Pathogenesis, Diagnosis and Risk Stratification in Arrhythmogenic Cardiomyopathy

Round 1

Reviewer 1 Report

The narrative review by Florio et al. reviews the current literature on arrythmogenic cardiomyopathies focussing on pathomechanisms as well as diagnostic and therapeutic modalities. The manuscript is well written however appears unusual lengthy and somewhat difficult to follow given the amount of information presented - rather in a descriptive than structured way. Given the title, it is surprising that the authors appear to give a complete overview of the disease instead of highlighting new developments and therapeutic options. It is concerning that some figure appear to be copied from already published manuscripts without editing and/or consent of the authors/journals.

In addition I have a few remarks: 

1. Given this extensive overview of arrythmogenic cardiomyopathies the authors should find a way to condense the information presented (tables, figures). In addition I am surprised that - in contrast to what the title implies - a complete overview of the disease is given. The authors should also consider brief summaries at the end of each section. 
2. It is concerning that many figure are taken from other journals without consent.  In addition, Figure 1 and 2 have a typical nature layout and may not be allowed to be published without editing (even when permission of the authors exist)
3. Figure 3 is not entierely clear to me. The upper part of the figure deals with potential cause of myocarditis and the lower part with a vicious circle that leads to fibrofatty replacement. Are these pathomechanisms connected? If not the authors should find a way to separate these mechanisms visually to demonstrate that there are different etiologies. 
4. Figure 4 is taken from Corrado D et al. Diagnosis of arrhythmogenic cardiomyopathy: The Padua criteria. Int. J. Cardiol. 2020.  however not referenced to it 
5. Since the authors refer to the ARVC diagnostic-criteria these should at least be presented in a table 

Author Response

1. Q: Given this extensive overview of arrythmogenic cardiomyopathies the authors should find a way to condense the information presented (tables, figures). In addition I am surprised that - in contrast to what the title implies - a complete overview of the disease is given. The authors should also consider brief summaries at the end of each section.

A: We have rearranged the section trying to focus our attention to the main topic and using some adjunctive tables and figures to explain and condense the information.

2. Q: It is concerning that many figure are taken from other journals without consent. In addition, Figure 1 and 2 have a typical nature layout and may not be allowed to be published without editing (even when permission of the authors exist).

A: A suggested, we remove the figures taken from other journals (Figure 1 and 2) and used instead an original Figure (Figure 1).

3. Q: Figure 3 is not entierely clear to me. The upper part of the figure deals with potential cause of myocarditis and the lower part with a vicious circle that leads to fibrofatty replacement. Are these pathomechanisms connected? If not the authors should find a way to separate these mechanisms visually to demonstrate that there are different etiologies.

A: By mistake we put into the original manuscript figure 4 instead of figure 3 and vice versa. That’s been solved into the reviewed manuscript.

4. Q: Figure 4 is taken from Corrado D et al. Diagnosis of arrhythmogenic cardiomyopathy: The Padua criteria. J. Cardiol. 2020. however not referenced to it.

A: Original Figure 4 (now renamed Figure 2) has been cited has suggested.

5. Q: Since the authors refer to the ARVC diagnostic-criteria these should at least be presented in a table.
   A: We presented the ARVC Diagnostic Criteria into Table 1.

Reviewer 2 Report

This manuscript has been designed to review the current state-of-the-art of arrhythmogenic cardiomyopathy regarding its pathophysiology, diagnostic issues and risk stratification strategies.

The herein presented work is exhaustive, updated, well written and illustrated with nice figures taken from the cited papers. Moreover, this review introduces valuable critical comments on some of the papers cited.

