Overestimating the Effects of Healthy Aging

Typical cognitive aging is often defined as aging free of dementia, and yet it does not seem to exclude dementia-related pathology. The most common type of dementia is Alzheimer’s disease (AD) – it is a progressive neurodegenerative disease that affects multiple cognitive domains, including memory, attention, reasoning, judgment, language, as well as behavior. AD accounts for 60–80% of dementia cases, currently affecting 35 million people worldwide and progressing toward an estimated 115 million by the year 2050 (Riedel, 2014). Since there is no cure for AD, the research focus has shifted from the symptomatic stage to the earlier stages of the disease, with the goal to develop treatments that would effectively delay the onset of symptoms and progression of this devastating disease (Selkoe, 2012; Bateman, 2015). The need to diagnose and begin treatment even before overt symptoms appear has generated in neurological researchers a strong interest in cognitively healthy people who are at risk for AD. Recent modifications of the research and diagnostic criteria for AD reflect this trend by including cognitively normal (CN) at-risk-for-AD persons as a preclinical stage in the AD continuum (Dubois et al., 2007, 2010; Sperling et al., 2011). Crucial in themodified criteria for AD is the concept of biomarkers of Alzheimer’s pathology, i.e., physiological and anatomical parameters that can be objectively measured to establish the presence of changes due to the disease (Jack and Holtzman, 2013). Since amyloid-β neuritic plaques and neurofibrillary tangles have been established as the pathological hallmarks of AD, biomarkers of amyloidosis and neurodegeneration are used as in vivo indicators of the presence of Alzheimer’s pathology (Dubois et al., 2007, 2010; Sperling et al., 2011; Jack and Holtzman, 2013). The most validated AD biomarkers are cerebrospinal fluid (CSF) amyloid-β1–42, total tau and phosphorylated tau as pathophysiological biomarkers of Alzheimer’s pathology, and brain regional structural and metabolic changes, as topographical biomarkers (Dubois et al., 2007; Jack and Holtzman, 2013). However, it has been recently recognized that the topographicalmarkers lack pathological specificity necessary for diagnosis, and that they are more useful for measuring the disease progression (Dubois et al., 2014). According to this view, in vivo evidence of Alzheimer’s pathology, and thus preclinical AD stage, would constitute increased cerebral amyloid burden/decreased CSF amyloid level together with increased levels of total tau and phospho-tau. The general idea is that if a person with Alzheimer’s pathology lives long enough, they will eventually progress to AD dementia. Thus, biomarkers-based diagnosis provides an opportunity to introduce early treatments that will attempt to delay appearance of dementia symptoms and disease development. While the value of early diagnosis is beyond any dispute, the question of how to disentangle typical aging and preclinical AD remains open. This issue is further complicated by the heterogeneous nature of the preclinical AD. According to onemodel, preclinical AD comprises three stages: at stage 1, only brain amyloidosis is evident; at stage 2, both cerebral amyloidosis and neurodegeneration are present; and at stage 3, these features are further combined with subtle cognitive changes (Sperling et al., 2011). Two additional stages have been added to thismodel: stage 0, which represents cognitively intact people without brain amyloidosis, neurodegeneration, or subtle cognitive changes, and suspected non-amyloid pathology (SNAP), which is characterized by normal amyloid markers and abnormal neurodegeneration markers and which is not necessarily related to AD (Jack et al., 2012). While it is clear that preclinical AD is characterized by the absence of cognitive impairment