu B, Boyer PJ, Johnston SA, Hynan LS & Rosenberg RN. Abeta42 gene vaccination reduces brain amyloid plaque burden in transgenic mice. Journal of the Neurological Sciences 2006; 244(1-2):151–158. https://doi.org/10.1016/j.jns.2006.02.006
93. Prins ND & Scheltens P. Treating Alzheimer’s disease with monoclonal antibodies: current status and outlook for the future. In Alzheimer’s Research & Therapy 2013; 5(6):56. https://doi.org/10.1186/alzrt220
94. Panza F, Solfrizzi V, Imbimbo BP, Tortelli R, Santamato A &LogroscinoG. Amyloid-based immunotherapy for Alzheimer’s disease in the time of prevention trials: the way forward. In Expert Review of Clinical Immunology 2014; 10(3):405–419. https://doi.org/10.1586/1744666x.2014.883921
95. Moreth J, Mavoungou C &Schindowski K. Passive anti-amyloid immunotherapy in Alzheimer’s disease: What are the most promising targets? In Immunity & Ageing 2013; 10(1):18. https://doi.org/10.1186/1742-4933-10-18
96. Morgan D. Immunotherapy for Alzheimer’s disease. In Journal of Internal Medicine 2011; 269(1):54–63). https://doi.org/10.1111/j.1365-2796.2010.02315.x
97. Bard F, Cannon C, Barbour R, Burke R-L, Games D, Grajeda H, Guido T, Hu K, et al. Peripherally administered antibodies against amyloid β-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease. Nature Medicine 2000; 6(8):916–919.
98. Montoliu-Gaya L & Villegas S. Aβ-Immunotherapeutic strategies: a wide range of approaches for Alzheimer’s disease treatment. Expert Reviews in Molecular Medicine 2016; 18. https://doi.org/10.1017/erm.2016.11
99. Frost JL, Liu B, Kleinschmidt M, Schilling S, Demuth H-U &Lemere CA. Passive Immunization against Pyroglutamate-3 Amyloid-β Reduces Plaque Burden in Alzheimer-Like Transgenic Mice: A Pilot Study. Neurodegenerative Diseases 2012; 10(1-4):265–270.
100. Venkataramani V, Wirths O, Budka H, Härtig W, Kovacs GG & Bayer TA. Antibody 9D5 Recognizes Oligomeric Pyroglutamate Amyloid-β in a Fraction of Amyloid-β Deposits in Alzheimer’s Disease without Cross-Reactivity with other Protein Aggregates. Journal of Alzheimer’s Disease 2012; JAD29(2):361–371.
101. Black RS, Sperling RA, Safirstein B, Motter RN, Pallay A, Nichols A & Grundman M. A single ascending dose study of bapineuzumab in patients with Alzheimer disease. Alzheimer Disease and Associated Disorders 2010; 24(2):198–203.
102. Sperling R, Salloway S, Brooks DJ, Tampieri D, Barakos J, Fox NC, Raskind M, et al. Amyloid-related imaging abnormalities in patients with Alzheimer’s disease treated with bapineuzumab: a retrospective analysis. Lancet Neurology 2012; 11(3):241–249.
103. Bagyinszky E, Youn YC, An SSA & Kim S. The genetics of Alzheimer’s disease. Clinical Interventions in Aging 2014; 9:535–551.
104. Blennow K, Zetterberg H, Rinne JO, Salloway S, Wei J, Black R, Grundman M, Liu E & for the AAB-001 201/202 Investigators Effect of Immunotherapy With Bapineuzumab on Cerebrospinal Fluid Biomarker Levels in Patients With Mild to Moderate Alzheimer Disease. Archives of Neurology 2012; 69(8):1002–1010
105. Sperling RA, Jack CR, Jr Black SE, Frosch MP, Greenberg SM, Hyman BT, Scheltens P. Amyloid-related imaging abnormalities in amyloid-modifying therapeutic trials: recommendations from the Alzheimer’s Association Research Roundtable Workgroup. Alzheimer’s &Dementia: The Journal of the Alzheimer’s Association 2011; 7(4):367–385. https://doi.org/10.1016/j.jalz.2011.05.2351
106. Aisen, P.S. Failure After Failure. What Next in AD Drug Development?. J Prev Alzheimers Dis 2019; 6:150. https://doi.org/10.14283/jpad.2019.2
107. Siemers ER, Sundell KL, Carlson C, Case M, SethuramanG, Liu-Seifert H, Dowsett SA, Pontecorvo MJ, Dean RA &Demattos R. Phase 3 solanezumab trials: Secondary outcomes in mild Alzheimer’s disease patients. Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association 2016; 12(2):110–120.
108. Dodart JC, Bales KR, Gannon KS, Greene SJ, DeMattos RB, Mathis C, DeLong CA, Wu S, Wu X, Holtzman DM & Paul SM. Immunization reverses memory deficits without reducing brain Aβ burden in Alzheimer’s disease model. Nature Neuroscience 2002; 5(5):452–457.
109. Mably AJ, Liu W, Mc Donald JM, Dodart JC, Bard F, Lemere CA, O’Nuallain B & Walsh DM. Anti-Aβ antibodies incapable of reducing cerebral Aβ oligomers fail to attenuate spatial reference memory deficits in J20 mice. In Neurobiology of Disease 2015; 82:372–384. https://doi.org/10.1016/j.nbd.2015.07.008
110. Farlow M, Arnold SE, van Dyck CH, Aisen PS, Joy Snider B, Porsteinsson AP, Friedrich S, et al. Safety and biomarker effects of solanezumab in patients with Alzheimer’s disease. In Alzheimer’s & Dementia 2012; 8(4):261–271. https://doi.org/10.1016/j.jalz.2011.09.224
111. Neurology TL & The Lancet Neurology. Solanezumab: too late in mild Alzheimer’s disease? In The Lancet Neurology 2017; 16(2):97. https://doi.org/10.1016/s1474-4422(16)30395-7
112. Topline Result for First DIAN-TU Clinical Trial: Negative on Primary 10 Feb 2020
113. https://www.alzforum.org/therapeutics/solanezumab updated May 2020
114. Bateman RJ, Benzinger TL, Berry S, Clifford DB, Duggan C, Fagan AM, Fanning K, et al. The DIAN-TU Next Generation Alzheimer’s prevention trial: Adaptive design and disease progression model. Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association 2017; 13(1):8–19.
115. Sperling RA, Rentz DM, Johnson KA, Karlawish J, Donohue M, Salmon DP & Aisen P. The A4 Study: Stopping AD Before Symptoms Begin? Science Translational Medicine 2014; 6(228):228fs13–fs228fs13.
116. Bohrmann B, Baumann K, Benz J, Gerber F, Huber W, Knoflach F, Messer J, Oroszlan K. Gantenerumab: A Novel Human Anti-Aβ Antibody Demonstrates Sustained Cerebral Amyloid-β Binding and Elicits Cell-Mediated Removal of Human Amyloid-β. Journal of Alzheimer’s Disease: JAD 2012; 28(1):49–69.
117. Ostrowitzki S, Lasser RA, Dorflinger E, Scheltens P, Barkhof F, Nikolcheva T, Ashford E, et al. A phase III randomized trial of gantenerumab in prodromal Alzheimer’s disease. Alzheimer’s Research & Therapy 2017; 9(1):1–15.
118. High-Dose Gantenerumab Lowers Plaque Load 13 Dec 2017
119. https://www.alzforum.org/therapeutics/gantenerumab updated 25 January 2021
120. Ultsch M, Li B, Maurer T, Mathieu M, Adolfsson O, Muhs A, Pfeifer A, Pihlgren M. Structure of Crenezumab Complex with Aβ Shows Loss of β-Hairpin. Scientific Reports 2016; 6:39374.
121. Adolfsson O, Pihlgren M, Toni N, Varisco Y, Buccarello AL, Antoniello K, Lohmann S, et al. An Effector-Reduced Anti-β-Amyloid (Aβ) Antibody with Unique Aβ Binding Properties Promotes Neuroprotection and Glial Engulfment of Aβ. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 2012; 32(28):9677–9689.
122. Cummings JL, Cohen S, van Dyck CH, Brody M, Curtis C, Cho W, Ward M, et al. ABBY: A phase 2 randomized trial of crenezumab in mild to moderate Alzheimer disease. Neurology 2018; 90(21):e1889–e1897.
123. Tariot PN, Lopera F, Langbaum JB, Thomas RG, Hendrix S, Schneider LS, Rios-Romenets S, et al. The Alzheimer’s Prevention Initiative Autosomal-Dominant Alzheimer’s Disease Trial: A study of crenezumab versus placebo in preclinical PSEN1 E280A mutation carriers to evaluate efficacy and safety in the treatment of autosomal-dominant Alzheimer’s disease, including a placebo-treated noncarrier cohort. Alzheimer’s & dementia (New York, N. Y.) 2018; 4:150–160. https://doi.org/10.1016/j.trci.2018.02.002
124. Porte SLL, La Porte SL, Bollini SS, Lanz TA, Abdiche YN, Rusnak AS, Ho W.H, et al. Structural Basis of C-terminal β-Amyloid Peptide Binding by the Antibody Ponezumab for the Treatment of Alzheimer’s Disease. In Journal of Molecular Biology 2012; 421(4-5):525–536). https://doi.org/10.1016/j.jmb.2011.11.047.
125. Landen JW, Zhao Q, Cohen S, Borrie M, Woodward M, Billing CB, Jr Bales K, Alvey C, McCush F, Yang J, Kupiec JW & Bednar MM. Safety and pharmacology of a single intravenous dose of ponezumab in subjects with mild-to-moderate Alzheimer disease: a phase I, randomized, placebo-controlled, double-blind, dose-escalation study. Clinical Neuropharmacology 2013; 36(1):14–23.
126. Topline Results: 18 Months of BAN2401 Might Work 7 Jul 2018
127. Logovinsky, V., Satlin, A., Lai, R., Swanson, C., Kaplow, J., Osswald, G., Basun, H., & Lannfelt, L. Safety and tolerability of BAN2401–a clinical study in Alzheimer’s disease with a protofibril selective Aβ antibody. Alzheimer’s Research & Therapy 2016 ;8(1), 14. https://doi.org/10.1186/s13195-016-0181-2
128. Vellas B, Aisen P. New Hope for Alzheimer’s Disease. J Prev Alzheimers Dis 2021; 8, 238–239. https://doi.org/10.14283/jpad.2021.26
129. Sevigny J, Chiao P, Bussière T, Weinreb PH, Williams L, Maier M, DunstanR, et al. The antibody aducanumab reduces Aβ plaques in Alzheimer’s disease. Nature 2016; 537(7618):50–56). https://doi.org/10.1038/nature19323.
130. Aducanumab, Solanezumab, Gantenerumab Data Lift Crenezumab, As Well 10 Aug 2015
131. Sevigny J, Chiao P, Bussière T, Weinreb PH, Williams L, Maier M, et al. The antibody aducanumab reduces Aβ plaques in Alzheimer’s disease. Nature 2016; 537(7618):50-6. PubMed.
132. Keep Your Enthusiasm? Scientists Process Brutal Trial Data 16 May 2019
133. ‘Reports of My Death Are Greatly Exaggerated.’ Signed, Aducanumab 24 Oct 2019
134. Biogen Asks FDA To Approve Aducanumab 8 Jul 2020
135. Advisory Committee Again Urges FDA to Vote No on Aducanumab 14 Apr 2021
136. Aducanumab Approved to Treat Alzheimer’s Disease 7 Jun 2021
137. Szabo P, Mujalli DM, Rotondi ML, Sharma R, Weber A, Schwarz H-P, Weksler ME & Relkin N. Measurement of anti-beta amyloid antibodies in human blood. Journal of Neuroimmunology 2010; 227(1-2):167–174
138. Relkin NR, Szabo P, Adamiak B, Burgut T, Monthe C, Lent RW, Younkin S, Younkin L, Schiff R & Weksler ME. 18-Month study of intravenous immunoglobulin for treatment of mild Alzheimer disease. Neurobiology of Aging 2009; 30(11):1728–1736.
139. Imbimbo BP, Ippati S, Ceravolo F, Watling M. Perspective: Is therapeutic plasma exchange a viable option for treating Alzheimer’s disease?. Alzheimer’s & Dementia: Translational Research & Clinical Interventions 2020; 6(1):e12004.
140. Loeffler DA. AMBAR, an Encouraging Alzheimer’s Trial That Raises Questions. Frontiers in Neurology, 2020; 11:459. https://doi.org/10.3389/fneur.2020.00459
141. Liu YH, Giunta B, Zhou HD, Tan J, Wang YJ. Immunotherapy for Alzheimer disease: the challenge of adverse effects. Nat Rev Neurol. 2012 Aug;8(8):465-9.

REPRESENTATION OF RACIAL AND ETHNIC MINORITY POPULATIONS IN DEMENTIA PREVENTION TRIALS: A SYSTEMATIC REVIEW

 

A.R. Shaw1, J. Perales-Puchalt1, E. Johnson1, P. Espinoza-Kissell1, M. Acosta-Rullan1, S. Frederick1, A. Lewis1, H. Chang2, J. Mahnken2, E.D. Vidoni1

 

1. University of Kansas Alzheimer’s Disease Center, University of Kansas Medical Center, Kansas City, USA; 2. Department of Biostatistics, University of Kansas, Medical Center, Kansas City, USA

Corresponding Author: Eric Vidoni, 4350 Shawnee Mission Parkway, Fairway, KS, 66205, USA, Email: evidoni@kumc.edu; Phone: 913-588-5312; Fax: 913-945-5035

J Prev Alz Dis 2021;
Published online August 26, 2021, http://dx.doi.org/10.14283/jpad.2021.49

 


Abstract

Despite older racial and ethnic minorities (REMs) being more likely to develop dementia they are underrepresented in clinical trials focused on neurological disorders. Inclusion of REMs in dementia prevention studies is vital to reducing the impact of disparities in dementia risk. We conducted a systematic review to characterize the number of REM enrolled in brain health and prevention randomized controlled trials (RCTs). RTCs published from January 1, 2004 to April 21, 2020 were included. Participants were normal cognitive adults aged 45 years and older who participated in a Phase II or Phase III U.S. based preventative trial. Analyses were performed to examine differences in trial characteristics between RCTs that did and those that did not report race/ethnicity and to calculate the pooled proportion of each racial/ethnic group in randomized brain healthy prevention trials. A total of 42 studies consisting of 100,748 participants were included in the final analyses. A total of 26 (62%) reported some racial/ethnic identity data. The pooled proportion of REM participants was 0.256 (95% CI, 0.191, 0.326). There is a lack of racial/ethnic reporting of participants and REMs remain underrepresented in brain health prevention RCTs.

Key words: Alzheimer’s disease, dementia, underrepresented minorities, aging, prevention.


 

Introduction

In 2021 it is estimated that 6.2 million older Americans are living with dementia (1). The population aged 65 years and older is projected to increase from 56 million to 94.7 million by 2060, with much of the growth being affiliated with the aging of Baby boomers (2). Additionally, the proportion of racial and ethnic minority (REM) older adults is expected to increase among African Americans (9% to 13%) (3) and Hispanics (8% to 21%) (4) by 2060. As the older population continues to rapidly grow it is projected that the number people living with dementia will also increase. Among people 65 years and older African Americans and Hispanics are disproportionally impacted. Previous research has indicated that older African Americans are twice as likely and older Hispanics are 1.5 times as likely to develop dementia compared to non-Hispanic Whites (1). Despite the growing number of racial/ethnic minorities in the United States and the current disproportionate impact of dementia among these populations, there is a poor representation of these groups in randomized control trials (RCTs) focused on neurological disorders (5-7). Furthermore, it is estimated that in dementia RCTs, minority participation rates are lower than 5% (8).
There are several possible reasons why most racial and ethnic minorities are underrepresented in RCTs, including historical unethical practices and cultural barriers. Unethical practices such as the Tuskegee syphilis study in which researchers infected and withheld treatment for syphilis to Black men (9) and more recently the Havasupai Tribe case, in which DNA samples initially collected for genetic markers of type 2 diabetes had been used in several unrelated studies such as schizophrenia, and migration without consent from tribal members (10) have led to mistrust in research field among REMs. Cultural barriers in RCTs have been documented as lack of tailoring to diverse communities, implicit bias, and investigators; limiting participation of REM within studies (11). Yet, previous research has indicated that REMs are willing to participate in clinical trials when presented with the opportunity and when trial objectives can be translated in a culturally relevant manner (12), which demonstrates that REMs are not necessarily hard to reach but are rarely reached.
Race is a socially constructed category and a proxy for unique psychosocial factors strongly related to dementia (13) that needs to be considered when designing dementia prevention interventions. Because dementia prevention trials have primarily focused on non-Hispanic Whites, progress in research related to characteristics of dementia among REM has been limited. Dementia prevention is likely to be a critical aspect in reducing racial and ethnic disparities (14). However, disease prevention is not a one-size fits all model and it is imperative that preventative approaches aimed at mitigating risk factors of dementia among REM incorporate culture.
Inclusion of REMs in dementia prevention studies is vital to reducing the impact of disparities and critical for addressing imperative gaps in knowledge. Therefore, we conducted a systematic review to characterize the number of REMs enrolled in brain health and prevention trials.

 

Methods

We searched Ovid MEDLINE, Embase, CINAHL Complete, and clinical trials.gov published in English from January 1, 2004 to April 21, 2020. Minimum race and ethnicity reporting standards were adopted by the National Institutes of Health in January 2002 (NOT-OD-01-053). Our search window allows for the completion and reporting of smaller clinical trial projects initiated following this directive. Additionally, we screened references from eligible studies to determine additional eligible articles to include in the review. A detailed search strategy of this review is available in the supplementary materials.
We included published RCTs that met the following eligibility criteria 1) enrollees with normal cognition ages 45 years and older, 2) Phase II or Phase III randomized controlled trials, 3) at least one explicitly identified cognitive outcome measure, and 4) United States-based trials. We excluded investigational medication trials seeking FDA approval, trials aimed at treatment of existing cognitive impairment, psychiatric-related cognitive trials (i.e. major depression as primary diagnosis), protocol related articles, clinical trials that did not provide results, RCTs with no cognitive outcome examined, retrospective, and secondary articles.
Eight reviewers (ARS, EDV, JPP, EDJ, SIF, PEK, MDA, AL) independently screened all titles and abstracts. Two reviewers (ARS and EDV) cross checked all titles, abstracts, and completed full text-review of all eligible studies following screening. Information was abstracted from all eligible studies: publication information (first author, title, journal, PubMed ID [PMID], year of publication); funding source; demographics of enrollees (total number of participants, average age, number of females, number and type of race or ethnicity as defined by the study); study design data (type of intervention, intervention components, primary language of intervention delivery, cognitive tests used); Percentages of ethno-racial groups in the eligible studies were only included in this review if it was specifically mentioned in the manuscript.
We examined the differences in trial characteristics between RCTs that did and those that did not report race/ethnicity of trial participants using Student’s t-test for continuous variables and X2 tests for categorical variables. Study characteristics examined included average proportion female, average age, sample size (mean), type of intervention (i.e. diet/supplement vs exercise vs. cognitive training vs. multi-domain), funding source, non-English language delivery (yes vs. no).
We conducted a meta-regression analysis to calculate the pooled proportion of each racial/ethnic group in randomized brain healthy prevention trials. Our primary outcome was non-White/non-Hispanic which included a composite of the following racial/ethnic groups African American, Hispanic or Latinx (hereafter referred to as Hispanic), Asian, American Indian or Alaskan Native, Native Hawaiian or Pacific Islander, and Other Race, following standard NIH reporting guidance. We conducted a subgroup analysis to assess the heterogeneity across different study factors including type of intervention (diet/supplement, exercise, cognitive training, multi-domain), cognitive tests in intervention (MMSE, Rey Auditory Verbal, other) and funding (public, private, or mixed). The pooled estimate of proportion and 95% confidence (15) interval were calculated using random effects meta-analyses with inverse variance weighting. The Freeman-Tukey double arsine transformation of the proportion was used in the estimation (16). Analyses were performed using SAS 9.4 and R 4.0.2. We used the metaprop function from the “meta” package in R to calculate the pooled estimate and confidence intervals (17).

 

Results

A total of 4,600 articles were screened: 4385 non-duplicate abstracts identified via our search strategy, and 215 articles manually added. After review of titles and abstracts, 49 underwent a full text review. Of the 49 articles reviewed; 7 were excluded for not being RCTs or being derivative of the primary report (n=6), or including participants younger than 45 years old (n=1). A total of 42 articles were eligible for inclusion. The study selection flow diagram is shown in Figure 1.

Figure 1. PRISMA flow diagram of reviewed publications and results

 

Of the 42 studies, 26 (62%) reported some ethno-racial identity data. The 42 eligible studies included a total of 100,748 participants, including the 76365 participants in studies that reported ethno-racial information. White race was reported in 23 trials, Black or African American in 15, Asian in 6, American Indian or Alaska Native in 3, Hawaii Native or Pacific Islander in 1. No studies reported bi-racial identity. In 9 of these trials, White race was explicitly centered and either no other race categories were listed, or all other races were combined into an “Other Race” category. In one instance, only the African American proportion of the sample was reported and all other races including White were captured under “Other Race.” Hispanic ethnicity was reported in 10 trials, including two trials for only Hispanic individuals. In all studies, Hispanic ethnicity appeared to be included as a separate ethno-racial category with no intersection with an identified racial identity. No studies reported enrolling exclusively White, non-Hispanic participants.
Table 1 shows the characteristics of the eligible studies stratified by whether the studies reported ethno-racial information or not. Studies that reported ethno-racial information did not have different sample sizes ( p=0.48, CI [-5459.7, 2633.4]), average age (p=0.75, CI [-5.4, 3.9]), or have a different percentage of women (p=0.24, CI [-13.9, 3.5]). Studies that reported ethno-racial information did not employ different intervention types (X2 = 8.0, p=0.33), or have significantly different funding (X2 = 9.8, p=0.08) but in general were more often funded by the National Institutes of Health (62% vs 19%). Only two studies were explicitly delivered in Spanish, both of which exclusively enrolled individuals who identified as Hispanic.

Table 1. Characteristics of Eligible Studies

Some combination of race and/or ethnicity was reported in 26 studies. Number or percentage of individuals who were white were reported in 23 trials, Black or African American in 15 trials, Asian in 6 trials, American Indian or Alaska Native in 3 trials, Hawaii Native or Pacific Islander in 1 trial. Hispanic ethnicity was reported in 10 trials and appeared to be included as a separate ethno-racial category with no overlap with race. Funding identification based on sources listed in text.

 

The overall and subgroup frequency of each level and the pooled estimated proportions of participants from ethno-racial minority groups are illustrated using forest plotting, Figure 2. The plot displays the estimate and confidence intervals for both overall and subgroup analyses. For studies that reported racial or ethnic identity of participants, the estimated pooled percentage of REM participation was 25.6%. Only the type of intervention demonstrated differences in the proportion of ethno-racial minorities in the sample, (p<0.003). Specifically, diet studies had a lower estimated pooled proportion of REM enrollees. There was insufficient evidence to conclude the proportion of minority enrollees different by funding source or cognitive measure employed.

Figure 2. Pooled Proportion of REM in brain healthy prevention trials between 2004 and 2020 using meta-analyses, and subgroup analyses to assess heterogeneity

 

Discussion

In the current study, we conducted a systematic review to characterize the number of REMs enrolled in brain health and prevention trials. We conducted this research because inclusion of REMs in dementia prevention studies is vital to reducing the impact of disparities in dementia risk and critical for addressing imperative gaps in knowledge. Our findings suggest that, between 2004 and 2020, one third of studies failed to report ethno-racial information. Of the studies that did report this information, we found the estimated pooled percentage of REM participation was 25.6%.
This estimate is higher than the representation of REMs in dementia pharmacological treatment RCTs as reported in a 2007 review(18), 10% in NIH and 3.2% in industry-funded RCTs. However, it is important to note that one third of studies in the current review did not report race or ethnicity, which suggests that the numbers of minorities are likely low. Also, 25.6% is lower than the 2019 the U.S. Census Bureau estimate of 30.4% people 45 years and older that identified as a REM, demonstrating that REM remain under-represented in preventative RCTs (19). Our study findings of the lack of representation of REM in brain health and prevention trials, aligns with current evidence demonstrating the underrepresentation of culturally and linguistical populations in clinical trials (20-22). Lack of racial/ethnic representation in clinical trials is problematic for generalizability of study findings and equity for REM in obtaining the benefits of participating in clinical trials (23). It is also troubling given that at a minimum, we know that individuals identifying as African American, Hispanic, and American Indians are at greater risk of developing cognitive change (24, 25).
Barriers to REM participation in clinical trials are well-documented (26, 27) due to past unethical research practices that have had a significant influence on the community resulting in mistrust of research, medical, and academic institutions as a whole (28, 29). Additionally, cultural and institutional barriers have impacted REM participation including primary use of passive recruitment strategies (e.g. flyers) that are not culturally tailored to REM communities. Study designs and inclusion criteria for clinical trials also present significant challenges for REM participation (e.g. medical/health eligibility criteria, time/duration of study, transportation requirements to go to study sites) (11, 13, 30).
The prevalence of dementia is high among REM, especially among older African Americans (13.8%) and Hispanics (12.2%) compared to Whites (10.3%) (31). Although evidence consistently indicates that disparities in dementia disproportionately impact REM, especially in terms of incidence, prevalence, diagnosis, and disease burden, the same populations have been historically underrepresented and nearly absent in dementia research (14, 18, 25, 32), in which less than 4% of ADRD prevention brain health trials are focused on REM communities.(33) However, this review found that an estimated 25.6% enrollment of REM in dementia prevention trials, which indicates there have been improvements made within research to enhance participation and inclusion of REM in clinical trials.
Effective approaches to increase recruitment among REM in research have been made, which emphasize the importance of forming sustainable partnerships with REM communities (e.g. community centers, churches, and trusted community leaders). Centering the community partnerships at all phases of the research in useful to acquiring buy-in from the community as a whole.(34) For example, a church-based HIV intervention used community based participatory research approach to increase HIV testing among African Americans resulting in increased HIV testing (59% vs. 42%, p = 0.008) and church-based testing (54% vs. 15%, p < 0.001) within 12 months (35). In another in trial, Promotoras de Salud delivered a diabetes prevention program for prediabetic Latina adults in Spanish. Participants reported overall high satisfaction with the culturally tailored program and results indicated significant reduction in weight loss (5.6% of initial body weight) and cardiovascular related risk factors (e.g. diastolic blood pressure, insulin, and LDL cholesterol) (36). These trials demonstrate that community driven trials that are culturally tailored result in feasibility, acceptability, and effectiveness in addressing health disparities among REM. Therefore, it is imperative that dementia prevention trials involve community partners to enhance recruitment and acceptability among REM. In addition, resources through the NIA’s Office of Special Populations are available to support recruitment and retention efforts for REM (37). Incorporating technology to support recruitment and participation has shown to be effective in reaching REM and reducing barriers to participation such as time. The Trial-Ready Cohort for Preclinical/Prodromal Alzheimer’s Disease program enhanced timely recruitment into trials by leveraging a cost-effective information technology infrastructure (38). Also, the internet-based platform Alzheimer’s Prevention Registry raised awareness about AD prevention trials and served a recruitment tool to connect members in the community to current enrolling trials (39). These studies have demonstrated that using technology can serve as an effective tool to support recruitment of REM who have traditionally been underrepresented in trials. Policy-related strategies might also help increase the representation of REMs in dementia prevention studies. The NIA is considering a funding strategy for Practice Based Research Networks, which have been reported to increase access to a more diverse population (40). Another policy may be ensuring accountability by means of contingencies upon achieving the quotas of REM participation proposed in the research design. The NIA Clinical Research Operations & Management System (CROMS), currently being piloted, may assist track recruitment and retention to make it more equitable (41). Providing incentives has also been shown to enhance participation in dementia clinical trials. African Americans reported that receiving financial compensation, cognitive and genetic tests results would make them more likely to enroll in dementia focused clinical trials (28), demonstrating that cultural alignment and the use of incentives can support strides towards achieving representation in dementia prevention trials. It is important to note that diversifying samples will lead to greater generalizability of prevention recommendations. Tailored and targeted interventions can address specific social, environmental, and in limited cases possible biological differences between REM groups in the US. These benefits vastly outweigh any challenges to efficacy assessment that may be experienced in inclusive trials due to attrition.
This study incorporated a comprehensive systematic methodological search using both electronic databases and gray literature. This study had a few limitations which merit discussion. First, we applied a U.S. based study limit to our review to focus on U.S. based REMs which limits generalization of findings to REMs outside of U.S. Second, we limited our search to studies reported in English, which has been argued to result in systematic bias (42). Third, we included only studies that reported change in cognitive scores, therefore excluding those that only reported dementia or MCI incidence as an outcome. Future research should explore those studies, although there are likely few, due to the required long time periods to assess those outcomes.
In conclusion, this systematic review highlights the lack of ethno-racial reported among participants in brain health prevention RCT trials. Representation of REM is dementia prevention trials is critical to reducing the disproportionate burden dementia has among these populations. Reporting of ethno-racial within dementia prevention trials is encouraged and use of effective recruitment including collaboration with community partners is suggested to enhance recruitment for future dementia prevention trials.

