Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors performed a retrospective analysis of a limited number of patients treated with methotrexate (MTX). The objective was to assess whether metabolic syndrome or the cumulative dose of MTX is likely a risk factor for liver injury and fibrosis.

1. The authors talk about methotrexate-induced liver injury. It would be useful to provide ALT/AST values for the patients. This information should be available for all patients, not just those undergoing biopsy.
2. Table 1: For age, BMI, and dose, the authors need to provide the mean and SE or range.
3. Did all patients receive the same dose of methotrexate? If not, the time from the start of treatment to the onset of liver injury should be reported.
4. Table II: the actual data should be given for the 2 groups (mean ± SE and n) and not just a p value.
5. If MTX can cause fatty liver disease and fibrosis, is it the drug alone, previous obesity and MASLD, or the combination that triggers the hepatotoxicity? Are baseline values of liver injury and fibrosis available for any of the patients to get an idea of how much these parameters increase with drug treatment and when?
6. Line 219: The authors claim that they described the utility of VCTE in detecting MASLD…in patients with MTX use (ref. 7). However, Ref. 7 is a review on MASLD epidemiology. What is the correct reference?
7. Line 237 -240: The authors claim that previous studies have shown similar results. These studies need to be cited.

Author Response

Comment 1: The authors talk about methotrexate-induced liver injury. It would be useful to provide ALT/AST values for the patients. This information should be available for all patients, not just those undergoing biopsy.

Response 1:We appreciate this suggestion. The primary aim of our study was to correlate histological findings and non-invasive tests (LSM, CAP, and histology) to define liver injury. For this reason, we focused on reporting histological and elastography data rather than biochemical markers. While ALT and AST are useful markers of hepatocellular injury in general, their levels may not reliably reflect underlying fibrosis or steatosis in chronic liver disease. The focus of our analysis is based on structural measures of chronic liver injury.

Comment 2: Table 1: For age, BMI, and dose, the authors need to provide the mean and SE or range.

Response 2:Thank you for this helpful suggestion. We have updated Table 1 to include the mean ± standard deviation (SD) and the range for age, BMI, and cumulative MTX dose. We chose to report SD rather than standard error (SE), as SD is the convention in clinical studies for describing the variability within a study population, whereas SE reflects the precision of the estimated mean, thus rarely used to describe demographics of populations.

Comment 3: Did all patients receive the same dose of methotrexate? If not, the time from the start of treatment to the onset of liver injury should be reported.

Response 3:We thank the reviewer for this important point. Patients in our cohort did not all receive the same MTX dose/regimen, as dosing varied depending on indication (e.g., rheumatologic, dermatologic, oncologic). For this reason, we focused on cumulative dose as a more clinically meaningful measure of exposure than an absolute weekly dose or duration of treatment FibroScan. Prior studies have also demonstrated cumulative dose (weekly dose x weeks of use) to be a more robust predictor of hepatic injury risk. In our analysis, cumulative dose was included as a covariate, and we found that higher LSM and CAP values were independent of cumulative MTX dosage. Thus, cumulative dose would be a good measure of both dose and duration of MTX use prior to liver disease assessment.

 

Comment 4: Table 2: the actual data should be given for the 2 groups (mean ± SE and n) and not just a p value.

Response 4: We appreciate this comment. The table was previously mislabeled and has now been relabeled as Table 2. We have updated the Table to present the mean ± standard deviation (SD) and the sample size for each group, alongside the p values. We also included 95% confidence intervals to provide additional clarity. We chose SD instead of SE, as SD more accurately reflects variability within the study population and is the convention in clinical cohort reporting.

Comment 5: If MTX can cause fatty liver disease and fibrosis, is it the drug alone, previous obesity and MASLD, or the combination that triggers the hepatotoxicity? Are baseline values of liver injury and fibrosis available for any of the patients to get an idea of how much these parameters increase with drug treatment and when?

Response 5: We appreciate the reviewer’s important question. Because this was a retrospective study, patients were already on MTX at the time of hepatology referral for evaluation, so pre-treatment histology or elastography data were not obtained. Liver enzymes were not included in the analysis as they do not reliably reflect fibrosis or steatosis status. Instead, our study focused on correlating liver histology and LSM in the context of cumulative MTX exposure and metabolic risk factors. We believe both MTX and metabolic syndrome contribute to hepatotoxicity, but in long term MTX uses the presence metabolic syndrome emerged as the stronger driver of steatosis and steatohepatitis in this cohort. We agree that future prospective studies with pre-treatment baseline assessments would help clarify the temporal contribution of MTX versus pre-existing metabolic liver disease.

Comment 6: Line 219: The authors claim that they described the utility of VCTE in detecting MASLD…in patients with MTX use (ref. 7). However, Ref. 7 is a review on MASLD epidemiology. What is the correct reference?

