A Proof-of-Concept Pilot Study of Contrast-Enhanced Ultrasound as a Potential Alternative to Contrast-Enhanced Magnetic Resonance Imaging in the Surveillance of Hepatocellular Adenoma and Focal Nodular Hyperplasia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript addresses an interesting and clinically relevant question, namely whether contrast-enhanced ultrasound (CEUS) can be comparable to MRI for longitudinal follow-up of focal nodular hyperplasia (FNH) and hepatocellular adenoma (HCA), with potential advantages in terms of cost. While the overall concept is reasonable, there are several important methodological and reporting issues that need to be addressed before the conclusions can be supported.

1. Characterization of subcentimeter nodules

In Table 2, nodules as small as 5 mm are included. It is unclear whether subcentimeter lesions can be reliably characterized as FNH or HCA using either MRI or CEUS. According to current guidelines and clinical practice, reliable imaging-based diagnosis of FNH or HCA generally requires larger lesions with typical imaging features. The authors should clarify how a 5 mm nodule was diagnosed and whether such lesions should be included in the analysis.

2. Clarification of the number of paired MRI–CEUS measurements
The manuscript reports 92 paired MRI–CEUS measurements, while the total number of lesions is stated to be 41. At baseline, 41 lesions underwent both MRI and CEUS, resulting in 41 paired measurements. Does the number 92 refer to follow-up MRI–CEUS examinations? If so, does this imply a total of 133 measurements (41 baseline + 92 follow-up)?
The authors should clearly explain the structure of the dataset, including:

 - the total number of follow-up examinations,

 - the average number of follow-up studies per lesion,

 - and how these repeated measurements were handled in the analysis.

3. Definition of “relative change of the sum of lesion diameters” (Table 2)
The calculation of the “relative change of the sum of lesion diameters between observations” requires more detailed explanation. It is particularly important to clarify how this metric was derived when lesions underwent more than two follow-up examinations.

Given that many lesions appear to have been followed more than twice, the repeated measurements within the same lesion are not independent. Comparing these percentage changes using the Wilcoxon signed-rank test may therefore be statistically inappropriate. A statistical approach that accounts for repeated measurements and clustering at the lesion level, such as a linear mixed-effects model, would likely be more suitable.

4. Follow-up scheme and intervals
The manuscript lacks sufficient detail regarding the follow-up protocol. The authors should report:

- the number of follow-up examinations per lesion (mean, range, and maximum),

- the follow-up intervals,

- and the maximum duration of follow-up.

This information is essential to properly interpret the longitudinal analysis.

5. Pooling of FNH and HCA into a single study group
Although combining FNH and HCA may increase statistical power, these two entities are biologically and clinically distinct. In particular, HCA carries malignant potential and warrants separate consideration. Subgroup analyses for FNH and HCA should therefore be provided.
In addition, graphical visualization of longitudinal size changes (e.g., spaghetti plots) would help readers better understand lesion behavior over time in each subgroup.

6. Discrepant assessment of progression between MRI and CEUS (Figure 3)
Figure 3 shows five cases in which lesions were considered stable on MRI but progressive on CEUS. In the Discussion, the authors suggest that CEUS measurements may overestimate lesion size due to oblique diameter measurements. However, this explanation is insufficient, as interval percentage changes were calculated using CEUS measurements alone.
The authors should further elaborate on the potential reasons for these discrepancies. Presenting one representative case as an illustrative figure would significantly strengthen the discussion and improve transparency.

 

In summary, while the study addresses a relevant clinical question and proposes a potentially useful role for CEUS in longitudinal follow-up, substantial clarification of the methodology, statistical analysis, and data presentation is required to support the authors’ conclusions.

 

Author Response

We thank the Reviewer for the thorough and constructive review. Detailed responses to all comments and corresponding revisions can be found in the attached document. All changes introduced in the manuscript have been highlighted in red.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses some interesting clinical issues, such as the need to improve differential diagnosis with relevant prognostic implications and the importance of optimizing patient monitoring in case of suspicious and/or potentially evolving lesions, using a non-invasive modality such as imaging integrated with standardized and validated clinical approaches.

The authors' critical analysis clearly highlighted that, although CEUS can be a practical aid in monitoring stable lesions, it is important to underline that MRI is currently the gold standard for evaluating suspicious progression in this context.

The English language and the organization of the sections are reported in a simple and clear way.

Overall, I would suggest clearly highlighting that this is a proof-of-concept pilot study, due to the limited sample size, as briefly discussed in manuscript limitations.

This observation would be consistent with limited, descriptive and/or nonparametric statistical analysis.

Typos: I suggest checking how some reference and figure numbers are reported in the manuscript.

Results:

• I would suggest integrating the anamnestic details on predisposing factors in accordance with those interestingly described in the introduction, in addition to data on gender and age prevalence.
• I would suggest integrating further representative examples of CEUS complementarity and limits compared to MRI, as reported in figure 1.

 

Discussion:

• I would suggest to deeply discuss about strenghts and limitations of the present study compared to reference [15] and [16], which compare MRI and CEUS performances on the same lesions type not serially but in a wider cohort and with histology confirmation, reaching slightly different or more detailed results.

Author Response

We thank the Reviewer for the thorough and constructive review. Detailed responses to all comments and corresponding revisions can be found in the attached document. All changes introduced in the manuscript have been highlighted in red.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have provided appropriate responses to the previously raised comments, and the manuscript has been revised accordingly. However, there is still a point that remains unclear from the reader’s perspective.

 

According to the manuscript, the dataset consists of 41 baseline examinations and 51 total follow-up examinations. If so, response categories such as regression, stable disease, and progression should be determined based on the follow-up examinations.

Therefore, the total number of cases presented in Figure 3 should sum to 51. However, the numbers shown in Figure 3 add up to 92, which is inconsistent with the stated number of follow-up examinations.

This discrepancy requires correction and clarification.

Author Response

The response to the review can be found in the appendix.

Author Response File: Author Response.pdf