A Phase IIa Randomized Control Trial of VEL015 (Sodium Selenate) in Mild-Moderate Alzheimer’s Disease

Abstract

Background: There is increasing interest in targeting hyperphosphorylated tau (h-tau) as a disease modifying approach for Alzheimer’s disease (AD). Sodium selenate directly stimulates the activity of PP2A, the main enzyme responsible for h-tau dephosphorylation in the brain.

Objective: This study assessed the safety and tolerability of 24-week treatment with VEL015 (sodium selenate) in AD. Investigating the effects of VEL015 on cognitive, CSF, and neuroimaging biomarkers of AD were secondary, exploratory objectives. Data were used to identify biomarkers showing most promise for use in subsequent efficacy trials.

Methods: A 24-week, multicenter, Phase IIa, double-blinded randomized controlled trial. Forty patients aged ≥55 y with mild-moderate AD (MMSE 14–26) were randomized to supranutritional (VEL015 10 mg tds [n = 20]) and control (VEL015 320μg tds [n = 10] or placebo [n = 10]) groups. Patients were regularly monitored for safety, adverse events (AEs), and protocol compliance. Exploratory biomarkers included cognitive tests, neuroimaging (diffusion MR), and CSF (p-tau, t-tau, and Aβ_1-42).

Results: Thirty-six (90%; [supranutritional n = 18, control/placebo n = 18]) patients completed the trial. There were no differences in the incidence of specific AEs between groups. Only one secondary biomarker, diffusion MR measures, showed group differences, with less deterioration in the supranutritional group (p < 0.05).

Conclusion: Treatment with VEL015 at doses up to 30 mg per day for 24 weeks was safe and well-tolerated in patients with AD. Diffusion MR measures appear to be the most sensitive biomarkers to assess disease progression over 24 weeks.

Keywords

Alzheimer’s disease clinical trial dementia sodium selenate

INTRODUCTION

Hyperphosphorylated tau (h-tau) is a potential target for disease-modifying therapies in Alzheimer’s disease (AD) [1]. One strategy involves the upregulation of PP2A, the major tau serine/threonine phosphatase in the brain, which has been implicated in the pathogenesis of AD [2, 3]. Sodium selenate (VEL015) stimulates the activity of the PP2A/PR55 heterotrimer consisting of the PR55 regulatory B-subunit, the form of PP2A directly associated with dephosphorylation of tau in the human brain [4]. High-dose VEL015 in animal models upregulates PP2A/PR55, reduces levels of total tau (t-tau) and phospho-tau (p-tau), and reverses memory deficits [4, 5]. Similar effects have been demonstrated in models of other tau-related conditions, including traumatic brain injury and epilepsy [6, 7].

No studies of VEL015 have been conducted in AD patients; however, a Phase I study in patients with prostate cancer demonstrated a favorable safety profile [5]. We report a Phase IIa double-blinded randomized controlled trial of a 24-week treatment with VEL015 in humans with mild-moderate AD. Given that selenate may be obtained nutritionally from a number of sources in very low concentrations, a supranutritional dose was compared to a combined nutritional dose and placebo. The primary objective was to assess the safety and tolerability of a supranutritional dose of VEL015. The exploratory objectives were to evaluate the effect of VEL015 on a range of cognitive, cerebrospinal fluid (CSF), and imaging biomarkers of AD disease progression, to identify those that show most promise for use as endpoints in subsequent efficacy.

MATERIALS AND METHODS

Participants

This was a double-blind, randomized, placebo-controlled, parallel group study conducted at four centers in Melbourne, Australia. The study was conducted between December 2011 (first screening visit) and June 2013 (last follow-up visit). Recruitment was conducted through four centers in Melbourne, Australia. Eligible patients were community dwelling, aged ≥55 y, and satisfied the following inclusion criteria: a diagnosis of probable AD according to NIA criteria [8]; a ‘mild’ to ‘moderate’ degree of dementia as defined by a Mini-Mental Status Examination (MMSE) of between 14 and 26 at screening; a modified Hachinski score ≤4; treatment with an acetylcholine esterase inhibitor at the time of the screening visit (receiving a stable dose for at least 4 months); and documented MRI scan performed within 14 days of baseline showing no gross structural abnormality indicative of a neurological disorder other than AD.

