AD-Smart: Alzheimer’s trial will test diabetes and ADHD drugs on patients
Source: View publication →
A news report announcing the launch of AD-Smart, a pioneering UK-based multi-arm adaptive platform trial that will simultaneously test multiple repurposed drugs—starting with metformin and atomoxetine—against a shared placebo group to treat Alzheimer's disease.
Key Findings
Study Design
Study Limitations
Clinical Significance
The AD-Smart trial represents a major methodological shift in dementia research. By utilizing an adaptive platform design, investigators can significantly cut research costs, reduce the necessary number of total trial participants by using a shared placebo arm, and accelerate the identification of effective therapies. Testing repurposed drugs with established safety profiles means that any successful candidate could be implemented into routine clinical care much faster than a novel therapeutic.
Historical Context
Historically, Alzheimer's disease clinical trials have tested single developmental drugs in isolation. This traditional approach exhausts large pools of potential trial participants and takes many years, contributing to a landscape where no current NHS treatments can halt or reverse Alzheimer's disease progression. The AD-Smart trial borrows the multi-arm, multi-stage platform design that proved highly successful in other fields (such as the RECOVERY trial for COVID-19) to introduce much-needed efficiency into neurodegenerative disease research.
Guided Discussion
High-yield insights from every perspective
What is the pathophysiological rationale for testing metformin and atomoxetine, drugs traditionally used for diabetes and ADHD respectively, in patients with Alzheimer disease?
Key Response
Metformin addresses brain insulin resistance (often termed type 3 diabetes) and AMPK activation, potentially reducing tau phosphorylation. Atomoxetine, a norepinephrine reuptake inhibitor, targets the early degeneration of the locus coeruleus in AD, aiming to improve attention and executive function while potentially offering neuroprotection.
If atomoxetine or metformin were integrated into the care plan for an elderly patient with Alzheimer disease, what specific adverse effects and drug interactions would you need to monitor closely?
Key Response
Metformin carries risks of gastrointestinal intolerance, B12 deficiency (which can worsen cognitive decline), and lactic acidosis. Atomoxetine can cause increased heart rate, hypertension, weight loss, and reduced seizure threshold, which are particularly concerning in frail, elderly patients with cardiovascular comorbidities.
Considering the recent approvals of anti-amyloid monoclonal antibodies, how might the AD-Smart trial incorporate these as background therapies, and what challenges does this pose for detecting the efficacy of repurposed drugs?
Key Response
As anti-amyloid therapies become standard of care, platform trials must decide whether to stratify by or allow concurrent use. This creates potential overlapping adverse effects (like ARIA complicating clinical assessments) and a heterogeneous baseline that could dilute the effect size of the repurposed investigational drugs.
How does the multi-arm adaptive platform design of the AD-Smart trial fundamentally shift the paradigm of Alzheimer disease research compared to traditional two-arm randomized controlled trials?
Key Response
Platform trials use a shared placebo group and allow for continuous introduction of new therapies and early cessation of futile arms. This maximizes patient resources, reduces the number of patients receiving placebo, accelerates drug discovery, and mimics a dynamic clinical practice environment.
Scholarly Review
Critical appraisal through the lens of expert reviewers and guideline development
In a multi-arm, multi-stage (MAMS) adaptive platform trial like AD-Smart, what are the primary statistical challenges regarding the shared control group over time, and how can researchers mitigate time-trend biases?
Key Response
A shared control group evaluated over years is susceptible to time-trend biases due to changes in standard of care or population health. Researchers must employ concurrent randomization and advanced Bayesian hierarchical models to adjust for temporal drift in the placebo arm.
Given that repurposed drugs like metformin and atomoxetine have distinct, recognizable side effect profiles, how should a peer reviewer critically evaluate the integrity of blinding in the AD-Smart trial?
Key Response
Unblinding due to recognizable adverse events (e.g., GI upset for metformin, tachycardia for atomoxetine) is a major threat to validity, especially with subjective cognitive endpoints. Reviewers must look for robust blinding assessments to ensure efficacy signals are not driven by expectation bias.
If AD-Smart demonstrates significant cognitive benefit for a repurposed drug like metformin, how would guideline committees weigh this evidence against newly approved, high-cost anti-amyloid therapies when determining first-line recommendations?
Key Response
Current guidelines balance modest benefits of cholinesterase inhibitors against the high cost and ARIA risks of anti-amyloids. A positive trial for a cheap, orally bioavailable drug would necessitate updating guidelines to prioritize cost-effectiveness and accessibility, potentially placing it ahead of or as an essential adjunct to mAb therapies.
Clinical Landscape
Noteworthy Related Trials
SNIFF Trial
Tested
Intranasal insulin 40 IU daily
Population
Adults with mild cognitive impairment or Alzheimer disease
Comparator
Placebo
Endpoint
Change in ADAS-Cog12 score
ADMET 2 Trial
Tested
Methylphenidate 20 mg daily
Population
Patients with Alzheimer disease and apathy
Comparator
Placebo
Endpoint
Change in NPI apathy subscale score
NorAD Trial
Tested
Atomoxetine up to 100 mg daily
Population
Patients with mild-to-moderate Alzheimer disease
Comparator
Placebo
Endpoint
Change in ADAS-Cog11 score
Tailored to your role
Want this tailored to you?
Add your specialty or training stage to get role-specific takeaways and more questions.
Personalize this analysis