Tramiprosate (Alzhemed) for Alzheimer's disease

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To clearly define the clinical efficacy and safety of Tramiprosate for the treatment of Alzheimer's disease.

Background

Description of the condition

Alzheimer’s disease (AD) is the most common form of dementia in old age. Almost five million individuals in the United States alone currently suffer from the condition and it affects over 24 million people worldwide (Ferri 2005). The UK Alzheimer's society (Alzheimer's Society 2008) states that there are 700,000 people with dementia in the UK. This will rise to over one million people by 2025 and 1.7 million by 2051. Worldwide, more than 24 million people currently have dementia with the numbers expected to double every 20 years to more than 80 million by 2040.

Alzheimer’s disease is a neurodegenerative disorder associated with amyloid deposition in the brain. This process starts in the hippocampus and progresses to the bilateral cortical areas. Growing evidence suggests that the underlying pathology precedes clinical disease onset by many years (Golde 2003 ), and, several studies suggest that individuals at risk for AD have elevated biomarkers ABeta 1‐40 (ABeta 1‐40) and/or ABeta 1‐42 many years before symptoms occur (Ertekin‐Taner 2008).

AD progressively affects affects many functions including memory, orientation and judgment, community functioning, activities of daily living and other domains and it is commonly accompanied by behavioural and psychiatric symptoms (BPSD) which increase the burden of caregivers (Cummings 2004; Marson 2000; Grossberg 2002).

A major strand of current research is targeted at modifying amyloid beta protein (Aß) deposition. This may be by reducing the generation and aggregation of Aß or by promoting clearance of amyloid deposits that are in development. Genetic studies of mutations that cause early AD provide support that preventing ABeta aggregation might eventually prevent or delay AD (Delacourte 2006). The importance of brain ABeta in the development of AD has increased interest in developing anti‐amyloid treatments; many small molecule inhibitors of ABeta aggregation have been identified (Soto 2005 ).

Description of the intervention

Tramiprosate (Alzhemed) (3‐aminopropane‐1‐sulphonic acid) (Homotaurine) (Cerebril) (VIVIMIND) is a low molecular weight agent found in transgenic mice studies to bind to soluble and insoluble ABeta (Gervais 2007). The drug is an orally administered, small organic molecule, that is is analogous to taurine, but with an extra carbon in its chain. Taurine is an analogue of the amine acid cysteine.

How the intervention might work

Tramiprosate has been specifically designed to modify the course of AD through its anti‐amyloid activity. As part of a "disease modifying" novel class of product candidates, Tramiprosate is expected to act at many levels:

• Tramiprosate reduces soluble and insoluble ABeta 40 and ABeta 42 fractions by 25 to 30%, and causes a 30% reduction in brain plaque deposition and up to 60% lower plasma ABeta levels in in TgCRND8 transgenic mice with early‐onset, aggressive brain amyloidosis (Gervais 2007).

• It also appears to protect against ABeta‐induced neurotoxicity in hippocampal culture (Kryzwkowski 2007).

• Another hypothesis is that Tramiprosate reduces the phosphorylation of tau through inhibition of the relevant kinase (Delacourte 2006).

Why it is important to do this review

The currently licensed treatments for Alzheimer’s disease are neurotransmitter‐based interventions, which act on the cholinergic and glutaminergic systems, and provide symptomatic relief. Memantine may also reduce excitotoxicity. If the deposition of ABeta is causal, a disease‐modifying drug targeting this mechanism would be a major advance.

Despite the fact that a Phase III trial using Tramiprosate for Alzheimer's disease was suspended in 2007, an identical agent Homotaurine was developed and marketed by Neurochem Pharmaceuticals (now Bellus Health Inc., Bellus Health Inc 2008) as a neutraceutical. This is now marketed as VIVIMIND for "protecting memory function" (OVOS Natural Health Inc., a subsidiary of Bellus Health Inc.) (OVOS Natural Health Inc. 2008).

The same drug, labelled Cerebril, was trialled for the treatment of hemorrhagic stroke due to cerebral amyloid angiopathy (Neurochem 2002; Neurochem 2004; Greenberg 2006). This Phase II trial involved 24 CAA patients with lobar cerebral haemorrhage who were randomized to receive three different daily doses of Cerebril for a period of 12 weeks. Neurochem 2004 and Greenberg 2006 reported that there were no safety concerns based on the most frequently reported adverse events (nausea and vomiting). Neurochem 2004 and Greenberg 2006 reported that "Cerebril at all dosages lowers the risk of secondary brain haemorrhage".

In the light of these contradictory findings, a systematic review is needed to clearly define the clinical efficacy and safety of Tramiprosate in patients with AD.

Objectives

To clearly define the clinical efficacy and safety of Tramiprosate for the treatment of Alzheimer's disease.

Methods

Criteria for considering studies for this review

Types of studies

Unconfounded, randomized, double‐blind, placebo‐controlled, parallel group trials will be included.