Exclusion criteria included: contraindication for lumbar puncture; history of alcohol and/or other substance abuse; known sensitivity to selenium or sodium selenate (or similar agent); any dementia syndrome other than AD, or evidence of current neurological, psychiatric, or other illness that could contribute to non-Alzheimer’s dementia; known history of familial AD; significant medical disease that is not adequately controlled; history of epilepsy; diabetes; impaired renal, hepatic, or hematological function; current or recent (within 6 weeks of screening) treatment with lithium, NMDA receptor antagonists, steroids, or injectable non-steroidal anti-inflammatory drugs; dietary supplements containing >26μg selenium per day; and current treatment with carbamazepine, digoxin, phenobarbitone, phenytoin, or warfarin.

Procedures

The study was approved by the local institutional ethics committee (Melbourne Health, Melbourne, Victoria, Australia) and registered with the Australian and New Zealand Clinical Trials Registry (ID: ACTRN12611001200976). An independent data and safety monitoring board (DSMB) was established to oversee the safety of the study. Written informed consent was obtained from the participant or legally authorized representative and the participant’s caregiver. The duration of the study was up to 33 weeks, consisting of a screening period of 4 weeks, 24 weeks of treatment, and a 5-week post-treatment follow-up period. During screening, the diagnosis was confirmed by an experienced clinical neurologist or neuropsychiatrist, which included a medical and cognitive history, MMSE, general physical examination, neurological examination, MRI, 12-lead ECG and hematology, biochemistry, and urine analyses.

Randomization and masking

Following screening, 40 eligible patients were randomized in a 2:1:1 ratio to receive supranutritional-dose VEL015 (10 mg tds), nutritional-dose VEL015 (320μg tds), or placebo. This sample size was determined based on the early stage of investigation (i.e., Stage IIa). A computer generated randomization list was used. All investigators, participants, and caregivers remained blinded to randomization status until the conclusion of the trial. The DSMB conducted an interim safety analysis after the first 12 patients had been treated for 8 weeks. As a favorable safety profile was demonstrated, enrollment continued.

Primary outcomes

Safety parameters included adverse events (unsolicited, and solicited via diary cards), vital signs, physical examination, neurological examination, laboratory evaluations (hematology, biochemistry, and urine analyses), and ECG. Vital signs, laboratory evaluations and ECG examination were performed at baseline (pre-treatment), and weeks 4, 8, 16, 24, and 28 (post-treatment). Adverse events (AEs) were defined as any untoward medical occurrence that was not necessarily causally related to the treatment. Serious adverse events (SAEs) were those that resulted in death, were life threatening, required or prolonged hospitalization, or resulted in persistent or significant disability.

Exploratory outcomes

Exploratory variables included CSF biomarker levels, cognitive test scores, and neuroimaging metrics.

CSF

CSF biomarkers were sampled via lumbar-puncture following an overnight fast undertaken at baseline and at 24 weeks of treatment, and included p-tau, t-tau, and Aβ_1-42. Assay of biomarkers from CSF samples were performed using validated enzyme-linked immunosorbent assay methods (specifically Innogenetics’ and Meso Scale Discovery’s commercially available kits: Innotest β-amyloid(1-42), Innotest hTau Ag, Innotest phosphor-tau, and/or MSD total tau and MSD phospho-tau).

Cognitive measures

Cognitive assessment scores included the total scores from the Alzheimer’s Disease Assessment Scale cognitive subscale (ADAS-Cog); MMSE; Controlled Oral Word Association Test (COWAT); and the Category Fluency Test (CFT) [9]. These tests were administered at baseline and week 24. Three tests were administered from the CogState computerized battery (CogState Ltd.): the one-card learning memory task (OCL), identification reaction time task (IDN), and the detection reaction time task (DET).

Structural MRI

Volumetric MRI was acquired at baseline and 24 weeks for all participants on the same 3 Tesla Siemens Trio Tim MRI scanner at the Royal Melbourne Hospital, Australia (MPRAGE, flip angle = 9°, TR = 1900 ms, RE = 2.13 ms, TI = 900 ms, FOV = 256 mm, matrix = 256×256, number of slices = 176, slice thickness = 1 mm). Hippocampal volumes were quantified by manual tracing using a predefined, validated protocol by a blinded single operator [10]. Intra-rater reliability was assessed via blind-retracing of 10 volumes and was excellent (Spearman’s rho = 0.99). Mean bilateral volume was computed. Bilateral entorhinal cortex thickness was automatically extracted from cortical models generated using the FreeSurfer software longitudinal pipeline [11].

