Dose–Volume Constraints fOr oRganS At risk In Radiotherapy (CORSAIR): An “All-in-One” Multicenter–Multidisciplinary Practical Summary

Round 1

Reviewer 1 Report

With great ambition the authors have undertaken the work of collecting a comprehensive set of dose constraint for Radiotherapy treatment planning, simplified into a single main table. A great number of publications have been consulted and such a table would be of great interest to radiotherapy professionals. However, in order to be comprehensive, critical compromises have been made which prevent the result from useful application. These compromises include:

1) A clear, but in practice simplified, categorisation of evidence levels. Here, international guidelines are considered the highest level of evidence, even where such guidelines might be outdated. For some organs at risk (OAR), updated validated evidence exists which is more applicable to modern treatment techniques, although not presented in the reviews or guidelines considered by the authors. Also, guidelines are sometimes based on evidence of level C.

2) In many Radiotherapy departments, tables of dose constraints are pragmatically applied as planning aims. While the treatment planners, reviewers and therapists are not generally familiar with the evidence behind those constraints, the selection of suitable dose constraints for the local context ideally ensures some local expertise to guide proper application. While the rational of sharing such efforts between centres makes sense, there is then a greater risk of inappropriate or suboptimal application of the dose constraints. Where such an endeavour is nonetheless undertaken, the reporting of important details for the application of constraints should be attempted. The format of the tables in this manuscript does not fulfil that need.

3) It is problematic to reference the consulted guidelines, rather than the original publications which the guidelines are based on. Unless each reference is verified, the current authors then rely on the correct citing of previous authors, which is a high-risk process. E.g. Noël et al. (2016) makes several references to dose constraints in Darby et al. (2013), relating to lung and breast as OAR, which I cannot find. Also, the context for the dose constraints is even less visible when hidden behind a chain of references.

In addition to the compromises in the selection of recommended dose constrains mentioned above, the manuscript suffers from a lack of specificity in the aim. While the primary aim appears to be regarding dose constraints, recommendations for OAR definitions have also been included in the table. Since these definitions do not genarally correspond to those used in the generation of dose constraints, these are really unrelated sets of recommendations and should not be presented together. It would be very easy to make the mistake of assuming that the dose constraints apply to OARs as defined in the table, which is not the case (e.g. for Bowel V45Gy, but presumably more generally given the different references). The two aims should be treated in separate manuscripts or at least in separate tables.

Further, please find below concerns and requests of a more specific nature:

A) Search criteria: What dates were covered in the search? I noticed e.g. that the old GEC-ESTRO guidelines for prostate brachytherapy were included rather than the latest (Henry et al. 2022). Which languages were considered? Only considering articles unfortunately excludes the book "Modelling Radiotherapy Side effects: practical applications for planning optimisation" edited by Rancati and Fiorino (2019), which provides well-informed conclusions regarding dose constraints. The HyTEQ report doesn't seem to have been consulted? Why were NCCN guidelines considered for four OAR specifically, and not for all? What type of evidence are these guidelines based on?

B) What were the criteria considered for application in low-medium resource settings? This needs to be specified. (Such critera could defend using older guidelines based on non-IMRT/IMPT treatments, but this principle doesn't appear to have prevailed.)

C) Definition of ‘conventional fractionation’ is missing. For some treatments this might be 25 fractions, for others 39. This can easily make a 10% difference in BED.

D) Inconsistent constraints for bladder, presumably for prostate and anal cancer treatments respectively? DMAX < 65 Gy quoted from Quantec, but I can't find this number in the original article.

E) The breast dose constraint is very low. It is unclear what indication and endpoint is considered. Also, as mentioned above, I haven't been able to verify this number in the original publication.

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors,

it was a pleasure for me to revise your manuscript. Even if it doesn’t add any new original findings, it has the merit to collect in a single document crucial information for planning and dosimetric purposes during the radiation oncologist’s daily clinical practice. I think that the provided information is quite comprehensive and needs to be revised for a plenty of minor issues before publication:

1)      “...such Hodgkin's lymphomas[11] , breast carcinomas [12–14] , and pediatric cancers [15–20].”, add “as” after “such”.

2)      Sometimes the tone used in the manuscript is a little too colloquial and doesn’t sound scientific enough: for example, “...Moreover, the bibliographic list of several publications was screened in order to identify other relevant sources.”, how many is “several” for? In a scientific paper, you should be as accurate as possible to ensure the methods are reproducible by the readers, if they want. You may indicate a flowchart for search strategy.

3)       “The results recorded during this search were independently checked by at least three authors”, see again the point 2). “at least”? how many authors? Specify them.

4)      “In addition, in order to provide users with a critical assessment of the DVCs, we categorized the levels of evidence as follows: [A] international guidelines; [B] literature reviews on clinical or planning studies; [C] data from the results of clinical or planning studies; [D] expert opinions or DVCs used in prospective trials. Moreover, when different sources presented different values of the same DVC, we included in the tables only the one with the highest level of evidence. Therefore, we included recommendations with “B-D” level of evidence only in case of lack of level "A" DVCs.”, I’m not sure about the correctness of such a nomenclature in this context: the “level of evidence” mainly refers to a context in which the quality of the retrieved findings influences the strength of the consequential recommendation and is fundamental to select the best therapeutic decisions. Your classification introduces a scale system (A better than the others, the level D being the worst), which could not be really representative of the so-called “level of evidence”: for example, a DVC deriving from a clinical study (level C) might not be worse than one deriving from international guidelines as these could be based even on a summary of “level D” findings. For these reasons, I believe that a more neutral nomenclature, such as “source of recommendation”, would be more appropriate for the purposes of this paper, leaving to the readers the possibility to check the quality/strength of findings by themselves. I suggest you revise the manuscript accordingly.

