Assessment of Tumor Infiltrating Lymphocytes in Predicting Stereotactic Ablative Radiotherapy (SABR) Response in Unresectable Breast Cancer

Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors present a pilot study investigating the computational analysis of tumor-infiltrating lymphocytes (TILs) to predict stereotactic ablative radiotherapy (SABR) response in patients with inoperable breast cancer (BC). Clinical and spatial features were collected and analyzed using machine learning (ML) classifiers, including K-nearest neighbor (KNN), support vector machine (SVM), and Gaussian Naïve Bayes (GNB), to predict SABR response. Models were evaluated using receiver operator characteristics (ROC) area under the curve (AUC) analysis. The highest-performing model using computationally derived graph features showed an AUC of 0.92, while the highest clinical model showed an AUC of 0.62 within unseen test sets. The authors conclude that spatial TIL models demonstrate strong potential for predicting SABR response in inoperable breast cancer. TILs indicate a higher independent predictive performance than clinical features alone.
The study is interesting, original, and well-conducted. I have nothing else to add. I believe it can be accepted for publication in your journal.
Author Response
Please see attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsIt is a very elegant and sophisticated study.
I have one question. What is the additional cost involved in this testing?
Does it require the level of sophistication in ML and AI that your group has? How reproducible and scalable will it be?
Because you will be able to do better patient selection, will it eventually reduce cost? You will not be treating patients who will not benefit from radiation.
Author Response
Please see attachement
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThis study describes the combination of images and TILs infiltration along with machine learning to predict responses to radiotherapy in patients with inoperable breast tumors. The response was evaluated in the context of clinical and graph features. I have a few comments to improve the manuscript.
For me, it was not clear if the identification of lymphocytes in the tumors was appropriately performed. The methodology has been used in another study or is widely used? If not, the authors reinforce their findings with histological analysis using monoclonal antibodies reactive to specific markers of lymphocytes?
Table 1 is confusing. I suggest that the two experiments are classified independently from each other in all parameters, including median age, sex, etc.
The Y axis in Figure 3 is very difficult to read.
Author Response
Please see attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors have answered all my concerns.