 

Funding: This work was supported by the National Institute of Aging grant number P30AG035982.

Acknowledgements: Not applicable.

Disclosures/Competing interests: The authors declare that they have no competing interests.

 

References

1. Association As. Facts and Figures 2021 [Available from: https://www.alz.org/alzheimers-dementia/facts-figures.
2. Vespa J, Armstrong DM, Medina L. Demographic turning points for the United States: Population projections for 2020 to 2060: US Department of Commerce, Economics and Statistics Administration, US …; 2018.
3. Living AfC. 2018 Profile of African Americans Age 65 and Over. Washington DC: U.S. Department of Health and Human Services; 2018.
4. Living AfC. 2018 Profile of Hispanic Americans Age 65 and Over. Washington DC: U.S. Department of Health and Human Services; 2018.
5. Burneo JG, Martin R. Reporting race/ethnicity in epilepsy clinical trials. Epilepsy & Behavior. 2004;5(5):743-5.
6. Zhang Y, Ornelas IJ, Do HH, Magarati M, Jackson JC, Taylor VM. Provider perspectives on promoting cervical cancer screening among refugee women. J Community Health. 2017;42(3):583-90.
7. Robbins NM, Bernat JL. Minority representation in migraine treatment trials. Headache: The Journal of Head and Face Pain. 2017;57(3):525-33.
8. George S, Duran N, Norris K. A systematic review of barriers and facilitators to minority research participation among African Americans, Latinos, Asian Americans, and Pacific Islanders. Am J Public Health. 2014;104(2):e16-31.
9. Scharff DP, Mathews KJ, Jackson P, Hoffsuemmer J, Martin E, Edwards D. More than Tuskegee: understanding mistrust about research participation. J Health Care Poor Underserved. 2010;21(3):879-97.
10. Garrison NA. Genomic Justice for Native Americans: Impact of the Havasupai Case on Genetic Research. Sci Technol Human Values. 2013;38(2):201-23.
11. Green-Harris G, Coley SL, Koscik RL, Norris NC, Houston SL, Sager MA, et al. Addressing Disparities in Alzheimer’s Disease and African-American Participation in Research: An Asset-Based Community Development Approach. Frontiers in Aging Neuroscience. 2019;11:125.
12. Forrester S, Jacobs D, Zmora R, Schreiner P, Roger V, Kiefe CI. Racial differences in weathering and its associations with psychosocial stress: The CARDIA study. SSM – Population Health. 2019;7:100319.
13. Konkel L. Racial and Ethnic Disparities in Research Studies: The Challenge of Creating More Diverse Cohorts. Environmental Health Perspectives. 2015;123(12):A297-A302.
14. Barnes LL, Bennett DA. Alzheimer’s disease in African Americans: risk factors and challenges for the future. Health Aff (Millwood). 2014;33(4):580-6.
15. Woodward M. Epidemiology: study design and data analysis: CRC press; 2013.
16. Lin L, Xu C. Arcsine-based transformations for meta-analysis of proportions: Pros, cons, and alternatives. Health Sci Rep. 2020;3(3):e178.
17. Schwarzer G. meta: An R package for meta-analysis. R news. 2007;7(3):40-5.
18. Faison WE, Schultz SK, Aerssens J, Alvidrez J, Anand R, Farrer LA, et al. Potential ethnic modifiers in the assessment and treatment of Alzheimer’s disease: challenges for the future. Int Psychogeriatr. 2007;19(3):539-58.
19. Bureau USC. Census estimates 2019 2019 [Available from: https://www2.census.gov/programs-surveys/popest/tables/2010-2019/national/asrh/nc-est2019-asr5h.xlsx.
20. McDougall GJ, Jr., Simpson G, Friend ML. Strategies for research recruitment and retention of older adults of racial and ethnic minorities. J Gerontol Nurs. 2015;41(5):14-23; quiz 4-5.
21. Stahl SM, Vasquez L. Approaches to improving recruitment and retention of minority elders participating in research: examples from selected research groups including the National Institute on Aging’s Resource Centers for Minority Aging Research. J Aging Health. 2004;16(5 Suppl):9s-17s.
22. Mody L, Miller DK, McGloin JM, Freeman M, Marcantonio ER, Magaziner J, et al. Recruitment and retention of older adults in aging research. J Am Geriatr Soc. 2008;56(12):2340-8.
23. Hughson J-a, Woodward-Kron R, Parker A, Hajek J, Bresin A, Knoch U, et al. A review of approaches to improve participation of culturally and linguistically diverse populations in clinical trials. Trials. 2016;17(1):263.
24. Perales-Puchalt J, Gauthreaux K, Shaw A, McGee JL, Teylan MA, Chan KCG, et al. Risk of mild cognitive impairment among older adults in the United States by ethnoracial group. Int Psychogeriatr. 2021;33(1):51-62.
25. Mayeda ER, Glymour MM, Quesenberry CP, Whitmer RA. Inequalities in dementia incidence between six racial and ethnic groups over 14 years. Alzheimer’s & Dementia. 2016;12(3):216-24.
26. Williams MM, Scharff DP, Mathews KJ, Hoffsuemmer JS, Jackson P, Morris JC, et al. Barriers and facilitators of African American participation in Alzheimer disease biomarker research. Alzheimer Dis Assoc Disord. 2010;24 Suppl(Suppl):S24-9.
27. Ballard EL, Gwyther LP, Edmonds HL. Challenges and opportunities: recruitment and retention of African Americans for Alzheimer disease research: lessons learned. Alzheimer Dis Assoc Disord. 2010;24 Suppl(0):S19-S23.
28. Zhou Y, Elashoff D, Kremen S, Teng E, Karlawish J, Grill JD. African Americans are less likely to enroll in preclinical Alzheimer’s disease clinical trials. Alzheimer’s & Dementia: Translational Research & Clinical Interventions. 2017;3(1):57-64.
29. Corbie-Smith G, Thomas SB, Williams MV, Moody-Ayers S. Attitudes and beliefs of African Americans toward participation in medical research. J Gen Intern Med. 1999;14(9):537-46.
30. Wendler D, Kington R, Madans J, Wye GV, Christ-Schmidt H, Pratt LA, et al. Are Racial and Ethnic Minorities Less Willing to Participate in Health Research? PLOS Medicine. 2005;3(2):e19.
31. Prevention CfDCa. U.S. burden of Alzheimer’s disease, related dementias to double by 2060 2018 [Available from: https://www.cdc.gov/media/releases/2018/p0920-alzheimers-burden-double-2060.html#:~:text=Among%20people%20ages%2065%20and,Pacific%20Islanders%20(8.4%20percent).
32. Gilmore-Bykovskyi AL, Jin Y, Gleason C, Flowers-Benton S, Block LM, Dilworth-Anderson P, et al. Recruitment and retention of underrepresented populations in Alzheimer’s disease research: A systematic review. Alzheimer’s & Dementia: Translational Research & Clinical Interventions. 2019;5:751-70.
33. Rayhan R, Rochelle, P., Holzaphel, D., Vradenburg, G. Non-Pharmacological Therapies in Alzheimer’s disease: A Systematic Review. 2019.
34. Samus QM, Amjad H, Johnston D, Black BS, Bartels SJ, Lyketsos CG. A Multipronged, Adaptive Approach for the Recruitment of Diverse Community-Residing Elders with Memory Impairment: The MIND at Home Experience. Am J Geriatr Psychiatry. 2015;23(7):698-708.
35. Berkley-Patton JY, Thompson CB, Moore E, Hawes S, Berman M, Allsworth J, et al. Feasibility and Outcomes of an HIV Testing Intervention in African American Churches. AIDS Behav. 2019;23(1):76-90.
36. O’Brien MJ, Perez A, Alos VA, Whitaker RC, Ciolino JD, Mohr DC, et al. The Feasibility, Acceptability, and Preliminary Effectiveness of a Promotora-Led Diabetes Prevention Program (PL-DPP) in Latinas:A Pilot Study. The Diabetes Educator. 2015;41(4):485-94.
37. Health NIo. Office of Special Populations [Available from: https://www.nia.nih.gov/research/osp.
38. Jimenez-Maggiora GA, Bruschi S, Raman R, Langford O, Donohue M, Rafii MS, et al. TRC-PAD: Accelerating Recruitment of AD Clinical Trials through Innovative Information Technology. J Prev Alzheimers Dis. 2020;7(4):226-33.
39. Langbaum JB, High N, Nichols J, Kettenhoven C, Reiman EM, Tariot PN. The Alzheimer’s Prevention Registry: A Large Internet-Based Participant Recruitment Registry to Accelerate Referrals to Alzheimer’s-Focused Studies. The journal of prevention of Alzheimer’s disease. 2020;7(4):242-50.
40. VideoCasting NIoH. Development of an NIA Practice-Based Research Network to Conduct Alzheimer’s and Related Dementias Clinical Research. 2021.
41. Aging NIo. NIA’s Clinical Research Operations & Management System (CROMS) 2021 [Available from: https://www.nia.nih.gov/research/grants-funding/nias-clinical-research-operations-management-system-croms.
42. Morrison A, Polisena J, Husereau D, Moulton K, Clark M, Fiander M, et al. THE EFFECT OF ENGLISH-LANGUAGE RESTRICTION ON SYSTEMATIC REVIEW-BASED META-ANALYSES: A SYSTEMATIC REVIEW OF EMPIRICAL STUDIES. International Journal of Technology Assessment in Health Care. 2012;28(2):138-44.

USING DIGITAL TOOLS TO ADVANCE ALZHEIMER’S DRUG TRIALS DURING A PANDEMIC: THE EU/US CTAD TASK FORCE

 

J. Kaye1, P. Aisen2, R. Amariglio3, R. Au4, C. Ballard5, M. Carrillo6, H. Fillit7, T. Iwatsubo8, G. Jimenez-Maggiora2, S. Lovestone9, F. Natanegara10, K. Papp3, M.E. Soto11, M. Weiner12, B. Vellas13, and the EU/US CTAD Task Force

 

*EU/US/CTAD TASK FORCE: Sandrine Andrieu (Toulouse); Matthew Barton (Raleigh); Randall Bateman (Saint Louis); Monika Baudler (Basel); Joanne Bell (Willmington); Kevin Biglan (Indianapolis); Adam Boxer (San Francisco); Sasha Bozeat (Basel); Claudine Brisard (Issy les Moulineaux); Miroslaw Brys (Indianapolis); Marc Cantillon (Gilbert); Bill Chan (Beijing); Ivan Cheung (Woodcliff Lake); Min Cho (Woodcliff Lake); Julia Coelho (San Francisco); Shobha Dhadda (Woodcliff Lake); Daniel Di Giusto (Basel); Michael Donohue (San Diego); Rachelle Doody (Basel); John Dwyer (Washington); Michael Egan (North Wales); Rianne Esquivel (Malvern); Wendy Galpern (New Jersey); Harald Hampel (Woodcliff Lake); Jason Hassenstab (St Louis); David Henley (New Jersey); Joseph Herring (North Wales); Carole Ho (South San Francisco); Michael Irizarry (Woodcliff Lake); Keith Johnson (Boston); Geoffrey Kerchner (South San Francisco); Gene Kinney (South San Francisco); Shailaja Korukonda (Woodcliff Lake); Lynn Kramer (Woodcliff Lake); Jaren Landen (Cambridge); Ishani Landri (Woodcliff Lake); Lars Lannfelt (Uppsala); Valérie Legrand (Nanterre); Manoj Malhotra (Woodcliff Lake); Eric McDade (St Louis); Francisco Nogueira (South San Francisco); Gerald Novak (New Jersey); Gunilla Osswald (Stockholm); Susanne Ostrowitzki (South San Francisco); Amanda Paley (New York); Martin Rabe (Woodcliff Lake); Rema Raman (San Diego); Elena Ratti (Cambridge); Laurie Ryan (Bethesda); Stephen Salloway (Providence); Peter Schüler (Langen); Hiroshi Sekiya (Malvern); Jiong Shi (Las Vegas); Melanie Shulman (Cambridge); Eric Siemers (Zionsville); John Sims (Indianapolis); Kaycee Sink (South San Francisco); Reisa Sperling (Boston); Joyce Suhy (Newark); Jina Swartz (London); Pierre Tariot (Phoenix); Edmond Teng (South San Francisco); Jacques Touchon (Montpellier); Martin Traber (Basel); Andrea Vergallo (Woodcliff Lake); Judy Walker (Singapore); Jon Walsh (San Francisco); Alette Wessels (Indianapolis); Haichen Yang (North Wales); Wagner Zago (San Francisco); Kenton Zavitz (Cambridge).

1. Layton Aging and Alzheimer’s Disease Center, School of Medicine, Oregon Health and Science University, Portland, OR, USA; 3. Alzheimer’s Therapeutic Research Institute, University of Southern California, San Diego, CA, USA; 3. Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA; 4. Boston University, Boston, MA, USA; 5. University of Exeter, Exeter, UK; 6. Alzheimer’s Association, Chicago IL, USA; 7. Alzheimer Drug Discovery Foundation, New York, NY, USA; 8. University of Tokyo, Tokyo, japan; 9. Janssen Pharmaceutical Company, High Wycombe, UK; 10. Eli Lilly and Company, Indianapolis IIN, USA; 11. Toulouse University Hospital, Toulouse, France; 12. University of California, San Francisco CA, USA; 13. Gerontopole, INSERM U1027, Alzheimer’s Disease Research and Clinical Center, Toulouse University Hospital, Toulouse, France

Corresponding Author: Jeffrey Kaye, Layton Aging and Alzheimer’s Disease Center, School of Medicine, Oregon Health and Science University, Portland, OR, USA, kaye@ohsu.edu

J Prev Alz Dis 2021;4(8):513-519
Published online June 28, 2021, http://dx.doi.org/10.14283/jpad.2021.36

 


Abstract

The 2020 COVID-19 pandemic has disrupted Alzheimer’s disease (AD) clinical studies worldwide. Digital technologies may help minimize disruptions by enabling remote assessment of subtle cognitive and functional changes over the course of the disease. The EU/US Clinical Trials in Alzheimer’s Disease (CTAD) Task Force met virtually in November 2020 to explore the opportunities and challenges associated with the use of digital technologies in AD clinical research. While recognizing the potential of digital tools to accelerate clinical trials, improve the engagement of diverse populations, capture clinically meaningful data, and lower costs, questions remain regarding the stability, validity, generalizability, and reproducibility of digital data. Substantial concerns also exist regarding regulatory acceptance and privacy. Nonetheless, the Task Force supported further exploration of digital technologies through collaboration and data sharing, noting the need for standardization of digital readouts. They also concluded that while it may be premature to employ remote assessments for trials of novel experimental medications, remote studies of non-invasive, multi-domain approaches may be feasible at this time.

Key words: Alzheimer’s disease, clinical outcomes, digital tools, remote assessments.


 

Introduction

The 2020 COVID-19 pandemic disrupted clinical trials worldwide as health care facilities became overwhelmed and lock-down conditions shuttered clinical trial sites, forcing both study staff and trial participants to stay home (1–3). Nearly every ongoing clinical study has felt the impact, although the degree of impact varies. Alzheimer’s disease (AD) studies have been acutely affected for multiple reasons, including: 1) the increased vulnerability to COVID-19 of people with or at risk of AD due to their advanced age and high prevalence of comorbid conditions; 2) the effects of even mild cognitive decline on the ability of participants to comply with the operational changes in how trials are conducted (e.g., social distancing from study staff and study partners, changes in study protocols); and 3) effects on clinical endpoints due to isolation, confusion, and behavioral and psychological symptoms of dementia (BPSD) such as agitation, apathy, and depression (1, 4). Other disruptions common to all clinical trials including those for AD interventions include: 1) reduced clinical capacity; 2) increased costs due to the need for increased personal protective equipment (PPE) and infrastructure for altered intervention delivery; 3) delays in ethics approvals as institutional review boards (IRBs) are swamped with COVID-19 related protocols; 4) reluctance of patients and participants to be in contact with clinics and health providers; and 5) suspension of patient screening and recruitment.
Study sponsors have responded to these challenges in different ways; for example, by mailing drugs to trial participants to ensure participants have access to interventions; limiting clinic visits through the establishment of alternative trial sites; allowing a brief hiatus from dosing; and implementing home infusions. Funders have also responded with innovations. In an analysis of their funding portfolio during the latter half of 2020, the Alzheimer’s Drug Discovery Foundation (ADDF) found that about 70% of grantees expected their studies to be delayed, and some requested no-cost extensions. Delays may harm not only budgets but data analyses as well, since data interpretation can be muddled due to long-term disruptions, missing data, and protocol deviations; with clinical studies more affected than preclinical studies. Observational studies have also been affected. For example, Swedish BioFINDER2 halted enrollment, and the Rush University aging cohort studies stopped performing autopsies due to lack of PPE. In response to these challenges, the Alzheimer’s Association introduced a Rapid Program in Dementia (RAPID) funding program to provide bridge funding for Association-funded research during the pandemic.
In March 2020, the U.S. Food and Drug Administration (FDA) issued guidelines on the conduct of clinical trials during the pandemic and updated this guidance in December 2020 (5). The guidance aims to protect the safety of trial participants while ensuring good clinical practice and trial integrity. It also recognized the likely use of remote assessments and new technologies for data acquisition.
In recent years, digital technologies that enable remote assessment and gather real-world granular data on a variety of cognitive and functional outcomes have received increased attention and interest in the AD drug development community (6). Recognizing the urgency of developing these technologies due to the pandemic, the EU/US CTAD Task Force met in November 2020 to address the topic, bringing together an international group of clinical investigators from academia and industry along with funders. They explored emerging digital technologies that are being used in clinical research as well as the challenges that have yet to be addressed to realize the potential of these technologies. First and foremost, they recognized that global challenges such as the COVID-19 pandemic require global solutions.

 

Opportunities and challenges to make trials more robust using digital technologies

Even if there were no pandemic, clinical trials as currently conducted are too slow and costly to adequately respond to a disease such as Alzheimer’s, which currently affects around 50 million people worldwide and for which there is no approved disease-modifying treatment (7). The difficulty of conducting clinical trials for AD treatments is exacerbated by the use of conventional outcome measures that are insensitive to change early in disease and sensitive to environmentally-induced variation. By capturing alternative outcome measures in real-world settings, digital devices have the potential to mitigate these problems, thereby expediting trials, lowering costs, and enabling more effective clinical trials. Digital tools can also promote inclusiveness and social justice, increase the pool of participants, decrease sample size, and increase trial effectiveness by improving the generalizability and representation of vulnerable populations. Smartphones hold particular promise in the realm of digital tools since they are among the most penetrating technologies available, with over three billion users worldwide (8) .
However, converting trial outcomes from in-person to remote measures is also associated with many challenges. Among the most critical aspects to be considered are usability and the reliability and fidelity of technology-related measures (9). Consumer-oriented devices and apps must be scaled for use in clinical trial settings and confounds introduced by remote assessment must be addressed. User training and support will be critical to reduce user errors. Technologies themselves should not drive their use in clinical trials; rather the use case should drive the technology.
In the AD space, digital tools have shown particular promise in capturing subtle cognitive and functional changes in preclinical AD. Yet there remain questions regarding their ability to serve as outcome measures due to questions around feasibility, reliability, and validity.

 

Implementing digital devices in AD clinical studies

As the COVID-19 pandemic has necessitated moving clinical trials from specialized centers to home-based virtual assessments using both passive and active measures, investigators are learning lessons that have the potential to accelerate all clinical trials for dementia. Indirect measures of behavior, function, and physiology are being explored as cognitive and functional endpoints for clinical trials by individual research groups as well as technology companies. For example, Elektra Labs has created the Atlas platform, which takes a data science approach to clinical trials through the application of a wide selection of connected sensors and physiological and behavioral markers.
Many research groups that have developed and/or implemented digital technologies in clinical studies presented data and plans to the Task Force, providing lessons learned as well as strategies for moving forward:

CART/ORCATECH Platform

The Collaborative Aging Research Using Technology (CART) Initiative funded by the US National Institutes of Health and the Department of Veterans Affairs, has developed a multi-functional digital assessment platform that is technology agnostic, use case flexible, sharable, and scalable (10). The platform, an extension of the Oregon Center for Aging & Technology (ORCATECH) assessment system at Oregon Health and Science University has been deployed to 301 older adults for up to three years in four US regions. It enables unobtrusive, long-term assessment across multiple domains (e.g., mobility, cognitive function, sleep) through the use of many different sensors and connected devices (e.g., electronic pillboxes, bed mats, actigraphs, online reporting on laptops, tablets or smartphones). The platform is currently deployed by multiple groups in the United States and Canada, with plans to extend the platform to France (in the community-wide INSPIRE study based at Toulouse University Hospital) and Australia. Testing in two cohorts — a low-income housing cohort in Portland, Oregon, and an African-American Minority Aging Research Study (MARS) cohort in Chicago – gathered data spanning the time period of the pandemic demonstrating increased loneliness and decreased physical activity (Figure 1). In a proof-of-concept trial, the platform has also demonstrated feasibility for monitoring and predicting agitation among residents of memory care facilities (11), suggesting that it may be possible to demonstrate the efficacy of treatments for agitation with sensors that monitor life-space activity.

Figure 1. Pervasive Computing and Continuous Monitoring Demonstrate Effects of COVID-19 Pandemic on Loneliness and Physical Activity

In two separate cohorts, digital data streams provided by the CART ORCATECH platform demonstrated changes in mood (through online self-reported questionnaires) and physical activity (through actigraphy) that coincided with the imposition of stay-at-home orders because of the pandemic.

 

RADAR-AD

The public private partnership Remote Assessment of Disease and Relapse – AD (RADAR-AD), funded by the Innovative Medicines Initiative (IMI) has embraced an “internet of things” (IOT) approach similar to the CART/ORCATECH system to measure early decline in cognitive function using low friction digital devices across a range of challenges. A study is planned that will enroll 200 participants over age 50 including those with preclinical AD, MCI due to AD, mild-to-moderate AD, and age and gender matched controls. RADAR earlier launched a multi-site prospective cohort study to determine the feasibility, usability, and acceptability of remote measurement technology (RMT) for major depressive disorder (MDD) (12).
Remote digital measurements that may be used in RADAR-AD include adaptations of multiple instruments to be administered using a table or smartphone interface such as the Amsterdam-instrumental activities of daily living questionnaire (A-IADL-Q) (13) to assess orientation, planning skills, and memory; and the Patient-Reported Outcomes Measurement Information Systems (PROMIS) general health measures to assess quality of life, disabilities, pain, and fatigue (14).
The RADAR base app, using a smartphone and passive sensors, will assess mobility, social communication, time at home, general activity, and sleep duration and quality. A residential movement detector will also assess gait and mobility at home and time spent in different areas of the home. The Mezurio smartphone app will capture high-frequency data related to cognition, including sleep, planning, memory, mood and anxiety, voice and speech, and typing and swiping (15).

Harvard Aging Brain Study (HABS)

In its search for sensitive measures of subtle cognitive decline in preclinical AD, The Harvard Aging Brain Study (HABS) developed an iPad-based Computerized Cognitive Composite (C3) and subsequently a smartphone-based tool for remote assessment of memory. The C3 on the iPAD comprises 3 components: a Behavioral Pattern Separation task, a Face-Name Associative Memory Exam, and the Cogstate Brief Battery. Development of the C3 included its use in the A4 study, a secondary prevention trial in clinically normal older adults with abnormal levels of amyloid who were randomized to receive placebo or solanezumab. This initial study showed that when the drug was administered in screening period, the amyloid-positive group performed worse on C3 compared to the Preclinical Alzheimer Cognitive Composite (PACC) (16, 17). The C3 has also been shown to provide reliable data when participants complete the assessments at home on an iPad (18). The remote use of C3 allows for more frequent assessments, which provide new information about subtle cognitive changes (19).
HABS has also developed the Boston Remote Assessment for Neurocognitive Health (BRANCH), which combines two paired associative learning tasks, an associated memory test, and a continuous visual recognition task accessible on a smartphone. In a recent study, BRANCH was shown to be correlated with scores on the paper-based preclinical Alzheimer’s cognitive composite (PACC5) (20).
HABS investigators have also been using BRANCH to assess learning curves using data from clinically normal older adults who completed 5 consecutive days of BRANCH. Early evidence suggests that a steeper learning curve is associated with better performance on the PACC, although the results were not statistically significant. Nonetheless, capturing learning curves using a smartphone over several days appears to be feasible and valid and may be useful to prescreen potential trial participants for biomarker status and identify those at greater risk for imminent cognitive decline and most likely to benefit from treatment. Cognitive learning curves may also serve as a paradigm for exploring psychopharmacokinetic profiles of medications to assess for efficacy over relatively short time periods.

The Trial-Ready Cohort for Preclinical and Prodromal Alzheimer’s Disease (TRC-PAD)

The Trial-Ready Cohort for Preclinical and Prodromal Alzheimer’s Disease (TRC-PAD) was established to accelerate drug development for AD by ensuring timely recruitment of targeted individuals into optimally designed preclinical and prodromal trials. TRC-PAD gathers multiple types of data including digital data from wearables and sensors through its informatics platform (TRC-PAD IP). In concert with TRC-PAD, the Alzheimer’s APT Webstudy registry was established along with an analytics platform, a referral management system, and a data management system (21).
The APT Webstudy has been working to optimize digital smartphone-based tools to assess cognition, such as the Cogstate Brief Battery. The study plans to implement this tool in the AHEAD 3-45 secondary prevention trials in early preclinical and preclinical AD and in the CT1812 trial in early AD (22). APT is also exploring alternative remote testing sensitive to amyloid elevation.

WHO – Integrated Care for Older People (ICOPE) Monitor

In its World Report on Healthy Ageing, the World Health Organization (WHO) redefined healthy aging in terms of functional ability, which is a product of an individual’s intrinsic capacity (IC) interacting with the environment (23). In order to implement integrated care for older people (ICOPE), an international partnership of researchers have developed a tool for assessing IC (24) as well as a mobile app available on smartphones that enable self-assessment by individuals to provide health care professionals with data on intrinsic capacity (25). Initial results from more than 5300 older individuals in the ICOPE Digital Cohort report high levels of IC impairments in multiple domains, including in vision, hearing, psychological health, and cognition. Recognizing that additional cognitive assessments are needed, ICOPE data will be combined with the Alzheimer’s Prevention Trial (APT) Webstudy data beginning in 2021. The ICOPE integrative approach from WHO provides the opportunity to detect memory complaints or decline, taking into consideration declines in other ICs, such as vision, hearing, mood, nutrition.

Brain Health Registry

With a goal of improving clinical trial recruitment, the Brain Health Registry (BHR) was established in 2014 as a website and online registry that collects digital data from potential clinical trial participants and study partners through self-report questionnaires and neuropsychological tests. Among the over 70,000 individuals enrolled in BHR, there have been over 106,000 referrals and 5,000 enrolled in other clinical studies. Through co-enrollment, BHR and clinical trial data can be linked.
BHR is also involved in methods development and validation of new online methods, including an electronic Clinical Dementia Rating (CDR) scale, a digital financial capacity instrument, and other digital assessment tools. In support of blood biomarkers studies, BHR has also collaborated with Quest Diagnostics to scale up blood collection using remote identification through the registry.

Other Studies Incorporating Digital Assessment Tools

Many other clinical research programs are exploring using digital measures of cognition and function (e.g., web- or mobile- device based cognitive testing, automated speech and language assessment, or use of sensor-embedded home-based device responsives). For example, the Framingham Heart Study (FHS) is developing a smartphone platform with multiple applications including digital voice. Voice (spoken utterances and language) is measured through device agnostic technologies and FHS research shows that linguistic and acoustic measures can be used as biomarkers of cognitive function (26–28). However, while speech and language assessments are likely to be very sensitive, additional work is needed to determine the best way to capture speech samples.
The Deep and Frequent Phenotyping (DFP) study in the United Kingdom has also incorporated digital measures from wearables and smartphones to capture highly granular data on gait, memory, navigational ability, and other measures that may serve as proxies for cognition and function (29). The study aims to identify an optimal set of markers for patient stratification.

 

Potential benefits and challenges associated with incorporating digital tools in trials

Digital devices offer the potential to expedite clinical trials, reduce the sample sizes or duration of studies, lower the cost of acquiring and analyzing data, enable the recruitment of more diverse and representative populations, reduce the need for specialized raters, reduce rater burden, and increase the clinical meaningfulness of outcome data collected to determine efficacy of an intervention. During a pandemic, digital tools also may provide the means for remote assessments, which could mitigate problems associated with closure of clinical trial sites and the safety concerns associated with in-person assessments and interventions.
Combining cognitive and functional endpoints in clinical trials is well established through the use of questionnaires in symptomatic stages of disease; and in preclinical disease, reports from both participants and study partners regarding high-level functional activities appear to provide valuable information on disease trajectory. Pairing digital cognitive measures with digital functional assessments may offer increased sensitivity in preclinical disease. The HABS team has shown that functional inventories can be administered remotely through the use of REDCap software, saving time and reducing data errors without sacrificing data quality.
The technologies employed in various digital devices provide additional potential advantages:
• They allow data to be time stamped, which enables assessment of everyday function.
• They can improve participant compliance by providing study participants and partners with reminders to take medications and engage in study-related activities.
• Audio and video recordings can provide a level of biometric validation to ensure that study participants themselves (rather than study partners) are providing data.
• Passive sensing of activities such as driving and computer use have demonstrated unique personalized patterns that can be captured using relatively simple algorithms.
• Passive sensors reduce participant burden, thus potentially reducing drop-outs and producing outcomes that are easily understandable and relevant in a real-world context.
• Meta-data such as time taken to complete a questionnaire or remember to take medication on time may also reflect changes in episodic memory, a key domain affected in AD.
• Passive assessment of sleep may prove useful since disrupted sleep has been shown to be a mediator of amyloid deposition in many studies.