Studies with participants on stable treatment for AD will also be considered, provided there is a placebo control.

Types of participants

Participants with clinical diagnose of Alzheimer’s disease with mild, moderate or severe dementia based on standardized diagnostic criteria such as NINCDS‐ADRDA (McKhann 1984), ICD‐10, and DSM.

Patients receiving other Alzheimer's disease treatments during the course of the study will be considered, provided there is a placebo control group.

Types of interventions

Tramiprosate at any dosage with parallel placebo control.

Types of outcome measures

Primary outcomes

1. cognitive function

2. clinical global impression

3. changes in global disease severity

4. activities of daily living

5. behavioural disturbance

6. quality of life

7. safety as measured by the incidence of mortality and adverse effects.

Secondary outcomes

1. Drug pharmacodynamics measuring ABeta plasma or cerebrospinal fluid (CSF) level changes.

2. Changes of brain volume using MRI, functional MRI or CT scan.

Search methods for identification of studies

Electronic searches

See Cochrane Dementia and Cognitive Improvement Group methods used in reviews.

The Specialized Register of the Cochrane Dementia and Cognitive Improvement Group was searched on 10 November 2008 using the terms: tramiprosate OR cerebral OR homotaurine OR vivimind OR "3‐amino‐1‐propanesulfonic acid" OR alzhemed OR cerebril. This register contains reports of trials from bibliographic databases such as MEDLINE, EMBASE, PsycINFO, CINAHL, CENTRAL, LILACS, as well as references from searches performed in many trial registers and grey literature sources. The Specialized Register is updated regularly and search strategies used for the retrieval of studies can be seen on the Group's module in The Cochrane Library. Searches will also be performed in other sources to ensure that as many relevant reports of trials are retrieved as possible. Table 1 shows the search strategy that was used for retrieving references to studies in MEDLINE (via the Ovid SP platform).

Table 1.

Search strategy for MEDLINE

Source

Search strategy

Medline (Ovid SP)

1. tramiprosate.mp. 2. cerebril.mp. 3. homotaurine.mp. 4. vivimind.mp. 5. “3‐amino‐1‐propanesulfonic acid”.mp. 6. alzhemed.mp. 7. (1 OR 2 OR 3 OR 4 OR 5 OR 6) 8. (alzheimer* OR AD).mp. 9. Alzheimer Disease/ 10. dement*.mp. 11. (8 OR 9 OR 10) 12. (7 AND 11) 13. randomized controlled trial.pt. 14. controlled clinical trial.pt. 15. randomized.ab. 16. “randomi?ed controlled trial”.tw. 17. placebo.ab. 18. drug therapy.fs. 19. randomly.ab. 20. trial.ab. 21. groups.ab. 22. (13 OR 14 OR 15 OR 16 OR 17 OR 18 OR 18 OR 20 OR 21) 23. humans.sh. 24. (22 AND 23) 25. (12 AND 24)

Searching other resources

Conference proceedings, correspondence, news release and lists of references will be searched. The drug company Bellus will be contacted for unpublished trials or missing data.

Data collection and analysis

Selection of studies

Abstracts and reports of the comprehensive search will be checked independently by both reviewers. The selection of studies for exclusion or inclusion will be based on the specified inclusion criteria in the review. Full text of potentially relevant studies will be retrieved. The trialists will be contacted to clarify study details as necessary. Any disagreements about studies for inclusion will be resolved by discussion with the contact editor.

Data extraction and management

Data will be extracted from published reports. Trialists will be approached for any missing data from studies included in the review. A data extraction form which contains data reported in the original paper and ultimately data calculated by the reviewers will be developed and used.

Assessment of risk of bias in included studies

The methodological quality of each selected trial will be assessed using the Cochrane domain‐based evaluation: No for low risk, yes for high risk, unclear for otherwise (Higgins 2008).

Sequence generation

• Sequence Generation (Yes, low risk of bias): using a "computer random number generator", "coin tossing" or Throwing dice".

• Sequence Generation (No, high risk of bias): sequence generation using non‐random approach such as "date of birth", "day of week", hospital patient number, "patient preference", or by availability of the intervention.

• Sequence Generation (Unclear): randomization with unclear sequence generation.

Allocation concealment

• Allocation Concealment (Yes, low risk of bias): both participants and trialists could not predict the assignment if the reports describe allocation of treatment by referring to a "random number table", use of "envelopes" or "sealed envelopes".

• Allocation Concealment (No, high risk of bias): Participants or trialists could possibly predict the intervention assignment if the report describes the allocation concealment by: alteration, reference to case record numbers, date of birth, day of week, or any other approach.

• Allocation Concealment (Unclear, uncertain risk of bias): When the allocation concealment method was not stated sufficiently to make the yes or no judgment.