Diffusion-weighted MRI

Advanced diffusion-weighted MR imaging was acquired for each participant at each time-point (baseline and 24 weeks) on the same scanner (TR/TE = 8700/92 ms, FOV = 240 mm, matrix 96×96, b = 1000 s/mm², voxel size 2.5×2.5×2.5 mm, 30 directions). The diffusion data were pre-processed using the FSL software [12] by skull stripping the raw diffusion images and fitting the diffusion tensor with weighted least squares. The DTI-TK software was then used to create a population-specific template and all of the participants’ diffusion tensor maps were registered to the template with affine and diffeomorphic alignments [13].

PET

Fluorodeoxyglucose (FDG) PET images were acquired at baseline and week 24 on a GE Discovery 690 (GE Medical Systems Milwaukee, WI). Fasted (≥6 hours) patients were injected with FDG (220 MBq) through an intravenous catheter and rested in a darkened quiet room for 45 min. Patients underwent a 15-min list mode brain PET acquisition. A 5×3-min dynamic series was reconstructed to qualitatively determine patient movement during the acquisition. All scans were processed using OSEM3D iterative reconstruction (8 iterations, 24 subsets, 5 mm filter, 192×192 matrix, and a 35 cm field of view). Images were then normalized to MNI space and five regions of interest were extracted (posterior cingulate cortex, left and right angular gyrus, and left and right inferior temporal gyrus) based on the Alzheimer’s Disease Neuroimaging Initiative meta-ROIs [14]. Mean uptake in these regions was computed and normalized to the mean uptake of the pons and cerebellar vermis.

Statistical analyses

All statistical analyses were conducted on the intention to treat (ITT) population. Data were included for all participants enrolled in the study who had complete data for the relevant analysis. For all outcome variables, the supranutritional group was compared to the control group (combined nutritional dose and placebo). Separate analyses were conducted comparing all three groups. The pattern of results was comparable to the analyses presented here.

For exploratory outcome analyses the absolute change from baseline for the variables was calculated and compared via a planned linear contrast. Standardized effect sizes with confidence intervals were computed for each comparison in the form of Bonett’s d [15]. Confidence intervals were calculated to indicate precision of the parameter estimate as well as statistical significance (95%). For interpretability, contrasts weights were specified such that a positive treatment outcome (i.e., effect contrary to the expected course of the disease) was always represented by a positive effect size. These effect sizes were used to calculate the sample sizes required to replicate the study with 80% power at the p < 0.05 level (two-tailed).

For the diffusion imaging outcomes, voxel-wise analysis of the diffusion data (fractional anisotropy [FA], mean diffusivity [MD], axial diffusivity [A_xD], and radial diffusivity [RD]) was carried out using the FSL Tract-Based Spatial Statistics software [16]. For each participant, a difference image was generated for each diffusion parameter representing absolute change over time. Between-groups statistics were calculated using a permutation-based approach with 10,000 permutations. Threshold-free cluster enhancement (TFCE) with 2D optimization was applied and the resulting statistical maps thresholded at p < 0·05 (Bonferroni corrected) [17].

RESULTS

Cohort

Forty patients were randomly assigned to either the supranutritional (n = 20) or the control group (nutritional dose, n = 10; placebo, n = 10). The groups did not differ on demographics or baseline secondary outcome variables (Table 1). Four participants withdrew prior to completion of the 28-week study period. The first (supranutritional group) was withdrawn after developing a skin rash. The second (supranutritional group) withdrew from the study due to nail changes. At the time of withdrawal, the participant completed the full battery of assessments scheduled for the 24-week visit, and so the subject was included in this ITT analysis. The third (control group) withdrew due to violation of inclusion criteria (diagnosis of normal pressure hydrocephalus). The fourth participant (control group) was withdrawn following commencement of a protocol-prohibited medication (NMDA antagonist). The final ITT analysis was performed on 37 participants. Full participant flow is shown in Fig. 1.

