Can Molecular Biomarkers Help Reduce the Overtreatment of DCIS?

Round 1

Reviewer 1 Report (Previous Reviewer 2)

To the Authors,

Thank you for your revision which very briefly (not comprehensively) addressed the concerns raised.

To the Authors,

I have been tasked with addressing the recycling report on your manuscript.

I believe there is room for rephrasing and revising many parts of the manuscript to reduce 'recycling' as highlighted by the journal's search.  For example, in line 38, one could say, 'Breast conserving surgery is the modality of treatment for the majority of women with DCIS.' & in lines 52-53, changing the order of the studies & references may result in the non-identification of language recycling.

Author Response

I believe there is room for rephrasing and revising many parts of the manuscript to reduce 'recycling' as highlighted by the journal's search.  For example, in line 38, one could say, 'Breast conserving surgery is the modality of treatment for the majority of women with DCIS.' & in lines 52-53, changing the order of the studies & references may result in the non-identification of language recycling.

Thank you for your suggestions for reducing similar language. Through the iterations of this paper and revisions, many partial phrases have already been changed in similar fashion to what the reviewer is suggesting in order to reduce the similarity index, despite having written original sentences. In fact, in the first review, several sections of the paper were highlighted for revision due to similarity of partial phrases and the example given for line 38 was not flagged initially, only in a subsequent revision request. Nevertheless, I will go through the paper again and revise many of the partial sentences with the goal of reducing the similarity index even though this is original writing, for the purposes of this revision. Similarity of partial sentences is expected when there is common terminology regarding a particular subject for which there is much writing in the literature, especially in reviewing literature. The order of references in lines 52-53 was because that was the order of when those studies came out and were published (1998, 2000, 2003, and 2014); these 4 studies are frequently cited together in that order since they collectively form a body of high-quality evidence for RT after BCS for DCIS. Other flagged phrases in the similarity report are clearly common terms, yet are flagged, for example:

• “breast radiotherapy (RT)”
• “The 10-year”
• “treated by BCS alone”
• “increased risk of breast cancer mortality”
• “single-arm study”

This list could go on. If all these partial phrases were not counted in the similarity index, then I believe the numerical value would be much lower and not of concern. These partial phrases are not of any concern in our opinion and seem to be a key issue in the submission of this paper.

I also would like to highlight that revising many sentences in this paper may not necessarily resolve the similarity of partial phrases. Looking though the iterations of revisions, I see that some revised sentences have now been highlighted as partially similar. Again, I see this as expected since there is much commonly used phrases and terminology when discussion DCIS, breast surgery, radiation, and outcomes.

Please see the tracked changes through the paper where I have attempted to rephrase like you have suggested.

Reviewer 2 Report (Previous Reviewer 3)

The authors have properly addressed all my questions. I consider that the manuscript is now suitable for publication.

Author Response

Thank you so much.

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

Evaluation scores and decision values have been considered the least valuable. 

Author Response

Thank you to the reviewers for your thoughtful comments and suggestions.

The manuscript has been edited with tracked changes in regards to the selected “Extensive editing of English language and style required” and wording related to other work.

Author Response File: Author Response.docx

Reviewer 2 Report

This is a review on the possibilities of applying molecular biomarkers to the treatment of DCIS, preferably with respect to de-escalation.

The points are well argued and presented.

On the basis of their presentation, do the authors have an opinion on whether surveillance for presumably low-risk DCIS is appropriate?  In the Japanese study that they cited, 12-25% of patients with DCIS on core biopsy were found to have invasive disease.  Do the authors think that this is significant and what role would molecular biomarker play in active surveillance?  Is there currently data on natural history of presumed low-risk disease?  It has been reported that DCIS can progress to the invasive form in a space of half a month.1  How would the authors view delayed diagnosis in such circumstances?

Thank you.

Reference 

1.  Tan KHX et al. Quantifying the natural history of breast cancer.  Br J Cancer 2013;109:2035-2043

Author Response

On the basis of their presentation, do the authors have an opinion on whether surveillance for presumably low-risk DCIS is appropriate?  In the Japanese study that they cited, 12-25% of patients with DCIS on core biopsy were found to have invasive disease.  Do the authors think that this is significant and what role would molecular biomarker play in active surveillance?  Is there currently data on natural history of presumed low-risk disease?  It has been reported that DCIS can progress to the invasive form in a space of half a month.1  How would the authors view delayed diagnosis in such circumstances?

Thank you to the reviewers for your thoughtful comments and suggestions.

We agree regarding the point that a patient with an undetected invasive breast cancer would not be the appropriate patient to select for an active surveillance study, especially with features of non-indolent behaviour. How biomarkers would help with selecting appropriate patients remains to be seen and work needs to be done prior to that point. The impact of delayed diagnosis on molecularly defined low risk patients is unknown and could be studied with a thoughtfully designed clinical trial. Based on your comments, we have updated that paragraph of the manuscript, which now reads:

“This study highlights the need to appropriately select patients with a very low risk for active surveillance on appropriately designed clinical trials in order to not enroll patients with invasive disease or aggressive biology. How molecular biomarkers may contribute to optimizing patient selection for possible active surveillance trials remains to be seen, and further work is needed to correlate molecular biomarkers on core biopsy to full surgical specimens on excision.”

