jpad journal

AND option

OR option



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:

J Prev Alz Dis 2021;4(8):414-424
Published online September 21, 2021,



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.



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.



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.



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.



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



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.


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).



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).



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 (, 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.





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.
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



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: (J.T. Yu); (L. Tan), Tel: +86 21 52888160; Fax: +86 21 62483421

J Prev Alz Dis 2021;
Published online September 20, 2021,



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.



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.




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.



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.



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.



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.
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.
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.
4. Iadecola C, Gottesman RF. Cerebrovascular Alterations in Alzheimer Disease. Circulation research. 2018;123(4):406-8.
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.
6. Iadecola C. The Neurovascular Unit Coming of Age: A Journey through Neurovascular Coupling in Health and Disease. Neuron. 2017;96(1):17-42.
7. Liesz A. The vascular side of Alzheimer’s disease. Science. 2019;365(6450):223-4.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
30. Pluta R, Ułamek-Kozioł M, Januszewski S, Czuczwar SJ. Tau Protein Dysfunction after Brain Ischemia. J Alzheimers Dis. 2018;66(2):429-37.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
45. Iadecola C. The pathobiology of vascular dementia. Neuron. 2013;80(4):844-66.
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.
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.
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.
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.
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.
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.
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.
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.
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.




E. Siemers1, P.S. Aisen2, M.C. Carrillo3


1. Siemers Integration LLC, Zionsville, IN, USA; 2. Alzheimer’s Therapeutic Research Center, University of Southern California, San Diego, CA, USA; 3. Alzheimer’s Association, Chicago, IL, USA

Corresponding Author: E. Siemers, Siemers Integration LLC, Zionsville, IN, USA,

J Prev Alz Dis 2021;
Published online September 17, 2021,



Very recently, the Food and Drug Administration (FDA) in the United States gave an “accelerated approval” to aducanumab, the first new drug to be available to patients with Alzheimer’s disease (AD) in nearly two decades and the first ever that targets the underlying neuropathology. The accelerated approval pathway is based on a biomarker effect, in this case reduction in brain amyloid as measured by PET scan, that is “reasonably likely” to predict clinical efficacy. While there were numerous complexities surrounding the approval, this event was nevertheless seminal for the treatment of AD and for the amyloid hypothesis.
The amyloid hypothesis is frequently discussed as a monolithic viewpoint; however, there are many important nuances within the broad theory. As noted vide infra, Aβ monomers may be targeted by both γ-secretase and β-Amyloid Cleavage Enzyme (BACE) inhibitors, as well as certain monoclonal antibodies. Amyloid plaques, composed of anti-parallel β-pleated sheets of Aβ monomers (primarily Aβ1-42) (1) are targeted by a number of monoclonal antibodies, including aducanumab. Aβ protofibrils and Aβ oligomers have been targeted less frequently by monoclonal antibodies but represent plausible targets within the amyloid framework.
Thus, the broad categorization of the amyloid hypothesis has important sub-types which will be discussed. Amyloid accumulation in brain is a defining feature of AD. Much evidence, particularly tight linkage of amyloid pathways to all genetic forms of AD, support amyloid as a therapeutic target. The relative value of targeting the various forms of amyloid is widely debated.
Very importantly, a growing consensus is forming that Aβ aggregation in the brain begins early and is followed by inflammation and the accumulation and spread of tau tangles in areas of the brain important for cognition (2, 3). Based on other biomarker and genetic data, a number of other targets for AD are clearly worth pursuing (4). These include tau, inflammatory mechanisms, and even other “non-amyloid non-tau” (“NANT”) mechanisms that should be investigated. An emerging consensus in the field of AD research is that that no single drug is likely to provide optimal treatment of AD, and that combination therapy using drugs with different mechanisms is most likely to provide the best therapy for the disorder (5). While this paper will focus on the amyloid hypothesis broadly, other mechanisms should continue to be pursued vigorously alone and in combination.


Gamma secretase inhibitors

Among the first potential disease-modifying drugs to be tested in clinical trials for AD are the γ-secretase inhibitors (6). γ-secretase is an aspartyl protease which cleaves the amyloid precursor protein (APP) following cleavage by BACE leading to the formation of the amyloid-β (Aβ) peptide (7). Inhibition of γ-secretase leads to reduction in the synthesis of Aβ in the central compartment (8). Despite this effect on Aβ synthesis, two γ-secretase inhibitors taken into the clinic did not cause slowing of disease progression and in fact caused slight cognitive worsening (9, 10). While unexpected and unfortunate, this worsening of cognition may have been related to multiple other substrates of γ-secretase and inhibition of their cleavage (7).


BACE inhibitors

BACE inhibitors collectively received a great deal of enthusiasm as several of these small molecules moved into Phase 2 and Phase 3 studies. This enthusiasm may have been due to robust reductions of Aβ in cerebrospinal fluid (CSF), and also a report of a polymorphism in the APP gene at the BACE cleavage site that reduced BACE cleavage of APP and had an apparent protective effect with regard to AD in an Icelandic population (11). Despite this promising background, unfortunately trials of several BACE inhibitors were stopped due to negative results, with cognitive worsening in some studies, or due to futility as reviewed by Imbimbo et al (12). Like γ-secretase, BACE has multiple substrates in addition to APP (12) which may be related to these disappointing results. The fact that the “Icelandic mutation” was in the APP gene means that the effect of BACE on its other substrates was unimpaired in that population, thus providing protection from AD without the adverse effects associated with BACE inhibitors. Alternatively, the similar cognitive worsening with γ-secretase and BACE inhibition raises the possibility that substantial reduction of Aβ levels adversely affects synaptic function.


Monoclonal antibodies

Despite the disappointments of the γ-secretase and BACE inhibitor studies, monoclonal antibodies targeting various forms of Aβ or amyloid plaque have led to more encouraging results. Monoclonal antibodies may be engineered to bind primarily to Aβ monomers, Aβ oligomers, protofibrils or deposited amyloid plaques. Many antibodies have some degree of binding to multiple forms of Aβ/amyloid.

Monoclonal antibodies primarily targeting amyloid plaques or protofibrils

Antibodies which were developed to primarily target deposited amyloid plaques include aducanumab (13-15), donanumab (16, 17), and gantenerumab (18-20). While these antibodies can lead to a substantial lowering of amyloid plaque load as assessed by amyloid positron emission tomography (PET), they are all accompanied by amyloid-related imaging abnormalities (ARIA) to some degree. While ARIA may be asymptomatic, it can also be accompanied by relatively minor symptoms such as headache, and can in some cases lead to hospitalization. Dose titration and surveillance with magnetic resonance imaging (MRI) is necessary when using these antibodies. Positive clinical data have been reported for aducanumab; however, statistical significance was not achieved for the primary outcome measure in one of two pivotal trials (13, 15) as noted in Table 1. Positive clinical data were also achieved for a Phase 2 trial of donanumab (17). Phase 3 trials using an increased dose of gantenerumab are currently ongoing.
The monoclonal antibody lecanemab (BAN2401) was developed to bind to protofibrils that have been associated with the “Arctic mutation” (21). Trial results show that the antibody is associated with substantial plaque reduction based on amyloid PET and the fact that it causes ARIA. As summarized in Table 1, clinical efficacy results from a Phase 2 study were also encouraging (22).
Bapineuzumab was one of the first monoclonal antibodies to enter the clinic and was the first to be associated with ARIA. Largely due to concerns about ARIA, doses were very limited compared to those now used with other antibodies and the amount of plaque reduction as determined by PET was very limited (23-27). In hindsight, given the small doses and small effects on plaque load, the lack of clinical efficacy is not unexpected.

Monoclonal antibodies targeting Aβ monomers

Solanezumab is a monoclonal antibody binding the mid-domain of Aβ and has binding largely restricted to Aβ monomers (28, 29). Given that solanezumab does not bind to amyloid plaques, it does not reduce plaque load based on PET and is not associated with ARIA (30, 31). Solanezumab was studied in two large pivotal trials in patients with mild-moderate dementia (EXPEDITION and EXPEDITION-2), and a third trial (EXPEDITION-3) that was limited to patients with mild dementia who were also known to be amyloid positive based on PET or CSF. While the EXPEDITION and EXPEDITION-2 studies did not meet their primary outcomes in the mild-moderate populations (30), planned secondary analyses did show promising results for patients with mild dementia only (32). The EXPEDITION-3 study also did not achieve statistical significance for the primary outcome measure, but consistent trends favoring a drug effect were present (31) as noted in Table 1.

Table 1. Impact of Therapy on Disease Progression in Recent Phase 2-3 AD Anti-Amyloid mAb Studies*


Crenezumab is a monoclonal antibody based on an IgG4 background that was developed in part as a safer alternative to IgG1 antibodies (33). Similar to solanezumab, this antibody binds to the mid-domain of Aβ and does bind Aβ monomers (34-36); however, it also binds to other Aβ species including Aβ oligomers (33, 36). Given the large excess of Aβ monomers compared to oligomers in brain, the significance of the binding to oligomers is unclear. In clinical trials, crenezumab did not demonstrate clinical benefit at doses up to 15 mg/kg, but like solanezumab also did not result in ARIA (34, 35). In January 2019 the Phase 3 trials of crenezumab using a higher dose of 60 mg/kg were stopped based on futility, but the data from these Phase 3 studies are not yet available.

Monoclonal antibodies primarily targeting Aβ oligomers

At this time, only one antibody with specificity for Aβ oligomers has entered Phase 1 clinical trials (37, 38). As reviewed by Cline et al (37) Aβ oligomers may target an Aβ species that has substantial toxicity, and targeting this Aβ species may not be associated with ARIA. Future clinical data will determine whether this target and antibody have important advantages over other antibodies as previously discussed.


Summary of clinical data for monoclonal antibodies showing possible clinical efficacy in Phase 2 or 3 clinical trials

Several monoclonal antibodies have shown probable efficacy with varying degrees of statistical significance. The general consistency of outcomes with various monoclonal antibodies as shown in Table 1 suggests strongly that these changes are biologically mediated. The obvious outliers in these studies are the results for the CDR-SB and MMSE for the aducanumab ENGAGE trial. The reasons for these discrepancies are not fully clear, but higher drug exposure in EMERGE than ENGAGE is a likely factor. Table 1 provides a comparison of these results from different monoclonal antibodies studied in different clinical trials and shows an overall consistency in drug effects.


Future directions in AD drug development

The accelerated approval by FDA of aducanumab marks a new era of AD treatment. Studies of four different antibodies indicate that substantial reduction of fibrillar amyloid or Aβ monomers in brain is feasible and is associated with slowing of cognitive/clinical progression. Aducanumab and the other antibodies in clinical development are unlikely to be a complete solution to the epidemic of AD. Nevertheless, there is now an opportunity to build upon this initial success. Other targets such as tau as well as microglia and other NANT targets may provide additional benefit alone or in combination. Many investigators in the field believe that earlier intervention, at the pre-symptomatic stage of the Alzheimer’s continuum, will lead to better outcomes (39, 40).
The treatment of Human Immunodeficiency Virus (HIV) has evolved from a modest effect on an ultimately fatal disease to potent combination therapies which have changed the infection to a manageable chronic disease (41). In AD, we have now seen the equivalent of the first serine protease inhibitor for the treatment of HIV. With further drug development in AD, this disease can be changed from an inexorable and fatal decline in cognition and function in late life, to a manageable condition that allows patients and families to enjoy their retirements, travel, and grandchildren.


Acknowledgements: The assistance of Karen Sundell BS in review of the manuscript and references is greatly appreciated.

Conflict of interests: Dr. Siemers reports personal fees from Acumen Pharmaceuticals Inc., personal fees from Acelot Inc., personal fees from Aquestive Therapeutics Inc., personal fees from Athira Pharma, Inc., personal fees from Biogen, Inc., personal fees from Cogstate, Ltd., personal fees from Cortexyme, Inc., personal fees from Gates Ventures, LLC, personal fees from Hoffman La-Roche, Ltd., personal fees from Indiana University, personal fees from LuMind Research Down Syndrome, personal fees from Partner Therapeutics, Inc., personal fees from Pinteon Therapeutics, Inc., personal fees from Prothena, Inc., personal fees from Vaccinex, Inc., personal fees from Washington University (St. Louis), outside the submitted work. Dr. Aisen reports grants from Janssen, grants from Lilly, grants from Eisai, grants from NIA, grants from the Alzheimer’s Association, grants from FNIH, personal fees from Biogen, personal fees from Roche, personal fees from Merck, personal fees from Abbvie, personal fees from Shionogi, personal fees from Immunobrain Checkpoint, outside the submitted work. Dr. Carrillo has nothing to disclose.



1. Fukumoto H A-OA, Suzuk N, Shimada H, Ihara Y, Iwatsubo T. Amyloid Beta Protein Deposition in Normal Aging Has the Same Characteristics as That in Alzheimer’s Disease. American Journal of Pathology;148(1):259-265.
2. Jack CR, Knopman DS, Jagust WJ, et al. Tracking pathophysiological processes in Alzheimer’s disease: an updated hypothetical model of dynamic biomarkers. The Lancet Neurology 2013;12(2):207-216. DOI: 10.1016/s1474-4422(12)70291-0.
3. Bateman RJ, Xiong C, Benzinger TL, et al. Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med 2012;367(9):795-804. DOI: 10.1056/NEJMoa1202753.
4. Siemers E. Commentary: Combination Therapy for Alzheimer’s Disease: Perspectives of the EU/US CTAD Task Force. J Prev Alzheimers Dis 2019;6(3):180-181. DOI: 10.14283/jpad.2019.19.
5. Hendrix JA, Bateman RJ, Brashear HR, et al. Challenges, solutions, and recommendations for Alzheimer’s disease combination therapy. Alzheimers Dement 2016;12(5):623-30. DOI: 10.1016/j.jalz.2016.02.007.
6. Panza F, Lozupone M, Logroscino G, Imbimbo BP. A critical appraisal of amyloid-beta-targeting therapies for Alzheimer disease. Nat Rev Neurol 2019;15(2):73-88. DOI: 10.1038/s41582-018-0116-6.
7. Steiner H, Fluhrer R, Haass C. Intramembrane proteolysis by gamma-secretase. J Biol Chem 2008;283(44):29627-31. DOI: 10.1074/jbc.R800010200.
8. Bateman RJ, Siemers ER, Mawuenyega KG, et al. A gamma-secretase inhibitor decreases amyloid-beta production in the central nervous system. Ann Neurol 2009;66(1):48-54. DOI: 10.1002/ana.21623.
9. Doody RS, Raman R, Farlow M, et al. A phase 3 trial of semagacestat for treatment of Alzheimer’s disease. N Engl J Med 2013;369(4):341-50. DOI: 10.1056/NEJMoa1210951.
10. Coric V, van Dyck CH, Salloway S, et al. Safety and tolerability of the gamma-secretase inhibitor avagacestat in a phase 2 study of mild to moderate Alzheimer disease. Arch Neurol 2012;69(11):1430-40. DOI: 10.1001/archneurol.2012.2194.
11. Jonsson T, Atwal JK, Steinberg S, et al. A mutation in APP protects against Alzheimer’s disease and age-related cognitive decline. Nature 2012;488(7409):96-9. DOI: 10.1038/nature11283.
12. Panza F, Lozupone M, Watling M, Imbimbo BP. Do BACE inhibitor failures in Alzheimer patients challenge the amyloid hypothesis of the disease? Expert Rev Neurother 2019;19(7):599-602. DOI: 10.1080/14737175.2019.1621751.
13. Sevigny J, Chiao P, Bussiere T, et al. The antibody aducanumab reduces Abeta plaques in Alzheimer’s disease. Nature 2016;537(7618):50-6. DOI: 10.1038/nature19323.
14. Budd Haeberlein S, O’Gorman J, Chiao P, et al. Clinical Development of Aducanumab, an Anti-Abeta Human Monoclonal Antibody Being Investigated for the Treatment of Early Alzheimer’s Disease. J Prev Alzheimers Dis 2017;4(4):255-263. DOI: 10.14283/jpad.2017.39.
15. Alexander GC, Emerson S, Kesselheim AS. Evaluation of Aducanumab for Alzheimer Disease: Scientific Evidence and Regulatory Review Involving Efficacy, Safety, and Futility. JAMA 2021. DOI: 10.1001/jama.2021.3854.
16. Demattos RB, Lu J, Tang Y, et al. A plaque-specific antibody clears existing beta-amyloid plaques in Alzheimer’s disease mice. Neuron 2012;76(5):908-20. DOI: 10.1016/j.neuron.2012.10.029.
17. Mintun MA, Lo AC, Duggan Evans C, et al. Donanemab in Early Alzheimer’s Disease. New England Journal of Medicine 2021. DOI: 10.1056/NEJMoa2100708.
18. Ostrowitzki S, Deptula D, Thurfjell L, et al. Mechanism of amyloid removal in patients with Alzheimer disease treated with gantenerumab. Arch Neurol 2012;69(2):198-207. DOI: 10.1001/archneurol.2011.1538.
19. 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. DOI: 10.3233/JAD-2011-110977.
20. Ostrowitzki S, Lasser RA, Dorflinger E, et al. A phase III randomized trial of gantenerumab in prodromal Alzheimer’s disease. Alzheimers Res Ther 2017;9(1):95. DOI: 10.1186/s13195-017-0318-y.
21. Lannfelt L, Relkin NR, Siemers ER. Amyloid-ss-directed immunotherapy for Alzheimer’s disease. J Intern Med 2014;275(3):284-95. DOI: 10.1111/joim.12168.
22. 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-Aβ protofibril antibody. Alzheimer’s Research & Therapy 2021;13(1). DOI: 10.1186/s13195-021-00813-8.
23. Rinne JO, Brooks DJ, Rossor MN, et al. 11C-PiB PET assessment of change in fibrillar amyloid-β load in patients with Alzheimer’s disease treated with bapineuzumab: a phase 2, double-blind, placebo-controlled, ascending-dose study. The Lancet Neurology 2010;9(4):363-372. DOI: 10.1016/s1474-4422(10)70043-0.
24. Sperling R, Salloway S, Brooks DJ, et al. Amyloid-related imaging abnormalities in patients with Alzheimer’s disease treated with bapineuzumab: a retrospective analysis. The Lancet Neurology 2012;11(3):241-249. DOI: 10.1016/s1474-4422(12)70015-7.
25. Vandenberghe R, Rinne JO, Boada M, et al. Bapineuzumab for mild to moderate Alzheimer’s disease in two global, randomized, phase 3 trials. Alzheimers Res Ther 2016;8(1):18. DOI: 10.1186/s13195-016-0189-7.
26. Salloway S, Sperling R, Gilman S, et al. A phase 2 multiple ascending dose trial of bapineuzumab in mild to moderate Alzheimer disease. Neurology 2009;73(24):2061-70. DOI: 10.1212/WNL.0b013e3181c67808.
27. Salloway S, Sperling R, Fox NC, et al. Two phase 3 trials of bapineuzumab in mild-to-moderate Alzheimer’s disease. N Engl J Med 2014;370(4):322-33. DOI: 10.1056/NEJMoa1304839.
28. Farlow M, Arnold SE, van Dyck CH, et al. Safety and biomarker effects of solanezumab in patients with Alzheimer’s disease. Alzheimers Dement 2012;8(4):261-71. DOI: 10.1016/j.jalz.2011.09.224.
29. Siemers ER, Friedrich S, Dean RA, et al. Safety and changes in plasma and cerebrospinal fluid amyloid beta after a single administration of an amyloid beta monoclonal antibody in subjects with Alzheimer disease. Clin Neuropharmacol 2010;33(2):67-73. DOI: 10.1097/WNF.0b013e3181cb577a.
30. Doody RS, Thomas RG, Farlow M, et al. Phase 3 trials of solanezumab for mild-to-moderate Alzheimer’s disease. N Engl J Med 2014;370(4):311-21. DOI: 10.1056/NEJMoa1312889.
31. Honig LS, Vellas B, Woodward M, et al. Trial of Solanezumab for Mild Dementia Due to Alzheimer’s Disease. N Engl J Med 2018;378(4):321-330. DOI: 10.1056/NEJMoa1705971.
32. Siemers ER, Sundell KL, Carlson C, et al. Phase 3 solanezumab trials: Secondary outcomes in mild Alzheimer’s disease patients. Alzheimers Dement 2016;12(2):110-120. DOI: 10.1016/j.jalz.2015.06.1893.
33. Adolfsson O, Pihlgren M, Toni N, et al. An effector-reduced anti-beta-amyloid (Abeta) antibody with unique abeta binding properties promotes neuroprotection and glial engulfment of Abeta. J Neurosci 2012;32(28):9677-89. DOI: 10.1523/JNEUROSCI.4742-11.2012.
34. Cummings JL, Cohen S, van Dyck CH, et al. ABBY: A phase 2 randomized trial of crenezumab in mild to moderate Alzheimer disease. Neurology 2018;90(21):e1889-e1897. DOI: 10.1212/WNL.0000000000005550.
35. Salloway S, Honigberg LA, Cho W, et al. Amyloid positron emission tomography and cerebrospinal fluid results from a crenezumab anti-amyloid-beta antibody double-blind, placebo-controlled, randomized phase II study in mild-to-moderate Alzheimer’s disease (BLAZE). Alzheimers Res Ther 2018;10(1):96. DOI: 10.1186/s13195-018-0424-5.
36. Ultsch M, Li B, Maurer T, et al. Structure of Crenezumab Complex with Abeta Shows Loss of beta-Hairpin. Sci Rep 2016;6:39374. DOI: 10.1038/srep39374.
37. Cline EN, Bicca MA, Viola KL, Klein WL. The Amyloid-beta Oligomer Hypothesis: Beginning of the Third Decade. J Alzheimers Dis 2018;64(s1):S567-S610. DOI: 10.3233/JAD-179941.
38. Cline E. VK, Klein W., Wang X., Bacskai B., Rammes G., Dodart J., Palop J., Siemers E., Jerecic J., Krafft G. Synaptic intervention in alzheimer’s disease: soluble aβ oligomer directed ACU193 monoclonal antibody therapeutic for treatment of early alzheimer’s disease. J Prevent Alzheimer’s Dis 2019;6:Supplement 1:S151.
39. Aisen PS, Cummings J, Jack CR, Jr., et al. On the path to 2025: understanding the Alzheimer’s disease continuum. Alzheimers Res Ther 2017;9(1):60. DOI: 10.1186/s13195-017-0283-5.
40. Sperling RA, Jack CR, Jr., Aisen PS. Testing the right target and right drug at the right stage. Sci Transl Med 2011;3(111):111cm33. DOI: 10.1126/scitranslmed.3002609.
41. Gunthard HF, Saag MS, Benson CA, et al. Antiretroviral Drugs for Treatment and Prevention of HIV Infection in Adults: 2016 Recommendations of the International Antiviral Society-USA Panel. JAMA 2016;316(2):191-210. DOI: 10.1001/jama.2016.8900.




