Morris GP, Clark IA, Vissel B.
Questions concerning the role of amyloid-β in the definition, aetiology and diagnosis of Alzheimer's disease.
Acta Neuropathol. 2018 Nov;136(5):663-689. Epub 2018 Oct 22
PubMed.
This is a detailed analysis of the role of Aβ in Alzheimer’s disease and the justification for defining Alzheimer’s disease by the presence of Aβ, with supplementary information to extend particular arguments. It traces the historical development of thinking regarding the disease(s) and calls for open-mindedness and objectivity in research. It suggests a way forward to avoid the logical trap that the field appears to have fallen into. In my opinion, this paper should be essential reading for postgraduate students or postdocs who are thinking about devoting their careers to understanding and combating Alzheimer’s disease.
The newly proposed NIA-AA research framework for AD represents a radical departure from how we conceive AD, removing the clinical syndrome from the definition of the disease. This paper provides an excellent examination of the assumptions underlying this framework. Specifically, the authors present the main lines of evidence for the amyloid cascade hypothesis, and critically review whether we are justified in basing AD diagnosis on the presence/absence of Aβ alone. The authors argue that this is premature—that we do not yet know the role of Aβ in AD, and there is not sufficient evidence that Aβ is the major player. This is an important contribution to the debate. It will certainly be negative if research is restricted by an incorrect hypothesis about the etiology of AD. Since we do not know the brain mechanisms that lead to AD, the authors advocate a less amyloid-centric, less-biased approach to AD, based on a view of AD as a complex disease and acknowledging the importance of a multitude of factors to disease initiation and progression.
A challenge with the debate about the NIA-AA framework is that we need to be able critically to examine whether the histopathological criteria for AD should set the premise. If Aβ (and tau) define the disease, then the only thing that can be discussed is whether the in vivo biomarkers are good enough and what the cut-off values should be. However, it is possible to go in the opposite direction of the framework, defining AD in terms of the cognitive and clinical symptoms instead of the biomarkers, and then try to understand the multiple pathways and mechanisms that cause them. While it can be argued that the clinical symptoms are not specific to AD, the argument can be turned on its head: The biomarkers are not sufficiently sensitive or specific for the clinical syndrome (McCleery et al., 2018). It makes sense to focus research on what we are interested in preventing or curing—the clinical syndrome—not a biomarker of unknown role. If Aβ were the main upstream cause of AD, it would be reasonable to target it. As convincingly argued by Morris et al., we have almost no clue about the role of Aβ in AD. In this situation, AD is likely better understood as a complex disease with multiple risk factors (Livingston et al., 2017) that cannot be reduced to a single protein, researching Aβ and Aβ-independent (Tse and Herrup, 2017) mechanisms alike.
This is the position advocated by Morris et al. It should be taken seriously in the discussion of how AD is to be defined and understood.
References:
McCleery J, Flicker L, Richard E, Quinn TJ.
When is Alzheimer's not dementia-Cochrane commentary on The National Institute on Ageing and Alzheimer's Association Research Framework for Alzheimer's Disease.
Age Ageing. 2018 Oct 17;
PubMed.
Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, Ballard C, Banerjee S, Burns A, Cohen-Mansfield J, Cooper C, Fox N, Gitlin LN, Howard R, Kales HC, Larson EB, Ritchie K, Rockwood K, Sampson EL, Samus Q, Schneider LS, Selbæk G, Teri L, Mukadam N.
Dementia prevention, intervention, and care.
Lancet. 2017 Jul 19;
PubMed.
Tse KH, Herrup K.
Re-imagining Alzheimer's disease - the diminishing importance of amyloid and a glimpse of what lies ahead.
J Neurochem. 2017 Nov;143(4):432-444. Epub 2017 Jun 21
PubMed.
Karl Herrup University of Pittsburgh School of Medicine
Posted:
This review and commentary is a must-read for anyone interested in entering the field of Alzheimer's research and a solid choice for anyone already working in the field. The documentation is extensive and the arguments are all worth considering.
We should be cautious in making Aβ a required part of the definition of AD.
We should diversify our approaches to studying and treating AD.
We should not abandon our interest in amyloid or tau, but should view them as two of the many facets of a truly multifaceted disease.
Rejecting the amyloid cascade hypothesis does not mean rejecting amyloid as one factor (of many) contributing to the etiology of AD.
