Drug-Induced Liver Injury: Role of Circulating Liver-Specific microRNAs and Keratin-18

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I read this study by Cardin et al. with great interest. The study aimed to evaluate the “Drug Induced Liver Injury: role of circulating liver-specific microRNAs and Keratin-18”. The result of this study postulated Liver specific miRNAs and K-18 could be promising serum biomarkers of Drug induced liver injury. Despite its great contribution to expand our current knowledge, it suffers from several weaknesses to be used to make a valid and trustworthy conclusion.

This manuscript MUST be entirely rewritten under the supervision of a professional writer, particularly the sections on the introduction and discussion.

The introduction should be cohesive. It must provide sufficient evidence to justify the study. The introduction section of this manuscript is quite unclear and contains many unrelated topics. Nevertheless, this section should be completely revised under the guidance of a professional writer.

 

Why is the sample size low in this study?

 

The number of individuals in the control group is 5. This number is unacceptable, and the results cannot be considered reliable.

 

The same recommendations for the introduction should also be applied to the discussion.

 

 

Comments on the Quality of English Language

Extensive editing of English language required.

Author Response

C: This manuscript MUST be entirely rewritten under the supervision of a professional writer, particularly the sections on the introduction and discussion.

R: As suggested by the Reviewer, the manuscript has been carefully revised by a native English speaker.

C: The introduction should be cohesive. It must provide sufficient evidence to justify the study. The introduction section of this manuscript is quite unclear and contains many unrelated topics. Nevertheless, this section should be completely revised under the guidance of a professional writer. The same recommendations for the introduction should also be applied to the discussion.

R: Introduction and discussion have been revised and improved according to Reviewer suggestions.

C: Why is the sample size low in this study? The number of individuals in the control group is 5. This number is unacceptable, and the results cannot be considered reliable.

R. We agree with the Reviewer that the sample size is low, however, we would like to highlight that our study is monocentric and the number of patients enrolled is in line with the incidence of acute liver diseases presenting at our Unit. We are also aware that statistical analysis might be influenced by the small number of patients included, however we believe that the results obtained are encouraging and that a larger scale analysis should be performed, including more centers, to confirm our preliminary data.

Reviewer 2 Report

Comments and Suggestions for Authors

This study proposed novel prognostic scores to discriminate drug induced liver injury (DILI) from other liver pathologies that displayed somewhat similar parameters and biomarkers. The authors aim to evaluate combined prognostic model of liver-specific miRNA and cellular apoptosis biomarker keratin-18 (K-18) by performing ROC analysis on three candidate miRNAs, which serves as circulating biomarkers and displays distinct median level among three patient groups. In the present study, the new combined prognostic model could potentially identify patients with DILI, thereby improving decision-making in clinical practice. There are a few aspects that could be further addressed to strengthen the overall clarity and impact of the study.

1)    To improve the flow of the paper, the writing of the manuscript needs to be revised, especially introduction and discussion section, to help the reader to follow the rationale underlying the knowledge gaps and objectives.

 

2)      Authors could further elaborate the underlying problem caused by inability to differentiate DILI patient from liver-related pathologies or other patterns. By discussing how this differentiation impacts treatment, this could strengthen the relevance of the study.

 

3)    The primary aim of the study is to differentiate DILI patients from other liver-related conditions. However, in figure 1, candidate miRNAs, especially miR-21-5p and miR-122-5p, display non-specific pattern between DILI and HBV group.

 

4)    In figure 2, it would be helpful to provide the rationale behind the dividing DILI patients with R-value. For example, providing more context on how this grouping might improve the diagnosis power of the combined model.

 

5)    Summarizing of key finding and ensuring that p-value are reported in consistent format could solidify the study’s completeness and persuasiveness.

 

6)    There are a few instances where typographical errors or possible incorrect numbers appear in the text. For instance, on line 240, the fold change of M65 in HBV group is listed as ‘189’, while figure 3 suggest it should be ‘1890’.

Comments on the Quality of English Language

Overall, the narrative flow could be improved to help reader better understand the knowledge gap and clearly identify the objectives of the study. Grammar could use some polishing to enhance readability. For instance, on line 62, the comparison of the combined model of K-18 with other biomarkers should be phrased as 'better than ALT alone' rather than 'better that ALT alone’. Similarly, on line 67, the phrase could be more clearly stated as 'these biomarkers were also evaluated'. Addressing these points could improve the overall readability and completeness of the manuscript.

Author Response

1)    To improve the flow of the paper, the writing of the manuscript needs to be revised, especially introduction and discussion section, to help the reader to follow the rationale underlying the knowledge gaps and objectives.

As suggested by the Reviewer, the manuscript has been carefully revised by a native English speaker. Introduction and discussion have been improved.