M.G.H.E. den Brok1,2, M.P. Hoevenaar-Blom2, N. Coley5,6, S. Andrieu5,6, J. van Dalen1,2, Y. Meiller7, J. Guillemont5, C. Brayne8, W.A. van Gool3, E.P. Moll van Charante3,4, E. Richard1,3 On behalf of the preDIVA and MAPT/DSA groups*


1. Radboud University Medical Center, Donders Institute for Brain, Cognition, and Behaviour, Department of Neurology, Nijmegen, the Netherlands; 2. Amsterdam University Medical Center, location AMC, Department of Neurology, Amsterdam, the Netherlands; 3. Amsterdam University Medical Center, location AMC, Department of Public and Occupational Health, Amsterdam, the Netherlands; 4. Amsterdam University Medical Center, location AMC, Department of General Practice, Amsterdam, the Netherlands; 5. Centre for Epidemiology and Research in Population health (CERPOP), INSERM-University of Toulouse UMR 1295, Toulouse, France; 6. Department of Epidemiology and Public Health, Toulouse University Hospital, Faculty of Medicine, Toulouse, France; 7. Department of Information and Operations Management, ESCP Europe, Paris, France; 8. Cambridge Public Health, University of Cambridge, Department of Psychiatry, Cambridge, United Kingdom; *The members of the preDIVA and MAPT/DSA study groups are listed at the end of the manuscript.

Corresponding Author: Melina G.H.E. den Brok, MD, Department of Neurology (910), Radboud University Medical Center, PO Box 9101, 6500 HB Nijmegen, the Netherlands,

J Prev Alz Dis 2021;
Published online September 15, 2021,



BACKGROUND: Cardiovascular risk factors and lifestyle factors are associated with an increased risk of cognitive decline and dementia in observational studies, and have been targeted by multidomain interventions.
OBJECTIVES: We pooled individual participant data from two multi-domain intervention trials on cognitive function and symptoms of depression to increase power and facilitate subgroup analyses.
DESIGN:Pooled analysis of individual participant data.
SETTING: Prevention of Dementia by Intensive Vascular Care trial (preDIVA) and Multidomain Alzheimer Preventive Trial (MAPT).
PARTICIPANTS: Community-dwelling individuals, free from dementia at baseline.
INTERVENTION: Multidomain interventions focused on cardiovascular and lifestyle related risk factors.
MEASUREMENTS: Data on cognitive functioning, depressive symptoms and apathy were collected at baseline, 2 years and 3-4 years of follow-up as available per study. We analyzed crude scores with linear mixed models for overall cognitive function (Mini Mental State Examination [MMSE]), and symptoms of depression and apathy (15-item Geriatric Depression Scale). Prespecified subgroup analyses were performed for sex, educational level, baseline MMSE <26, history of hypertension, and history of stroke, myocardial infarction and/or diabetes mellitus.
RESULTS: We included 4162 individuals (median age 74 years, IQR 72, 76) with a median follow-up duration of 3.7 years (IQR 3.0 to 4.1 years). No differences between intervention and control groups were observed on change in cognitive functioning scores and symptoms of depression and apathy scores in the pooled study population. The MMSE declined less in the intervention groups in those with MMSE <26 at baseline (N=250; MD: 0.84; 95%CI: 0.15 to 1.54; p<0.001).
CONCLUSIONS: We found no conclusive evidence that multidomain interventions reduce the risk of global cognitive decline, symptoms of depression or apathy in a mixed older population. Our results suggest that these interventions may be more effective in those with lower baseline cognitive functioning. Extended follow-up for dementia occurrence is important to inform on the potential long-term effects of multidomain interventions.

Key words: Multidomain intervention trials, cognition, depression, apathy, pooled analysis.



The global prevalence of dementia is expected to triple in the coming decades. Over 50 million individuals were living with dementia in 2019, and this number might rise to 152 million by 2050 (1). Around 30-40% of dementia cases might be attributable to potentially modifiable risk factors such as midlife hypertension, depression and physical inactivity (2–4). However, evidence from randomized controlled trials targeting these risk factors is inconsistent (5, 6).
Several large multidomain intervention studies using cognitive functioning or dementia as primary outcome have been performed in older persons from the general population free from dementia at baseline (7–9). Two major trials, the prevention of dementia by intensive vascular care trial (preDIVA) and Multidomain Alzheimer Preventive Trial (MAPT) reported no significant effect on their respective primary outcomes dementia and cognitive decline (10, 11). The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability trial (FINGER) is the only study so far which reported a modest excess improvement of cognitive functioning in the intervention group compared to an improvement in the control group (12).
Depression is a potentially modifiable risk factor for dementia (3), and there is a bidirectional association between depression and cardiovascular risk factors and events (13-17). Apathy is associated with cardiovascular risk factors and increased risk of dementia (18, 19). A multidomain intervention targeting the same risk factors as those for dementia, could potentially reduce depressive symptoms and apathy. However, none of the trials reported significant effects of the interventions on symptoms of depression, and apathy was not reported.
Individual multidomain trials may have suffered from lack of power, due to the improvement in those in the control condition. In the preDIVA study, subgroup analyses suggested a potential beneficial effect in those with untreated hypertension, those with no history of cardiovascular disease at baseline, and in males (10). A more personalized approach, with interventions tailored to specific subgroups, could potentially lead to better intervention effects than a one-size-fits-all approach.
We pooled individual participant data from two large multidomain intervention trials targeting cardiovascular and lifestyle related risk factors to increase power and allow for subgroup analyses to detect possible intervention effects on cognition, symptoms of depression and apathy.



Study design and participants

We combined individual participant data from two large multidomain intervention trials targeting vascular and lifestyle-related risk factors in older people (10, 11). Both studies were European multicenter randomized controlled trials and included 3526 (preDIVA) and 1679 (MAPT) community dwelling individuals free from dementia at baseline, recruited from either general practices (preDIVA) or memory centers (MAPT). Individuals were aged 70 years or older and participants in MAPT were at increased risk for cognitive decline, operationalized as at least one of the three following criteria: spontaneous memory complaint expressed, limitation in one instrumental activity of daily living, or slow gait speed. The multidomain interventions consisted of individual or group sessions providing lifestyle advice concerning diet, physical activity and vascular risk factors. The MAPT study included cognitive training in the intervention. If needed, participants were advised to contact their general practitioner for optimization or initiation of drug treatment for cardiovascular risk factors. Control groups received usual care (10, 11). Main characteristics of the trials are specified in table 1. Study protocols have been published previously (20, 21). The MAPT study had a factorial design, with omega-3 supplementation in an additional arm (but there was no effect of this intervention on the trial’s primary or secondary outcomes). In this analysis we only evaluate the multidomain intervention. The ethics committees in the respective medical centers approved both trials and all individuals gave written informed consent. Both trials were registered, respectively in the ISRCTN registry (preDIVA: ISRCTN29711771) and on (MAPT: NCT00672685).

Table 1. Main characteristics of pooled trials


Global cognitive function was assessed with the Mini-Mental State Examination (MMSE). MMSE items were divided into anterograde episodic memory, evaluated with the delayed recall item (item 5, max 3 points) or ‘other cognitive functions’ (all other items). Both studies used different neuropsychological tests for memory (Visual Association Test (22) in preDIVA and Free and Cued Selective Reminding (23) in MAPT), precluding the possibility to use these more extensive memory tests in our analyses. As measure of subjective memory loss we used question 10 of the 15-item Geriatric Depression Scale (GDS-15): ‘Do you feel you have more problems with memory than most?’. Depressive symptoms were quantified using the crude scores of the 15-item Geriatric Depression Scale (GDS-15) and using a dichotomized cut-off (GDS-15 >5: indicative of depression; ≤5: not indicative of depression). We additionally assessed the effect on apathy, operationalized as the three apathy items from the GDS-15 (GDS-3A), as has previously been shown to be an appropriate screening instrument for symptoms of apathy (24).

Statistical Analysis

We included participants with at least one follow-up visit to analyze the effect of the intervention on cognitive decline, symptoms of depression and apathy. Mean differences in change between the intervention and control group were calculated using linear mixed regression models. Crude continuous and binary scores were used as outcome, adjusted for baseline scores of the specific tests using analysis of covariance (ANCOVA). We used follow-up measures of the outcome of interest as outcome and treatment allocation as predictor, including random intercepts for participant and health center level. Model fit was estimated using Akaike Information Criterion values, and more complex models (i.e. with additional random slope for randomization allocation or the time difference between randomization and the follow-up visit) did not result in a better fit. Analyses were adjusted for the baseline measure of the outcome of interest, study, age, sex, time in study and also for LDL cholesterol and systolic blood pressure because of significant differences in baseline values between randomization groups. Furthermore, linear interaction for time in study was assessed by means of an interaction term. Predefined subgroup analyses were performed for 1) sex, since multidomain interventions may have differential effects in men and women, 2) baseline hypertension status defined as history of hypertension and/or systolic blood pressure of >140 mmHg, 3) history of myocardial infarction, stroke and/or diabetes at baseline, since there is a high probability that individuals with a history of cardiovascular disease already receive a form of intervention and therefore, there might be more room for improvement in participants without a history of cardiovascular disease, 4) baseline MMSE <26, since individuals with lower baseline cognitive functioning might benefit more from a multidomain intervention, 5) educational level (low and high) because this is an important risk factor for dementia and is also associated with low socioeconomic status (3). Additional analyses were performed by study, to assess heterogeneity between the different studies. A p-value for interaction <0.05 was considered to reflect a significant interaction. Post-hoc analyses included exploration of differential dropout by comparing baseline characteristics of individuals included in the current study to individuals without any follow-up visits, and subgroup analyses for different cutoff values of baseline MMSE. Analyses were conducted in Rstudio (version 3.6.1, package “lme4”) (25).



Of a total of 5205 individuals in the two studies, 4162 (80%) individuals had at least one follow-up visit and were included in the present analysis (median follow-up duration: 3.7 years; IQR 3.0 to 4.14 years). The median age at baseline was 74 years (IQR 72, 76 years) and slightly more women were included (57.7%). No significant between-group differences in baseline characteristics were found in the pooled population, except for mean systolic blood pressure (control 149.3, SD 20.9; intervention 151.2, SD 22.1; p=0.006) and mean LDL cholesterol (control 3.19, SD 0.97; intervention 3.10, SD 0.95; p=0.003) (Table 2). eTable 1 shows baseline characteristics of included individuals per study. Participants who did not have a follow-up visit were slightly older, had a lower educational level, a slightly lower MMSE score, and a higher mean systolic blood pressure (eTable 2).
There were no differences in change from baseline to 3-4 year follow-up in MMSE and GDS scores between the control and intervention groups: Total MMSE score deteriorated by 0.09 points in the control group versus 0.05 points in the intervention group (mean difference in change [MDc] between intervention and control group: 0.03; 95% confidence interval [95%CI] -0.06 to 0.13). Total GDS score deteriorated by 0.21 points in the intervention group versus 0.22 points in the control group (MDc between intervention and control group: -0.04, 95%CI -0.16 to 0.07). The GDS-apathy score deteriorated by 0.12 points in both randomization groups (MDc between intervention and control group: -0.007 (-0.05 to 0.04). There was no time by treatment interaction for any of the outcome variables.

Table 2. Baseline characteristics of the pooled sample

All individuals with at least one follow-up visit, and comparison of the intervention and control group. Differences between randomization groups in spite of the large numbers, may partly be due to cluster randomization in preDIVA. P-value for comparison between control and intervention. Abbreviations: BMI: Body Mass Index; IQR: interquartile range; LDL cholesterol: low-density lipoprotein cholesterol; SD: standard deviation

Table 3. Mean difference in effect of multidomain interventions on cognition, symptoms of depression and apathy after 3-4 years of follow-up

* high score indicates better results; † low score indicates better results. Abbreviations: GDS: Geriatric Depression Scale; GDS15-Q10: question 10 of the Geriatric Depression Scale: ‘Do you feel you have more problems with memory than most?’; MDc: mean difference in change between the intervention and control group; MMSE: Mini-Mental State Examination; 95%CI: 95% confidence interval


Subgroup Analysis

In individuals with a baseline MMSE score <26, total MMSE score improved over time in both randomization groups, but more in the intervention group (MDc 0.84 points, 95%CI 0.15 to 1.54). Similar effects were seen for anterograde episodic memory (MDc 0.21, 95%CI -0.01 to 0.43) and other MMSE items (MDc 0.74, 95%CI 0.26 to 1.21). In those with a baseline MMSE score ≥26, some deterioration or no change in MMSE scores (total, memory or other items) was seen with similar effects in both randomization groups (Table 5, eTable 3). No significant differences between randomization groups in change in test scores for cognitive functioning, depressive symptoms or apathy were found for sex, educational level, baseline hypertension status and history of cardiovascular disease or diabetes at baseline (Table 4, Table 5). In MAPT, stronger deterioration on other MMSE items was seen in the control group compared to the intervention group, but there was no effect in preDIVA. None of the other outcomes were significantly different between both studies (eTable 4).
To assess the consistency of the effects favoring the intervention in individuals with a low baseline MMSE, we performed post-hoc subgroup analyses stratified for various baseline MMSE cutoff scores (eTable 5). The highest MDc were found with a baseline MMSE cut-off score <26, and with increasing baseline MMSE scores, MDc between the intervention and control group gradually decreased. Individuals with MMSE score <26 at baseline were significantly lower educated and had a higher Body Mass Index at baseline (eTable 6). Additional adjustment for these covariates did not significantly change the MDc between the intervention and control group in total MMSE score (baseline MMSE ≤26: MMSE MDc 0.77, 95%CI 0.07 to 1.47; p<0.001). There was no differential dropout between both randomization groups in individuals with a baseline MMSE score <26 and > 26 points.

Table 4. Mean difference in effect of multidomain interventions on cognition, symptoms of depression and apathy in subgroups (biological and clinical factors) after 3-4 years of follow-up

The given numbers represent the number of individuals in the different subgroups in the full cohort. These numbers slightly differed in each analysis. * high score indicates better results; † low score indicates better results. Abbreviations: GDS: Geriatric Depression Scale; GDS15-Q10: question 10 of the Geriatric Depression Scale: ‘Do you feel you have more problems with memory than most?’; MDc: mean difference in change between the intervention and control group; MMSE: Mini-Mental State Examination; 95%CI: 95% confidence interval

Table 5. Mean difference in effect of multidomain interventions on cognition, symptoms of depression and apathy in subgroups (baseline cognition and educational level) after 3-4 years of follow-up

The given numbers represent the number of individuals in the different subgroups in the full cohort. These numbers slightly differed in each analysis. * high score indicates better results; † low score indicates better results. Abbreviations: GDS: Geriatric Depression Scale; GDS15-Q10: question 10 of the Geriatric Depression Scale: ‘Do you feel you have more problems with memory than most?’; MDc: mean difference in change between the intervention and control group; MMSE: Mini-Mental State Examination; 95%CI: 95% confidence interval



This pooled analysis of two large randomized controlled trials in community-dwelling individuals over 60 years old did not show an overall effect of multidomain interventions on cognitive function or symptoms of depression or apathy after 3-4 years follow-up. Subgroup analyses suggests that multidomain interventions may improve cognition in those with lower cognitive scores at baseline. We observed no interaction of the effect of the interventions with sex, history of stroke, diabetes mellitus and/or myocardial infarction, hypertension and educational level.

Strengths and limitations

The major strength of this study is that we pooled data on individual participant level from two large randomized controlled trials, providing more power to detect possible intervention effects on cognition, symptoms of depression and apathy, and to better allow for subgroup analyses to explore whether interventions may be more effective in specific subgroups. Furthermore, variation in inclusion criteria and multidomain interventions between the different trials improved the external validity of our overall results.
Several limitations should be noted. First of all, the use of the MMSE as outcome measure for cognitive studies has limitations. This time-honored test was designed as a cognitive screening test and does not measure cognitive function as comprehensively or sensitively, or with such detailed quantification, as a full neuropsychological evaluation does. Assessments using cognitive screening instruments such as the MMSE are known to show substantial variation over time, depending on conditions such as e.g. illness, stress or sleep deprivation, particularly in people without cognitive impairments (26), although the same holds for full neuropsychological evaluation. A possible random error caused by these fluctuations may have resulted in bias towards the null. Another disadvantage of the MMSE in populations such as under study here, is its ceiling effect. It lacks the ability to differentiate well between healthy individuals and early signs of dementia in individuals with MMSE scores in the range of 24-30. However, despite its limitations, as with any other screening test, the MMSE is characterized by an unprecedented dissemination and appreciation among the scientific epidemiological and dementia community. Moreover, these interventions were not designed to boost cognition, but to prevent cognitive decline. Less decline in MMSE was a more likely hypothesis than more increment in MMSE due to the intervention – nuancing the potential ceiling effect. Secondly, attrition bias could have influenced our results, since those who dropped out were significantly older, had a lower educational level, higher mean systolic blood pressure, and lower MMSE scores; i.e. they were at higher risk of cognitive decline. This could have biased our results to the null, since those with lower cognitive function appear to potentially benefit most from intervention. Lastly, while both studies were designed to test the efficacy of multidomain interventions in elderly, there were important differences in study design and populations between both trials which could have impacted the results of this study. The recruitment strategy of individuals in preDIVA was population-based through general practices, whereas MAPT recruited individuals at risk for cognitive decline through memory centers. Cognitive training was not part of the intervention in preDIVA, but the intervention in MAPT strongly focused on cognitive training in the intervention group, which was given in supervised sessions, in contrast to the physical activity and nutrition components which were simply based on advice. Moreover, cardiovascular risk factors in the intervention group in preDIVA were assessed every four months by a practice nurse, and if necessary, medical interventions such as drug treatment were advised according to a detailed protocol. The MAPT intervention assessed cardiovascular risk factors annually, and there was less focus on drug treatment. These differences might improve the external validity of our results, but they are likely to also cause heterogeneity. However, there were no structural differences in intervention effects between the MAPT and preDIVA intervention.
Although overall this study did not show beneficial effects of multidomain interventions on symptoms of depression and apathy and cognitive function, several methodological challenges associated with dementia prevention trials complicate the interpretation of these results. First, the development of cognitive impairment and dementia is a slow and insidious process, and risk factors in midlife have a stronger association with incident dementia than risk factors in later life (27–30). However, since inclusion of participants in midlife requires an unrealistically long follow-up (31), these trials included participants in later life, far beyond the stage of life in which interventions are expected to have their optimal effect. Second, follow-up durations for the current analyses were too short to detect an effect on incident dementia, as most clear clinically relevant outcome. Long-term interventions starting at a younger age might be needed to achieve clinically relevant effects.
Although observational studies have shown consistent evidence supporting the association of cardiovascular risk factors and lifestyle related factors with cognitive decline and dementia, and some interventions targeting these factors have shown to improve cognitive function (3, 4, 32), trials evaluating multidomain interventions report inconsistent results (10–12,33). Variation in intensity, components of the multidomain intervention, and characteristics of the control condition in the individual trials may partly explain these inconsistent results. Alternative explanations include that the associations found in observational studies do not reflect a causal relationship or that the harmful effects have been effectuated by the time these interventions started.
Subgroup analyses for individuals with low baseline MMSE score yielded consistent results with beneficial effects of the multidomain intervention on cognitive function: total MMSE score, including anterograde episodic memory and other MMSE items increased more in the intervention group. Similar effects on memory were seen in a multidomain intervention trial in black individuals with MCI (34). The size of these groups (MMSE <26) was small within the overall studies (n=296). However, the gradual decrease in MDc with increasing baseline MMSE score appeared to be consistent (Table 4), and the results did not change with additional adjustment for differences in baseline characteristics. Furthermore, results were still statistically significant after multiple-comparison correction (p < 0.01 after Bonferroni correction). These findings should be confirmed in the multidomain intervention trials that are currently underway (35–37).
The underlying etiology of dementia is heterogeneous, and a multidomain intervention tailored to specific subgroups could potentially lead to better intervention effects than a one-size-fits-all approach. The results of this study, showing that, in later life, this type of intervention may be more effective in those with lower baseline cognitive scores, should be considered in the design of future dementia prevention trials.



This study with pooled data at the individual participant level from two large, randomized controlled trials did not show conclusive evidence that multidomain interventions can reduce the risk of cognitive decline or symptoms of depression and apathy in a mixed older population. These interventions may be more effective in those with lower baseline cognitive function. Extended follow-up for dementia outcomes is important to evaluate whether multidomain interventions can indeed have beneficial effects crossing the threshold of minimal clinically important difference. These extended follow-ups are planned and ongoing in both trials.


Funding: The present study was financially supported by the European Union’s Seventh Framework Program (FP7, 2007- 2013, grant nr. 305374), as part of the HATICE project. The preDIVA study was financially supported by the Dutch Ministry of Health, Welfare and Sport (grant number 50-50110-98-020), the Dutch Innovation Fund of Collaborative Health Insurances (grant number 05-234), and the Netherlands Organisation for Health Research and Development (grant number 62000015). The MAPT study was supported by grants from the Gérontopôle of Toulouse, the French Ministry of Health (PHRC 2008, 2009), Pierre Fabre Research Institute (manufacturer of the omega-3 supplement), ExonHit Therapeutics SA, and Avid Radiopharmaceuticals Inc. The promotion of this study was supported by the University Hospital Center of Toulouse. The data sharing activity was supported by the Association Monegasque pour la Recherche sur la maladie d’Alzheimer (AMPA) and the UMR 1027 Unit INSERM-University of Toulouse III. 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.

Acknowledgements: The authors would like to thank all the participants to the preDIVA and MAPT studies, and the members of the preDIVA and MAPT trial teams for their help in data collection and management.

Declaration of Helsinki: The studies included comply with the Declaration of Helsinki, locally appointed ethics committees have approved the research protocols and informed consent has been obtained from all the subjects.

Conflict of interest disclosure: M.G.H.E. den Brok, M.P. Hoevenaar-Blom, N. Coley, J. van Dalen, C. Brayne, W.A. van Gool, E.P. Moll van Charante, E. Richard have nothing to disclose. Y. Meiller, J. Guillemont and S. Andrieu report grants from the European Union, during the conduct of the study (FP7, 2007- 2013, grant nr. 305374). S. Andrieu additionally reports grants from the French Ministry of health (PHRC 2008, 2009), Pierre Fabre Research Institute, and AMPA during the conduct of the study.