What is the alternative to defining AD by plaques and tangles? Opponents of the NIA AA research framework typically take the position that the disease should be defined by the presence of symptoms, not by biomarker (or neuropathologic) evidence of β-amyloid and tau tangles. But precisely what phenotypic presentation should then “define AD”? Even a cursory walk down this path leads quickly to logical dead ends.
If AD were defined as a progressive cognitive impairment that led to dementia, regardless of the domain(s) involved, then many diseases would become “Alzheimer’s disease.” Progressive multiple sclerosis can lead to dementia. Would this then be the “MS variant of Alzheimer’s disease”? The FTLD diseases would be variants of Alzheimer’s disease; carriers of MAPT mutations who became demented would have the “MAPT mutation variant of Alzheimer’s disease”?
A more restricted cognitive definition fares no better in the logic department. If AD were defined as an amnestic impairment with progression to multidomain involvement that led to dementia, then disorders that no reasonable person would label AD would fall under the AD label. An individual with amnestic impairment, hippocampal sclerosis/TDP 43 but with few plaques and Braak 1-2 NFTs would then have the “TDP43 variant of Alzheimer’s disease”? An individual with atrial fibrillation who experienced multiple cerebral embolic events would have the “atrial fibrillation variant of Alzheimer’s disease.” And the opposite side of the argument—a 55-year-old individual with a dysexecutive syndrome and intact memory with florid plaques and tangles (whom most would label EOAD) would not have AD by the amnestic clinic syndrome definition.
Dementia is multifactorial. Alzheimer’s disease (defined as plaques and tangles) is one of several common diseases that can lead to dementia. Most elderly people who are demented have more than one disease. The path forward to discovering cures for dementia is to identify the individual contributing diseases, understand the biology underlying each, and identify treatments for each disease individually. Effective treatments for AD (when they are discovered) will not be the same as treatments for atrial fibrillation, even though both commonly contribute to cognitive impairment in the elderly.
Finally, a common misinterpretation of the NIA AA framework is the idea that β-amyloidosis alone defines AD. This represents a misreading of the NIA AA document. The NIA AA research framework is a rational extension of the NIA AA neuropathologic guidelines published in 2012. Both plaques and tangles are required to define AD, whether that is done at autopsy or in vivo using biomarkers.
Imagine a special, hypothetic type of forest fire. During the fire, black ash is produced. This ash builds up, initially slowly as the fire starts with a few trees, but then more rapidly as more trees catch fire and burn and ignite neighbor trees. At a certain point, the amount of fire (and ash production) saturates as fewer trees ignite than burn out. This produces a sigmoidal curve of ash over time.
The cause of the fire remains unknown. However, some genetically specific trees (let’s say conifers) are known to produce ash more easily, and they also burn more easily. Hence many people assume that the ash could cause the burning.
Anyone who measures the amount of ash during a fire will notice that only fires produce ash. However, sometimes, there is ash without fire, for a variety of reasons (including small fires that have stopped). Some people also notice that the amount of ash is sigmoidal in time, as explained above, and correlates with the progression of the fire. Thus the ash is used as a marker of the forest fire.
The most important risk factor of these special forest fires turns out to be the monotonic increase in heat during the summer. Only warm forests in late summer catch fire, and the risk of catching fire grows rapidly with the temperature.
Any model of the forest fire would need to include the heating as a persistent background parameter, as this phenomenon only arises upon heating. Any model neglecting the heating would be lacking in explanatory power and the quality of the model’s output.
Now AD is the forest fire. The ash is the Aβ plaques. The conifer predisposition is the FAD genes relating to Aβ. The heating during summer is aging.
If one wants to understand what causes forest fire and also measure the onset and progression of the fire more accurately in individual cases, it would be useful not just to monitor the ash, but also the dryness of the soil in the forest, the sun intensity, and the ability of the individual trees to prevent and battle the fire, in particular, the water content of the trees.
The water content of the trees depends on the lifestyle and genetics of the tree in ways that we still need to figure out, but there is lots of emerging evidence of many processes that affect it. Thus forest fires are apparently caused by many factors. It turns out that problems with water content and use of water is the earliest marker of any measure of the forest fire.
What is the water?
References:
Kepp KP.
Ten Challenges of the Amyloid Hypothesis of Alzheimer's Disease.
J Alzheimers Dis. 2017;55(2):447-457.
PubMed.
Kepp KP.
Alzheimer's disease due to loss of function: A new synthesis of the available data.
Prog Neurobiol. 2016 Aug;143:36-60. Epub 2016 Jun 18
PubMed.