5)      “dose received by β% of the organ volume, D = dose received by γ cm3 of the organ volume”, I suggest you delete the greek letters, leaving only the “%” symbol and adding “the cubic centimeters” before (cm³) between brackets.

6)      “Over 650 papers/documents/websites were examined.”, again, how many?

7)      In bladder DVCs for conventional fractionation, you indicate two conflicting values: Dmax 65 Gy and V65 Gy < 50%. How is it possible to agree with a V80Gy < 15%, for example, equal to 5-10%, while not exceeding a 65 Gy Dmax? Please, clarify this issue. Moreover, the DVCs for hypofractionation refer to the bladder wall or for the bladder including its content contour. Please, clarify also this issue.

8)      “BrachialPlexs, sum of two volumes, may also be considered”, add “the” after “sum of”.

9)      In brain DVCs for single shot treatments, “Whole brain - GTV:”, replace “-“ with “less” .

10)   In brainstem DVCs for conventional fractionation, please, clarify if the 59 Gy limit refers to the peripheral edge of the brainstem while 54 Gy is for the inner part.

11)   In the esophagus constraint “V50Gy <40 ;” add “%” after 40.

12)   In the heart constraint “Contour the heart along with the pericardial sac. Cranial edge (or base): at the bifurcation the pulmonary trunk and right pulmonary artery. Caudal edge: apex of the heart. Exclude major vessels from the contour.” Add “of” after “the bifurcation”.

13)   The hippocampus constraints have not a clear sense and could generate misinterpretation. Please, specify if the first listed (see Dmax 6 Gy), which are stricter than the second ones (see “hippocampal avoidance volume:”), refer to particular treatment sites (i.e. head and neck vs brain tumors). Moreover, “DMAX <25.2 Gy; DMAX <12 Gy;”, what constraint I should consider in my daily clinical practice? The first or the second? Why these two are so different? Does their exceeding produce different risk rates of a particular adverse event or a different type of adverse event? Clarify also this issue.

14)   “Kidneys may be considered as a sum of the two volumes can be created to calculate total kidney dose.” correct as follows: “...of the two volumes, which can be created...”.

15)   “Located upon the hypophysial fossa of the sphenoid bone. Surrounded by a small bony cavity (sella turcica) covered by a Dural fold (diaphragm sellae). Connected to the hypothalamus by its pituitary stalk. [28,46] [α] Fossa_Pituitary, can be contoured as an alternative structure. [28] [α]”, “Dural” does not need the first capital letter; delete “,” after “Fossa_Pituitary”. In this line of the table or in the other one, please, cite PMID: 35349010 .

16)   In the LAD line, you should cite PMID: 33937532 and PMID: 33788746 for Dmax_LAD ≤ 20 Gy as an optimal constraint for conventional fractionation with a level C in terms of source of recommendation.

17)   In the supplementary table for emerging OARs, you can add a line for Bichat fat pad by citing PMID: 35833962 , which didn’t find any dose-volume constraints as no OAR modification occurred after H&N irradiation.

18)   In the supplementary table for emerging OARs, you should add a line for long bones, especially for femurs, since some papers (cite PMID: 15687153 and PMID: 20108335 as level C) proved that a dose lower than 50 Gy is less risky than a dose of 60 Gy for radio-induced fracture. For the description of the anatomical contour, you can write “The diaphysis cross-sectioned by the beam entrances. Reduce the femur dose using VMAT [*].” and in [*] cite PMID: 34946324 .

19)   In the lung line, replace “V05Gy ≤ 50%;” with “V5Gy ≤ 50%;”.

20)   In the discussion section, “We subsequently decided to share this work and to involve a multidisciplinary and international team of authors to write this publication.”, replace “publication” with “paper”.

21)   In the discussion section, immediately after “Therefore, great caution is required when using "new" RT techniques with "old" DVCs.”, please, add the following paragraph: “Similarly, Ferini et al. effectively summarized a large series of ever-changing dosimetric issues about the rectum, including contradictory perspectives on how to consider such an OAR (parallel vs serial) and, consequently, conflicting DVCs [PMID: 33910815]. In this scenario, there is also the need to understand whether some medical interventions can improve the rectal tolerance to high radiation exposure [PMID: 33813420 and https://doi.org/10.3390/gastroent12040043].”

22)   At the end of the discussion section, please, add the following paragraph: “Additionally, no DVC indication is made for RT schedules combining different radiation dose sizes, as in the cases of a stereotactic boost to a limited tumor site after a more extended normofractionated irradiation [PMID: 27686231 and PMID: 35034325] or of a large spatially fractionated dose administration before a homogeneously delivered hypofractionated palliative course [DOI:10.3390/cancers14163909].” as the last sentence.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Thank you for well-juged and carefully considered answers to my comments and corresponding changes to the manuscript! I look forward to seeing it in print. I do, however, have a few further minor change requests/recommendations in the attached document. I have inserted these in the history of comments and answers, in a different colour of text.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf