Does Dementia Have a Microbial Cause?
Round 1
Reviewer 1 Report
This review summarized about the relationship between microbiota and Alzheimer’s disease (AD). This review is well organized. I have some comments.
The introduction is relatively long. Authors had better describe mainly the amyloid, inflammatory, and microbial hypotheses.
The relationship between brain-gut axis dysfunction and dementia has attracted attention in recent years. Shouldn’t the relationship be described?
Please spell out the words “CDC”.
Author Response
We would like to thank the reviewer for insightful comments. The responses to each concern are shown in italics.
The introduction is relatively long. Authors had better describe mainly the amyloid, inflammatory, and microbial hypotheses.
Thank you. We have now reduced the length of the Introduction, without losing any essential content.
The relationship between brain-gut axis dysfunction and dementia has attracted attention in recent years. Shouldn’t the relationship be described?
We agree and thank you for pointing this out. It may actually bridge the hypotheses. We have now included a section on this with appropriate references.
Please spell out the words “CDC”. This has been corrected.
Reviewer 2 Report
First of all, I thank the authors for reviewing a controversial etiologic hypothesis of Alzheimer's disease.
I point out that from my point of view the quality of the article differs significantly between the explanations provided for the infectious hypothesis of Alzheimer's disease (high) and the rest of the paper. I detail below the aspects that I believe must be improved in order to be able to consider the option of publishing this work:
1) In the title: the logical thing would be for Alzheimer's disease to appear. There is hardly any mention of other etiologies of neurodegenerative dementias. It would be more realistic with the text. In turn, I would propose deleting mentions of other neurodegenerative diseases or the opposite, to increase the mention of them, but I consider that the information provided in the actual form is not useful for the readers.
2) The methodology section: more information about search strategy could be of interest. Exclusion and inclusion criteria of the papers? MesH terms were used? Pubmed? WOS?... It is not a systematic review but this information could increase the perception of the quality of the paper.
3) Too many statements don't have enough references such us epidemiological statements in the introduction section.
4) Terminology used could be improvable and there are some error in its application. For instance: Early Onset Alzheimer's disease (EOAD) refers to cases with onset earlier than 65 years, could be not only due to Autosomal Dominant Alzheimer's disease (ADAD) but also due to sporadic cases or duet to DSAD (Down syndrome related Alzheimer's disease). On the other hand, a better explanation of the differences among risk factors and well known causative mutations or APP triplication should be provided.
5) Fortunately, the diagnosis strategy in Mild Cognitive Impairment (MCI) context has been modified. Currently the diagnosis using biomarkers is advisable. No references for CSF biomarkers are done and they are wildly used in clinical practice for AD diagnosis with a very high negative diagnostic predictive value (higher than other diagnostic tools referred in the text, cheaper,... and is not so clearly mentioned in the manuscript the utility not only for research purposes but also in clinical practice of the other diagnostic tools used in the MCI etiological diagnosis).
6) Figure 1 must be improved. The AD pathology is diffuse, check Braak and Braak stages for instance. The diagnosis in symptomatic stages does not occur when the amyloid and tau pathology is limited to an unique brain region. A picture of the microscopic changes could be of interest to try to improve the picture as well to add a summary of the diagnostic tools and the relation with macro and microscopic changes in AD.
7) The most frequent form of present of AD is progressive multidomain amnesic cognitive impairment impairment, unidomain amnesic is much more infrequent. Atypical presentations (not mentioned in the manuscript) include: posterior cortical atrophy related, progressive aphasia...
8) A better explanation than brain dysfunction terms is advisable. Explain better the disturbances described in neuronal signaling for instance.
9) The statement that the protein deposited in the brain is a specific of clinical phenotype is not true. This must be corrected. For instance, the coexistence of AD (amyloid plaques and neurofibrillary tangles) with Lewy bodies is very frequent; MSA and PD are synucleinopathies; progressive aphasia could be due to AD pathology, FTD and the latter due to tau or not tau pathology...).
10) Even if the amyloid etiologic hypothesis has its limitations it is more plausible (at least with the current information) that infectious hypothesis of AD and this is not so clearly mentioned in the manuscript. The first drug that seems to be able to modify the AD pathology checked by PET amyloid (the clinical benefit is not so clear) is Aducanumab. A mention that it has been recently approved by the FDA (not by EMA) and its suggested potential benefits, as well as the same temporal changes in ADAD, DSAD and LOAD in different biomarkers also suggest that amyloid is crucial in the AD progression. Mentioning the long preclinical stage AD should be done.
11) Other etiological hypothesis are ignored. Why? Vascular hypothesis hypotheses? (Try to improve or explain the selection criteria better in the methodology section).
12) Reorganize the evidence of the infectious etiologies as possible risk factors for the development of AD. A logical distribution could be: a) information from cell cultures or animal models; b) humans without diagnose performed using core biomarkers; c) from humans diagnosed with AD biomarkers or with anatomophatological confirmation; d) information about their suggested potential benefits? (examples from animal models and later in humans, for instance antibiotics have been approved as clinical trials for AD pathology). A summary table with all these information could be of interest.
