Prognostic Value of Performance Status, Albumin, and CRP in Last-Line Chemotherapy for Pancreatic vs. Other Gastrointestinal Cancers—Simple Tools Matter

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Congratulations to Westgaard et al. on their insightful study. The authors conducted a thorough secondary analysis within the PALLiON trial, focusing on the prognostic value of baseline factors like performance status (PS), albumin (alb), C-reactive protein (CRP), and body mass index (BMI) for overall survival in patients with advanced gastrointestinal cancers. The study’s strengths lie in its robust methodology, based on a post hoc analysis of a prospective trial. This work will certainly be of great interest to the readers. However, I have some suggestions to improve the manuscript.  

 

Clarify if this current study was initially planned in the trial protocol or not (is this study a post hoc analysis?).

 

Title: The current title, "Differential prognostic value of performance status, albumin, and CRP in last-line chemotherapy for pancreatic vs. other gastrointestinal cancers – why simple tools matter," is informative but a bit lengthy. A more concise version could be considered.

 

In the text, use italics and a dot for "vs."

 

Clarification is needed on whether all patients were followed until death or if there were any dropouts. Understanding the completeness of the follow-up is crucial for interpreting the study's findings accurately.

 

The study does not specify the types of pancreatic cancers included. It is important to distinguish whether neuroendocrine tumors or other less frequent cancer types besides adenocarcinoma were part of the analysis, as these can have significantly different survival rates. Including this information would enhance the study’s applicability and generalizability.

 

Serum albumin is closely related to the nutrition status. Do the authors have any additional clinical information related to nutritional status, such as weight loss, sarcopenia, body mass index, etc?

 

The determination of cutpoints for continuous variables like albumin and CRP is not fully explained. It would be beneficial for the authors to provide more details on this process. I suggest reading the following article DOI: 10.1590/0100-6991e-20223346-em to provide you with better insights for choosing the cutpoint.

 

The discussion section is short. The discussion could be expanded to better address the clinical implications of these findings.

 

Overall, this study offers significant contributions to the field, and with these minor revisions, it will be even more impactful.

Author Response

Comment 1: "Clarify if this current study was initially planned in the trial protocol or not (is this study a post hoc analysis?)."

Response 1: Thank you for pointing out this, which was not stated specifically in the manuscript. We have added this information now under "Study design", see line 89 of the revised manuscript: "The present study was a post hoc analysis comparing PAN vs. other GI cancers."

Comment 2: "The current title, "Differential prognostic value of performance status, albumin, and CRP in last-line chemotherapy for pancreatic vs. other gastrointestinal cancers – why simple tools matter," is informative but a bit lengthy. A more concise version could be considered."

Response 2: We agree and have now omitted two unnecessary words. The new title is: "Prognostic value of performance status, albumin, and CRP in last-line chemotherapy for pancreatic vs. other gastrointestinal cancers – simple tools matter".

Comment 3: "In the text, use italics and a dot for 'vs.'"

Response 3: Done, changed throughout the entire manuscript.

Comment 4: "Clarification is needed on whether all patients were followed until death or if there were any dropouts. Understanding the completeness of the follow-up is crucial for interpreting the study's findings accurately."

Response 4: The Study design secton of the manuscript has now been supplemented with this information. Thus, from line 89, we have added: "The present study was a post hoc analysis comparing PAN vs. other GI cancers, with complete follow up of the patients until their death. More information on the original sample can be found in Hjermstad et al[9]."

Comment 5: "The study does not specify the types of pancreatic cancers included. It is important to distinguish whether neuroendocrine tumors or other less frequent cancer types besides adenocarcinoma were part of the analysis, as these can have significantly different survival rates. Including this information would enhance the study’s applicability and generalizability."

Response 5: "We agree that this would be valuable information, but unfortunately, we do not have this information except for patients included at one of the institutions (Oslo University Hospital), where all patients were adenocarcinomas. We have not added this information in the manuscript.

Comment 6: "Serum albumin is closely related to the nutrition status. Do the authors have any additional clinical information related to nutritional status, such as weight loss, sarcopenia, body mass index, etc?"

Response 6: We added information on BMI at baseline, and included this in the analyses (see supplementary Tables 2 and 3. Longitudinal measures of weight loss, nutritional status, and sarcopenia were not included in the present analysis, but we are including such analyses in an ongoing study. 

Comment 7: "The determination of cutpoints for continuous variables like albumin and CRP is not fully explained. It would be beneficial for the authors to provide more details on this process. I suggest reading the following article DOI: 10.1590/0100-6991e-20223346-em to provide you with better insights for choosing the cutpoint."

Response 7: Thank you for pointing this out. We agree. The adequate threshold to define a true positive sample relies on methodology and disease prevalence, which may vary from one dataset to another),  and optimisation of cut-points to maximise the difference between “positive” and “negative” samples increases the risk of reporting significant results by chance. We therefore defined a priori the cutpoints for albumin and CRP, aligned with the established cutpoints for mGPS (modified Glasgow Prognostic Score), to avoid potential bias (see Altman et al, "Dangers of using 'optimal' cutpoints in the evaluation of prognostic factors' [note, we got an error message when trying to retrieve the article with the provided DOI refered to by the reviewer]). In the manuscript, we have added the following sentence to the section on "Prognostic factors", line 99-102: "We defined a priori the cutpoints for albumin and CRP, aligned with the established cutpoints for mGPS, to avoid potential bias by maximizing the difference between “positive” and “negative” samples and thereby increasing the risk of reporting significant results by chance."

Comment 8: "The discussion section is short. The discussion could be expanded to better address the clinical implications of these findings."

Response 8: Thank you for this comment. We have now expanded the discussion to better address the clinical implications of these findings (with track changes in word).

Reviewer 2 Report

Comments and Suggestions for Authors

This a clear and well written paper on an hot topic. There are few data in the literature about the last weeks of GI and PANC cancer and with this paper some light starts.

No other comments, all the sections are adequate and complete informations are given

Author Response

Comments 1: "This a clear and well written paper on an hot topic. There are few data in the literature about the last weeks of GI and PANC cancer and with this paper some light starts. No other comments, all the sections are adequate and complete informations are given."

Response 1: Thank you for your positive feedback and for recognizing the relevance of this work; I am grateful for your evaluation.