To achieve the promise of digital devices will require addressing many challenges including those related to reliability, validation, user acceptability, regulatory approval, privacy and security. While regulators have indicated a willingness to consider digital data, much work remains to be completed before digital measures are accepted as primary or secondary outcomes. Indeed, the FDA issued guidance on the conduct of trials during a pandemic which includes increased flexibility around trial procedures intended to maintain participant safety, but less flexibility around efficacy measures. Task Force members also questioned whether regulators would accept remotely acquired data where there is no assurance that study participants and not study partners completed questionnaires or cognitive outcome assessments.
While digital registries may enable outreach to otherwise under-represented populations, engagement and retention of trial participants through digital means alone may be insufficient. Indeed, studies have reported high compliance rates that they attribute to personalized rather than anonymous digital engagement of study participants (30).
Transitioning cognitive assessments to smartphones may enable more frequent and sensitive measurement among large numbers of participants in a short period of time. However, smartphones may introduce additional challenges with regard to data fidelity and security. Other technological challenges related to digital data include the difficulty of analyzing continuous data and deconvoluting diverse data captured by devices. Addressing these challenges will be advanced by wider collaboration with the computer science and data science community.
Digital readouts are diverse and non-standardized, which results in impediments to data aggregation and sharing of data, both of which are essential to advance the field. Collaboration among technology developers is needed to ensure that raw data from which readouts on different devices are derived will be made available to the research community. Privacy concerns represent an additional roadblock to data sharing. Federated data management platforms have been developed that provide researchers with access to multiple types of data from multiple sources (31). The Alzheimer’s Disease Data Initiative has also established the AD Workbench (ADWB) platform to promote data, information, and tool sharing across the AD research community (32), although no digital biomarker data are yet incorporated in this relatively new effort.

 

Moving Forward

Despite the challenges of conducting clinical studies during the pandemic, the AD clinical research community remains committed to advancing current studies and launching new studies. Indeed, Task Force participants suggested that the COVID-19 experience may fuel preparations for better clinical trials that mitigate both the risks associated with the pandemic and more common complications of clinical trials for example by broadening the reach of recruitment efforts.
Among the possibilities discussed by the Task Force was the potential to design completely digital or hybrid trials combining virtual/telemedicine and physical visits. For example, telemedicine consultations in the pre-screening phase of a trial could help verify inclusion and exclusion criteria, provide information to potential trial participants, and obtain electronic consent. Virtual visits could also be used for trial components that do not require technically invasive procedures. The INSPIRE study team at Toulouse University Hospital tested this strategy following the first COVID lockdown in France and found a high rate of acceptance among participants (33).
Digital devices may provide new analytical methods as well, for example, by enabling assessment of intraindividual change in a high frequency way. Digital tools implemented during the pandemic may also provide increased understanding to the effects of daily stress, anxiety, and poor sleep on cognitive outcomes. They may also enable continued assessment of cognitive function in participants post-infection with COVID-19.
To ensure that clinical studies continue through the pandemic and its aftermath, the Task Force recommended that regulators and institutional review boards provide guidance yet maintain flexibility with regard to changes in applications/devices and statistical plans and allow for virtual and home-based visits and remote monitoring. Developers must consider alternate or additional trial sites, and funders should expect unanticipated modifications to protocols and funding needs.
The Task Force concluded its session with a discussion of the field’s readiness to conduct remote clinical trials. Task Force members urged caution in employing remote assessments for trials of experimental medications; the digital measures at this time may be most productively embedded in parallel with conventional measures in early-stage Phase 1 and 2 trials, demonstrating their fidelity or even superiority to current commonly used clinical outcome measures. Enthusiasm was expressed for using digital technologies in remote studies of non-invasive interventions such as studies of vitamins, exercise, and mind-body or multidomain approaches. Trials of repurposed drugs for which safety data exists may also be feasible as entirely remote trials. For example, because of the COVID-19 pandemic, the Alzheimer’s Disease Cooperative Study (ADCS) is attempting to conduct the PEACE-AD (prazosin for disruptive agitation in Alzheimer’s disease) trial (34) as a remote-only trial.

 

Acknowledgements: The authors thank Lisa J. Bain for assistance in the preparation of this manuscript.

Conflicts of interest: The Task Force was partially funded by registration fees from industrial participants. These corporations placed no restrictions on this work. Dr. Kaye reports grants from Merck, Eisai, Genentech, and Abbvie; and serves on a Data Safety Monitoring Committee at Eli Lilly. Dr. Aisen reports grants from Janssen, Eli Lilly, Eisai, NIA, the Alzheimer’s Association, and FNIH; and consulting fees from Biogen, Roche, Merck, Abbvie, Lundbeck, Proclara, and Immunobrain Checkpoint. Dr. Amariglio declares there are no conflicts. Dr. Au is on the scientific advisory board of Signant Health and a consultant to Biogen. Dr. Ballard reports grants from Arcadia pharmaceutical company and Lundbeck; personal fees from Arcadia, Lundbeck, Roche, Otsuka, Novartis, Eli Lilly, Suven, Sunovion, ADDEX, and Exciva; personal fees and other from Synexus, Novo Nordisk; and consulting fees from Biogen. Dr. Carillo declares there are no conflicts. Dr. Fillit declares there are no conflicts. Dr. Iwatsubo declares there are no conflicts. Dr. Jimenez-Maggiora declares there are no conflicts. Dr. Lovestone is an employee of Janssen Medical Ltd and a co-founder of Akrivia Health Ltd . Dr. Natanegara is an employee of Eli Lilly and Company. Dr. Papp declares there are no conflicts. Dr. Soto declares payment as consultant or advisor from Avanir, Acadia. Dr. Weiner reports grants from Siemens, Biogen, and Johnson & Johnson, NIH, California Department of Health, University of Michigan, Hillblom Foundation, Alzheimer’s Association, State of California, Kevin and Connie Shanahan, GE, VUMc, American Catholic University, The Stroke Foundation, and the Veteran’s Administration; has served on advisory boards for Eli Lilly, Cerecin/Accera, Roche, Alzheon Inc., Merck Sharp & Dohme Corp, Nestle/Nestec; and has consulted with or acted as a speaker to Cerecin/Accera Inc., BioClinica, Nestle/Nestec, Genentech, FUJIFILM-Toyama Chemical, and T3D Therapeutics; and holds stock options with Alzheon Inc, Alzeca, and Anven. Dr. Vellas reports grants from Lilly, Merck, Roche, Lundbeck, Biogen, grants from Alzheimer’s Association, European Commission, personal fees from Lilly, Merck, Roche, Biogen, outside the submitted work.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

 

References

1. Morris MC, Tangney CC, Wang Y, et al. MIND diet slows cognitive decline 1. van Dorn A. COVID-19 and readjusting clinical trials. The Lancet. 2020 Aug 22;396(10250):523–4.
2. Fleming TR, Labriola D, Wittes J. Conducting Clinical Research During the COVID-19 Pandemic: Protecting Scientific Integrity. JAMA. 2020 Jul 7;324(1):33–4.
3. Sathian B, Asim M, Banerjee I, Pizarro AB, Roy B, van Teijlingen ER, et al. Impact of COVID-19 on clinical trials and clinical research: A systematic review. Nepal J Epidemiol. 2020 Sep;10(3):878–87.
4. Brown EE, Kumar S, Rajji TK, Pollock BG, Mulsant BH. Anticipating and Mitigating the Impact of the COVID-19 Pandemic on Alzheimer’s Disease and Related Dementias. Am J Geriatr Psychiatry. 2020 Jul;28(7):712–21.
5. Food and Drug Administration. Conduct of Clinical Trials of Medical Products During the COVID-19 Public Health Emergency Guidance for Industry, Investigators, and Institutional Review Boards. U.S. Department of Health and Human Services,; 2020.
6. Gold M, Amatniek J, Carrillo MC, Cedarbaum JM, Hendrix JA, Miller BB, et al. Digital technologies as biomarkers, clinical outcomes assessment, and recruitment tools in Alzheimer’s disease clinical trials. Alzheimers Dement (N Y). 2018;4:234–42.
7. World Health Organization. Dementia [Internet]. 2020 [cited 2021 Jan 6]. Available from: https://www.who.int/news-room/fact-sheets/detail/dementia
8. Statista. Number of smarphone users worldwide from 2016 to 2021 [Internet]. Statista. 2020 [cited 2021 Jan 26]. Available from: https://www.statista.com/statistics/330695/number-of-smartphone-users-worldwide/
9. Lindauer A, Seelye A, Lyons B, Dodge HH, Mattek N, Mincks K, et al. Dementia Care Comes Home: Patient and Caregiver Assessment via Telemedicine. Gerontologist. 2017 Oct 1;57(5):e85–93.
10. Beattie Z, Miller LM, Almirola C, Au-Yeung W-TM, Bernard H, Cosgrove KE, et al. The Collaborative Aging Research Using Technology Initiative: An Open, Sharable, Technology-Agnostic Platform for the Research Community. Digit Biomark. 2020;4(Suppl 1):100–18.
11. Au-Yeung W-TM, Miller L, Beattie Z, Dodge HH, Reynolds C, Vahia I, et al. Sensing a problem: Proof of concept for characterizing and predicting agitation. Alzheimers Dement (N Y). 2020;6(1):e12079.
12. Matcham F, Barattieri di San Pietro C, Bulgari V, de Girolamo G, Dobson R, Eriksson H, et al. Remote assessment of disease and relapse in major depressive disorder (RADAR-MDD): a multi-centre prospective cohort study protocol. BMC Psychiatry. 2019 Feb 18;19(1):72.
13. Stringer G, Leroi I, Sikkes SAM, Montaldi D, Brown LJE. Enhancing ‘meaningfulness’ of functional assessments: UK adaptation of the Amsterdam IADL questionnaire. International Psychogeriatrics. undefined/ed;1–12.
14. Tatsuoka C, DeMarco L, Smyth KA, Wilkes S, Howland M, Lerner AJ, et al. Evaluating PROMIS Physical Function Measures in Older Adults at Risk for Alzheimer’s Disease. Gerontol Geriatr Med [Internet]. 2016 Sep 5 [cited 2021 Jan 8];2. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5590694/
15. Lancaster C, Koychev I, Blane J, Chinner A, Wolters L, Hinds C. Evaluating the Feasibility of Frequent Cognitive Assessment Using the Mezurio Smartphone App: Observational and Interview Study in Adults With Elevated Dementia Risk. JMIR Mhealth Uhealth [Internet]. 2020 Apr 2 [cited 2021 Jan 8];8(4). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7163418/
16. Papp KV, Rentz DM, Maruff P, Sun C-K, Raman R, Donohue MC, et al. The Computerized Cognitive Composite (C3) in A4, an Alzheimer’s Disease Secondary Prevention Trial. J Prev Alzheimers Dis. 2021;8(1):59–67.
17. Papp KV, Rentz DM, Orlovsky I, Sperling RA, Mormino EC. Optimizing the preclinical Alzheimer’s cognitive composite with semantic processing: The PACC5. Alzheimers Dement (N Y). 2017 Nov;3(4):668–77.
18. Rentz D, Dekhtyar D, Sherman J, Burnham S, Blacker D, Aghjayan S, et al. The Feasibility of At-Home iPad Cognitive Testing For Use in Clinical Trials. 2015 [cited 2021 Jan 8]; Available from: http://www.jpreventionalzheimer.com/all-issues.html?article=148
19. Samaroo A, Amariglio RE, Burnham S, Sparks P, Properzi M, Schultz AP, et al. Diminished Learning Over Repeated Exposures (LORE) in preclinical Alzheimer’s disease. Alzheimers Dement (Amst). 2020;12(1):e12132.
20. Papp KV, Samaroo AH, Chou H-CL, Buckley RF, Rentz D, Sperling RA, et al. Repeated memory-based assessments: Implications for clinical trials and practice. Alzheimer’s & Dementia. 2020;16(S6):e038143.
21. Jimenez-Maggiora GA, Bruschi S, Raman R, Langford O, Donohue M, Rafii MS, et al. TRC-PAD: Accelerating Recruitment of AD Clinical Trials through Innovative Information Technology. J Prev Alzheimers Dis. 2020;7(4):226–33.
22. Grundman M, Morgan R, Lickliter JD, Schneider LS, DeKosky S, Izzo NJ, et al. A phase 1 clinical trial of the sigma-2 receptor complex allosteric antagonist CT1812, a novel therapeutic candidate for Alzheimer’s disease. Alzheimers Dement (N Y). 2019 Jan 23;5:20–6.
23. Beard JR, Officer A, de Carvalho IA, Sadana R, Pot AM, Michel J-P, et al. The World report on ageing and health: a policy framework for healthy ageing. Lancet. 2016 May 21;387(10033):2145–54.
24. Takeda C, Guyonnet S, Sumi Y, Vellas B, Araujo de Carvalho I. Integrated Care for Older People and the Implementation in the INSPIRE Care Cohort. J Prev Alzheimers Dis. 2020;7(2):70–4.
25. Sanchez-Rodriguez D, Annweiler C, Gillain S, Vellas B. Implementation of the Integrated Care of Older People (ICOPE) App in Primary Care: New Technologies in Geriatric Care During Quarantine of COVID-19 and Beyond. J Frailty Aging. 2020 May 6;1–2.
26. Thomas JA, Burkhardt HA, Chaudhry S, Ngo AD, Sharma S, Zhang L, et al. Assessing the Utility of Language and Voice Biomarkers to Predict Cognitive Impairment in the Framingham Heart Study Cognitive Aging Cohort Data. J Alzheimers Dis. 2020;76(3):905–22.
27. Thomas NWD, Beattie Z, Marcoe J, Wright K, Sharma N, Mattek N, et al. An Ecologically Valid, Longitudinal, and Unbiased Assessment of Treatment Efficacy in Alzheimer Disease (the EVALUATE-AD Trial): Proof-of-Concept Study. JMIR Res Protoc. 2020 May 27;9(5):e17603.
28. Lin H, Karjadi C, Ang TFA, Prajakta J, McManus C, Alhanai TW, et al. Identification of digital voice biomarkers for cognitive health. Exploration of Medicine. 2020 Dec 31;1(6):406–17.
29. Koychev I, Lawson J, Chessell T, Mackay C, Gunn R, Sahakian B, et al. Deep and Frequent Phenotyping study protocol: an observational study in prodromal Alzheimer’s disease. BMJ Open. 2019 Mar 23;9(3):e024498.
30. Sano M, Egelko S, Zhu CW, Li C, Donohue MC, Ferris S, et al. Participant satisfaction with dementia prevention research: Results from Home-Based Assessment trial. Alzheimers Dement. 2018;14(11):1397–405.
31. Toga AW, Bhatt P, Ashish N. Global data sharing in Alzheimer’s disease research. Alzheimer Dis Assoc Disord. 2016;30(2):160–8.
32. Alzheimer’s Disease Data Initiative. AD Workbench [Internet]. ADDI. [cited 2021 Jan 11]. Available from: https://www.alzheimersdata.org/ad-workbench
33. Takeda C, Guyonnet S, Ousset PJ, Soto M, Vellas B. Toulouse Alzheimer’s Clinical Research Center Recovery after the COVID-19 Crisis: Telemedicine an Innovative Solution for Clinical Research during the Coronavirus Pandemic. J Prev Alzheimers Dis. 2020;7(4):301–4.
34. Alzheimer Disease Cooperative Study. PEACE-AD [Internet]. ADCS PEACE-AD. [cited 2021 Jan 11]. Available from: https://www.adcs.org/peace-ad/

 

A JAPANESE MULTICENTER STUDY ON PET AND OTHER BIOMARKERS FOR SUBJECTS WITH POTENTIAL PRECLINICAL AND PRODROMAL ALZHEIMER’S DISEASE

 

M. Senda1, K. Ishii2, K. Ito3, T. Ikeuchi4, H. Matsuda5, T. Iwatsubo6, A. Iwata7, R. Ihara7, K. Suzuki8, K. Kasuga4, Y. Ikari1,9, Y. Niimi6, H. Arai10, A. Tamaoka11, Y. Arahata3, Y. Itoh12, H. Tachibana13, Y. Ichimiya14, S. Washizuka15, T. Odawara16, K. Ishii17, K. Ono18, T. Yokota19, A. Nakanishi20, E. Matsubara21, H. Mori12, H. Shimada12

 

1. Kobe City Medical Center General Hospital, Japan; 2. Tokyo Metropolitan Institute of Gerontology, Japan; 3. National Center for Geriatrics and Gerontology, Japan; 4. Niigata University, Japan; 5. National Center of Neurology and Psychiatry, Japan; (currently, Southern Tohoku Drug Development and Cyclotron Research Center, Japan); 6. The University of Tokyo, Japan; 7. The University of Tokyo, Japan; (currently, Tokyo Metropolitan Geriatric Hospital, Japan); 8. The University of Tokyo, Japan; (currently, National Defense Medical College, Japan); 9. Osaka University, Japan; 10. Tohoku University, Japan; 11. University of Tsukuba, Japan; 12. Osaka City University, Japan; 13. Kobe University, Japan; 14. Juntendo Tokyo Koto Geriatric Medical Center, Japan; 15. Shinshu University, Japan; 16. Yokohama City University, Japan; 17. Kindai University, Japan; 18. Showa University, Japan; 19. Tokyo Medical and Dental University, Japan; 20. Osaka City Kosaiin Hospital, Japan; 21. Oita University, Japan

Corresponding Author: Michio Senda, Division of Molecular Imaging Research Kobe City Medical Center General Hospital (KCGH), 2-1-1 Minatojima-Minamimachi, Chuo-ku, Kobe 650-0047 Japan, E-mail: michio_senda@kcho.jp, Phone: 81-78-304-5212, Fax: 81-78-304-5201.

J Prev Alz Dis 2021;4(8):495-502
Published online June 30, 2021, http://dx.doi.org/10.14283/jpad.2021.37

 


Abstract

BACKGROUND: PET (positron emission tomography) and CSF (cerebrospinal fluid) provide the “ATN” (Amyloid, Tau, Neurodegeneration) classification and play an essential role in early and differential diagnosis of Alzheimer’s disease (AD).
OBJECTIVE: Biomarkers were evaluated in a Japanese multicenter study on cognitively unimpaired subjects (CU) and early (E) and late (L) mild cognitive impairment (MCI) patients.
MEASUREMENTS: A total of 38 (26 CU, 7 EMCI, 5 LMCI) subjects with the age of 65-84 were enrolled. Amyloid-PET and FDG-PET as well as structural MRI were acquired on all of them, with an additional tau-PET with 18F-flortaucipir on 15 and CSF measurement of Aβ1-42, P-tau, and T-tau on 18 subjects. Positivity of amyloid and tau was determined based on the positive result of either PET or CSF.
RESULTS: The amyloid positivity was 13/38, with discordance between PET and CSF in 6/18. Cortical tau deposition quantified with PET was significantly correlated with CSF P-tau, in spite of discordance in the binary positivity between visual PET interpretation and CSF P-tau in 5/8 (PET-/CSF+). Tau was positive in 7/9 amyloid positive and 8/16 amyloid negative subjects who underwent tau measurement, respectively. Overall, a large number of subjects presented quantitative measures and/or visual read that are close to the borderline of binary positivity, which caused, at least partly, the discordance between PET and CSF in amyloid and/or tau. Nine subjects presented either tau or FDG-PET positive while amyloid was negative, suggesting the possibility of non-AD disorders.
CONCLUSION: Positivity rate of amyloid and tau, together with their relationship, was consistent with previous reports. Multicenter study on subjects with very mild or no cognitive impairment may need refining the positivity criteria and cutoff level as well as strict quality control of the measurements.

Key words: Alzheimer’s disease, PET, CSF biomarker, amyloid, tau.


 

 

Introduction

Early and differential diagnosis of Alzheimer’s disease (AD) has been drawing more and more attention these days as the target population of the therapeutic trials has shifted toward the early phases of the AD continuum. Biomarkers including PET, MRI and cerebrospinal fluid(CSF)/plasma play an essential role in such early phases, where clinical manifestation and behavioral findings are limited. Jack et al (1) extracted three markers, i.e., amyloid (A), tau (T) and neurodegeneration (N), and proposed the “ATN” classification for differential diagnosis of AD continuum. PET provides imaging and quantification of amyloid and tau deposition as well as neurodegeneration evaluable with 18F-fluorodeoxyglucose(FDG)-PET. Amyloid and tau can also be evaluated with CSF sampling, and recently with plasma as well, and MRI volumetry has also been used as a marker of neurodegeneration.
In Japan, a large-scale prospective observational study called J-ADNI (Japanese Alzheimer’s Disease Neuroimaging Initiative) was completed (2), in which a total of 537 subjects were enrolled, comprising 154 cognitively unimpaired subjects (CU), 234 MCI and 149 AD patients.
Then, a new version of J-ADNI was designed by the same group, named “AMED Preclinical AD Study”, which focused on CU and MCI and acquired amyloid-PET and FDG-PET on all subjects. Part of the subjects also underwent a tau-PET scan and/or a CSF sampling. The objective of the study was to evaluate PET and MRI images and CSF biomarkers in CU and MCI subjects in Japan, compare those biomarkers between modalities to explore their reliability and usefulness in such early-phase subjects, and obtain a rough idea of the fractions of ATN-based classifications. This report summarizes the results of the study that was recently completed.

 

Methods

Subjects

The study was a non-randomized prospective observational study, and was designed and conducted in accordance with the ethical principles as proclaimed in the Declaration of Helsinki. The study protocol was first approved by Ethical Committee of Osaka City University Graduate School of Medicine (site of leading PI) and registered as UMIN000019926, and was later re-approved by Osaka City University Hospital Certified Review Board when the Japanese Law on Clinical Research was enacted, and was registered as jRCTs051180239. The protocol was also approved by each participating site according to the Japanese regulations and ethics guidance. The tau-PET portion of the study was designed as a nominally separate add-on study when tau-PET became available, though limited, later in the course of the research project, and was approved and registered as jRCTs051190065.
The subjects were enrolled at a total of 14 clinical sites and consisted of 26 CU subjects and 12 MCI patients (7 early (E) MCI and 5 late (L) MCI as classified below) based on the neuropsychological tests.
Written informed consent was obtained from each subject and the study partner such as a family member of the subject.
The CU subjects were 65-82 years old without any memory problem and CDR-J=0. The MMSE-J score was 25-30 (higher than or equal to 24), and the delayed recall score of WMS-R logical memory (WMS-R LM II) ranged from 4 to 19 except for one subject (See footnote of Table 1).
The MCI subjects were 65-84 years old with objective persistent memory impairment reported by the study partner and CDR-J=0.5 with memory box score being 0.5 or higher. The MMSE-J score was 27-30 (higher than or equal to 24). The WMS-R LM II was used to classify the MCI subjects into EMCI (WMS-R LM II = 3-6, 5-9, 9-11) and LMCI (≤2, ≤4, ≤8) depending on the educational years (0-7, 8-15, ≥16 years, respectively), and was 6-18 for EMCI and 0-8 for LMCI.

CSF measurements and genotyping

CSF was collected from 18 subjects by lumbar puncture and stored in polypropylene tubes at -80℃ until biochemical analysis. CSF concentration of Aβ1-42 was analyzed using V-PLEX Aβ Peptide Panel 1 kit with MESO QuickPlex SQ120 (MesoScale Discovery, Rockville, MD). CSF phosphorylated tau (P-tau) and total tau (T-tau) were measured using commercially available ELISA kits, INNOTEST hTAU and PHOSPHO-TAU (181P) (Fujirebio Europe, Belgium), respectively, according to the manufacturer’s instructions. Stability of the results was monitored in the Alzheimer’s Association QC program. Cutoff values (Aβ42<378.7 pg/mL, P-tau>29.1 pg/mL, and T-tau>88.8 pg/mL) that best discriminated PiB-PET positive AD patients from PiB-PET negative CU subjects were determined using independent J-ADNI cohort (2). Because the CSF assays used in this study were different from those used in J-ADNI study, calibration between two assays were performed.
APOE genotyping (rs429358 and rs7412) was performed by Taq-Man based assay using blood samples.

MRI imaging

The brain MRI was acquired for each subject using a 3-Tesla or 1.5-Tesla scanner. The structural 3D-T1 images (MP-RAGE or IR-SPGR) were analyzed with FreeSurfer (Ver. 6.0) to measure the regional cerebral gray-matter volumes. Because the absolute volumetry depends on the version of the software and other conditions, the regional atrophy of the subject was derived as z-score using mean and SD of the baseline scan for the 26 CU subjects of this study. The volume of 8 regions in the temporal lobe (right and left entorhinal cortex, parahippocampal gyrus, hippocampus, and amygdala) were summed up and the z score was derived as a measure of the temporal lobe atrophy for each subject.

PET image acquisition

All subjects underwent an amyloid-PET and an FDG-PET. Each PET imaging site, together with the PET camera, was qualified, in which the reconstruction parameters were determined for each PET camera so that all the PET cameras satisfied the image quality criteria with the Hoffman 3D brain phantom and the uniform cylindrical phantom (3).
For amyloid PET, either 11C-PiB (PiB), 18F-florbetapir (FBP) or 18F-flutemetamol (FMM) was used for 23, 13, and 2 subjects, respectively. The injection activity was 555MBq, 370MBq, 185MBq, the uptake time (start of emission scan post injection) was 50min, 50min, 90min, and the scan duration was 20min, 20min, 30min, for PiB, FBP, and FMM, respectively.
For the FDG scan, after at least 4 hours of fasting, the subject was administered with 185 MBq of 18F-FDG in a quiet, dimly lit room while resting in a reclining chair or bed, and the subject remained in the condition until several minutes before the start of the scanning session. The PET emission data was acquired for 30 minutes starting at 30 minutes post injection.
Tau-PET was performed with 18F-flortaucipir (FTP) on 15 subjects. Because tau-PET was not ready until late in the course of the research project, the time span from amyloid to tau-PET ranged from 1.0 to 2.0 (mean 1.56) years. The subject was administered with 240.5 MBq of FTP and a 30 min emission scan started 75 min post injection.
No adverse effects were observed at the PET scans of this study.

PET image analysis

The amyloid PET images were binary interpreted visually in a blind manner by the readers who were qualified for this study, and the adjudicator (K.I.) confirmed them. The PiB images were interpreted visually using the criteria adopted in J-ADNI (4), and the FMM and FBP images were interpreted with each vendor’s criteria.
As a quantitative analysis of the amyloid PET, mean cortical standardized uptake value ratio (mcSUVR) of PiB images was computed using the cerebellar cortex as a reference based on the method of J-ADNI, and the cutoff value of 1.5 was used to determine the quantitative positivity (4). The FMM images were analyzed with CORTEX ID (GE Healthcare) to derive mcSUVR using the pons as a reference, for which the cutoff value of 0.58 was used for the quantitative positivity (5). The FBP images were analyzed with MIMneuro (MIM Software) to derive mcSUVR using the whole cerebellum as a reference, for which the cutoff value of 1.10 was used for the quantitative positivity (6).
The FDG images together with the semiquantitative 3D-SSP results were visually interpreted by three independent readers followed by a consensus read in the same way as J-ADNI (7), and the images were classified into N1 (normal), N2 (reflecting atrophy), N3, P1 (AD pattern), P2 (FTD pattern), P3, and P1+ (DLB pattern) (8). No one presented N3 or P3 in this study. The DLB pattern criteria was interpreted in a broader sense to include cases with occipital hypometabolism extending to neighboring areas even if typical temporoparietal hypometabolism was not observed. The FDG images were also quantified with AD t-sum (9) using the module PALZ in the PMOD software package (Ver. 3.2; PMOD Technologies, Zurich, Switzerland), which were then converted into PET score [10] that reflects the severity of temporoparietal hypometabolism (AD pattern).
The FTP images were interpreted and classified into AD negative, AD+ and AD++, according to the vendor’s criteria that regards cortical uptake except anterior temporal as AD-related (https://pi.lilly.com/us/tauvid-uspi.pdf). The FTP-PET was also analyzed with MUBADA-PERSI method to derive SUVR over the area affected by AD process (posterior temporal, occipital, parietal and part of frontal cortex) with white matter as a reference (11, 12).

Follow up

Whenever possible, each subject was followed up every year with a general clinical interview with neuropsychological tests, an MRI scan, and an FDG-PET scan. As a result, 1-year follow-up data were acquired on 32 subjects, and 2-year follow-up on 5 subjects.

Statistical methods

Because the number of subjects was small, descriptive results were presented in general. Proportion of positivity was compared between groups using chi-square tests, in which EMCI and LMCI were combined to increase the number of observations. Statistical tests were also performed on the Pearson correlation coefficient between two variables.

