Pericoronary Radiomics Signature for Non-Culprit Lesion Progression and Revascularization Decision in NSTE-ACS

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I am grateful to the editor for the opportunity to review the manuscript by Haidan Zhang et al., "Assessment of Non-Culprit Lesion Progression via Pericoronary Radiomics Signature Refines Revascularization Decision in Non-ST-Segment Elevation Acute Coronary Syndrome." In this article, the authors examined an important question: the ability to identify vulnerable coronary plaques using an original technique (CCTA-based radiomics model of perivascular adipose tissue - PCAT). The authors' results are impressive: the resulting lesion-specific PCAT radiomics signature identifies non-coronary lesions with a high risk of accelerated progression and future major cardiovascular events in patients with NSTE-ACS. The study is well designed, the data obtained are compelling, and it naturally raises the desire to see further research in this area. While my overall impression of the article is very favorable, I do have a few comments. My suggestions arising from reviewing this manuscript: 1. It would be useful to examine in more detail recent articles on the use of Pericoronary Adipose Tissue Radiomics from Coronary Computed Tomography Angiography in identifying vulnerable atherosclerotic plaques (refs. 1-2, see below). For example, the relationship between the radiomic characteristics of coronary CT angiography images and the characteristics of vulnerable plaques identified using intravascular optical coherence tomography was studied. I believe it would be appropriate to present this analysis and compare it with the authors' data in the Discussion section. 2. I believe Table S3 (Baseline Characteristics Between Patients with and without NCL-related MACE) and Table S4 (Univariable Cox Proportional Hazards Analysis for the Prediction of NCL-related MACE in the Training and Validation Cohorts) should be moved from the Supplementary Materials to the main text of the article; this would be more helpful for understanding the authors' results. Conversely, Table 1. Baseline Characteristics of the Training and Validation Cohorts could be moved from the main text to the Supplementary Materials. References: 1. Kim JN, Gomez-Perez L, Zimin VN, Makhlouf MHE, Al-Kindi S, Wilson DL, Lee J. Pericoronary Adipose Tissue Radiomics from Coronary Computed Tomography Angiography Identifies Vulnerable Plaques. Bioengineering (Basel). 2023 Mar 14;10(3):360. doi: 10.3390/bioengineering10030360. 2. Pan J, Huang Q, Zhu J, Huang W, Wu Q, Fu T, Peng S, Zou J. Prediction of plaque progression using different machine learning models of pericoronary adipose tissue radiomics based on coronary computed tomography angiography. Eur J Radiol Open. 2025 Feb 15;14:100638. doi: 10.1016/j.ejro.2025.100638.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Review report attached.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript presents a prospective cohort study investigating the role of a CCTA-derived pericoronary adipose tissue (PCAT) radiomics signature in predicting non-culprit lesion (NCL) progression and major adverse cardiovascular events (MACE) in patients with NSTE-ACS. The topic is highly relevant, and the study addresses a clinically important gap, namely the identification of biologically high-risk non-culprit lesions beyond traditional anatomical and functional assessment. Overall, the manuscript is well structured, methodologically sound, and supported by appropriate statistical analyses. The integration of radiomics with clinical variables to improve risk stratification represents a novel and potentially impactful contribution to the field. 
-the Methods section of the Abstract contains excessive technical detail, which may reduce accessibility for a broader clinical readership.
-the Introduction section is somewhat lengthy and contains minor redundancies, particularly in the discussion of the limitations of invasive coronary angiography, intravascular imaging, and FFR. A more concise presentation would improve readability.
-In the Results, some borderline statistical findings are interpreted with strong clinical implications, which should be moderated. 
-In the discussion, the clinical implications, especially the suggestion that the model may guide revascularization strategies, should be presented more cautiously, as this remains a hypothesis-generating finding rather than a validated clinical application.
-Table 1: typo "hyperlipemia" change with "Hyperlipidemia"
-Line 308-311: There is a duplication in the text, the same sentence is repeated twice.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this interesting manuscript. The topic is of clinical interest, as the management of non-culprit lesions remains a challenging and important aspect of decision-making in patients with acute coronary syndromes. The study proposes an innovative CCTA-based radiomics approach focused on pericoronary adipose tissue. Overall, the work is promising, but I think there are some points to be better clarified:

1. The manuscript aims to identify high-risk non-culprit lesions, yet the patient-level analysis seems to rely on the score of the single highest-risk non-culprit lesion. This transition from lesion-level assessment to patient-level prediction should be described more clearly, including its methodological rationale and its implications for clinical interpretation.
2. One of the main limitations of the study is the absence of any comparison with intracoronary imaging techniques such as OCT or IVUS. Such a comparison would have considerably strengthened the biological and clinical interpretation of the proposed radiomics signature by allowing correlation with direct markers of plaque vulnerability. If these data are not available, this limitation should be acknowledged more explicitly in the Discussion.
3. Authors should clarify exactly how the different filtering and selection steps were performed and whether all of these steps were carried out strictly within the training cohort. This is important to exclude the possibility of information leakage and to allow readers to better assess the robustness of the model.
4. The authors should better explain what measures were taken to reduce this risk and how stable they believe the final model is likely to be in an external population.
5. In some sections the model is described as consisting of 8 features, whereas elsewhere it is described as including 9. This discrepancy should be carefully checked and corrected throughout the manuscript.
6. More details about the CCTA follow-up patients is needed since follow-up imaging was performed for clinical indications and the possibility of selection bias should be discussed more explicitly.
7. Since this is a single-center study with internal validation only, statements regarding the use of the model to guide revascularization decisions or personalized treatment strategies should be moderated.

Comments on the Quality of English Language

The manuscript would benefit from careful language revision to improve clarity and to correct a few minor inconsistencies in terminology and presentation across the text.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors responded to my comments and made corrections to the manuscript. I have no other comments.

Reviewer 2 Report

Comments and Suggestions for Authors

Responses to all review comments have been addressed.

Reviewer 4 Report

Comments and Suggestions for Authors

The revised version addresses my main concerns. No further comments.

Comments on the Quality of English Language

The manuscript would benefit from careful language revision to improve clarity and to correct a few minor inconsistencies in terminology and presentation across the text.