Circulating Cell-Free DNA Analysis for Diagnostic and Prognostic Assessment of Hepatocellular Carcinoma in Cirrhosis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

I'm pleased to let you know that I highly evaluated your research, its very smart rationale, hypothesis, experimental details, and statistics. There is no doubt that your research and, in particular, your current manuscript in such an important field as liver cancer, greatly contributes to the methodologies of early liver cancer detection, and, as a final goal, longer cancer patient survivals. I have no questions or critique as concerns your scientific part of the manuscripts. However, I offered some corrections in the next section below. 

Please incorporate the following comments: 

• 26, 27: for some reason, some characters are underlined; the same in 139, 140
• 88: Please explain the CRP abbreviation and following abbreviations too. In general, the manuscript contains a lot of abbreviations without explanations. And this list of abbreviations is not complete; the manuscript contains more abbreviations
• 100: Should be concentration, mononucleosomes
• Table 1, page 4, 'Early Advanced A B C D - needs formatting
• 104: Should be : 'mean ± SD' 
• 143: Please make spaces
• 164: Should be: concentration, mononucleosomes
• 211: Should be: concentration, mononucleosomes
• 237: 'with an early detection'
• 280, 282, and 283: in e.g. 'the CMAC model" the article 'the' is a must
• 312-315: The phrase 'The inclusion criterion for cirrhotic patients was diagnosis of Child-Pugh A cirrhosis without HCC with follow-up in a screening program and the inclusion criterion for HCC patients was diagnosis of naive HCC at any BCLC (Barcelona Clinic Liver Cancer) stage.' should be reworded. It definitely sounds confusing
• 355: Should be: '[in (%)]'
• 365: Should be: 'the Mann-Whitney U test'
• 413-414: This reference needs formatting

Sincerely,

Reviewer 

Author Response

Dear Authors,

I'm pleased to let you know that I highly evaluated your research, its very smart rationale, hypothesis, experimental details, and statistics. There is no doubt that your research and, in particular, your current manuscript in such an important field as liver cancer, greatly contributes to the methodologies of early liver cancer detection, and, as a final goal, longer cancer patient survivals. I have no questions or critique as concerns your scientific part of the manuscripts. However, I offered some corrections in the next section below. 

Please incorporate the following comments: 

• 26, 27: for some reason, some characters are underlined; the same in 139, 140
• 88: Please explain the CRP abbreviation and following abbreviations too. In general, the manuscript contains a lot of abbreviations without explanations. And this list of abbreviations is not complete; the manuscript contains more abbreviations
• 100: Should be concentration, mononucleosomes
• Table 1, page 4, 'Early Advanced A B C D - needs formatting
• 104: Should be : 'mean ± SD' 
• 143: Please make spaces
• 164: Should be: concentration, mononucleosomes
• 211: Should be: concentration, mononucleosomes
• 237: 'with an early detection'
• 280, 282, and 283: in e.g. 'the CMAC model" the article 'the' is a must
• 312-315: The phrase 'The inclusion criterion for cirrhotic patients was diagnosis of Child-Pugh A cirrhosis without HCC with follow-up in a screening program and the inclusion criterion for HCC patients was diagnosis of naive HCC at any BCLC (Barcelona Clinic Liver Cancer) stage.' should be reworded. It definitely sounds confusing
• 355: Should be: '[in (%)]'
• 365: Should be: 'the Mann-Whitney U test'
• 413-414: This reference needs formatting

We are very grateful to the reviewer for the detailed and careful reading of our manuscript. We have addressed every one of the orthographic, grammatical, and formatting suggestions provided. Specifically, we have:

• Removed the unintentional underlining and corrected spacing issues throughout the text.
• Defined all abbreviations upon their first mention and ensured the list of abbreviations is now complete and consistent.
• Corrected terminology and ensured the use of 'the' before 'CMAC model' as requested.
• Standardized formatting in Table 1, including the staging labels and the use of 'mean ± SD'.
• Reworded the inclusion criteria (lines 320-324) for both LC and HCC patients to improve clarity.
• Updated the statistical nomenclature (e.g., 'the Mann-Whitney U test') and fixed the formatting of the references.

All these corrections have been implemented, and for the convenience of the editor and reviewers, they are highlighted in yellow in the revised manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a well-conducted study evaluating plasma cfDNA concentration and ‎fragmentomics as diagnostic and prognostic biomarkers for hepatocellular carcinoma (HCC) in ‎patients with liver cirrhosis (LC). The development of the CMAC model (incorporating cfDNA ‎concentration, mononucleosome proportion, AFP, and CRP) represents a novel and clinically ‎relevant approach. The study addresses an important clinical need, improving early HCC ‎detection in high-risk cirrhotic patients, where current screening methods, particularly AFP, ‎demonstrate limited sensitivity. However, several issues need to be addressed before the work ‎can be considered for publication, including:‎