Study group: The members of the preDIVA study group: Eric P. Moll van Charante, Edo Richard, Lisa S. Eurelings, Jan-Willem van Dalen, Suzanne A. Ligthart, Emma F. van Bussel, Marieke P. Hoevenaar-Blom, Marinus Vermeulen, Willem A. van Gool (Amsterdam University Medical Center, location AMC, Amsterdam, the Netherlands). The members of the MAPT/DSA study group: MAPT study group: Principal investigator: Bruno Vellas (Toulouse); Coordination: Sophie Guyonnet; Project leader: Isabelle Carrié; CRA: Lauréane Brigitte; Investigators: Catherine Faisant, Françoise Lala, Julien Delrieu, Hélène Villars; Psychologists: Emeline Combrouze, Carole Badufle, Audrey Zueras; Methodology, statistical analysis, and data management: Sandrine Andrieu, Christelle Cantet, Christophe Morin; Multidomain group: Gabor Abellan Van Kan, Charlotte Dupuy, Yves Rolland (physical and nutritional components), Céline Caillaud, Pierre-Jean Ousset (cognitive component), Françoise Lala (preventive consultation). The cognitive component was designed in collaboration with Sherry Willis from the University of Seattle, and Sylvie Belleville, Brigitte Gilbert, and Francine Fontaine from the University of Montreal. Coinvestigators in associated centers: Jean-François Dartigues, Isabelle Marcet, Fleur Delva, Alexandra Foubert, Sandrine Cerda (Bordeaux); Marie-Noëlle Cuffi, Corinne Costes (Castres); Olivier Rouaud, Patrick Manckoundia, Valérie Quipourt, Sophie Marilier, Evelyne Franon (Dijon); Lawrence Bories, Marie-Laure Pader, Marie-France Basset, Bruno Lapoujade, Valérie Faure, Michael Li Yung Tong, Christine Malick-Loiseau, Evelyne Cazaban-Campistron (Foix); Françoise Desclaux, Colette Blatge (Lavaur); Thierry Dantoine, Cécile Laubarie-Mouret, Isabelle Saulnier, Jean-Pierre Clément, Marie-Agnès Picat, Laurence Bernard-Bourzeix, Stéphanie Willebois, Iléana Désormais, Noëlle Cardinaud (Limoges); Marc Bonnefoy, Pierre Livet, Pascale Rebaudet, Claire Gédéon, Catherine Burdet, Flavien Terracol (Lyon), Alain Pesce, Stéphanie Roth, Sylvie Chaillou, Sandrine Louchart (Monaco); Kristel Sudres, Nicolas Lebrun, Nadège Barro-Belaygues (Montauban); Jacques Touchon, Karim Bennys, Audrey Gabelle, Aurélia Romano, Lynda Touati, Cécilia Marelli, Cécile Pays (Montpellier); Philippe Robert, Franck Le Duff, Claire Gervais, Sébastien Gonfrier (Nice); Yannick Gasnier and Serge Bordes, Danièle Begorre, Christian Carpuat, Khaled Khales, Jean-François Lefebvre, Samira Misbah El Idrissi, Pierre Skolil, Jean-Pierre Salles (Tarbes). Biological sample collection: Bertrand Perret, Claire Vinel, Sylvie Caspar-Bauguil (Toulouse). Safety management: Pascale Olivier-Abbal. DSA Group: Sandrine Andrieu, Christelle Cantet, Nicola Coley.
Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, 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.





1. Alzheimer’s Disease International (ADI). World Alzheimer Report 2019: Attitudes to Dementia.; 2019. Accessed 07-12-2020
2. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. Lancet. 2017;390(10113):2673–734
3. Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396(10248):413–46
4. 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(8):788–94
5. Coley N, Andrieu S, Gardette V, et al. Dementia prevention: Methodological explanations for inconsistent results. Epidemiol Rev. 2008;30(1):35–66
6. Yaffe K. Modifiable risk factors and prevention of dementia what is the latest evidence? JAMA Intern Med. 2018;178(2):281–2
7. Espeland MA, Luchsinger JA, Baker LD, et al. Effect of a long-term intensive lifestyle intervention on prevalence of cognitive impairment. Neurology. 2017;88(21):2026–35
8. Chen LK, Hwang AC, Lee WJ, et al. Efficacy of multidomain interventions to improve physical frailty, depression and cognition: data from cluster-randomized controlled trials. J Cachexia Sarcopenia Muscle. 2020;11(3):650–62
9. Hafdi M, Hoevenaar-Blom MP, Richard E. Multi-domain interventions for the prevention of dementia and cognitive decline (Protocol). Cochrane Database Syst Rev. 2020;(4)
10. Moll van Charante E, Richard E, Eurelings LS, et al. Effectiveness of a 6-year multidomain vascular care intervention to prevent dementia (preDIVA): a cluster-randomised controlled trial. Lancet. 2016;388(10046):797–805
11. Andrieu S, Guyonnet S, Coley N, et al. Effect of long-term omega 3 polyunsaturated fatty acid supplementation with or without multidomain intervention on cognitive function in elderly adults with memory complaints (MAPT): a randomised, placebo-controlled trial. Lancet Neurol. 2017;16(5):377–89
12. 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. Lancet (London, England). 2015 Jun;385(9984):2255–63
13. Anderson RJ, Freedland KE, Clouse RE, Lustman PJ. The Prevalence of Comorbid Depression in Adults with Diabetes. Diabetes Care. 2001;24(6):1069–78
14. Gan Y, Gong Y, Tong X, et al. Depression and the risk of coronary heart disease: A meta-analysis of prospective cohort studies. BMC Psychiatry. 2014;14(1):1–11
15. Mezuk B, Eaton WW, Albrecht S, Golden SH. Depression and type 2 diabetes over the lifespan: A meta-analysis. Diabetes Care. 2008;31(12):2383–90
16. Meng L, Chen D, Yang Y, Zheng Y, Hui R. Depression increases the risk of hypertension incidence: A meta-analysis of prospective cohort studies. J Hypertens. 2012;30(5):842–51
17. Wu Q, Kling JM. Depression and the Risk of Myocardial Infarction and Coronary Death. Med (United States). 2016;95(6):1–9
18. Ligthart SA, Richard E, Fransen NL, et al. Association of vascular factors with apathy in community-dwelling elderly individuals. Arch Gen Psychiatry. 2012;69(6):636–42
19. Van Dalen JW, Van Wanrooij LL, Moll Van Charante EP, Brayne C, Van Gool WA, Richard E. Association of Apathy with Risk of Incident Dementia: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2018;75(10):1012–21
20. Richard E, Den Heuvel E Van, Moll Van Charante EP, et al. Prevention of dementia by intensive vascular care (prediva): A cluster-randomized trial in progress. Alzheimer Dis Assoc Disord. 2009;23(3):198–204
21. Vellas B, Carrie I, Gillette-Guyonnet S, et al. MAPT study: a multidomain approach for preventing Alzheimer’s disease: design and baseline data. J Prev Alzheimers Dis. 2014;1(1):13–22
22. Lindeboom J, Schmand B, Tulner L, Walstra G, Jonker C. Visual association test to detect early dementia of the Alzheimer type. J Neurol Neurosurg Psychiatry. 2002;73(2):126–33
23. Grober E, Buschke H, Crystal H, Bang S, Dresner R. Screening for dementia by memory testing. Neurology. 1988;38(6):900–3
24. van der Mast RC, Vinkers DJ, Stek ML, et al. Vascular disease and apathy in old age. The Leiden 85-plus Study. Int J Geriatr Psychiatry. 2008;23(3):266–71
25. Bates D, Mächler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models Using lme4. J Stat Softw. 2015;67(1):1–48
26. Richard E, Brayne C. Mild cognitive impairment – not always what it seems. Nat Rev Neurol. 2014;10(3):130–1
27. Kivimäki M, Luukkonen R, Batty GD, et al. Body mass index and risk of dementia: Analysis of individual-level data from 1.3 million individuals. Alzheimer’s Dement. 2018;14(5):601–9
28. Pedditizi E, Peters R, Beckett N. The risk of overweight/obesity in mid-life and late life for the development of dementia: A systematic review and meta-analysis of longitudinal studies. Age Ageing. 2016;45(1):14–21
29. Abell JG, Kivimäki M, Dugravot A, et al. Association between systolic blood pressure and dementia in theWhitehall II cohort study: Role of age, duration, and threshold used to define hypertension. Eur Heart J. 2018;39(33):3119–25
30. McGrath ER, Seshadri S. Author response: Blood pressure from mid- to late life and risk of incident dementia. Neurology. 2018;91(3):149
31. Richard E, Andrieu S, Solomon A, et al. Methodological challenges in designing dementia prevention trials – The European Dementia Prevention Initiative (EDPI). J Neurol Sci [Internet]. 2012;322(1–2):64–70. Available from:
32. Andrieu S, Coley N, Lovestone S, Aisen PS, Vellas B. Prevention of sporadic Alzheimer’s disease: Lessons learned from clinical trials and future directions. Lancet Neurol. 2015;14(9):926–44
33. Richard E, Moll van Charante EP, Hoevenaar-Blom MP, et al. Healthy ageing through internet counselling in the elderly (HATICE): a multinational, randomised controlled trial. Lancet Digit Heal. 2019;1(8):e424–34 33.
34. Rovner BW, Casten RJ, Hegel MT, Leiby B. Preventing Cognitive Decline in Black Individuals With Mild Cognitive Impairment: A Randomized Clinical Trial. JAMA Neurol. 2018;75(12):1487–1493
35. PRODEMOS consortium. Prevention of Dementia using Mobile Phone Applications (PRODEMOS). Accessed 16-02-2021.
36. Rosenberg A, Mangialasche F, Ngandu T, Solomon A, Kivipelto M. Multidomain Interventions to Prevent Cognitive Impairment, Alzheimer’s Disease, and Dementia: From FINGER to World-Wide FINGERS. J Prev Alzheimer’s Dis. 2020;7(1):29–36
37. de Souto Barreto P, Pothier K, Soriano G, et al. A Web-Based Multidomain Lifestyle Intervention for Older Adults: The eMIND Randomized Controlled Trial. J Prev Alzheimer’s Dis. 2021;8(2):142–50



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:;, Tel: +91-7488019194, ORCID:; Dr. Shreesh Ojha, Department of Pharmacology and Therapeutics, College of Medicine and Health Sciences, UAE University, PO Box – 17666, Al Ain, UAE, E-mail:, Tel: +971-3-7137524, ORCID:

J Prev Alz Dis 2021;4(8):534-551
Published online September 15, 2021,



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.



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 ( Identifier: NCT00479557), US ( 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 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 (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 (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 (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 (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).


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).


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.



1. LoGiudice D, & Watson R. Dementia in older people: an update. Internal Medicine Journal 2014; 44(11):1066–1073.
2. Mayeux R, & Stern Y. Epidemiology of Alzheimer disease. Cold Spring Harbor Perspectives in Medicine 2012; 2(8):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.
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.
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.
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.×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.
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.
10. Reitz C, Mayeux R. Alzheimer disease: epidemiology, diagnostic criteria, risk factors and biomarkers. Biochemical Pharmacology 2014; 88(4): 640–651.
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.
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.
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.
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.
15. Haass C. New hope for Alzheimer disease vaccine. Nature medicine 2002; 8(11):1195–1196.
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.
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.
20. Mattson MP. Pathways towards and away from Alzheimer’s disease. Nature 2004; 430(7000):631–639.
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.<358::aid-ana12>;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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
39. van Dyck CH. Anti-Amyloid-β Monoclonal Antibodies for Alzheimer’s Disease: Pitfalls and Promise. Biological Psychiatry 2018; 83(4):311–319.
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.
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.
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.
44. Zlokovic BV. Clearing amyloid through the blood-brain barrier. Journal of Neurochemistry 2004; 89(4):807–811.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
70. 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.
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.
73. 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.
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.
76. Grüninger F. Invited review: Drug development for tauopathies. Neuropathology and Applied Neurobiology 2015; 41(1):81–96.
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.
78. RobinsonHL & Pertmer TM. DNA vaccines for viral infections: basic studies and applications. Advances in Virus Research 2000; 55:1–74.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
96. Morgan D. Immunotherapy for Alzheimer’s disease. In Journal of Internal Medicine 2011; 269(1):54–63).
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.
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.
106. Aisen, P.S. Failure After Failure. What Next in AD Drug Development?. J Prev Alzheimers Dis 2019; 6:150.
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.
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.
111. Neurology TL & The Lancet Neurology. Solanezumab: too late in mild Alzheimer’s disease? In The Lancet Neurology 2017; 16(2):97.
112. Topline Result for First DIAN-TU Clinical Trial: Negative on Primary 10 Feb 2020
113. 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. 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.
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).
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.
128. Vellas B, Aisen P. New Hope for Alzheimer’s Disease. J Prev Alzheimers Dis 2021; 8, 238–239.
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).
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.
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.




T. Kawada


Department of Hygiene and Public Health, Nippon Medical School

Corresponding Author: Tomoyuki Kawada, Department of Hygiene and Public Health, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-Ku, Tokyo 113-8602, Japan,

J Prev Alz Dis 2021;4(8):552
Published online September 13, 2021,


Key words: Secondhand smoke, dementia, Alzheimer’s disease; stroke; risk assessment.


Dear Editor,

I read the paper by Zhou et al. with great interest (1). The authors conducted a long-term follow-up study to investigate the associations between exposure to secondhand smoke and the subsequent risk of dementia, Alzheimer’s disease (AD), and stroke (1). The adjusted hazard ratios (HRs; 95% confidence intervals [CIs]) of the highest exposure to secondhand smoke compared to individuals with no exposure for dementia, AD, and stroke were 2.86 (2.00–4.09), 3.13 (1.80–5.42), and 1.89 (1.37–2.61), respectively. In addition, a dose–response relationship was also observed in this study. I would like to present some information relevant to their study.
First, Rovio et al. also conducted a long-term follow-up study to evaluate the association between exposure to parental smoking during childhood/adolescence and midlife cognitive function (2). They classified the participants into three groups: G1) nonsmoking parents with the participants’ serum cotinine level <1.0 ng/mL; G2) 1–2 smoking parents with the participants’ serum cotinine level <1.0 ng/mL; and G3) 1–2 smoking parents with the participants’ serum cotinine level ≥1.0 ng/mL. Compared to participants in group G1, the relative risk (RR; 95% CI) of those in group G3 for midlife episodic memory and associative learning was 1.38 (1.08–1.75), and they were advised to avoid exposure to secondhand smoke during childhood/adolescence to maintain cognitive function in adulthood. This report was based on valid data with exposure to parental smoking evaluated by assessment of serum cotinine levels, and the RR (95% CI) of participants in group G3 for short-term and spatial working memory was 1.25 (0.98–1.58), presenting a tendency of the association. Although this report did not select dementia and AD as clinical outcomes, midlife cognitive impairment is closely related to the subsequent risk of dementia and AD.
Second, Pistilli et al. evaluated the association between exposure to secondhand smoke during childhood and the subsequent risk of stroke or coronary heart disease (CHD) among never-smokers (3). The adjusted HRs (95% CI) of ≥2 childhood household smokers for stroke and CHD were 1.66 (1.29–2.13) and 1.15 (0.82–1.59), respectively. Although the type of stroke could not be specified, a long-term effect of secondhand smoke on the risk of stroke was observed, which was consistent with the data presented by Zhou et al. Further studies are required to verify the association between stroke and the risk of dementia and AD.
The pathology of AD begins a few decades before onset of the clinical symptoms, and there are some modifiable risk and protective factors (4). In any case, comprehensive studies are required to verify the associations of lifestyle factors with dementia, AD, and stroke subtypes.


Conflict of interest: None declared.



1. Zhou S, Wang K. Childhood secondhand smoke exposure and risk of dementia, Alzheimer’s disease and stroke in adulthood: A prospective cohort study. J Prev Alzheimers Dis 2021;8(3):345-350,
2. Rovio SP, Pihlman J, Pahkala K, et al. Childhood exposure to parental smoking and midlife cognitive function. Am J Epidemiol 2020;189(11):1280-129,
3. Pistilli M, Howard VJ, Safford MM, et al. Association of secondhand tobacco smoke exposure during childhood on adult cardiovascular disease risk among never-smokers. Ann Epidemiol 2019;32:28-34.e1,
4. Zhang XX, Tian Y, Wang ZT, et al. The epidemiology of Alzheimer’s disease modifiable risk factors and prevention. J Prev Alzheimers Dis 2021;8(3):313-32,



J.D. Grill1,2,3,4, A. Kind5,6,7, D. Hoang1, D.L. Gillen1,8


1. Institute for Memory Impairments and Neurological Disorders, University of California Irvine, Irvine CA, USA; 2. Department of Psychiatry and Human Behavior, University of California Irvine, Irvine, California, USA; 3. Department of Neurobiology and Behavior, University of California Irvine, Irvine, California, USA; 4. Institute for Clinical and Translational Science, University of California Irvine, Irvine, California, USA; 5. Center for Health Disparities Research, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin, USA; 6. Department of Medicine, Division of Geriatrics, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin, USA; 7. Madison VA Geriatrics Research Education and Clinical Center, Middleton VA Hospital, Madison, Wisconsin, USA; 8. Department of Statistics, University of California Irvine, Irvine, California, USA

Corresponding Author: Joshua Grill, PhD, 3204 Biological Sciences III, University of California Irvine, Irvine, CA 92697,USA,, t: (949) 824-5905,
f: (949) 824-0885

J Prev Alz Dis 2021;
Published online August 26, 2021,



BACKGROUND: Disparities in clinical research participation perpetuate broader health disparities. Recruitment registries are novel tools to address known challenges in accrual to clinical research. Registries may accelerate accrual, but the utility of these tools to improve generalizability is unclear.
Objective: To examine the diversity of a local on-line recruitment registry using the Area Deprivation Index (ADI), a publicly available metric of neighborhood disadvantage.
Design: Retrospective analysis.
Setting: Data were collected in the University of California Irvine Consent-to-Contact Registry.
Participants: We categorized N=2,837 registry participants based on the ADI decile (collapsed into quintiles) using a state-based rankings.
Measurements: We examined the proportion of enrollees per ADI quintile and quantified the demographics of these groups. We assessed willingness to participate in studies involving unique research procedures among the ADI groups.
Results: Although registry enrollees represented the full spectrum of the ADI, they disproportionately represented less disadvantaged neighborhoods (lowest to highest quintiles: 42%, 30%, 15%, 6%, 7%). Compared to participants from less disadvantaged neighborhoods, participants from more disadvantaged neighborhoods were more often female, of non-white race, and Hispanic ethnicity. Despite demographic differences, ADI groups were observed to have similar willingness to participate in research studies.
Conclusions: People from more disadvantaged neighborhoods may be underrepresented in recruitment registries, increasing the risk that they will be underrepresented when using these tools to facilitate prospective recruitment to clinical research. Once enrolled in registries, participants from more disadvantaged neighborhoods may be equally willing to participate in research. Efforts to increase representation of participants from disadvantaged neighborhoods in registries could be an important first step toward increasing the generalizability of clinical research.

Key words: Registry, recruitment, neighborhood, diversity, disparities.




Clinical research is rarely inclusive of populations that reflect the full US population (1, 2). To address disparities in participation, researchers should consider applying recruitment approaches that are responsive to the mechanistic lens provided by the breadth of the National Institute on Aging Health Disparities Framework ( One of these considerations is socioeconomic contextual disadvantage, or “neighborhood disadvantage.” This construct can be measured using the census block-group level Area Deprivation Index (ADI), a marker for social determinants of health within these discrete geo-areas that may promote or impair human health such as employment, income, education and housing quality factors (3, 4). The most disadvantaged neighborhoods in the US as measured by ADI tend to have higher proportions of African American/Black, Hispanic and Native American residents; are often located within inner city urban or highly rural areas; and tend to have higher rates of complex chronic medical conditions like heart disease, diabetes and chronic pulmonary disease. Higher ADI is associated with poorer late life health outcomes (5-8), including risk for Alzheimer’s disease and related dementias (ADRD) (9-11). ADI is available freely to the public through the University of Wisconsin Neighborhood Atlas, a customizable mapping and data platform that makes this information easily accessible to investigators recruiting to prospective clinical research studies (3, 12).
Recruitment registries are relatively new tools meant to address a crisis in ADRD clinical research recruitment (13). Registries enroll large populations of potentially eligible and willing participants for research studies, in an effort to accelerate accrual to new studies once they begin. Several questions remain about the effectiveness of these tools (14), especially as it relates to diversifying research populations (15-17). We developed the UC Irvine Consent-to-Contact (C2C) Registry, a local on-line recruitment registry in Orange County, California.(18) Multiple strategies have been applied to enroll participants in the C2C Registry, including community, direct mail, and electronic outreach (18, 19). As yet, these strategies have not included specific geo-targeted approaches. In this exploratory study, we examined the representativeness of C2C Registry, based on state ADI deciles. To our knowledge, this is the first assessment of ADI within a recruitment registry. We also assessed whether specific recruitment techniques were more frequently sources of high ADI participants and whether ADI was associated with stated research preferences when enrolling in the C2C Registry.



Data source and participants

We performed an exploratory descriptive analysis of the C2C Registry for outcomes related to ADI using data from participants enrolled on or before 09/29/2020. This local on-line recruitment registry was developed and launched in 2016 to accelerate accrual to clinical research studies at the University of California Irvine (UCI), with a particular emphasis on preclinical AD trials.(18) To be eligible for the C2C, participants must be 18 years of age or older. All participants provided informed consent electronically. Registry enrollment requires completion of demographic and clinical questionnaires on-line, estimated to take approximately 20 minutes to complete. Demographic and clinical information is self-reported and has been described previously (18). Participants self-described race and ethnicity, queried as separate categories, each offering a “prefer not to answer” option. Recruitment sources were captured at enrollment and included earned media, community outreach activities, postcard mailings, e-mail, Internet, social media, and referrals from physicians or others. Nine questions determined enrollee’s willingness to be contacted for studies that involve: (1) modification of diet or physical activity, (2) cognitive testing, (3) blood draws, (4) magnetic resonance imaging (MRI), (5) positron emission tomography (PET) imaging, (6) FDA approved medications, (7) investigational medications, (8) lumbar puncture (LP), and (9) autopsy. Research attitudes were assessed by the validated Research Attitude Questionnaire (RAQ(20)), which uses Likert scales to examine participants’ agreement with 7-items, scored 1-5 (Range: 7-35), with higher scores indicating more positive research attitudes. Enrollees also complete the Cognitive Function Instrument (CFI), a 14-item measure of subjective cognitive performance (Range: 0-14), with higher scores indicating more complaints (21, 22).

ADI assessment

We used C2C enrollees’ permanent addresses to determine their ADI. The ADI incorporates 17 measures originally drawn from the long-form Census related to education, employment, housing-quality, and poverty (7), to rank the deprivation of US census block groups (~1500 people). From these, an index ranking is created to compare a specific census block to state or national norms, typically presented as deciles (3).
The ADI can be used for research purposes. For example, using the ADI based on 2000 Census data, Kind et al (7) found that the risk of living in a disadvantaged neighborhood is similar to that of having a chronic lung disease, like emphysema, and worse than that of health conditions such as diabetes when it comes to readmission risk. Using the ADI, Joynt Maddox and colleagues (23) added social risk factors including neighborhood disadvantage to models used to calculate penalties under the CMS’s Hospital Readmission Reduction Program. They found that accounting for these factors had a major impact on safety-net hospitals that serve patients from the most disadvantaged neighborhoods; over half would have seen a decline in their readmission penalty if such an adjustment had been applied. Most recently, the ADI has also been employed for COVID vaccine allocation in a number of US states as a means by which to most efficiently and effectively allocate resources to areas of greatest need (24).
To use the ADI, we downloaded the data through the Neighborhood Atlas ( and linked to C2C enrollee addresses using 12-digit Federal Information Processing Standards (FIPS) code via the US Census Bureau Geocoder ( C2C records with a 12-digit FIPS code were then matched to a locally download California 2015 ADI v2 dataset where ADI scores were obtained.
All ADI were calculated at the block group level. We examined C2C enrollees’ ADI using state-based norms. Adequate information to determine ADI was missing for N=1284 records (e.g. providing a PO Box, rather than a street address at enrollment). We also compared the C2C ADI distributions to the larger Orange County population, using data from the 2019: American Communities Survey 5-Year Estimates Detailed Tables (


The Institutional Review Board at UCI approved this study.


We assessed the relative representation of ADI deciles among enrolled C2C participants. We hypothesized that high ADI participants are underrepresented in this recruitment registry. We used geocoding maps to illustrate the distribution of ADI decile representation among C2C enrollees. We used descriptive statistics (mean and standard deviation for continuous responses, and frequency and percentage for discrete responses) to summarize the demographic characteristics of C2C enrollees by ADI, discretized into California state-specific quintiles. We further quantified willingness to participate across ADI quintiles. To do so, we characterized the frequency with which individuals from the differing ADI categories agreed to be contacted about studies that required the nine research procedures noted above. Given the descriptive nature of the research presented, inferential statements are not presented to avoid over-interpretation of exploratory results.