 

Results

Findings of each subject

Table 1 describes findings of each subject as classified according to the ATN concept. In this study, amyloid (A) was interpreted as positive (A+) when either PET or CSF Aβ was positive. Tau (T) was interpreted as positive (T+) when either PET or CSF P-tau was positive; negative (T-) when either of them was obtained and neither of them were positive; and was “na” (not available) (Tna) when neither of them were obtained. Neurodegeneration (N) was interpreted as positive (N+) when the consensus visual read of FDG-PET showed a progressive pattern (P1, P2, or P1+), and negative (N-) when it was a non-progressive pattern (N1 or N2).

Table 1. Findings for each subject and ATN classification

 

The amyloid positivity rate was 13/38 overall (6/26 CU, 4/7 EMCI, 3/5 LMCI, p>0.05 between CU and MCI), while it was 8/38 based on the PET alone (3/26 CU, 2/7 EMCI, 3/5 LMCI, p>0.05 between CU and MCI).
Tau was positive for 7, negative for 2 and not available for 4 out of the 13 A+ subjects, being 2, 2, 2 and 5, 0, 2 out of the 6 A+ CU and 7 A+ MCI subjects, respectively.
FDG-PET showed a progressive pattern in 6/13 A+ subjects (3/6 CU, 1/4 EMCI, 2/3 LMCI) as compared to 3/25 A- subjects (0/20 CU, 1/3 EMCI, 2/2 LMCI). Significant difference was observed in the FDG-PET positivity (N+) proportion between A+ and A- (p<0.05) as well as between CU (3/26) and MCI (6/12) (p<0.05).
Of interest, tau was positive for as many as 8 (negative for 8, not available for 9) out of the 25 amyloid negative subjects, indicating tau deposition without AD pathological process. It should be noted that all the 8 A-T+ subjects was tau positive due to CSF test, in spite of negative tau PET for two of them.
Association of APOE genotypes with amyloid PET (p>0.6) or CSF Aβ (p>0.5), or with any other biomarkers, was not observed for the presence of E4, probably due to the small number of subjects.

Representative cases

Figures 1 (#24, LMCI) and 2 (#22, CU) depict a case with prodromal AD (A+T+N+) and preclinical AD (A+T+N-), respectively. PET and CSF were discordant for “A” and/or “T” in both cases, which may be related to visually equivocal images and near-cutoff level quantified values. In the case of Figure 2, CSF P-tau was positive while tau PET was negative, consistent with the report of earlier and more sensitive positivity of CSF P-tau than tau-PET in the AD continuum (13).
Four cases (1 CU, 1 EMCI, 2 LMCIs) showed a mild/partial DLB pattern in FDG-PET marked with “P1+” in Table 1, featuring hypometabolism in the occipital cortex extending into surrounding areas but not showing a typical AD pattern of temporo-parietal hypometabolism. Amyloid was positive for 3/4 and tau was positive for 4/4. Figure 3 (#26, CU) depicts one of them.

Figure 1. PiB, FTP and FDG-PET of a female LMCI patient in her 70s (#24) interpreted as prodromal AD

Amyloid PET with PiB was visually negative, as the left parietal mild accumulation did not reach the cortical surface (arrow). However, the subject was classified as “A+” because quantitative analysis revealed SUVR (1.57) above cutoff. The CSF Aβ was negative (399.8 pg/mL). The FTP-PET showed abnormal tau accumulation in the left posterior temporal lobe (arrow), typical of AD process. Note off-target uptake of FTP in choroid plexus (arrowheads), substantia nigra, and striatum. The FDG-PET was read as temporo-parietal hypometabolism indicating AD pattern in the baseline that progressed in two years (arrows). PET score and MRI z-score also increased in two years: from 0.76 to 1.08 and from 2.7 to 3.2, respectively.

Figure 2. PiB, FTP and FDG-PET of a female CU subject of her 70s (#22) interpreted as preclinical AD

PiB-PET revealed positive amyloid accumulation in the left temporal and parietal areas (arrows). Tau PET with FTP acquired 1 year later was negative, because mild activity along the cortical rim was interpreted as off-target uptake by the meninges (short arrows) and that the left anterior temporal uptake was considered non-pathological within the AD continuum (long arrow). CSF P-tau was positive. FDG-PET showed a normal pattern.

Figure 3. FBP, FTP and FDG-PET of a female CU subject of her 70s (#26)

FBP-PET presented negative amyloid, and tau was negative in FTP-PET, although CSF showed positive Aβ (317.9pg/mL) and P-tau (38.2pg/mL). FDG-PET revealed a DLB pattern, presenting occipital hypometabolism (long arrows) extending to the right temporal and parietal cortex (short arrows), which progressed 1 year later. Note cingulate island sign denoting preserved metabolism in the posterior cingulate cortex (arrowheads).

 

Association between PET and CSF

For the 18 subjects, in which CSF data were obtained, amyloid positivity by CSF agreed with that by PET in 12 cases while 6 showed a discordance (Table 1). The rate of discordance was consistent with previous reports and may be caused by various factors (13).
Quantified tau uptake (SUVR) measured with FTP-PET using MUBADA-PERSI method was significantly correlated with CSF P-tau (r=0.92, p<0.001, n=8) (Figure 4). Although the cutoff for SUVR with MUBADA-PERSI SUVR is not established yet, the visual read of the FTP-PET was positive only for two of them. In CSF P-tau, however, 7 out of the 8 subjects showed P-tau above the cutoff level, indicating a discordance in the tau positivity between PET and CSF. This is consistent with recent investigations that reported earlier or more sensitive positivity of CSF P-tau than tau-PET in the AD continuum (i.e. in amyloid positive subjects), because secretion of soluble p-tau to CSF is increased by Aβ pathology before tau begins to accumulate in the brain (14).

Figure 4. Scatter plot of tau uptake (SUVR) quantified with FTP-PET and CSF P-tau

Red marks indicate PET-positive cases by visual read. Arrow indicates cutoff value for CSF P-tau.

 

Discussion

PET/CSF discordance for amyloid and tau

This study suffers limitations such as the small number of subjects, poor follow-up records, and lack of tau-PET and CSF measurement for a large fraction of the subjects. However, some findings are notable.
The rate of amyloid positivity based on the combination of PET and CSF (6/26=23% for CU, 7/12=58% for MCI) was consistent with previous reports including J-ADNI. Discordance of positivity between amyloid-PET and CSF Aβ was observed in 6 subjects (5 PET-/CSF+, 1 PET+/CSF-), suggesting higher sensitivity of CSF, which was also consistent with the ADNI data on CU and MCI (13).
The rate of tau positivity was 4/6 for amyloid PET-/CSF Aβ+ or amyloid PET+/CSF Aβ-, and 2/2 for amyloid PET+/CSF Aβ+ in this study (Table 1). This was agreeable with the above ADNI data, in which the former two groups presented significantly lower CSF p-tau and PET-measured tau deposition than the latter and suggested earlier manifestations of AD process (13).
It is known that CSF p-tau is quantitatively associated with PET-measured tau deposition, especially in the AD continuum, and that CSF p-tau rises in the earlier stage than the pathological uptake of FTP-PET accrues (14). The result of the present study indicated a similar association in spite of the small number of subjects (Figure 4), in which FTP-PET was quantified with MUBADA, and that CSF p-tau was more sensitive than visual read of FTP-PET. In the early phase of AD continuum, CSF p-tau and FTP-PET could be considered to reflect different pathological changes, as the former may indicate excess secretion of p-tau to CSF and the latter represents tissue tau deposition. The present study adopted the criteria of visual FTP-PET interpretation to determine the positivity, by which the anterior temporal FTP uptake was considered insignificant. Because MUBADA VOI covers wide cortical areas, it may not be a sensitive measure of early tau deposition in the AD continuum that begins in the temporal cortex. Although a recent study suggests the earliest tau deposition at the rhinal cortex located in the anteromedial temporal lobe (15), it was hard to quantify the pathological FTP uptake in that region due to off-target binding to the choroid plexus in the present study (Figure 2). In that sense, use of FTP was another limitation of this study. Tau PET drugs with little off-target binding such as 18F-MK-6240 or 18F-PI-2620 may be more suitable for evaluation of the earliest stage of AD.

Neurodegeneration marker

In this study only visual assessment of FDG-PET was used to evaluate the “N” (neurodegeneration) marker, and quantitative PET score was not used so that one “N-” subject presented a high PET score (#4). Although CSF T-tau and MRI-volumetry are also regarded as an N-marker, their association with FDG-PET remains to be investigated as they represent different pathophysiology.
Hypometabolism depicted by FDG-PET reflects reduced neuronal activity in general, regardless of pathophysiology. No subjects showed A+T-N+ in this study, which agrees with the concept of tau deposition leading to neurodegeneration in AD continuum, although such manifestation, if occurred, might have suggested combined AD and non-AD processes. Outside the AD continuum (A-), however, FDG-PET neurodegeneration was positive in 3/5 MCI (2/2 LMCI) subjects and was not observed in CU subjects (0/20), which is consistent with the above notion and agrees with previous reports (16).

Binary criteria

A large number of subjects presented quantitative measures and/or visual read that are close to the borderline of binary positivity in the present study, which caused, at least partly, discordance between PET and CSF in amyloid and/or tau, such as the cases in Figures 1 and 2. This is understandable because most of the subjects in the present study were in the early phase of AD continuum or of a non-AD disease if any, and that the current criteria and cutoff level have been derived from differential diagnosis of AD patients from CU subjects. To deal with early-phase subjects having no or very mild cognitive impairment, the criteria and cutoff level of the biomarkers may need refining, and the data acquisition may need strict quality control.

Non-AD disorders

The present study revealed 8 A-T+ subjects. Because two of them (#37, #38) were LMCI patients and showed AD or DLB pattern in FDG-PET, they are considered SNAP and to have cognitive impairment due to non-AD disorders (17). The other six are CU subjects and had non-progressive FDG pattern, and may suggest a very early stage of various non-AD tauopathy such as PART or normal aging process (17). There is also a possibility of false-positive CSF P-tau as the value was close to the cutoff, ranging 31.8-38.2 pg/mL for 5 of the 6 subjects. Since biomarkers in non-AD tauopathies are not well understood, further investigations are needed.
Another EMCI subject (#36) presented “A-T-N+” with FTD pattern in FDG-PET, and was suspected of early stage of non-amyloid non-tau FTLD.
The present study also revealed 4 subjects presenting DLB pattern in FDG-PET, with occipital hypometabolism extending to surrounding areas (Figure 3). Three of them were amyloid positive and all were tau positive. It is not clear whether they were preclinical or prodromal stage of atypical AD, or DLB with or without amyloid deposition.
These findings suggest that a significant fraction of the subjects in this study might be related to non-AD disorders such as DLB, SNAP, PART, argyrophilic grain disease (AGD), and TDP-43 proteinopathy (like LATE) (17). Even if they are amyloid positive, there is a possibility of incidental amyloid deposition. Therefore, possibility of non-AD disorders should always be considered when clinical trials targeting preclinical or prodromal AD are designed and subjects are selected based on the biomarkers.

 

Conclusion

In conclusion, this study confirmed the known changes of PET and CSF biomarkers in preclinical and prodromal AD, and at the same time, suggested difficulties of determining the criteria and cutoff level of those biomarkers to evaluate such subjects as well as the possibility of unsolicited inclusion of early-phase non-AD disorders.

 

Acknowledgements: We are grateful for the materials and technical supports for the PET imaging by Fujifilm Toyama Chemical, Avid Radiopharmaceuticals/Eli Lilly Japan, and GE Healthcare. PET centers that imaged the subjects but did not belong to the clinical site that enrolled the subjects are also acknowledged, including Tohoku University Cyclotron and Radioisotope Center (CYRIC), Tsukuba Advanced Imaging Center (AIC), Tokyo Metropolitan Institute of Gerontology (TMIG), Aizawa Hospital, MI Clinic, and Kobe City Medical Center General Hospital (KCGH). We thank all the people who participated in this study in the clinical and imaging sites as well as in the Core sites.

Conflict of interest: The following conflicts of interest are disclosed by the authors. Senda reports provision of devices, cassettes, and precursors from Avid/Eli Lilly Japan and GE, funding as PI of clinical trials sponsored by Eli Lilly, Eisai, Biogen, Cerveau and Merck, as well as leadership role in the Japanese Society of Nuclear Medicine as board member, congress chair and committee chair. Ikeuchi reports grants from AMED (JP19dk0207020, JP20dk0202028, JP20dm0207073). Matsuda reports a grant from AMED (19dk0207020h0005), intramural grants from National Center of Neurology and Psychiatry, and an entrusted research fund from Nihon Medi-Physics Co. Ltd. Iwatsubo reports a grant from an anonymous Foundation. Iwata reports grants from AMED (19dk0207020h0005, 16dk0207028h0001). Ikari is a full time employee of CMIC Inc. as well as graduate student of Osaka University. Washizuka reports research funding from AMED and pharmaceutical companies including Otsuka, Eisai, Pfizer, Daiichi-Sankyo, Tsumura, Mochida, Astellas, Shionogi, Takeda, Sumitomo-Dainippon, as well as honoraria from such pharmaceutical companies. Kazunari Ishii reports honoraria from Nihon Medi-Physics. Yokota reports licensing and collaboration research with Takeda Pharmaceutical Company. Nakanishi reports research funding from Eisai and Elli Lilly Japan as well as leadership role as a director in the Japan Society for Dementia Research. Shimada reports grants from AMED (19dk0207020h0005, 20dk0207028h0005). The other authors have nothing to disclose.

Funding: This study was financially supported by grants from Japan Agency for Medical Research and Development (AMED) 19dk0207020h0005, 20dk0202028h0005 and 20dm0207073h003, as well as by an anonymous Foundation.

 

References

1. Jack CR, Jr., Bennett DA, Blennow K, et al. A/T/N: An unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology 2016;87:539-547.
2. Iwatsubo T, Iwata A, Suzuki K, et al. Japanese and North American Alzheimer’s Disease Neuroimaging Initiative studies: Harmonization for international trials. Alzheimers Dement 2018;14:1077-1087.
3. Ikari Y, Akamatsu G, Nishio T, et al. Phantom criteria for qualification of brain FDG and amyloid PET across different cameras. EJNMMI physics 2016;3:23.
4. Yamane T, Ishii K, Sakata M, et al. Inter-rater variability of visual interpretation and comparison with quantitative evaluation of (11)C-PiB PET amyloid images of the Japanese Alzheimer’s Disease Neuroimaging Initiative (J-ADNI) multicenter study. Eur J Nucl Med Mol Imaging 2017;44:850-857.
5. Thurfjell L, Lilja J, Lundqvist R, et al. Automated quantification of 18F-flutemetamol PET activity for categorizing scans as negative or positive for brain amyloid: concordance with visual image reads. J Nucl Med 2014;55:1623-1628.
6. Clark CM, Pontecorvo MJ, Beach TG, et al. Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-β plaques: a prospective cohort study. The Lancet Neurology 2012;11:669-678.
7. Yamane T, Ikari Y, Nishio T, et al. Visual-statistical interpretation of (18)F-FDG-PET images for characteristic Alzheimer patterns in a multicenter study: inter-rater concordance and relationship to automated quantitative evaluation. AJNR Am J Neuroradiol 2014;35:244-249.
8. Silverman DHS, Small GW, Chang CY, et al. Positron Emission Tomography in Evaluation of DementiaRegional Brain Metabolism and Long-term Outcome. JAMA 2001;286:2120-2127.
9. Herholz K, Salmon E, Perani D, et al. Discrimination between Alzheimer dementia and controls by automated analysis of multicenter FDG PET. NeuroImage 2002;17:302-316.
10. Herholz K, Westwood S, Haense C, Dunn G. Evaluation of a calibrated (18)F-FDG PET score as a biomarker for progression in Alzheimer disease and mild cognitive impairment. J Nucl Med 2011;52:1218-1226.
11. Southekal S, Devous MD, Sr., Kennedy I, et al. Flortaucipir F 18 Quantitation Using Parametric Estimation of Reference Signal Intensity. J Nucl Med 2018;59:944-951.
12. Devous MD, Sr., Joshi AD, Navitsky M, et al. Test-Retest Reproducibility for the Tau PET Imaging Agent Flortaucipir F 18. J Nucl Med 2018;59:937-943.
13. Reimand J, Collij L, Scheltens P, Bouwman F, Ossenkoppele R, Alzheimer’s Disease Neuroimaging I. Association of amyloid-beta CSF/PET discordance and tau load 5 years later. Neurology 2020;95:e2648-e2657.
14. Mattsson-Carlgren N, Andersson E, Janelidze S, et al. Aβ deposition is associated with increases in soluble and phosphorylated tau that precede a positive Tau PET in Alzheimer’s disease. Science advances 2020;6:eaaz2387.
15. Sanchez JS, Becker JA, Jacobs HIL, et al. The cortical origin and initial spread of medial temporal tauopathy in Alzheimer’s disease assessed with positron emission tomography. Science translational medicine 2021;13:eabc0655.
16. Coutinho AM, Busatto GF, de Gobbi Porto FH, et al. Brain PET amyloid and neurodegeneration biomarkers in the context of the 2018 NIA-AA research framework: an individual approach exploring clinical-biomarker mismatches and sociodemographic parameters. Eur J Nucl Med Mol Imaging 2020;47:2666-2680.
17. Jack CR, Jr., Knopman DS, Chételat G, et al. Suspected non-Alzheimer disease pathophysiology–concept and controversy. Nature reviews Neurology 2016;12:117-124.

COMMUNICATING PERSONAL RISK PROFILES OF ALZHEIMER’S DISEASE TO OLDER ADULTS: A PILOT TRIAL

 

I. Choi1, H. La Monica1, S.L. Naismith2, A. Rahmanovic2, L. Mowszowski2, N. Glozier1

 

1. Central Clinical School, Brain and Mind Centre, Faculty of Medicine and Health, University of Sydney, Australia; 2. Charles Perkins Centre, Faculty of Science, School of Psychology and the Brain and Mind Centre, University of Sydney, Australia

Corresponding Author: Dr Isabella Choi, 94 Mallett Street, Camperdown, NSW 2050, Australia, isabella.choi@sydney.edu.au, +612 8627 7240.

J Prev Alz Dis 2021;
Published online June 23, 2021, http://dx.doi.org/10.14283/jpad.2021.34

 


Abstract

Communicating personal Alzheimer’s disease risk profiles based on validated risk algorithms may improve public knowledge about risk reduction, and initiate action. This proof of concept pilot trial aimed to test whether this is feasible and potentially effective and/or harmful. Older at-risk adults (N=24) were provided with their personal Alzheimer’s disease risk profile online, which contained information on their personal risk level, scores and tailored recommendations to manage modifiable risk factors. After receiving the risk profile, participants were significantly more accurate in identifying risk and protective factors, and revised their perceived risk to be lower than their initial estimate. There was no apparent harm seen in psychological distress or dementia-related worry. This shows preliminary support for the feasibility of delivering personal dementia risk profiles to low risk, help-seeking older adults in an online format. A definitive trial examining behavioural outcomes and testing in groups with higher risk profiles is now warranted.

Key words: Risk communication, health literacy, psychological distress, prevention, Alzheimer’s disease.


 

Introduction

Up to a third of Alzheimer’s disease cases can be prevented through improved education and reduction of modifiable risk factors (1). Growing evidence from multi-domain interventions shows that targeting modifiable risk factors can reduce risk of Alzheimer’s disease (AD) and improve cognition (2, 3). However, lack of knowledge about dementia and its risk factors among the public is a major barrier to individuals implementing behavioural and lifestyle change and, in turn, to dementia risk reduction (4).
Having an accurate understanding of one’s personal risk of future disease is considered essential for engaging in behaviours for risk reduction. Most health behaviour change models, including the Health Belief Model, identify four major constructs that surround health behaviour: health literacy, perceived susceptibility, motivation to change, and perceived barriers to change (5). However, there is poor health literacy of dementia risk, risk factors, and prevention strategies among the public. A systematic review found that almost half of the general public agreed that dementia was a normal and non-preventable part of ageing (6). Mental activity, healthy diet, physical exercise and social engagement were the most commonly nominated ways to reduce risk, but other well established risk factors such as vascular risk (including smoking, high blood pressure, high cholesterol, obesity), low education, poor mental health, brain trauma, and environmental toxins were rarely mentioned (6-8). Only around 25% of Australians were confident they could reduce risk (8). However, there is support that improving dementia health literacy has a positive impact on risk reduction. People who had a strong belief that dementia risk could be reduced, had moderate to high knowledge of risk-reduction behaviours, or had high confidence that risk reduction can be achieved were almost twice as likely to take action to reduce risk (9).
Communicating personal dementia risk level and risk factors to at-risk adults, based on validated AD risk algorithms, may be one way to improve dementia risk knowledge and engage people in risk reduction behaviours. A number of dementia risk algorithms have been validated for the general population and can identify those with high risk with acceptable predictive ability (10). For instance, the Australian National University Alzheimer’s Disease Risk Index (ANU-ADRI) includes various modifiable risk factors that have been validated for middle-age and older adults that are easily assessed via self-report (11, 12). These evidence-based algorithms, along with personalised risk factor feedback and recommendations to reduce risk, can be delivered online for wide access, allowing people to screen for their risk without having had to first consult with a physician. Older adults are able to use technology proficiently and over 90% of help-seeking older adults with some degree of cognitive impairment use the Internet at home, most commonly for emails (13). Such online dementia risk algorithms already exist and are made freely available to the public, for instance, the CAIDE Risk Score App allows users to detect their dementia risk, obtain information on modifiable risk factors, and receive suggestions on how to modify their risk (14).
Surveys have found that there is high interest among older adults in knowing their risk of AD (15, 16) While there may be potential benefits to disclosing risk, there are also concerns that this could cause negative psychological outcomes. For instance, about 30% of older adults who were interested in knowing their risk also actively worried about developing AD (17). Evidence from a systematic review suggests that disclosure of increased AD risk was not associated with anxiety or depression, but did lead to heightened test-related distress, long-term care insurance uptake and health-related behaviour changes (18).
This pilot trial explores the feasibility and acceptability of communicating online personal dementia risk profiles to at-risk older adults and the impact on dementia health literacy, motivation to engage in behaviour change, and potential harmful psychological effects.

 

Method

Design

This is an uncontrolled proof of concept pilot trial. The study was approved by the University of Sydney Human Research Ethics Committee (protocol number: 2019/669) and registered with the Australian New Zealand Clinical Trials Registry (ANZCTRN12619001242112p).

Participants

People attending the Healthy Brain Ageing Clinic, a specialised memory research clinic for people aged 50 years or older in Sydney, Australia, who were clinically diagnosed with mild cognitive impairment (MCI) or subjective cognitive decline (SCD) between October 2019 to May 2020 were recruited. Those who had dementia or pre-existing severe cognitive impairment due to neurological conditions, psychosis, intellectual disability, substance misuse, stroke, or acquired brain injury were excluded.
As part of standard clinic procedure, all clinic attendees completed self-report measures and were assessed by a geriatrician or neurologist, a psychologist, and a neuropsychologist for a review of medical and psychiatric history, mood, and cognitive functioning. Diagnoses were rated according to consensus, including at least two neuropsychologists and one specialist, and were used to exclude those with a dementia diagnosis and other exclusion criteria. Within two weeks of the clinic assessment, attendees received neuropsychological test result feedback over the phone from a neuropsychologist. The clinic does not provide treatment but refers attendees to suitable clinical investigations, e.g. sleep studies, if warranted, as well as clinical trials for which they are deemed eligible.
After receiving neuropsychological test feedback from a neuropsychologist, eligible attendees were invited via a telephone call to participate in the current study. Interested people had to have an email account and were emailed a survey link via REDCap (Research Electronic Data Capture), a web-based research management platform, with the Participant Information Statement and Consent Form. Participants gave consent to extract relevant data collected from their recent clinic assessment to populate their personal dementia risk profile.

Procedures

Participants completed self-report baseline measures online via REDCap. Participants’ demographic and risk information were extracted from their standard clinic assessment to compile their personal dementia risk profile using the ANU-ADRI (11). Risk factors in the model included: age, gender, highest level and total number of years of education completed, body mass index, diabetes, depression, high cholesterol, traumatic brain injury, smoking status, alcohol intake, social engagement, physical activity, cognitive activity, and diet.
Within two weeks of completing the baseline self-report measures, participants were emailed a pdf document with their personal dementia risk profile. The risk profile contained standard information about dementia, an explanation of their personal dementia risk profile, and information about the ANU-ADRI risk model. Participants were presented with a visual representation of their risk level in the form of a thermometer showing their risk from 0 to 100, along with an explainer “Your risk of developing dementia is low/ moderate/ high. It is estimated that XX out of 100 people with your risk factors will develop dementia in their lifetime” (Figure 1). They were reminded that this is an estimate based on their risk factors rather than a definitive guarantee, and that there are some risk factors they cannot change but some they could potentially work on to reduce their risk. Participants also received a summary of the dementia risk factors included in the risk model and their scores on each risk factor. They were told why the risk factor was important for brain functioning and were given tailored recommendations to manage each risk factor based on their risk factor score, as well as links for more information.
One week after receiving their risk profile, participants received an email asking them to complete the online post-intervention measures in REDCap. After completing all study measures, participants were reimbursed with a $20 gift card in return for their time.

Figure 1. Example of the risk level and risk factor feedback provided in the personal risk profile

 

 

Measures

Primary outcome: Dementia health literacy

Participants were asked “How likely do you think that you will get Alzheimer’s disease in your lifetime?” to assess perceived risk on a scale, where 0%=certain not to happen and 100%=certain to happen. To examine accuracy of perceived risk, the participant’s perceived risk was subtracted from their ANU-ADRI risk estimate. We recoded the difference (D) into a categorical variable, with <−10% indicating overestimation, >10% indicating underestimation, and accurate if −10% ≤ D ≤ 10%, in accordance with previous studies (19). Similarly, to examine change in perceived risk, participants’ perceived risk at post-intervention was subtracted from their perceived risk at baseline (d), and <-10% indicates increased perceived risk, >10% indicates reduced perceived risk, and −10% ≤ d ≤ 10% indicates no change.
We adapted the dementia risk and protective factors questionnaire in the MijnBreincoach survey (20) to assess for knowledge of dementia risk factors. We included five additional modifiable risk and protective factors that were identified in the ANU-ADRI (11) and the Lancet Commission for dementia prevention (21) (i.e. traumatic brain injury, social activity, sleep, education, and age), totalling 19 risk factors. Additional questions asked about barriers to improving brain health, confidence in risk reduction (8), and worry about getting dementia (7).

Secondary outcome: Motivation to Change Lifestyle and Health Behaviours for Dementia

The Motivation to Change Lifestyle and Health Behaviours for Dementia Risk Reduction (MCLHB-DRR) Scale is designed to assess beliefs and attitudes about lifestyle and health behavioural changes for dementia risk reduction among middle-aged and older adults (22). The scale includes (27) items matched onto seven subscales that reflect the seven concepts of the Health Behaviour Model. All items are rated on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). The scale has moderate to high internal consistencies for the seven subscales, and moderate test-retest reliability (18). Cronbach’s alpha for each of the subscales in this study were: perceived susceptibility (.916), perceived severity (.331), perceived benefits (.715), perceived barriers (.878), cues to action (.656), general health motivation (.638), and self-efficacy (.615).

Secondary outcome: Psychological distress

The K10 is a commonly used screening scale for non-specific psychological distress validated for use in Australia (23). The K10 has also been demonstrated as having moderate sensitivity to symptom change in an Australian sample (24). Scores on the K10 range from 10 to 50, and a score of 30 or more indicates a severe level of distress. Cronbach’s alpha in this study was 0.841.

Secondary outcome: Dementia-related worry

The Dementia Worry Scale was used to assess dementia-related worry (25). It has strong internal consistency and test-retest reliability. It consists of 12 items with scores ranging from 15 to 60. Cronbach’s alpha in this study was 0.908.

User evaluation

We adapted a five-point scale (from 1= not at all to 5= completely) previously used to assess user experience of a dementia information website (26). Participants were asked whether the information provided was engaging and easy to understand as well as how helpful they found the risk profile and how much they felt they had learned (from 1=nothing at all to 5=a great deal). Additionally, participants were asked if they required more information about their dementia risk profile and were given the option to discuss their experience of using the risk profile with a researcher in a telephone interview.

Data analysis

Data was analysed using SPSS version 23.0. Descriptive statistics regarding participant and baseline characteristics were analysed. Fischer’s exact tests and paired samples t-tests were used to test for differences between outcome measures pre- and post- receiving the dementia risk profile. All p-values were two-sided with an alpha of 0.05 to test for significance.

 

Results

Demographics and baseline characteristics

Overall, 24 eligible participants participated in the trial (Figure 2). Participants’ ages ranged from 53 to 87, with a mean age of 69.54 years (SD 7.69). Over half of the participants (54%) were female and majority spoke English as a first language (83%). Majority were tertiary educated (75%), 21% had completed a trade certificate, and 4% had completed high school. Majority were retired (58%), 29% were employed, and 13% were unemployed. Over half were married or in a de facto relationship (58%), 25% were widowed or divorced, and 17% had never married. The majority (75%) of participants had MCI and 25% had SCD.
Almost all participants (96%; 23/24) were considered to have Low Risk of developing AD (ANU-ADRI score of less than 17), and one participant (4%) was considered as having High Risk (ANU-ADRI score of greater than 27). Participants’ perceived risk of developing AD ranged from 10-99 (M=51.63, SD=23.85), with the majority of participants overestimating their personal risk (87.5%; 21/24).

