Can STEreotactic Body Radiation Therapy (SBRT) Improve the Prognosis of Unresectable Locally Advanced Pancreatic Cancer? Long-Term Clinical Outcomes, Toxicity and Prognostic Factors on 142 Patients (STEP Study)

Round 1

Reviewer 1 Report

In the result section the reporting of subgroups should be related to the same population. The denominator for the subgroups should always be the same. e.g. how many patients suffered from pain, how many experienced an improvement of their symptoms. How many patients received neoadjuvant chemotherapy and so on.

In the figures 3-4 censored patients should be marked. In Fig. 4 confidence intervals should be given.

The authors should comment in the discussion why their survival date have been so much more favourable that the reported data in the literature.

The retrospective design of the study should be pointed out in the Conclusions.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors performed a retrospective study on the value of SBRT in patients with LAPC. The cohort presented is relatively large as compared to what has been published previously and this study might therefore hold interesting information for readers, however, overall the study is not presented very well and I have some major concerns that need to be addressed to improve the paper:

Abstract:

- mentions inoperable disease, this should be changed to unresectable or locally advanced (as metastatic disease might also be considered inoperable)

- the methods section of the abstract says nothing about the statistical methods used. The conclusion states that OS and LC is 'better' when combined with chemotherapy, but from the methods it is unclear what comparison has been made. Also, it is unclear what is considered 'effective'.

- I don't think median follow-up is calculated correctly, as this is shorter than OS. Median fu should be calculated as follow-up in alive patients only.

Introduction:

- PDAC in general is typically considered a systemic disease, not LAPC per se. LAPC initially seems to be more of a local problem than a systemic one, as the disease is primarily manifested only locally, so I would change this statement of 'LAPC is typically considered a systemic disease' to 'PDAC is typically considered a systemic disease'.

- Considering that the authors aim to assess efficacy of SBRT in LAPC patients, I would recommend to present some number of survival and local/distant control in LAPC according to current literature, so that the results of this study can be interpreted in the right context and the rationale of the study becomes more clear. It is now difficult to interpret the results properly. For instance, it is stated that these findings are promising, but for the reader this isn't clear because the context is missing and there is no clear background/numbers about the problem.

Methods:

- Was PA not available for all patients? How many patients had PA proven PDAC and how many had radiologically proven PDAC?

- Prognostic features is not a (secondary) outcomes. Local control is an outcome, and you perform an analysis to identify prognostic factors associated with this outcome. This should be rewritten.

- The authors describe a multivariate analysis in the methods section but don't present the outcomes of the analysis in the results? If they performed a multivariable analysis (multivariable is the correct term rather than multivariate), this should be more clearly described: which prognostic factors where included? How was the analysis performed, what analysis was used? There is multicollinearity between for instance CTV-PTV-tumor size, so this should be taken into account. Also, as 142 patients where included, only 14 variables can be included (with a categorical variable with 2 categories accounting for 2 variables, with 3 categories for 3 variables etc). Also pre- and post SBRT CA 19-9 is related, so you might better choose a change (%) in CA 19-9?

- Overall, I think the data can be analyzed and presented much better, as this cohort holds very relevant and interesting information as the right analysis are performed with the right methods, and presentation and discussion of results is improved. 

Results:

- The results section should also include median follow-up (calculated as follow-up in alive patients)

- for local control, it is unclear whether patients were censored from the date of death? This should be described more clearly in the methods section

Discussion:

The discussion and conclusion section mentions that the results are promising and local control is effective, but this is very arbitrary. It is very hard to interpret these numbers, because little context is given. How are these numbers compared to LAPC without local treatment? This should be elaborated on more clearly in the manuscript. What are these numbers measures against/compared to? The authors provide two nice tables with retrospective/prospective studies on this subject, but the differences between their outcomes and these studies should be better discussed and explained to be able to interpret their findings better.

Overall, the quality of English is okay, but can be improved throughout the manuscript to improve readability. 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Interesting work due to the large and homogeneous single-centre series: being a retrospective work, however, it does not add certain evidence to the use of the SBRT technique in the treatment of LAPC, even if it identifies its good tolerability and efficacy in association with standard systemic therapy.

The manuscript is well structured and full of good iconographic synthesis, correctly inserted in the text.

The bibliography is up-to-date and relevant.

In my opinion, however, it is necessary to define the rationale and objectives of the study more precisely, both in the abstract and in the introduction of the manuscript, where these are not clearly identifiable.

The results are exhaustive and detailed, the conclusion is consistent with the data analysis.

Therefore the article can be considered for publication after a minor revision.

Author Response

Please see the attachment

Author Response File: Author Response.docx