Among 4315 participants enrolled in the C2C Registry as of 09/29/2020, sufficient data were available for 2759 to link to the ADI. The supplementary table compares those with ADI information to those lacking it. Though no major differences were apparent between these groups, the group lacking ADI information was less often of white race (78% vs 83%), less often had two or more comorbidities (37% vs. 42%), and less often took three or more concomitant medications (40% vs. 50%). Among those with available ADI information, each of the ADI deciles was represented, though the distribution of enrollees was skewed toward lower deprivation. Forty-two percent of enrollees resided in the lowest ADI quintile (i.e., least neighborhood disadvantage), compared to only 7% and 6% in the highest and second-highest ADI quintiles (Figure 1A). In contrast, the distribution of ADI strata among all Orange County residents was skewed toward more disadvantaged neighborhoods, in particular for the tenth ADI decile. Figure 1B illustrates the geographic spread of C2C enrollees, coded by their ADI decile.

Figure 1. (A) Histogram plots of the relative proportions of ADI categories for C2C Registry enrollees (in orange, right-hand y-axis) and for the overall population in Orange County (blue, left hand y-axis). (B) Geocoded map of enrollees in the C2C Registry based on their state ADI index. Illustrated dots represent individual enrollees with added noise, using the R Jitter function, to protect participant confidentiality. ADI, Area Deprivation Index; C2C, Consent-to-Contact


Individuals from the highest ADI quintiles were observed to be more often female and to more often self-report being from a non-white race or Hispanic ethnicity (Table 1). Participants from the lowest ADI quintile had the highest average level of education. The lowest ADI quintile had the lowest proportion of participants self-reporting three or more comorbid medical conditions. Recruitment sources were similar across the ADI groups, although email produced less than half of the registrants in the highest ADI quintile, compared to 51-61% of the lower quintiles. CFI scores were observed to be lowest in the lowest ADI quintile and highest in the highest ADI quintile. RAQ scores were similar across the ADI groups.
The proportions willing to be contacted about studies among the ADI quintiles were highly consistent for each research procedure (Table 2). Across ADI categories, the proportions willing to participate were highest for research requiring cognitive testing and lowest for research requiring lumbar puncture.

Table 1. Demographic and clinical characteristics of C2C enrollees across ADI categories

*includes American Indian or Alaska Native, Native Hawaiian or Pacific Islander, those who refused, and missing.


Table 2. Willingness to participate across ADI categories in the C2C Registry



ADRD research faces critical challenges in recruiting samples that ensure generalizable results. Participants are consistently young, well educated, and from high socioeconomic status, compared to the general population (25). In this study, we examined socioeconomic diversity using the ADI in our local on-line recruitment registry, an example of an increasingly utilized tool to accelerate clinical research accrual. We found that participants in our registry were representative of all strata of the ADI but, as we hypothesized, were disproportionately from the lowest ADI strata (least disadvantaged neighborhoods). As has been noted in previous studies (12), participants from high ADI strata were more frequently of non-white races and Hispanic ethnicity. Education levels were notably high among all ADI strata and we observed no differences in the overall willingness to participate in research. This finding may suggest that registries, in particular those registries that are effective in recruiting high ADI participants, may offer a valuable opportunity to diversify ADRD studies.
There are numerous important implications to these findings. Ensuring the diversity of clinical research studies, especially clinical trials of new therapies, is a critical area of need. Relatively few research participants are non-white race or Hispanic ethnicity (26, 27), despite African Americans and Hispanics being at greatest risk for dementia (28). Biased homogeneous samples limit generalizability and risk misunderstanding of effect modification of treatment safety or efficacy (29). Barriers to registry recruitment may be lower than barriers to participation in clinical studies (30), since the risks and requirements are generally modest. This may create an opportunity to enroll diverse populations in registries for the purpose of engaging them and increasing participation in clinical research studies (15). The current results may suggest that geotargeted recruitment efforts will be essential to increasing the diversity of registry participants and that successfully doing so may permit careful selection for recruitment to prospective studies to ensure representation of different socioeconomic groups, since ADI strata were equally willing to be contacted about types of research studies. Further research is needed, however, to understand whether barriers to recruitment to registries may differ among ADI strata. While we observed associations between ADI and race and ethnicity, other social determinants of health, such as direct measures of socioeconomic status, acculturation, and racism all may be critical to understand and address (25, 31).
The inclusion of more disadvantaged neighborhoods is an important consideration to ADRD recruitment efforts. The digital divide is narrowing, with most US adults having smartphones (32). This may create opportunities to use social media and electronic campaigns to better reach people from disadvantaged communities (33). To date, we have engaged in minimal effort to recruit to the C2C Registry through digital tools, but other work points to the potential utility of social media and other on-line recruitment strategies (34-37). Alternatively, more traditional recruitment approaches such as direct mail (38) and grassroots education (39, 40) present clear opportunities to target specific neighborhoods. Though our previous direct mail campaigns produced lower yield than expected, this method has been successful in other registries (13) and we did not previously test for potential effect modification by ADI (38). We also note new opportunities to enhance the use of direct mail, such as “quick response (QR)” codes that enable recipients to open a link or install an application using the camera on their cell phone or tablet device as a barcode-reader.
Although our registry does not perform objective cognitive testing, as do some others (41, 42), it has other important strengths. Participants in our registry provide self-reported data on cognitive performance using the CFI, which has been shown to differ among preclinical AD participants and biomarker negative controls (21, 43). Intriguingly, CFI scores were elevated among the high ADI group in the C2C Registry. This observation is similar to previous cross sectional (12) as well as longitudinal studies of cognitive performance (11). Although we have no data to consider potential mediators of these subjective complaints, it is conceivable that complaints could be driven by differences among the groups in brain volume (10, 44) or even AD neuropathology (9), reaffirming the potential importance of recruiting these groups to prospective research studies, such as preclinical AD trials.
Lack of differences among ADI strata were also important, including the lack of differences in willingness to be contacted about studies and for the RAQ. Work at other academic researcher centers engaged in community outreach has found that RAQ scores were lower among diverse communities, compared to more traditional research populations (45). Previous analyses of the diverse racial and ethnic groups that make up C2C Registry observed similar differences (17), but we found no such differences based on neighborhood disadvantage here. Future work should aim to elucidate relationships among other social determinants of health and research attitudes.
We note some important limitations of the current study. We did not have sufficient data to assess ADI on every registrant due to data missingness and some participants including only a PO Box address at enrollment. We are unable to assess whether missingness due to this factor is at random or more disproportionately affects specific ADI strata. If missingness were more prominent among high ADI strata, it might suggest that these data overestimate the underrepresentation of high ADI participants, but create more uncertainty about the examination of these participants in particular (e.g., their willingness to participate). We also acknowledge that self-reported willingness to be contacted about studies is not equivalent to the behavior of participating in a study. From our registry, we have referred participants to a large variety of studies and consistently achieve >30% enrollment of referred individuals. We cannot rule out that, despite similarities in indicated willingness, differences among ADI strata in actual study enrollment could still exist. Similarities in RAQ scores across ADI strata may argue against this possibility, however, and future research will examine this question.
In conclusion, people from more disadvantaged neighborhoods may be underrepresented in recruitment registries, increasing the risk that they will be similarly underrepresented when using these tools to facilitate prospective recruitment to clinical research. Once enrolled in a registry, these data suggest that participants from more disadvantaged neighborhoods may be equally willing to participate in research. Efforts to increase representation of participants from disadvantaged neighborhoods in registries could therefore be an important intervention to increase generalizability in clinical research studies.


Acknowledgements: The authors would like to acknowledge all participants in the C2C Registry. This registry was made possible by a donation from HCP, Inc. and is supported by NIA AG066519 and NCATS TR001414. Dr. Kind’s time is supported by R01AG070883, RF1AG057784 and P30AG062715.

Funding: NIA, Grant/Award Number: AG066519, AG070883, AG057784 and AG062715; NCATS, Grant/Award Number: UL1 TR001414

Conflicts of interest: Dr. Grill reports research support from Biogen, Eli Lilly, Genentech, and the NIH. He reports personal fees from SiteRx, outside the submitted work. Dr. Kind reports grants from NIH during the conduct of the study; grants from NIH, grants from VA, outside the submitted work; Mr. Hoang has nothing to disclose. Dr. Gillen reports service on Data Safety Monitoring Boards for Pfizer, Biomarin, Novo Nordisk, Novartis, Amgen, Celgene, CRISPR, AstraZeneca, Merck Serano, Array, Seattle Genetics, Genentech/Roche, UCB, Acerta, Juno Therapeutics, Medivation outside the submitted work. He has provided consulting services for Eli Lilly, ChemoCyntrix, FibroGen, GlaxoSmithKline, ProventionBio, Biom’Up outside the submitted work.





1. Watson JL, Ryan L, Silverberg N, Cahan V, Bernard MA. Obstacles and opportunities in Alzheimer’s clinical trial recruitment. Health affairs (Project Hope). 2014;33(4):574-9.
2. Fargo KN, Carrillo MC, Weiner MW, Potter WZ, Khachaturian Z. The crisis in recruitment for clinical trials in Alzheimer’s and dementia: An action plan for solutions. Alzheimers Dement. 2016;12(11):1113-5.
3. Kind AJH, Buckingham WR. Making Neighborhood-Disadvantage Metrics Accessible – The Neighborhood Atlas. The New England journal of medicine. 2018;378(26):2456-8.
4. Singh GK. Area deprivation and widening inequalities in US mortality, 1969-1998. American journal of public health. 2003;93(7):1137-43.
5. Arias F, Chen F, Fong TG, Shiff H, Alegria M, Marcantonio ER, et al. Neighborhood-Level Social Disadvantage and Risk of Delirium Following Major Surgery. Journal of the American Geriatrics Society. 2020 Dec;68(12):2863-2871. doi: 10.1111/jgs.16782. Epub 2020 Aug 31.
6. Durfey SNM, Kind AJH, Buckingham WR, DuGoff EH, Trivedi AN. Neighborhood disadvantage and chronic disease management. Health services research. 2019;54 Suppl 1:206-16.
7. Kind AJ, Jencks S, Brock J, Yu M, Bartels C, Ehlenbach W, et al. Neighborhood socioeconomic disadvantage and 30-day rehospitalization: a retrospective cohort study. Annals of internal medicine. 2014;161(11):765-74.
8. Sheets L, Petroski GF, Jaddoo J, Barnett Y, Barnett C, Kelley LEH, et al. The Effect of Neighborhood Disadvantage on Diabetes Prevalence. AMIA Annual Symposium proceedings / AMIA Symposium AMIA Symposium. 2017;2017:1547-53.
9. Powell WR, Buckingham WR, Larson JL, Vilen L, Yu M, Salamat MS, et al. Association of Neighborhood-Level Disadvantage With Alzheimer Disease Neuropathology. JAMA Netw Open. 2020;3(6):e207559.
10. Hunt JFV, Buckingham W, Kim AJ, Oh J, Vogt NM, Jonaitis EM, et al. Association of Neighborhood-Level Disadvantage With Cerebral and Hippocampal Volume. JAMA neurology. 2020;77(4):451-60.
11. Hunt JFV, Vogt NM, Jonaitis EM, Buckingham WR, Koscik RL, Zuelsdorff M, et al. Association of Neighborhood Context, Cognitive Decline, and Cortical Change in an Unimpaired Cohort. Neurology. 2021 May 18;96(20):e2500-e2512. doi: 10.1212/WNL.0000000000011918.Epub 2021 Apr 14.
12. Zuelsdorff M, Larson JL, Hunt JFV, Kim AJ, Koscik RL, Buckingham WR, et al. The Area Deprivation Index: A novel tool for harmonizable risk assessment in Alzheimer’s disease research. Alzheimers Dement (N Y). 2020;6(1):e12039.
13. Aisen P, Touchon J, Andrieu S, Boada M, Doody RS, Nosheny RL, et al. Registries and Cohorts to Accelerate Early Phase Alzheimer’s Trials. A Report from the E.U./U.S. Clinical Trials in Alzheimer’s Disease Task Force. The journal of prevention of Alzheimer’s disease. 2016;3(2):7.
14. Grill JD. Recruiting to preclinical Alzheimer’s disease clinical trials through registries. Alzheimers Dement (N Y). 2017;3(2):205-12.
15. Cocroft S, Welsh-Bohmer KA, Plassman BL, Chanti-Ketterl M, Edmonds H, Gwyther L, et al. Racially diverse participant registries to facilitate the recruitment of African Americans into presymptomatic Alzheimer’s disease studies. Alzheimers Dement. 2020 Aug;16(8):1107-1114. doi: 10.1002/alz.12048. Epub 2020 Jun 16.
16. Ashford MT, Eichenbaum J, Williams T, Camacho MR, Fockler J, Ulbricht A, et al. Effects of sex, race, ethnicity, and education on online aging research participation. Alzheimers Dement (N Y). 2020;6(1):e12028.
17. Salazar CR, Hoang D, Gillen DL, Grill JD. Racial and ethnic differences in older adults’ willingness to be contacted about Alzheimer’s disease research participation. Alzheimers Dement (N Y). 2020;6(1):e12023.
18. Grill JD, Hoang D, Gillen DL, Cox CG, Gombosev A, Klein K, et al. Constructing a Local Potential Participant Registry to Improve Alzheimer’s Disease Clinical Research Recruitment. J Alzheimers Dis. 2018;63(3):1055-63.
19. Gombosev A, Salazar CR, Hoang D, Cox CG, Gillen DL, Grill JD. Direct Mail Recruitment to a Potential Participant Registry. Alzheimer Dis Assoc Disord. 2021 Jan-Mar 01;35(1):80-83.doi: 10.1097/WAD.0000000000000368.
20. Rubright JD, Cary MS, Karlawish JH, Kim SY. Measuring how people view biomedical research: Reliability and validity analysis of the Research Attitudes Questionnaire. J Empir Res Hum Res Ethics. 2011;6(1):63-8.
21. Amariglio RE, Donohue MC, Marshall GA, Rentz DM, Salmon DP, Ferris SH, et al. Tracking Early Decline in Cognitive Function in Older Individuals at Risk for Alzheimer Disease Dementia: The Alzheimer’s Disease Cooperative Study Cognitive Function Instrument.JAMA Neurol. 2015 Apr;72(4):446-54. doi: 10.1001/jamaneurol.2014.3375.
22. Walsh SP, Raman R, Jones KB, Aisen PS. ADCS Prevention Instrument Project: the Mail-In Cognitive Function Screening Instrument (MCFSI). Alzheimer Dis Assoc Disord. 2006;20(4 Suppl 3):S170-8.
23. Joynt Maddox KE, Reidhead M, Hu J, Kind AJH, Zaslavsky AM, Nagasako EM, et al. Adjusting for social risk factors impacts performance and penalties in the hospital readmissions reduction program. Health services research. 2019;54(2):327-36.
24. Schmidt H, Gostin LO, Williams MA. Is It Lawful and Ethical to Prioritize Racial Minorities for COVID-19 Vaccines? JAMA. 2020 Nov 24;324(20):2023-2024. doi: 10.1001/jama.2020.20571.
25. Brewster P, Barnes L, Haan M, Johnson JK, Manly JJ, Napoles AM, et al. Progress and future challenges in aging and diversity research in the United States. Alzheimers Dement. 2019;15(7):9.
26. Shin J, Doraiswamy PM. Underrepresentation of African-Americans in Alzheimer’s Trials: A Call for Affirmative Action. Front Aging Neurosci. 2016;8:123.
27. 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. International psychogeriatrics / IPA. 2007;19(3):539-58.
28. Mehta KM, Yeo GW. Systematic review of dementia prevalence and incidence in United States race/ethnic populations. Alzheimers Dement. 2017;13(1):72-83.
29. Oh SS, Galanter J, Thakur N, Pino-Yanes M, Barcelo NE, White MJ, et al. Diversity in Clinical and Biomedical Research: A Promise Yet to Be Fulfilled. PLoS Med. 2015;12(12):e1001918.
30. Rogers JL, Johnson TR, Brown MB, Lantz PM, Greene A, Smith YR. Recruitment of women research participants: the Women’s Health Registry at the University of Michigan. Journal of women’s health (2002). 2007;16(5):721-8.
31. Williams DR. Race and health: basic questions, emerging directions. Annals of epidemiology. 1997;7(5):322-33.
32. Perrin A, Anderson M. Share of U.S. adults using social media, including Facebook, is mostly unchanged since 2018
33. Pew Research Center. 2019 Pew Research Center;
34. Whitaker C, Stevelink S, Fear N. The Use of Facebook in Recruiting Participants for Health Research Purposes: A Systematic Review. J Med Internet Res. 2017;19(8):e290.
35. Hough D. Social media and participant recruitment: What we’ve learned so far
36. Dobkin RD, Amondikar N, Kopil C, Caspell-Garcia C, Brown E, Chahine LM, et al. Innovative Recruitment Strategies to Increase Diversity of Participation in Parkinson’s Disease Research: The Fox Insight Cohort Experience. J Parkinsons Dis. 2020;10(2):665-75.
37. Shaver LG, Khawer A, Yi Y, Aubrey-Bassler K, Etchegary H, Roebothan B, et al. Using Facebook Advertising to Recruit Representative Samples: Feasibility Assessment of a Cross-Sectional Survey. J Med Internet Res. 2019;21(8):e14021.
38. Waltman NL, Smith KM, Kupzyk KA, Lappe JM, Mack LR, Bilek LD. Approaches to Recruitment of Postmenopausal Women for a Community-Based Study. Nurs Res. 2019;68(4):307-16.
39. Ballard EL, Nash F, Raiford K, Harrell LE. Recruitment of black elderly for clinical research studies of dementia: the CERAD experience. The Gerontologist. 1993;33(4):561-5.
40. Carr SA, Davis R, Spencer D, Smart M, Hudson J, Freeman S, et al. Comparison of recruitment efforts targeted at primary care physicians versus the community at large for participation in Alzheimer disease clinical trials. Alzheimer Dis Assoc Disord. 2010;24(2):165-70.
41. Walter S, Clanton TB, Langford OG, Rafii MS, Shaffer EJ, Grill JD, et al. Recruitment into the Alzheimer Prevention Trials (APT) Webstudy for a Trial-Ready Cohort for Preclinical and Prodromal Alzheimer’s Disease (TRC-PAD). The journal of prevention of Alzheimer’s disease. 2020;7(4):219-25.
42. Weiner MW, Nosheny R, Camacho M, Truran-Sacrey D, Mackin RS, Flenniken D, et al. The Brain Health Registry: An internet-based platform for recruitment, assessment, and longitudinal monitoring of participants for neuroscience studies. Alzheimers Dement. 2018;14(8):1063-76.
43. Sperling RA, Donohue MC, Raman R, Sun CK, Yaari R, Holdridge K, et al. Association of Factors With Elevated Amyloid Burden in Clinically Normal Older Individuals. JAMA neurology. 2020;77(6):11.
44. Meeker KL, Wisch JK, Hudson D, Coble D, Xiong C, Babulal GM, et al. Socioeconomic Status Mediates Racial Differences Seen Using the AT(N) Framework. Ann Neurol. 2021;89(2):254-65.
45. Neugroschl J, Sewell M, De La Fuente A, Umpierre M, Luo X, Sano M. Attitudes and Perceptions of Research in Aging and Dementia in an Urban Minority Population. J Alzheimers Dis. 2016;53(1):69-72.



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:; Phone: 913-588-5312; Fax: 913-945-5035

J Prev Alz Dis 2021;
Published online August 26, 2021,



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.



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.



We searched Ovid MEDLINE, Embase, CINAHL Complete, and clinical 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).



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



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.



1. Association As. Facts and Figures 2021 [Available from:
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:
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:,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:
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:
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.



I. Yunusa1, N. Rashid2, V. Abler2, K. Rajagopalan3


1. Center for Outcomes Research & Evaluation, University of South Carolina College of Pharmacy, Columbia, South Carolina, USA; 2. Acadia Pharmaceuticals, Inc., San Diego, California, USA; 3. Anlitiks Inc., Dover, Massachusetts, USA

Corresponding Author: Krithika Rajagopalan, PhD, Anlitiks Inc., 18 Old Colony Dr, Dover, Massachusetts, 02030, USA, Phone: 508-314-8158, Email:

J Prev Alz Dis 2021;4(8):520-533
Published online August 6, 2021,



Objectives: To evaluate the comparative efficacy, safety, tolerability, and effectiveness of atypical antipsychotics (AAPs) for the treatment of dementia related psychosis (DRP) in older adults.
Methods: In this systematic literature review (SLR), we qualitatively synthesized evidence on the comparative efficacy (based on neuropsychiatric inventory), tolerability (weight gain), and safety (cerebrovascular adverse events [CVAE], cardiovascular events, mortality, somnolence, extrapyramidal symptoms [EPS]) of AAPs used to treat DRP. We also assessed effectiveness based on all-cause discontinuations and discontinuations due to lack of efficacy or adverse events (AE). Published articles from through March 2021 from PubMed, EMBASE, PsycINFO, and Cochrane databases evaluated. We included double-blind, active-comparator/placebo-controlled randomized trials, open-label trials, and observational studies.
Results: This qualitative synthesis included 51 eligible studies with sample size of 13,334 and mean age of 79.36 years. Risperidone, olanzapine, quetiapine, and aripiprazole demonstrated numerically small improvement in psychotic symptoms among patients with DRP. Somnolence was the most reported AE for all the AAPs, with weight gain and tardive dyskinesia more common with olanzapine and risperidone, respectively. These AAPs are associated with falls, EPS, cognitive declines, CVAE, and mortality. Aripiprazole and olanzapine had lower odds of discontinuation due to lack of efficacy, with olanzapine having greater discontinuation odds due to AEs.
Conclusion: This SLR demonstrated that AAPs used off-label to treat DRP are associated with small numerical symptom improvement but with a high risk of AEs, including cognitive decline and potentially higher mortality. These results underscore the need for new treatments with a favorable benefit-risk profile for treating DRP.

Key words: Dementia-related psychosis, antipsychotics, atypical antipsychotics, hallucinations, delusions, safety, tolerability, efficacy, effectiveness.