Figure 2. Participant flow

Pre-post change on dementia health literacy

All participants completed the post-intervention questionnaires. After receiving their personal dementia risk profile, participants’ perceived risk of developing AD ranged from 3-81 (M=38.83, SD=25.02). There was a significant decrease in perceived risk among the group (p = .010). A total of 18 participants (75%) still overestimated their level of risk whereas the remaining 6 correctly identified their level of risk (25%). Eleven participants (45.8%) reported a reduction in their perceived risk, eleven (45.8%) reported no change, and two participants’ (8.3%) perceived risk had increased.
The average number of correctly identified risk and protective factors increased from a mean of 11.42 items (SD= 4.50) at baseline to 13.96 items (SD= 3.98) (t1,23= -3.839, p=.001) at follow-up.

Pre-post change on motivation to engage in behaviour change and psychological effects

There was a significant reduction on the perceived susceptibility subscale of the MCLHB (t1,23=4.416, p<.001) from baseline (M=12.86; SD=3.30) to 1-week follow up (M=10.29; SD=4.21), but no change on the other subscales. There was no change on the K10 or Dementia Worry Scale.
Participants’ self-reported worry about getting dementia was significantly reduced at follow-up, from 2.75 (SD=.85) to 2.29 (SD=.91) (t1,23= 3.412, p= .002). The most common barriers to reducing risk factors at baseline were lack of knowledge (45.8%; 11/24), followed by health problems (25%; 6/24). At 1-week follow up, the most common barriers were lack of motivation (29.2%; 7/24), health problems (29.2%; 7/24), and difficulty with organisation (25%; 6/24). There was no change in confidence to take action to change risk.

User evaluation

Overall, 70.9% (17/24) of participants agreed that the information in the personal risk profile was engaging, 79.2% (19/24) agreed the information was easy to understand, 79.2% (19/24) agreed it was helpful, and 79.2% (19/24) reported they learned a good deal from their personal risk profile.
Two participants participated in the optional telephone interview with a researcher. Both expressed surprise at their lower than expected AD risk feedback, and identified difficulty addressing some of their risk factors (e.g. getting motivated to exercise). One participant agreed that seeking guidance from a health professional may support them to work on their risk factors.

 

Discussion

This pilot study aimed to explore the feasibility, acceptability and potential impact of providing an online personal dementia risk profile to help-seeking older adults at risk of developing AD on risk knowledge, motivation to change health-related behaviours, and psychological effects. To our knowledge, this is the first study focusing on the effects of communicating personal risk profile using risk algorithms based on epidemiological risk factors. Communication of the personal dementia risk profile led to more accurate knowledge of AD risk factors and improved understanding of perceived susceptibility among patients with MCI and SCD. Importantly, there was no negative effect of communicating the personal risk profile online on psychological distress or dementia-related worry among our participants. Participants mostly had a low risk of developing AD, but still reported reduced worry about getting dementia after receiving their risk profile. These findings support the feasibility and acceptability of using dementia risk algorithms to deliver personal risk profiles to low risk older adults in an online format, and indicate that providing this information can improve AD health literacy without a negative impact on psychological wellbeing.
However, there was little evidence in this study that providing personal risk profiles as a standalone intervention was sufficient to motivate change in behaviours to address AD risk factors. Although providing the dementia risk profile addressed one main barrier for risk reduction at baseline (i.e. lack of knowledge of dementia risk factors), participants reported that lack of motivation, health problems, and difficulty with organisation became the main barriers after receiving their risk profile. This suggests that older adults need extra support to effect behavioural change. The personal dementia risk profile could potentially be used as part of a collaborative, shared decision-making approach to address these barriers by guiding and engaging users, carers and clinicians to choose several high impact or easy-to-change risk factors to focus on, and by providing feedback on the change in risk level if risk factors are modified. Trials are underway to test the impact of a tailor-made online lifestyle programme targeting modifiable risk factors on risk score and health behaviours and compliance to health advice (27). There may also be a role for clinicians to follow up with specific guidance on addressing health problems and to assist the older adult to develop a personalised risk reduction plan. A recent rapid review on approaches to healthy ageing interventions for older adults demonstrated that optimal interventions are those that incorporate collaborative approaches with shared decision-making and behavioural change techniques (28). In this regard, the personal dementia risk profile represents a useful tool that clinicians can draw on to present evidence-based, tailored, health and risk information, which in turn can stimulate a collaborative decision-making process around which health/lifestyle factors to target and how best to achieve long-lasting behaviour change.
An interesting finding was that most participants overestimated their risk of developing AD, even after receiving their personal risk profile. This is possibly reflective of our cohort which was composed of help-seeking older adults seeking an evaluation at a memory clinic and were concerned with developing dementia. The continuing high levels of perceived risk at follow up is unsurprising given that previous research has found that even among individuals who accurately recalled their communicated AD risk, over 50% did not fully adjust their perceived risk to match the communicated risk, and that high baseline AD risk perception was the strongest predictor of overestimation of risk (29). It is also possible that participants may not have readily accepted the communicated risk after receiving lower than expected risk feedback, as seen by interview participants expressing surprise at their communicated risk.
Nonetheless, this has implications for supporting clinicians to communicate AD risk information to people with MCI. A survey found that 90% of neurologists said they counselled MCI patients on their risk of AD in general terms but only 60% communicated AD risk in numeric terms (30). Our findings provide preliminary support that patients with MCI or SCD understood numeric AD risk information and risk factors even when it is communicated online without support, and that the multifaceted approach with a clear visual representation and accompanying explanatory text may have facilitated understanding. Clinicians are encouraged to discuss numeric risk estimates with patients with visual aids, explain how these are estimated from risk models, and explore reasons for discord to improve risk appraisal.
There are several important considerations in interpreting the findings. The study sample was a well-educated inner-city cohort who have expressed concern about their memory and were highly motivated to seek help. Participants already knew that they did not have dementia. Their relatively positive reactions to their personal risk profile may reflect their personal interest in managing their brain health or because they were reassured of having low risk. It is unclear whether a general population or primary care sample, who had subjective memory concerns and not assessed for AD, would have similar reactions. It should be noted that a number of older adults approached to take part in the trial declined due to not wanting to know their risk or because they felt overwhelmed. Further research is needed to explore these concerns about knowing one’s personal risk.
The majority of our sample had low risk of developing AD, and it is unknown how moderate or high risk adults would respond to their personal dementia risk profile. There is some indication that high risk individuals, such as those who screen positive to genetic biomarkers, have heightened test-related distress (18). In order for dementia risk profiles to be widely and safely distributed to older adults in public health programs, particularly if they are to be delivered online in the absence of immediate clinical support, it is important to understand how moderate or high risk older adults react to their personal dementia risk profile and to monitor any potential adverse reactions. Finally, this was a pilot trial with a small sample size and short-term follow up. Longer-term follow up and randomised controlled trials to examine effects of communication of personal risk of developing AD are required.
The application of dementia risk algorithms to identify those at risk and to promote and encourage risk reduction behaviour is still in its early stages. This study provides preliminary support for the utility of using risk models that incorporate accessible and potentially modifiable risk factors to communicate personal dementia risk profiles to at-risk older adults.

 

Acknowledgements: The authors would like to acknowledge Professor Kaarin J. Anstey and Dr Sarang Kim for permission to use the ANU Alzheimer’s Disease Risk Index and their advice on adapting it to the Healthy Brain Ageing clinic measures. We thank the participants who have helped make this research possible.

Funding: This study was supported by a Dementia Australia Research Foundation project grant award. The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.

Ethical standards: The study was approved by the University of Sydney Human Research Ethics Committee (protocol number: 2019/669) and registered with the Australian New Zealand Clinical Trials Registry (ANZCTRN12619001242112p).

Conflict of interest: The authors confirm that there are no known conflicts of interest associated with this publication.

 

References

1. Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer’s disease: an analysis of population-based data. The Lancet Neurology. 2014;13(8):788-794 http://dx.doi.org/10.1016/S1474-4422(14)70136-X.
2. Isaacson RS, Hristov H, Saif N, et al. Individualized clinical management of patients at risk for Alzheimer’s dementia. Alzheimer’s & dementia : the journal of the Alzheimer’s Association. 2019;15(12):1588-1602 https://doi.org/10.1016/j.jalz.2019.08.198.
3. Ngandu T, Lehtisalo J, Solomon A, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. The Lancet. 2015;385(9984):2255-2263 http://dx.doi.org/10.1016/ S0140-6736(15)60461-5.
4. Kim S, Sargent-Cox KA, Anstey KJ. A qualitative study of older and middle-aged adults’ perception and attitudes towards dementia and dementia risk reduction. J Adv Nurs. 2015;71(7):1694-1703 https://doi.org/10.1111/jan.12641.
5. Janz NK, Becker MH. The Health Belief Model: A Decade Later. Health Educ Q. 1984;11:1-47 https://doi.org/10.1177/109019818401100101.
6. Cations M, Radisic G, Crotty M, Laver KE. What does the general public understand about prevention and treatment of dementia? A systematic review of population-based surveys. PLoS One. 2018;13(4):e0196085 https://doi.org/10.1371/journal.pone.0196085.
7. Low L-F, Anstey KJ. Dementia literacy: Recognition and beliefs on dementia of the Australian public. Alzheimer’s & Dementia. 2009;5(1):43-49 https://doi.org/10.1016/j.jalz.2008.03.011.
8. Smith BJ, Ali S, Quach H. Public knowledge and beliefs about dementia risk reduction: a national survey of Australians. BMC Public Health. 2014;14(1):661 https://doi.org/10.1186/1471-2458-14-661.
9. Smith BJ, Ali S, Quach H. The motivation and actions of Australians concerning brain health and dementia risk reduction. Health Promot J Austr. 2015;26(2):115-121 http://dx.doi.org/10.1071/HE14111.
10. Hou X-H, Feng L, Zhang C, Cao X-P, Tan L, Yu J-T. Models for predicting risk of dementia: a systematic review. J Neurol Neurosurg Psychiatry. 2018:jnnp-2018-318212 http://dx.doi.org/10.1136/jnnp-2018-318212.
11. Anstey KJ, Cherbuin N, Herath PM. Development of a New Method for Assessing Global Risk of Alzheimer’s Disease for Use in Population Health Approaches to Prevention. Prevention Science. 2013;14(4):411-421 http://dx.doi.org/10.1007/s11121-012-0313-2.
12. Anstey KJ, Cherbuin N, Herath PM, et al. A Self-Report Risk Index to Predict Occurrence of Dementia in Three Independent Cohorts of Older Adults: The ANU-ADRI. PLoS One. 2014;9(1):e86141 https://doi.org/10.1371/journal.pone.0086141.
13. LaMonica HM, English A, Hickie IB, et al. Examining Internet and eHealth Practices and Preferences: Survey Study of Australian Older Adults With Subjective Memory Complaints, Mild Cognitive Impairment, or Dementia. J Med Internet Res. 2017;19(10):e358 https://doi.org/10.2196/jmir.7981.
14. Sindi S, Calov E, Fokkens J, et al. The CAIDE Dementia Risk Score App: The development of an evidence-based mobile application to predict the risk of dementia. Alzheimer’s & Dementia : Diagnosis, Assessment & Disease Monitoring. 2015;1(3):328-333 http://dx.doi.org/10.1016/j.dadm.2015.06.005.
15. Sheffrin M, Stijacic Cenzer I, Steinman MA. Desire for predictive testing for Alzheimer’s disease and impact on advance care planning: a cross-sectional study. Alzheimers Res Ther. 2016;8(1):55. https://doi.org/10.1186/s13195-016-0223-9
16. Wikler EM, Blendon RJ, Benson JM. Would you want to know? Public attitudes on early diagnostic testing for Alzheimer’s disease. Alzheimers Res Ther. 2013;5(5):43 http://dx.doi.org/10.1186/alzrt206.
17. Roberts JS, McLaughlin SJ, Connell CM. Public beliefs and knowledge about risk and protective factors for Alzheimer’s disease. Alzheimer’s & Dementia. 2014;10(5):S381-S389 http://dx.doi.org/10.1016/j.jalz.2013.07.001.
18. Bemelmans SASA, Tromp K, Bunnik EM, et al. Psychological, behavioral and social effects of disclosing Alzheimer’s disease biomarkers to research participants: a systematic review. Alzheimers Res Ther. 2016;8(1):46 http://dx.doi.org/10.1186/s13195-016-0212-z.
19. Rimer BK, Halabi S, Sugg Skinner C, et al. Effects of a mammography decision-making intervention at 12 and 24 months. Am J Prev Med. 2002;22:247-257 https://doi.org/10.1016/S0749-3797(02)00417-8.
20. Heger I, Deckers K, van Boxtel M, et al. Dementia awareness and risk perception in middle-aged and older individuals: baseline results of the MijnBreincoach survey on the association between lifestyle and brain health. BMC Public Health. 2019;19(1):678 https://doi.org/10.1186/s12889-019-7010-z.
21. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. The Lancet. 2017;390(10113):2673-2734 https://doi.org/10.1016/S0140-6736(17)31363-6.
22. Kim S, Sargent-Cox K, Cherbuin N, Anstey KJ. Development of the motivation to change lifestyle and health Behaviours for Dementia Risk Reduction Scale. Dement Geriatr Cogn Dis Extra. 2014;4(2):172-183 https://doi.org/10.1159/000362228.
23. Kessler R, Andrews G, Colpe L, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med. 2002;32(06):959-976 https://doi.org/10.1017/S0033291702006074.
24. Perini SJ, Slade T, Andrews G. Generic effectiveness measures: Sensitivity to symptom change in anxiety disorders. J Affect Disord. 2006;90:123-130 https://doi.org/10.1016/j.jad.2005.10.011.
25. Kinzer A, Suhr JA. Dementia worry and its relationship to dementia exposure, psychological factors, and subjective memory concerns. Applied Neuropsychology: Adult. 2016;23(3):196-204 https://doi.org/10.1080/23279095.2015.1030669.
26. Farrow M. User perceptions of a dementia risk reduction website and its promotion of behavior change. JMIR research protocols. 2013;2(1):e15-e15 https://doi.org/10.2196/resprot.2372.
27. Vrijsen J, Abu-Hanna A, Maeckelberghe ELM, et al. Uptake and effectiveness of a tailor-made online lifestyle programme targeting modifiable risk factors for dementia among middle-aged descendants of people with recently diagnosed dementia: study protocol of a cluster randomised controlled trial (Demin study). BMJ Open. 2020;10(10):e039439 http://dx.doi.org/10.1136/bmjopen-2020-039439.
28. Owusu-Addo E, Ofori-Asenso R, Batchelor F, Mahtani K, Brijnath B. Effective implementation approaches for healthy ageing interventions for older people: A rapid review. Arch Gerontol Geriatr. 2021;92:104263 https://doi.org/10.1016/j.archger.2020.104263.
29. Linnenbringer E, Roberts JS, Hiraki S, Cupples LA, Green RC. “I know what you told me, but this is what I think:” Perceived risk of Alzheimer disease among individuals who accurately recall their genetics-based risk estimate. Genet Med. 2010;12(4):219-227 https://doi.org/10.1097/GIM.0b013e3181cef9e1.
30. Roberts JS, Karlawish JH, Uhlmann WR, Petersen RC, Green RC. Mild cognitive impairment in clinical care: A survey of American Academy of Neurology members. Neurology. 2010;75(5):425-431 https://doi.org/10.1212/WNL.0b013e3181eb5872.

TAUROURSODEOXYCHOLIC ACID ATTENUATES DIET-INDUCED AND AGE-RELATED PERIPHERAL ENDOPLASMIC RETICULUM STRESS AND CEREBRAL AMYLOID PATHOLOGY IN A MOUSE MODEL OF ALZHEIMER’S DISEASE

 

T. Ochiai1,2, T. Nagayama1, K. Matsui1, K. Amano1, T. Sano1, T. Wakabayashi1,3, T. Iwatsubo1

 

1. Department of Neuropathology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; 2. Pharmacology Department, Drug Research Center, Kaken Pharmaceutical Co., LTD., Kyoto, Japan; 3. Department of Innovative Dementia Prevention, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan

Corresponding Author: Tomoko Wakabayashi, Takeshi Iwatsubo, Department of Neuropathology, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan, Tel: +81-3-5841-3541, Fax: +81-3-5841-3613 tomoko-wakabayashi@umin.ac.jp, iwatsubo@m.u-tokyo.ac.jp

J Prev Alz Dis 2021;4(8):483-494
Published online June 17, 2021, http://dx.doi.org/10.14283/jpad.2021.33

 


Abstract

BACKGROUND: Obesity and diabetes are well-established risk factors of Alzheimer’s disease (AD). In the brains of patients with AD and model mice, diabetes-related factors have been implicated in the pathological changes of AD. However, the molecular mechanistic link between the peripheral metabolic state and AD pathophysiology have remained elusive. Endoplasmic reticulum (ER) stress is known as one of the major contributors to the metabolic abnormalities in obesity and diabetes. Interventions aimed at reducing ER stress have been shown to improve the systemic metabolic abnormalities, although their effects on the AD pathology have not been extensively studied.
OBJECTIVES: We examined whether interventions targeting ER stress attenuate the obesity/diabetes-induced Aβ accumulation in brains. We also aimed to determine whether ER stress that took place in the peripheral tissues or central nervous system was more important in the Aβ neuropathology. Furthermore, we explored if age-related metabolic abnormalities and Aβ accumulation could be suppressed by reducing ER stress.
METHODS: APP transgenic mice (A7-Tg), which exhibit Aβ accumulation in the brain, were used as a model of AD to analyze parameters of peripheral metabolic state, ER stress, and Aβ pathology in the brain. Intraperitoneal or intracerebroventricular administration of taurodeoxycholic acid (TUDCA), a chemical chaperone, was performed in high-fat diet (HFD)-fed A7-Tg mice for ~1 month, followed by analyses at 9 months of age. Mice fed a normal diet were treated with TUDCA by drinking water for 4 months and intraperitoneally for 1 month in parallel, and analyzed at 15 months of age.
RESULTS: Intraperitoneal administration of TUDCA suppressed ER stress in the peripheral tissues and ameliorated the HFD-induced obesity and insulin resistance. Concomitantly, Aβ levels in the brain were significantly reduced. In contrast, intracerebroventricular administration of TUDCA had no effect on the Aβ levels. Peripheral administration of TUDCA was also effective against the age-related obesity and insulin resistance, and markedly reduced amyloid accumulation.
CONCLUSIONS: Interventions that target peripheral ER stress might be beneficial therapeutic and prevention strategies against brain Aβ pathology associated with metabolic overload and aging.

Key words: ER stress, Aβ, obesity, diabetes, Alzheimer’s disease.


 

Introduction

A growing body of evidence suggests that a certain proportion of the causative factors of dementia (1), and also those of Alzheimer’s disease (AD) (2), may be attributable to lifestyle-related risks. Therefore, it may be possible to prevent AD by identifying and reducing those risk factors (3). Epidemiological and biological evidence that support the association of obesity and type 2 diabetes mellitus (T2DM) with AD is increasingly accumulating. In particular, meta-analyses of prospective cohort studies showed that the presence of diabetes increases the risk of AD by ~1.5 times (4, 5). Various biological mechanisms have been suggested to explain the link between these lifestyle-related risks and AD: pathological conditions such as chronic inflammation, hyperglycemia, insulin resistance, and vascular complications underlie cognitive dysfunctions (6, 7). Notably, it has been established that peripheral insulin resistance is associated with higher amyloid plaque loads in the brains of patients with AD (8). This has been recapitulated in the mouse models of AD, in which the degrees of diet-induced insulin resistance and metabolic abnormalities correlated well with brain Aβ accumulation, and moreover, the effects were reversible (9–13). Because Aβ accumulation is considered to play a causative role in the pathophysiology of AD (14), deterioration of the peripheral metabolic state may adversely affect both the pathological and symptomatic manifestations of AD. However, it remains unresolved what kind of molecular abnormalities observed in the pathophysiology of diabetes are causative to the AD pathogenesis in the brain.
One of the key factors that plays a central role in the pathogenesis of obesity and diabetes is the endoplasmic reticulum (ER) stress. Metabolic overload, e.g., high-fat diet (HFD) feeding and overnutrition, is a primary cause of obesity and T2DM, in which the action of insulin on peripheral tissues is attenuated. ER stress due to metabolic overload is considered to be a major culprit in the insulin resistance in metabolic organs, e.g., liver and adipose tissue (15). Under the ER stress conditions, accumulation of misfolded proteins and loss of homeostasis in the ER are initially detected by sensor proteins. This results in the activation of three major unfolded protein response (UPR) signaling pathways, i.e., IRE1-XBP1, PERK-eIF2α-ATF4, and ATF6, which in turn induce the expression of chaperone proteins and repress overall protein translations (16). Several mechanisms that link UPR to insulin resistance through direct inhibition of the insulin signaling pathway have been postulated. It has been shown that IRE1-dependent activation of JNK1 phosphorylates serine residues of insulin receptor substrate 1 (IRS-1) and attenuates insulin receptor signaling (17, 18).
Molecular signatures of ER stress and UPR activation have also been documented in postmortem AD brains (19–22). Accumulation of Aβ and tau, pathogenic proteins in AD, has been proposed to cause ER stress, leading to synaptic dysfunction and neurodegeneration (22–24). These observations have highlighted ER stress as a common pathophysiological feature of the metabolic disorders and AD. However, the question as to whether ER stress elevated in the peripheral tissues or in the central nervous system is causative to the AD changes, remains elusive.
ER stress can be suppressed in vivo by administration of chemical chaperones, e.g. tauroursodeoxycholic acid (TUDCA) and 4-phenyl butyric acid (4-PBA). Treatment with TUDCA and 4-PBA has been shown to improve the systemic metabolic abnormalities in obesity/diabetes models, e.g., ob/ob mice and HFD-fed mice (25–28), supporting the notion that ER stress is causally involved in the pathogenesis of metabolic disorders. Given the correlation between metabolic status and AD pathology, interventions targeting ER stress induced by metabolic overload may also be a potential preventive target for the concomitant worsening of AD pathology.
In this study, we investigated the effects of metabolic improvement by administration of TUDCA on the HFD-induced increase in ER stress and exacerbation of AD pathophysiology in mice overexpressing the amyloid precursor protein (APP) in the brain (A7-Tg mice). Peripheral administration of TUDCA counteracted HFD-induced ER stress and metabolic abnormalities in the periphery, and decreased Aβ levels in the brain, whereas direct brain administration of TUDCA had no effect. Furthermore, age-related exacerbation of ER stress and metabolic abnormalities also were improved by TUDCA, resulting in a marked suppression of brain amyloid accumulation. These results support the view that interventions targeting peripheral ER stress might be effective in reducing Aβ accumulation in the brain caused by obesity and diabetes that are associated with metabolic overload or by aging.

 

Materials and Methods

Mice

A7 transgenic mice (A7-Tg) overexpressing human APP695 harboring KM670/671NL and T714I familial AD mutations under the control of the Thy-1.2 promoter were backcrossed and maintained on a C57BL/6J background (11). Mice were maintained on a 12 h light/dark cycle and provided ad libitum access to water. In the experiments with a high-fat diet, mice were fed standard chow diet (CRF-1, Oriental Yeast Co., Ltd.) until 3 months of age. Thereafter, they were either maintained on the standard chow or switched to a high-fat diet (HFD32, CLEA Japan Inc.) containing 32% fat. The animal care and experimental procedures were approved by the animal experiment committee of The University of Tokyo Graduate School of Medicine.

TUDCA treatment

We performed three TUDCA administration experiments. In the first experiment, 8-month-old HFD-fed A7-Tg mice were intraperitoneally administered TUDCA (250 mg/kg, Merck) or saline twice per day for 30 days, as previously reported (25). In the second experiment, 7- to 8-month-old HFD-fed A7-Tg mice received intracerebroventricular administration of TUDCA (10 µg/day) or phosphate buffered saline (PBS) for 28 days, in accordance with previous reports (29, 30). For intracerebroventricular administration of TUDCA, cannulas (ALZET Brain Infusion Kit 3, ALZET Osmotic Pumps) were stereotaxically implanted into the lateral ventricle (from bregma: anteroposterior -0.5 mm, mediolateral -1.1 mm, dorsoventral -2.5 mm) under anesthesia (0.3 mg/kg of medetomidine, 4.0 mg/kg of midazolam, and 5.0 mg/kg of butorphanol). Then a catheter tube was connected to an osmotic minipump flow moderator (model 2004, ALZET Osmotic Pumps). The minipump was inserted in a subcutaneous pocket on the dorsal surface of the mouse, and the incision was closed. In the third experiment, TUDCA was administered by drinking water to 11-month-old A7-Tg mice for 120 days. The concentration of TUDCA was increased gradually to avoid taste aversion (1 mg/mL for 60 days, 3 mg/mL for 7 days, 5 mg/mL for 14 days, 6.5 mg/mL for 39 days). For the last 31 days, TUDCA was administered intraperitoneally (250 mg/kg) twice per day, 6 days per week, in parallel.

Antibodies

The following antibodies were used: anti-phospho-eIF2α (3398, Cell Signaling Technology (CST)), anti- eIF2α (5324, CST), anti-phospho-JNK (4668, CST), anti-JNK (9258, CST), anti-phospho-PERK (3179, CST), anti-PERK (3192, CST), anti-BiP (610978, BD biosciences), anti-CHOP (ab11419, Abcam), anti-Human APP (C) (18961, IBL), anti-ADAM10 (ab124695, Abcam), anti-BACE1 (5606, CST), anti-α-tubulin (DM1A, Merck), anti-LC3B (3868, CST), anti-Beclin1 (612113, BD biosciences), anti-Sirt1 (2028, CST), anti-PGC-1α (SC-518025, Santa Cruz Biotechnology), anti-Aβ (82E1, IBL), peroxidase-conjugated AffiniPure anti-rabbit IgG (111-035-003, Jackson ImmunoResearch), and peroxidase-conjugated AffiniPure anti-Mouse IgG (515-035-003, Jackson ImmunoResearch).

Metabolic measurements

Blood glucose was measured using Glutest sensor (Sanwa Kagaku Kenkyusho Co., LTD.). For an insulin tolerance test (ITT), 3-hour-fasted mice were intraperitoneally injected with human insulin (Humulin R, Eli Lilly) at 0.75 U/Kg body weight, and blood glucose levels were measured every 20 min for 120 min.

Western blot analysis

Epididymal adipose tissues or periovarian adipose tissues were obtained and weighed. Brains were harvested, dissected into the hypothalamus, hippocampus, and cerebral cortex. These tissue samples were snap frozen in liquid nitrogen and stored at -80 °C until use. Tris-buffered saline (TBS)-soluble fractions were obtained as the supernatant from homogenizing tissues in a 1:10 (w/v) volume of TBS and centrifuging at 347,600 x g for 20 min at 4 °C. TBS-insoluble pellets were homogenized in a 1:10 (w/v) volume of 2% Triton X-100 (TX) in TBS and centrifuged at 347,600 x g for 20 min at 4 °C and supernatants were saved as TX-soluble fractions. RIPA-soluble fcations were obtained as the supernatant from homogenizing tissues in a 1:10 (w/v) volume of RIPA buffer (1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS) in TBS) and centrifuging at 17,800 x g for 30 min at 4 °C. Protein concentration was determined with BCA protein assay kit (Takara Bio Inc.). All the buffers were supplemented with cOmplete protease inhibitor and PhosSTOP phosphatase inhibitor cocktails (Merck). TBS-, TX-, and RIPA-soluble fractions were used for immunoblotting.
For immunoblotting, samples were separated by SDS-polyacrylamide gel electrophoresis under a reducing condition using a Tris-Glycine gel system, transferred to polyvinylidene difluoride membranes (Merck), and incubated with antibodies. The immunoblots were developed using ImmunoStar reagents (Wako) and SuperSignal (Thermo Fisher), and visualized by LAS-4000 mini (Fujifilm).

ELISA quantitation of Aβ

For the measurement of soluble and insoluble Aβ, the TBS-soluble and SDS-insoluble/formic acid-soluble fractions were used, respectively. For extraction of the insoluble fraction, brains were homogenized in a 1:10 (w/v) volume of RIPA buffer, centrifuged at 347,600 x g for 20 min at 4 °C. Resulting pellets were homogenized in a 1:10 (w/v) volume of 2% SDS in TBS, incubated for 30 min at 37 °C and centrifuged at 347,600 x g for 20 min at 20 °C. SDS-insoluble pellets were dissolved in 70% formic acid using a sonicator (Branson), centrifuged at 347,600 x g for 20 min at 4 °C and supernatants were desiccated by Speed-Vac followed by resuspension in dimethyl sulfoxide (DMSO). Levels of Aβ were quantitated by BNT77/BA27 or BNT77/BC05 Human/Rat β Amyloid ELISA kit (Wako). Prior to the measurement of soluble Aβ, an equal volume of 1 M guanidine hydrochloride was added and incubated for 30 min at room temperature.

