The Tomosynthesis Broken Halo Sign: Diagnostic Utility for the Classification of Newly Diagnosed Breast Tumors

Round 1

Reviewer 1 Report

Dear Authors,

I congratulate you for the interesting topic provided. Nevertheless, I have some concerns about your paper.
In particular, my main concern is about the correct definition of "the broken halo sign" and, consequently, the sample of your study. I think that the analysis of the impact of the identification of thin interruptions of the radiolucent ring surrounding a lesion, especially in dense breasts, can be of relevant interest because it can be useful to predict malignity. At the same time, It is already well known that, for example, a spiculate mass is usually malignant. Consequently, is it correct to state that this kind of lesion has really a "broken halo"? We can say the same for architectural distortions. Indeed, one of the cited work [23] mentioned a "vertical trabecula sign" that I think it is quite different from the broken halo sign. Furthermore, as you correctly report in the limitations paragraph, the "halo" is influenced by DBT and the device used, while AD and spiculae are usually visible, especially with DBT. Consequently, I suggest to better clarify this aspect and to focus on the small halo changes that can really be significant for an early identification of breast cancer. If not possible, I think that this is a limitation of the study that needs to be discussed. 
Another concern is about the BI-RADS edition used to define the breast density. As you reported on lines 311-314, it seems that you used the BI-RADS 4th Edition instead of the latest one (5th). This aspect also can have an impact on the the conclusion of your work since it is well known that a lower proportion of dense breasts can be classified when using 4th edition instead of 5th. I think that this aspect should be discussed and not only mentioned as limitation (please, check 10.2214/AJR.16.17525 about 4th and 5th differences and 10.3390/jpm13040609 for the impact of artificial intelligence that can support the correct classification).

Finally, why did you perform DBT only on MLO orientation? Have you evaluated the halo sign only on these projections? I think that this aspect should me clarify and discussed and maybe considered another limitation to the study.

Please, clarify this sentence: lines 31-32 "per 100.000" what?

Please, clarify this sentence: lines 46-47. Architectural Distortion is one of the presentation of breast cancer but it is not always associated to it. Furthermore, a spiculate opacity and an AD are two different kind of lesion. Please, I think that is mandatory to clarify this sentence.

Please, clarify the acronyms.

Please, clarify this sentence: line 105 "these lesions were reported twice...", what does it means?

Please, clarify this sentence: lines 130-131 "each...to be definitively proven", what does it means?

Please, clarify this sentence: line 141 "described above": I'm sorry but I can't find what you refer to?

Please, clarify this sentence: line 157 "on average rated as 2.4..": what does it mean? 

Please, clarify this sentence: line 161 "intermediate differentiation", I don't think it is the correct term. Do you mean "category"?

Please, revise table 5: ranges are necessary? if yes, please, use a different lines to make more understandable the table.

 

The paper is well written even if there are some typos and incorrect verb tenses (e.g. lines: 194, 189, 153, 120). Consequently, I suggest a revision of the English language.

Author Response

Dear reviewer, thank you for the kind words.

Please see the attachment for our Reply to the Review Report. 

Kind regards, 

Michael Swoboda

Author Response File: Author Response.pdf

Reviewer 2 Report

In general, the Paper is well-written and structured/designed. However, I have a few comments to improve the manuscript. The comments are specified as follows:

The abstract is also well written, but it seems to be missing overall conclusion sentences at the end of the abstract.

 

Good and well-written introduction with good citations from previously conducted studies. Thus, I have very optional remarks.
-The halo sign and the broken halo sign are not commonly valid concepts, so it would be very good if the halo-sign (peritumoral halo and broken halo) concepts were described theoretically.

 

The following  abbreviations in the materials and methods chapter need to be defined:

Line 66: RIS

Line 73: PGMI

Line 75: MLO 

 

In line 82 the vendor name is Siemens healthneers (Siemens Medical any more).

 

In the criteria selection of the patients, I think some important parameters need to be mentioned, including a range of breast tissue thicknesses and breast densities. These parameters are very crucial in considering the image quality and tumour-identifying

 

In lines 83-84 the author mentions that the image acquisition parameters are based on international guidelines, this needs to be either referenced or a list of exposure parameters. 

 

The number of patients/cases used in this study should be mentioned in the materials and methods.

 

In the Figure 1 description, the acronyms (SNT) in lines 111 and 112 should be defined.

Author Response

Dear reviewer, thank you for the kind words.

Please see the attachment for our Reply to the Review Report. 

Kind regards, 

Michael Swoboda

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors,

You have addressed the most of my concerns.

Consequently, the paper can be accepted in present form.

Sincerely,