Stefano Sensi CAST, Center for Advanced Studies and Technologynal Medicine University “G. d’Annunzio” Chieti-Pescara
Posted:
The paper is excellent reading material for anybody interested in working in the dementia field. A must. The review provides an outstanding array of facts that should be taken into consideration when raising doubts on the central role of amyloid in the disease.
However, one key point should be further emphasized. The reason why a controversial construct like the "amyloid hypothesis" has dominated the field for so long relies on the more general issue of how modern medicine has failed to produce a full transition from the Oslerian reductionist approach of the past century. The poor heuristic value of the amyloid construct depends on a more general faulty assumption of what a "disease" is. A growing and exciting body of evidence indicates that diseases result from the unique/individual default of multiple systems and networks that are in control of the health status. In other words, each one of us will eventually be affected by a "personalized" disease that results from the distinct contribution of factors that depend on our own genetic blueprint and its age-dependent modulation promoted by the environment and our life experience (Greene et al., 2017). Thus, in line with a more modern, personalized, and realist view, we need to put to rest the whole concept of magic "silver bullets." We need to reconcile ourselves to the fact that complex, non-transmissible chronic conditions cannot be treated with a single drug approach. This new scenario is definitely challenging, as the complexity of diagnosis and treatments reach levels that are difficult to master by the individual clinician, but offers the possibility of exploring entirely different avenues. Artificial intelligence and unbiased approaches will be of great help in venturing into these new territories. The future of medicine is there.
References:
Greene JA, Loscalzo J.
Putting the Patient Back Together - Social Medicine, Network Medicine, and the Limits of Reductionism.
N Engl J Med. 2017 Dec 21;377(25):2493-2499.
PubMed.
The debate over the criteria for “Alzheimer’s disease” recalls the evolution of those for two other ailments, “pneumonia” and “diarrhea.” At first each was defined mainly by its clinically notable signs and symptoms. Hence each broad term provided interim satisfaction to clinicians and families, but postponed identifying the several agents and mechanisms which could produce similar effects in the lung or intestine.
When laboratory tests became available, additional criteria were added to increase specificity. Stool and sputum samples are cultured to ascertain the particular pathogen, or the absence of any. Thus, we now distinguish diarrhea of salmonella from that of shigella and cholera, and the pneumonia of influenza from that of the pneumococcus and tuberculosis.
It seems that affixing precise criteria and boundaries to the definition of Alzheimer’s and its stages will await the development of generally accepted laboratory tests which reflect initial and progressive brain damage. Meanwhile, why not agree on rough-hewn interim criteria, while recognizing their limitations?
Comments
The University of Adelaide
This is a detailed analysis of the role of Aβ in Alzheimer’s disease and the justification for defining Alzheimer’s disease by the presence of Aβ, with supplementary information to extend particular arguments. It traces the historical development of thinking regarding the disease(s) and calls for open-mindedness and objectivity in research. It suggests a way forward to avoid the logical trap that the field appears to have fallen into. In my opinion, this paper should be essential reading for postgraduate students or postdocs who are thinking about devoting their careers to understanding and combating Alzheimer’s disease.
University of Oslo, Norway
The newly proposed NIA-AA research framework for AD represents a radical departure from how we conceive AD, removing the clinical syndrome from the definition of the disease. This paper provides an excellent examination of the assumptions underlying this framework. Specifically, the authors present the main lines of evidence for the amyloid cascade hypothesis, and critically review whether we are justified in basing AD diagnosis on the presence/absence of Aβ alone. The authors argue that this is premature—that we do not yet know the role of Aβ in AD, and there is not sufficient evidence that Aβ is the major player. This is an important contribution to the debate. It will certainly be negative if research is restricted by an incorrect hypothesis about the etiology of AD. Since we do not know the brain mechanisms that lead to AD, the authors advocate a less amyloid-centric, less-biased approach to AD, based on a view of AD as a complex disease and acknowledging the importance of a multitude of factors to disease initiation and progression.
A challenge with the debate about the NIA-AA framework is that we need to be able critically to examine whether the histopathological criteria for AD should set the premise. If Aβ (and tau) define the disease, then the only thing that can be discussed is whether the in vivo biomarkers are good enough and what the cut-off values should be. However, it is possible to go in the opposite direction of the framework, defining AD in terms of the cognitive and clinical symptoms instead of the biomarkers, and then try to understand the multiple pathways and mechanisms that cause them. While it can be argued that the clinical symptoms are not specific to AD, the argument can be turned on its head: The biomarkers are not sufficiently sensitive or specific for the clinical syndrome (McCleery et al., 2018). It makes sense to focus research on what we are interested in preventing or curing—the clinical syndrome—not a biomarker of unknown role. If Aβ were the main upstream cause of AD, it would be reasonable to target it. As convincingly argued by Morris et al., we have almost no clue about the role of Aβ in AD. In this situation, AD is likely better understood as a complex disease with multiple risk factors (Livingston et al., 2017) that cannot be reduced to a single protein, researching Aβ and Aβ-independent (Tse and Herrup, 2017) mechanisms alike.