Immunohistochemical analysis and morphometry

Mouse brains were fixed with 4% paraformaldehyde in PBS for 24 h, dehydrated, and embedded in paraffin. Serial sections were cut at 4-µm thickness. Deparaffinized sections were treated with microwave (700 W) in citrate buffer pH 6.0 for 20 min, followed by digestion with 100 µg/ml proteinase K (Worthington) in TBS for 6 min at 37 °C. After blocking by incubation with 10% calf serum in TBS, the sections were incubated with an anti-Aβ antibody 82E1 and then a biotinylated anti-mouse IgG antibody (Vector Laboratories), followed by visualization by avidin-biotin complex method (ABC elite, Vector Laboratories) using diaminobenzidine as chromogen. The percentage area covered by Aβ immunoreactivity in the parietal cortex/cingulate gyrus, hippocampus, and piriform cortex was measured using Image J software (NIH) as previously described (31).

Quantitative reverse transcription PCR

Total RNA was isolated using TRIzol Plus RNA Purification Kit and PureLink RNA Mini kit (Thermo Fisher). RNA purity and concentration were measured by NanoDrop (ThermoFisher). Total RNA was reverse-transcribed into cDNA using ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO). Real-time PCR was performed with LightCycler 480 system (Roche) using THUNDERBIRD SYBR qPCR Mix (TOYOBO). Threshold cycle values were normalized to Gapdh. The primer pairs used in this study are as follows: 5′- AACGACCCCTTCATTGAC -3′ and 5′- GAAGACACCAGTAGACTCCAC -3′ for Gapdh; 5′- AAGCTATTTCAGTCCCCAGTGG -3′ and 5′- AAGAGCAACCCGAACATGAC -3′ for Map1lc3a; 5′- GACGTGGAGAAAGGCAAGATTG -3′ and 5′- TTGAGCGCTTTTGTCCACTG -3′ for Becn1; 5′- TTGACCGATGGACTCCTCAC -3′ and 5′- AACAAAAGTATATGGACCTATCCGC -3′ for Sirt1.

Statistical analysis

Quantitative data were analyzed statistically by unpaired t test for two-group data, or one-way ANOVA followed by Tukey’s multiple comparisons test for three-group data using GraphPad Prism 7. In figures, all data are represented by mean ± SEM. Statistical significance is indicated by *p < 0.05, ** p < 0.01, and *** p < 0.001.

 

Results

Intraperitoneal administration of TUDCA attenuated diet-induced peripheral ER stress and improved systemic metabolic abnormalities in A7-Tg mice

To study the link between the peripheral metabolic state and brain AD pathology, we have used A7-Tg mice expressing human APP harboring the Swedish and Austrian mutations in neurons. A7-Tg mice develop progressive Aβ deposition in the brain starting at ~12 months of age, and we have previously shown that inducing obesity/diabetes by feeding HFD accelerates amyloid pathology in the brains of A7-Tg mice (11). Furthermore, HFD-induced acceleration of Aβ pathology was reversibly suppressed in dietary intervention experiments that switched from HFD to normal diet (ND) either starting at the age of 9 months (11) or 15 months (Figure S1), as the systemic metabolic abnormalities were improved. These results indicate that metabolic improvement is effective in reversing the HFD-induced Aβ deposition.
We investigated the effects of mitigating ER stress on metabolic state and consequently amyloid pathology using a chemical chaperone TUDCA, which has been shown to be effective in improving HFD-induced metabolic abnormalities (25). TUDCA was intraperitoneally administered to HFD-fed A7-Tg mice for 30 days starting at 8 months of age (Figure 1a). HFD feeding increased body weights and wet weights of adipose tissues compared to normal diet (ND) feeding, whereas TUDCA treatment significantly alleviated HFD-induced obesity in A7-Tg mice (Figure 1b-d). HFD-fed A7-Tg mice exhibited significant hyperglycemia, which was associated with a decreased insulin sensitivity as shown by the insulin tolerance test (ITT) (Figure 1e-g). Administration of TUDCA also improved the impaired glucose metabolism to a level comparable to that of ND-fed mice (Figure 1e-g). These data suggest that TUDCA could ameliorate diet-induced obesity and diabetes-like metabolic abnormalities in A7-Tg mice.

Figure 1. Intraperitoneal administration of TUDCA improved diet-induced metabolic abnormalities and reduced peripheral ER stress

(a) Schematic diagram of the study design. Male A7-Tg mice were fed with normal diet (ND) or high-fat diet (HFD) from 3 months of age, and TUDCA (250 mg/kg) or saline (vehicle) was administered intraperitoneally twice per day for 30 days from 8 months of age. (b) The time course of body weight changes during TUDCA treatment. (c-d) Body weight (c) and adipose tissue wet weight (d) on day 31 after starting TUDCA treatment. (e-g) Changes in the blood glucose levels (e) and area under the curve (AUC) (f) during the insulin tolerance test (ITT) on day 15 or 16, and blood glucose levels before the test (g). (h-i) Immunoblot analyses of phosphorylation levels of eIF2α and JNK in the RIPA-soluble fractions of liver (h) and adipose tissue (i) (upper panels). The ratios of phosphorylation to total protein content were measured by densitometry (lower panels). Data are mean ± SEM (ND: n = 8, HFD: n = 5, HFD-TUDCA: n = 6). *, p < 0.05; **, p < 0.01; ***, p < 0.001, one-way ANOVA with Tukey’s post-hoc test.

 

Previous studies have shown that TUDCA treatment ameliorates insulin resistance and abnormal glucose metabolism via the reduction of ER stress in the liver and adipose tissue of obese mice (25, 28). We therefore examined the effects of HFD feeding and TUDCA treatment on UPR in the liver and adipose tissues in A7-Tg mice. HFD feeding induced phosphorylation of eIF2α and JNK in the liver, which was significantly decreased by TUDCA treatment, in HFD-fed A7-Tg mice (Figure 1h). In the adipose tissue, phospho-eIF2α showed an increasing trend with HFD feeding, which was not inhibited by TUDCA treatment (Figure 1i). On the other hand, phospho-JNK was markedly increased by HFD and decreased significantly by TUDCA treatment, as in the liver (Figure 1i).

Intraperitoneal administration of TUDCA prevented the HFD-induced exacerbation of Aβ accumulation in brains

We next examined the effects of TUDCA on HFD-induced exacerbation of Aβ pathology in the cerebral cortices of A7-Tg mice. As we previously reported (11), HFD-feeding significantly increased the levels of both Aβ40 and Aβ42 at 9 months of age (Figure 2a). Administration of TUDCA reduced the Aβ levels in the HFD group to a similar level to that of control ND group, indicating that the effect of metabolic overload on Aβ pathology was totally reversed by TUDCA (Figure 2a). The levels of APP fragments were not altered by HFD feeding, whereas the levels of CTFβ, a carboxy-terminal fragment produced upon β-cleavage, were decreased by TUDCA treatment (Figure 2b). Although the protein levels of ADAM10 and BACE1, corresponding to α- and β-secretases, respectively, did not change in any of the experimental groups (Figure 2c), the reduced levels of CTFβ might be related to the inhibitory effect of TUDCA on Aβ production.

Figure 2. Intraperitoneal administration of TUDCA prevented the HFD-induced increase of brain Aβ accumulation

(a) The levels of TBS-soluble Aβ40 and Aβ42 in the cerebral cortices of 9-month-old A7-Tg mice in the three experimental groups indicated in Figure 1a (ND, HFD, and HFD-TUDCA) were analyzed by ELISA. (b) Immunoblot analyses of full-length APP (APP FL), APP-CTFα, APP-CTFβ, and α-tubulin in the TX-soluble fractions of cerebral cortex. The lower graph shows the results of densitometry. The amount of protein was expressed as a relative value to the ND group. (c) Immunoblot and densitometric analyses of ADAM10 and BACE1 in the TX-soluble fraction of cerebral cortex. (d) Immunoblot analyses of peIF2α, pJNK, total eIF2α, total JNK, Grp78/BiP, CHOP, and α-tubulin in the TBS-soluble fractions of cerebral cortex (left panels). The right panel shows the results of densitometry. The levels of peIF2α and pJNK were normalized to total eIF2α and JNK, respectively; those of Grp78/BiP and CHOP were normalized to α-tubulin. Data are mean ± SEM (ND: n = 8, HFD: n = 5, HFD-TUDCA: n = 6). *, p < 0.05; **, p < 0.01; ***, p < 0.001, one-way ANOVA with Tukey’s post-hoc test.

 

To investigate the relationship between changes in Aβ levels and ER stress in the brain, we analyzed the expression of ER stress marker proteins phospho-eIF2α, phospho-JNK, Grp78/Bip, and CHOP in the cerebral cortex. In contrast to the peripheral tissues, no increase in the expression of ER stress marker proteins was observed upon HFD feeding in the cerebral cortices (Figure 2d), and TUDCA treatment did not alter the levels of any of these proteins (Figure 2d). We also examined the mRNA expression of TNFα in the hippocampus to investigate the possibility that TUDCA treatment affects inflammatory signals in the brain. The results showed that there was no significant difference in relative expression levels among the ND (1.000 ± 0.151), HFD (0.866 ± 0.115), and HFD-TUDCA (1.070 ± 0.176) groups. These results suggest that metabolic improvement through reduction of the peripheral ER stress by TUDCA may prevent HFD-induced exacerbation of Aβ pathology in brains.

Intracerebroventricular administration of TUDCA had no effect on the HFD-induced increase in Aβ levels in the cerebral cortex of A7-Tg mice

Previous studies have shown that TUDCA is able to cross the blood-brain barrier and exert the effects on the central nervous system tissues (32, 33). This suggests that the effects of intraperitoneally administered TUDCA on Aβ pathogenesis may be due to either an indirect effect of TUDCA in the periphery or a direct effect of TUDCA translocated to the brain. To examine the latter possibility, we directly administered TUDCA into the cerebral ventricules (i.c.v.) of HFD-fed A7-Tg mice for 28 days (Figure 3a).
Intraventricular administration of TUDCA decreased the blood glucose levels, but did not improve the HFD-induced obesity (Figure 3b-d). No change in UPR of the liver or adipose tissue of HFD-fed A7-Tg mice were observed (Figure 3e-f). In addition, UPR in the cerebral cortex also was not altered by intraventricular administration of TUDCA (Figure 3g). To confirm whether TUDCA reached effective concentrations in the brain, we evaluated the ER stress markers in the hypothalamus and found that the levels of phospho-JNK and Grp78/BiP were significantly reduced (Figure S2). Given that HFD-induced hypothalamic ER stress is known to affect the peripheral metabolic state (26, 34), this reduction of ER stress by intraventricular administration of TUDCA might have resulted in a decrease in blood glucose levels. Under these conditions, intraventricular administration of TUDCA did not reduce the levels of Aβ40 and Aβ42 in HFD-fed A7-Tg mice (Figure 3h). Taken together with the fact that HFD feeding did not enhance ER stress in the brain (Figure 2d), we reasoned that the central effect of TUDCA did not contribute much to the suppression of Aβ levels in the cerebral cortex of HFD-fed A7-Tg mice, and speculated that the effect on ER stress in the periphery was more important.

Figure 3. Intracerebroventricular administration of TUDCA did not affect ER stress and Aβ levels in HFD-fed A7-Tg mice

(a) Schematic diagram of the study design. Female A7-Tg mice were fed with HFD from 3 months of age, and TUDCA (10 µg/day) or PBS (vehicle) was intracerebroventricularly administered using ALZET Osmotic Pumps for 28 days from 7-8 months of age. (b-d) Body weight (b), adipose tissue wet weight (c), and blood glucose levels (d) on day 28 after starting TUDCA treatment. (e-f) Immunoblot and densitomeric analyses of phosphorylation levels of eIF2α and JNK in the RIPA-soluble fractions of liver (e) and adipose tissue (f). (g) Immunoblot and densitomeric analyses of phosphorylation levels of eIF2α and JNK, and the levels of Grp78/BiP, CHOP in the TBS-soluble fractions of cerebral cortex. The results were analyzed as in Figure 2d. (h) The levels of TBS-soluble Aβ40 and Aβ42 in the cerebral cortices at 9 months of age were analyzed by ELISA. Data are mean ± SEM (HFD: n = 8, HFD-TUDCA: n = 9). *, p < 0.05, unpaired t-test.

 

Peripheral administration of TUDCA improved age-related ER stress and metabolic abnormalities in A7-Tg mice

Both in humans and animal models, increasing adiposity and insulin resistance have been documented as the characteristics of aging. Furthermore, age-related decline in the UPR has also been suggested (35). We therefore wondered whether aging-related ER stress and the associated metabolic abnormalities might have contributed to the aggravation of amyloid pathology in the brain. To test this hypothesis, we examined the effects of TUDCA in A7-Tg mice raised on normal diet. Because A7-Tg mice require a long period of time, i.e., >12 months, by the accumulation of amyloid plaques, a long-term administration method was adopted. We administered TUDCA orally (1-6.5 mg/ml in drinking water) starting at 11 months of age and intraperitoneally in parallel starting at 14 months of age, and analyzed the mice at 15 months (Figure 4a).
TUDCA treatment significantly decreased body weight and wet weight of adipose tissue in these aged mice (Figure 4b-c). Insulin sensitivity also was improved as demonstrated by the ITT, but blood glucose levels were not significantly decreased by TUDCA treatment (Figure 4d-f). Notably, the metabolic status of TUDCA-treated 15-month-old A7-Tg mice (body weight: 33.6 ± 1.0 g, adipose tissue wet weight: 0.73 ± 0.06 g, area under the curve (AUC) for ITT: 8691 ± 374 mg/dl) was improved to levels equivalent to those in 9-month-old ND-fed A7-Tg mice (body weight: 31.8 ± 1.1 g, adipose tissue wet weight: 0.76 ± 0.10 g, AUC for ITT: 9755 ± 599 mg/dl, see Figure 1).
We next examined the ER stress in the peripheral tissues of these mice. In contrast to the results of HFD-fed A7-Tg mice (Figure 1), TUDCA treatment on 15-month-old A7-Tg mice did not alter the expression of UPR markers in the liver (Figure 4g). In contrast, the levels of phospho-JNK were significantly decreased, and phospho-eIF2α also showed a decreasing trend in the adipose tissue (Figure 4h). These results suggest that ER stress in adipose tissue, rather than that in liver, contributed to the age-related metabolic abnormalities, which can be ameliorated by TUDCA treatment.

Figure 4. Peripheral administration of TUDCA improved age-related metabolic abnormalities and reduced peripheral ER stress

(a) Schematic diagram of the study design. TUDCA was administered to 11-month-old male A7-Tg mice by drinking water for 120 days with a gradual increase in concentration to avoid taste aversion (1 mg/ml for 60 days, 3 mg/ml for 7 days, 5 mg/ml for 14 days, 6.5 mg/ml for 39 days). For the last 31 days, TUDCA (250 mg/kg) or saline was administered intraperitoneally twice per day, 6 days per week, in parallel. (b-c) Body weight (b) and adipose tissue wet weight (c) on day 120 after starting TUDCA treatment. (d-f) Changes in the blood glucose levels (d) and area under the curve (AUC) (e) during ITT on day 97 or 100, and blood glucose levels before the test (f). (e-f) Immunoblot and densitometric analyses of phosphorylation levels of eIF2α and JNK in the RIPA-soluble fractions of liver (g) and adipose tissue (h). Data are mean ± SEM (ND: n = 8, ND-TUDCA: n = 7). *, p < 0.05; **, p < 0.01; ***, p < 0.001, unpaired t-test.

 

Peripheral administration of TUDCA reduced amyloid deposition in the brain of 15-month-old A7-Tg mice

We then evaluated the effects of TUDCA on age-dependent amyloid accumulation in the brains of 15-month-old A7-Tg mice. Biochemical analyses revealed that the levels of insoluble Aβ40 and Aβ42 in the cerebral cortex were significantly decreased by TUDCA treatment (Figure 5a). Immunohistochemical analyses showed that Aβ plaques were significantly reduced in the piriform and cerebral cortices (Figure 5b-c). Aβ deposition in the hippocampus also showed a tendency to decrease (Figure 5d). Analyses of proteins related to Aβ production in the brains of this experimental group showed that TUDCA treatment caused a downward trend in the level of CTFβ, but no significant change was observed (Figure S3a). The UPR activities of the cerebral cortex was not changed by TUDCA treatment (Figure 5e). Overall, the reduction of peripheral ER stress and improvement of metabolism during aging by TUDCA treatment also attenuated the formation of amyloid pathology in the brain.

Figure 5. Peripheral administration of TUDCA reduced amyloid deposition in 15-month-old A7-Tg mice

(a) The levels of SDS-insoluble/formic acid-soluble Aβ40 and Aβ42 in the cerebral cortices of 15-month-old A7-Tg mice. (b-d) Immunohistochemical analyses of 15-month-old A7-Tg mouse brains using an anti-Aβ (82E1) antibody. Representative images of brain regions including the piriform cortex (b), parietal cortex and cingulate gyrus (c), and hippocampus (d) are shown. The accompanying graphs represent the quantitative results of amyloid deposition in percentage of the area covered by Aβ immunoreactivity. (e) Immunoblot and densitomeric analyses of phosphorylation levels of eIF2α and JNK, and the levels of Grp78/BiP, CHOP in the RIPA-soluble fractions of cerebral cortex. Data are mean ± SEM (ND: n = 8, ND-TUDCA: n = 7). *, p < 0.05; ***, p < 0.001, unpaired t-test. Scale bars: 1.0 mm.

 

TUDCA treatment caused dietary restriction-like expression changes both in the peripheral tissues and brain

The effects of TUDCA observed in this present study, i.e., inhibition of metabolic abnormalities and amyloid pathology associated with obesity/T2DM and aging, were similar to those caused by dietary restriction (11–13). Because dietary restriction has been documented to ameliorate the age-related abnormalities in several species through increased sirtuin expression and authophagy (36, 37), we investigated the expression of Sirt1 and genes related to autophagy, e.g. Map1lc3a and Becn1, in 15-months-old A7-Tg mice.
In the liver, peripheral TUDCA treatment showed a tendency to increase Sirt1 mRNA (Figure S3b). In adipose tissue, mRNA expression of Becn1 and Sirt1 was increased, and Map1lc3a showed an increasing trend (Figure S3c). In the hippocampus, TUDCA treatment increased the mRNA expression levels of Map1lc3a, Becn1, and Sirt1 (Figure S3d). Furthermore, protein expression analysis showed that the LC3B-II/LC3B-I ratio, which indicates activation of autophagy, and the levels of Sirt1 and its substrate, PGC-1α (Figure S3e) were increased in the cerebral cortex of 15-month-old A7-Tg mice. Taken together, we reasoned that reducing ER stress in the peripheral tissues by systemic administration of TUDCA has effects that mimic dietary restriction both on the peripheral tissues and the brain, which in turn might reduce brain Aβ accumulation.

 

Discussion

Besides causing metabolic disturbances systemically, HFD feeding has been shown to promote Aβ accumulation in the brain in mouse models of AD (9, 11, 12). Here, we showed for the first time that intraperitoneal administration of TUDCA ameliorates not only the metabolic abnormalities caused by HFD, but also the concomitant increase in brain Aβ pathology, in A7-Tg mice. The most probable mechanism of TUDCA action would be that its chaperone activity reduced ER stress in the peripheral tissues, thereby ameliorating metabolic abnormalities such as obesity and insulin resistance, and consequently alleviating brain amyloid pathology. This hypothesis is consistent with the previous observations that Aβ accumulation is reversibly suppressed when HFD-dependent peripheral metabolic abnormalities are improved by dietary switching (11–13). Importantly, we showed that even under continuous HFD feeding, TUDCA treatment improved metabolism and lowered Aβ to a level similar to that in the ND diet group.
Since TUDCA is a brain-penetrating compound that has been shown to be neuroprotective through its anti-apoptotic effects (33, 38), another possibility for the mechanism of action is that central ER stress may have been targeted to reduce Aβ levels. Some studies in human AD postmortem brains have reported findings suggestive of increased ER stress, and the link between Aβ-induced toxicity and ER stress has been reported in various experimental models (39, 40). Thus, central ER stress may be augmented as a feedback mechanism for the pathological progression of AD, which may further contribute to the exacerbation of AD pathology. Furthermore, metabolic overload has been suggested to increase brain ER stress, especially in the hypothalamus, the latter being exposed to peripheral circulation (26, 34). Several studies have reported HFD-dependent UPR enhancement in brain regions including the hippocampus and cerebral cortex (41, 42). In our experiment, HFD feeding increased ER stress in the liver and adipose tissue, which was lowered by TUDCA. However, neither HFD feeding nor TUDCA treatment altered the UPR signaling activities in the cerebral cortex. Our results may have differed from those previously reported due to multiple factors, e.g., differences in the nutritional composition of the diet, the rearing environment, and the rate of pathological progression that varies among model animals. Nevertheless, our data may suggest that the cerebral cortex is less prone to elevated ER stress due to metabolic overload. Furthermore, even in conditions where the UPR activity was not increased, the Aβ pathology was exacerbated by HFD feeding. Thus, it may be reasonable to speculate that cortical ER stress is not the major culprit for the HFD-induced Aβ accumulation. Moreover, the finding that central administration of TUDCA did not alter brain Aβ levels supports the lack of interdependence between ER stress in the brain and HFD-induced Aβ accumulation. Although the UPR in the hypothalamus was significantly reduced, it cannot be ruled out that TUDCA diffused into the cerebral cortex in this administration paradigm may not have reached a sufficient concentration to exert an effect on AD pathology.
Aging is the greatest risk factor of AD. In our study, TUDCA ameliorated not only the diet-dependent but also the age-related metabolic deterioration, and suppressed amyloid accumulation. In 15-month-old A7-Tg mice, unlike HFD-fed A7-Tg mice, TUDCA did not alter the expression of UPR marker proteins in the liver, but reduced the levels of phospho-JNK and phspho-eIF2α in adipose tissues. It has been documented that diet-induced obese mice had a greater infiltration of macrophages and more pro-inflammatory immune cells in the liver than in aged obese mice, whereas adipose tissues showed similar levels of cytokine changes (43), which is consistent with our findings, prompting us to speculate that suppression of stress in the adipose tissue may have ameliorated the age-related, systemic metabolic deterioration.
In a series of studies using APP/PS1 mice, Rodrigues et al. have shown that treatment with TUDCA reduced Aβ production, suppressed amyloid accumulation, and prevented cognitive impairments (44–46). However, a reduction in amyloid deposits by TUDCA has been observed in APP/PS1 mice at a young age, without significant changes in body weight (45). Furthermore, it has previously been reported that intraperitoneal administration of TUDCA to young lean mice did not affect peripheral metabolic parameters (25). These results do not support the hypothesis that the inhibitory effect of TUDCA on Aβ accumulation is due to amelioration of the systemic metabolic abnormalities. However, it should be noted that ER stress might have been upregulated only in the brains of AD model mice that overexpress presenilin 1 together with APP, including APP/PS1 mice, raising serious concerns about its use in the study of ER stress (47, 48). Accordingly, upregulation of ER stress has not been described in other AD model mice, e.g. Tg2576, APP23, and AppNL-G-F mice (48–51). Thus, the effect of TUDCA on amyloid reduction in APP/PS1 mice might be attributable, at least in part, to the reduction of central ER stress. Future studies on the effects of TUDCA in young A7-Tg mice will better address these issues.
We found that the ratio of LC3BII/LC3BI and the expression of Sirt1 and PGC-1α are factors that are altered both in the periphery and brain upon TUDCA-treatment in A7-Tg mice. These molecules are involved in the molecular pathways that play an important role in the dietary restriction that delays the onset of many chronic diseases such as obesity, diabetes, and AD, as well as the anti-aging effects of its mimetic, resveratrol (52, 53). Sirt1 activation, or acetylation of PGC-1α by Sirt1, has been suggested to alter Aβ levels in the brain (54–56). In addition, increased autophagy has been suggested to reduce Aβ levels (57). This suggests that TUDCA may regulate Aβ pathology by suppressing ER stress in the periphery, thereby exerting a dietary restriction-like effect. Which of the processes underlying Aβ accumulation, i.e., production, clearance or aggregation, were altered by TUDCA is yet to be clarified. The levels of CTFβ, an indicator of β-secretase activity, tended to be decreased by TUDCA in HFD-fed A7-Tg mice, suggesting that reduced Aβ production may contribute at least in part to the anti-amyloid effect. On the other hand, we previously showed that HFD feeding decreases the clearance of Aβ in the brain interstitial fluid (11). Since TUDCA ameliorated the adverse effects of HFD on the systemic metabolism, it is also possible that Aβ clearance was improved accordingly. Further studies are needed to elucidate the molecular link between the peripheral metabolism and signal changes in the brain, and brain Aβ dynamics.
Overall, our results suggest a therapeutic potential of TUDCA in suppressing obesity/diabetes-induced Aβ accumulation. Recent progress in imaging biomarker research has revealed that amyloid accumulation occurs in the brains of AD patients decades before the onset of dementia (59). Thus, preclinical stage of pathological progression may be a critical period for disease prevention. Our results showed that interventions in the periphery at the early stage of AD pathology formation may be effective in counteracting Aβ accumulation in the brain. Considering the biosafety and its ability to inhibit apoptosis and neuroinflammation (38), TUDCA is expected to be a multifaceted prevention strategy of AD.

 

Funding: This work was supported by AMED under Grant Number JP20dm0107056, and JSPS KAKENHI Grant Number JP20H00525.

Declaration of Competing Interest: T.O. is an employee of Kaken Pharmaceutical Co., LTD. All other authors declare no conflict of interests.

