Pain Relief after Stereotactic Radiotherapy of Pancreatic Adenocarcinoma: An Updated Systematic Review

Round 1

Reviewer 1 Report

The authors have set out to update a previous analysis on the role of SBRT in pancreatic cancer pain control that was published in 2018.  The new analysis incorporated publications from Jan 2018-Dec 2020, enriching their analysis by 6 additional reports. The topic of pain control in pancreatic cancer is very important and with no clinical trial data to discuss the efficacy of SBRT in pancreatic cancer pain mgmt, reviews like these are important to give pancreatic cancer treaters more information about what to do with their patients.

Several questions did arise while I was reading this:

1. I think it is important to differentiate the pancreatic cancer (ie are you speaking about adenocarcinoma or are you including pNETs). Generally pNETs are not treated in this way (as pain is not necessarily an issue). Most of what is being discussed is about adeno or adenosquamous. As pNETs were excluded, but this is buried in the methods, I think this should be highlighted earlier
2. In the search results, figure 1 and what is written do not seem to flow. 7 new papers were found, and added to the 14 reports (which is 21). You mention 20 papers were included (as 1 study reported the same data as a paper already included in the study) which gives you 20 (21-1). But then in figure 1 it says only 19 studies were reviewed and 19 are listed in Table 1. So the numbers do not match.
3. figure 2 states rate of reduction OR suspension, but on line 124, it says analgesics reduction PLUS suspension.  Is this meant to be different?
4. 615 patients are analyzed in 19(?) studies of which 296 had pain scores before SBRT. However figure 2 only refers to 95 patients, figure 3 refers to 64 patients, and figure 4 has 146 patients over 9 studies. Less than half of the patients were analyzed. As well, only 14 unique studies were analyzed over the 3 figures. I am thus unclear how 19 or 20 studies were analyzed when not all of them were used in some form
5. If a paper does not report a side effect, it does not mean it was not there. It may not have been recorded or there could be patient recall bias, or there might not have been any side effects to report. I think all has to be acknowledged.
6. It is mentioned 2 studies reported negative effect of SBRT. They accounted for 45 patients. I feel this the authors dismissed this as not amounting to much of nothing. But their analysis is based on 95 ,65 and 146 patients. 45 patients is equivalent to 47%, 69% and 31% of their sample size and thus it is not insignificant if they are basing their conclusions on such small samples. I feel this could be acknowledged a bit more.
7.  I think the sample sizes need to be acknowledged a bit more as patient selection bias could play a big role in what they are seeing (which patients were selected for SBRT versus other pain modalities) and how that could affect the results
8. Combined modality patients (chemo+SBRT) was allowed (as the Polistina study is a neoadjuvant chemo+rads study). there is no mention how a chemo/rads approach would be the same or differ from just straight SBRT. Could this affect your results?
9. Some studies are for locally advance PDAC, some are for metastatic and some are neoadjuvant. Is all of the SBRT focused on the pancreatic lesion or anywhere else? How many were pancreatic head/body masses vs tails? This could affect the results and the conclusions. I think some more thought needs to be given here.

A couple of personal thoughts:

1. Usually pancreatic cancer is referred to PDAC (for the adenocarcinoma version). I have seen PCa refer to prostate cancer but not necessarily to pancreatic cancer. That is more of a preference and should not necessarily detract from the work itself.
2. I am not use to seeing the methods come after the discussion. It definitely ruined the flow as I had to skip the results to go looking for the methods. Again, that is more of a preference and should not necessarily detract from the work itself.
3. There is a lot of dosing analysis done in the discussion. I would prefer this to be in the results

Author Response

Please see attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present an update on their previously published systematic review analyzing studies that report pain response after stereotactic body radiotherapy for unresectable pancreatic cancer.

Please see my comments below

• It was noted that no formal prediction tool was used to trigger the update.
• Would suggest using Risk of Bias in Non-Randomized Studies - of Interventions (ROBINS-I) to assess risk of bias in this update.
• The methods section is after the “results and Discussion” sections. This needs to be switched as appears to be a typographic error
• Some studies demonstrated worsening of pain; it is interesting that was not captured in the forest plot
• Adhering to current standards of systematic review reporting, a column with quality of evidence (GRADE) needs to be added to the analysed studies table
• More is warranted in the discussion section. Rather than repeating the reason for the analysis, the authors can do more in expanding on each result of this study such as what was the effectiveness (or lack of it) and the toxicity separately. The discussion focusses on the side-effects rather than the effectiveness findings of this update.

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

The authors provide an updated meta-analysis of pain relief following pancreatic SBRT. The text of the article is well written, the approach is appropriate, and the conclusions are supported by the data/literature. Enthusiasm is damped by the frequent grammatical/layout inconsistencies which raise concern about the rigor as well as the limited novelty given the short time interval since prior the groups own work published in this area.

Regarding novelty, could the authors, even if only at a study level, provide an clinically actionable graph showing correlation (or lack thereof) between pain response and SBRT dose to guide clinically decision making regarding dose when pain is the primary indication for treatment? I think the inclusion of such an analysis could increase impact/novelty/utility. If the authors do not wish to carry out this analysis, I would strongly recommend alternative analyses to improve novelty.

I think the failed Alliance trial would also be important to mention and/or discuss.

 

Specific Comments.

Figure 1 Diagram has crooked arrows, please correct. Further, in the introduction, the authors say 180 new pprs identified in this area, yet the figure says 42 new papers. This is internally inconsistent.

Table 1, please capitalize headers

Figure 2 and 3, columns on right do not line up with studies listed on left.

Figure 5 is missing substantial text in the figure and I do not feel this needs to be included in the article.

 

 

While I understand that pain in pancreatic patients can have multiple etiologies, I’m not sure I understand whether they stress in the same sentence in the introduction and discussion that stenting of obstructive biliary jaundice resolves pain similar to neurolysis/SBRT. I don’t think the other mentioned local approaches are treating pain from the same etiology, and I think clarification of this point in the text would be helpful.

 

 

Author Response

Please see attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have appropriately reviewed my comments and critiques and have made sufficient changes to the manuscript that I am fine with it. Thank you

Reviewer 2 Report

The authors present an update on their previously published systematic review analyzing studies that report pain response after stereotactic body radiotherapy for unresectable pancreatic cancer. The information presented will be interesting to the GI radiation oncologists. 

Reviewer 3 Report

The revised manuscript is substantially improved. All of my concerns have been addressed. I commend the authors on their responsiveness. This is now an outstanding clinically relevant manuscript with significant novelty worthy of publication in its current format.