Overdetection of Breast Cancer

Round 1

Reviewer 1 Report

Dear Authors,

Overdetection is one of the most challenging phenomenon in breast cancer screening and prevention, especially for its ethical concerns. It refers to the diagnosis of breast cancers that remain occult through an individual’s life because they are slow growing or indolent. The consequence is the overdiagnosis and overtreatment of the disease, especially in older women, with no real benefits. One of the main problem is represented by in situ cancer. An early detection is the key point for a good screening program. Nevertheless, an excess of in situ cancer could represent a significant number of indolent cancers. I really appreciate the balanced evaluation of this topic. Your work is mainly based on the evaluation of the two Canadian studies but it could represent a stimulus towards a more comprehensive overview of overdiagnosis in general.

I really appreciate your work, the discussion section and the conclusions provided.

Given these considerations, the contribution of your paper, at this stage, is sufficiently developed such that it could be accepted for publication in Current Oncology with the following minor revisions:

1. I really appreciate your in-depth evaluation of the phenomenon, as the argument has an intrinsic complexity. Nevertheless, I think that with a better spell-check and a rearrangement of some phrases, the introduction could be better summarized due to some redundancies.
2. In “Materials and Methods”, please, merge the two paragraphs avoiding the subdivision that is redundant.
3. Figure 2. What the letter (c) refers to?

Author Response

Please note: reviewers' comments are indicated in bold type

I really appreciate the balanced evaluation of this topic. Your work is mainly based on the evaluation of the two Canadian studies but it could represent a stimulus towards a more comprehensive overview of overdiagnosis in general.

I really appreciate your work, the discussion section and the conclusions provided.

Thank You for your kind remarks.

1. I really appreciate your in-depth evaluation of the phenomenon, as the argument has an intrinsic complexity. Nevertheless, I think that with a better spell-check and a rearrangement of some phrases, the introduction could be better summarized due to some redundancies.
2. In “Materials and Methods”, please, merge the two paragraphs avoiding the subdivision that is redundant.Thanks. We have done this.
3. Figure 2. What the letter (c) refers to?Thanks for pointing this out. There should be a Part c in both Figures 1 and 2, containing a graph illustrating situations of no overdetection (Fig 1) and overdetection (Fig 2). We have added labels to both figures and have provided updated versions of the figures.

Reviewer 2 Report

1) I found the introduction to be overall very extensive. In the first paragraph it is stated the the term 'overdetection' will be used throughout the remainder of this article, however 'overdiagnosis' is used later on talking about pathological diagnosis (line 109). Is this intentional?

Paragraph starting at line 89 describing false positive results is not necessary.

In the section 1.2 the description of an ideal randomized trial might better suit the discussion rather than introduction. 

Line 150 and 159, 2 references listed as Error! Bookmark not identified

2. Line 501/discussion the indications of low risk are stated. This is a highly variable subject and the uptodate reference (36) also states low risk description is not clear-cut (especially clear resection margins). Would recommend rephrasing this.

3. The last paragraph of conclusion section is an abrupt end to the article with a reference, consider revising this paragraph (to end with the first sentence of the paragraph). 

Author Response

Please note: reviewers' comments are indicated in bold type

1) In the first paragraph it is stated the the term 'overdetection' will be used throughout the remainder of this article, however 'overdiagnosis' is used later on talking about pathological diagnosis (line 109). Is this intentional?

Use of the terms "overdetection" and "overdiagnosis" at various locations in the article is intentional.  We have tried to be specific in the use of these terms in our article. By overdetection, we refer to cancers that in an ideal world would not need to be detected because of their lack of aggressiveness. This is a limitation of the screening test. By overdiagnosis we refer to a limitation of the pathology tests in determining if a cancer is aggressive or not. Overdetection can lead to overdiagnosis, but it doesn't necessarily do so.  Unfortunately, many authors use the term "overdiagnosis" to refer to both phenomena. Therefore, when we refer to or quote the findings reported by others, we use the terminology that those authors used.

We have also added a few words at the end of the first paragraph of the introduction to clarify.

Paragraph starting at line 89 describing false positive results is not necessary.

We agree that many readers understand the difference between false positive findings on screening and overdetection, we have often found that others confuse these concepts or treat them as the same phenomenon. Therefore, we included this clarification that overdetected cancers are real cancers. I would prefer if the sentence remained, but am happy to leave this to the discretion of the editor.  If it is desired to remove the explanation, please just delete the sentence, "Overdetected cancers are real cancers and should not confused with so-called false positive results of screening, where further imaging or biopsy,  triggered by an equivocal screening examination, demonstrates that suspicious results on screening are not cancer."

In the section 1.2 the description of an ideal randomized trial might better suit the discussion rather than introduction. 

This description is not focussed so much on what is an ideal randomized trial as what would be the ideal design of a randomized trial whose data would be used to measure overdetection. We thought that it would be important for the reader to have an explanation of the principles and a sense of how the data would be expected to behave before we presented data and/or the results of computer simulations.

Line 150 and 159, 2 references listed as Error! Bookmark not identified

Unfortunately, the template does not allow me to display line numbers; nor can I see the "Error! Bookmark not identified" warning. If you supply me with the wording in the sentences containing these two references I can provide the reference. It is possible that the discrepancy occurs because we are using different versions of Word.

2. Line 501/discussion the indications of low risk are stated. This is a highly variable subject and the up to date reference (36) also states low risk description is not clear-cut (especially clear resection margins). Would recommend rephrasing this.

We agree and have rephrased

3. The last paragraph of conclusion section is an abrupt end to the article with a reference, consider revising this paragraph (to end with the first sentence of the paragraph). 

Thanks for the comment. We agree and have revised the paragraph.