 

SUPPLEMENTARY MATERIAL

 

References

1. Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396, 413–446
2. Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer’s disease: an analysis of population-based data. Lancet Neurol. 2014;13, 788–794
3. Barnes DE, Yaffe K. The projected effect of risk factor reduction on Alzheimer’s disease prevalence. Lancet Neurol. 2011;10, 819–828
4. Gudala K, Bansal D, Schifano F, Bhansali A. (2013) Diabetes mellitus and risk of dementia: A meta-analysis of prospective observational studies. J Diabetes Investig. 2013;4, 640–650
5. Zhang J, Chen C, Hua S, Liao H, Wang M, Xiong Y, Cao F. (2017) An updated meta-analysis of cohort studies: Diabetes and risk of Alzheimer’s disease. Diabetes Res Clin Pract. 124, 41-47
6. Kandimalla R, Thirumala V, Reddy PH. Is Alzheimer’s disease a Type 3 Diabetes? A critical appraisal. Biochim Biophys Acta Mol Basis Dis. 2017;1863, 1078–1089
7. Biessels GJ, Despa F. Cognitive decline and dementia in diabetes mellitus: mechanisms and clinical implications. Nat Rev Endocrinol. 2018;14, 591–604
8. Matsuzaki T, Sasaki K, Tanizaki Y, et al. Insulin resistance is associated with the pathology of Alzheimer disease: The Hisayama Study. Neurology. 2010;75, 764–770
9. Ho L, Qin W, Pompl PN, et al. Diet-induced insulin resistance promotes amyloidosis in a transgenic mouse model of Alzheimer’s disease. FASEB J. 2004;18, 902–904
10. Cao D, Lu H, Lewis TL, Li L. Intake of sucrose-sweetened water induces insulin resistance and exacerbates memory deficits and amyloidosis in a transgenic mouse model of Alzheimer disease. J Biol Chem. 2007;282, 36275–36282
11. Wakabayashi T, Yamaguchi K, Matsui K, et al. Differential effects of diet- and genetically-induced brain insulin resistance on amyloid pathology in a mouse model of Alzheimer’s disease. Mol Neurodegener. 2019;14, 15
12. Maesako M, Uemura K, Kubota M, et al. Exercise is more effective than diet control in preventing high fat diet-induced β-amyloid deposition and memory deficit in amyloid precursor protein transgenic mice. J Biol Chem. 2012;287, 23024–23033
13. Walker JM, Dixit S, Saulsberry AC, May JM, Harrison FE. Reversal of high fat diet-induced obesity improves glucose tolerance, inflammatory response, β-amyloid accumulation and cognitive decline in the APP/PSEN1 mouse model of Alzheimer’s disease. Neurobiol Dis. 2017;100, 87–98
14. Selkoe DJ, Hardy J. The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol Med. 2016;8, 595–608
15. Villalobos-Labra R, Subiabre M, Toledo F, Pardo F, Sobrevia L. Endoplasmic reticulum stress and development of insulin resistance in adipose, skeletal, liver, and foetoplacental tissue in diabesity. Mol Aspects of Med. 2019;66, 49–61
16. Walter P, Ron D. The unfolded protein response: from stress pathway to homeostatic regulation. Science. 2011;334, 1081–1086
17. Hirosumi J, Tuncman G, Chang L, et al. (2002) A central role for JNK in obesity and insulin resistance. Nature. 2002;420, 333–336
18. Özcan U, Cao Q, Yilmaz E, et al. Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes. Science. 2004;306, 457–461
19. Hoozemans JJM, Veerhuis R, Van Haastert ES, et al. The unfolded protein response is activated in Alzheimer’s disease. Acta Neuropathol. 2005;110, 165–172
20. Hoozemans JJM, van Haastert ES, Nijholt DAT, et al. The unfolded protein response is activated in pretangle neurons in Alzheimer’s disease hippocampus. Am J Pathol. 2009;174, 1241–1251
21. Stutzbach LD, Xie SX, Naj AC, et al. The unfolded protein response is activated in disease-affected brain regions in progressive supranuclear palsy and Alzheimer’s disease. Acta Neuropathol Commun. 2013;1, 31
22. Yoon SO, Park DJ, Ryu JC, et al. JNK3 perpetuates metabolic stress induced by Aβ peptides. Neuron. 2012;75, 824–837
23. Duran-Aniotz C, Martinez G, Hetz C. Memory loss in Alzheimer’s disease: are the alterations in the UPR network involved in the cognitive impairment? Front. Aging Neurosci. 2014;6, 8
24. Gerakis Y, Hetz C. Emerging roles of ER stress in the etiology and pathogenesis of Alzheimer’s disease. FEBS J. 2018;285, 995–1011
25. Özcan U, Yilmaz E, Özcan L, et al. Chemical chaperones reduce ER stress and restore glucose homeostasis in a mouse model of type 2 diabetes. Science. 2006;313, 1137–1140
26. Ozcan L, Ergin AS, Lu A, et al. Endoplasmic reticulum stress plays a central role in development of leptin resistance. Cell Metab. 2009;9, 35–51
27. Vettorazzi JF, Kurauti MA, Soares GM, et al. Bile acid TUDCA improves insulin clearance by increasing the expression of insulin-degrading enzyme in the liver of obese mice. Sci Rep. 2017;7, 14876
28. Zhang Z, Wang X, Zheng G, et al. Troxerutin attenuates enhancement of hepatic gluconeogenesis by inhibiting NOD activation-mediated inflammation in high-fat diet-treated mice. Int J Mol Sci. 2017;18, 31
29. DeVos SL, Miller TM. Direct intraventricular delivery of drugs to the rodent central nervous system. J Vis Exp. 2013;e50326
30. Contreras C, González-García I, Seoane-Collazo P, et al. Reduction of hypothalamic endoplasmic reticulum stress activates browning of white fat and ameliorates obesity. Diabetes. 2017;66, 87–99
31. Yamamoto K, Tanei Z, Hashimoto T, et al. Chronic optogenetic activation augments Aβ pathology in a mouse model of Alzheimer disease. Cell Rep. 2015;11, 859–865
32. Kaemmerer WF, Rodrigues CMP, Steer CJ, Low WC. Creatine-supplemented diet extends Purkinje cell survival in spinocerebellar ataxia type 1 transgenic mice but does not prevent the ataxic phenotype. Neuroscience. 2001;103, 713–724
33. Keene CD, Rodrigues CMP, Eich T, et al. Tauroursodeoxycholic acid, a bile acid, is neuroprotective in a transgenic animal model of Huntington’s disease. Proc Natl Acad Sci U S A. 2002;99, 10671–10676
34. Zhang X, Zhang G, Zhang H, et al. Hypothalamic IKKβ/NF-κB and ER stress link overnutrition to energy imbalance and obesity. Cell. 2008;135, 61–73
35. Naidoo N, Brown M. The endoplasmic reticulum stress response in aging and age-related diseases. Front Physiol. 2012;3, 263
36. Fontana L, Partridge L. Promoting health and longevity through diet: from model organisms to humans. Cell. 2015;161, 106–118
37. Van Cauwenberghe C, Vandendriessche C, Libert C, Vandenbroucke RE. Caloric restriction: beneficial effects on brain aging and Alzheimer’s disease. Mamm Genome. 2016;27, 300–319
38. Zangerolamo L, Vettorazzi JF, Rosa LRO, Carneiro EM, Barbosa HCL. The bile acid TUDCA and neurodegenerative disorders: An overview. Life Sci. 2021;272, 119252
39. Uddin MS, Tewari D, Sharma G, et al. Molecular mechanisms of ER stress and UPR in the pathogenesis of Alzheimer’s disease. Mol Neurobiol. 2020;57, 2902–2919
40. Halliday M, Mallucci GR. Review: Modulating the unfolded protein response to prevent neurodegeneration and enhance memory. Neuropathol Appl Neurobiol. 2015;41, 414–427
41. Liang L, Chen J, Zhan L, et al. Endoplasmic reticulum stress impairs insulin receptor signaling in the brains of obese rats. PLoS One. 2015;10, e0126384
42. Binayi F, Zardooz H, Ghasemi R, et al. The chemical chaperon 4-phenyl butyric acid restored high-fat diet- induced hippocampal insulin content and insulin receptor level reduction along with spatial learning and memory deficits in male rats. Physiol Behav. 2021;231, 113312
43. Krishna KB, Stefanovic-Racic M, Dedousis N, Sipula I, O’Doherty RM. Similar degrees of obesity induced by diet or aging cause strikingly different immunologic and metabolic outcomes. Physiol Rep. 2016;4, e12708
44. Nunes AF, Amaral JD, Lo AC, et al. TUDCA, a bile acid, attenuates amyloid precursor protein processing and amyloid-β deposition in APP/PS1 mice. Mol Neurobiol. 2012;45, 440–454
45. Lo AC, Callaerts-Vegh Z, Nunes AF, Rodrigues CMP, D’Hooge R. Tauroursodeoxycholic acid (TUDCA) supplementation prevents cognitive impairment and amyloid deposition in APP/PS1 mice. Neurobiol Dis. 2013;50, 21–29
46. Dionísio PA, Amaral JD, Ribeiro MF, et al. Amyloid-β pathology is attenuated by tauroursodeoxycholic acid treatment in APP/PS1 mice after disease onset. Neurobiol Aging. 2015;36, 228–240
47. Hashimoto S, Ishii A, Kamano N, et al. Endoplasmic reticulum stress responses in mouse models of Alzheimer’s disease: Overexpression paradigm versus knockin paradigm. J Biol Chem. 2018;293, 3118–3125
48. Hashimoto S, Saido TC. Critical review: involvement of endoplasmic reticulum stress in the aetiology of Alzheimer’s disease. Open Biol. 8, 180024
49. Lee JH, Won SM, Suh J, et al. Induction of the unfolded protein response and cell death pathway in Alzheimer’s disease, but not in aged Tg2576 mice. Exp Mol Med. 2010;42, 386–394
50. Ma T, Trinh MA, Wexler AJ, et al. Suppression of eIF2α kinases alleviates Alzheimer’s disease–related plasticity and memory deficits. Nat Neurosci. 2013;16, 1299–1305
51. Barbero-Camps E, Fernández A, Baulies A, et al. Endoplasmic reticulum stress mediates amyloid β neurotoxicity via mitochondrial cholesterol trafficking. Am J Pathol. 2014;184, 2066–2081
52. Guarente L. Calorie restriction and sirtuins revisited. Genes Dev. 2013;27, 2072–2085
53. Madeo F, Zimmermann A, Maiuri MC, Kroemer G. Essential role for autophagy in life span extension. J Clin Invest. 2015;125, 85–93
54. Katsouri L, Parr C, Bogdanovic N, Willem M, Sastre M. PPARγ co-activator-1α (PGC-1α) reduces amyloid-β generation through a PPARγ-dependent mechanism. J Alzheimers Dis. 2011;25, 151–162
55. Wang R, Li JJ, Diao S, et al. Metabolic stress modulates Alzheimer’s β-secretase gene transcription via SIRT1-PPARγ-PGC-1 in neurons. Cell Metab. 2013;17, 685–694
56. Qin W, Yang T, Ho L, et al. Neuronal SIRT1 activation as a novel mechanism underlying the prevention of Alzheimer disease amyloid neuropathology by calorie restriction. J Biol Chem. 2006;281, 21745–21754
57. Menzies FM, Fleming A, Caricasole A, et al. Autophagy and neurodegeneration: pathogenic mechanisms and therapeutic opportunities. Neuron. 2017; 93, 1015–1034
58. Pluvinage JV, Wyss-Coray T. Systemic factors as mediators of brain homeostasis, ageing and neurodegeneration. Nat Rev Neurosci. 2020;21, 93–102
59. Jack CR, Knopman DS, Jagust WJ, et al. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol. 2010;9, 119–128

PERIPHERAL BLOOD BRCA1 METHYLATION POSITIVELY CORRELATES WITH MAJOR ALZHEIMER’S DISEASE RISK FACTORS

 

T. Mano1, K. Sato1, T. Ikeuchi2, T. Toda1, T. Iwatsubo3, A. Iwata4, Japanese Alzheimer’s Disease Neuroimaging Initiative5

 

1. Department of Neurology, Graduate School of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo, Japan; 2. Department of Molecular Genetics, Brain Research Institute, Niigata University, Chuo-ku, Niigata, Japan; 3. Department of Neuropathology, Graduate School of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo, Japan; 4. Tokyo Metropolitan Geriatric Hospital and Institute of Gerontology, Itabashi-ku, Tokyo, Japan; 5. The full list of members of the Japanese Alzheimer’s Disease Neuroimaging Initiative is provided in the supplementary file, «J-ADNI co-investigators».

Corresponding Author: Tatsuo Mano, Department of Neurology, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo Bunkyo-ku, Tokyo 113-8655, Japan, Email: tatsuomano@me.com, Phone +81-3-5800-8672, Fax +81-3-5800-6548

J Prev Alz Dis 2021;4(8):477-482
Published online June 9, 2021, http://dx.doi.org/10.14283/jpad.2021.31

 


Abstract

BACKGROUND: Recent biomarker studies demonstrated that the central nervous system (CNS) environment can be observed from peripherally-derived samples. In a previous study, we demonstrated significant hypomethylation of the BRCA1 promoter region in neuronal cells from post-mortem brains of Alzheimer’s disease patients through neuron-specific methylome analysis. Thus, we investigate the methylation changes in the BRCA1 promoter region in the blood samples.
OBJECTIVES: To analyze the methylation level of the BRCA1 promoter in peripheral blood from AD patients and normal controls.
DESIGN, SETTING, PARTICIPANTS: Genomic DNA samples from peripheral blood were obtained from the J-ADNI repository, and their biomarker data were obtained J-ADNI from the National Bioscience Database Center. Genomic DNA samples from an independent cohort for validation was obtained from Niigata University Hospital (Niigata, Japan). Amyloid positivity was defied by visual inspection of amyloid PET or a CSF Aβ42 value ≤ 333 pg/mL at the baseline.
MEASUREMENTS: Methylation level of the BRCA1 promoter was analyzed by pyrosequencing.
RESULTS: Compared to normal controls, methylation of the BRCA1 promoter in AD patients was not significantly changed; however, in AD patients, it showed a positive correlation with AD risk factors.
CONCLUSIONS: Our data confirmed the importance of cell-type specific methylome analysis and also suggested that environmental changes in the CNS can be detected by observing the peripheral blood, implying that the peripheral BRCA1 methylation level can be a surrogate for AD.

Key words: Alzheimer’s disease, methylome, BRCA1, peripheral blood.

Abbreviations: AD: Alzheimer’s disease; Aβ: amyloid β; CNS: central nervous system; NC: normal control; NFT: neurofibrillary tangle; PET: positron emission tomography; CSF: cerebrospinal fluid.


 

 

Introduction

BRCA1 is a nuclear DNA repair protein, and its loss-of-function causes breast and ovarian cancers. In familial cases, insufficient DNA repair caused by loss-of-function mutations leads to genomic instability and contributes to cancer development (20). In sporadic breast cancers, hypermethylation of BRCA1 leads to its downregulation and results in insufficient DNA repair, causing cancer (1, 19). Thus, proper BRCA1 function is thought to be crucial in maintaining genomic DNA homeostasis.
In our previous study, aberrant hypomethylation of the BRCA1 promoter was observed in brains from Alzheimer’s disease (AD) patients (13). In contrast to cancer, BRCA1 was upregulated in association with promoter demethylation. A series of experiments showed that BRCA1 upregulation was a cellular protective response to amyloid β (Aβ)-induced DNA double strand breaks. Despite its upregulation, BRCA1 was sequestered to neurofibrillary tangles (NFTs) and mis-localized to the cytoplasm. This suppressed its function and led to the accumulation of significant DNA damage in AD neurons. Aberrant hypomethylation was observed in neurons as well as glial cells, suggesting that Aβ toxicity affected all types of cells in the central nervous system (CNS).
In AD brains, neurons are far more vulnerable than glial cells to Aβ toxicity. This difference could be explained by the absence of NFTs in glial cells. Thus, functional BRCA1 provided resistance against DNA damage. Hypomethylation was also observed in the cerebellum, suggesting that Aβ was affecting the entire brain. However, its effect on other peripheral tissues and cells is not yet known.
The blood-brain-barrier was once thought to separate the CNS from peripheral blood, with the exception of several small molecules. However, there are a number of studies showing that the CNS environment can be observed from peripherally-derived samples (8, 12). Among patients with bipolar disorders and schizophrenia, there is evidence that genes encoding molecules reported to be involved in these diseases show an altered epigenome in peripheral blood (2, 9, 18). A series of recent studies on AD strongly indicates that fibrillar Aβ accumulation in the CNS can be detected by measuring Aβ levels in peripheral blood samples (5, 15, 16, 21). Thus, we believed that it was worthwhile to analyze the methylation level of the BRCA1 promoter in peripheral blood from AD patients.
Here, we report the levels of BRCA1 promoter methylation using peripheral blood DNA derived from clinically diagnosed AD patients. We also analyzed the methylation levels in J-ADNI participants who underwent either Pittsburgh compound B amyloid positron emission tomography (PET) or cerebrospinal fluid (CSF) Aβ42 analysis that identified them as amyloid pathology-positive, and examined the relationship between the methylation level and clinical features.
Methods

Ethics

All participants provided written informed consent. This study was approved by the ethics committee of the University of Tokyo (approvals G2183-18 and 11628-(1)). All experiments were performed in accordance with the principles of the Declaration of Helsinki.

Genomic DNA samples

Peripheral blood samples were obtained from the J-ADNI repository upon approval from the sample sharing committee. We obtained J-ADNI biomarker data from the National Bioscience Database Center with approval from its data access committee (https://humandbs.biosciencedbc.jp/en/hum0043-v1). We also validated the results of the J-ADNI samples in an independent cohort of patients admitted at Niigata University Hospital (Niigata, Japan). Patients were diagnosed clinically as AD or NC.

Amyloid positivity

Amyloid positivity was defined clinically by visual inspection of PET images using Pittsburgh Compound B as a ligand by trained radiologists who were blinded to any clinical information (22) and/or a CSF Aβ42 value ≤ 333 pg/mL at the baseline, using the cut-off value determined in a previous report (11). When both data were available, we defined a sample as «positive» when at least one test met the «positive» criterion.

Pyrosequencing

Bisulfite-conversion of DNA was performed by applying 100 ng of genomic DNA to an EpiTect Bisulfite Kit (Qiagen, Hilden, Germany), following the manufacturer’s instructions; the final product was eluted with 50 μL buffer. The primers used in this study were published previously (13) (Supplementary Figure 1). Samples were applied to PyroMark Q24 using PyroMark Gold Q24 Reagents (Qiagen), following amplification of bisulfite-converted DNA by the polymerase chain reaction using a PyroMark PCR Kit (Qiagen). The results were analyzed using PyroMark Q24 software (Qiagen). Primer set designs were verified using universal methylated and de-methylated human DNA standards.

Statistical analyses

Prism 6 (GraphPad, San Diego, CA, USA) and R software were used for statistical analyses. Unless otherwise noted, the significance of differences between groups was determined using the t-test. P < 0.05 was considered significant.

 

Results

We analyzed the methylation level of the BRCA1 promoter in peripheral blood from normal control (NC) and clinically diagnosed AD patients. The demographics of the subjects are shown in Table 1. We performed pyrosequencing analysis of genomic DNA samples derived from 76 subjects. In total, we analyzed seven CpGs of the BRCA1 promoter region that were differentially methylated in AD brains (13). The linearity of all primer sets was validated using universal methylated and unmethylated human genomic DNA (Figure S1). There were no statistically significant differences in the methylation levels between NC and AD patients in all the probes analyzed (Figure 1). Because these subjects were diagnosed based only on clinical features without knowing brain Aβ accumulation, there remains a possibility that insufficient diagnostic accuracy could have affected the results. Therefore, amyloid positivity should be confirmed for an accurate diagnosis of AD.

Table 1. Demographics of normal control (NC) and clinically diagnosed Alzheimer’s disease (AD) patients

* Fisher’s exact test; †The APOE ε4 genotype was not available in five NC samples.

Figure 1. Pyrosequenced methylation levels of each BRCA1 probe

The methylation levels of CpG probes in BRCA1 were analyzed using the validation cohort. Black dots and gray boxes represent normal control (NC) and Alzheimer’s disease (AD) patients, respectively. Error bars represent means + SD. Statistical significance was determined using a two-tailed t-test. n = 40 (NC) and 36 (AD).

 

Because the previous study on the neuron-specific methylome in AD brains demonstrated that the presence of Aβ was essential for aberrant methylation of the BRCA1 promoter region (13), we examined methylation in peripheral blood from NC and AD subjects with known amyloid pathology from the J-ADNI cohort. The demographics of this cohort are shown in Table 2. Among the 537 subjects enrolled in that study, there were 51 NC and 49 AD subjects for whom amyloid data from amyloid PET or CSF Aβ42 analysis were available. No significant change in the methylation level of the BRCA1 promoter region was observed (Figure 2).

Table 2. Demographics of participants from the J-ADNI study

CSF: cerebrospinal fluid

Figure 2. Differentially methylated BRCA1 CpG probes in neuronal cells

The CpG probes in BRCA1 that were differentially methylated in neuronal cells (13])were analyzed by pyrosequencing. Black dots and the gray boxes represent normal control (NC) and Alzheimer’s disease (AD) patients, respectively. Error bars represent means + SD. Statistical significance was determined using a two-tailed t-test. n = 51 (NC) and 49 (AD).

 

Because the methylation levels of some CpGs correlate with age, and the age of the participants examined in this study differed between NC and AD subjects, we considered the possibility that age-related effects might have diminished the difference in methylation levels of BRCA1 between NC and AD patients. However, the methylation level of the CpGs in the BRCA1 promoter region were not affected by age (Figure 3). Thus, we concluded that the methylation level of the BRCA1 promoter region derived from peripheral blood did not reflect the accumulation of fibrillar Aβ in the brain.

Figure 3. Relationship between the pyrosequenced methylation level of each BRCA1 probe and age at death

Correlation plots of the pyrosequenced methylation levels of each BRCA1 probe and age at death. For normal control (NC) patients, blue dots and lines represent individual data and regression lines, respectively. For Alzheimer’s disease (AD) patients, red dots and lines represent individual data and regression lines, respectively. Gray bands represent the 95% confidence interval of each linear regression. Pearson’s product-moment correlation coefficient (r) values between the methylation level and age of NC and AD patients, and subsequent significance, are shown in each graph.

 

In the brain, methylation levels of the BRCA1 promoter region in neuronal cells were associated with the number of APOE ε4 alleles only in the AD cohort (13). Thus, we analyzed the relationship between sex differences or APOE ε4 and the methylation level of this region in the NC and AD cohorts. As shown in Figure 4A, female sex had a significant positive effect on the methylation level of the BRCA1 promoter region only in the AD group. The number of APOE ε4 alleles had opposite effects on the methylation levels in NC and AD groups; in NCs, the methylation level was negatively correlated with the APOE ε4 allele number, while in AD the level had a positive correlation.

Figure 4. Relationship between the pyrosequenced methylation level of each BRCA1 probe and age at death

(A, B) Comparison of the methylation level of each differential methylation position based on sex difference (A) and the number of APOE ε4 alleles (B). (A) Orange, green, blue, and purple represent normal control (NC) males, NC females, Alzheimer’s disease (AD) males, and AD females, respectively. Significance was determined using a two-way analysis of variance (ANOVA) followed by the post hoc Sidak method. (B) Orange, yellow, light green, green, blue, and purple represent NC without the APOE ε4 allele, NC with one APOE ε4 allele, NC with two APOE ε4 alleles, AD without APOE ε4 allele, AD with one APOE ε4 allele, and AD with two APOE ε4 alleles, respectively. Whiskers were defined by Tukey’s boxplot method. Significance was determined for NC and AD subjects using a two-way ANOVA followed by the post hoc Tukey’s (for NC) or Sidak (for AD) method. Boxes extend from the 25th–75th percentile. Lines in the boxes are medians.

 

Discussion

In this study, we clearly demonstrated that the methylation level of the BRCA1 promoter in peripheral blood was unchanged between NC and AD patients. This was in stark contrast to the global methylation change in the CNS. In the brain, toxic Aβ induces DNA damage regardless of the region or the cell type, resulting in BRCA1 up-regulation through promoter hypomethylation. This allows sufficient DNA repair only in the absence of cytoplasmic aggregated tau because BRCA1 co-aggregates with NFTs and loses its proper function exclusively in neurons. In peripheral blood cells that are apparently free from NFTs, even if peripheral toxic Aβ had any effect on DNA damage, proper repair should occur. The absence of different methylation levels in NC and AD subjects could be attributed to a low Aβ concentration in the peripheral blood leading to a less prominent response of peripheral cells towards Aβ toxicity. Another explanation is that, in peripheral blood, Aβ toxicity could differ from that of the CNS, even if blood biomarker studies show that CNS Aβ accumulation can be detected by an analysis of the peripheral blood. This suggests that the CNS and peripheral blood share common properties of Aβ species.
A sub-analysis revealed several interesting results. Upon stratifying by sex, AD females showed higher methylation levels than males. One of the CpGs showed significantly higher methylation levels even after multiple post-hoc comparisons (Figure 4A, probe cg18372208). As the methylation level of the promoter region generally shows an inverse relationship with the expression of downstream genes (1, 19), the peripheral response to Aβ could be insufficient in females, resulting in vulnerability to toxic Aβ. This response could be related to the fact that being female is a risk factor for AD (4, 17) and also for a rapid decline in cognitive function (3, 6, 7, 10, 14).
In our previous study, we did not observe any sex differences in the BRCA1 promoter methylation level in the CNS (13). This discrepancy could be explained by differences of the organs we analyzed or the disease stage. Specifically, J-ADNI AD patients exhibited mild to moderate dementia, while autopsy patients usually suffer from severe dementia.
When comparing the number of APOE ε4 alleles and BRCA1 methylation levels, the presence of APOE ε4 alleles was negatively associated in NC and positively associated in AD subjects. These opposing results could be explained by a potential protective effect of BRCA1 upregulation in response to peripheral Aβ exposure; APOE ε4 carriers are protected from Aβ toxicity when BRCA1 is upregulated (i.e., a lower methylation level), whereas they are not protected when BRCA1 is downregulated (i.e., higher methylation in AD patients).
The cause of these different responses is not fully clear. Recent studies have shown that Aβ accumulation in the CNS could be detected by analyzing the peripheral blood (5, 15, 16). One study even showed that patients with CNS Aβ accumulation had increased Aβ oligomerization activity in peripheral blood (21). These results imply that aberrant Aβ metabolism could be occurring both in central and peripheral tissues, and could drive BRCA1 upregulation in certain conditions.
In summary, no significant changes in methylation of the BRCA1 promoter were observed in NC and AD patients. Thus, we concluded that BRCA1 promoter methylation cannot be a biomarker for diagnosing AD. Nevertheless, we found distinctive hypomethylation of the BRCA1 promoter in the brain through a neuron-specific methylome analysis. These data emphasized the importance of analyzing specific cell types directly involved in the disease process. However, only in the AD group were risk factors for AD (i.e., female sex and APOE allele ε4 number) correlated with BRCA1 promoter methylation in the peripheral blood, implying that the response to toxic Aβ in terms of BRCA1 expression was shared in the peripheral blood. This could be attributed to recent findings that the brain environment can be partially detected by observing peripheral blood, and provides insights into how the CNS and peripheral blood cross-talk in terms of Aβ accumulation (5, 15, 16, 21). Increased hypomethylation of the BRCA1 promoter has a potentially protective effect against Aβ toxicity. Therefore, while its relevance remains unclear, higher levels of BRCA1 promoter methylation might indicate AD risk, suggesting that peripheral methylation could be a potential biomarker for AD progression.

 

Author Contributions: Conceptualization, TM and AI; Methodology, TM and KS; Resource, TI and TI; Supervision, TT; Writing – Original Draft Preparation, TM; Writing – Review & Editing, AI; Funding, TM and AI.

Funding: This study was supported by AMED under grant number 17dm0107069h0002, 18dk0207028h0003, 18dk0207020h0004, JP21dk0207057h0001, 21dk0207042h0003, 21dk0207046h0001. JSPS KAKENHI grant numbers 16H05316 and 19K17027, the Cell Science Research Foundation (Osaka, Japan), the Ichiro Kanehara Foundation for the Promotion of Medical Sciences and Medical Care (Tokyo, Japan), the Takeda Science Foundation (Osaka, Japan), The Mochida Memorial Foundation for Medical and Pharmaceutical Research (Tokyo, Japan), Janssen Pharmaceutical K.K. (Tokyo, Japan), and Eisai Co. (Tokyo, Japan).

Acknowledgements: We are grateful for the technical support provided by Yuki Inukai-Mizutani.

Conflicts of interest: None.

 

SUPPLEMENTARY MATERIAL1

SUPPLEMENTARY MATERIAL2

 

References

1. Baldwin RL, Nemeth E, Tran H, et al., BRCA1 promoter region hypermethylation in ovarian carcinoma: a population-based study, Cancer research 2000;60:5329-5333.
2. Booij L, Szyf M, Carballedo A, et al., DNA methylation of the serotonin transporter gene in peripheral cells and stress-related changes in hippocampal volume: a study in depressed patients and healthy controls, PLoS One 2015;10:e0119061.
3. Butchart J, Birch B, Bassily R, Wolfe L, Holmes C, Male sex hormones and systemic inflammation in Alzheimer disease, Alzheimer Dis Assoc Disord 2013;27:153-156.
4. Caracciolo B, Palmer K, Monastero R, et al., Occurrence of cognitive impairment and dementia in the community: a 9-year-long prospective study, Neurology 2008;70:1778-1785.
5. Fandos N, Perez-Grijalba V, Pesini P, et al., Plasma amyloid beta 42/40 ratios as biomarkers for amyloid beta cerebral deposition in cognitively normal individuals, Alzheimers Dement (Amst) 2017;8:179-187.
6. Ferretti MT, Iulita MF, Cavedo E, et al., Sex differences in Alzheimer disease – the gateway to precision medicine, Nat Rev Neurol 2018;14:457-469.
7. Fyfe I, Alzheimer disease: Sex-specific inflammatory link to early Alzheimer pathology, Nat Rev Neurol 2017;13:5.
8. Gaiottino J, Norgren N, Dobson R, et al., Increased neurofilament light chain blood levels in neurodegenerative neurological diseases, PLoS One 2013;8:e75091.
9. Ikegame T, Bundo M, Murata Y, et al., DNA methylation of the BDNF gene and its relevance to psychiatric disorders, J Hum Genet 2013;58:434-438.
10. Iwata A, Iwatsubo T, Ihara R, et al., Effects of sex, educational background, and chronic kidney disease grading on longitudinal cognitive and functional decline in patients in the Japanese Alzheimer’s Disease Neuroimaging Initiative study, Alzheimers Dement (N Y) 2018;4:765-774.
11. Iwatsubo T, Iwata A, Suzuki K, et al., Japanese and North American Alzheimer’s Disease Neuroimaging Initiative studies: Harmonization for international trials, Alzheimer’s & Dementia 2018.
12. Lu CH, Macdonald-Wallis C, Gray E, et al., Neurofilament light chain: A prognostic biomarker in amyotrophic lateral sclerosis, Neurology 2015;84:2247-2257.
13. Mano T, Nagata K, Nonaka T, et al., Neuron-specific methylome analysis reveals epigenetic regulation and tau-related dysfunction of BRCA1 in Alzheimer’s disease, Proceedings of the National Academy of Sciences 2017.
14. Mosconi L, Berti V, Quinn C, et al., Sex differences in Alzheimer risk: Brain imaging of endocrine vs chronologic aging, Neurology 2017;89:1382-1390.
15. Nakamura A, Kaneko N, Villemagne VL, et al., High performance plasma amyloid-β biomarkers for Alzheimer’s disease, Nature 2018.
16. Ovod V, Ramsey KN, Mawuenyega KG, et al., Amyloid beta concentrations and stable isotope labeling kinetics of human plasma specific to central nervous system amyloidosis, Alzheimers Dement 2017;13:841-849.
17. Roberts RO, Geda YE, Knopman DS, et al., The incidence of MCI differs by subtype and is higher in men The Mayo Clinic Study of Aging, Neurology 2012;78:342-351.
18. Turecki G, Ota VK, Belangero SI, Jackowski A, Kaufman J, Early life adversity, genomic plasticity, and psychopathology, The Lancet Psychiatry 2014;1:461-466.
19. Turner NC, Reis-Filho JS, Russell AM, et al., BRCA1 dysfunction in sporadic basal-like breast cancer, Oncogene 2007;26:2126-2132.
20. Venkitaraman AR, Cancer susceptibility and the functions of BRCA1 and BRCA2, Cell 2002;108:171-182.
21. Wang MJ, Yi S, Han JY, et al., Oligomeric forms of amyloid-beta protein in plasma as a potential blood-based biomarker for Alzheimer’s disease, Alzheimers Res Ther 2017;9:98.
22. Yamane T, Ishii K, Sakata M, et al., Inter-rater variability of visual interpretation and comparison with quantitative evaluation of (11)C-PiB PET amyloid images of the Japanese Alzheimer’s Disease Neuroimaging Initiative (J-ADNI) multicenter study, Eur J Nucl Med Mol Imaging 2017;44:850-857.