This is the position advocated by Morris et al. It should be taken seriously in the discussion of how AD is to be defined and understood.
References:
McCleery J, Flicker L, Richard E, Quinn TJ. When is Alzheimer's not dementia-Cochrane commentary on The National Institute on Ageing and Alzheimer's Association Research Framework for Alzheimer's Disease. Age Ageing. 2018 Oct 17; PubMed.
Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, Ballard C, Banerjee S, Burns A, Cohen-Mansfield J, Cooper C, Fox N, Gitlin LN, Howard R, Kales HC, Larson EB, Ritchie K, Rockwood K, Sampson EL, Samus Q, Schneider LS, Selbæk G, Teri L, Mukadam N. Dementia prevention, intervention, and care. Lancet. 2017 Jul 19; PubMed.
Tse KH, Herrup K. Re-imagining Alzheimer's disease - the diminishing importance of amyloid and a glimpse of what lies ahead. J Neurochem. 2017 Nov;143(4):432-444. Epub 2017 Jun 21 PubMed.
University of Pittsburgh School of Medicine
This review and commentary is a must-read for anyone interested in entering the field of Alzheimer's research and a solid choice for anyone already working in the field. The documentation is extensive and the arguments are all worth considering.
Mayo Clinic
What is the alternative to defining AD by plaques and tangles? Opponents of the NIA AA research framework typically take the position that the disease should be defined by the presence of symptoms, not by biomarker (or neuropathologic) evidence of β-amyloid and tau tangles. But precisely what phenotypic presentation should then “define AD”? Even a cursory walk down this path leads quickly to logical dead ends.
If AD were defined as a progressive cognitive impairment that led to dementia, regardless of the domain(s) involved, then many diseases would become “Alzheimer’s disease.” Progressive multiple sclerosis can lead to dementia. Would this then be the “MS variant of Alzheimer’s disease”? The FTLD diseases would be variants of Alzheimer’s disease; carriers of MAPT mutations who became demented would have the “MAPT mutation variant of Alzheimer’s disease”?
A more restricted cognitive definition fares no better in the logic department. If AD were defined as an amnestic impairment with progression to multidomain involvement that led to dementia, then disorders that no reasonable person would label AD would fall under the AD label. An individual with amnestic impairment, hippocampal sclerosis/TDP 43 but with few plaques and Braak 1-2 NFTs would then have the “TDP43 variant of Alzheimer’s disease”? An individual with atrial fibrillation who experienced multiple cerebral embolic events would have the “atrial fibrillation variant of Alzheimer’s disease.” And the opposite side of the argument—a 55-year-old individual with a dysexecutive syndrome and intact memory with florid plaques and tangles (whom most would label EOAD) would not have AD by the amnestic clinic syndrome definition.
Dementia is multifactorial. Alzheimer’s disease (defined as plaques and tangles) is one of several common diseases that can lead to dementia. Most elderly people who are demented have more than one disease. The path forward to discovering cures for dementia is to identify the individual contributing diseases, understand the biology underlying each, and identify treatments for each disease individually. Effective treatments for AD (when they are discovered) will not be the same as treatments for atrial fibrillation, even though both commonly contribute to cognitive impairment in the elderly.
Finally, a common misinterpretation of the NIA AA framework is the idea that β-amyloidosis alone defines AD. This represents a misreading of the NIA AA document. The NIA AA research framework is a rational extension of the NIA AA neuropathologic guidelines published in 2012. Both plaques and tangles are required to define AD, whether that is done at autopsy or in vivo using biomarkers.
Technical University of Denmark
The forest fire analogy
Imagine a special, hypothetic type of forest fire. During the fire, black ash is produced. This ash builds up, initially slowly as the fire starts with a few trees, but then more rapidly as more trees catch fire and burn and ignite neighbor trees. At a certain point, the amount of fire (and ash production) saturates as fewer trees ignite than burn out. This produces a sigmoidal curve of ash over time.