Abbreviations: NPS:Neuropsychiatric symptoms; DRP: dementia-related psychosis; AP: antipsychotics; SLRs: systematic literature reviews; FDA: Food and Drug Administration; AAPs: atypical antipsychotics; EPS: extrapyramidal symptoms; APA: American Psychiatric Association; NPI: neuropsychiatric inventory; NPI-NH: neuropsychiatry index-nursing home; BPRS: Brief Psychiatric Rating Scale; CVAE: cerebrovascular events; AD: Alzheimer’s disease; VaD: vascular dementia; DLB: dementia with Lewy bodies; MD: mixed dementia; PD: Parkinson’s disease; NH: nursing home; LTC: long-term care; CGI: clinical global improvement; CATIE-AD: clinical antipsychotic trials of intervention effectiveness- Alzheimer’s disease; CGI-S: clinical global impression scale- severity; CGI-C: clinical global impression scale- change; PANSS: Positive and Negative Syndrome Scale; SAS: Simpson-Angus Scale; TEAEs: treatment-emergent adverse events; MMSE: Mini-Mental State Examination; BEHAVE-AD: Behavioral Symptoms in Alzheimer’s Disease; SUCRA: surface under cumulative ranking curve




Worldwide, an estimated 50 million people currently live with dementia, which results in progressive loss of cognitive function severe enough to cause a decline in the patient’s ability to perform activities of daily living (1). It is estimated that 7.5M individuals have dementia of different types in the United States, and this number is expected to double by 2030, given the increase in the elderly population and rising life expectancy (2). While cognitive decline is the hallmark of dementia, neuropsychiatric symptoms (NPS) are common and can dominate its clinical presentation (3). NPS in patients with dementia includes dementia-related psychosis (DRP) that manifests as delusions (false fixed beliefs) or hallucinations (seeing or hearing things that others do not see or hear), and other behavioral symptoms such as agitation, aggression, depression, apathy, elation, anxiety, disinhibition, irritability, and aberrant motor behavior (4). Notably, about 2.4M people in the U.S. are estimated to have DRP (5). Data on patients with Alzheimer’s disease (AD) dementia suggest that 94% of patients experience delusions while 70% experience hallucinations (6, 7). DRP prevalence is also anticipated to grow significantly with the increasing rates of dementia, exerting significant distress to individuals and their families and potentially imposing an enormous economic burden to the society (8). Thus, interventions aimed at treating DRP could tremendously improve the health outcomes of patients, their families, and caregivers (3).
At the time of this analysis, there was no Food and Drug Administration (FDA) approved treatment for DRP. However, antipsychotics (APs), both typical antipsychotics (first-generation) and atypical antipsychotics [AAPs] (second-generation) are used off-label to treat hallucinations and delusions associated with DRP. Available evidence on the comparative efficacy of currently used AAPs versus placebo suggests that AAPs only offer a numerically small improvement in psychotic symptoms among patients with DRP (3). On the other hand, they are known to be associated with significant safety risks related to treatment-emergent cerebrovascular adverse events (CVAE) such as stroke, and a higher risk of mortality (9, 10). In recognition of the unfavorable benefit-risk profile of current off-label AAPs, they all carry an FDA boxed warning on the increased risk of mortality among the elderly with dementia (9). Furthermore, years of research on these medications suggest that each of them report a unique side effect profile that ranges from extrapyramidal symptoms (EPS) (more associated with typical APs) to weight gain, hyperlipidemia, impaired glucose metabolism including potential insulin resistance, diabetes, tardive dyskinesia, sedation, hyperprolactinemia, orthostatic hypotension, sexual dysfunction, increased rate of fractures and cognitive deterioration, among others (11-13). Considering this, the American Psychiatric Association (APA) guidelines recommend the short-term use of pharmacological interventions only after nonpharmacological interventions such as cognitive, behavioral, and environmental therapies have been attempted first to treat NPS such as DRP (14). Additionally, they also provide instructions for gradual dose reduction or taper.
This study focused on assessing the outcomes of the off-label use of AAPs in treating DRP. Previously published SLRs and network meta-analyses (NMA) assessed AAPs’ relative benefits and safety on a broader NPS population from randomized controlled trials (RCTs) which assessed fewer outcomes (15, 16). As they evaluated the overall NPS of dementia, previous studies assessed total neuropsychiatric inventory (NPI) or neuropsychiatry inventory-nursing home (NPI-NH) score as opposed to their psychosis subscale which would be more informative in assessing DRP. The SLRs demonstrated that the trade-offs between the benefits of treating NPS did not adequately offset the risks associated with AAPs. While Yunusa et al. 2019 (15) found significant benefits for total NPI score, and Brief Psychiatric Rating Scale (BPRS) with some AAPs compared to placebo; there were a greater risk of adverse outcomes of CVAE. A recently published SLR and NMA by Watt et al. (16) suggest that, in subgroups of persons with dementia, AAPs are associated with greater harm (i.e., falls, fractures, and CVAE) than antidepressants and anticonvulsants (medications used in place of AAPs for treating NPS of dementia). Other agent (anticonvulsants, dextromethorphan-quinidine) combinations were also associated with an adverse safety profile compared to placebo. However, no single comprehensive SLR has been conducted to comparatively assess efficacy, tolerability, safety, and, most of all, AAPs’ effectiveness in treating symptoms of DRP. Furthermore, it is crucial to consider that additional clinical trials have been published since the most recent SLR (17). This SLR aimed to comprehensively compare the efficacy, effectiveness, safety, and tolerability of different AAPs used to treat DRP by including open label and non-double blind studies along with RCTs to assess a broader gamut of outcomes to reflect real-world data and address critical knowledge gaps and provide the most recent review to date.



This systematic literature review followed the preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2020 guidelines (18). The study selection process is illustrated in Figure 1. A protocol for this SLR was developed internally by the study team before starting the review, and it guided the conduct of the study.

Figure 1. PRISMA Study Selection Flowchart


Eligibility Criteria

The Patients/Population, Intervention, Comparator, Outcome, and Study design (PICOS) framework, was used as an eligibility criterion to search, select, and review relevant studies. Included participants from the studies (age ≥ 40, those living in the community or nursing home [NH]) had to have dementia of the following type: AD dementia, frontotemporal dementia, vascular dementia (VaD), dementia with Lewy bodies (DLB) and Parkinson’s disease (PD) dementia. The interventions included were both typical and atypical antipsychotics with or without multiple comparator groups and outcomes related to efficacy, tolerability, safety, and effectiveness. Finally, we excluded studies other than double-blind, active-comparator or placebo-controlled RCTs, open-label trials, or observational studies. Efficacy was assessed as improvement in DRP symptoms related to psychosis (i.e., hallucinations and delusions) and the NPI-psychosis subscale, NPI-NH psychosis subscale, BPRS psychosis factor subscale, and BEHAVE-AD (Behavioral Symptoms in Alzheimer’s Disease) psychosis subscale were used to measure improvement in psychotic symptoms in patients. The psychosis subscale is a combination of the delusion and hallucination subscales. Additionally, measures of adverse effects were assessed as tolerability (i.e., weight gain due to AAPs) and safety outcomes (i.e., somnolence, EPS including tardive dyskinesia, cognition, CVAE, falls, and mortality, among others). Effectiveness was summarized from all-cause discontinuations and discontinuations due to lack of efficacy or safety.

Data Sources and Literature Search Strategy

The literature search was conducted in MEDLINE/PubMed (Appendix. 1), PsycINFO, EMBASE, and Cochrane Central Register of Controlled Trials from January 2000 to March 2021. The search was limited to articles published in English. Databases were searched using predefined search terms to identify published studies evaluating the safety and effectiveness of antipsychotics used to treat DRP of any type (PD dementia, VaD, DLB, AD dementia, and frontotemporal dementia). Search strategies were developed using medical subject headings (MeSH) terms (PubMed and Cochrane library), Emtree terms (EMBASE and PsycINFO), and text words related to antipsychotic treatment in DRP.
Key search terms to define the patient population included dementia, NPS, DRP of any type including psychosis related to PD dementia, VaD, DLB, AD dementia and frontotemporal dementia, hallucinations, delusions, agitation, and aggression. Search terms for interventions (medications) included typical antipsychotics, haloperidol, acetophenazine, carbamazepine, chlorpromazine, chlorprothixene, fluphenazine, loxapine, mesoridazine, molindone, perphenazine, prochlorperazine, promazine, thioridazine, thiothixene, trifluoperazine, atypical antipsychotics, aripiprazole, clozapine, ziprasidone, risperidone, quetiapine, olanzapine, pimavanserin, asenapine, brexpiprazole, cariprazine, paliperidone, and lurasidone.

Search, Study selection, Data Extraction

Two researchers independently searched the indexed electronic databases through the Anlitiks SLR platform, a proprietary internal search engine within Anlitiks that allows integrated searches of selected index databases such as Medline/PubMed and allows the searches for the index databases such as Cochrane review separately, as applicable. Articles were also independently screened against predefined eligibility criteria in two phases, title/abstract screening (Phase 1) and full-text screening (Phase 2). References of all eligible articles were searched to identify the possibility of any missing articles. Subsequently, we extracted data from eligible articles that passed Phase 2 screening using an apriori standardized data extraction form in Microsoft Excel. A third reviewer resolved any disagreement in the information extracted from the articles by both researchers. Data were extracted from both the secondary analysis and original trials. In case of missing information, authors were contacted where necessary. Outcomes related to efficacy, safety, tolerability, and effectiveness were reviewed. Effectiveness outcome was assessed as AP withdrawal or discontinuations, time to AP discontinuations, AP switches or augmentation, as well as relapses of NPS. Also, the availability of data on other effectiveness measures such as hospitalizations, ER visits, and other health resource use was evaluated. Lastly, outcomes like measures of cardiometabolic disturbances (e.g., hyperglycemia, dyslipidemia), sedation, somnolence, and cardiovascular events (e.g., heart attacks), CVAE (e.g., stroke), EPS including tardive dyskinesia, falls and fractures, urinary incontinence, urinary tract infection, as well as measures of cognitive decline, were assessed. In cases where outcomes were reported as a composite measure (e.g., CVAE), we also tried to extract data for the individual outcomes (i.e., stroke, transient ischemic attack, etc.) constituting the composite measure, if available in the published articles. Outcome data were extracted regardless of the reporting format, i.e., either as a dichotomous, categorical, or continuous variable.

Risk of bias and Study Quality Assessment

The Cochrane Risk of Bias tool (19) was used to assess the risk of bias in 49 original trials using seven domains (namely sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting and ‘other sources of bias’s). Accordingly, the risk of bias was categorized on a three-level categorical scale viz. – ‘Low risk’ of bias; ‘High risk’ of bias, or ‘Unclear risk’ of bias. The summary of the risk of bias assessment is shown in Figure 2, and the study-level risk of bias is shown in a tabular form in Appendix 2. The observational studies were assessed using the Newcastle-Ottawa scale for the assessment of the quality of observational studies (20) as shown in Appendix 3.

Figure 2. Risk of Assessment Bias

The figure depicts pooled study quality assessment of all the RCT included in the study.



Study Selection

The initial search yielded a total of 1,388 citations, and after removing the duplicates, 1,315 titles and abstracts were screened for eligibility against pre-defined inclusion/exclusion criteria. A total of 199 articles were eligible for full-text review. Overall, 51 studies 17, 21-77) published between January 2000-March 2021 were included in the qualitative synthesis that met the inclusion/exclusion criteria.

Study Characteristics

Figure 1 illustrates the search yield and study attrition for the selection of eligible studies. Both checklist (Appendix 4,5) (18) and flow diagram from the PRISMA ensured transparency in selecting articles, and the quality standards were followed. Of the total included 51 original studies, 39 were randomized trials, 10 were open-label trials, and 2 were observational studies. Information from 8 post-hoc analyses of the randomized trials were also extracted. Among the identified studies, 21 were conducted in institutionalized settings such as the NH or long-term care (LTC) facilities, while the rest were conducted among community-dwelling, or outpatient settings. Based on assessment of study quality, it was found that 10% of the studies included in the SLR were of low quality.
From the included studies ranging from 4 to 52 weeks, there were a total of 13,334 patients (sample size ranged from 10 to 4499) with mean age of 79.36 years. Most patients in the included studies were diagnosed with AD dementia, with some studies on VaD, DLB, PD-related dementia, or mixed dementia (MD). There were two studies that had PD dementia-related psychosis as their primary inclusion criteria. Of the 42 parallel-group studies, 21 were placebo-controlled studies, and 21 had active-controls. Overall, there were 30 trials of risperidone, 14 of quetiapine, 10 of olanzapine, 3 of aripiprazole. There was one trial each for ziprasidone, tiapride, and brexpiprazole. The summary of all included study characteristics is described in Table 1, and the Table 2 depicts the outcome measures assessed for individual AAPs.
While the SLR was intended to include outcome measures such as quality of life (QoL), functioning and caregiver burden outcomes, and hospital admissions and emergency department visits, interestingly, little has been reported in the literature about the role of APs and their impact on these outcomes. Of note, the review resulted in a reasonable number of available publications for the six commonly used AAPs (i.e., olanzapine, risperidone, quetiapine, aripiprazole, ziprasidone, and brexpiprazole) and only one AP (i.e., haloperidol). Therefore, this SLR primarily focused on the qualitative assessment of comparative efficacy, tolerability, safety, and effectiveness of AAPs in the treatment of DRP.

Table 1. Summary of individual study characteristics

Abbreviations and full names; RCT- Randomized control trials; NR- No reported findings relevant to the study; NH- Nursing home; DLB – dementia with Lewy bodies; BPSD – behavioral and psychological symptoms of dementia; VaD- Vascular dementia; AD- Alzheimer’s disease; PD- Parkinson’s Disease; BEHAVE-AD- Behavioral Symptoms in Alzheimer’s Disease; AE- Adverse Events; EPS- Extrapyramidal symptoms; UTI- Urinary tract infection; CVAE- Cerebrovascular adverse events; Olz- Olanzapine; N/A- Not applicable

Table 2. Table of evaluated outcome measures for AAPs

The table depicts the individual outcomes assessed for every atypical antipsychotic (AAP); Abbreviations: NPI (Neuropsychiatric inventory); NPI-NH (Neuropsychiatric inventory- nursing home subscale); BEHAVE-AD (Behavioral Symptoms in Alzheimer’s Disease); CVAE (Cerebrovascular adverse events); EPS (Extrapyramidal symptoms); UTI (Urinary tract infection); CGI-C (Clinical Global Impression – Corrections). The ‘X’ represent outcome measures that were studied for corresponding AAPs. The blank cells represent the outcome measures that were not studied for corresponding AAPs.



A total of 10 studies (4 placebo-controlled and 6 active comparator [e.g., risperidone, haloperidol, and promazine] studies) were included in the review (27, 29, 31, 33, 38, 54, 56, 57, 60, 64, 65, 69, 70, 75, 77). Placebo comparison studies suggest that lower doses of olanzapine have a greater effect on improving psychotic symptoms than higher doses (27, 29, 31, 57, 60). RCT of individuals with AD dementia who were retrospectively identified as meeting the DLB criteria (N =29) found that individuals treated with 5 mg olanzapine/day (N =10) showed greater reductions in NPI delusion subscale (−3.8 points) and hallucination (−5.9 points) subscale scores when compared with placebo (N =10) (29). As per De Deyn et al. 2004, in a double-blind study of 652 patients with delusions or hallucinations associated with AD dementia, olanzapine 7.5 mg/day significantly decreased psychotic symptoms (31). While according to Street et al., another randomized placebo comparison study (N =206) of olanzapine (5,10, or 15mg/day) among AD patients with psychosis and/or agitation/ aggression, it was reported that low-dose olanzapine (5 and 10 mg/d) showed significant improvement on the NPI-NH psychosis total , subscale compared to placebo; on the other hand, improvement with olanzapine (15 mg/day) was not significantly greater than placebo for this subscale (57). As per the study published by Verhey et al. 2006 (64), olanzapine was found to improve the NPI psychosis subscale score compared to placebo.
Reported AEs associated with olanzapine were high risk of CVAE (33, 60), somnolence (27, 57, 69), weight gain (31, 38), and EPS (56, 57). For example, Street et al. 200057 found that somnolence was significantly more common among all patients receiving olanzapine, but gait disturbance occurred in those receiving 5 or 15 mg/d (57) compared to placebo and 10mg/d of olanzapine. Additionally, the effects of olanzapine on cognitive symptoms were found to be inconsistent. Although Vigen et al. 2011 (65) found that olanzapine and other AAPs (risperidone) were associated with the worsening of cognitive symptoms, Clark et al. 2001 (27) reported that olanzapine was not associated with a decline in cognition compared to placebo. According to Deberdt et al. (33), olanzapine was associated with a significant increase in CVAE and mortality in comparison with placebo.
For effectiveness outcomes, De Deyn et al. 2004 (31) reported that olanzapine had significantly lower rate of discontinuation due to lack of efficacy as compared to placebo. According to the CATIE-AD trial (56, 60, 65), which compared olanzapine, quetiapine, and risperidone to placebo, olanzapine had the highest rate of discontinuation due to AEs.


A total of 30 studies (8 placebo-controlled, 18 active-comparator [i.e., quetiapine, haloperidol, olanzapine, and rivastigmine, escitalopram, galantamine, promazine, topiramate, amisulpride, citalopram, tiapride], and 4 single-arm) were included in the review (21, 23-26, 28, 33-36, 38-40, 42-45, 47, 50-52, 54, 56, 59, 60, 63, 65-67, 70, 71, 73-76). Placebo-comparison efficacy studies of risperidone found that in most cases, risperidone was associated with improved psychosis symptoms, compared to placebo (25, 51, 60). In the clinical antipsychotic trials of intervention effectiveness- Alzheimer’s disease (CATIE-AD) study, 421 outpatients, were included with AD and psychosis or agitated/aggressive behavior. Compared to placebo, greater improvements were seen with risperidone in the BPRS psychosis factor subscale (60). In a randomized placebo comparison study of risperidone (n=345), a significant reduction in BEHAVE-AD psychotic symptoms subscale (p=0.004) was seen with risperidone (25). In a secondary analysis of a 12-week, randomized controlled trial of individuals with AD, mean change at endpoint in BEHAVE-AD psychosis subscale was higher in risperidone group compared to placebo (-5.2 vs. -3.3; p=0.039) (24). Another secondary exploratory analysis of data on 479 nursing-home patients with psychosis of AD from three 12-week, double-blind, placebo-controlled clinical trials reported risperidone to be effective on the BEHAVE-AD delusion and hallucination subscales (51). In a multicenter, randomized, double-blind placebo-controlled trial of nursing home residents diagnosed with AD and psychosis (45), both risperidone group and placebo groups showed significant improvements on the BEHAVE-AD psychosis subscale (45). Overall, these studies reported that risperidone is moderately effective in treating various symptoms associated with psychosis. Interestingly, in a study by Deberdt et al. 2005 (33), olanzapine, risperidone, and placebo treatment reported improved NPI-NH psychosis subscale scores, though no significant changes emerged across treatments, including placebo-comparisons (33). Furthermore, significant reductions were found in the NPI hallucination and delusion subscales scores for amisulpride and risperidone according to Lim et al. 2006 (44).
The AE profile of risperidone demonstrated inconsistent results on the various safety outcomes depending on the number and type of measures. Jeste et al. (40) observed that the incidence of persistent tardive dyskinesia with risperidone appeared to be much lower than that seen in elderly patients treated with conventional neuroleptics (40). In a comparative study of risperidone vs. haloperidol, Suh et al. 2006 (59) reported that the risk of antipsychotic-induced Parkinsonism was significantly lower with risperidone (59). Yoon et al. (67) found that risperidone treatment was generally well tolerated, although EPS were noted in a dose-dependent manner (67). On the other hand, Teranishi et al. 2013 (63) found that drug-induced EPS increased significantly in the risperidone group (63) As far as cognitive decline, there were no differences between placebo and the APs, i.e., risperidone, thioridazine, haloperidol, chlorpromazine, and trifluoperazine in the DART-AD trial (21).
In a study of participants with DLB, risperidone experienced higher overall neurologic effects and worsening of neuropsychiatric symptoms (28). Other reported AEs for risperidone included falls,59,63 EPS (26, 38, 44, 56, 59) somnolence (25, 38, 44, 45, 59) and CVAE (42). In the long-term follow-up of the DART-AD trial, Kaplan-Meier estimates of mortality showed a significantly increased risk of mortality for patients among patients randomized to continue antipsychotic treatment on risperidone compared with those randomized to placebo.
In terms of effectiveness measures, results with risperidone were mixed for outcomes such as discontinuations, time to discontinuations, and or treatment augmentation and switches. In a study by Culo et al. 2010 (28), a significantly higher proportion of participants with DLB (68%) discontinued risperidone prematurely than AD (50%) patients on risperidone, and discontinuation rates were comparable in DLB participants with psychosis that were treated with citalopram (71%) or risperidone (65%). According to the CATIE-AD trials (56, 60, 65), it was reported that risperidone had higher rate of all cause discontinuation and discontinuation due to lack of efficacy compared to placebo. In terms of relapse prevention, Devanand et al. 2012 (34) found that risperidone was associated with a lower rate of psychotic relapse than placebo (60% vs. 33%). To our knowledge, apart from the Devanand study, there have been no other studies of psychotic relapse prevention in patients with DRP.


A total of 14 studies (6 placebo-controlled, 4 active-comparator [e.g., risperidone, rivastigmine, quetiapine, haloperidol], and 4 single-arm) were reviewed for quetiapine (22, 32, 37, 41, 47-49, 52, 55, 56, 60-62, 65, 68, 72). Efficacy studies for quetiapine in improving psychosis, have had mixed reports; while some studies reported favorable effects on symptom improvements, others showed quetiapine to be ineffective in improving psychotic symptoms. For example, Scharre et al. 2002 (55) found that patients on quetiapine showed a significant reduction in NPI-NH delusion subscale scores, after receiving doses of 50 to 150 mg (55). In another 10-week, double-blind, fixed-dose study, elderly institutionalized patients with dementia and agitation randomized to quetiapine 200mg/day, 100mg/day, or placebo, quetiapine 200mg was associated with clinically greater improvements in the NPI-NH psychosis subscale scores.68 Fujikawa et al. 2005 (37) found significant improvements with quetiapine in the BEHAVE-AD subscales of delusions.
However, other studies found that quetiapine did not improve psychosis compared with placebo (41, 62). For example, Tariot et al. 2006 showed that quetiapine, haloperidol, and placebo demonstrated similar levels of improvement in psychotic symptoms (i.e., no difference between placebo), as reported by mean NPI-NH2 (i.e. hallucination and delusion) subscale, in patients with possible AD from baseline to week 10 (62).
The most commonly reported adverse effects for quetiapine were somnolence (61, 62) death (61, 62) CVAE (52), and EPS (37, 49, 55, 56). While Zhong et al. 2007 (68) found that incidence of CVAE, postural hypotension, and falls were similar among quetiapine and placebo groups while mortality was numerically higher in the quetiapine group; however, these rates were not statistically significant (68).
In terms of effectiveness, Tariot et al. 2000 (61) reported that only 89 (48%) patients (n=184) on quetiapine completed treatment through 52 weeks. The main reasons for antipsychotic withdrawal or discontinuations included lack of efficacy (19%), AEs (15%), failure to return for follow-up (13%). Somnolence (31%), dizziness (17%), postural hypotension (15%) and EPS (13%).61 Onor et al. 2007 (48) found that clinically significant orthostatic hypotension (for patients on quetiapine) led to the discontinuation of 5 patients from their observational study (n =41) (48).