13) Herpes virus could also increase the probability of AD dementia due to modifications in excitatory-inhibitory regulation. Risk of epilepsy is increased in the context of herpes virus infection and the epilepsy itself it is well known risk factor for AD? Indeed the presence of APOE E4 alleles a risk factor for both epilepsy (including late onset) and AD.
14) A table that summarizes the potential new explanations of this suggested etiological hypothesis compared to others could be also of interest.
Author Response
To begin, we would like to thank this reviewer for thoughtful comments and clear expertise in the topic area. We have made substantial changes to the manuscript and it is now much improved. We have responded to each concern with the description of changes shown in italics.
I point out that from my point of view the quality of the article differs significantly between the explanations provided for the infectious hypothesis of Alzheimer's disease (high) and the rest of the paper. I detail below the aspects that I believe must be improved in order to be able to consider the option of publishing this work:
1) In the title: the logical thing would be for Alzheimer's disease to appear. There is hardly any mention of other etiologies of neurodegenerative dementias. It would be more realistic with the text. In turn, I would propose deleting mentions of other neurodegenerative diseases or the opposite, to increase the mention of them, but I consider that the information provided in the actual form is not useful for the readers.
The purpose of the review certainly begs more clarity. As such, we have limited discussion of non-dementia neurodegenerative disease and focused on the types of dementia. Given the importance of evidence supporting microbial involvement in non -AD dementia (eg Lewy body), we have refined terms throughout where appropriate to include "dementia" versus Ad or simply "Alzheimers and related dementias." In addition, we have described clearly the objective of the review.
2) The methodology section: more information about search strategy could be of interest. Exclusion and inclusion criteria of the papers? MesH terms were used? Pubmed? WOS?... It is not a systematic review but this information could increase the perception of the quality of the paper.
After revising the manuscript to better conform to the specifications of the journal for a Review article, we have eliminated the Methods section. In addition, this is not a systematic review, so any confusing language has been removed for transparency.
3) Too many statements don't have enough references such us epidemiological statements in the introduction section.
This is understood, and as such, we have included more relevant references into the Introduction section. Examples are references 5 and 6.
4) Terminology used could be improvable and there are some error in its application. For instance: Early Onset Alzheimer's disease (EOAD) refers to cases with onset earlier than 65 years, could be not only due to Autosomal Dominant Alzheimer's disease (ADAD) but also due to sporadic cases or duet to DSAD (Down syndrome related Alzheimer's disease). On the other hand, a better explanation of the differences among risk factors and well known causative mutations or APP triplication should be provided.
Thank you for pointing this out. The descriptions of case types and risks are now described in lines 114-122.
5) Fortunately, the diagnosis strategy in Mild Cognitive Impairment (MCI) context has been modified. Currently the diagnosis using biomarkers is advisable. No references for CSF biomarkers are done and they are wildly used in clinical practice for AD diagnosis with a very high negative diagnostic predictive value (higher than other diagnostic tools referred in the text, cheaper,... and is not so clearly mentioned in the manuscript the utility not only for research purposes but also in clinical practice of the other diagnostic tools used in the MCI etiological diagnosis).
Thank you. This is important and more relevant discussion of diagnosis is now included. This can be found in lines 81-84.
6) Figure 1 must be improved. The AD pathology is diffuse, check Braak and Braak stages for instance. The diagnosis in symptomatic stages does not occur when the amyloid and tau pathology is limited to an unique brain region. A picture of the microscopic changes could be of interest to try to improve the picture as well to add a summary of the diagnostic tools and the relation with macro and microscopic changes in AD.
Agreed. The Figure is too simplified, esp. for AD. As such, we have included a panel of the Braak stages and temporal changes associated with affected areas.
7) The most frequent form of present of AD is progressive multidomain amnesic cognitive impairment impairment, unidomain amnesic is much more infrequent. Atypical presentations (not mentioned in the manuscript) include: posterior cortical atrophy related, progressive aphasia...
This is indeed important to include. The typical and atypical presentations are now included in lines 63-72.
8) A better explanation than brain dysfunction terms is advisable. Explain better the disturbances described in neuronal signaling for instance.
This is an important inclusion, especially where the pathophysiology is concerned. . We have added a description and reference in lines 145-148.
9) The statement that the protein deposited in the brain is a specific of clinical phenotype is not true. This must be corrected. For instance, the coexistence of AD (amyloid plaques and neurofibrillary tangles) with Lewy bodies is very frequent; MSA and PD are synucleinopathies; progressive aphasia could be due to AD pathology, FTD and the latter due to tau or not tau pathology...).
