Review Reports - Optimal Puncture Number and Tissue Evaluation Method in Endoscopic Ultrasound-Guided Fine-Needle Biopsy for Patients with Malignant Neoplasm of Pancreas

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper is well written but i think lacks of novelty. Other limitations:

1) Small sample size and retrospective design

2) The article is overall short. The discussion should be improved with several comments, for example the authors should comment on the potential role of ancillary techniques on tissue sampling (cite PMID: 33481633)

3) All these concepts are widely described in the latest ESGE technical review on tissue sampling that the authors should acknowledge

4) Some endoscopic images would improve the quality of the manuscript

5) There are papers summarizing the suction techniques, for example a network meta-analysis published a couple of years ago in Gastrointestinal Endoscopy.

Author Response

Comments Reviewer 1

We are grateful for this reviewer’s positive response and have addressed the raised issues below.

Small sample size and retrospective design

We acknowledge that this study is a retrospective, single-center analysis with a relatively limited sample size, which is an inherent limitation. However, several points merit clarification.

First, the study population represents a consecutive, real-world cohort of patients with confirmed pancreatic malignancy who underwent EUS-guided fine-needle biopsy under uniform procedural conditions during a defined period. To minimize heterogeneity and enhance internal validity, we excluded cases diagnosed by non-EUS modalities, metastatic-site biopsies, and procedures with fewer than three punctures from initially included 285 patients, resulting in a well-characterized and clinically relevant cohort.

Second, although the absolute sample size may appear modest, the number of included patients (n = 106) is comparable to that of several prior single-center studies evaluating EUS-FNB techniques. Also, our study focuses on within-patient comparisons across sequential needle passes, which reduces inter-individual variability and allows meaningful assessment of diagnostic performance trends despite a limited cohort size.

Finally, the retrospective design reflects real-world clinical practice in centers without routine access to ROSE, which remains the case in many institutions worldwide. As such, our findings provide pragmatic evidence supporting a simplified, two-pass EUS-FNB strategy guided by gross-eyed evaluation. We agree that prospective, multicenter studies are warranted to further validate these results, and this has been explicitly acknowledged as a limitation in the revised Discussion section.

We have revised the manuscript to more clearly emphasize these limitations. This issue has been mentioned in the revised version. Please see line 322 to 325.

The article is overall short. The discussion should be improved with several comments, for example the authors should comment on the potential role of ancillary techniques on tissue sampling (cite PMID: 33481633)

We thank the reviewer for this valuable comment. The Discussion section has been expanded to address the potential role of ancillary techniques in EUS-guided tissue sampling, including contrast-enhanced harmonic EUS–guided sampling, as suggested (PMID: 33481633). We discuss how such techniques may complement modern FNB needles and gross-eyed assessment in selected cases. These revisions strengthen the Discussion and improve the overall depth of the manuscript. We have added this issue in the revised manuscript. Please see line 285 to 299

All these concepts are widely described in the latest ESGE technical review on tissue sampling that the authors should acknowledge

We thank the reviewer for this important remark. We have revised the Discussion to explicitly acknowledge the latest ESGE technical and technology review on EUS-guided tissue sampling and have discussed how the concepts evaluated in our study align with these recommendations. This reference has been added to place our findings within the context of current ESGE guidance. We have added this issue in the revised manuscript. Please see line 229 to 231.

Some endoscopic images would improve the quality of the manuscript

We thank the reviewer for pointing out this issue. We have added the figure of specimen acquired by EUS-FNB using fanning technique. Please see the revised version. (Figure 2).

There are papers summarizing the suction techniques, for example a network meta-analysis published a couple of years ago in Gastrointestinal Endoscopy.

We thank the reviewer for this helpful comment. We have revised the manuscript to acknowledge prior studies summarizing suction techniques, including a network meta-analysis by Giri et al. which found no single suction method to be superior in terms of sample adequacy or diagnostic accuracy in EUS-guided tissue acquisition (Endosc Int Open 2023). Please see line 266 to 274.

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for allowing me to review your article. This is very interesting topic. Few comments as noted below

Please clarify your definitions.

I strongly believe defining "Diagnostic accuracy" requires comparison between both malignant and benign cases. This study includes only confirmed pancreatic malignancies, with no benign comparator group. Therefore, use of the term diagnostic accuracy is methodologically seems to be inappropriate.

Diagnostic yield definition Is Inconsistent with the study design - Diagnostic yield is defined as a definitive benign or malignant diagnosis, yet no benign cases were included.

