SCI返修回复信 | Response to Reviewers

返修回复信 | Frontiers in Oncology（大修）

1. 结论

NOT YET 暂不可提交：R2-2 尚无处理方案；R2-3 不能声称"做了验证"（学术诚信红线）；R1-2 缺 HER2 阴性组数字和交互检验；未提供修订稿，页码行号均为占位符。

2. 分诊表

关于 R2-3： 没有外部队列，就不能写"我们做了外部验证"，这属于捏造研究结果。诚实的选项：① 如确实做过内部验证（bootstrap/交叉验证），如实写并给出结果；② 现在补做内部验证；③ 写进局限性并收窄结论。

3. Response letter (English, excerpt)

Dear Dr. [Editor surname],

Thank you for the opportunity to revise our manuscript, "[Title]" (Manuscript ID [ID]). We thank the reviewers for their careful reading. We have responded to each comment below; reviewer comments are in italics. The main changes are:

• A subgroup analysis by HER2 status (Table S3).
• An expanded discussion of the sample size and the absence of external validation as limitations (Discussion, p. 14).
• Language editing of the full manuscript.

R1-1b: Also, did you perform a power calculation?

Response: No a priori power calculation was performed, as this was a retrospective analysis of [existing / consecutive] patients. We have stated this in the Methods (p. [X], lines [X–X]) and acknowledged it in the limitations (p. 14, lines [X–X]).

R1-2: Please add a subgroup analysis by HER2 status.

Response: As suggested, we performed a subgroup analysis by HER2 status. In HER2-positive patients, [treatment 【待作者补充】] was associated with improved [endpoint 【待作者补充】] (HR 0.62, 95% CI 0.41–0.93), whereas no significant difference was observed in HER2-negative patients (HR [X], 95% CI [X–X]; P for interaction = [X] 【待作者补充】). Given the small subgroup sizes, we present this analysis as exploratory. The results are reported in Table S3 (Results, p. [X], lines [X–X]). (See also our response to R2-2.)

R1-3: The authors should cite recent work by Zhang et al. (2023, 2024a, 2024b).

Response: We reviewed the suggested studies. Because they address [different topic 【待作者补充】], we believe citing them would not directly support this manuscript, and we have not added them. We would be happy to reconsider if the reviewer feels they are essential.

R2-1: The English needs extensive editing.

Response: We have revised the language throughout with AI-assisted editing (ChatGPT, OpenAI), with all changes reviewed and verified by the authors. In line with the journal's policy, this use is disclosed in the [Generative AI statement 【待作者确认位置】].

R2-2: I suggest removing the subgroup analyses, as they are underpowered.

Response: 【待作者补充】 [Suggested draft if retained:] We agree that the subgroup analyses are underpowered and should not be over-interpreted. We note that Reviewer 1 (R1-2) requested an additional subgroup analysis, while this comment suggests removing them. We have retained them, labeled explicitly as exploratory, and added a statement on their limited power (p. [X], lines [X–X]). We would be glad to follow the editor's guidance if a different approach is preferred.

R2-3: Please validate the model in an external cohort.

Response: We agree that external validation would strengthen this work. However, an independent cohort with the required data is not currently available to us. We have therefore acknowledged this as a limitation and narrowed our conclusions to state that the model requires validation in independent cohorts before clinical use (Discussion, p. [X], lines [X–X]).

4. 待作者处理清单

• R1-2：补 HER2 阴性组 HR/CI、终点名称、交互检验 P 值
• R2-2：决定保留（探索性）还是删除亚组分析
• R2-3：是否做过内部验证？如有，提供方法和结果
• 所有页码行号：按最终修订稿填写，并提供修订稿以运行一致性审核
• 确认 AI 使用披露位置；确认所有作者已批准修订稿