Analyses of CNS Response to Osimertinib in Patients with T790M-Positive Advanced NSCLC from ASTRIS Korean Subset, Open-Label Real-World Study

Round 1

Reviewer 1 Report

This is an interesting study aiming to analyze response to Osimertinib in patients with T790-positive advanced NSCLC, trough a clear study design, statistical method, defined inclusion criteria and imaging review by independent radiologists. The table summarizing patients’ characteristics and consort diagram are complete and helpful in interpreting the results. Minor remarks/changes could be: a) Add few details such as how many patients were evaluated with MRI vs. CT scan and if the independent radiologists considered these two different techniques equivalent. b) Introduce a better definition of LM diseases and be consisted with it in all parts of the paper (methods, results and discussion) as the sentence at pag 3: ….19 patients were suspected to have LM…. Is misleading. I would suggest to write only 19 cases showed LM disease on imaging. c) Consider to include an image showing one of the 2 cases with LM disease who presented CR. d) For the 38 cases with prior RT it should be indicated at which median interval from end of RT CNS imaging was performer (this could be relevant as there is a difference in response between cases with no prior RT and cases with no RT; furthermore it could be indicated if cases with prior RT had clear intracranial progression either of lesions in irradiated fields or in not irradiated fields. e) Pages should be checked for numeration f) Non published material is not written in English!

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

This study demonstrates the potential role of intracranial efficacies of osimertinib  administration in T790M-positive advanced NSCLC with or without brain metastases. The paper is well written and represents a real world approach to this widely known clinical problem; originality is not a strong point of the study as well as the limited serie of patients. However there are few real world analyses in literature; the main part of the studies reflect clinical trials and not daily practice. As the Authors pointed out conventional cancer clinical trials can be slow and costly, often produce results with limited external validity, and are difficult for patients to participate in.  The statistical approach is correct even if difficult to read in the present form.
I suggest to make the paragraphs 3.2 and 3.4 easier to read. The mass of abbreviations makes reading heavy.

The colors of the figure 3B are confusing and should be changed.

The dramatic problem of the leptomeningeal diffusions should be developed and the diagnostic difficulties described. The lack  of cerebrospinal fluid study unfortunately widespread in many Oncological Hospitals deserves a specific comment

Author Response

Please see the attachment

Author Response File: Author Response.docx