Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Referee’s comment

 

Article n° IJMS 4604598

 

Title: Advanced Cellular Models for Drug Testing and Discovery to 2 Target Cardiac Fibrosis Progression

 

Authors: F. Icolaro, I. Chimenti, C. Cozzolino, A. Bordin, V. Pontecorvi, G. Frati, E. De Falco, E. Floris, V Picchio, F. Pagano

 

General comment

The manuscript by Icolaro et al. tackles an interesting and timely topic, focusing on 3D systems for modeling cardiac fibrosis and testing potential antifibrotic drugs. Incorporating a mechanobiological perspective is a clear strength that gives the review a distinct angle. However, the current draft is still fairly descriptive and lacks a truly critical comparison between the platforms. In particular, concepts like model complexity, biological relevance, and predictive capacity are often used as if they were equivalent. The paper would be significantly strengthened by a more rigorous critical discussion around model validation, practical limitations, reproducibility, and actual translational value.

Specific points

1) Introduction: The criteria used to select the studies cited in the review are not clearly stated. It’s difficult to tell whether the chosen examples represent a comprehensive overview of the field or just a curated selection. I suggest adding a short paragraph to outline the literature search strategy and inclusion/exclusion criteria for better clarity.

2) 2. Mimicking Cardiac Fibrosis: 3D Models as Platforms for Evaluating Antifibrotic Compounds: The text tends to imply that greater model complexity automatically brings higher predictive power. However, a more complex platform isn't necessarily more predictive unless its performance has been rigorously cross-validated against relevant biological or clinical outcomes. I recommend clearly separating the concepts of model complexity, biological fidelity, and actual demonstrated predictive value.

3) 2.1. Three-Dimensional Models for Comparative Drug Evaluation: Antifibrotic efficacy is compared across different experimental platforms, but these setups vary wildly in cell composition, extracellular matrix, mechanical cues, and readout endpoints. Direct comparison of drug performance across such heterogeneous studies can be misleading. It would be valuable to explicitly discuss how platform-specific conditions influence and potentially bias the observed drug responses.

4) 2.3. Next Generation of Biomimetic Platforms: Beyond Drug Screening: The manuscript places heavy emphasis on the predictive potential of these advanced models, yet most cited studies rely purely on internal, in vitro readouts. Only a few offer validation in independent in vivo systems. The authors should explicitly highlight the specific level of validation achieved by each model discussed.

5) Table 1: Table 1 serves as a helpful initial summary, but it doesn't allow for a true head-to-head comparison. Key technical parameters are missing. I strongly suggest adding dedicated columns detailing cell sources, mechanical conditions, throughput, validation level, and major limitations.

6) 3. Mechanobiology, How Cells Generate, Sense and Respond to Mechanical Cues: The mechanobiology section is informative, but at times it feels a bit disconnected from the core focus of the paper. The link between specific mechanotransduction pathways and their direct utility in drug screening needs better development. The authors should tie each main mechanical mechanism more directly to assay design and pharmacological testing.

7) 3.1. Mechanotransduction in Cardiac Fibrosis and Fibroblast Responses: Findings from pulmonary, dermal, and other tissue models are often used to illustrate mechanobiological principles. While helpful, these observations don't always translate directly to cardiac fibroblasts. It would be better to clearly distinguish findings established in cardiac models from those derived from non-cardiac fibrotic tissues.

8) Conclusions and Future Perspectives (1): The current limitations of 3D cardiac fibrosis platforms aren't covered in sufficient depth. Critical issues such as cell immaturity, donor-to-donor variability, missing immune/vascular components, matrix inconsistency, and overall reproducibility deserve a much more detailed breakdown. Adding a dedicated section on these biological and technical hurdles would make the paper much stronger.

9) Conclusions and Future Perspectives (2): The manuscript maintains a very optimistic tone about the ability of 3D models to improve preclinical translation. However, robust evidence showing that these setups consistently outperform simpler models in predicting human clinical responses is still quite sparse. These claims should be toned down, and the authors should outline what benchmarks would actually be required to prove true predictive superiority.