Safety and tolerability

36 patients (90%) experienced at least one treatment emergent adverse event (TEAE): 19 (95%) in the supranutritional group and 17 (85%) in the control group. Of these, 27 (68%) patients experienced at least one drug-related TEAE: 16 (80%) from the supranutritional group and 11 (50%) from the control group. Most reported adverse events were rated as mild, and did not affect the patient’s willingness to continue in the trial. One SAE of pre-syncope was reported by a participant in the supranutritional group. The SAE was mild and possibly related to the study drug, and resolved within 24 h of onset. As outlined above, two patients (both from the supranutritional group) withdrew due to TEAEs that were considered causally related to the study drug. The breakdown of TEAEs by type is shown in Table 2.

Exploratory outcomes

Absolute change over the 24-week treatment period for CSF biomarkers, neuroimaging metrics, and cognition are shown in Table 3. Figure 2 shows the standardized effect sizes for this comparison for all outcome variables. No significant differences between the supranutritional and control groups were observed for the CSF, cognition, or neuroimaging biomarkers except for the diffusion imaging measures.

On diffusion imaging, regional analysis revealed several clusters where significantly greater increase was observed in the control group compared to the supranutritional group (Fig. 3). The difference between the supranutritional and control groups in MD change was d = 1.95 (1.11, 2.79), for A_xD was d = 1.57 (0.79, 2.35), and for RD was d = 1.92 (1.04, 2.79). The locations of these clusters are included in Table 4.

Based on the observed effect sizes, power calculations for a subsequent study to demonstrate an efficacy effect of VEL015 treatment using each of the CSF, imaging, and cognitive biomarkers assessed in this study are shown in Table 5.

DISCUSSION

This Phase IIa double-blinded RCT study explored the use of VEL015 as a novel therapy for the treatment of mild-moderate AD. The primary objective of the study was to assess the safety and tolerability of 24 weeks of treatment with VEL015 at a supranutritional dose, compared with a nutritional dose or placebo. The results demonstrated that the treatment was generally safe and well-tolerated with similar rates of TEAEs reported in each group. The most common adverse events were headache, lethargy, fatigue, and nausea occurring in >25% of patients in the supranutritional group. Ninety percent of patients completed the 24-week treatment phase. Two patients randomized to the supranutritional group withdrew from the study experiencing AEs that were not typical of the entire cohort. Only one SAE was reported by one patient in the supranutritional group. The SAE of pre-syncope was judged to be possibly related to the study drug.

The safety and tolerability findings of this current study are consistent with those of an earlier open-label, Phase I dose-escalation study of VEL015 in 19 patients with castration-resistant prostate cancer [5]. In that study treatment with VEL015 was found to be safe and well-tolerated at doses of 5 to 60 mg per day, with the most frequently reported TEAEs being nausea, diarrhea, fatigue, muscle spasms, alopecia, and nail disorders, most frequent in the higher dose group.

The exploratory objectives of this trial were to study the effects of VEL015 on a range of cognitive, CSF, and imaging biomarkers of AD progression. Of the biomarkers assessed, the diffusion MR measures appear the most promising. These measures demonstrated significantly greater increase (i.e., worsening) in the control group for three key diffusivity variables: MD, A_xD, and RD. These increases were possibly indicative of deterioration in axonal structural integrity [18], and were widespread, being most prominent in the corpus callosum and cingulate regions. In contrast, the increase in diffusivity markers was significantly lower in the subjects in the supranutritional group. In these significant regions, A_xD and RD significantly decreased (i.e., improved) in the supranutritional group— possibly indicating some repair in cellular membrane integrity over the treatment period.

The diffusion results in the control group supports the utility of diffusion imaging for measuring disease progression in longitudinal studies of AD. Studies in animal models of AD have found a relationship between these diffusion MR measures and axonal damage with changes in A_xD related to axonal damage and RD with demyelination [19, 20]. The results of the present study suggest that VEL015 might have positively affected neuronal or axonal integrity. Serial diffusion MRI is a promising approach for future studies to establish definitively whether VEL015 has a disease-modifying effect in AD.

Hippocampal volume measurements and FDG-PET relative uptake measures showed small, but significant, decreases on the serial assessments in the control group. In addition, there was a trend for a decrease in the entorhinal cortex thickness in these patients. In contrast to the encouraging diffusion MRI results, annualized hippocampal volume atrophy rates were 4.3% for the supranutritional group and 3.0% for the control group, both of which are in keeping with previous longitudinal studies [21]. The annualized decrease in entorhinal cortex thickness of 1% in the control group and 3.9% for the supranutritional group found in this study were somewhat less than reported in previous work [22].

