本文已被：浏览 127次 下载 171次
投稿时间：2025-11-15 网络发布日期：2026-06-28
投稿时间：2025-11-15 网络发布日期：2026-06-28
中文摘要: 目的 采用生存树对女性弥漫型胃癌人群的预后进行风险分组，为临床治疗决策提供依据。方法 提取美国监测、流行病学和最终结果（SEER）数据库中2004年至2015年635例女性弥漫型胃癌患者的临床资料。采用Cox风险比例回归分析探究影响患者总生存期（OS）和肿瘤特异生存期（CSS）的相关因素。采用生存树获取患者不同OS和CSS下的预后分层分组，继而根据各组5年和10年预后的相似性进行整合，最终各获得四个风险递增的OS预测组（组1~组4）和四个CSS预测组（组A~组D），形成风险分层系统。采用随机生存森林分析计算各临床参数对OS和CSS的影响。结果 Cox风险比例回归结果显示，年龄>65岁、未经手术治疗、肿瘤直径>5cm、局域期和远处转移的患者OS和CSS较差，亚洲或太平洋岛民及经过化疗的患者预后较好。风险分层系统分析显示，组1、组2、组3、组4 患者的5年OS率分别为76.9%、40.8%、17.0%和2.2%，10年OS率分别为68.0%、26.1%、9.9%和2.2%。组A、组B、组C、组D患者5年CSS率分别为75.7%、42.8%、15.5%和2.4%，10年CSS率分别为68.0%、33.5%、8.3%和2.4%；四组间的5年、10年OS率、CSS率差异均有统计学意义（P＜0.01）。随机生存森林分析发现，肿瘤分期、手术、化疗、肿瘤直径、年龄及放疗是OS和CSS重要的影响因子。结论 肿瘤分期、手术、化疗、肿瘤直径、年龄及放疗等因素会影响OS和CSS，通过生存树确定的女性弥漫型胃癌预后的风险分层，可为该类患者的临床治疗决策和长期监测提供帮助。
Abstract:Objective To obtain the prognostic risk stratification of female patients with diffuse gastric cancer by the survival tree, aiming to provide basis for clinical treatment decision. Methods The clinical data of 635 female patients with diffuse gastric cancer were extracted from the Surveillance, Epidemiology and End Results（SEER）database from 2004 to 2015. Cox proportional hazard regression was used to identify factors affecting overall survival（OS）and cancer-specific survival（CSS）. The survival tree was used to obtain the prognosis stratification grouping of patients under different OS and CSS, after integrating according to the similarity of 5-year and 10-year prognosis of each group, 4 OS prediction groups（Group 1 to Group 4）and 4 CSS prediction groups（Group A to Group D）with increasing risks were finally obtained, forming a risk stratification system. Random survival forest was used to calculate the impact of clinical parameters on OS and CSS. Results Cox proportional hazard regression showed that patients with age>65 years, not receiving surgery, tumor diameter>5 cm, regional and distant stage had worse OS and CSS, while Asian or Pacific patients and those received chemotherapy had better OS and CSS. Risk stratification system analysis showed that the patients in Group 1, Group 2, Group 3 and Group 4 had the 5-year OS rates of 76.9%, 40.8%, 17.0%, and 2.2%, and the 10-year OS rates of 68.0%, 26.1%, 9.9%, and 2.2%, respectively. While the patients in Group A, Group B, Group C and Group D had the 5-year CSS rates of 75.7%, 42.8%, 15.5%, and 2.4%, and the 10-year CSS rates of 68.0%, 33.5%, 8.3%, and 2.4%, respectively. The differences in 5-year and 10-year OS rates and CSS rates among the four groups were all significant（P＜0.01）.Random survival forest analysis found that tumor stage, surgery, chemotherapy, tumor diameter, age, and radiotherapy were the important factors affecting OS and CSS. Conclusion Tumor stage, operation, chemotherapy, tumor diameter, age and radiotherapy can influence OS and CSS. The risk stratification of female diffuse gastric cancer prognosis determined by survival tree can provide assistance for clinical decision-making can help the clinical treatment decision and long-term monitoring of diffuse gastric cancer in women.
keywords: Diffuse gastric cancer Female Survival tree analysis Clinical prognosis Risk stratification
文章编号： 中图分类号：R735.2 文献标志码：A
基金项目:山东省自然科学基金资助项目（ZR2024QH365）；山东省立医院科研孵育基金（2023FY001）
附件
| 作者 | 单位 |
| 蔡磊 | 山东第一医科大学附属省立医院病理科,山东 济南 250021 |
| 吕蓓蓓 | 山东第一医科大学附属省立医院病理科,山东 济南 250021 |
| 苏文敬 | 山东第一医科大学附属省立医院病理科,山东 济南 250021 |
| 孙也淇 | 上海交通大学医学院附属新华医院病理科,上海 200092 |
引用文本：
蔡磊,吕蓓蓓,苏文敬,等.基于生存树分析的女性弥漫型胃癌预后分层研究[J].中国临床研究,2026,39(6):896-902.
蔡磊,吕蓓蓓,苏文敬,等.基于生存树分析的女性弥漫型胃癌预后分层研究[J].中国临床研究,2026,39(6):896-902.