10) Conclusions and Future Perspectives (3): Standardization is almost entirely absent from the current discussion. Without common benchmark controls, reference compounds, and proven inter-laboratory reproducibility, routine adoption of these platforms in drug discovery will remain out of reach. I suggest adding a discussion on the need for standardized reference drugs, shared endpoints, and multi-center validation protocols.

11) Entire manuscript: The manuscript requires a thorough linguistic and editorial polish. There are several grammatical inaccuracies, occasional inconsistencies in terminology, and a few incomplete template elements. Please perform a detailed line-by-line proofreading and double-check formatting prior to resubmission.

Author Response

Please see attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This review discusses the use of three-dimensional (3D) cellular models for studying cardiac fibrosis and testing antifibrotic treatments. The authors describe several platforms, including fibrosis-on-a-chip, engineered heart tissues, and mechanically stimulated models. The topic is relevant, and the examples show how the tissue microenvironment can affect drug responses. However, the manuscript currently reads more as a description of individual studies than a critical review. I suggest major revision to strengthen its contribution.

1. A previous review by some of the authors already discussed multicellular 3D models of cardiac fibrosis (reference [15]). The authors should explain more clearly what new perspective the present manuscript provides. A focus on model selection for drug testing could be valuable, but this idea needs to be developed throughout the paper.
2. Table 1 summarizes the main findings, but it does not show the limitations of each platform. Please compare cell sources, methods used to induce fibrosis, mechanical stimulation, functional readouts, screening capacity, and reproducibility. It would also be helpful to explain which models are more suitable for studying post-infarction scar formation or chronic interstitial fibrosis.
3. In Section 2.1, the manuscript states that 22 compounds were identified as potentially antifibrotic in the 2D model. However, reference [20] describes 22 compounds with different results between the 2D and 3D screening models. These are not the same meaning. Please correct this statement. More generally, the claims that 3D models reduce false-positive and false-negative results or have better predictive value should be presented with caution. The evidence from selected compounds should not be extended to all 3D platforms.
4. Section 3 gives useful information on stiffness and mechanosensitive pathways, but the authors should explain more directly how these factors change the choice of a model or the interpretation of a drug response. This would make the review more useful for researchers planning antifibrotic studies.

Author Response

please see attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have substantially addressed my previous comments. The revised manuscript now provides a clearer focus on model selection for antifibrotic drug testing. The expanded Table 1, the new discussion linking mechanobiology to assay design, and the sections on validation and standardization improve its practical value. The incorrect interpretation of the compound screening results has also been corrected, and claims about the predictive superiority of 3D models are more cautious. Although the contribution remains moderate and mainly based on qualitative synthesis, the manuscript is considerably improved.

1. The distinction between model complexity, biological fidelity, and predictive value is useful. Please add the proposed validation level for each platform in Table 1, together with a brief justification. Please also clarify that these levels are a practical framework proposed in this review rather than an established classification. Confirmation in animal models should remain clearly distinguished from validation of clinical predictive performance.
2. Section 2.4 would benefit from a concise decision matrix linking disease context, drug mechanism, required cell types, and recommended endpoints. This would strengthen the main contribution of the review and help readers apply the recommendations when planning experiments.
3. In Section 4.3, please clarify that pirfenidone is a clinically used antifibrotic drug for pulmonary fibrosis and an investigational comparator in cardiac fibrosis, rather than an approved treatment for cardiac fibrosis. In Table 2, effects on cardiomyocytes or endothelial cells should not automatically be described as “off-target effects,” because they may reflect on-target activity in other cell types.
4. The manuscript is generally understandable, but some repetitive or awkward expressions remain. For example, “Also, the growing recognition … also has …” should be simplified, and “Noteworthy, the mechanosensitive pathways …” should be changed to “Notably, the mechanosensitive pathways …”.

Author Response

Please se attached file

Author Response File: Author Response.pdf