Valve-Dependent Regional Heterogeneity of Wall Mechanics and Collagen Remodeling in Ascending Thoracic Aortic Aneurysms

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In the manuscript entitled “Valve-dependent regional heterogeneity of wall mechanics and collagen remodeling in ascending thoracic aortic aneurysms” the authors test the hypothesis that current assessment of the diseased vessel diameters does not match more detailed information regarding mRNA and collagen content, etc. They present data from an analysis of human samples using tensile tests, histology, proline assays, and pcr. Overall, the text is clear and logical. The figures are easy to understand. The authors are thanked for their efforts. A few comments and suggestions follow.

Novelty and rationale: State clearly at the start i.e. the abstract and introduction what the clinical hurdles the work would be expected to overcome and how care might be improved by so doing. Will these data help stage and or apply corrective surgery or hypertension heart failure therapies. Parenthetically, as rapid search of PubMed using the phrase ‘valve heterogeneity thoracic aortic aneurysm’ returned 47 papers. Thus, there is some novelty in the topic. The size of the tissue bank is a real plus, although not all enrolled individuals had sufficient biomaterial for the full testing profile.

Abstract: Please spell out abbreviations and plainly state the hypothesis.

Introduction: short and to the point. But explain what knowledge deficit this work overcomes of if heterogeneity is known, as it is, how the data extends what is known. Just have a larger cohort size may be helpful but does not mean a real advance.

Methods: Well described.

Ethics: Add in how human data was protected and archived; and how consent was obtained. Include the IRB protocol number.

Data: A good panel of information. Actual protein levels of several genes would help. Some left off that might be considered include the MMPs and matricellular proteins such as thrombospondins. Also maybe try and look for immune cells in the vessel sections. These ideas would expand the data which as it is did not tell us all that much beyond what was known. 

Discussion: Appropriate length and with a thoughtful limitations section.

Figures: Nicely done but the tissue microscope images were a little blurry.

Author Response

We thank Reviewer 1 for the thorough and constructive evaluation of our manuscript. The comments have helped us substantially improve the clarity and scientific rigor of the work. Below, we provide point-by-point responses to each comment. All changes are highlighted in the revised manuscript using tracked changes.

Comment 1: Novelty and rationale: State clearly at the start i.e. the abstract and introduction what the clinical hurdles the work would be expected to overcome and how care might be improved by so doing. Will these data help stage and or apply corrective surgery or hypertension heart failure therapies. Parenthetically, as rapid search of PubMed using the phrase ‘valve heterogeneity thoracic aortic aneurysm’ returned 47 papers. Thus, there is some novelty in the topic. The size of the tissue bank is a real plus, although not all enrolled individuals had sufficient biomaterial for the full testing profile.

Response 1: We thank the reviewer for this important suggestion. We agree that the clinical relevance was insufficiently foregrounded. We have revised the abstract and introduction to explicitly state the clinical problem (limitations of diameter-based risk stratification) and how regional wall assessment could improve surgical decision-making. Specifically, we now reference data showing that the majority of aortic dissections occur below current surgical thresholds (Kreibich et al.), and we articulate how regional heterogeneity might serve as a complementary parameter for risk stratification. The text now clearly states that the findings do not provide immediate clinical tools but motivate further investigation of wall-based endpoints in prospective studies.

Comment 2: Abstract: Please spell out abbreviations and plainly state the hypothesis.

Response 2: We have spelled out all abbreviations at first use in the abstract (ATAA, IC, OC, TAV, BAV, HTx, RT-qPCR, ELN). The hypothesis is now explicitly stated in the abstract: We hypothesized that regional differences between the outer and inner curvature of the ascending aorta are valve-morphology dependent and reflect distinct remodeling mechanisms in bicuspid versus tricuspid aortic valve-associated aortopathy.

Comment 3: Introduction: short and to the point. But explain what knowledge deficit this work overcomes of if heterogeneity is known, as it is, how the data extends what is known. Just have a larger cohort size may be helpful but does not mean a real advance.