Response 6: We thank the reviewer for catching this. The correct citation should be Brotherton et al., EMJ 2024, which specifically studied VCTE in MTX users. Line 232:“We have previously described the utility of VCTE in detecting MASLD and correlated the LSM and CAP scores with steatosis and fibrosis noted on liver biopsies in these patients with chronic MTX use [14].”

 

Comment 7: Line 237 -240: The authors claim that previous studies have shown similar results. These studies need to be cited.

Response 7: We have now cited Bedoui et al. (2019), Lertnawapan et al. (2023), and Bauer et al. (2017), which reported similar findings on MTX hepatotoxicity and metabolic risk factors.

Line 256: “We acknowledge that several studies have previously examined the association between MetS and liver injury in chronic MTX users. While prior studies have explored dose-dependent MTX hepatotoxicity, recent evidence increasingly highlights the importance of metabolic risk factors [1,3,8].”

Reviewer 2 Report

Comments and Suggestions for Authors

The topic is cutting-edge but i think the authors should improve the quality of their papers.

1) In table 2, report the point estimates, not only the p values

2) Refrain from using Roman numbers but use Arabic ones

3) The authors should improve the Discussion that is too short. Comment more on the state of the art, for example on what is known about the impact of diabetes and metabolic syndrome on the clinical course of liver patients from the diagnosis till the final complications, for example HCC (in this regard cite the relevant paper PMID: 23845075)

4) The retrospective design and the limited sample size represent major limitations and should be discussed in the paper

5) Report the 95% CIs along with the summary estimates

Author Response

Comment 1: In table 2, report the point estimates, not only the p values

Response 1: We appreciate this comment. For clarity, the biopsy results are now designated as Table 3 (previously mislabeled). This table has been updated to report the point estimates mean +/- SD for each biopsy characteristic in addition to p values.

Comment 2: Refrain from using Roman numbers but use Arabic ones.

Response2 : We thank the reviewer for this observation. We have revised the manuscript to use Arabic numerals consistently instead of Roman numerals.

Comment 3: The authors should improve the Discussion that is too short. Comment more on the state of the art, for example on what is known about the impact of diabetes and metabolic syndrome on the clinical course of liver patients from the diagnosis till the final complications, for example HCC (in this regard cite the relevant paper PMID: 23845075)

Response 3: We appreciate this feedback. We have expanded the Discussion section to provide a broader overview of the current state of knowledge, including the impact of diabetes and metabolic syndrome on liver disease progression and outcomes such as hepatocellular carcinoma. We have incorporated the suggested citation (PMID: 23845075) into this section.

LINE 208 “This finding challenges the traditional belief that MTX hepatotoxicity is primarily dose-dependent, highlighting instead the critical role of metabolic risk factors. Our results align with broader evidence showing that diabetes and MetS accelerate the progression of chronic liver disease from steatosis to fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) [19]. This is consistent with studies demonstrating that patients with MetS or diabetes develop more rapid fibrosis progression, have higher rates of hepatic decompensation, and experience worse long-term survival compared to those without metabolic risk factors. In this context, MTX exposure may act as an additional insult on an already vulnerable liver, compounding the injury driven by underlying metabolic dysfunction. Importantly, this paradigm shift emphasizes that liver injury in MTX users may not be explained solely by cumulative drug exposure, but by the synergistic interplay between MTX and the metabolic milieu. These observations underscore the need for objective, reliable measures such as VCTE and biopsy to disentangle the relative contributions of MTX and metabolic dysfunction, and to guide risk-stratified monitoring strategies.”

Line 224 “The hypothesis that MetS may mediate drug-induced liver injury is further supported by studies of other hepatotoxic medications such as tamoxifen, valproic acid, and lamotrigine, where metabolic risk factors modify risk [10,19]”

Comment 4: The retrospective design and the limited sample size represent major limitations and should be discussed in the paper

Response 4: We agree with the reviewer and have explicitly added these points to the limitations section of the Discussion.

Line 239: “This study has several limitations that should be considered when interpreting its findings. The retrospective design introduces potential biases, including incomplete or inconsistent medical records, and the single-center nature of the study limits the gener-alizability of results to broader populations. Additionally, patient selection bias may have occurred since only those referred to hepatology clinics were included, potentially overrepresenting more severe disease. The overall sample size was modest, and only a subset (n=32) underwent biopsy, which reduces statistical power and limits generalizability. The reliance on secondary data, such as medical records and pharmacy refill re-ports, may lead to inaccuracies in assessing MetS and cumulative MTX dosage. Variability in LSM and CAP values, particularly in the MetS group, suggests unmeasured hetero-geneity, and the retrospective design precludes causal inferences. While we used es-tablished criteria to define metabolic syndrome, we acknowledge that reliance on chart documentation may have introduced some misclassification. Finally, the study's focus on MetS and MTX dosage does not account for other potential contributors, such as lifestyle factors or genetic predispositions, and the predominance of rheumatologic indications and lack of demographic diversity further limit its applicability to other patient populations.”

