Commentary on “Diagnosis of Alzheimer's disease: Two decades of progress”

Dr. Zaven Khachaturian has provided an excellent account of the history of Alzheimer's disease (AD) research in the world, particularly in the United States, in the last few decades. His perspective is unique because he was instrumental in many of the key activities that took place during this time. I am not entirely certain that he did not also have a role in Kraeplin's decision to name the disease after Alois Alzheimer in 1910. The role of the National Institute on Aging (NIA) in this evolutionary process has been quite significant through the development of the Alzheimer's Disease Centers program as well as through several other key initiatives including the Alzheimer's Disease Patient Registries, Alzheimer's Disease Cooperative Study, and the recent Alzheimer's Disease Neuroimaging Initiative. I would like to comment on 2 issues raised by Dr. Khachaturian regarding this period in history: (1) the characterization of “normal” aging and (2) the emphasis on an earlier diagnosis of the AD process, in particular, mild cognitive impairment (MCI). Many investigators typically state that the condition they are studying differs from normal aging. However, in reality, one area that we have not addressed adequately concerns the characterization of the cognitive changes of normal aging. This remains a significant challenge for us, because many of our investigations are contingent on this distinction. For example, as I will discuss below, the construct of MCI is predicated on the differentiation of cognitive changes of normal aging from the very early impairments seen in MCI. However, it is equally important to distinguish MCI from early AD, and both of these demarcations present prominent challenges. On the normal aging boundary, we need to know “How much forgetfulness is too much?” Most investigators believe that some loss of cognitive efficiency is typical for aging insofar as many individuals become somewhat more forgetful for recent events and experience signs of slowing of cognitive processing as well as some loss of cognitive flexibility. Are these changes normal? These questions involve semantic issues concerning normality. Do we mean statistically normal or typically normal? Or, do we mean optimally normal with no loss of function? Some investigators believe that any loss of function represents the presence of disease, albeit occult. These are difficult issues but are important when dealing with aging and topics concerning the boundaries and thresholds for disease. The issue of normal performance pertains not only to cognition but also to functional activities. For example, many of the commonly used rating scales presume a meaningful functional change is secondary to alterations in cognition. However, this assumes that we know how much of a functional change one can allow as a part of aging before we can make an adequate judgment regarding abnormalities. The same issue that arose with cognition arises once again with regard to functionality. That is, are any functional changes allowable as a part of normal aging? Or, do we allow for some slowing of physical activity as a result of the cognitive aging process? The discussion of normal aging pertains to all aspects of AD research and even involves the fundamental underpinnings of the disorder itself. For example, the “gold standard” for AD has always been the neuropathologic confirmation of the clinical syndrome of dementia. However, beginning with the hallmark “Khachaturian criteria” for the neuropathologic diagnosis of AD in 1985, through the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) neuropathology criteria in the late 80s and early 90s to the present, NIA-Reagan criteria, there often has been the reference to age-adjusted criteria implying that some aspects of AD pathology might be a part of normal aging. More recently, the NIA-Reagan criteria have avoided age-adjusted criteria but have retreated to a probabilistic scheme requiring the judgment that the neuropathologic changes seen are reflective of an underlying AD process. A lingering problem for these criteria has been the issue of how much pathology is to be expected as a part of aging. This conundrum likely influences the difficulties in developing an absolute biomarker for AD. One has to compare the levels of cerebrospinal fluid β-amyloid (Aβ) and tau with comparable samples of aged normal individuals to determine appropriate thresholds. Once again, the issue of the boundaries with aging emerges and the tacit notion that some expression of tau and Aβ in the spinal fluid can be tolerated as a part of the aging process. A similar argument arises in the domain of neuroimaging. Does one assume that any atrophy of the hippocampal formation, entorhinal cortices, frontal lobes, or other cerebral structures is believed to be abnormal? Again, are we dealing with a statistical or an optimal aging process? Virtually all brains shrink with aging, and consequently, is this normal? What about glucose metabolism on PET and aging? If we see any