Response 3: We fully agree. The introduction has been revised to more clearly delineate the specific knowledge gap. While regional biomechanical heterogeneity in ATAA has been described, prior studies have not systematically examined whether these differences are valve-morphology dependent, and no study has simultaneously characterized biomechanical, histological, and transcriptional OC/IC differences in a single well-characterized cohort with n > 100 paired samples. The novelty claim is therefore not simply sample size, but the multimodal, valve-stratified characterization of OC/IC heterogeneity and the discovery of a valve-specific discordance between collagen gene expression and protein content.

Comment 4: Methods: Well described.

Response 4: We thank the reviewer for this positive assessment.

Comment 5: Ethics: Add in how human data was protected and archived; and how consent was obtained. Include the IRB protocol number.

Response 5: The IRB protocol number (22-0280, Ethics Committee of Ludwig-Maximilians-Universität München) was already stated in the Methods section (4.1). We have now added a sentence on data protection and consent: all identifiable data were pseudonymized and stored in a password-protected institutional database accessible only to study personnel. Written informed consent was obtained from all participants prior to tissue collection. The HTx control tissue was collected anonymously under institutional approval. These details are now included in section 4.1.

Comment 6: Data: A good panel of information. Actual protein levels of several genes would help. Some left off that might be considered include the MMPs and matricellular proteins such as thrombospondins. Also maybe try and look for immune cells in the vessel sections. These ideas would expand the data which as it is did not tell us all that much beyond what was known.

Response 6: We thank the reviewer for these thoughtful suggestions. MMP expression profiling, matricellular protein quantification (e.g., thrombospondin-1, osteopontin), and immune cell quantification (e.g., macrophage, T-cell infiltration) would indeed add important mechanistic depth. However, these analyses were beyond the scope of the present study, which focused on primary structural ECM components (fibrillar collagens, elastin) in parallel with biomechanical characterization. We have added these as explicit future research directions in the in the Limitations section. Importantly, the collagen transcript findings are supported at the biochemical level by the hydroxyproline assay, which provides independent protein-level evidence for collagen content. We note that protein-level quantification of individual collagen subtypes from aortic tissue homogenates is methodologically challenging due to the high insolubility and extensive crosslinking of fibrillar collagens, which limits efficient protein extraction for Western blot or ELISA-based approaches. This is particularly relevant for the minor collagens COL5A1 and COL11A1, which are deeply embedded within the fibrillar network. Similarly, elastin represents one of the most insoluble structural proteins and requires specialized solubilization protocols for quantitative protein-level assessment.Comment 7: Discussion: Appropriate length and with a thoughtful limitations section.

Response 7: We thank the reviewer for this positive assessment.

Comment 8: Figures: Nicely done but the tissue microscope images were a little blurry.

Response 8: We apologize for the reduced image quality in the submitted version. The histology images were downsampled during PDF conversion of the submission. We have replaced Figure 3 with a high-resolution TIFF export (≥300 dpi), submitted as a separate figure file alongside the revised manuscript

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a comprehensive analysis of regional (outer vs. inner curvature) biomechanical, histological, biochemical, and transcriptional differences in ascending thoracic aortic aneurysms (ATAA), stratified by valve morphology (BAV vs. TAV). I have these suggestions to follow:

1) TAV patients had significantly larger aortic diameters and were significantly older than BAV patients (Table 1). Both diameter and age are known to influence, Aortic wall stiffness, Collagen/elastin composition and Cellular density and medial degeneration. Although this difference is acknowledged in the limitations, no adjustment or sensitivity analysis is presented. I suggest performing multivariable analyses, linear regression adjusting for age and maximal aortic diameter when comparing TAV and BAV groups.

2) Clarify whether regional (OC vs. IC) comparisons are robust to diameter stratification.

3) A central conclusion is that TAV shows transcriptional upregulation without net collagen accumulation, whereas BAV shows “functionally productive” transcription leading to increased collagen deposition. This is not concluded properly as hydroxyproline measures total collagen content, not subtype-specific accumulation. Furthermore, increased hydroxyproline does not necessarily imply biomechanically effective fibrillar remodeling.

4) The HTx group is small (n=8 for biomechanics) and lacks clinical data. The manuscript states that absence of regional differences in HTx suggests pathology-specific heterogeneity. However, donor and recipient aortas may have variable exposure to catecholamines, inflammation, or cardiomyopathy-related remodeling.