Comment 5: Report the 95% CIs along with the summary estimates

Response 5: Thank you for this suggestion. We have added 95% confidence intervals alongside summary estimates throughout the results tables.

Reviewer 3 Report

Comments and Suggestions for Authors

The present study evaluated the association between MetS and MTX-associated liver injury using vibration-controlled transient elastography (VCTE) and liver biopsy in patients suspected to have MTX related liver injury. The results showed that MetS is strongly associated with liver injury in chronic MTX users, independent of cumulative MTX dosage. The study was overall well conducted and the results are interesting. There are some points that should be further addressed by the authors.

1. The sample size is relatively small, and only a subset (n=32) underwent liver biopsy. This may limit the statistical power and generalizability of the results.
2. The diagnostic criteria of metabolic syndrome should be supported by appropriate references.
3. The retrospective design may introduce patients selection bias.

Author Response

Comment 1: The sample size is relatively small, and only a subset (n=32) underwent liver biopsy. This may limit the statistical power and generalizability of the results.

Response 1: We acknowledge this important limitation. We have explicitly discussed the small sample size and limited biopsy subgroup in the Discussion section, noting that it may reduce statistical power and generalizability of our findings.

Line 242: “The retrospective design introduces potential biases, including incomplete or inconsistent medical records, and the single-center nature of the study limits the generalizability of results to broader populations. Additionally, patient selection bias may have occurred since only those referred to hepatology clinics were included, potentially overrepresenting more severe disease. The overall sample size was modest, and only a subset (n=32) underwent biopsy, which reduces statistical power and limits generalizability.”

Comment 2: The diagnostic criteria of metabolic syndrome should be supported by appropriate references.

Response 2: We agree with the reviewer. We have added appropriate references to support the diagnostic criteria of metabolic syndrome used in this study.

“Patients were classified as having metabolic syndrome if they met three or more of the following criteria: (1) history of DM or treatment for DM, (2) history of HTN or treatment for HTN, (3) elevated triglycerides or treatment for elevated triglycerides, (4) low high-density lipoproteins (HDL) or treatment for low HDL, and (5) elevated BMI [15]. Traditionally, waist circumference is traditionally used for diagnosis, but because this is not traditionally a portion of vital signs taken in clinic, BMI was used as a surrogate.”

Comment 3: The retrospective design may introduce patients selection bias.

Response 3: We concur with the reviewer. We have highlighted in the Discussion that the retrospective design introduces potential selection bias, particularly since the cohort was derived from patients referred to hepatology clinics. Please see comment 1.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript still needs to be improved, particularly the Discussion. The impact of diabetes was not discussed as suggested and reference 20 is not relevant in that context. For sake of clarity i repeat my previous comment: 

The authors should improve the Discussion that is too short. Comment more on the state of the art, for example on what is known about the impact of diabetes and metabolic syndrome on the clinical course of liver patients from the diagnosis till the final complications, for example HCC (in this regard cite the relevant paper PMID: 23845075)

Author Response

Comment 1: The manuscript still needs to be improved, particularly the Discussion. The impact of diabetes was not discussed as suggested and reference 20 is not relevant in that context. For sake of clarity, I repeat my previous comment: The authors should improve the Discussion that is too short. Comment more on the state of the art, for example on what is known about the impact of diabetes and metabolic syndrome on the clinical course of liver patients from the diagnosis till the final complications, for example HCC (in this regard cite the relevant paper PMID: 23845075).

Response 1: Thank you for this valuable feedback. We agree with this comment and have taken careful time to restructure and expand the Discussion to provide a more comprehensive integration of the current evidence. We intentionally incorporated new citations that we believe best represent the state of the art on diabetes mellitus (DM), metabolic syndrome (MetS), and their influence on the progression of chronic liver disease. Therefore, we have done the following to improve the quality of the article.

1. Replaced the prior reference with Younossi et al., Hepatology 2013 (PMID: 23845075) [24] as suggested by the reviewer.
2. Added additional mechanistic and clinical references, please see the references [25-28] for the updated references.
3. Expanded the Discussion significantly to include the impact of diabetes and MetS on liver disease progression from steatosis through fibrosis and HCC, and clarified the biologic mechanisms, such as mitochondrial dysfunction, oxidative stress, autophagy impairment, and hepatic stellate cell activation, that underlie this process.
4. Discussed the cytotoxic pathways of methotrexate (MTX) and their potential synergy with metabolic dysfunction, emphasizing how this “two-hit” interaction may accelerate hepatocellular injury.
5. Added new text on the clinical implications of these findings, emphasizing that individuals with MetS or diabetes on long-term MTX should undergo more frequent hepatic monitoring (e.g., VCTE, laboratory follow-up), while those without metabolic risk factors may not require as frequent assessments—improving both patient safety and healthcare resource utilization.

Please see the re-worked discussion at Lines 203 – 268.

Author Response File: Author Response.docx