Table 1.
Issues of RCTs of AD.
| Issue | Possible solution |
|---|---|
| Subjects | |
| Target group selection: patients with AD have various types of neuropathology (i.e., amyloid plaques, NFTs, infarcts, and Lewy bodies) | Criteria for identifying subgroups with more homogeneous biomarker evidence of AD pathology are needed to facilitate RCTs |
| In patients with mild-to-moderate AD, the disease could be too advanced for a disease-modifying effect of a specific drug (e.g., immunotherapy) | RCTs that include patients with early AD might enable detection of disease-modifying effects; investigation into which stage of the AD process a therapeutic strategy is more effective is warranted |
|
| |
| Agents | |
| Choosing the right drug: compounds with positive results in preclinical and early clinical testing failed in large phase III RCTs, with costly losses (e.g., tramiprosate) | Robust proof-of-concept studies should be mandatory Investigators should take into account class efficacy |
| Use of drug-related biomarkers in preclinical and early clinical stages can help to confirm the target engagement and to assure early withdrawal of ineffective drugs | |
| Some RCTs were likely hindered by the inability to reach a therapeutic dosage (e.g., tarenflurbil) or short treatment duration | Optimization of drug dosage and treatment duration based on pharmacokinetics |
| Genetics: polymorphisms (e.g., APOE,) might affect drug response | Personalized therapeutic approach: considering genetic polymorphisms that affect drug response can help to optimize drug dosage (e.g., increased doses for individuals with a rapid metabolism) |
|
| |
| Outcome measurements | |
| Measuring effects: many RCTs are developed according to the design of AChEI RCTs, an approach that has indicated the AChEI symptomatic effect but is not sensitive in detecting the efficacy of disease-modifying drugs, rating scales used may have low sensitivity for changes and/or the drug type assessed and these tools have a subjective component | Development and use of relevant, reliable, multidimensional measures for clinical (cognitive and functional) endpoints are key factors, as well the use of biomarkers (neuroimaging, CSF, or blood molecules) that reliably and quantitatively correlate with disease progression; collection of baseline data (clinical, biomarkers) that can be used as reference to interpret later findings is advisable; for early AD (i.e., mild cognitive impairment), self-rated and observer-rated assessments of activities of daily living, instrumental activities of daily living, and quality of life are recommended |
| Unreliable evaluation of patients by RCT raters | Adequate training and monitoring of RCT raters to maximize homogeneous recruitment of patients, reduce variance, and guarantee a more accurate rating; effective implementation of quality control on data at research sites |
|
| |
| Optimization of resources | |
| Consistency: multicenter RCTs done in several countries can have cultural and linguistic issues with assessment scales (e.g., translation, validation), as well as infrastructure problems (technological disparities between centers) | Multicenter trials should use centers of excellence that are already experienced in RCTs to minimize intersite and intercountry variability |
| Unsuccessful preclinical and clinical studies are often not published leading to repetition of unsuccessful trials or errors | More collaboration between pharmaceutical companies and clinical researchers, with information sharing, can lead to more standardized RCT protocols, reduction of errors, and decreased costs |