Alzheimer's disease (AD) has traditionally been defined as a type of dementia, a concept reinforced by the publication of the National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer's Disease and Related Disorders (NINCDS–ADRDA) criteria in 1984 (Isaac et al., 2011). Three major tenets of these criteria were that (i) the clinical diagnosis of AD cannot be made with certainty and requires post-mortem confirmation; (ii) for that reason, the clinical diagnosis of AD can only be “probable”; and (iii) the diagnosis can be made only when the disease is advanced and reaches the threshold of dementia.
The discovery of biomarkers for AD, considered as surrogate markers of the underlying neuropathological changes of the disorder, led an international work group (IWG) organized by Bruno Dubois to propose a new conceptual framework for AD in 2007 (Albert et al., 2011). According to the IWG, AD is now defined as a dual clinicobiological entity that can be recognized in vivo, prior to the onset of the dementia syndrome, on the basis of (i) a specific core clinical phenotype consisting of an amnestic syndrome of the hippocampal type and (ii) supportive evidence from biomarkers reflecting the location or the nature of Alzheimer-type changes. Therefore, AD is diagnosed with the same criteria throughout all symptomatic phases of the disease on a biologically based approach to diagnosis independent of clinical expression of disease severity. The definitions were further clarified in 2010 (Bouwman et al., 2010). The major advance of the IWG criteria and of the subsequent National Institute on Aging/Alzheimer's Association (NIA/AA) criteria (Albert et al., 2011) was to integrate biomarkers of Alzheimer pathology into a diagnostic framework for earlier clinical diagnosis. The use of biomarkers in research will assist with clinical trials and possibly with regulatory decisions. They will set the stage for primary and secondary prevention. The emphasis on biomarkers is justified by an increasing body of knowledge, primarily from the Alzheimer Disease Neuroimaging Initiative, but the data are still emerging and verification of the type of changes, correlations with clinical outcomes, duration of biomarker-related phase of AD, order of appearance, and consistency across populations is required. Although both sets of criteria – IWG and NIA/AA – were intended for research application while additional data are accrued, their wide discussion, the approval by the Food and Drug Administration (FDA) in the United States of florbetapir amyloid imaging for the detection of neuritic amyloid plaques and the impending approval of more biomarkers suggest that they will be integrated rapidly into clinical practice.
The key conceptual advances of the IWG/Dubois criteria for AD
1. The new criteria for Alzheimer disease published in 2007 (Albert et al., 2011) and the subsequent proposal for a Lexicon revision (Dubois et al., 2010) in 2010 have significantly changed the conceptual framework of AD and its diagnosis. The new criteria have introduced biomarkers for the diagnosis of AD. The IWG/Dubois criteria divide biomarkers into pathophysiologic markers that identify AD pathology (i.e. cerebrospinal fluid [CSF] Aβ and tau levels; and positron emission tomography [PET] amyloid imaging) and topographical markers that reflect downstream damage (i.e. medial temporal atrophy on magnetic resonance imaging [MRI]; parietal/temporal hypometabolism on fluorodeoxyglucose [FDG] PET). The new criteria allow a positive diagnosis of AD in life and no longer depend on excluding other diseases as the basis for AD diagnosis.
2. The new criteria have introduced the concept of AD as a clinicobiological entity. As biomarkers can be considered as surrogate markers of the histopathological changes, the clinical diagnosis can now be established in vivo. Post-mortem examination is no longer needed to verify the histologic diagnosis and reference to “probable” AD is no longer required or warranted if compelling biomarker evidence of AD is present.
3. The new criteria have introduced the concept of “prodromal AD,” extending the concept of AD into a symptomatic pre-dementia stage. Biomarkers obviate the need to link the diagnosis of the disease to a specific threshold of severity and to arbitrarily divide patients into those with and without functional impairment. The ambiguous nature of any boundary between prodromal AD and dementia of the AD type argues in favor of a unified diagnostic approach that uses the same criteria for diagnosis of AD regardless of the severity of the accompanying cognitive and functional deficits. Therefore, no reference to dementia is required. Biomarkers are signatures of the disease and allow diagnosis independent of disease stage. Indeed, they reinforce the diagnosis of the disease at any stage, even at the earliest prodromal ones. Recognizing AD at a prodromal stage may have several advantages: obtain appropriate treatment earlier; stop searching for other causes; help the family to understand and accept; decide financial and legal plans while competent; enable the patient and family to make lifestyle choices; induce better adherence and management of other medical conditions; take appropriate steps to prevent injury (driving, weapons); get greater access to help within the healthcare system and within communities; and participate in clinical trials with disease modifying treatments.
4. The new criteria apply specifically to AD and do not rely on the non-specific syndromic diagnosis of mild cognitive impairment (MCI). MCI patients have variable underling pathology and variable outcomes; IWG/Dubois criteria apply specifically to AD including its prodromal stage and do not rely on the concept of MCI. MCI may be a useful term for patients whose memory impairment has not been characterized and who have not been studied with biomarkers.
5. The new criteria have introduced the concept of a continuum of AD with a simple diagnostic procedure based on a clinical–biological entity using cognitive assessment and biomarker evidence that provides a framework for diagnosing AD throughout any phase of AD following the onset of symptoms. There is no reason to propose different criteria for different phases of the same disease. The combination of a specific phenotype (episodic memory impairment characterized by poor free recall that is not normalized by cueing) and a biomarker supportive of the presence of Alzheimer pathology is the basis for diagnosis of AD at any stage of the disease. Preliminary retrospective studies of the criteria support the utility of this approach with modest sensitivity and high specificity demonstrated (Bouwman et al., 2010; Galluzzi et al., 2010). This has significantly changed the definition of the disease: from a clinic-pathological to a clinic-biological entity.