5) Figure numbering inconsistency: There are two “Figure 2” captions (correlation plots and workflow). Please correct numbering. And there is no figure 1 in the manuscript.

6) Clarify whether R² values are from simple linear regression and whether slope significance was tested.

7) In Table 3, clarify whether collagen ratios (EvG vs. hydroxyproline) were compared statistically between valve groups.

Author Response

We thank Reviewer 2 for the thorough and constructive evaluation of our manuscript. The comments have helped us substantially improve the clarity and scientific rigor of the work. Below, we provide point-by-point responses to each comment. All changes are highlighted in the revised manuscript using tracked changes.

Comment 1: TAV patients had significantly larger aortic diameters and were significantly older than BAV patients (Table 1). Both diameter and age are known to influence, Aortic wall stiffness, Collagen/elastin composition and Cellular density and medial degeneration. Although this difference is acknowledged in the limitations, no adjustment or sensitivity analysis is presented. I suggest performing multivariable analyses, linear regression adjusting for age and maximal aortic diameter when comparing TAV and BAV groups.

Response 1: We thank the Reviewer for this important point. We agree that the age and diameter differences between TAV and BAV patients (TAV: 64 ± 10 years, 54.8 ± 7.7 mm; BAV: 56 ± 11 years, 50.9 ± 5.1 mm; both p < 0.001) represent potential confounders for between-group comparisons.

For biomechanical and histological parameters (Sections 2.2 and 2.3), formal multivariable adjustment was not applied, as no inferential statistical comparisons between valve groups are made for these outcomes. Cross-group observations in these sections are described narratively without hypothesis testing. All primary statistical analyses for biomechanical and histological data are paired within-patient comparisons between OC and IC, in which each patient serves as their own control and aortic diameter is by definition identical for both specimens within each pair, precluding diameter-related confounding by design.

The RT-qPCR OC/IC ratios represent the only formal inferential BAV vs. TAV comparisons in this manuscript, and multivariable adjustment was therefore applied exclusively to these data. We performed multivariable linear regression for all six RT-qPCR outcomes in all 137 aneurysmal patients with complete data (TAV n=69, BAV n=68), modeling the OC/IC expression ratio as dependent variable with valve morphology (BAV = 1 vs. TAV = 0), age, and maximal aortic diameter as independent variables. Results are presented in Supplementary Table 3.

After adjustment, the direction of all between-group differences was preserved. For COL5A1, the OC/IC ratio remained nominally lower in BAV than TAV (β = −0.404, 95% CI [−0.743, −0.065], p(adj) = 0.020); however, this association did not survive Benjamini-Hochberg correction across all six genes (p(FDR) = 0.121). Similarly, ELN showed a nominally higher OC/IC ratio in BAV than TAV in unadjusted analysis (p = 0.011), but this difference was no longer significant after adjustment (β = +0.304, 95% CI [−0.148, 0.756], p(adj) = 0.185, p(FDR) = 0.554). Accordingly, no between-group difference in OC/IC expression ratios reached statistical significance after full adjustment. We have revised Results Section 2.4 and the Discussion to reflect this more cautious interpretation, explicitly noting that the previously reported COL5A1 and ELN differences between valve groups should be considered exploratory, pending independent validation in cohorts with matched baseline characteristics. The multivariable regression results have been added as Supplementary Table 3, and relevant passages in Results Section 2.4, Discussion, and the Limitations paragraph have been updated accordingly.

Comment 2: Clarify whether regional (OC vs. IC) comparisons are robust to diameter stratification.

Response 2: We thank the Reviewer for raising this point. All regional comparisons between OC and IC in this manuscript are paired within-patient analyses. Since OC and IC specimens are obtained from the same aortic ring of the same patient, aortic diameter does not differ between paired specimens by design and is therefore not expected to confound within-patient regional comparisons. This applies equally to biomechanical, histological, and RT-qPCR analyses. Formal covariate adjustment for diameter was therefore not performed for these paired comparisons.