This is also very important--thank you. We have now included a description of Lewy body dementia, including the pathologic characteristics, and included a more comprehensive description of the protein deposition. Relevant text includes lines 100-113 and 129-134.
10) Even if the amyloid etiologic hypothesis has its limitations it is more plausible (at least with the current information) that infectious hypothesis of AD and this is not so clearly mentioned in the manuscript. The first drug that seems to be able to modify the AD pathology checked by PET amyloid (the clinical benefit is not so clear) is Aducanumab. A mention that it has been recently approved by the FDA (not by EMA) and its suggested potential benefits, as well as the same temporal changes in ADAD, DSAD and LOAD in different biomarkers also suggest that amyloid is crucial in the AD progression. Mentioning the long preclinical stage AD should be done.
This has been included under the amyloid hypothesis. it is interesting to consider the potential of "amyloid-busters" for their prophylactic potential.
11) Other etiological hypothesis are ignored. Why? Vascular hypothesis hypotheses? (Try to improve or explain the selection criteria better in the methodology section).
While the focus is intended to be on the microbial hypothesis, this has been indicated by 2 reviewers. As such, we have included a section on the gut-brain axis and the permeability of the blood brain barrier. This ties in the microbial and inflammatory hypotheses, actually , and is important to include.
12) Reorganize the evidence of the infectious etiologies as possible risk factors for the development of AD. A logical distribution could be: a) information from cell cultures or animal models; b) humans without diagnose performed using core biomarkers; c) from humans diagnosed with AD biomarkers or with anatomophatological confirmation; d) information about their suggested potential benefits? (examples from animal models and later in humans, for instance antibiotics have been approved as clinical trials for AD pathology). A summary table with all these information could be of interest.
We have certainly made an effort to include the lines of evidence for the microbial hypothesis. Currently, the manuscript organizes by pathogen because the experimental evidence differs for each. The goal of Figure 6 is to synthesize the hypothesis in a way that the suggested Table would do.
13) Herpes virus could also increase the probability of AD dementia due to modifications in excitatory-inhibitory regulation. Risk of epilepsy is increased in the context of herpes virus infection and the epilepsy itself it is well known risk factor for AD? Indeed the presence of APOE E4 alleles a risk factor for both epilepsy (including late onset) and AD.
This is important to include, and can now be found in lines 302-307.
14) A table that summarizes the potential new explanations of this suggested etiological hypothesis compared to others could be also of interest.
We have utilized figures instead of Tables for this purpose, but the point is well taken.
Reviewer 3 Report
1) The introduction section about AD and its pathogenesis is too long, I suggest to give space just to microbial pathogenesis.
2) A precise objective section is lacking
3) The materials and methods section does not clarify: i) the database used for the research; ii) the keywords used for the research; iii) inclusion and exclusion criteria for the articles selected. If not willing to add such information, it should be better to change the review type to "narrative" review.
4) The results section appears like a discussion one. I suggest to add a table showing the articles authors found for each pathogenetic microbe. There is no mention about the quality of the evidence among the articles included.
5) The conclusion section is too long and includes new hypothesis, not mentioned during the result section. It is better to add a "discussion" section where to write such hypothesis.
Author Response
We would like to thank the reviewer for thoughtful comments. We have responded to each and the suggestions made have improved the manuscript significantly. All responses are shown in italics.
1) The introduction section about AD and its pathogenesis is too long, I suggest to give space just to microbial pathogenesis.
Thank you for this suggestion. We have reduced the Introduction and confined the Clinical and Pathological Characteristics to a separate section.
2) A precise objective section is lacking
Indeed, and thank you for pointing this out. The objective is now stated at the end of the Introduction, lines 74-76.
3) The materials and methods section does not clarify: i) the database used for the research; ii) the keywords used for the research; iii) inclusion and exclusion criteria for the articles selected. If not willing to add such information, it should be better to change the review type to "narrative" review.
After revising the manuscript to better conform to the specifications of the journal for a Review article, we have eliminated the Methods section. In addition, this is not a systematic review, so any confusing language has been removed for transparency.
4) The results section appears like a discussion one. I suggest to add a table showing the articles authors found for each pathogenetic microbe. There is no mention about the quality of the evidence among the articles included.
This is a valid point and well-received. We elected to organize the review by pathogens implicated, and further confining it to those with multiple studies to support the microbial interaction. The pathogens are listed in Figure 3, with Figure 4 showing the possible mechanism for HSV because this virus has the largest body of supportive evidence for AD development.
5) The conclusion section is too long and includes new hypothesis, not mentioned during the result section. It is better to add a "discussion" section where to write such hypothesis.
Thank you. We have re-organized the manuscript accordingly, to include separate sections for Conclusions and Future directions.
Round 2
Reviewer 2 Report
I thank the authors for modifying the manuscript according to the suggestions that reviewers including myself have done to them. Currently, I believe that could be considered to be published in the present form.
Reviewer 3 Report
I feel that the authors have made a good revision and now the review is fine
