Alzheimer's disease prevention: a reality check

Abstract

Sometimes only a crisis produces real change, said the famous US economist Milton Friedman. This saying would be a suitable preface for both a recent National Institutes of Health (NIH) Conference Statement on Preventing Alzheimer's Disease and Cognitive Decline and the latest report of the Alzheimer's Association (AA). The first document—still a draft version as The Lancet Neurology went to press—highlights the crisis of evidence in dementia prevention and provides guidance on how to tackle this situation. The figures in the AA report bluntly remind us why that effort is crucial. The NIH-sponsored document presents the conclusions of an independent panel of health professionals and public representatives, who were not experts on Alzheimer's disease (AD). Their unbiased judgments were based on presentations by and discussions with AD experts and on the findings from an exhaustive evidence report, commissioned by the Agency for Healthcare Research and Quality (AHRQ). The panel concluded that recommendations cannot be made for disease prevention because the available evidence is not robust enough for safe advice to be given. Before any interventions are implemented, the underlying risk factors should be characterised; however, according to the rigorous systematic review produced for the panel by the AHRQ, no definitive evidence exists on the association of modifiable risk factors with either cognitive decline or AD. Only for genetic factors, particularly the APOE ɛ4 allele, is there strong evidence of an association with increased risk of late-onset AD. There is moderate evidence for an association of risk of AD with some non-steroidal anti-inflammatory drugs and, in women, with conjugated equine oestrogen with methylprogesterone treatment. The evidence is weaker still when addressing relations between factors thought to be associated with increased risk of both cognitive decline and AD. These factors include diabetes mellitus, current smoking, and depression, but the level of evidence seems moderate at best. Findings are not conclusive on the association of cognitive engagement and physical activity with decreasing risk of cognitive decline and AD, and data are also limited on the protective effects of a Mediterranean diet. If decades of research can support only scant conclusions, how could the necessary evidence be obtained in the future? The panel proposes a set of challenging but fairly uncontroversial priorities, and the good news is that some clinical researchers are already working on these areas. The panel recommends the development of consensus-based diagnostic criteria. For a disease with a long asymptomatic prodromal phase, moving from syndromic to pathophysiological definitions is crucial, and several efforts are ongoing in this respect, such as those by the Dominantly Inherited Alzheimer Network to track the prodromal phase and by an NIH Task force to update diagnostic criteria. But diagnostic criteria rely on understanding the natural history of the disease. To accomplish this better understanding, the panel recommends the standardisation of outcomes and measures of exposure to potential risks; only reliable and validated measures can provide the evidence required to track the progression of cognitive decline and AD in large long-term population-based studies. The NIH Toolbox and the new Alzheimer's Disease Neuroimaging Initiative (ADNI), both works in progress, are good examples of attempts to harmonise assessments. These recommendations can be put into context by the AA report, which explores the consequences of a hypothetical advance that could lead to a preventive intervention. It is estimated that 13·5 million US citizens aged 65 and older will have AD in 2050. Moreover, the annual costs for their care will increase from US$172 billion in 2010 to $1·08 trillion in 2050. A hypothetical breakthrough intervention that could become available in 2015 and would delay onset by 5 years could reduce the number of patients in 2050 by over 40%, to less than 8 million. The cost of their care would then be reduced to $631 billion. The AA report remarks that this hypothetical scenario is based on assumptions that have been translated into real achievements for other chronic disorders, such as heart disease and HIV/AIDS; hence, it provides an optimistic counterpoint to the down-to-earth conclusions of the NIH Conference Statement. In crisis, Friedman thought, the actions taken depend on the ideas that are prevalent at the time. Both reports, released in the space of just a few weeks, have unambiguoushly exposed the crisis in AD prevention and laid down a useful set of ideas; now action must be taken.