Survival Machine Learning Methods Improve Prediction of Histologic Transformation in Follicular and Marginal Zone Lymphomas

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Kim and coworkers reported the development and validation of survival-based and traditional classification machine-learning models to predict histologic transformation in cohorts.  They evaluate the model performance across independent FL and MZL cohorts, assess the added value of incorporating NGS data, and provide practical insights into optimizing HT risk prediction strategies. The data presented support the conclusion drawn, which is useful for others in the related research field. In view of this, the manuscript may be accepted for publication after minor revise:

1. Page 3, please confirm 2026 is right in “Patients diagnosed
   between January 2011 and February 2026 were included.”.
2. Table 1, please set P in italics in the table and wherever applicable.
3. Table 1, add N for the Total group.

Author Response

Reviewer 1 Comments for the Author:

Kim and coworkers reported the development and validation of survival-based and traditional classification machine-learning models to predict histologic transformation in cohorts.  They evaluate the model performance across independent FL and MZL cohorts, assess the added value of incorporating NGS data, and provide practical insights into optimizing HT risk prediction strategies. The data presented support the conclusion drawn, which is useful for others in the related research field. In view of this, the manuscript may be accepted for publication after minor revise:

Page 3, please confirm 2026 is right in “Patients diagnosed between January 2011 and February 2026 were included.”.

à Thank you for pointing this out. The year 2026 was a typographical error. It has been corrected to February 2025 throughout the manuscript to ensure consistency.

2) Table 1, please set P in italics in the table and wherever applicable.

Table 1, add N for the Total group.

à We appreciate the suggestion. We have formatted P in italics in Table 1 and consistently wherever applicable across the manuscript, and added N for the total cohort in Table 1.

Table 1 updates:

P values are now formatted as P.

“Total” row now includes N = 1,068.

We also ensured that table notes specify the statistical tests used and the data presentation format, following the journal’s style.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript is clear, logically structured, and relevant to the field of haemato-oncology. The topic of predicting histological transformation in FL and MZL lymphoma is of both clinical and scientific importance. The introduction, methods, results, and discussion are well-connected. The study is methodologically well designed. 

The year 2026 was entered by mistake: Patients diagnosed between January 2011, and February 2026 were included.

Author Response

Reviewer 2 Comments for the Author:

Comments and Suggestions for Authors

The manuscript is clear, logically structured, and relevant to the field of haemato-oncology. The topic of predicting histological transformation in FL and MZL lymphoma is of both clinical and scientific importance. The introduction, methods, results, and discussion are well-connected. The study is methodologically well designed.

The year 2026 was entered by mistake: Patients diagnosed between January 2011, and February 2026 were included.

à We sincerely thank the reviewer for the positive evaluation of our work and for the helpful comment.