There were 3 studies of aripiprazole compared with placebo. Efficacy results for aripiprazole appear inconsistent across studies (Streim et al. 2008; De Deyn et al. 2005; Mintzer et al. 2007) (30, 46, 58). In a randomized, double-blind, placebo-controlled multicenter trial of 487 institutionalized AD patients with psychosis, Mintzer et al. 2007 (46) found that Aripiprazole 10 mg/day showed significantly greater improvements than placebo on the NPI-NH Psychosis Subscale for baseline scores compared to Week 10 scores (-6.87 versus -5.13; p =0.013) and NPI-NH Psychosis response rate (65 versus 50; p =0.019). However, in the study reported by De Deyn et al., (30) compared to placebo aripiprazole showed similar improvements in psychotic symptoms as assessed by NPI psychosis subscale scores but significantly greater improvements from baseline in BPRS psychosis subscale scores at the study endpoint (30). Additionally, Streim et al. 2008 (58) also found conflicting results to that reported by Mintzer (46), with no significant differences in mean change from baseline score on the NPI-NH Psychosis Subscale between aripiprazole and placebo.
As it relates to the safety and tolerability of aripiprazole, Streim et al, reported comparable rates for treatment-emergent adverse events (TEAEs) between aripiprazole and placebo, except for somnolence (aripiprazole, 14%; placebo, 4%) (58). While in the study reported by De Deyn et al. 2005 (30), only mild somnolence was observed. Moreover, the AEs were generally mild to moderate in severity and included (aripiprazole vs. placebo): urinary tract infection (8% vs. 12%), accidental injury (8% vs. 5%), somnolence (8% vs. 1%), and bronchitis (6% vs. 3%) (30). There were no significant differences from placebo in EPS, or clinically significant ECG abnormalities, vital signs, or weight (30). In a study by Mintzer et al. 2007 (46), CVAE was a reported outcome for the aripiprazole-treated population while no patients from the placebo group suffered from the same. Other AEs seen in the aripiprazole group were asthenia, agitation, and EPS (46).
Effectiveness outcomes reported in these studies did not show any clear patterns. In the De Deyn et al. study (30), the number of patients discontinuing due to AEs, lack of efficacy, or withdrawal of consent was similar in the aripiprazole and placebo groups (30). AEs leading to greater than 2% discontinuation in the aripiprazole or placebo group were asthenia (4%, aripiprazole 10mg/dl) and agitation (4%, placebo), respectively (46).


There was one reported publication of brexpiprazole that reported the results of two separate studies Grossberg et al. 2020 (17) assessed the efficacy, safety, and tolerability of brexpiprazole in patients with agitation in Alzheimer’s dementia (AAD) in two 12-week, randomized, double-blind, placebo-controlled, parallel-arm studies. While one study was a fixed-dose study (Study 1: 433 randomized), the second one was a flexible-dose study (Study 2: 270 randomized) of patients with AAD; in a care facility or community-based setting. Since the main focus of our study was psychosis and Grossberg et a., 2020 focused on patients with agitation, only safety outcomes reported in the study were considered for review.
TEAEs among patients receiving brexpiprazole were headache, insomnia, and somnolence. In general, most TEAEs were mild or moderate in severity. The studies found that brexpiprazole 2 mg/day has the potential to be efficacious, safe, and well-tolerated in the treatment of agitation in AD dementia (17). All cause discontinuations reported across both studies did not show numerical differences between brexpiprazole and placebo groups for both studies. However, discontinuation due to adverse events was reported to be higher for brexpiprazole as compared to placebo.


Rocha et al. 2006 (53) evaluated the efficacy and tolerability of ziprasidone in a 7-week open-label trial. For the patients included in the study, the mean NPI delusion subscale score fell significantly from 4.88 to 2.28 i.e., -53% from baseline to day 49 (p < 0.01). However, of the 25 patients who participated, 10 discontinued the study. The main reason for discontinuation was AEs. The most frequent AEs were somnolence, gastrointestinal symptoms, and parkinsonism (39, 53).



Published SLRs of AAP use among dementia patients largely focused on the gamut of NPS (i.e., psychosis – delusions and hallucinations, agitation/aggression, depression, anxiety, and irritability, among others) and selected safety parameters (such as CVAE, stroke, and mortality) from RCTs or observational studies. This qualitative synthesis adds to previously published SLRs in many ways. First, it is the most comprehensive, recent review of APs as a treatment of dementia related psychosis. Second, this SLR is intended to review real-world effects by including publications of RCTs, open-label trials, and observational studies, including single-arm or comparator-arm trials. Third, this is a comparative review of placebo-AP or AP-AP differences in efficacy, safety, tolerability, and effectiveness in treating DRP. The final and most important difference is that this SLR examined effectiveness outcomes such as differences in all-cause discontinuations and discontinuations due to lack of efficacy or discontinuation due to AEs, and time to relapses between the different AAPs.
This SLR showed that off-label therapies such as risperidone, olanzapine, quetiapine, and aripiprazole demonstrated numerically small psychotic symptom improvements among DRP patients; however, only risperidone was reported to have symptom improvements consistently while also showing a significant increase in EPS. This is supported by the fact that although it is not approved by the FDA, for short-term treatment of aggression in AD, risperidone is the only licensed drug in countries like UK, Canada, and Australia if aggression poses a risk or the person has not responded to non-drug approaches (78, 79). Both quetiapine and aripiprazole reported mixed results, and lower dose olanzapine showed greater symptom improvements than higher doses. These results are consistent with the previously published NMA of 17 studies (5373 patients) conducted by Yunusa et al. 2019 (15), comparing different AAPs (risperidone, olanzapine, aripiprazole, and quetiapine) that reported no statistically significant differences between the AAPs in terms of NPI symptoms scores. While no drug-drug differences were found across measures of efficacy and safety among aripiprazole, olanzapine, quetiapine, and risperidone, placebo-drug differences were found for some drugs for specific outcomes. The surface under the cumulative ranking curve estimated relative ranking of treatments from Yunusa’s study suggested that aripiprazole might be the most effective and safe AAP and that olanzapine provides the least effect overall; however, these results should be interpreted with caution where point estimates (OR and SMD) show that there is no statistically significant difference between placebo-drug and drug-drug comparisons from the synthesis of the 17 trials.
Interestingly, the current review, consistent with other studies, suggested that olanzapine may potentially exhibit a dose-response relationship in symptom improvements despite demonstrating only slight symptom improvements with pooled doses. Specifically, doses <10mg of olanzapine were significantly better than placebo in terms of neuropsychiatric symptom improvements (NPI), and doses >10mg were not different from placebo according to two studies (31, 57). While a dose range was found to be effective, no single effective dose was reported in these primary studies. It is not clear why higher doses of olanzapine had similar effects to placebo; it is plausible that the higher rates of AEs reported for olanzapine may have tempered with the effectiveness of olanzapine in psychotic symptom improvements. An NMA along with a meta-regression of these outcomes by dose level differences may be contemplated in the future to test the hypothesis.
While published SLRs and NMAs by Yunusa (15) and Watt (16) suggest that APs may be associated with a significant risk of strokes, falls, fractures, CVAEs, or death, and the umbrella review by Papola et al. (80) suggest an association between the use of APs and fractures, stroke, and cardiac death, the current review of AAP tolerability and safety outcomes included more outcomes such as weight gain, metabolic disturbances (e.g., hyperglycemia, dyslipidemia), sedation, somnolence, and cardiovascular events, CVAE, EPS including tardive dyskinesia, falls and fractures, urinary incontinence, urinary tract infection and cognitive decline. Not surprisingly, our review showed that somnolence was the most reported AE for all the major APs, with weight gain and tardive dyskinesia being more commonly reported for olanzapine and risperidone, respectively. Other AEs reported for all AAPs were EPS, falls, and CVAEs, except for brexpiprazole. Furthermore, studies also show that these APs may be associated with greater cognitive declines (38, 41, 60, 81) and potentially increased mortality (21, 33, 68) in patients with DRP. Although our study was qualitative in nature, Maust et al., reported that, about 27-50 patients need to be treated with AAPs in order for one person to die (9). In alignment with a previous study, this review also suggests that other AEs among patients receiving olanzapine include CVAE and EPS (81). While falls, EPS, and CVAE were also reported for risperidone. In addition to somnolence, commonly reported adverse effects of quetiapine were EPS, dizziness, postural hypotension, and death. AEs associated with aripiprazole are somnolence, urinary tract infection, accidental injury, somnolence, bronchitis, CVAE, akathisia, asthenia, agitation, and EPS. All these qualitative findings suggest the major AAPs that are currently used off-label for treating DRP may have an unfavorable benefit-risk based on multiple outcome measures, including CVAE, and mortality. These findings suggest that a quantitative review through an NMA may be needed.
The current study is an extension of previous reviews by examining additional measures of effectiveness defined as AP withdrawal or discontinuations, time to discontinuations, AP switches or augmentation, as well as psychotic relapses. While our review attempted to review data on other effectiveness measures such as hospitalizations, ER visits, and other health resource use, it was limited by the paucity of data on these outcomes. Available data suggest that compared to placebo, odds of all-cause discontinuations were lower with aripiprazole while olanzapine, quetiapine, risperidone, and brexpiprazole reported no differences. While aripiprazole and olanzapine had lower discontinuation odds due to lack of efficacy, olanzapine had higher discontinuation odds due to lack of safety. Compared to placebo, odds of discontinuation were found to be higher for aripiprazole, olanzapine, risperidone, quetiapine and brexpiprazole. Based on the reported information, it appears that risperidone had a lower likelihood of relapse of NPS than placebo. Studies of olanzapine, quetiapine, brexpiprazole did not report treatment effects based on relapse of psychosis.
For multiple drug comparator studies from the CATIE-AD trial, effectiveness for olanzapine, quetiapine and risperidone was measured through discontinuation or effects of AAP withdrawal. Ruths et al. 2004 (54) reported that for olanzapine and risperidone most patients’ behavioral scores remained stable after the withdrawal of APs from nursing home (NH) patients with dementia. However, impairment in the patient’s nighttime and daytime activity was reported as sleep problems and restlessness (54). In another double-blind, placebo-controlled trial of 421 outpatients with AD and psychosis or aggression/agitation, Schneider et al. 2006 (56) found no significant differences in time to all-cause discontinuation (i.e., discontinuation for any reason) among olanzapine, quetiapine and risperidone.
As with any research, this SLR has a few limitations. Specifically, the SLR included trials beyond the gold standard for double-blind, randomized trials and thus resulted in 10% of studies of low-quality being included with high risk of bias for blinding of participants and personnel. This could be attributed to the inclusion of non-blinded and open label studies in the SLR. It is anticipated that the small proportion of low-quality studies included in this qualitative review of the SLR are not likely to change the overall conclusions. However, a sensitivity analysis may be considered during quantitative NMA, if any, based on this SLR’s findings to evaluate the impact of low-quality studies included in the SLR. It is recommended that the NMA ought to consider issues of heterogeneity, transitivity, and other inherent problems to overcome any intrinsic study-related biases. Notwithstanding these few limitations, this review represents the most comprehensive analysis to-date.



Consistent with previous findings, the overall evidence from this SLR suggest that currently used AAPs described in this study confer non-significant benefits in treating dementia related hallucinations and delusions. Additionally, they are associated with a high risk of significant AEs, accelerated cognitive decline, and potentially higher mortality among patients with DRP. Furthermore, antipsychotic effectiveness may be poor, given the potentially high rates of all-cause discontinuations and discontinuations due to AEs reported in the studies. Overall, these qualitative findings suggest that these AAPs used as off-label treatments for the vulnerable DRP population may have an unfavorable benefit-risk profile and require quantitative confirmation through an NMA of data derived from this SLR. These results also underscore the potential unmet need for new treatment options with an improved benefit-risk profile for the treatment of DRP.


Ethics approval and consent to participate: Not applicable.

Consent for publication: Not applicable.

Data availability statement: Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Competing interests: NR and VA are employees of Acadia Pharmaceuticals, Inc. which has funded the research. The authors confirm that funding has in no way influenced the outcome. There are no conflicts of interest associated with this publication.

Funding: The research was funded by Acadia Pharmaceuticals Inc. Employees of Acadia Pharmaceuticals Inc. were a part of the research team for development of concept, study design, study selection and interpretation of the results.

Authors’ contributions: IY, NR, VA, KR development of concept and study design; IY, and KR literature search; IY, NR, VA, KR study selection, and interpretation of data; IY, NR, VD, KR preparation of manuscript; All authors critically reviewed and approved the final version of the paper.

Acknowledgements: Author KR is an employee, while IY is a former employee of Anlitiks Inc which is a part of the funded research group.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, 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.





1. Association Asd. Alzheimer’s disease facts and figures report. Accessed accessed June 26 2020.
2. Prince M, Ali GC, Guerchet M, Prina AM, Albanese E, Wu YT. Recent global trends in the prevalence and incidence of dementia, and survival with dementia. Alzheimers Res Ther. 2016;8(1):23.
3. Sink KM, Holden KF, Yaffe K. Pharmacological treatment of neuropsychiatric symptoms of dementia: a review of the evidence. Jama. 2005;293(5):596-608.
4. Kales HC, Gitlin LN, Lyketsos CG, Detroit Expert Panel on A, Management of Neuropsychiatric Symptoms of D. Management of neuropsychiatric symptoms of dementia in clinical settings: recommendations from a multidisciplinary expert panel. J Am Geriatr Soc. 2014;62(4):762-769.
5. Small GW. Managing the Burden of Dementia-Related Delusions and Hallucinations. Supplement to The Journal of Family Practice.69(7).
6. Paulsen JS, Salmon DP, Thal LJ, et al. Incidence of and risk factors for hallucinations and delusions in patients with probable AD. Neurology. 2000;54(10):1965-1971.
7. Wilson RS, Gilley DW, Bennett DA, Beckett LA, Evans DA. Hallucinations, delusions, and cognitive decline in Alzheimer’s disease. J Neurol Neurosurg Psychiatry. 2000;69(2):172-177.
8. Scarmeas N, Brandt J, Albert M, et al. Delusions and Hallucinations Are Associated With Worse Outcome in Alzheimer Disease. Archives of Neurology. 2005;62(10):1601-1608.
9. Maust DT, Kim HM, Seyfried LS, et al. Antipsychotics, other psychotropics, and the risk of death in patients with dementia: number needed to harm. JAMA Psychiatry. 2015;72(5):438-445.
10. Hsu WT, Esmaily-Fard A, Lai CC, et al. Antipsychotics and the Risk of Cerebrovascular Accident: A Systematic Review and Meta-Analysis of Observational Studies. J Am Med Dir Assoc. 2017;18(8):692-699.
11. Mittal V, Kurup L, Williamson D, Muralee S, Tampi RR. Risk of cerebrovascular adverse events and death in elderly patients with dementia when treated with antipsychotic medications: a literature review of evidence. Am J Alzheimers Dis Other Demen. 2011;26(1):10-28.
12. Citrome L, Eramo A, Francois C, et al. Lack of tolerable treatment options for patients with schizophrenia. Neuropsychiatric disease and treatment. 2015;11:3095.
13. Papola D, Ostuzzi G, Thabane L, Guyatt G, Barbui C. Antipsychotic drug exposure and risk of fracture: a systematic review and meta-analysis of observational studies. Int Clin Psychopharmacol. 2018;33(4):181-196.
14. Reus VI, Fochtmann LJ, Eyler AE, et al. The American Psychiatric Association Practice Guideline on the Use of Antipsychotics to Treat Agitation or Psychosis in Patients With Dementia. Focus (Am Psychiatr Publ). 2017;15(1):81-84.
15. Yunusa I, Alsumali A, Garba AE, Regestein QR, Eguale T. Assessment of Reported Comparative Effectiveness and Safety of Atypical Antipsychotics in the Treatment of Behavioral and Psychological Symptoms of Dementia: A Network Meta-analysis. JAMA Netw Open. 2019;2(3):e190828.
16. Watt JA, Goodarzi Z, Veroniki AA, et al. Safety of pharmacologic interventions for neuropsychiatric symptoms in dementia: a systematic review and network meta-analysis. BMC Geriatr. 2020;20(1):212.
17. Grossberg GT, Kohegyi E, Mergel V, et al. Efficacy and safety of brexpiprazole for the treatment of agitation in Alzheimer’s dementia: two 12-week, randomized, double-blind, placebo-controlled trials. The American Journal of Geriatric Psychiatry. 2020;28(4):383-400.
18. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Int J Surg. 2021:105906.
19. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928.
20. Wells G, Shea B, O’Connell D, et al. The Newcastle–Ottawa Scale (NOS) for Assessing the Quality of Non-Randomized Studies in Meta-Analysis. ᅟ. 2000;ᅟ.
21. Ballard C, Lana MM, Theodoulou M, et al. A randomised, blinded, placebo-controlled trial in dementia patients continuing or stopping neuroleptics (the DART-AD trial). PLoS Med. 2008;5(4):e76.
22. Ballard C, Margallo-Lana M, Juszczak E, et al. Quetiapine and rivastigmine and cognitive decline in Alzheimer’s disease: randomised double blind placebo controlled trial. Bmj. 2005;330(7496):874.
23. Barak Y, Plopski I, Tadger S, Paleacu D. Escitalopram versus risperidone for the treatment of behavioral and psychotic symptoms associated with Alzheimer’s disease: a randomized double-blind pilot study. International psychogeriatrics. 2011;23(9):1515-1519.
24. Brodaty H, Ames D, Snowdon J, et al. Risperidone for psychosis of Alzheimer’s disease and mixed dementia: results of a double-blind, placebo-controlled trial. International Journal of Geriatric Psychiatry: A journal of the psychiatry of late life and allied sciences. 2005;20(12):1153-1157.
25. Brodaty H, Ames D, Snowdon J, et al. A randomized placebo-controlled trial of risperidone for the treatment of aggression, agitation, and psychosis of dementia. The Journal of clinical psychiatry. 2003.
26. Chan Wc, Lam LCw, Choy CNp, Leung VPy, Li Sw, Chiu HFk. A double-blind randomised comparison of risperidone and haloperidol in the treatment of behavioural and psychological symptoms in Chinese dementia patients. International journal of geriatric psychiatry. 2001;16(12):1156-1162.
27. Clark WS, Street JS, Feldman PD, Breier A. The effects of olanzapine in reducing the emergence of psychosis among nursing home patients with Alzheimer’s disease. The Journal of clinical psychiatry. 2001.
28. Culo S, Mulsant BH, Rosen J, et al. Treating neuropsychiatric symptoms in dementia with Lewy bodies: a randomized controlled-trial. Alzheimer Disease & Associated Disorders. 2010;24(4):360-364.
29. Cummings JL, Street J, Masterman D, Clark WS. Efficacy of olanzapine in the treatment of psychosis in dementia with Lewy bodies. Dementia and geriatric cognitive disorders. 2002;13(2):67-73.
30. De Deyn P, Jeste DV, Swanink R, et al. Aripiprazole for the treatment of psychosis in patients with Alzheimer’s disease: a randomized, placebo-controlled study. Journal of clinical psychopharmacology. 2005;25(5):463-467.
31. De Deyn PP, Carrasco MM, Deberdt W, et al. Olanzapine versus placebo in the treatment of psychosis with or without associated behavioral disturbances in patients with Alzheimer’s disease. International journal of geriatric psychiatry. 2004;19(2):115-126.
32. De Deyn PP, Eriksson H, Svensson H. Tolerability of extended-release quetiapine fumarate compared with immediate-release quetiapine fumarate in older patients with Alzheimer’s disease with symptoms of psychosis and/or agitation: a randomised, double-blind, parallel-group study. Int J Geriatr Psychiatry. 2012;27(3):296-304.
33. Deberdt WG, Dysken MW, Rappaport SA, et al. Comparison of olanzapine and risperidone in the treatment of psychosis and associated behavioral disturbances in patients with dementia. The American journal of geriatric psychiatry. 2005;13(8):722-730.
34. Devanand D, Mintzer J, Schultz SK, et al. Relapse risk after discontinuation of risperidone in Alzheimer’s disease. New England Journal of Medicine. 2012;367(16):1497-1507.
35. Ellingrod VL, Schultz SK, Ekstam-Smith K, Kutscher E, Turvey C, Arndt S. Comparison of risperidone with olanzapine in elderly patients with dementia and psychosis. Pharmacotherapy. 2002;22(1):1-5.
36. Freund-Levi Y, Jedenius E, Tysen-Bäckström AC, Lärksäter M, Wahlund L-O, Eriksdotter M. Galantamine versus risperidone treatment of neuropsychiatric symptoms in patients with probable dementia: an open randomized trial. The American journal of geriatric psychiatry. 2014;22(4):341-348.
37. Fujikawa T, Takahashi T, Kinoshita A, et al. Quetiapine treatment for behavioral and psychological symptoms in patients with senile dementia of Alzheimer type. Neuropsychobiology. 2004;49(4):201-204.
38. Gareri P, Cotroneo A, Lacava R, et al. Comparison of the efficacy of new and conventional antipsychotic drugs in the treatment of behavioral and psychological symptoms of dementia (BPSD). Archives of gerontology and geriatrics Supplement. 2004(9):207.
39. Holmes C, Wilkinson D, Dean C, et al. Risperidone and rivastigmine and agitated behaviour in severe Alzheimer’s disease: a randomised double blind placebo controlled study. Int J Geriatr Psychiatry. 2007;22(4):380-381.
40. Jeste DV, Okamoto A, Napolitano J, Kane JM, Martinez RA. Low incidence of persistent tardive dyskinesia in elderly patients with dementia treated with risperidone. American Journal of Psychiatry. 2000;157(7):1150-1155.
41. Kurlan R, Cummings J, Raman R, Thal L. Quetiapine for agitation or psychosis in patients with dementia and parkinsonism. Neurology. 2007;68(17):1356-1363.
42. Kurz A, Schwalen S, Schmitt A. Effects of risperidone on behavioral and psychological symptoms associated with dementia in clinical practice. International psychogeriatrics. 2005;17(4):605.
43. Laks J, Engelhardt E, Marinho V, et al. Efficacy and safety of risperidone oral solution in agitation associated with dementia in the elderly. Arquivos de neuro-psiquiatria. 2001;59(4):859-864.
44. Lim H-K, Pae C-U, Lee C, Lee C-U. Amisulpride versus risperidone treatment for behavioral and psychological symptoms in patients with dementia of the Alzheimer type: a randomized, open, prospective study. Neuropsychobiology. 2006;54(4):247-251.
45. Mintzer J, Greenspan A, Caers I, et al. Risperidone in the treatment of psychosis of Alzheimer disease: results from a prospective clinical trial. The American journal of geriatric psychiatry. 2006;14(3):280-291.
46. Mintzer JE, Tune LE, Breder CD, et al. Aripiprazole for the treatment of psychoses in institutionalized patients with Alzheimer dementia: a multicenter, randomized, double-blind, placebo-controlled assessment of three fixed doses. The American Journal of Geriatric Psychiatry. 2007;15(11):918-931.
47. Mullen J, Jibson MD, Sweitzer D. A comparison of the relative safety, efficacy, and tolerability of quetiapine and risperidone in outpatients with schizophrenia and other psychotic disorders: the quetiapine experience with safety and tolerability (QUEST) study. Clin Ther. 2001;23(11):1839-1854.
48. Onor ML, Saina M, Aguglia E. Efficacy and tolerability of quetiapine in the treatment of behavioral and psychological symptoms of dementia. American Journal of Alzheimer’s Disease & Other Dementias®. 2007;21(6):448-453.
49. Paleacu D, Barak Y, Mirecky I, Mazeh D. Quetiapine treatment for behavioural and psychological symptoms of dementia in Alzheimer’s disease patients: a 6-week, double-blind, placebo-controlled study. International Journal of Geriatric Psychiatry: A journal of the psychiatry of late life and allied sciences. 2008;23(4):393-400.
50. Pollock BG, Mulsant BH, Rosen J, et al. A double-blind comparison of citalopram and risperidone for the treatment of behavioral and psychotic symptoms associated with dementia. The American Journal of Geriatric Psychiatry. 2007;15(11):942-952.
51. Rabinowitz J, Katz I, De Deyn PP, Greenspan A, Brodaty H. Treating behavioral and psychological symptoms in patients with psychosis of Alzheimer’s disease using risperidone. International psychogeriatrics. 2007;19(2):227-240.
52. Rainer M, Haushofer M, Pfolz H, Struhal C, Wick W. Quetiapine versus risperidone in elderly patients with behavioural and psychological symptoms of dementia: efficacy, safety and cognitive function. European psychiatry. 2007;22(6):395-403.
53. Rocha FL, Hara C, Ramos MG, et al. An exploratory open-label trial of ziprasidone for the treatment of behavioral and psychological symptoms of dementia. Dementia and geriatric cognitive disorders. 2006;22(5-6):445-448.
54. Ruths S, Straand J, Nygaard HA, Bjorvatn B, Pallesen S. Effect of antipsychotic withdrawal on behavior and sleep/wake activity in nursing home residents with dementia: a randomized, placebo-controlled, double-blinded study The Bergen District Nursing Home Study. J Am Geriatr Soc. 2004;52(10):1737-1743.
55. Scharre DW, Chang S-I. Cognitive and behavioral effects of quetiapine in Alzheimer disease patients. Alzheimer Disease & Associated Disorders. 2002;16(2):128-130.
56. Schneider LS, Tariot PN, Dagerman KS, et al. Effectiveness of atypical antipsychotic drugs in patients with Alzheimer’s disease. New England Journal of Medicine. 2006;355(15):1525-1538.
57. Street JS, Clark WS, Gannon KS, et al. Olanzapine treatment of psychotic and behavioral symptoms in patients with Alzheimer disease in nursing care facilities: a double-blind, randomized, placebo-controlled trial. Archives of General Psychiatry. 2000;57(10):968-976.
58. Streim JE, Porsteinsson AP, Breder CD, et al. A randomized, double-blind, placebo-controlled study of aripiprazole for the treatment of psychosis in nursing home patients with Alzheimer disease. The American Journal of Geriatric Psychiatry. 2008;16(7):537-550.
59. Suh GH, Greenspan AJ, Choi SK. Comparative efficacy of risperidone versus haloperidol on behavioural and psychological symptoms of dementia. International Journal of Geriatric Psychiatry: A journal of the psychiatry of late life and allied sciences. 2006;21(7):654-660.
60. Sultzer DL, Davis SM, Tariot PN, et al. Clinical symptom responses to atypical antipsychotic medications in Alzheimer’s disease: phase 1 outcomes from the CATIE-AD effectiveness trial. American Journal of Psychiatry. 2008;165(7):844-854.
61. Tariot PN, Salzman C, Yeung PP, Pute J, Rak IW. Long-term use of quetiapine in elderly patients with psychotic disorders. Clinical Therapeutics. 2000;22(9):1068-1084.
62. Tariot PN, Schneider L, Katz IR, et al. Quetiapine treatment of psychosis associated with dementia: a double-blind, randomized, placebo-controlled clinical trial. The American journal of geriatric psychiatry. 2006;14(9):767-776.
63. Teranishi M, Kurita M, Nishino S, et al. Efficacy and tolerability of risperidone, yokukansan, and fluvoxamine for the treatment of behavioral and psychological symptoms of dementia: a blinded, randomized trial. Journal of clinical psychopharmacology. 2013;33(5):600-607.
64. Verhey FR, Verkaaik M, Lousberg R. Olanzapine versus haloperidol in the treatment of agitation in elderly patients with dementia: results of a randomized controlled double-blind trial. Dementia and geriatric cognitive disorders. 2006;21(1):1-8.
65. Vigen CL, Mack WJ, Keefe RS, et al. Cognitive effects of atypical antipsychotic medications in patients with Alzheimer’s disease: outcomes from CATIE-AD. American Journal of Psychiatry. 2011;168(8):831-839.
66. Yang X-P, Wang Y-T, Liu D-C, et al. Comparison of the effects of tiapride and risperidone on psycho-behavioral symptoms of senile dementia. Int J Clin Exp Med. 2016;9(6):9593-9597.
67. Yoon JS, Kim JM, Lee H, Shin IS, Choi SK. Risperidone use in Korean patients with Alzheimer’s disease: optimal dosage and effect on behavioural and psychological symptoms, cognitive function and activities of daily living. Human Psychopharmacology: Clinical and Experimental. 2003;18(8):627-633.
68. Zhong KX, Tariot P, Mintzer J, Minkwitz M, Devine N. Quetiapine to treat agitation in dementia: a randomized, double-blind, placebo-controlled study. Current Alzheimer Research. 2007;4(1):81-93.
69. Mintzer J, Faison W, Street JS, Sutton VK, Breier A. Olanzapine in the treatment of anxiety symptoms due to Alzheimer’s disease: a post hoc analysis. International journal of geriatric psychiatry. 2001;16(S1):S71-S77.
70. Mulsant BH, Gharabawi GM, Bossie CA, et al. Correlates of anticholinergic activity in patients with dementia and psychosis treated with risperidone or olanzapine. J Clin Psychiatry. 2004;65(12):1708-1714.
71. Rainer MK, Masching AJ, Ertl MG, Kraxberger E, Haushofer M. Effect of risperidone on behavioral and psychological symptoms and cognitive function in dementia. J Clin Psychiatry. 2001;62(11):894-900.
72. Savaskan E, Schnitzler C, Schröder C, Cajochen C, Müller-Spahn F, Wirz-Justice A. Treatment of behavioural, cognitive and circadian rest-activity cycle disturbances in Alzheimer’s disease: haloperidol vs. quetiapine. Int J Neuropsychopharmacol. 2006;9(5):507-516.
73. Weiser M, Rotmensch HH, Korczyn AD, Hartman R, Cicin-Sain A, Anand R. A pilot, randomized, open-label trial assessing safety and pharmakokinetic parameters of co-administration of rivastigmine with risperidone in dementia patients with behavioral disturbances. Int J Geriatr Psychiatry. 2002;17(4):343-346.
74. Mowla A, Pani A. Comparison of topiramate and risperidone for the treatment of behavioral disturbances of patients with Alzheimer disease: a double-blind, randomized clinical trial. J Clin Psychopharmacol. 2010;30(1):40-43.
75. Ruths S, Straand J, Nygaard HA, Aarsland D. Stopping antipsychotic drug therapy in demented nursing home patients: a randomized, placebo-controlled study–the Bergen District Nursing Home Study (BEDNURS). Int J Geriatr Psychiatry. 2008;23(9):889-895.
76. Katz IR, Jeste DV, Mintzer JE, Clyde C, Napolitano J, Brecher M. Comparison of risperidone and placebo for psychosis and behavioral disturbances associated with dementia: a randomized, double-blind trial. Risperidone Study Group. J Clin Psychiatry. 1999;60(2):107-115.
77. Kinon BJ, Stauffer VL, McGuire HC, Kaiser CJ, Dickson RA, Kennedy JS. The effects of antipsychotic drug treatment on prolactin concentrations in elderly patients. J Am Med Dir Assoc. 2003;4(4):189-194.
78. Antipsychotic drugs for behavioural and psychological symptoms. Accessed March 27, 2021.
79. Corbett A, Burns A, Ballard C. Don’t use antipsychotics routinely to treat agitation and aggression in people with dementia. BMJ. 2014;349:g6420.
80. Papola D, Ostuzzi G, Gastaldon C, et al. Antipsychotic use and risk of life-threatening medical events: umbrella review of observational studies. Acta Psychiatr Scand. 2019;140(3):227-243.
81. Ballard C, Howard R. Neuroleptic drugs in dementia: benefits and harm. Nat Rev Neurosci. 2006;7(6):492-500.