• The manuscript indicates that early-stage (ES) patients comprised 38.5% (n=15) but does not ‎clearly specify the criteria for ES versus AS within the BCLC system. BCLC 0 (very early) and ‎A (early) have different prognostic implications. This should be clarified.‎
• In the developed CMAC model, the unit of each variable should be specified.‎
• Table 1 indicates that 3 deaths occurred in the LC group. The causes of death should be ‎specified, and it should be clarified whether these patients were excluded from any analyses, as ‎excluding them from analysis might bias results.‎
• CRP is an acute-phase reactant that can be elevated in cirrhosis without HCC due to infections or ‎ The manuscript does not address whether LC patients with elevated CRP (due to ‎non-malignant causes) might be misclassified by the CMAC model.‎
• Lymphovascular invasion was detected in 12.8% of HCC patients. The authors should clarify ‎how LVI was assessed (e.g., Histopathology of resected specimens or Imaging).‎

 

Author Response

This manuscript presents a well-conducted study evaluating plasma cfDNA concentration and ‎fragmentomics as diagnostic and prognostic biomarkers for hepatocellular carcinoma (HCC) in ‎patients with liver cirrhosis (LC). The development of the CMAC model (incorporating cfDNA ‎concentration, mononucleosome proportion, AFP, and CRP) represents a novel and clinically ‎relevant approach. The study addresses an important clinical need, improving early HCC ‎detection in high-risk cirrhotic patients, where current screening methods, particularly AFP, ‎demonstrate limited sensitivity. However, several issues need to be addressed before the work ‎can be considered for publication, including:‎

• The manuscript indicates that early-stage (ES) patients comprised 38.5% (n=15) but does not ‎clearly specify the criteria for ES versus AS within the BCLC system. BCLC 0 (very early) and ‎A (early) have different prognostic implications. This should be clarified.‎

We thank the reviewer for this important observation. We agree that the distinction between BCLC 0 and BCLC A is crucial due to their different prognostic implications. We would like to clarify that our study did not include any patients in BCLC stage 0 (very early stage). Therefore, the early stage (ES) group in our cohort consists exclusively of patients in BCLC stage A, while the advanced stage (AS) group encompasses BCLC stages B, C, and D. We have revised the 'Patient population' section (page 11, line 333) to explicitly state the absence of BCLC 0 patients and to reinforce the definition of our staging criteria to ensure there is no ambiguity.

• In the developed CMAC model, the unit of each variable should be specified.‎

We thank the reviewer for this suggestion. We have ensured that the units for all parameters included in the CMAC model are clearly specified to facilitate reproducibility. Specifically, we have added the units directly into the model equation section in the revised manuscript (page 5, line 145) to ensure there is no ambiguity when applying the formula. We believe this addition clarifies the technical application of the CMAC model.

• Table 1 indicates that 3 deaths occurred in the LC group. The causes of death should be ‎specified, and it should be clarified whether these patients were excluded from any analyses, as ‎excluding them from analysis might bias results.

We appreciate the reviewer’s attention to detail regarding the mortality in the liver cirrhosis (LC) group. The causes were subdural hematoma (n=1), pneumonia (n=1), and cardiorespiratory arrest (n=1). We have included the causes of death in a footnote for Table 1 for better transparency. Regarding the analysis, we would like to clarify that none of these patients were excluded. All 3 patients were maintained in the baseline characterization and all analyses to avoid any potential selection bias.

• CRP is an acute-phase reactant that can be elevated in cirrhosis without HCC due to infections or ‎ The manuscript does not address whether LC patients with elevated CRP (due to ‎non-malignant causes) might be misclassified by the CMAC model.

We appreciate the reviewer’s comment. It is indeed true that CRP is an acute-phase reactant and can be elevated in patients with cirrhosis due to various non-malignant inflammatory conditions. However, the strength of the CMAC model lies precisely in its multi-parametric nature. By integrating CRP with other independent variables, the model does not rely on a single non-specific marker. Instead, it assigns a balanced weight to each parameter, allowing for high diagnostic accuracy even when one of the components, such as CRP, might be influenced by the underlying cirrhosis or minor systemic inflammation. In our cohort, the model maintained its robust performance without significant misclassification in the LC group, demonstrating that the combination of variables effectively filters the 'noise' of non-malignant CRP elevations. We have added a sentence in the 'Discussion' section (page 9, lines 274-277) to highlight this advantage of the multi-parametric approach.

• Lymphovascular invasion was detected in 12.8% of HCC patients. The authors should clarify ‎how LVI was assessed (e.g., Histopathology of resected specimens or Imaging).‎

We thank the reviewer for pointing out the need for clarification on this diagnostic procedure. In our study, the presence of vascular invasion was assessed through cross-sectional imaging at the time of diagnosis. We have updated the 'Methods' section (page 10, line 324) to explicitly state that LVI was determined via imaging criteria to avoid any misunderstanding regarding the use of surgical pathology.