Analysis of Spatial Distribution and Prognostic Value of Different Pan Cytokeratin Immunostaining Intensities in Breast Tumor Tissue Sections
, Tijana Vujasinović 2, Nemanja Rajković 3, Ksenija Kanjer 2, Nebojša T. Milošević 3 and Marko Radulovic 2,*Round 1
Reviewer 1 Report
This paper analyzes the spatial distribution of immunostainings of pan-cytokeratins in breast cancer. The authors took a group of patients (102 female Caucasian women) that underwent only local treatment by surgery or radiation and analyzed representive histopathology sections using a pan-cytokeratin antibody by immunohistochemistry. They previously reported that low intensity of pan-cytokeratin staining associate with higher metastatic risk. Within this paper they separate immunostaining intensities in risk categories and associated moderate intensities with metastasis outcomes while strong intensities produced no prognostic value. The data they reported also confirmed the intratumoural heterogeneity of the staining.
The paper is well written and the analysis convincing. However, I find the results obtained not completely novel and limited considering that the prognostic value of cytokeratin staining is known and even the same authors published recently that low intensity of pan-cytokeratin staining determines high risk of metastasis. Nevertheless, the study is useful for estimating the prognostic value.
Author Response
We found the Referee's comments very helpful. Below we address the provided comment in detail. The changes within the revised manuscript text which were introduced as a response to suggestions by the Reviewer are marked in red.
Referee 1 notes that the paper is well written and the analysis convincing. However, I find the results obtained not completely novel and limited considering that the prognostic value of cytokeratin staining is known and even the same authors published recently that low intensity of pan-cytokeratin staining determines high risk of metastasis. Nevertheless, the study is useful for estimating the prognostic value.
Author response: In response to this remark of Referee 1, we further clarify our opinion that pan-cytokeratin immunostaining requires even further detailed investigation to identify its most valuable prognostic clues and by that improve the prognostic performance that is currently achievable by analysis of the growth patterns of tumour malignant cells. This has now been mentioned in the manuscript text, lines 393-395.
Reviewer 2 Report
The manuscript entitled "Analysis of spatial distribution and prognostic value of different pan-cytokeratin immunostaining intensities in breast tumour tissue sections" describes reveals the intratumoural heterogeneity of the pan-cytokeratin immunostaining, together with the prognostic evaluation and spatial distribution of its discrete intensities. The current work should be accepted with minor revisions as follows:
Comments:
Please increase the quality of figures.
In materials section, H2O2 shoul be changed to H2O2
AUC is area under the curve, please correct in the manuscript.
Why an AUC=0.5 is good? please explain and introduce the respective reference.
Author Response
We found the comments of Referee 2 very helpful. Below we address the provided comments in detail. The changes within the revised manuscript text which were introduced as a response to suggestions by the Reviewer are marked in red.
Changes according to the comments of Reviewer 2
- Please increase the quality of figures.
Author response: All three figures have now been improved in quality.
- In materials section, H2O2 should be changed to H2O2
Author response: This error has now been corrected in line 327.
- AUC is area under the curve, please correct in the manuscript.
Author response: In agreement with this suggestion of Referee 2, we now introduce AUC abbreviation, at its first mention in line 54.
- Why an AUC=0.5 is good? please explain and introduce the respective reference.
Author response: We fully agree with Referee 2 that AUC=0.5 does not indicate good discrimination. After a careful check of the manuscript text, we could not find a statement that AUC=0.5 is good. We thus assume a misunderstanding.
The original manuscript provides four explanations of AUC:
- In line 143 of the revised manuscript: “AUC=0.5 represents chance discrimination, while perfect discrimination equals 0.0 or 1.0”.
- Legend to Table 1, line 159: ”AUC in the 0.0 - 0.5 range indicates an association with low metastasis risk and in the 0.5 - 1.0 range with high risk. AUC values between 0.3 - 0.4 and 0.6 - 0.7 are considered as fair discrimination performance, 0.2 - 0.3 and 0.7 - 0.8 as good, 0.1 - 0.2 and 0.8 - 0.9 as excellent and 0.0 - 0.1 and 0.9 - 1.0 as almost perfect”.
- Line 169: ”distances of significant AUC values from the null hypothesis value of 0.5”.
- Line 377: ”AUC values farther away from its random performance mid-point at 0.5 indicate an improved discrimination efficiency.”
We hope that you will find our manuscript now after its revision acceptable for publication in IJMS. We look forward to your response at your earliest convenience.
Yours sincerely,
Marko Radulovic