N. Costa1,2,3, M. Mounié1,2, A. Pagès2,4,5, H. Derumeaux1, T. Rapp6, S. Guyonnet2,3,5, N. Coley2,3,7, C. Cantet2,3,5, I. Carrié5, S. Andrieu2,3,7, L. Molinier1,2,3 and on behalf of the MAPT/DSA Group*


1. Health Economic Unit of the University Hospital of Toulouse, Toulouse 31059, France; 2. INSERM-UMR 1027, Toulouse 31000, France; 3. University of Toulouse III, Toulouse 31330, France; 4. Department of Pharmacy, Toulouse University Hospital, Toulouse, 31000, France; 5. Gérontopôle, Department of Geriatrics, University Hospital of Toulouse, Toulouse 31059, France; 6. Université de Paris (LIRAES EA4470), Paris 75006, Paris, France; 7. Department of Epidemiology and Public Health, university Hospital of Toulouse, Toulouse 31059, France

Corresponding Author: Nadège Costa, Health economist, PhD, Health Economic Unit of the University Hospital of Toulouse, Hôtel Dieu Saint-Jacques, 2, rue viguerie, 31059 Toulouse Cedex 9, France, Email :, Tel: +335 61 77 83 72

J Prev Alz Dis 2021;4(8):425-435
Published online August 2, 2021,



BACKGROUND: To date, no curative treatment is available for Alzheimer’s disease (AD). Therefore, efforts should focus on prevention strategies to improve the efficiency of healthcare systems.
Objective: Our aim was to assess the cost-effectiveness of three preventive strategies for AD compared to a placebo.
Design: The Multidomain Alzheimer Preventive Trial (MAPT) study was a multicenter, randomized, placebo-controlled superiority trial with four parallel groups, including three intervention groups (one group with Multidomain Intervention (MI) plus a placebo, one group with Polyunsaturated Fatty Acids (PFA), one group with a combination of PFA and MI) and one placebo group.
Setting: Participants were recruited and included in 13 memory centers in France and Monaco.
Participants: Community-dwelling subject aged 70 years and older were followed during 3 years.
Interventions: We used data from the MAPT study which aims to test the efficacy of a MI along PFA, the MI plus a placebo, PFA alone, or a placebo alone.
Measurement: Direct medical and non-medical costs were calculated from a payer’s perspective during the 3 years of follow-up. The base case incremental Cost-Effectiveness Ratio (ICER) represents the cost per improved cognitive Z-score point. Sensitivity analyses were performed using different interpretation of the effectiveness criteria.
Results: Analyses were conducted on 1,525 participants. The ICER at year 3 that compares the MI + PFA and the MI alone to the placebo amounted to €21,443 and €21,543 respectively, per improved Z score point. PFA alone amounted to €111,720 per improved Z score point.
Conclusion: Our study shows that ICERS of PFA combined with MI and MI alone amounted to €21,443 and €21,543 respectively per improved Z score point compared to the placebo and are below the WTP of €50,000 while the ICER of PFA alone amounted to €111,720 per improved Z score point. This information may help decision makers and serve as a basis for the implementation of a lifetime decision analytic model.

Key words: Cost-effectiveness, economics, Alzheimer disease, prevention.



According to the 2019 World Alzheimer report, 50 million people worldwide and 1.2 millions in France suffer from Alzheimer’s disease (AD) (1). Associated costs of care are consistent and vary from €24,140 for mild and moderate stages to €44,171 for the severe stage at 18 months (2).
According to the latest meta-analyses, specific drugs in the treatment of AD have a low and time-limited efficacy on symptoms, quality of life, institutionalization, mortality and the burden of caregivers (3, 4). In 2016, the French High Authority for Health (HAS) considered that the benefit of these medicines was insufficient to justify reimbursement by the French National Health Insurance (FNHI) (5, 6).
As no curative treatment is available, efforts should focus on prevention strategies. Current evidence suggests that nutrition, physical exercise, cognitive activity and social stimulation may improve cognitive health (7). Results from the Multidomain Alzheimer Preventive Trial (MAPT), which test the effect of Multidomain Intervention (MI) and supplementation using omega 3 polyunsaturated fatty acids (PFA) alone or in combination on cognitive decline alongside a large randomized controlled trial show no significant differences in 3-year cognitive decline between any of the three intervention groups and the placebo group (8). Nevertheless, this trial shows a trend in z-score differences in favor of MI + PFA and MI alone groups.
Published cost-effectiveness analyses of primary prevention strategies for AD show cost-saving results. Nevertheless, these studies are only based on simulated models and hypothetical interventions indicating potential cost-effectiveness results (9). Interventions tested were physical activity, management of cardiovascular risk factors, vitamin supplementation, and multidomain cardiovascular disease prevention programs (10-13). Currently in France, these interventions are not reimbursed specifically for the prevention of Alzheimer’s disease but they can be offered to the patient for the maintenance of their overall health. More randomized control trials (RCT) are required to reinforce the results cost-effectiveness study of prevention programs for AD.
In the framework of the large MAPT study, we aim to assess the cost-effectiveness of PFA supplementation alone, MI (nutritional counseling, physical exercise, and cognitive stimulation) alone or a combination of both interventions compared to a placebo.



Design, setting and participants

The MAPT study was a multicenter, randomized, placebo-controlled superiority trial with four parallel groups, including three intervention groups (one group with MI plus a placebo, one group with PFA, one group with a combination of PFA and MI) and one placebo group. Community-dwelling subjects, followed during 3 years, aged 70 years and older were recruited at 13 memory centers in France and Monaco. In France, memory centers are outpatient structures that performed diagnostic workup and follow-up of elderly subjects. Participants met at least one of three criteria: spontaneous memory complaint, limitation in one instrumental Activity of Daily Living (ADL), or slow gait were eligible to be included in the study. Participants with a Mini Mental State Examination (MMSE) score below 24, those who were diagnosed with dementia, those with any difficulty in basic ADL and those taking PFA supplements at baseline were excluded. Full methods have been previously described elsewhere (8, 14). The trial protocol was approved by the French Ethics Committee in Toulouse (CPP SOOM II) and was authorized by the French Health Authority (8).


Participants were randomly assigned to one of the following four groups:
– Multidomain Intervention: consisted of 2 h group sessions focusing on three domains (cognitive stimulation, physical activity, and nutrition) and a preventive consultation (at baseline, 12 months, and 24 months). MI was done twice a week during the first month, once a week during the second month, and one per month for the remainder of the three years study,
– Omega 3 Polyunsaturated Fatty Acids: two capsules per day with 400 mg docosahexaenoic acid (DHA) and no more than 112·5 mg eicosapentaenoic acid (EPA),
– Combined intervention: Multidomain intervention and Omega-3 PFA,
– Placebo: two capsules per day containing flavoured paraffin oil.

More details are given elsewhere (8).


All costs were recorded throughout the MAPT trial at 6, 12, 18, 24, 30 and 36 months using a Case Report Form and were analysed from the FNHI perspective. All monetary values are in 2018 Euros. Costs taken into account were direct medical (hospitalizations, consultations, medical and paramedical procedures and drugs) and non-medical (transportation) costs.
Valuation was based on several sources of unit costs (Appendix 1. Table A1). Hospital stays were valued using the French Disease Related Groups (DRG) including extra charges if applicable (e.g. the cost of days of intensive care) (15). We used mean DRG rates calculated from the national hospitalization database for patients aged 70 or over, according to the medical unit to which the participant was admitted. Rehabilitation and psychiatric hospitalizations were valued using per diem costs. Consultations were valued using the General French Nomenclature for Medical Procedures according to the specialization (16). Medical procedures were valued using the Medical Classification of Clinical Procedures (CCAM) (Version 54.10) and the Nomenclature of Clinical Biological Procedures (NABM) according to the type of medical procedure (imaging, biology, other) (17, 18). Each consultation and medical procedure was valued using the appropriate FNHI reimbursement rate.
No details, except the frequency, were available in the database on transportation and paramedical procedures, therefore valuation was based on means estimate from a sample of the FNHI database, the General Sample of Beneficiaries database (EGB) (19, 20). The gamma distribution shape and scale parameters were derived from the mean and variance observed in the 2018 EGB database for each cost component for the population aged 70 years or older.
For drugs reimbursed by the FNHI, we assumed that the daily dosage was equal to the Daily Defined Dose (DDD) (21). If there was no recommended DDD, we calculated an average daily dose according to the Summary of Product Characteristics (SmPC) (22). We then multiplied the reimbursement price per unit by the daily dosage and the treatment duration (23). For hospital drugs, only the costs of very expensive drugs were taken into account because the others were included in the DRG rate (24).
MI was valued by the mean wage rate for a psychologist, dietician and physical activity facilitator (40€) multiplied by the intervention period (2.30 hours) and the number of prescribed sessions (46) during three years. PFA was valued using the retail price per capsule (€0.50 cents) multiplied by the number of prescribed capsules taken per participant per year (365.5/year), multiplied by 3 years.
The primary efficacy outcome used to determine the ICER consisted of the change from baseline to 36 months in a composite Z score (8). It combines four cognitive tests (free and total recall of the Free and Cued Selective Reminding Test, ten MMSE orientation items, the Digit Symbol Substitution Test score from the Revised Wechsler Adult Intelligence Scale, and the Category Naming Test [2 min category fluency in animals]) (8). Z-scores represent the number of standard deviations above or below the mean. Coley et al estimated that a 0.3-point decrease in Z score is the minimum clinically significant difference, which predicts dementia (25). We used this cut-off, in addition to the Z-score, to define whether a participant presented an aggravation in memory function in order to make the ICER more comprehensive for clinicians and decision makers. Other variables (age, gender, comorbidities, Fried frailty phenotype, educational level and Z score) were collected at baseline.

Statistical analyses

Description and comparison of baseline characteristics were made using mean and standard deviation and occurrences and percentages for continuous and qualitative variables on one hand and using Kruskal Wallis or Chi-squared on the other hand.
Cost components for participants who had a complete follow-up were summarized for each group. Three-year cumulative costs were expressed in terms of mean costs per participant and their bias-corrected and accelerated (BCa) bootstrap 95% confidence intervals (CI). Cost differences between groups where tested using a global non-parametric Kruskal-Wallis test.
Missing data on total cumulative cost at 3 years were accounted for by multiple imputation and predictive mean matching methods (26). Age group tercile, gender, intervention groups, initial frailty score tercile and pooled occurrences of medical history tercile were used in the imputation. We assumed that missing cost data are “Missing At Random” and we used Hausman test to verify whether our results were subject to attrition bias issues (27). Efficacy data used in our analysis were smoothed by a mixed model as described elsewhere (8). The fixed effects used in this model were intervention group, time, and interactions between intervention groups and each time. The random effects used were center-specific and participant-specific variables. In order to include adjusted outcomes in both the numerator and denominator of the ICER, we used a Generalized Linear Model (GLM) with a gamma distribution and a log link that allowed the use of fitted cost data (28). The same variables used for imputation were also used for adjustment. Fitted and imputed costs as well as fitted Z scores were then described using mean and BCA bootstrap CI.
We used non-parametric bootstrap outputs to graphically determined the 95% confidence ellipses and illustrate the uncertainty around the ICER (29, 30). ICERs with a positive value were compared to a Willingness-To-Pay (WTP) threshold set up at 50,000€ per Quality Adjusted Life Years (31-33). Additionally, the cost-effectiveness acceptability curve (CEAC) showed the probability that an intervention was cost-effective compared with the alternative according to a range of WTP thresholds (34). Moreover, sensitivity analysis was conducted using the data for patient with a complete follow-up.



Patient’s characteristics

Between 30 May 2008 and 24 February 2011, 1,680 participants were enrolled and randomly assigned to four arms. Participants were excluded from the modified intention-to-treat efficacy analysis because no cognitive assessment was available after baseline for 154 participants, and one participant in the PFA group withdrew their consent. One thousand two hundred and eighty-six participants completed the final visit and economic data were available for 1,320 participants (Figure 1). Missing economic data accounted for 12% to 15% in each group. A full description of the population was provided in prior work (14). The baseline characteristics of our sample are summarized in Table 1. No substantial difference was noted in any demographic or clinical characteristics between the arms.

Figure 1. Flow chart for patient’s selection

*The intention-to-treat analysis included assigned participants with a composite score at baseline who had at least one post-baseline visit.

Table 1. Participants’ characteristics at baseline

*p-value of khi2 or Kruskal Wallis test according to variables; † Gastrointestinal; ‡Ear Nose and Throat; §Polyunsaturated Fatty Acids; || Multidomain intervention

Three years costs description

The observed costs for the three-year follow-up period for 1,320 participants with complete economic data are presented in table 2. Total costs without intervention amounted to €7,702; €7,951; €7,845 and €7,106 for PFA + MI, PFA, MI and the placebo group, respectively (p=0.77). When the intervention cost was included in the total costs, they were significantly different and amounted to €9,171; €8,500; €8,765 and €7,106, respectively (p=0.001). The main cost driver in each group was inpatient stays which accounted for approximately 50% of the total cost in all groups. The second cost driver was medication, which accounted for 24% to 30% of the total cost depending on the group.
At 3 years, the placebo group had the lowest inpatient costs of the three groups, and particularly for psychiatric hospitalizations that were higher in the three other groups. The PFA group had significantly higher GP, cardiologist and lab test costs than others groups (p= 0.026, p= 0.018 and p=0.090, respectively). Finally, cardiovascular medication costs were higher in the PFA + MI group (p=0.018).

Table 2. Total costs over the three-year follow-up period

*Confidence Interval; †p-value of Kruskal-Wallis rank sum test; ‡Medical Surgical and Obstetrics; §Polyunsaturated Fatty Acids; ||Multidomain Intervention


Three years costs analysis

Detailed costs for every 6 months show a substantial increase in total costs for each group between 24 and 36 months of follow-up, which was mainly due to a substantial increase in hospital costs (Appendix 2. Table A2).
Table 3 presents the results of the GLM for the whole population (1,525) and show that total costs including intervention costs increased with age, number of medical conditions and the type of intervention. It was significantly higher in the PFA + MI group and the MI group. The GLM regression of total costs without intervention costs shows that only age and the number of medical conditions increased healthcare costs significantly.

Table 3. Multivariate analysis of total cost with and without intervention over the 3-year follow-up period (N= 1,525)

*Relative Risk; †Confidence Interval; ‡p-value; §Polyunsaturated Fatty Acids; ||Multidomain Intervention


Cost-effectiveness analysis

Differences in total costs including intervention costs between the intervention groups and the placebo group were €1,237, €1,705 and €1,986 for the PFA, MI and PFA + MI groups, respectively ( Appendix 3. Table A3). Changes in Z scores between the intervention groups and the control group were 0.011 for PFA, 0.079 for MI and 0.093 for PFA + MI, respectively (Appendix 3. Table A3).
As presented in the base case ICER scatter plot (Figure 2-a), the ICER comparing combined intervention and MI alone with placebo amounted to €21,443 and €21,543 per improved Z score point, respectively. The confidence ellipses of the ICERs comparing the PFA + MI and MI strategies overlap. All dots that represent the results of the 1,000 replications of ICERs for the PFA + MI and the MI strategies alone vs. placebo are concentrated in the northeast quadrant. As presented in the CEAC (Appendix 4. Figure A4.a), PFA + MI and MI alone have a probability of 95% to be cost-effective at a €50,000 WTP threshold. When the percentage of patients with no aggravation of cognitive functions between baseline and year 3 (Figure 2.b) is used, it can be noted that all the bootstrapped ICERs of the PFA + MI strategy vs. placebo are located in the northeast quadrant. The probability that PFA + MI and MI alone are cost-effective at a €50,000 threshold is 90% and 65%, respectively. (Appendix 4. Figure A4.b).
Results for the sensitivity analysis using the complete data set show an ICER amounting to €19,638 and €20,595 per improved Z score point for combined intervention and MI alone compared to placebo. All dots that represent the results of the 1,000 replications of ICERs for the PFA + MI and the MI strategies alone vs. placebo are concentrated in the northeast quadrant (Appendix 5).

