Potential Utility of Practice Effects in Preventive Trials.

By the time an individual fulfills clinical cr i ter ia for Alzheimer’s disease (AD) dementia, extensive and likely irreversible neurodegeneration has already occurred. While we might be able to slow disease progression at this late stage, there is minimal likelihood of completely restoring brain and cognitive function once extensive neurodegeneration is present. Accordingly, most clinical trials for AD have shifted toward the prodromal (before dementia) or even the preclinical (before symptom onset) phase of the disease. With this shift towards AD prevention, new challenges have emerged, the most important of these being i) to better identify individuals on the path to AD dementia and ii) to develop optimal outcome measures that can capture the earliest changes related to AD. Studies involving individuals with autosomal dominant forms of AD (ADAD) suggest that diseaserelated changes occur in a predictable order, with a reduction of cerebrospinal fluid (CSF) amyloid levels appearing 25 years before expected symptom onset, accumulation of brain amyloid deposition assessed using positron emission tomography (PET) imaging and hippocampal volume loss 15 years before expected symptom onset, and memory deficits 10 years before expected symptom onset (1). While the time course of sporadic AD might be slightly different and more variable from one individual to another than what is found in ADAD, the order of biomarker changes in both forms of the disease is hypothesized to be very similar (2, 3). Based on this assumption, an international working group convened by the National Institute on Aging and the Alzheimer’s Association (NIA-AA) proposed a conceptual framework to define the preclinical (4) and the prodromal (5) phases of AD. Both phases are characterized by the presence of amyloidosis, which can be found alone (preclinical stage 1), and in association with neurodegeneration (preclinical stage 2), subtle cognitive decline (preclinical stage 3) and/or mild cognitive impairment (MCI, prodromal stage), reflecting increasing disease severity. Assessing the presence of amyloidosis using PET imaging or CSF assays might therefore be the optimal approach for identifying individuals in the preclinical and prodromal phases of the disease. When used alone however, these expensive and somewhat invasive procedures must be performed on a substantial number of individuals, as only about a third of cognitively normal older adults and half of individuals with MCI exhibit amyloidoisis (6). More importantly, because amyloid accumulation starts about two decades before dementia onset (1, 7), enrolling participants in prevention trials based solely on markers such as amyloid status may result in heterogeneous samples that include some individuals close to symptom onset while others remain decades away. Those who are close to symptom onset often show cognitive and functional decline, while those who are decades away are less likely to decline over a relatively short follow-up. Accordingly, effects of preventive interventions may not be apparent in individuals who are decades away from symptom onset, especially when conventional cognitive and functional outcomes are used. There is therefore a crucial need to develop better approaches for identifying individuals on the path to AD dementia who are at a similar severity stages.