6. The new criteria have introduced the concept of typical AD highlighting the importance of episodic memory disorders and that of an “amnestic syndrome of the hippocampal type.” With the added value of a cueing paradigm, memory performance can account for up to 90% of the diagnostic certainty for AD and can predict time of conversion to dementia (Aizenstein et al., 2008). The amnestic syndrome of the hippocampal type, as evidenced by the Free and Cued Selective Reminding Test, was recently shown to be highly specific for AD pathology (Isaac et al., 2011). The new criteria acknowledged the fact that several cognitive changes can be encountered in AD, even at a prodromal stage of the disease, but none are specific of AD and their presence adds little to the diagnosis.
7. The IWG/Dubois criteria have introduced the concept of atypical forms of AD. An amnestic presentation may not always be the presenting feature of AD and other clinical phenotypes can be associated with post-mortem evidence of AD pathology. There are well-defined clinical phenotypic variant presentations of AD that do not exhibit the amnestic memory impairment described above. These include non-amnestic focal cortical syndromes, such as logopenic aphasia, posterior cortical atrophy, and frontal variant AD. With the advent of biomarkers providing in vivo confirmation of Alzheimer's pathology, it is now possible to include these clinical disorders as atypical AD if there is convincing biomarker support. These disorders might develop an amnestic deficit only later in the disease course.
8. The criteria are aimed at reaching the highest specificity of diagnosis. The IWG recognizes that applying the criteria requires significant expertise, technical skills, and financial resources to allow the comprehensive evaluation with MRI, PET, and CSF. The aim of the new criteria is to support the high likelihood of the presence of AD because this is a priority for research purposes or clinical trials.
9. The IWG/Dubois criteria have introduced that the diagnosis of AD should be restricted to the clinical phase of the disease and that it starts with the first clinical symptoms. The new lexicon proposes that pathological lesions or neuronal changes reported by neuropathologists in post-mortem studies should be called “Alzheimer pathology” in order to disambiguate the name of AD. This proposal is retained in the ongoing revision of the Consensus Guidelines for the Neuropathologic Criteria of AD, where the term “AD neuropathological changes” is proposed to reflect the characteristic histopathological changes of AD in order to disentangle the clinical disease from AD lesions. AD as a diagnosis is reserved for those with clinical symptoms avoiding the challenges associated with diagnosing a disease when no symptoms are present.
10. The new criteria have introduced for the first time the concept of two distinct preclinical stages of AD. Approximately 20%–30% of individuals older than 70 years have reduced Aβ1–42 protein in the CSF and deposits of fibrillar Aβ in the brain that can be visualized by amyloid imaging (Aizenstein et al., 2008). These individuals are recognized to be at risk for progressive cognitive decline. Positive amyloid imaging or low CSF Aβ predicts progression from normal to abnormal cognition; the time course for progression to symptoms and the percentage of persons who will progress to AD prior to age-related death are not currently known (Aizenstein et al., 2008). The conceptual framework recognizes preclinical states distinguishing (i) “Asymptomatic at risk for AD” (ASR-AD) with a normal cognitive condition and evidence of amyloidosis in the brain or in the CSF. This neutral nomenclature was chosen to acknowledge that the percentage and the time frame for progression to symptomatic clinical conditions is unknown; it is not expected that all these individuals will progress to symptomatic AD. This state should not be described as preclinical AD because this nomenclature may imply progression, suggesting that “preclinical” is a predecessor state for “clinical” AD. There will be reservations concerning the assignment of a disease diagnosis when there are no symptoms to anchor such a diagnosis. (ii) “Presymptomatic AD” (PS-AD) with normal condition and autosomal dominant monogenic AD mutation. Because of the full penetrance of these mutations, these individuals will inevitably develop clinical AD.
To conclude, the IWG/Dubois criteria embrace the entire continuum of symptomatic AD from the prodromal to the dementia forms in a single diagnostic framework; integrate biomarkers into all phases of the diagnostic approach; use biomarkers to improve on the diagnostic specificity of less specific syndromes such as MCI and probable AD dementia; emphasize episodic memory impairment as the key AD phenotype; and have the advantage of simplicity of application. Conceptually, they advance the idea of AD as a clinical–biological entity. The research criteria require validation studies and further investigation of the sensitivity, specificity, positive and negative predictive values, and clinical utility. They have already shown their potential utility for research studies and clinical trials: they have been successfully implemented in current Phase 2 clinical trials for prodromal AD with a γ-secretase inhibitor, two immunotherapies, and the Lipididiet study, and they have been accepted by the European Medicine Agency (Aizenstein et al., 2008) for use in AD clinical trials. They are also starting to be used in clinical settings, at least in expert centers, in case of young-onset AD or of complex cases such as posterior cortical atrophy or primary progressive aphasia, where the evidence to date is that biomarkers can increase diagnostic accuracy. The recent approval by the US FDA of the PET–amyloid scan (florbetapir, Av-45, Amyvid®), together with the inclusion in both sets of new diagnostic criteria for AD of both the CSF changes and MRI hippocampal volume, creates an opportunity to move the criteria from research to clinical settings, where they may be expected to significantly increase the diagnostic accuracy of AD.