Figure 2. Confidence ellipses of intervention strategies versus placebo



This study provides first time evidence on the cost-effectiveness of preventive interventions for AD. Our study showed that PFA + MI and MI alone have an ICER of €21,443 and €21,543 respectively per improved Z score point compared to the placebo and are below the WTP of €50,000. Clinical results from the MAPT study showed that in the modified intention-to-treat population (n=1525), there were no significant differences in 3-year cognitive decline between any of the three intervention groups and the placebo group, explaining the impossibility to conclude that an intervention was most efficient than another (8). Between-group differences compared with the placebo were 0.093 (95% CI 0.001 to 0.184; adjusted p=0.142) for the combined intervention group, 0.079 (-0.012 to 0.170; adjusted p=0.179) for the MI plus placebo group, and 0.011 (-0.081 to 0.103; adjusted p=0.812) for the PFA group. Although the clinical results do not show any significant differences in efficacy between the different interventions studied, we can note a trend regarding the increase in efficacy for the combined intervention and MI groups in comparison with placebo, with a p value less than 0.2. In this context, the implementation of a cost-minimisation analysis was not appropriate, because interventions effectiveness were not strictly equivalent, that is why we choose to implement cost-effectiveness analyses. Additionally, clinical efficacy and efficiency are different measurement tools that have different aims. Efficiency measurement provides information on whether healthcare resources are used to get the best value for money while efficacy measurement determines whether an intervention produces the expected result under ideal circumstances (35).
Two efficiency studies with interventions to reduce risk factors for dementia showed cost-saving results (11, 13). The Lin et al. study used a cohort-based simultaneous equation system in United States with a lifetime time horizon. The intervention (disease management of overweight, diabetes, hypertension and other cardiovascular diseases) was cost saving (-9,259 US$ for a gain of 0.03 LY without dementia) (11). The Zhang et al. study used a Markov model in Sweden and Finland with a 20-year time horizon. The intervention (health promotion program combined with pharmacological treatment of cardiovascular risk factors) was cost saving (-21,974 SEK for a gain of 0.0511 QALY) (13). In another study on physical activity, van Baal and colleagues used a Markov model in United Kingdom with a lifetime time horizon. They calculated incremental costs of -4600 GBP to 1500 GBP depending on the scenarios (physical activity levels and adherence to recommendations), the interventions were cost saving or cost-effective depending on the context, and the maximum ICER was £2,777/LY (10). Finally, an economic evaluation of nutritional supplementation (B-vitamins) in the prevention of dementia based on stochastic decision model in United Kingdom with a lifetime time horizon was tested. Contrary to our study, the intervention was cost saving (-502 GBP for 0.008 QALY gained) (12). However, this supplementation was based on B-vitamins and not PFA. Caution should be exercised in comparing because all these studies were based on hypothetical interventions in decision models and were not RCT like our study (9).
Three years total costs amounted from €7,106 for the placebo group to €7,951 for the PFA group. Costs differences between groups were not statistically significant when interventions costs are not included and becomes significant after the inclusion of these costs. This results show that intervention costs is the main cost component, which explain the difference in total costs. Nevertheless, we can note a cost difference of at least €596 between placebo group and the other three groups. This difference is mainly lead by psychiatric hospitalizations (p=0.012). We can explain this difference by the fact that few patients are hospitalized for psychiatric reasons in each group. In the placebo group, only two psychiatric hospitalizations were found during the three years follow-up period while between four and eight psychiatric hospitalizations were found for the other groups. Moreover, we can note a significant cost difference between groups for consultation costs. This is led by the cost of general practitioner cardiologist, which were higher for PFA group compared with other groups. However, this cost difference from a clinical perspective, correspond between 0.5 to 1.5 consultations in terms of frequency during the three years follow-up period. Annualized costs amounted to €2,567; €2,650; €2,618 and €2,369, for PFA + MI, PFA alone, MI alone and placebo groups, respectively. As shown in the original clinical paper, 45% of the participants included in the MAPT study had at least one Fried frailty criterion and the other participants had none of those criteria. The sample of participants included in the MAPT study was considered as pre-frail or robust [8]. A meta-analysis published in 2019 showed that annual healthcare costs for the elderly varied from €1,217 to €2,056 for a Spanish study and from €9,193 and to €18,525 for a study performed in the USA, for robust and pre-frail older adults, respectively [36]. All the studies included in this meta-analyse took into account inpatient stays, ER and outpatient care. Total costs for Mexican and German studies, which also included formal and informal care costs, varied from €1,248 to €1,775 and from €2,568 to €3,284 for robust and pre-frail older adults, respectively (36). In a French study, the authors demonstrated that annual costs for outpatient care were €1,254 for a robust population of older adults (37). This cost was higher for participants 70-74 years of age and amounted to €1,432. In our study, the mean annualized outpatient care costs amounted to €1,315. A comparison with other studies shows that our cost results are consistent with results in published papers.
The efficacy of MI and/or PFA supplementation was estimated using a Z score. Some countries, such as the UK (NICE), recommend the use of QALY to inform decision-makers for resource allocation. We chose not to use QALY in our study because it is very limited for the elderly. Health related quality of life instruments such as EQ-5D-5L measures do not capture the maintenance of independence or the social effects of interventions, which are particularly important dimensions for the elderly [38]. The QALY metric has also been criticized for being insufficiently sensitive to measure small but clinically meaningful changes in health status or utility (39, 40). In order to provide an ICER that can be informative for clinicians and decision makers, sensitivity analyses were performed on different interpretation of effectiveness using the 0.3-point Z-score as a cut-off (25). ICERs were €434/percent of participants with no aggravation of cognitive functions between baseline and 3 years for PFA + MI compared to the placebo. The use of different interpretation of effectiveness did not change the results and confirmed that the PFA + MI strategy present an ICER under the WTP threshold.
Informal and formal care costs were not included in our analysis. Unlike for demented people for which formal and informal care costs can constitute more than 50% of total costs, these type of costs are very low in older people without dementia (41, 42). As stated in the Panaponaris et al study, only 10% of the older people without dementia needed assistance with ADL and less than 25% needed assistance with IADL (42).
Healthcare consumption was measured using ad hoc questionnaires and was based on declarative data. In order to take into account the uncertainty, we implemented probabilistic sensitivity analyses. We used Probabilistic Sensitive Analysis (PSA) instead of Determinist Sensitive Analysis (DSA) because in DSA analysis the analyst himself chooses parameters and their variation (which leads to selection bias); it only allows the simultaneous variation of a few parameters and cannot take into account the interaction between parameters (43-46). Moreover, the Missing at Random characteristics of our data were verified and the issue of missing data was addressed through multiple imputation. In addition, the sensitivity analysis of the ICERs calculated with the 1,320 participants for whom economic data were available was performed and confirmed our results. Nevertheless, results are based on individual data recorded alongside an RCT, which provided us with robust data analysed using adapted statistical methods.



Results show ICERS of PFA combined with MI and MI alone amounted to €21,443 and €21,543 respectively per improved Z score point compared to the placebo and are below the WTP of €50,000 while the ICER of PFA alone amounted to €111,720 per improved Z score point. These results are consolidated through the sensitivity analyses performed on different effectiveness criteria and ICER calculated using observed data on 1320 participants. The article provides additional information to strictly medical data and may serve as a basis for decision making for the FNHI and more widely for other relevant policy makers. Further results, using lifetime horizon analytical model, are necessary to complete information provided as part of this RCT based study. This study was the first to collect economic and clinical data of older people probably at risk of developing AD during a three years follow-up period. This study may help the scientific community to access additional economic data in the field of AD prevention which can be used on the one hand to build lifetime model and on the other hand, to compare results between different countries and finally contribute to improve the economic research on AD.


* MAPT/DSA Group: Principal investigator: Bruno Vellas (Toulouse); Coordination: Sophie Guyonnet; Project leader: Isabelle Carrié; CRA: Lauréane Brigitte; Investigators: Catherine Faisant, Françoise Lala, Julien Delrieu, Hélène Villars; Psychologists: Emeline Combrouze, Carole Badufle, Audrey Zueras; Methodology, statistical analysis and data management: Sandrine Andrieu, Christelle Cantet, Christophe Morin; Multidomain group: Gabor Abellan Van Kan, Charlotte Dupuy, Yves Rolland (physical and nutritional components), Céline Caillaud, Pierre-Jean Ousset (cognitive component), Françoise Lala (preventive consultation) (Toulouse). The cognitive component was designed in collaboration with Sherry Willis from the University of Seattle, and Sylvie Belleville, Brigitte Gilbert and Francine Fontaine from the University of Montreal. Co-Investigators in associated centres: Jean-François Dartigues, Isabelle Marcet, Fleur Delva, Alexandra Foubert, Sandrine Cerda (Bordeaux); Marie-Noëlle-Cuffi, Corinne Costes (Castres); Olivier Rouaud, Patrick Manckoundia, Valérie Quipourt, Sophie Marilier, Evelyne Franon (Dijon); Lawrence Bories, Marie-Laure Pader, Marie-France Basset, Bruno Lapoujade, Valérie Faure, Michael Li Yung Tong, Christine Malick-Loiseau, Evelyne Cazaban-Campistron (Foix); Françoise Desclaux, Colette Blatge (Lavaur); Thierry Dantoine, Cécile Laubarie-Mouret, Isabelle Saulnier, Jean-Pierre Clément, Marie-Agnès Picat, Laurence Bernard-Bourzeix, Stéphanie Willebois, Iléana Désormais, Noëlle Cardinaud (Limoges); Marc Bonnefoy, Pierre Livet, Pascale Rebaudet, Claire Gédéon, Catherine Burdet, Flavien Terracol (Lyon), Alain Pesce, Stéphanie Roth, Sylvie Chaillou, Sandrine Louchart (Monaco); Kristelle Sudres, Nicolas Lebrun, Nadège Barro-Belaygues (Montauban); Jacques Touchon, Karim Bennys, Audrey Gabelle, Aurélia Romano, Lynda Touati, Cécilia Marelli, Cécile Pays (Montpellier); Philippe Robert, Franck Le Duff, Claire Gervais, Sébastien Gonfrier (Nice); Yannick Gasnier and Serge Bordes, Danièle Begorre, Christian Carpuat, Khaled Khales, Jean-François Lefebvre, Samira Misbah El Idrissi, Pierre Skolil, Jean-Pierre Salles (Tarbes). MRI group: Carole Dufouil (Bordeaux), Stéphane Lehéricy, Marie Chupin, Jean-François Mangin, Ali Bouhayia (Paris); Michèle Allard (Bordeaux); Frédéric Ricolfi (Dijon); Dominique Dubois (Foix); Marie Paule Bonceour Martel (Limoges); François Cotton (Lyon); Alain Bonafé (Montpellier); Stéphane Chanalet (Nice); Françoise Hugon (Tarbes); Fabrice Bonneville, Christophe Cognard, François Chollet (Toulouse). PET scans group: Pierre Payoux, Thierry Voisin, Julien Delrieu, Sophie Peiffer, Anne Hitzel, (Toulouse); Michèle Allard (Bordeaux); Michel Zanca (Montpellier); Jacques Monteil (Limoges); Jacques Darcourt (Nice). Medico-economics group: Laurent Molinier, Hélène Derumeaux, Nadège Costa (Toulouse). Biological sample collection: Bertrand Perret, Claire Vinel, Sylvie Caspar-Bauguil (Toulouse). Safety management: Pascale Olivier-Abbal. DSA Group: Sandrine Andrieu, Christelle Cantet, Nicola Coley.

Acknowledgments: The present study called ECO-MAPT study was supported by grants from the French Ministry of Health (PHRC 2008) and the France Alzheimer Association (Doctoral Scholarship 2010). The MAPT study was supported by grants from the Gérontopôle of Toulouse, the French Ministry of Health (PHRC 2008, 2009), the Pierre Fabre Research Institute (manufacturer of the polyunsaturated fatty acid supplement), Exhonit Therapeutics, and Avid Radiopharmaceuticals. No sponsor placed any restriction on this study or had any role in the design of the study, data collection, data analyses or interpretation, or in the preparation, review, or approval of the manuscript. The promotion of this study was supported by the University Hospital Center of Toulouse. We are indebted to the investigators from CHU de Toulouse, Centre Hospitalier Lyon-Sud, Hôpital de Tarbes, Hôpital de Foix, Hôpital de Castres, CHU de Limoges, CHU de Bordeaux, Hôpital de Lavaur, CHU de Montpellier, Hôpital Princesse Grace, Hôpital de Montauban, CHU de Nice, and CHU de Dijon for their participation in this study.

Funding: The present study called ECO-MAPT study was supported by grants from the French Ministry of Health (PHRC 2008) and the France Alzheimer Association (Doctoral Scholarship 2010). The original MAPT study was supported by grants from the Toulouse Geriatric Center, the French Ministry of Health (PHRC 2008, 2009), Pierre Fabre Research Institute (manufacturer of the omega-3 supplement), ExonHit Therapeutics SA, and Avid Radiopharmaceuticals Inc. Trial registration number: NCT00672685, first registered in May 6, 2008

All authors meet criteria for authorship as stated in the COI form, as well as their contributions to the manuscript. All authors’ specific areas of contributions is listed, using categories below: – Study concept and design: Nadège Costa, Hélène Derumeaux, Sophie Guyonnet, Isabelle Carrié, Sandrine Andrieu, Laurent Molinier; – Acquisition of data: Nadège Costa, Hélène Derumeaux, Sophie Guyonnet, Isabelle Carrié, Sandrine Andrieu, Laurent Molinier; – Analysis and interpretation of data: Nadège Costa, Michael Mounié, Arnaud Pagès, Hélène Derumeaux, Nicola Coley, Chrsitelle Cantet, Sandrine Andrieu, Laurent Molinier; – Drafting of the manuscript: Nadège Costa, Michael Mounié, Arnaud Pagès, Hélène Derumeaux, Laurent Molinier; – Critical revision of the manuscript for important intellectual content: Thomas Rapp, Nicola Coley, Sandrine Andrieu.

Ethics approval and consent to participate: The investigating physicians, who verified inclusion and exclusion criteria and obtained written informed consent, recruited all participants. The trial protocol was approved by the French Ethical Committee located in Toulouse (CPP SOOM II) and was authorised by the French Health Authority.

Conflict of interest: Authors report no conflict of interest.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, 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.









1. International Alzheimer’s Disease. World Alzheimer Report 2019: Attitudes to dementia. Accessed on March 13, 2020.
2. Rapp T, Andrieu S, Chartier F, et al. Resource Use and Cost of Alzheimer’s Disease in France: 18-Month Results from the GERAS Observational. Value in Health 2018;21(3):295-303.
3. Bond M, Rogers G, Peters J, et al. The effectiveness and cost-effectiveness of donepezil, galantamine, rivastigmine and memantine for the treatment of Alzheimer’s disease (review of Technology Appraisal No. 111): a systematic review and economic model. Health Technol Assess Winch Engl 2012;25(1):470.
4. Birks JS, Harvey RJ. Donepezil for dementia due to Alzheimer’s disease. Cochrane Database Syst Rev 2018;6:CD001190.
5. French National Authority for Health (HAS) – Transparency Committee. Reassessment of the Medical Benefit of acetylcholinesterase inhibitors and non-competitive NMDA-receptor antagonist in the treatment of Alzheimer’s disease. s.l. : French National Authority for Health, 2016.
6. Pouryamout L, Dams J, Wasem J, et al. Economic evaluation of treatment options in patients with Alzheimer’s disease: a systematic review of cost-effectiveness analyses. Drugs 2012;72:789-802.
7. Andrieu S, Coley N, Lovestone S, et al. Prevention of sporadic Alzheimer’s disease: lessons learned from clinical trials and future directions. Lancet Neurol 2015;14:926-44.
8. Andrieu S, Guyonnet S, Coley N, et al. Effect of long term omega 3 polyinsaturated fatty acid supplementation with or without multidomain intervention on cognitive function in elderly adults with memory complaints (MAPT): a randomised, placebo-controlled trial. Lancet Neurol 2017;16(5):377-389.
9. Handels R, Wimo A. Challenges and recommendations for the health-economic evaluation of primary prevention programmes for dementia. Aging Ment Health 2019;23(1):53-59.
10. Baal PHM van, Hoogendoorn M, Fischer A. Preventing dementia by promoting physical activity and the long-term impact on health and social care expenditures. Prev Med 2016;85:78-83.
11. Lin P-J, Yang Z, Fillit HM, et al. Unintended Benefits: The Potential Economic Impact Of Addressing Risk Factors To Prevent Alzheimer’s Disease. Health Aff (Millwood) 2014;33(4):547-554.
12. Tsiachristas A, Smith AD. B-vitamins are potentially a cost-effective population health strategy to tackle dementia: Too good to be true? Alzheimers Dement Transl Res Clin Interv 2016;2(3):156-161.
13. Zhang Y, Kivipelto M, Solomon A, Wimo A. Cost-Effectiveness of a Health Intervention Program with Risk Reductions for Getting Demented: Results of a Markov Model in a Swedish/Finnish Setting. J Alzheimers Dis 2011 ;26(4) :735-744.
14. Vellas B, Carrie I, Gillette-Guyonnet S, et al. Mapt study: a multidomain approach for preventing Alzheimer’s disease: design and baseline data. J Prevent Alzheimer 2014;(1):13-22.
15. Boudemaghe T, Belhadj I . Data Resource Profile: The French National Uniform Hospital Discharge Data Set Database (PMSI). International Journal of Epidemiology 2017 ;46(2) :392-392d.
16. French Health Insurance (Assurance Maladie). The general nomenclature of professional acts (La nomenclature générale des actes professionnels (NGAP)). Accessed on January 17, 2019.
17. French Health Insurance (Assurance Maladie). Common Classification of Medical Acts (Classification Commune des Actes Médicaux (CCAM)). 3A%221.1%22%3B%7D&add=1.1.13#chapitre_1.1.13.Ameli. Accessed on January 17, 2019.
18. French Health Insurance (Assurance Maladie). National Biology Coding Table (Table Nationale de codage de Biologie). Accessed on January 17, 2019.
19. Bezin J , Duong M, Lassalle R, et al. The national healthcare system claims databases in France, SNIIRAM and EGB: Powerful tools for pharmacoepidemiology. Pharmacoepidemiol Drug Saf 2017;26(8):954-962.
20. Moulis G, Lapeyre-Mestre M, Palmaro A, et al. French health insurance databases: What interest for medical research? Rev Med Interne 2015;36(6):411-417.
21. World Health Organisation (WHO). Collaborating Centre for Drug Statistics Methodology. Anatomical Therapeutic Chemical (ATC) classification system and the Defined Daily Dose (DDD) [Internet]. World Health Organisation (WHO). Accessed on June 28, 2020.
22. European Medicine Agency. Committee for Medicinal Products for Human Use (CHMP). Accessed June 28, 2020.
23. French Health Insurance (Assurance Maladie). Drug database and pricing information (Base des médicaments et informations tarifaires). presentation.php?p_site=AMELI. Accessed on June 10, 2020.
24. Gridchyna I, Aulois-Griot M, Maurain C, Bégaud B. How innovative are pharmaceutical innovations?: the case of medicines financed through add-on payments outside of the French DRG-based hospital payment system. Health Policy 2012;104(1):69-75.
25. Coley N, Gallini A, Ousset JP, et al. Evaluating the Clinical Relevance of a Cognitive Composite Outcome Measure: An Analysis of 1414 Participants From the 5-year GuidAge Alzheimer’s Prevention Trial. Alzheimers Dement 2016;12(12):1216-1225.
26. Burton A, Billingham LJ, Bryan S. Cost-effectiveness in clinical trials: using multiple imputation to deal with incomplete cost data. Clinical Trials 2007;4(2):154-161.
27. Contoyannis P, Jones A, Rice N. The Dynamics of Health in the British Household Panel Survey. Journal of Applied Econometrics 2004;19:473-503.
28. Nixon RM, Thompson SG. Parametric modelling of cost data in medical studies. Statist Med. 2004, Vol. 23, pp. 1311-1331.
29. Briggs AH, Gray AM. Handling uncertainty when performing economic evaluation of healthcare interventions. Health Technol Assess 1999;3(2):1-134.
30. Briggs AH, Wonderling DE, Mooney CZ. Pulling cost-effectiveness analysis up by its bootstraps: a non-parametric approach to confidence interval estimation. Health Econ 1997;6(4):327-340.
31. Anderson JL, Heidenreich PA, Barnett PG, et al. ACC/AHA statement on cost/value methodology in clinical practice guidelines and performance measures: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures and Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63(21):2304-2322.
32. Claxton K, Martin S, Soares M, et al. Methods for the estimation of the National Institute for Health and Care Excellence cost-effectiveness threshold. Health Technol Assess 2015 ;19(14):1-503.
33. French National Authority for Health (HAS). Reference values for health economic evaluation (Valeurs de réference pour l’évaluation économique en santé). Paris : s.n., 2014. Accessed on July 23, 2020.
34. Fenwick E, Byford S. A guide to cost-effectiveness acceptability curves. Br J Psychiatry 2005;187:106-8.
35. lo Storto C, Goncharuk AG. Efficiency vs Effectiveness: a Benchmarking Study on European Healthcare Systems. Economics and Sociology 2017;10(3):102-115.
36. Gotaro Kojima. Increased healthcare costs associated with frailty among community-dwelling older people: A systematic review and meta-analysis. Arch Gerontol Geriatr 2019:82:103898.
37. Sirven N, Rapp T. The cost of frailty in France. Eur J Health Econ 2017;18(2):243-253.
38. Huter K, Kocot E,Kissimova-Skarbek K,Dubas-Jakóbczyk K, Rothgang H. Economic evaluation of health promotion for older people-methodological problems and challenges. BMC Health Serv Res 2016;16(S5):328.
39. Knapp M, Mangalore R. The trouble with QALYs…. Epidemiol Psichiatr Soc 2007 ;16 : 289-293.
40. Dennis A Revicki. Measuring health outcomes for cost-effectiveness studies: Are all quality adjusted life years created equal? Drug information journal 1995;29:1459-1467.
41. Costa N, Ferlicoq L, Derumeaux-Burel H, Rapp T, Garnault V, Gillette-Guyonnet S, Andrieu S, Vellas B, Lamure M, Grand A, Molinier L. Comparison of informal care time and costs in different age-related dementias: a review. Biomed Res Int. 2013;2013:852368. doi: 10.1155/2013/852368. Epub 2012 Dec 5. PMID: 23509789; PMCID: PMC3591240.
42. Paraponaris A, Davin B. Economics of the Iceberg: Informal Care Provided to French Elderly with Dementia. Value Health 2015;18(4):368-375.
43. Doubilet P, Begg CB, Weinstein M.C, Braun P, McNeil BJ. Probabilistic Sensitivity Analysis Using Monte Carlo Simulation: A Practical Approach. Medical Decision Making 1985;5(2):157-177.
44. Baltussen RM, Hutubessy RC, Evans DB, Murray CJ . Uncertainty in cost-effectiveness analysis. Probabilistic uncertainty analysis and stochastic league tables. Int J Technol Assess Health Care 2002;18(1):112-119.
45. Adalsteinsson E, Toumi M. Benefits of probabilistic sensitivity analysis – a review of NICE decisions. J Mark Access Health Policy 2013;1:10.
46. Claxton K, Sculpher M, McCabe C, et al. Probabilistic sensitivity analysis for NICE technology assessment: Not an optional extra. Health Econ 2005;14(4):339-349.