Define "surgical pathology" - are you referring to pathology from EUS biopsy or true surgical resection of pancreatic cancer with a final surgical pathology? This is important because - many patients (Stage IV) would not be a surgical candidate to have surgical pathology

As this is a retrospective study, it is unclear how diagnostic performance was determined for individual passes. Please state whether pass-specific outcomes were assessed retrospectively. If so, how - is this via sequential, cumulative, or independent specimen review.

PDAC, NET, GIST, and other malignancies were pooled together but they do differ in diagnostic yields. Provide clearer breakdown of tumor types/number of patients and justify pooling or include subgroup analysis.

Statements such as “significant improvement” between passes and for others in the results section are made without reporting p-values, confidence intervals, or specifying statistical tests. This is very important for a reader to understand if there was real statistical significance. If not, kindly revise language to remove claims of significance.

There are many confounder we have for this study, was there a multivariate analysis performed, if not why?

Minor spelling error

Line 17: “techique” - change it to technique

Author Response

Comments Reviewer 2

We are grateful for this reviewer’s positive response and have addressed the raised issues as follows:

Comment 1:

I strongly believe defining "Diagnostic accuracy" requires comparison between both malignant and benign cases. This study includes only confirmed pancreatic malignancies, with no benign comparator group. Therefore, use of the term diagnostic accuracy is methodologically seems to be inappropriate

Response:

We agree that the definition of diagnostic accuracy could be improved. We have changed this term to Diagnostic rate of malignancy in figure 4.

Comment 2:

Diagnostic yield definition Is Inconsistent with the study design - Diagnostic yield is defined as a definitive benign or malignant diagnosis, yet no benign cases were included.

Response:

We agree that only malignant lesions were included and have changed the term to yield rate of malignancy

Comment 3:

Response:

Thank you for pointing out this important information. This final diagnosis is based on histology diagnosis including surgical and/or EUS biopsy result.

Comment 4:

Response:

Thank you for the informative question. Every pass was regarded as individual specimen and analyzed cumulatively. For example, if patient A had three EUS samples, we regarded patient A had 3 independent samples.

For patient A, if the first specimen is negative but the second specimen is positive, we will present negative for 1 puncture and positive for 2 punctures. If the third puncture is still negative, because the second puncture is positive, the result for 3 punctures is still positive.

Comment 5:

Response:

Thank you for the constructive comment. NET, GIST, and other malignancy present different molecular biology with PDAC even with each other. However, malignant tumors have risk of tumor seeding even for NET and other malignancy in EUS-FNB. They all shared the similarity of tumor seeding and bleeding risk. Therefore, we decided to pool them together. To answer this question, we collected more cases to check and found that we saw the trend of increased diagnostic accuracy by increased puncture numbers. However, the difference is not significant.

Please see the figure below.

Comment 6:

Response:

By using Chi-square test, the p value is 0.03 between first pass and two passes in diagnostic rate of malignancy.

Comment 7:

There are many confounder we have for this study, was there a multivariate analysis performed, if not why?

Response:

Thank you for pointing this out, and we have added multivariate analysis, which showed no significant difference either. Please see Table 2.

Minor spelling error

Line 17: “techique” - change it to technique

Response:

Thank you for the informative correction, we have sent our manuscript for english editage.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This is a contribution to the "tissue is the issue". concept of EUS-guided aspiration or biopsy of solid pancreatic masses (SPM).

In case the authors provide a brief suggested algorithm for the biopsy technique for SPM with illustration, if possible, it may be useful for the readers' comprehension.

Further, it may be useful to provide a picture showing the appearance of obtained white and red tissue.

Author Response

Comments Reviewer 3

We are grateful for this reviewer’s positive response and have addressed the raised issues below.

In case the authors provide a brief suggested algorithm for the biopsy technique for SPM with illustration, if possible, it may be useful for the readers' comprehension.

We have revised the manuscript to more clearly illustrate biopsy technique for SPM with a picture. Please see Figure 1.

Further, it may be useful to provide a picture showing the appearance of obtained white and red tissue.

We thank the reviewer for this valuable comment. We have added a picture of obtained white or red tissue in the revised manuscript. Please see Figure 2.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised manuscript is OK. Thank you!

Reviewer 2 Report

Comments and Suggestions for Authors

Appreciate for making the necessary changes/revisions based on my comments. It looks overall well presented now. All the best

Reviewer 3 Report

Comments and Suggestions for Authors

Thanks to authors who added the relevant figures.