2)      Authors could further elaborate the underlying problem caused by inability to differentiate DILI patient from liver-related pathologies or other patterns. By discussing how this differentiation impacts treatment, this could strengthen the relevance of the study.

 We thank the Reviewer for the comment. A paragraph has been added in the discussion to further elaborate this specific topic.

3)    The primary aim of the study is to differentiate DILI patients from other liver-related conditions. However, in figure 1, candidate miRNAs, especially miR-21-5p and miR-122-5p, display non-specific pattern between DILI and HBV group.

We agree with the Reviewer’s comment. Since the analysis of only miRNAs failed to discriminate DILI patients from other etiologies, we evaluated different disease markers together and developed a score with greater sensitivity and specificity than that expressed by markers considered separately. Combinations of several miRNAs or other types of markers may have the potential of being more specific to differentiate pathologies. We have now better specified this aspect in the text.

4)    In figure 2, it would be helpful to provide the rationale behind the dividing DILI patients with R-value. For example, providing more context on how this grouping might improve the diagnosis power of the combined model.

We thank the Reviewer for the comment. The rationale behind dividing DILI patients with R-value is to further classify DILI patients according to the pattern of liver disease, which has been clearly shown in the literature to be associated with severity of liver disease and therefore with potential death/liver transplantation or spontaneous resolution. We added a sentence in the methods section.

5)    Summarizing of key finding and ensuring that p-value are reported in consistent format could solidify the study’s completeness and persuasiveness.

We thank the Reviewer for this suggestion. The discussion has been improved by adding a summary of the principal findings of the study. In particular, we reported that:

1. The miRNAs evaluated in this study when considered alone are significantly higher in patients with acute liver diseases compared to controls;
2. Necrosis and apoptosis indexes were significantly higher than controls and correlated with miR122 and miR21;
3. MiR-122 and M30, as apoptosis index, had the highest sensitivity and specificity in identifying DILI from other groups;
4. A model comprising different miRNA (miR122 and miR34), M30 and hepatocellular pattern (H) of liver disease was able to differentiate DILI from other pathologies.

 6)    There are a few instances where typographical errors or possible incorrect numbers appear in the text. For instance, on line 240, the fold change of M65 in HBV group is listed as ‘189’, while figure 3 suggest it should be ‘1890’.

Errors and typos have now been corrected.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript titled "Drug Induced Liver Injury: Role of Circulating Liver-Specific miRNAs and Keratin-18" investigates the potential use of microRNAs (miRNAs) and Keratin-18 (K-18) as biomarkers for drug-induced liver injury (DILI). The study focuses on miR-21-5p, miR-34a-5p, and miR-122-5p in conjunction with K-18 markers M30 (apoptosis) and M65 (necrosis). The authors analyze the biomarker levels in patients with DILI, HBV-related hepatitis, and alcohol-related hepatitis, with an aim to develop a novel scoring model for differentiating DILI from other causes of acute hepatitis.

 

While the study presents intriguing findings, there are several areas that could be strengthened or clarified to enhance the impact and clinical relevance of the manuscript.

 

1. The manuscript discusses K-18 as a marker for apoptosis and necrosis in DILI, but further elaboration on the mechanistic link between K-18 and liver injury would be beneficial. Specifically, the manuscript should detail how K-18 levels fluctuate in response to cellular damage and why these markers are more specific or sensitive than other liver biomarkers in distinguishing between apoptosis and necrosis.

 

2. While the study focuses on miR-21-5p, miR-34a-5p, and miR-122-5p, the criteria for selecting these specific microRNAs among the numerous possible candidates remains unclear. The manuscript would benefit from a deeper explanation of the biological mechanisms that support their selection as markers for DILI. This could include insights into their roles in hepatocyte injury, regulation, or their previous validation in similar contexts.

 

3. The tables currently have excessive borders, making them visually cluttered. Simplifying the table design by reducing the number of borders or using shading for emphasis would improve readability and clarity. Consider reformatting the tables to highlight the key data in a more straightforward and concise manner.

 

4. In the bar graphs that summarize patient data, adding individual dots to represent each patient's data point would provide a more nuanced view of the data distribution. This could also help in identifying outliers or clusters that might otherwise be missed.

 

5. The ROC curves are crucial for demonstrating the predictive power of the model compared to existing methods. It would be beneficial to explicitly highlight how the proposed model outperforms traditional diagnostic tools, both in terms of sensitivity and specificity. This could strengthen the argument for the clinical utility of the new biomarkers.

 

6. The Discussion section would be stronger with an expanded commentary on how the proposed biomarkers could be validated in larger and more diverse cohorts. Furthermore, discussing potential pathways toward commercialization, such as steps required for regulatory approval and integration into clinical practice, would provide a more forward-looking perspective on the translational potential of the study findings.

