Bone Metastases from Intrahepatic Cholangiocarcinoma Confer Worse Prognosis

Round 1

Reviewer 1 Report (Previous Reviewer 2)

1. Please descript more about the definition of bone metastasis in the text.

2. Please reform the Table 1 by the way of changing the item in X-axis into Y-axis like the Table 2. If possible, try to mix table 1 and 2 into one table. Some data may not enrolled and be overcome by descript in the text. The Table 1 looks like the reseach note which need to reform for publication.

3. Please add more references for reader and enrich the content of discussion.

4. Reference format need to match the reguirement of this journal.

Author Response

1. Please descript more about the definition of bone metastasis in the text.

More about the diagnosis/definition of bone metastasis was added in the text (“Materials and Methods”).

2. Please reform the Table 1 by the way of changing the item in X-axis into Y-axis like the Table 2. If possible, try to mix table 1 and 2 into one table. Some data may not enrolled and be overcome by descript in the text. The Table 1 looks like the reseach note which need to reform for publication.

The Table 1 and Table 2 were united, thank you for this suggestion. We extended description in the text.

3. Please add more references for reader and enrich the content of discussion.

We added more references. We tried to cite all important studies related to our research, though there are not so many.

4. Reference format need to match the reguirement of this journal.

Reference format was corrected.

Reviewer 2 Report (New Reviewer)

This is a retrospective study which retrospectively investigated the prognostic role of bone metastases in patients affected by intrahepatic cholangiocarcinoma. The authors showed that around 11% of ICC patients developed bone metastases. Furthermore, the authors demonstrated that the presence of bone metastases, poor performance status and no subsequent second-line treatment was associated with worse outcome.

This study was conducted well, and the methods are appropriate.

The results will be of interest to clinicians in the field.

However, the following minor issues require clarification:

Minor

1.     Please provide regimens used in 2^nd line treatment, which can influence the outcomes.

2.     Number of patients in each group should be also described in the main text.

3.     (Table 1) The explanations of Group A-C and the word of “ECOG PS” are redundant. The range of OS and PFS should be provided.

4.     There are some previous studies which showed similar results. I recommend that the authors emphasize the novelty and originality in this study.

Author Response

This is a retrospective study which retrospectively investigated the prognostic role of bone metastases in patients affected by intrahepatic cholangiocarcinoma. The authors showed that around 11% of ICC patients developed bone metastases. Furthermore, the authors demonstrated that the presence of bone metastases, poor performance status and no subsequent second-line treatment was associated with worse outcome.

This study was conducted well, and the methods are appropriate.

The results will be of interest to clinicians in the field.

However, the following minor issues require clarification:

Minor

1. Please provide regimens used in 2^ndline treatment, which can influence the outcomes.

All patients received gemcitabine-based chemotherapy in the first-line. In general, the combination treatment (gemcitabine and cisplatin or oxaliplatin) was reserved for patients in good clinical conditions (ECOG PS 0-1) without contraindications. The patients with poor clinical conditions (ECOG PS 2 or 3) or with contraindications were treated with gemcitabine in monotherapy. We did not include in this study patients who were treated with other type of chemotherapy in the first line (in particular, no addition of durvalumab, as it was not available at the time of the study). All 90 patients who underwent the second line treatment were treated with 5FU-based therapy.

2. Number of patients in each group should be also described in the main text.

We specify the number of patients also in the written text in “Results” chapter (not only in the Table 1) to be more clear.

3. (Table 1) The explanations of Group A-C and the word of “ECOG PS” are redundant. The range of OS and PFS should be provided.

We united Table 1 and Table 2. The range of OS and PFS was provided.

4. There are some previous studies which showed similar results. I recommend that the authors emphasize the novelty and originality in this study.

The novelty and originality of this study has been emphasized in the discussion (in red).

Reviewer 3 Report (New Reviewer)

Dear authors,

I've read your paper with great interest and although the topic is of great clinical importance, the paper has various weak points which must be fixed before the paper can be published. Please find my comments in the attached pdf file.

Briefly:

materials and methods are scarce

patients are insufficiently characterized

Statistics must be thoroughly revised

Table and figure captions are insufficiently detailed.

Best of luck!

Looking forward to your revised work!

Comments for author File: Comments.pdf

Author Response

Dear authors,

I've read your paper with great interest and although the topic is of great clinical importance, the paper has various weak points which must be fixed before the paper can be published. Please find my comments in the attached pdf file.

Briefly:

materials and methods are scarce

patients are insufficiently characterized

Statistics must be thoroughly revised

Table and figure captions are insufficiently detailed.

We appreciated your suggestions and your time, we are happy to better our study thanks to your observations (in red in the text).

Round 2

Reviewer 3 Report (New Reviewer)

Dear Authors,

I am glad you were able to incorporate in your work most of my suggestions. However, it is a good general practice to provide a point-by-point response to my comments, especially to those you have not applied.

Looking forward to your answers.