A UK-WIDE STUDY EMPLOYING NATURAL LANGUAGE PROCESSING TO DETERMINE WHAT MATTERS TO PEOPLE ABOUT BRAIN HEALTH TO IMPROVE DRUG DEVELOPMENT: THE ELECTRONIC PERSON-SPECIFIC OUTCOME MEASURE (EPSOM) PROGRAMME

 

S. Saunders1, G. Muniz-Terrera1, S. Sheehan2, C.W. Ritchie1,3, S. Luz2

 

1. Centre for Clinical Brain Sciences, University of Edinburgh, UK; 2. Usher Institute of Population Health Sciences and Informatics; Molecular, Genetic and Population Health Sciences, University of Edinburgh, UK; 3. Brain Health Scotland, UK

Corresponding Author: Stina Saunders, University of Edinburgh, Centre for Clinical Brain Sciences, UK, stina.saunders@ed.ac.uk

J Prev Alz Dis 2021;4(8):448-456
Published online June 9, 2021, http://dx.doi.org/10.14283/jpad.2021.30

 


Abstract

BACKGROUND: It is important to use outcome measures for novel interventions in Alzheimer’s disease (AD) that capture the research participants’ views of effectiveness. The electronic Person-Specific Outcome Measure (ePSOM) development programme is underpinned by the need to identify and detect change in early disease manifestations and the possibilities of incorporating artificial intelligence in outcome measures.
Objectives: The aim of the ePSOM programme is to better understand what outcomes matter to patients in the AD population with a focus on those at the pre-dementia stages of disease. Ultimately, we aim to develop an app with robust psychometric properties to be used as a patient reported outcome measure in AD clinical trials.
Design: We designed and ran a nationwide study (Aug 2019 – Nov 2019, UK), collecting primarily free text responses in five pre-defined domains. We collected self-reported clinical details and sociodemographic data to analyse responses by key variables whilst keeping the survey short (around 15 minutes). We used clustering and Natural Language Processing techniques to identify themes which matter most to individuals when developing new treatments for AD.
Results: The study was completed by 5,808 respondents, yielding over 80,000 free text answers. The analysis resulted in 184 themes of importance. An analysis focusing on key demographics to explore how priorities differed by age, gender and education revealed that there are significant differences in what groups consider important about their brain health.
Discussion: The ePSOM data has generated evidence on what matters to people when developing new treatments for AD that target secondary prevention and therein maintenance of brain health. These results will form the basis for an electronic outcome measure to be used in AD clinical research and clinical practice.

Key words: Clinically meaningful change, electronic patient reported outcome measures, Alzheimer’s disease, outcome measures, brain health.


 

 

Introduction

Attempts to develop disease modifying therapies for Alzheimer’s disease (AD) started over 20 years ago with little success to date. A recent estimate of the costs of AD was US$818B, which is equivalent to the combined GDP of Indonesia, The Netherlands, and Turkey (1).
The lack of progress in finding a pharmacological treatment for AD is however at odds with a rapid development in the understanding of the pathology of AD suggesting that clinical trial design and delivery may partially account for a lack of progress with insensitive outcome measures lacking clinical meaningfulness also playing a part in this lack of progress. It has been shown that the disease process starts long before an individual becomes symptomatic or eventually, the dementia syndrome manifests (2, 3). Increasingly, we are exploring AD processes at earlier disease stages through examining at-risk populations in mid-life which helps identify the earliest manifestation of declining brain health. In the absence of pharmacological interventions, it is estimated that approximately 40% of dementia cases could be prevented by targeting epidemiologically derived modifiable risk factors (4). Changes occurring years earlier than dementia develops have been observed in at-risk populations using exploratory and sensitive computerised tests assessing e.g. allocentric and egocentric spatial processing (5). These test results correlate with brain imaging findings in hippocampal subfields known to be sensitive to amyloid derived neurotoxicity (6); as well as in changes to brain β-amyloid in at risk populations aged between 63-81 years old who did not have dementia (7).
Whilst there are global initiatives focusing on dementia prevention through risk factor modification (8, 9), there remains a major and complementary need for effective AD pharmacological interventions. Irrespective of the type of intervention to reduce incident dementia rates, the fact is that these studies will engage at risk populations who will be, to the most part, in mid-life and healthy. Currently, there are 31 AD drugs being tested in Phase III clinical trials (19 of which are disease modifying) (10). We argue that using outcome measures assessing clinical symptoms and functioning in earlier disease stages is less valid than biological measures of disease and what the individual considers personally meaningful from a treatment. A treatment’s success should therefore be determined not only by the impact on the individual’s disease (as evidenced by biomarker change) but also by its effect on related well-being (as measured by patient reported outcomes).
To this end, whilst it is currently proposed by regulators that AD trials measure cognition as the primary outcome, as trials move to an earlier disease stage it could be argued that many commonly used (cognitive) measures lack ecological validity and are not sensitive enough to detect changes in the earlier stages of the AD continuum where the ideal intervention should take place (11). Moreover, it is recommended by both the US Food and Drug Administration (FDA) (12) and European Medicines Agency (EMA) (13) that AD trials incorporate measures which capture clinically meaningful results to the individual. Patient reported outcome measures (PROMs) are developed for the incorporation of the person’s own perspective regarding their treatment, though these measures are currently not used in AD clinical trials (14). PROMs reflect an individual’s view on what they define as an effective treatment and consider a meaningful change. Notably, PROMs are already more widely used in other disease areas. For example, a recent study of nearly 100,000 clinical trials published on clinicaltrials.gov found that a PROM had been used in 27% of all trials, primarily in oncology (15).
In light of the drive towards early detection, looking at younger at-risk populations and the main regulators’ recommendation for clinically meaningful outcome measures, we have established the electronic Person Specific Outcome Measure (ePSOM) development programme. As the target population in dementia prevention research is an at-risk population, our group took the view that what matters to people when developing new treatments for AD is approached by way of maintenance of brain health (16, 17). The ePSOM programme consists of four sequential steps, ultimately aiming to employ new technology to create an outcome measure to be used in AD clinical research and practice. This will be in the form of an outcome app used on any screen-based device which will assess aspects specific to the individual using it. At the start of the programme, we reviewed literature around PROMs in AD clinical trials which informed our focus group study with people with memory concerns, healthy volunteers and health care professionals (18). The focus group study yielded five domains of importance for what matters to people about brain health. These domains formed the basis for the next stage of the ePSOM development programme. In this paper, we report on a large UK-wide population-based study to understand what matters to people when developing new treatments for Alzheimer’s disease. We consider the respondents to the ePSOM study a representative population of individuals who may be enrolled in dementia primary and secondary prevention clinical trials and characterise what matters to people about brain health focusing on key demographic groups.

 

Methods

We designed and ran a UK-wide population-based online study collecting primarily free text answers (see Appendix 1). The study built on a previously run focus group study which yielded five domains of importance about brain health. The study obtained ethics approval from the ACCORD Medical Research Ethics Committee in Edinburgh, Scotland. The ePSOM study ran from Aug 2019 – Dec 2019 and was divided into sections, starting with an introductory video and informed consent.
Free text answers were collected across five pre-defined domains. These answers were clustered, leading to specific themes of what matters to people about brain health

The study was open to anyone over the age of 18 and was launched primarily via Alzheimer’s Research UK media channels through e-mails to individuals registered on their mailing lists and a social media campaign (with social media support from other dementia related organisations). We collected key sociodemographic and clinical data such as having been seen by a doctor because of any brain health issues though the primary method of the survey used a qualitative approach. Respondents were presented with the five domains derived from the earlier focus group work to orientate and channel free text responses: [1] Everyday functioning; [2] Sense of Identity; [3] Relationships and Social Connections; [4] Enjoyable Activities and [5] Thinking problems. They were then asked to provide free text answers on what they would like to retain or keep being able to do in those domains if their brain health got worse. At the end of the study, respondents were asked to identify five answers across all the answers they had given which they consider the most important. We used Natural Language Processing (NLP) techniques to analyse the free text data (see Figure 1).

Figure 1. Natural Language Processing techniques used to analyse the survey data

 

Free text answers were collected across five pre-defined domains. These answers were clustered, leading to specific themes of what matters to people about brain health

Step 1: Natural Language Processing to create clusters

We used NLP to create clusters of semantically similar free text answers. These clusters were then manually annotated with appropriate labels. We refer to the finally labelled clusters derived from this stepped NLP-manual annotation process as “themes”.
NLP employed word embeddings trained on vast amounts of text data to achieve fine-grained representation of semantic regularities in text. We were thus able to build robust representations of free text answers. To begin, “stop words” (i.e. words that occur very frequently and contribute little to semantic content) and punctuation were removed from the free text answers. The resulting texts were then converted to numerical vector representations by using GloVe vectors (19) to generate sentence embeddings. These vectors encode semantic relationships between words and can be used to cluster semantically similar text segments. This allowed us to use automated methods to identify words, and thus answers, of a similar “theme” or meaning. The K-means clustering algorithm was used to cluster the answer embeddings within each of the five domains. The K parameter, that is, the desired number of automatic clusters per domain, was determined analytically. The goal was to generate fine grained clusters which contain semantically similar answers while avoiding overfitting or creating so many clusters that important themes are not revealed. We found when the number of important items in the largest cluster changes by less than 10, between each of the previous five increments of K, that the majority of the clusters also exhibited minor changes in the number of important items. Using this criterion, we chose a value of 151 clusters across all five domains. This method resulted in a total of 755 clusters of free text answers, or 151 clusters for each of the five domains.

Step 2: Manual Annotation to create themes

The clusters that emerged within each of the five domains were reordered so that semantically similar clusters appeared close together. This was achieved using hierarchical clustering on the cluster centroids. We used the reordered clusters for manual annotation in each of the five domains. Each cluster was represented by the 200 most frequent unique answers, after punctuation and stopwords were removed. The annotation goals were to combine any clusters which fit together, exclude uninterpretable clusters and label the final clusters thus deriving the final themes. Six authors of the current paper annotated two domains each, ensuring two separate people analysed a single domain, which helped ensure inter-rater reliability between domains. Finally, two of the authors did quality control across the five domains and homogenised the labels across domains.

Statistical analyses

In this paper, we focus our analyses on key demographic groups: age (up to the age of 64 / age 65 and older); gender (men / women) and education (no degree / degree and higher). We present the largest themes for each of these demographic groups as well as themes which were identified as particularly important most often in the final question on the study forms. For both of these analyses, we report percentages for each theme by key demographic groups. As the demographic groups are unbalanced in terms of the number of respondents we use percentages rather than the absolute number of answers in the statistical analyses. The percentages are derived by dividing the count of answers within the demographic group by the total number of answers in that demographic group, thus providing proportions for comparison when dealing with imbalanced demographics. It should be noted that respondents were not bounded by a minimum or maximum number of free text answers they could give in each domain.
Finally, we conducted a Chi-squared test to analyse whether the differences in percentages between demographic groupings’ answers within each theme were statistically significant. A p-value of <0.01 was used in statistical significance testing.

 

 

Results

The study was completed by 5,808 people from across the UK. They provided a total of 82,514 free text answers. These were clustered using automated NLP techniques resulting in 151 clusters in each of the five pre-defined domains, a total of 755 automated clusters across all domains, as described. Subsequent analysis reduced the number of clusters to 334 (due to a cluster being represented in two or more domains) which were all manually annotated by the research team. Many of the same themes emerged from different domains (e.g., the theme of Walking in the “Enjoyable activities” domain as well as the “Everyday activities” domain). After merging themes with the same label in different domains, the final number of unique themes was 184. Some respondents used more generic language (e.g., “Maintaining independence”) whereas others were more specific (e.g., “Driving”). Using NLP methods for free text analysis means that, in this example, the “Maintaining independence” theme contains 1100 answers, most containing either the word “independent” or “independence”. Analytically, this is therefore not a general theme for answers which relate to the concept of independence, but a cluster of answers in which the respondents are directly referencing the word independence as something which is important for them to maintain. This has therefore resulted in themes which are more or less specific but directly reflect the language used by the respondents.
Pre-defined answers: Characteristics of the ePSOM survey sample

The characteristics of the 5,808 respondents are presented above (see Table 1).

Table 1. ePSOM survey respondent characteristics

We used NLP techniques and manual annotation to group individual free text answers into clusters and then themes respectively. The most frequent themes across all demographics were reading, driving, friendships and following a storyline (Figure 2). We also calculated the proportion of answers within each key demographic, expressed as percentages of the total answers given by that demographic group.

Figure 2. What matters to people about brain health? The survey received 82,514 free text answers which were clustered into 184 themes

This figure shows themes which were mentioned the most, broken down by key demographics. Full figure of the survey themes in Appendix 2.

At the end of the survey, respondents were asked to identify the five most important answers to them across all their answers. We used this metric to rank the themes in terms of being selected as particularly important and observed a subtle difference between the largest themes (themes mentioned the most frequently) and themes which are identified as the most important. The 5 top important themes across all demographics were family connections, driving, socializing, reading and friendships (Figure 3).

Figure 3. What matters to people about brain health?

This figure shows themes with the highest number of answers selected as particularly important by key demographics. Full figure of themes with the most important answers in Appendix 3.

 

Cross-Tabulations of key demographics

The following tables show statistically significant proportional differences in theme sizes (Table 2) and identifying themes as particularly important (Table 3), focusing on demographic group dyads (younger vs older; men vs women; individuals with no degree vs individuals with a degree or higher).
Table 2 Top 10 themes selected as particularly important which had the highest Chi square values representing greater differentiation between demographic groups by age (younger/older); gender (male/female) and education (no degree / degree and higher). A full list of particularly important themes which were significantly different across key demographics can be found in Appendix 4.

Table 2. Top 10 themes selected as particularly important which had the highest Chi square values representing greater differentiation between demographic groups by age (younger/older); gender (male/female) and education (no degree / degree and higher)

Full list of particularly important themes which were significantly different across key demographics in Appendix 4.

Table 3. Top 10 largest themes which had the highest Chi square values representing greater differentiation between demographic groups by age (younger/older); gender (male/female) and education (no degree / degree and higher)

Full list of largest themes which were significantly different across key demographics in Appendix 5.

 

Discussion

Building on the scientific foundation provided by previous stages of the ePSOM research programme, we designed and ran a nationwide study with open ended questions to derive free text answers exploring what matters to people about maintaining their brain health within five focus group-derived domains. To our knowledge, this is the first study collecting free and systematically analysing text responses from a very large number of respondents on what is important to them about brain health. The themes and granularity derived from our study are in line with the FDA’s guidance for capturing aspects relevant to AD research participants “e.g., [assessing] facility with financial transactions, adequacy of social conversation” (12).
As AD drug development moves to an earlier phase of the neurodegenerative disease spectrum and clinical research targets an earlier, younger population, it is crucial any outcomes are meaningful and relevant to that trial population. Additionally, as upcoming AD treatments are hoped to be disease modifying rather than reducing symptoms, the cognitive domains which respond to the medication may not be the same as with symptomatic treatments measured at a later disease stage (20). We also know from a recent review that lifestyle factors may influence brain health in midlife (21) so it is apposite to examine what matters to people about brain health including lifestyle dependant factors as this will be increasingly relevant in Brain Health Clinics which are developing throughout the UK (16) and Europe (17).
There has been other work collecting evidence on important outcomes focusing on the point of view of people living with dementia (22). The focus of the ePSOM programme though is the maintenance of brain health. As the majority of the individuals in our study had not received a diagnosis of neurodegenerative disease, the findings from our study provide evidence for what matters to people about brain health in normal lived experience which may include people at the earlier (asymptomatic) stages of disease rather than once the dementia syndrome develops. Our findings are supported by literature recognising that AD trials currently do not measure outcomes which are relevant to the patient themselves. Tochel et al. (23) carried out a literature review extracting data from studies where participants described outcomes which matter to them. Their review concluded by demonstrating an array of outcomes which are not commonly captured in clinical trials of new treatments (23).
Changes at the early stages of the AD continuum are currently detected by biomarker assessments, with functional measures used increasingly towards the more symptomatic and advanced stage of the continuum where ultimately impairment is evidenced in basic activities of daily living. However, dementia prevention cohorts have found differences in more than just biomarker assessed pathology, e.g. there is evidence that middle-aged adults at risk of dementia have poorer cognitive performance, principally in visuospatial functions (24) and memory (25). Lau et al. (26) concluded that observing early functional limitations at baseline in the at-risk population had prognostic value in identifying older adults at risk for developing functional disability a few years later (26).
A recent review also concluded that in the pre-dementia stages of AD, executive functions (such as inhibitory abilities), attentional and visuospatial functions can already be impacted (27). A PROM therefore could be viewed as an ecologically valid instrument for cognitive assessment measures which are proxies for what matters to people, especially if the PROM relates to a cognitive process affected early in the course of AD (e.g. activities requiring planning, judgement or navigation/orientation like confidence driving). The key questions here is: if an individual’s score changes on a particular domain using a cognitive assessment measure, does this correlate with a change of score in a PROM and is therefore a change meaningful (by definition) to the patient? While functional or Activities of Daily Living scales measure a more direct or practical effect a drug may have, these measures have limitations such as poor psychometric properties (28) and as evidenced by the analysis of key demographic groups in the ePSOM survey, what matters to people about brain health and their function is different depending on age, sex and education levels. By capturing data specific to the individual who in effect derives their own outcome measure, the ePSOM app in development would present an outcome measure for clinical trials that captures changes noticed by and meaningful to the person themselves and therefore more likely to be correlated to their own specific functional outcomes than generic outcomes which were derived by homogenising population level data. Ultimately, employing more meaningful, ecologically valid and sensitive measures will facilitate more drugs to be approved by regulatory bodies which will actually impact on well-being and not just impact on cognition and function ‘on average’ between groups (29). Moreover – ePSOMs are immune to cultural, educational and language variability as each outcome is unique to that individual and bears no reference to an external ‘population norm’.
We used an online study design as it was important to allow for free text answers and reach a large number of people. However, this is also a limitation in the study leading to inevitable sampling bias of individuals who are able to access an online survey. There was also a demographic imbalance among the survey respondents with reference to the UK population as a whole, but appropriate analysis focusing on proportions rather than absolute values of this relatively large sample mitigates the effects of the imbalances in the data. The main strength of the study was collecting free text answers and using NLP techniques in the data analysis. Employing NLP techniques to gather evidence for what outcomes matter in AD drug development is unique and we are not aware of any similar studies. Free text answers offer insights which go beyond rating themes on a scale which have been predefined as important by the researchers and are culturally biased and limited. Moreover, the open character of the questions may motivate respondents to reveal more (31). In some regards, our study results may be considered comparable to hundreds of focus group studies, though by using NLP techniques, we are able to extract patterns in answers by key demographic at a scale and level of detail not feasible using traditional qualitative methodologies.

 

Conclusion

There is a growing consensus that PROMs should be used in AD trials so that the patient can assess if they observe a change in their well-being which is meaningful and specific to them. Including the patient’s perspective is also recommended by regulatory bodies such as the EMA with whom we collaborated in the initial phases of this project, and the FDA. In our study, we included a large number of people collecting free text responses to understand what matters to people about their brain health – our analyses focussed on key demographic groups. This approach is novel in so much as it uses NLP approaches to create a range of outcomes from a theoretically limitless range of possible responses and then can apply these into quantifiable and ecologically valid outcomes. The main criticism and in many ways fatal flaw of current approaches to PROMs is that they are derived at a population level and therefore have to incorporate the characteristics of the population they were derived from. These populations will hold certain language, cultural and ethnic characteristics making their use in other limited in other populations. The ePSOM app will ultimately be used by people in earlier stages of neurodegenerative disease before dementia develops in populations across the globe, in clinical trials with seamless translation into clinical practice.

 

Acknowledgement: We thank Alison Evans from Alzheimer’s Research UK for her intellectual contribution to this study. We also thank the individuals who took part in the ePSOM study.

Conflicts: The authors declare no conflict of interest.

Funding Sources: The ePSOM survey was funded by Alzheimer’s Research UK.

Declarations of interest: none (all authors).

Ethical standards: The study obtained ethics approval from the ACCORD Medical Research Ethics Committee in Edinburgh, Scotland.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

 

SUPPLEMENTARY MATERIAL1

SUPPLEMENTARY MATERIAL2

SUPPLEMENTARY MATERIAL3

SUPPLEMENTARY MATERIAL4

SUPPLEMENTARY MATERIAL5

 

References

1. Wimo A, Guerchet M, Ali G-C, Wu Y-T, Prina AM, Winblad B, et al. The worldwide costs of dementia 2015 and comparisons with 2010. Alzheimer’s & dementia : the journal of the Alzheimer’s Association. 2017;13(1):1-7. doi: 10.1016/j.jalz.2016.07.150
2. DeTure MA, Dickson DW. The neuropathological diagnosis of Alzheimer’s disease. Molecular Neurodegeneration. 2019;14(1):32. doi: 10.1186/s13024-019-0333-5
3. Association As. Alzheimer’s disease facts and figures. Alzheimer’s & Dementia. 2020;16(3):391-460. doi: 10.1002/alz.12068
4. Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396(10248):413-46. doi: 10.1016/s0140-6736(20)30367-6
5. Ritchie K, Carrière I, Howett D, Su L, Hornberger M, O’Brien JT, et al. Allocentric and Egocentric Spatial Processing in Middle-Aged Adults at High Risk of Late-Onset Alzheimer’s Disease: The PREVENT Dementia Study. Journal of Alzheimer’s Disease. 2018;65:885-96. doi: 10.3233/JAD-180432
6. Brien JT, Firbank MJ, Ritchie K, Wells K, Williams GB, Ritchie CW, et al. Association between midlife dementia risk factors and longitudinal brain atrophy: the PREVENT-Dementia study. Journal of Neurology, Neurosurgery &amp;amp; Psychiatry. 2020;91(2):158. doi: 10.1136/jnnp-2019-321652
7. Martikainen IK, Kemppainen N, Johansson J, Teuho J, Helin S, Liu Y, et al. Brain β-Amyloid and Atrophy in Individuals at Increased Risk of Cognitive Decline. AJNR Am J Neuroradiol. 2019;40(1):80-5. doi: 10.3174/ajnr.A5891
8. Ritchie CW, Muniz-Terrera G, Kivipelto M, Solomon A, Tom B, Molinuevo JL, et al. The European Prevention of Alzheimer’s Dementia (EPAD) Longitudinal Cohort Study: Baseline Data Release V500.0. J Prev Alzheimers Dis. 2020;7(1):8-13. doi: 10.14283/jpad.2019.46
9. Cummings J, Aisen P, Barton R, Bork J, Doody R, Dwyer J, et al. Re-Engineering Alzheimer Clinical Trials: Global Alzheimer’s Platform Network. J Prev Alzheimers Dis. 2016;3(2):114-20. doi: 10.14283/jpad.2016.93
10. Devenish SRA. The current landscape in Alzheimer’s disease research and drug discovery. Drug Discov Today. 2020;25(6):943-5. doi: 10.1016/j.drudis.2020.04.002
11. Snyder PJ, Kahle-Wrobleski K, Brannan S, Miller DS, Schindler RJ, DeSanti S, et al. Assessing cognition and function in Alzheimer’s disease clinical trials: do we have the right tools? Alzheimers Dement. 2014;10(6):853-60. doi:10.1016/j.jalz.2014.07.158
12. FDA UFaDA. Early Alzheimer’s Disease: Developing Drugs for Treatment Guidance for Industry. 2018.
13. EMA EMA. Guideline on the clinical investigation of medicines for the treatment of Alzheimer’s disease. Committee for Medicinal Products for Human Use (CHMP). 2018:1-36.
14. Saunders S, Muniz-Terrera G, Watson J, Clarke CL, Luz S, Evans AR, et al. Participant outcomes and preferences in Alzheimer’s disease clinical trials: The electronic Person-Specific Outcome Measure (ePSOM) development program. Alzheimers Dement (N Y). 2018;4:694-702. doi:10.1016/j.trci.2018.10.013
15. Vodicka E, Kim K, Devine EB, Gnanasakthy A, Scoggins JF, Patrick DL. Inclusion of patient-reported outcome measures in registered clinical trials: Evidence from ClinicalTrials.gov (2007-2013). Contemp Clin Trials. 2015;43:1-9. doi:10.1016/j.cct.2015.04.004
16. Ritchie CW, Russ, T. C., Banerjee, S., Barber, B., Boaden, A., Fox, N. C., . . . Burns, A. The Edinburgh Consensus: preparing for the advent of disease-modifying therapies for Alzheimer’s disease. Alzheimers Res Ther. 2017;9(1):85. doi:doi:10.1186/s13195-017-0312-4
17. Frisoni GB, Molinuevo JL, Altomare D, Carrera E, Barkhof F, Berkhof J, et al. Precision prevention of Alzheimer’s and other dementias: Anticipating future needs in the control of risk factors and implementation of disease-modifying therapies. Alzheimers Dement. 2020;16(10):1457-68. doi:10.1002/alz.12132
18. Watson J, Saunders S, Muniz Terrera G, Ritchie C, Evans A, Luz S, et al. What matters to people with memory problems, healthy volunteers and health and social care professionals in the context of developing treatment to prevent Alzheimer’s dementia? A qualitative study. Health Expect. 2019;22(3):504-17. doi:10.1111/hex.12876
19. Pennington J, Socher, R., Manning, C. Glove: Global vectors for word representation Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP). 2014:1532–43. doi: 10.3115/v1/D14-1162
20. Hendrix SB. Measuring clinical progression in MCI and pre-MCI populations: enrichment and optimizing clinical outcomes over time. Alzheimers Res Ther. 2012;4(4):24-. doi:10.1186/alzrt127
21. Topiwala H, Terrera GM, Stirland L, Saunderson K, Russ TC, Dozier MF, et al. Lifestyle and neurodegeneration in midlife as expressed on functional magnetic resonance imaging: A systematic review. Alzheimer’s & Dementia: Translational Research & Clinical Interventions. 2018;4(1):182-94. doi:10.1016/j.trci.2018.04.001
22. Morbey H, Harding AJE, Swarbrick C, Ahmed F, Elvish R, Keady J, et al. Involving people living with dementia in research: an accessible modified Delphi survey for core outcome set development. Trials. 2019;20(1):12. doi:10.1186/s13063-018-3069-6
23. Tochel C, Smith M, Baldwin H, Gustavsson A, Ly A, Bexelius C, et al. What outcomes are important to patients with mild cognitive impairment or Alzheimer’s disease, their caregivers, and health-care professionals? A systematic review. Alzheimers Dement (Amst). 2019;11:231-47. doi:10.1016/j.dadm.2018.12.003
24. Ritchie K, Carrière I, Su L, O’Brien JT, Lovestone S, Wells K, et al. The midlife cognitive profiles of adults at high risk of late-onset Alzheimer’s disease: The PREVENT study. Alzheimer’s & Dementia. 2017;13(10):1089-97. doi: 10.1016/j.jalz.2017.02.008
25. Caselli RJ, Locke DEC, Dueck AC, Knopman DS, Woodruff BK, Hoffman-Snyder C, et al. The neuropsychology of normal aging and preclinical Alzheimer’s disease. Alzheimer’s & dementia : the journal of the Alzheimer’s Association. 2014;10(1):84-92. doi:10.1016/j.jalz.2013.01.004
26. Lau KM, Parikh M, Harvey DJ, Huang CJ, Farias ST. Early Cognitively Based Functional Limitations Predict Loss of Independence in Instrumental Activities of Daily Living in Older Adults. J Int Neuropsychol Soc. 2015;21(9):688-98. doi:10.1017/s1355617715000818
27. Guarino A, Favieri F, Boncompagni I, Agostini F, Cantone M, Casagrande M. Executive Functions in Alzheimer Disease: A Systematic Review. Front Aging Neurosci. 2019;10:437-. doi:10.3389/fnagi.2018.00437
28. Weintraub S, Carrillo MC, Farias ST, Goldberg TE, Hendrix JA, Jaeger J, et al. Measuring cognition and function in the preclinical stage of Alzheimer’s disease. Alzheimer’s & dementia (New York, N Y). 2018;4:64-75. doi:10.1016/j.trci.2018.01.003
29. Sabbagh MN, Hendrix S, Harrison JE. FDA position statement “Early Alzheimer’s disease: Developing drugs for treatment, Guidance for Industry”. Alzheimer’s & dementia (New York, N Y). 2019;5:13-9. doi:10.1016/j.trci.2018.11.004
30. Ashour HM, Elkhatib WF, Rahman MM, Elshabrawy HA. Insights into the Recent 2019 Novel Coronavirus (SARS-CoV-2) in Light of Past Human Coronavirus Outbreaks. Pathogens. 2020;9(3). doi:10.3390/pathogens9030186
31. Rohrer JM, Brümmer M, Schmukle SC, Goebel J, Wagner GG. “What else are you worried about?” – Integrating textual responses into quantitative social science research. PLOS ONE. 2017;12(7):e0182156. doi:10.1371/journal.pone.0182156s