Cognitive and CSF biomarker measures of tau, p-tau, and Aβ_1-42 assessed in this study showed no significant changes over the 24-week treatment period between groups. The cognition results are similar to those reported previously in a Phase IIa study investigating PBT2 as a disease-modifying treatment in patients with mild-moderate AD, which also found no group differences in scores on ADAS-Cog or MMSE [23]. More sensitive cognitive testing of reaction time and memory functions failed to detect a significant difference between the groups, and similarly there was no strong evidence for changes in measures of executive function.

CSF levels of p-tau and t-tau trended to increase less in the supranutritional group over the treatment period. This trend could be consistent with the proposed effect of VEL015 to increase p-tau clearance; however, in an animal AD model, p-tau and t-tau were decreased following treatment with sodium selenate [5]. It is possible that a longer treatment window may be required to observe a significant treatment effect on CSF biomarkers in humans with AD.

One of the secondary aims of this study was to generate data that could be used to estimate sample size requirements for future studies with VEL015. The most powerful measures were found to be those of diffusion MR, with only 12–16 participants being estimated to be required to demonstrate efficacy of VEL015. Of the other neuroimaging measures assessed, entorhinal cortex thickness was the most powerful, with an estimated 246 subjects being required. Of the CSF biomarkers, Aβ_1-42 and t-tau were the most powerful, requiring between 232 and 252 patients.

Limitations of this study include a relatively short study duration compared to other therapeutic trials in this population. Given the rate of disease progression, 18 months has been suggested as the optimal treatment duration in disease-modifying therapies for AD [24]. In addition, inclusion of participants with moderately severe AD coupled to the small number of persons in the trial are likely to have militated against observing benefits in exploratory outcomes of cognition and other biomarkers. Acknowledging, the doses used in a previous Phase 1 study of sodium selenate in prostate cancer [5], a higher dose might be investigated in future work.

In conclusion, this Phase IIa RCT provides Level 1 evidence for the safety and tolerability of VEL015 taken at a dose of 30 mg per day for 24 weeks in patients with mild-moderate AD. Diffusion MR measures (MD, A_xD, and RD) appear to be the most sensitive of a range of potential neurocognitive, neuroimaging, and CSF biomarkers assessed, to evaluate disease progression over 24 weeks, with evidence of a potential protective effect of VEL015 treatment. These findings can inform future trials designed to definitively obtain evidence of whether VEL015 is an effective disease-modifying treatment for patients with AD, as well as potentially other neurodegenerative diseases characterized by h-tau.

Footnotes

ACKNOWLEDGMENTS

The authors would like to thank the patients and their caregivers for participating in this study. We also thank the research staff at the various clinical sites: Anthony Ang, Jennifer Bortoli, Darren Germaine, Christopher Godden, Jack Germaine, Sean Hosking, Lucas Litewka, Elaine Lui, David Murphy, and Zofia Ross. Charles Malpas is the recipient of an Alzheimer’s Australia Viertel Foundation postgraduate research scholarship.

This study was funded by Velacor Therapeutics. The data were analyzed, and manuscript written, independently by the authors.

Authors’ disclosures available online ().

References

Gong

, Iqbal

(2008) Hyperphosphorylation of microtubule-associated protein tau: A promising therapeutic target for Alzheimer disease, Curr Med Chem 15, 2321.

Goedert

, Jakes

, Qi

, Wang

, Cohen

(1995) Protein phosphatase 2A is the major enzyme in brain that dephosphorylates τ protein phosphorylated by proline-directed protein kinases or cyclic AMP-dependent protein kinase, J Neurochem 65, 2804–2807.

Gong

C-X

, Lidsky

, Wegiel

, Zuck

, Grundke-Iqbal

, Iqbal

(2000) Phosphorylation of microtubule-associated protein tau is regulated by protein phosphatase 2A in mammalian brain implications for neurofibrillary degeneration in Alzheimer’s disease, J Biol Chem 275, 5535–5544.

van Eersel

, Ke

, Liu

, Delerue

, Kril

, Götz

, Ittner

(2010) Sodium selenate mitigates tau pathology, neurodegeneration, and functional deficits in Alzheimer’s disease models, Proc Natl Acad Sci U S A 107, 13888–13893.