 

7. The sample size of this study is relatively small, and it might limit the minimize the meaning of the results. It would be helpful to mention that larger-scale studies would be necessary to validate the findings in discussion section. 

Author Response

1. The manuscript discusses K-18 as a marker for apoptosis and necrosis in DILI, but further elaboration on the mechanistic link between K-18 and liver injury would be beneficial. Specifically, the manuscript should detail how K-18 levels fluctuate in response to cellular damage and why these markers are more specific or sensitive than other liver biomarkers in distinguishing between apoptosis and necrosis.

 We have better explained this mechanism in the Introduction. Full-length K18 and caspase-cleaved K18 fragments in serum reflect the degree of necrotic liver injury and/or apoptosis. K18 is the principal cytoskeletal protein of the hepatocytes and it is very important in the maintenance of cellular integrity. The cleavage of K18 at two distinct sites during cell death results in the release of two different peptides in the systemic circulation: M30 and M65.

The principal advantage of utilizing these markers is that they can diagnose early-stage of DILI. Identifying the type and mechanism of hepatic injury allows the severity of DILI to be evaluated and have a strong relationship with injury. Several studies have used these markers to identify DILI and the severity of injury.

2. While the study focuses on miR-21-5p, miR-34a-5p, and miR-122-5p, the criteria for selecting these specific microRNAs among the numerous possible candidates remains unclear. The manuscript would benefit from a deeper explanation of the biological mechanisms that support their selection as markers for DILI. This could include insights into their roles in hepatocyte injury, regulation, or their previous validation in similar contexts.

 We reported in the discussion published data demonstrating the role of the three considered miRNA in both DILI and other hepatic pathologies. In particular, miR-122 is a representative liver-specific miRNA candidate with potential clinical significance, miR-21 is one of the most frequently upregulated miRNAs in liver diseases, such as MASLD and HCC, and it is associated with poor overall survival. MiR-21 is reported to induce hepatic inflammation through promotion of inflammatory gene expression via the STAT3 signalling pathways, leading to liver disease. Lastly, miR-34 family regulates a number of different signalling pathways, including those linked to cancer, the immune system, metabolism and cellular structure.

We have added some references in the text to better support the selection of those 3 miRNAs.

3. The tables currently have excessive borders, making them visually cluttered. Simplifying the table design by reducing the number of borders or using shading for emphasis would improve readability and clarity. Consider reformatting the tables to highlight the key data in a more straightforward and concise manner.

 The tables have been simplified, as suggested by the Reviewer.

4. In the bar graphs that summarize patient data, adding individual dots to represent each patient's data point would provide a more nuanced view of the data distribution. This could also help in identifying outliers or clusters that might otherwise be missed.

 As suggested, individual dots have been added in Figure 3.

5. The ROC curves are crucial for demonstrating the predictive power of the model compared to existing methods. It would be beneficial to explicitly highlight how the proposed model outperforms traditional diagnostic tools, both in terms of sensitivity and specificity. This could strengthen the argument for the clinical utility of the new biomarkers.

 We thank the Reviewer for the comment. In the conclusions the following sentence has been added: “We concluded that a subset of miRNAs was altered in the plasma of DILI patients, showing specific differences depending on the type of liver injury (hepatocellular vs cholestatic). Moreover, when a model was created using this subset of miRNAs in association with other markers it was able to differentiate DILI from other pathologies.

6. The Discussion section would be stronger with an expanded commentary on how the proposed biomarkers could be validated in larger and more diverse cohorts. Furthermore, discussing potential pathways toward commercialization, such as steps required for regulatory approval and integration into clinical practice, would provide a more forward-looking perspective on the translational potential of the study findings.

The discussion has been modified adding the considerations suggested by the Reviewer.

7. The sample size of this study is relatively small, and it might limit the minimize the meaning of the results. It would be helpful to mention that larger-scale studies would be necessary to validate the findings in discussion section. 

 We are aware that the sample size is low, however, we would like to highlight that our study is monocentric and the number of patients enrolled is in line with the incidence of acute liver diseases. Nonetheless, the results obtained with this number of patients is encouraging, and we are working to involve other Italian and European centres, to confirm our preliminary data in a larger cohort.

In the Discussion section this consideration has been added: “Currently we are carrying out other studies to validate and confirm these preliminary findings, considering the diagnostic and prognostic potential of these markers.”

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript has been significantly improved in response to the previous round of comments, with all key points addressed thoughtfully and comprehensively. The revisions have enhanced the clarity and scientific rigor of the study. I have no further suggestions for improvement. I recommend the manuscript for publication.