diminution of glucose metabolism in the temporoparietal region, does that constitute prima facia evidence for AD, or can some of these glucose metabolic changes be considered a part of aging? I suspect the new amyloid imaging ligands are going to face the same dilemma. That is, although it is anticipated that amyloid imaging measures may give us an index of pathologic burden, this does not help us distinguish who will develop the disease in life versus who can tolerate a certain degree of amyloid load in the brain without becoming clinically demented. Dr. Khachaturian mentions the initial charge to the NIA was “to develop and support interdisciplinary research on healthy ‘normal’ aging as well as …” I suggest that numerous studies have contributed a great deal toward addressing this issue, but much more work needs to be done. Clearly, a benefit of the Alzheimer's Disease Centers program has been to encourage investigators to develop control groups of normal aged individuals. The Alzheimer's Disease Patient Registry program also encouraged investigators to evaluate cohorts of normal aging individuals, and one can argue that this aspect of that program was as beneficial as the study of the disease cohorts themselves. In fact, in the Mayo Alzheimer's Disease Patient Registry, numerous studies of cognition, imaging, and neuropathology on normal individuals have resulted from this effort. Although certain arguments can be entertained regarding the appropriateness of certain normative neuropsychology databases, the NIA-sponsored registries have produced sets of normative data on cognitive function on a wide spectrum of commonly used neuropsychological instruments that have served as the basis for the next generation of published neuropsychological instruments. As such, these NIA-sponsored programs have had an enormous effect on the practice of clinical neuropsychology in this country. That is not to say that all of these issues have been resolved, but this represents a significant stride toward the characterization of individuals who are aging typically. In spite of this progress, a great deal of additional research needs to be carried out on characterization of the normal aging process. Perhaps the term normal should be abandoned and typical aging studied to allow more accurate characterization of what individuals can expect with and without various comorbidities. Studies of optimal aging can represent the upper limits of the aging process. The NIA has been a leader in this area of research and will likely continue to promote healthy aging as well as the study of disease processes. This spinoff of the Alzheimer's disease research program has been extremely important for the mission of the NIA. All these issues surrounding normal aging bear upon the construct of MCI, which has come to represent an intermediate stage between the changes of aging and those that manifest in the very earliest stages of the dementias such as AD. But the construct clearly begs the question of “what is normal?” Nevertheless, although these issues remain unresolved, the study of MCI has contributed important data to the literature. As Dr. Khachaturian indicates, these topics involving intermediate stages between the changes of normal aging and early dementia would not have been conceivable 15 to 20 years ago, but only through progress in the field have they become central areas of investigation. The construct of MCI has influenced many areas of study including epidemiology, clinical diagnostic criteria, neuropsychology, neuroimaging, neuropathology, mechanisms of disease, and treatment. These inroads represent advancements in each of these areas and have influenced the direction of research in AD toward a slowing of the onset of the pathologic processes and ultimately toward prevention. However, in reality, considering the previous discussion on normal aging, “prevention” may represent a postponement of some of the changes that can be manifestations of the aging process. Nevertheless, either slowing these processes or preventing them would have a significant impact on the quality of life of many individuals as well as having a significant impact on society and the health care system. Dr. Khachaturian has captured the essence of research on aging and dementia over the last several decades in this country. This work had a profound influence on our understanding of normal and pathologic processes as well as provided the substrate for the development of the careers of many prominent academicians in this country. The Alzheimer's Disease Centers program itself has provided an environment for the fostering and maturation of numerous prominent academicians and clinicians in this country. Many individuals have risen through the academic ranks at many prominent academic institutions in this country because of the rich environment provided by this research support. The impact of these programs on the lives of many individuals cannot be underestimated. Therefore, Dr. Khachaturian's account of the contribution of AD research to the academic community is particularly apropos.