Corcoran

, Hovens

, Michael

, Rosenthal

, Costello

(2010) Open-label, phase I dose-escalation study of sodium selenate, a novel activator of PP2A, in patients with castration-resistant prostate cancer, Br J Cancer 103, 462–468.

Jones

, Nguyen

, Corcoran

, Velakoulis

, Chen

, Grundy

, O’Brien

, Hovens

(2012) Targeting hyperphosphorylated tau with sodium selenate suppresses seizures in rodent models, Neurobiol Dis 45, 897–901.

Shultz

, Wright

, Zheng

, Stuchbery

, Liu

, Sashindranath

, Medcalf

, Johnston

, Hovens

, Jones

, O’Brien

(2015) Sodium selenate reduces hyperphosphorylated tau and improves outcomes after traumatic brain injury, Brain 138, 1297–1313.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease, Neurology 34, 939–939.

Lezak

, Howieson

, Loring

, Hannay

, Fischer

(2004) Neuropsychological Assessment, Oxford University Press, Oxford.

10.

Cook

, Fish

, Shorvon

, Straughan

, Stevens

(1992) Hippocampal volumetric and morphometric studies in frontal and temporal lobe epilepsy, Brain 115, 1001–1015.

11.

Reuter

, Fischl

(2011) Avoiding asymmetry-induced bias in longitudinal image processing, Neuroimage 57, 19–21.

12.

Jenkinson

, Beckmann

, Behrens

, Woolrich

, Smith

(2012) FSL, Neuroimage 62, 782–790.

13.

Bach

, Laun

, Leemans

, Tax

, Biessels

, Stieltjes

, Maier-Hein

(2014) Methodological considerations on Tract-Based Spatial Statistics (TBSS), Neuroimage 100, 358–369.

14.

Landau

, Harvey

, Madison

, Koeppe

, Reiman

, Foster

, Weiner

, Jagust

(2011) Associations between cognitive, functional, and FDG-PET measures of decline in AD and MCI, Neurobiol Aging 32, 1207–1218.

15.

Bonett

(2008) Confidence intervals for standardized linear contrasts of means, Psychol Method 13, 99.

16.

Smith

, Jenkinson

, Johansen-Berg

, Rueckert

, Nichols

, Mackay

, Watkins

, Ciccarelli

, Cader

, Matthews

(2006) Tract-based spatial statistics: Voxelwise analysis of multi-subject diffusion data, Neuroimage 31, 1487–1505.

17.

Smith

, Nichols

(2009) Threshold-free cluster enhancement: Addressing problems of smoothing, threshold dependence and localisation in cluster inference, Neuroimage 44, 83–98.

18.

Mori

, Zhang

(2006) Principles of diffusion tensor imaging and its applications to basic neuroscience research, Neuron 51, 527–539.

19.

Song

, Sun

, Ju

, Lin

, Cross

, Neufeld

(2003) Diffusion tensor imaging detects and differentiates axon and myelin degeneration in mouse optic nerve after retinal ischemia, Neuroimage 20, 1714–1722.

20.

Sun

, Liang

, Trinkaus

, Cross

, Armstrong

, Song

(2006) Noninvasive detection of cuprizone induced axonal damage and demyelination in the mouse corpus callosum, Magn Reson Med 55, 302–308.

21.

Dickerson

, Sperling

(2005) Neuroimaging biomarkers for clinical trials of disease-modifying therapies in Alzheimer’s disease, NeuroRx 2, 348–360.

22.

, Schuff

, Kramer

, Ganzer

, Zhu

, Jagust

, Miller

, Reed

, Mungas

, Yaffe

(2004) Higher atrophy rate of entorhinal cortex than hippocampus in AD, Neurology 62, 422–427.

23.

Lannfelt

, Blennow

, Zetterberg

, Batsman

, Ames

, Harrison

, Masters

, Targum

, Bush

, Murdoch

, Wilson

, Ritchie

, PBT2-201-EURO study group (2008) Safety, efficacy, and biomarker findings of PBT2 in targeting Aβ as a modifying therapy for Alzheimer’s disease: A phase IIa, double-blind, randomised, placebo-controlled trial, Lancet Neurol 7, 779–786.

24.

Vellas

, Andrieu

, Sampaio

, Wilcock

(2007) Disease-modifying trials in Alzheimer’s disease: A European task force consensus, Lancet Neurol 6, 56–62.