Blinding

• Blinding (Yes, low risk of bias): (1) Both participants and outcome assessors were blind, (2) No blinding, but the outcome measures are unlikely to be influenced by the lack of blinding, (3) Some key study personnel and participants were not blind, but the study assessors are still blinded, and when blinding is not possible due to the nature of the intervention.

• Blinding (No, high risk of bias): (1) No blinding and the outcome measures are affected by the lack of blinding, (2) there is chance that the blinding could have been broken

• Blinding (Unclear, uncertain risk of bias): no information to make the judgment of yes or no.

Incomplete data outcomes

• Incomplete data outcomes (Yes, low risk of bias): (1) all outcome data are reported in the study, (2) missing data will not affect the clinical impact of the intervention effect estimated, or (3) balanced numbers of missing data across the study groups.

• Incomplete data outcomes (No, high risk of bias: (1) imbalance in numbers of missing data across the study groups, or (2) missing data have a clinical impact on the estimated intervention effects.

• Incomplete data outcomes (Unclear, uncertain risk bias): when the outcome is not addressed or unable to clarify yes or no.

Free of selective reporting (using the ORBIT assessment tool)

• Free of selective reporting (Yes, low risk of bias): All the primary and secondary outcomes are reported.

• Free of selective reporting (No, high risk of bias): (1) not all the prespecified outcomes are reported, (2) Some primary outcomes are reported using some not prespecified method, (3) Incomplete outcome data, or (4) failure to reporting some key outcomes that would be expected to have been reported such a study.

• Free of selective reporting (Unclear, uncertain risk of bias): unclear information to fall into yes or no categories.

Other bias

• Other bias (Yes, low risk of bias): the repot is free from any other elements of bias

• Other bias (No, high risk of bias): such as baseline characteristic imbalance across the study groups, or inappropriate study design.

• Other bias (Unclear, uncertain risk of bias): No sufficient information in the report to make the yes or no judgement.

Measures of treatment effect

The outcomes measures in dementia and cognitive impairment trials arise from ordinal rating scales. Where the rating scales used in the trials have a reasonably large number of categories (more than 10) the data will be treated as continuous variables arising from a normal distribution.

The summary statistics required for each trial and each outcome for continuous data are the mean change from baseline, the standard deviation of the mean change, and the number of patients for each treatment group at each assessment. Where changes from baseline were not reported, the mean, standard deviation and the number of people in each treatment group at each time point will be extracted if available.

For binary data, the number in each treatment group and the numbers experiencing the outcome of interest will be used.

The baseline assessment is defined as the latest available assessment prior to randomization, but no longer than two months before randomization.

For intention‐to‐treat analysis, for each outcome measure, data will be sought on every person assessed, the data will be used irrespective of compliance, whether or not the person was subsequently deemed ineligible, or otherwise excluded from treatment or follow‐up. If intention‐to‐treat data were not available in the publications, "on‐treatment" or the data of those who complete the trial will be sought and indicated as such.

Unit of analysis issues

For cross‐over design studies, only data from the first treatment phase after randomization will be eligible for inclusion. Data from 'open‐label' follow‐on phases after the randomized study will not be used to assess safety or efficacy because patients are usually not randomized, nor are treatments concealed. For long‐term studies, outcomes at different time points will be assessed using separate analyses.

Dealing with missing data

For unreported data, trialists will be approached first; if the data is not provided, we will perform statistical calculations to obtain the missing data. A sensitivity analyses will be performed to assess the sensitivity of the results to the assumptions that are made.

Assessment of heterogeneity

When pooling trials in meta‐analyses, chi‐square value, the P value and I² will be used to identify heterogeneity across studies. When I² > 40%, then sensitivity analyses will be performed to investigate possible sources of heterogeneity.

Assessment of reporting biases

Risk of bias will be assessed by completing the risk of bias tables for each included study. A summary of findings tables will be completed for the primary outcome measures.

Data synthesis

Individual study analyses will be combined using a fixed effects model to provide a pooled estimated effect from continuous and binary data.

Subgroup analysis and investigation of heterogeneity

Subgroup analysis, where data are available, will be investigated based on dosage, duration of treatment, and severity of AD. If there is evidence of heterogeneity, homogenous trials will be pooled or a random effects model will be used.

Sensitivity analysis

If the overall review conclusion is influenced by a particular decision or missing information, sensitivity analysis will be applied across studies.

What's new

Date	Event	Description
3 August 2018	Amended	This protocol, published in 2009, was withdrawn from the Cochrane Library in August 2018 due to lack of progress.

Contributions of authors

RM: developing title and drafting protocol; all correspondence

HC: commenting on and editing protocol.

Sources of support

Internal sources

• Cochrane Dementia and Cognitive Improvement Group, UK.

External sources

• Alzheimer's Association, USA.

Declarations of interest

None known.

Notes

This protocol, published in 2009, was withdrawn from the Cochrane Library in August 2018 due to lack of progress.

Withdrawn from publication for reasons stated in the review