Several minor points could be further improved: 

• At the Pathophysiology section (page 2 onwards) no mention on Wnt and Hippo pathways is provided, only a too brief comment in the Figure 1 Footnote. Since GSK3B pathway is mentioned when reviewing the role of inflammation (page 9, line 348) readers may not understand the consequences of GSK3B inhibition unless some sort of explanation connecting GSK3B and Wnt pathway is briefly included.
• Also at the Pathophysiology section, when dealing with the different genes causing ACM a short comment on how often big rearrangements are found in ACM patients will be interesting since they might not be picked up in routine sequencing studies. This comment would underline the usefulness of CNVs analyses and/or MLPA studies in some cases.
• There is controversy regarding the real pathogenic role of mutations in RyR2 gene in the scenario of ACM.
  From a clinical stand point, no clearly pathogenic mutation has already been identified in ACM families with a clear co-segregation study in several generations demonstrating its connection to the ACM phenotype, as far as I am concerned. The authors first introduce this gene as a cause of ACM (page 3-4, lines 118 onwards) whereas later on in the manuscript they acknowledge RyR2 may actually cause ACM phenocopies (not real ACM phenotypes) (page 4 lines 149-150). Actually, the re-consideration of RyR2 mutations as ACM phenocopy and not ACM-causing gene in 2011 came from the same group of researchers who in 2001 released the first original paper pointing to the RyR2 gene in connection to ACM (Thiene G and Basso C and their group) (1,2). They state that they actually caused CPVT phenotypes with only minor contraction abnormalities, not enough to maintain the initial diagnosis of ACM. These two papers (2001 and 2011) are cited by Campuzano et al which is the paper cited by the authors in the herein presented manuscript as the source of the comment in the Footnote of Figure 2. Additionally, other papers have searched for RyR2 mutations in ACM probands yielding in 9% of RyR2 rare variants but no co-segregation are provided (3), and maybe not identified variants in other genes might be playing a significant role. All in all, not compelling evidence can now be gathered to confirm RyR2 as an ACM-causing gene, and probably it should be considered as a candidate gene until more clinical evidence is released. 
• The authors of the herein presented manuscript give compelling in vitro evidence about the abnormalities in CPVT important proteins and calcium homeostasis caused by ACM mutations. In keeping with that, recent evidence confirm some sort of RyR2 dysfunction in ACM in vitro models (4). Thus, these pieces of evidence may confirm an overlap in the pathogenesis of ventricular arrhythmia with undistinguishable polymorphic stress-triggered ventricular tachycardia since calcium mishandling is present in both entities but it doesn’t mean that RyR2 mutations are capable of inducing an ouvert structural ACM phenotype. Mutations in other structural genes have also been found to cause calcium misshandling, such as RBM20.

In order to avoid more confusion, the authors should be consistent on the pathogenic role they give to RyR2 mutations to produce an ACM phenotype throughout their manuscript and, in case it is not fully clarified, it should be clearly stated.

• When describing the current evidence regarding autoimmunity (page 9 line 358 onwards) in ACM pathophysiology, besides autoantibodies against DSG2, anti heart and anti intercalated disks autoantibodies, also some short comment could be included regarding antimitochondrial autoantibodies. 
• Cites should be introduced in page 4 line 121 when the authors mention that some papers have linked RyR2 mutations R176Q, L433P, N2386I and T2504M to a ACM phenotype, unless all of them have been studied in the reference 23 later on introduced.
• Typing mistakes: page 3 line 108 celladhesion (istead of cell adhesion, with or without -), page 4 line 129 hIPSC-CMs (instead of hiPSC-CMs), page 4 line 150 FNLC (instead of FLNC), Page 13 line 551 comlication instead of complication.
• Some specific mutations are fully mentioned throughout the manuscript but, when mentioning PLN and TMEM43 genes only the genes are cited even though only scarce mutations have been clearly associated with the ACM phenotype and should be fully mentioned, such as PLN R14del and TMEM43 S358L.
• The Footnote of Figure 4 does not have much to do wit the figure itself, it appears to be some mistake here. 
• Several times throughout the manuscript appears misspelled the term triggerring instead of the correct version triggering.
• Page 10 line 421, when dealing this the electrocardiographical signs of ACM and thee diagnostic tools, it will be welcome a brief comment on thee usefulneess of SAECG and Fontaine leads nowadays.
• When the authors refer to epsilon waves Page 11 line 426 onwards) they sometimes employ other terms such as “epsilon potentials” and epsilon potentials. It is desirable to use always the same term.
• Page 11 line 436-438, some sort of explanation is needed here, at least a cite.
• Page 11 line 483, similarly to what has been mentioned with echocardiography, a short comment on the advances on strain evaluation with CMR in ACM will be interesting to be added here, both on RV and LV strain.
• Throughout the manuscript International Task Force Criteria are abbreviated as ITFC, except in page 11 line 454 where is is used TF.
• Page 15 line 584, authors should make clear that the abovementioned risk score model and its validation by Aquaro were designed ONLY on patients fulfilling definite ITFC of ACM and thus, it remains unknown whether this score is similarly accurate for LD-ACM forms which do not fulfill ITFC as definite ACM. Indeed, Aquaro et al also published another paper in 2020 stressing that this score is valid for lone-RV ACM but it understimated the risk in ACM with LV involvement.