Comment 3: A central conclusion is that TAV shows transcriptional upregulation without net collagen accumulation, whereas BAV shows “functionally productive” transcription leading to increased collagen deposition. This is not concluded properly as hydroxyproline measures total collagen content, not subtype-specific accumulation. Furthermore, increased hydroxyproline does not necessarily imply biomechanically effective fibrillar remodeling.

Response 3: We thank the Reviewer for this important methodological point and agree that our original interpretation was overstated. We have revised the relevant passages in the Discussion accordingly.

We acknowledge that hydroxyproline quantification measures total collagen content and does not allow subtype-specific conclusions. The observation that BAV shows elevated OC hydroxyproline content therefore reflects net total collagen accumulation and cannot be attributed to specific collagen subtypes. The inference that the transcriptional upregulation of individual collagen genes (COL1A1, COL3A1, COL4A1, COL5A1, COL11A1) translates into subtype-specific protein accumulation is not supported by the available data.

Furthermore, we agree that increased total collagen content does not necessarily imply biomechanically effective fibrillar remodeling. Elevated hydroxyproline content provides no information on collagen fibril diameter, cross-linking density, or spatial organization within the extracellular matrix, all of which determine the mechanical contribution of collagen to tissue integrity.

We have therefore replaced the term "functionally productive transcription" and revised the interpretation to state that BAV shows elevated total collagen content at the OC compared to the IC, while the relationship between this finding and fibrillar architecture or biomechanical efficacy cannot be resolved with the present data. Future studies incorporating collagen subtype-specific protein quantification and assessment of fibril ultrastructure will be needed to further characterize these valve-specific remodeling pathways.

Comment 4: The HTx group is small (n=8 for biomechanics) and lacks clinical data. The manuscript states that absence of regional differences in HTx suggests pathology-specific heterogeneity. However, donor and recipient aortas may have variable exposure to catecholamines, inflammation, or cardiomyopathy-related remodeling.

Response 4: We thank the Reviewer for this point. We agree with the Reviewer's assessment. The HTx group is described in the Limitations paragraph as a non-aneurysmal reference rather than a physiological baseline, acknowledging potential confounding by catecholamine exposure, cardiomyopathy-associated aortopathy, and other subclinical pathology. In response to this comment, we have additionally added a clarifying sentence to the Discussion to explicitly qualify the interpretation that the absence of regional differences in HTx reflects pathological rather than physiological heterogeneity, acknowledging that this conclusion rests on a small and clinically heterogeneous control group and should therefore be interpreted with caution.

Comment 5: Figure numbering inconsistency: There are two “Figure 2” captions (correlation plots and workflow). Please correct numbering. And there is no figure 1 in the manuscript.

Response 5: We thank the Reviewer for identifying this oversight. The figure numbering has been corrected throughout the manuscript. The experimental workflow figure has been renumbered as Figure 1, and the remaining figures have been renumbered consecutively. All in-text references to figures have been updated accordingly.

Comment 6: Clarify whether R² values are from simple linear regression and whether slope significance was tested.

Response 6: We thank the Reviewer for this clarifying question. The R² values reported in Section 2.2 were derived from simple linear regression of paired OC versus IC measurements within each group. The associated p-values reflect the significance of the regression slope, i.e. the test of the null hypothesis that the slope equals zero, which is equivalent to testing the significance of the Pearson correlation coefficient. This has been clarified in the Statistical Analysis section of the revised manuscript.
Comment 7: In Table 3, clarify whether collagen ratios (EvG vs. hydroxyproline) were compared statistically between valve groups.

Response 7: The collagen ratios presented in Table 3 (EvG-based and hydroxyproline OC/IC ratios) were not formally compared between valve groups. All statistical comparisons in Table 3 are paired within-patient OC vs. IC comparisons. The between-group differences in collagen ratios visible in Table 3 are reported descriptively, as the primary aim of this analysis was to characterize regional heterogeneity within each valve group rather than to compare absolute remodeling magnitude between groups. This has been clarified in the revised manuscript.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

A revised draft and response letter are provided. The authors addressed my concerns. They are thanked for the additional effort.

Reviewer 2 Report

Comments and Suggestions for Authors

Authors have sufficiently addressed my concerns, and I have no further comments.