The cause of the fire remains unknown. However, some genetically specific trees (let’s say conifers) are known to produce ash more easily, and they also burn more easily. Hence many people assume that the ash could cause the burning.
Anyone who measures the amount of ash during a fire will notice that only fires produce ash. However, sometimes, there is ash without fire, for a variety of reasons (including small fires that have stopped). Some people also notice that the amount of ash is sigmoidal in time, as explained above, and correlates with the progression of the fire. Thus the ash is used as a marker of the forest fire.
The most important risk factor of these special forest fires turns out to be the monotonic increase in heat during the summer. Only warm forests in late summer catch fire, and the risk of catching fire grows rapidly with the temperature.
Any model of the forest fire would need to include the heating as a persistent background parameter, as this phenomenon only arises upon heating. Any model neglecting the heating would be lacking in explanatory power and the quality of the model’s output.
Now AD is the forest fire. The ash is the Aβ plaques. The conifer predisposition is the FAD genes relating to Aβ. The heating during summer is aging.
If one wants to understand what causes forest fire and also measure the onset and progression of the fire more accurately in individual cases, it would be useful not just to monitor the ash, but also the dryness of the soil in the forest, the sun intensity, and the ability of the individual trees to prevent and battle the fire, in particular, the water content of the trees.
The water content of the trees depends on the lifestyle and genetics of the tree in ways that we still need to figure out, but there is lots of emerging evidence of many processes that affect it. Thus forest fires are apparently caused by many factors. It turns out that problems with water content and use of water is the earliest marker of any measure of the forest fire.
What is the water?
References:
Kepp KP. Ten Challenges of the Amyloid Hypothesis of Alzheimer's Disease. J Alzheimers Dis. 2017;55(2):447-457. PubMed.
Kepp KP. Alzheimer's disease due to loss of function: A new synthesis of the available data. Prog Neurobiol. 2016 Aug;143:36-60. Epub 2016 Jun 18 PubMed.
CAST, Center for Advanced Studies and Technologynal Medicine University “G. d’Annunzio” Chieti-Pescara
The paper is excellent reading material for anybody interested in working in the dementia field. A must. The review provides an outstanding array of facts that should be taken into consideration when raising doubts on the central role of amyloid in the disease.
However, one key point should be further emphasized. The reason why a controversial construct like the "amyloid hypothesis" has dominated the field for so long relies on the more general issue of how modern medicine has failed to produce a full transition from the Oslerian reductionist approach of the past century. The poor heuristic value of the amyloid construct depends on a more general faulty assumption of what a "disease" is. A growing and exciting body of evidence indicates that diseases result from the unique/individual default of multiple systems and networks that are in control of the health status. In other words, each one of us will eventually be affected by a "personalized" disease that results from the distinct contribution of factors that depend on our own genetic blueprint and its age-dependent modulation promoted by the environment and our life experience (Greene et al., 2017). Thus, in line with a more modern, personalized, and realist view, we need to put to rest the whole concept of magic "silver bullets." We need to reconcile ourselves to the fact that complex, non-transmissible chronic conditions cannot be treated with a single drug approach. This new scenario is definitely challenging, as the complexity of diagnosis and treatments reach levels that are difficult to master by the individual clinician, but offers the possibility of exploring entirely different avenues. Artificial intelligence and unbiased approaches will be of great help in venturing into these new territories. The future of medicine is there.
References:
Greene JA, Loscalzo J. Putting the Patient Back Together - Social Medicine, Network Medicine, and the Limits of Reductionism. N Engl J Med. 2017 Dec 21;377(25):2493-2499. PubMed.
The debate over the criteria for “Alzheimer’s disease” recalls the evolution of those for two other ailments, “pneumonia” and “diarrhea.” At first each was defined mainly by its clinically notable signs and symptoms. Hence each broad term provided interim satisfaction to clinicians and families, but postponed identifying the several agents and mechanisms which could produce similar effects in the lung or intestine.
When laboratory tests became available, additional criteria were added to increase specificity. Stool and sputum samples are cultured to ascertain the particular pathogen, or the absence of any. Thus, we now distinguish diarrhea of salmonella from that of shigella and cholera, and the pneumonia of influenza from that of the pneumococcus and tuberculosis.
It seems that affixing precise criteria and boundaries to the definition of Alzheimer’s and its stages will await the development of generally accepted laboratory tests which reflect initial and progressive brain damage. Meanwhile, why not agree on rough-hewn interim criteria, while recognizing their limitations?
Make a Comment
To make a comment you must login or register.