Reviewer 3 Report

The present paper provides an overview of current progress in development of the criteria to exclude radiotherapy after DCIS surgery, without increasing the risk of recurrence and/or development of invasive cancer in the future. The overview covers the studies that evaluate single clinicopathological factors ( tumor size, nuclear grade and resection margins) as the decision-making criteria, as well as the studies that deal with the effect of including other therapies (adjuvant endocrine therapy) after BCS. Finally, the overview presents studies that deal with combination of multiple biomarkers and clinicopathological factors to provide scoring scales that help decision making. The studies in progress are also mentioned.

The review is precise, clearly written and provides a good insight into the topics. The topics is interesting and important, since safely made decision on avoiding of unnecessary RT would bring undoubted benefits.

Some minor questions:

- The decision-making based on biomarkers is of general interest in cancer research. It would be interesting to mention the pioneering studies in the field – when it started to develop? Also, are there any biomolecular patterns that are already used in wider clinical practice for decision making in treatment of any CIS, or any type of malignant tumors?

- From the provided overview, it can be concluded that there are no simple molecular patterns that can guide decision making, and that would be simply and willingly used by clinical oncologists. On the other hand, more precise scoring scales, provided in the studies, also increase the complexity in future clinical implementation. Could you briefly discuss?

-How common is the mastectomy in the treatment of DCIS? Are there any studies dealing with DCIS induced mastectomy and the risk of future cancer events?

Author Response

Thank you to the reviewers for your thoughtful comments and suggestions.

The decision-making based on biomarkers is of general interest in cancer research. It would be interesting to mention the pioneering studies in the field – when it started to develop? Also, are there any biomolecular patterns that are already used in wider clinical practice for decision making in treatment of any CIS, or any type of malignant tumors?

One commonly used biomarker already in routine use is the oncotype dx recurrence score (RS). This RS is mentioned in the paper due to its potentially applicability to DCIS. However, its common use could be more explicitly stated with some history. We therefore added the following text to the manuscript to reflect this in section 4.

The 21-Gene Oncotype DX Breast Recurrence Score (RS) is a validated biomarker currently used in routine clinical practice, associated with the risk of developing metastases and chemotherapy benefit in patients with early, invasive breast cancer [59, 60]. This biomarker started being incorporated into society guidelines in 2007 with increasing adoption over the following years.

From the provided overview, it can be concluded that there are no simple molecular patterns that can guide decision making, and that would be simply and willingly used by clinical oncologists. On the other hand, more precise scoring scales, provided in the studies, also increase the complexity in future clinical implementation. Could you briefly discuss?

The reviewer raises a good point. The following was added to the first paragraph of section 4.

“… Finally, implementation of such biomarkers should be feasible to attain in the clinical workflow without onerous complexity, such as a report that is clinician and patient facing that can be used in clinical decision making.”

How common is the mastectomy in the treatment of DCIS? Are there any studies dealing with DCIS induced mastectomy and the risk of future cancer events?

The majority of woman with DCIS are treated with breast conserving surgery, however recurrences are often treated with mastectomy. Regarding this point we have the following text (which has been modified) in the introduction.

“More contemporary data also show the benefit of adjuvant RT in reducing the risk of local recurrence after BCS for DCIS. The reduction in local recurrence risk, importantly, also includes reduction in recurrence as an invasive breast cancer, referred to as an invasive local recurrence. Such recurrences are psychologically distressing and require further treatment that often includes further surgery, including mastectomy, and chemotherapy if indicated, due to the associated increased risk of breast cancer mortality. [9, 11]. Adjuvant RT has also been shown to lead to a lower risk of mastectomy. While some local recurrences can be managed with another breast conserving surgery,  randomized and population-based observational data report that most local recurrences are treated with mastectomy. In a population-based analysis based in Ontario, Canada, receipt of adjuvant RT was associated with a higher 10-year mastectomy free survival (87.3%) as compared to treatment with BCS alone (82.7%) (p<0.01), with an associated decreased hazard ratio for mastectomy of 0.71 (95% CI: 0.60, 0.84, p<0.0001)”

Reviewer 4 Report

Thank you for the opportunity to review this article. The theme is of extremely importance

The article respond to the question in the title.

The subject is comprehensive presented

In order to increase the value of the article may I suggest some modification

I think the molecular test must be presented as separate paragraphs ( for example line 194)

A personal interpretation of the result may be useful for the reader.

Regarding Predictive modeling the introduction of this section talks about Receive Operation Curve AUC but the  result talks about percent of LR and IBC.

If date are available please present the AUC.

The references are appropriate but may be a little updated to new articles.

Author Response

I think the molecular test must be presented as separate paragraphs ( for example line 194).

Thank you, the manuscript has been updated accordingly.

A personal interpretation of the result may be useful for the reader.

Interpretation of results in a clinically meaningful way is challenging due to the lack of external validation and assessment of biomarker performance in that setting. In the last paragraph of section 4, the following text has been modified.

“Given the current lack of prospective validation, it is challenging to interpret biomarker study results generalizing various scores and associated risks outside of the study in which it was conducted and internal validation reported. Treatment guidelines, therefore, do not currently recommend the routine use of molecular assays in the management of DCIS.”

Regarding Predictive modeling the introduction of this section talks about Receive Operation Curve AUC but the result talks about percent of LR and IBC. If date are available please present the AUC.

The c statistic (AUC) has been added to the manuscript where presenting the results of the model incorporating the DCIS score.

The references are appropriate but may be a little updated to new articles.