1. Tiso N, Stephan DA, Nava A, Bagattin A, Devaney JM, Stanchi F, Larderet G, Brahmbhatt B, Brown K, Bauce B, Muriago M, Basso C, Thiene G, Danieli GA, Rampazzo A. Identification of mutations in the cardiac ryanodine receptor gene in families affected with arrhythmogenic right ventricular cardiomyopathy type 2 (ARVD2). Hum Mol Genet 2001;10:189–194.
2. Basso C, Bauce B, Corrado D, Thiene G. Pathophysiology of arrhythmogenic cardiomyopathy. Nat Rev Cardiol 2011;9:223–33.
3. Roux-Buisson N et al. Prevalence and significance of rare RYR2 variants in arrhythmogenic right ventricular cardiomyopathy/dysplasia: results of a systematic screening. Heart Rhythm 2014 Nov;11(11):1999-2009. 
4. Wang Y, et al. Integrin beta1D Deficiency-Mediated RyR2 Dysfunction Contributes to Catecholamine-Sensitive Ventricular Tachycardia in Arrhythmogenic Right Ventricular Cardiomyopathy. Circulation. 2020 May 5;141(18):1477-1493

Author Response

1. A: At the Pathophysiology section (page 2 onwards) no mention on Wnt and Hippo pathways is provided, only a too brief comment in the Figure 1 Footnote. Since GSK3B pathway is mentioned when reviewing the role of inflammation (page 9, line 348) readers may not understand the consequences of GSK3B inhibition unless some sort of explanation connecting GSK3B and Wnt pathway is briefly included.

Q: We included the Wnt and Hippo pathway into the section and provided a dedicated figure (Figure 1).

2. A: Also at the Pathophysiology section, when dealing with the different genes causing ACM a short comment on how often big rearrangements are found in ACM patients will be interesting since they might not be picked up in routine sequencing studies. This comment would underline the usefulness of CNVs analyses and/or MLPA studies in some cases.

Q: A comment on big rearrangements have been inserted into paragraph 2.3.

3. A: There is controversy regarding the real pathogenic role of mutations in RyR2 gene in the scenario of ACM: From a clinical stand point, no clearly pathogenic mutation has already been identified in ACM families with a clear co-segregation study in several generations demonstrating its connection to the ACM phenotype, as far as I am concerned. The authors first introduce this gene as a cause of ACM (page 3-4, lines 118 onwards) whereas later on in the manuscript they acknowledge RyR2 may actually cause ACM phenocopies (not real ACM phenotypes) (page 4 lines 149-150). Actually, the re-consideration of RyR2 mutations as ACM phenocopy and not ACM-causing gene in 2011 came from the same group of researchers who in 2001 released the first original paper pointing to the RyR2 gene in connection to ACM (Thiene G and Basso C and their group) (1,2). They state that they actually caused CPVT phenotypes with only minor contraction abnormalities, not enough to maintain the initial diagnosis of ACM. These two papers (2001 and 2011) are cited by Campuzano et al which is the paper cited by the authors in the herein presented manuscript as the source of the comment in the Footnote of Figure 2. Additionally, other papers have searched for RyR2 mutations in ACM probands yielding in 9% of RyR2 rare variants but no co-segregation are provided (3), and maybe not identified variants in other genes might be playing a significant role. All in all, not compelling evidence can now be gathered to confirm RyR2 as an ACM-causing gene, and probably it should be considered as a candidate gene until more clinical evidence is released. The authors of the herein presented manuscript give compelling in vitro evidence about the abnormalities in CPVT important proteins and calcium homeostasis caused by ACM mutations. In keeping with that, recent evidence confirm some sort of RyR2 dysfunction in ACM in vitro models (4). Thus, these pieces of evidence may confirm an overlap in the pathogenesis of ventricular arrhythmia with undistinguishable polymorphic stress-triggered ventricular tachycardia since calcium mishandling is present in both entities but it doesn’t mean that RyR2 mutations are capable of inducing an ouvert structural ACM phenotype. Mutations in other structural genes have also been found to cause calcium misshandling, such as RBM20. In order to avoid more confusion, the authors should be consistent on the pathogenic role they give to RyR2 mutations to produce an ACM phenotype throughout their manuscript and, in case it is not fully clarified, it should be clearly stated.

Q: we clarify into Pathophysiology section that there isn’t a clear pathogenic role of RyR2 gene in ACM.

4. A: When describing the current evidence regarding autoimmunity (page 9 line 358 onwards) in ACM pathophysiology, besides autoantibodies against DSG2, anti heart and anti intercalated disks autoantibodies, also some short comment could be included regarding antimitochondrial autoantibodies.

Q: We wrote a short commentary on antimitrochondrial Abs in ACM pathophysiology into paragraph 2.2.

5. A: Cites should be introduced in page 4 line 121 when the authors mention that some papers have linked RyR2 mutations R176Q, L433P, N2386I and T2504M to a ACM phenotype, unless all of them have been studied in the reference 23 later on introduced.

Q: These part of the paragraph has been erased after reconsidering the pathogenetic role of RyR2 gene in ACM.

6. Q: Typing mistakes: page 3 line 108 celladhesion (istead of cell adhesion, with or without -), page 4 line 129 hIPSC-CMs (instead of hiPSC-CMs), page 4 line 150 FNLC (instead of FLNC), Page 13 line 551 comlication instead of complication.

A: Typing mistakes corrected.

7. Q: Some specific mutations are fully mentioned throughout the manuscript but, when mentioning PLN and TMEM43 genes only the genes are cited even though only scarce mutations have been clearly associated with the ACM phenotype and should be fully mentioned, such as PLN R14del and TMEM43 S358L.

A: PNL R14del mentioned into paragraph 2.4

8. Q: The Footnote of Figure 4 does not have much to do with the figure itself, it appears to be some mistake here.

A: By mistakes, we inserted Figure 3 instead of Figure 4 into the original manuscript. Figures have been modified: original Figure 2 have been eliminated, original Figure 4 (now Figure 2) has been correctly inserted instead of original Figure 3 (now Figure 3).

9. Q: Several times throughout the manuscript appears misspelled the term triggerring instead of the correct version triggering.

A: Misspelling corrected

10. A: Page 10 line 421, when dealing this the electrocardiographical signs of ACM and thee diagnostic tools, it will be welcome a brief comment on the usefulneess of SAECG and Fontaine leads nowadays.

Q:  A comment on SAECG is present in Paragraph 3

11. A: When the authors refer to epsilon waves Page 11 line 426 onwards) they sometimes employ other terms such as “epsilon potentials” and epsilon potentials. It is desirable to use always the same term.

Q: thanks for the suggestion we decided to used only the term  “epsilon wave”

12. A: Page 11 line 483, similarly to what has been mentioned with echocardiography, a short comment on the advances on strain evaluation with CMR in ACM will be interesting to be added here, both on RV and LV strain.

Q: A brief comment on CMR-strain analysis is presente at the end of paragraph 3

13. Throughout the manuscript International Task Force Criteria are abbreviated as ITFC, except in page 11 line 454 where is is used TF.

Q: Thanks to your suggestion we decided to use always the abbreviation ITFC

14. A: Page 15 line 584, authors should make clear that the abovementioned risk score model and its validation by Aquaro were designed ONLY on patients fulfilling definite ITFC of ACM and thus, it remains unknown whether this score is similarly accurate for LD-ACM forms which do not fulfill ITFC as definite ACM. Indeed, Aquaro et al also published another paper in 2020 stressing that this score is valid for lone-RV ACM but it understimated the risk in ACM with LV involvement.

Q: We followed your advices mentioning the possible limit of this risk score for LD-ACM score not fulfilling ITC criteria for definite ACM.