Comments for author File: Comments.pdf

Author Response

Dear reviewer:

• In Material and Methods – Statistical Analysis: all suggested corrections were implemented, in red

• Figure 1: Figures are still to be improved. Fonts are really small: that is standard size, proportional to the size of the Figures.

• Figure 2: all suggested corrections were implemented, in red

• Table 3, Multivariate COX analysis of overall survival: how are they in the univariate analysis? add this data in the table. Moreover, you previously had liver meta here as well. Why leave it out now? you should have all 3 groups here. maybe have group A as a ref for B and C and build your model like that. Moreover, you have not explained how you chose your variables for the multivariate analysis. Clinically relevant variables? only significant ones in the univariate? p value <0.1 or 0.2 in in the univariate?

We used your suggestion and done the new cox multivariate analysis: as a reference Group A was chosen, so it was reference for Group B and Group C. Cox proportional hazards model analysis showed that poorer ECOG PS, absence of second-line treatment and the presence of bone metastases were associated with a significantly higher HR for death, see Table 4.

We chose just clinically relevant variables for the multivariate analysis.

We added also univariate analysis for variables, as suggested (Table 3). Univariate analysis revealed that ECOG PS 1 or higher, the absence of second-line treatment and the presence of bone metastases, were associated with a significantly higher HR for death) (Table 3).

• References: I am sorry but I forgot to ask last time why you have so few references? No problem. Because there are not so many papers published for this topic. We cited all relevant studies. Though, we extended references and enriched the introduction and discussion part, all in red.

Round 3

Reviewer 3 Report (New Reviewer)

Dear Authors,

Thank you for your time and your extensive improvements. A few minor considerations:

Table 1. there's a lot of white space. Also you can add second line chemotherapy.

Figures are too small, especially the text in the labels. If you don't know how to fix this you can consult the SPSS manual regarding graphs and templates. You can pretty much adjust everything.

Figure 3 c is smaller than a and b.

For the univariable and multivariable anaysis please add the ref groups for each variable tested.

Abbrev must be explained in the captions of all tables and figures so that they are self standing a can be understood without having to read all the paper. please provide the brief description of the groups as well (ex: Group A: patients with liver metastases only, Group B etc etc).

Add that you used Chi Square test in mat and methods.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The authors performed a retrospective chart review of 186 patients with intrahepatic cholangiocarcinoma. Their aim was to compare patients with bone metastases against those who have intrahepatic metastases only. They find no difference in progression-free survival, but a difference in overall survival. Given the short survival of 4 months with bone metastases, they reflect on the futility of chemotherapy and conclude that palliative care should be recommended.  They also analyze the cases based on ECOG performance status and find that lower PS is associated with earlier death, especially for more advanced stages of disease.

MAJOR POINTS

Intrahepatic multifocal disease is AJCC/UICC category T2, stage II disease while bone metastases are category M1, stage IV.   That distant mets are an adverse prognostic factor is well known for all cancer types  and probably does not need to be demonstrated again. If it does, this would be better accomplished with large population-level database studies rather than single-institution experience.

Criteria for defining liver metastases versus multifocal involvement in the liver should be discussed and described. I am accustomed to thinking of these simply as multifocal involvement. Generally we do not think of involvement within the same organ as a form of metastasis, but if the authors are doing this, they should define it.

A power analysis is not performed. There is evidence that the study may be underpowered and at risk of Type II errors (specifically, they fail to find a significant survival difference between groups A and B although group B has distant metastases).

Did any patients have surgical resection? If they did, the role of this factor must be explored as it could confer a survival advantage on those who had it (presumably all group A).

In the analysis in Table 2, I am no statistician, but it seems like group B has been chosen as the baseline since a hazard ratio is shown for membership in group C (bone mets) and group A (liver-only mets). This is a bit counterintuitive. It would make more sense to use group A as the baseline and report HR for membership in group B and C.

MINOR POINTS

The manuscript is well prepared but minor editing for correct English usage is needed.

Line 76, metastatic “sides” -> “sites”

In Table 1, if the purpose is to enable a comparison between groups, it would be more standard to have three columns for groups A, B, C. This gives the authors a place to show a statistical test for difference between the groups in a fifth column. The purpose of this paper is to show the differences between groups so this should be done for the demographics.

The authors place patients in three groups: group A is liver involvement only, group C is bone mets with or without other metastatic sites, and group B is all patients who are not in group A or C (in other words they have mets outside the liver that are not in bone).  In Table 1 and Figure 1-2 they refer to group B as “all ICC group” which is somewhat misleading since the group is NOT consisting of all the ICC.

Reviewer 2 Report

1. please provide the definition and profiles of diagnosis of bone metastasis?

2. What's chemo-agents used for the fisr-line and second-line treatment methods ? and  guide-line?

3, Table 1 needed to be re-form

4. It is better to identify the name of subgroup in the figures samed as the text, Group A, B, and C.

5. Coding from references were not enough and missing for some data in the section of discussion.

6. Format of reference needed to be completed according to the journal regulation for author.