Risk factors associated with overall survival in patients with cervical cancer: a prospective cohort study in Western China comparing random survival forest and Cox proportional hazards models.
Zejia Mao, Ling Long, Li Yuan, Qianjie Xu, Misi He, Haike Lei, Dongling Zou
{"title":"Risk factors associated with overall survival in patients with cervical cancer: a prospective cohort study in Western China comparing random survival forest and Cox proportional hazards models.","authors":"Zejia Mao, Ling Long, Li Yuan, Qianjie Xu, Misi He, Haike Lei, Dongling Zou","doi":"10.3802/jgo.2026.37.e2","DOIUrl":null,"url":null,"abstract":"<p><strong>Objective: </strong>Cervical cancer (CCa) significantly affects female fertility and quality of life. This study aimed to construct and validate a random survival forest (RSF) model to identify the factors that affect the overall survival (OS) in patients with CCa in China and compare its performance with that of the Cox proportional hazards model (Cox model).</p><p><strong>Methods: </strong>Data on CCa patients were collected from Chongqing University Cancer Hospital. The performance and discrimination ability of the models were evaluated via the C-index, integrated Brier score (IBS), accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). The Kaplan-Meier (K-M) survival curve was used to analyze the difference in OS between patients with high and low risk predicted by RSF model.</p><p><strong>Results: </strong>A total of 3,982 patients were included in this study. Comparing to Cox model, the RSF model ranked important variables and identified radiotherapy (RT) as an important treatment measure. A comprehensive analysis of the evaluation indices confirmed that the RSF model outperformed the Cox model (IBS: 0.152 vs. 0.162, C-index: 0.863 vs. 0.764). The RSF model metrics for the validation cohort (VC) were as follows: 1-, 3-, and 5-year AUC (0.908, 0.884, and 0.869), sensitivity (0.746), specificity (0.825), and accuracy (0.808). The OS of low-risk patients predicted by RSF was greater than that of high-risk patients.</p><p><strong>Conclusion: </strong>The RSF model demonstrated excellent discrimination, calibrated predictions, and stratified risk for CCa patients. Furthermore, it outperformed the Cox model in predicting risks, thus enabling the delivery of personalised treatment and follow-up strategies.</p>","PeriodicalId":15868,"journal":{"name":"Journal of Gynecologic Oncology","volume":" ","pages":""},"PeriodicalIF":3.7000,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Gynecologic Oncology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.3802/jgo.2026.37.e2","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OBSTETRICS & GYNECOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Objective: Cervical cancer (CCa) significantly affects female fertility and quality of life. This study aimed to construct and validate a random survival forest (RSF) model to identify the factors that affect the overall survival (OS) in patients with CCa in China and compare its performance with that of the Cox proportional hazards model (Cox model).
Methods: Data on CCa patients were collected from Chongqing University Cancer Hospital. The performance and discrimination ability of the models were evaluated via the C-index, integrated Brier score (IBS), accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). The Kaplan-Meier (K-M) survival curve was used to analyze the difference in OS between patients with high and low risk predicted by RSF model.
Results: A total of 3,982 patients were included in this study. Comparing to Cox model, the RSF model ranked important variables and identified radiotherapy (RT) as an important treatment measure. A comprehensive analysis of the evaluation indices confirmed that the RSF model outperformed the Cox model (IBS: 0.152 vs. 0.162, C-index: 0.863 vs. 0.764). The RSF model metrics for the validation cohort (VC) were as follows: 1-, 3-, and 5-year AUC (0.908, 0.884, and 0.869), sensitivity (0.746), specificity (0.825), and accuracy (0.808). The OS of low-risk patients predicted by RSF was greater than that of high-risk patients.
Conclusion: The RSF model demonstrated excellent discrimination, calibrated predictions, and stratified risk for CCa patients. Furthermore, it outperformed the Cox model in predicting risks, thus enabling the delivery of personalised treatment and follow-up strategies.
目的：宫颈癌（CCa）显著影响女性生育能力和生活质量。本研究旨在构建并验证随机生存森林（RSF）模型，以确定影响中国CCa患者总生存期（OS）的因素，并将其与Cox比例风险模型（Cox模型）的性能进行比较。方法：收集重庆大学肿瘤医院CCa患者资料。通过c指数、综合Brier评分（IBS）、准确性、敏感性、特异性和受试者工作特征曲线下面积（AUC）评价模型的性能和识别能力。采用Kaplan-Meier （K-M）生存曲线分析RSF模型预测的高、低风险患者的OS差异。结果：本研究共纳入3982例患者。与Cox模型相比，RSF模型对重要变量进行排序，并将放疗（RT）确定为重要的治疗措施。综合评价指标分析，RSF模型优于Cox模型（IBS: 0.152 vs. 0.162, C-index: 0.863 vs. 0.764）。验证队列（VC）的RSF模型指标如下：1年、3年和5年AUC（0.908、0.884和0.869）、敏感性（0.746）、特异性（0.825）和准确性（0.808）。RSF预测低危患者的OS大于高危患者。结论：RSF模型对CCa患者具有出色的鉴别、校准预测和分层风险。此外，它在预测风险方面优于Cox模型，从而能够提供个性化治疗和后续策略。
期刊介绍：
The Journal of Gynecologic Oncology (JGO) is an official publication of the Asian Society of Gynecologic Oncology. Abbreviated title is ''J Gynecol Oncol''. It was launched in 1990. The JGO''s aim is to publish the highest quality manuscripts dedicated to the advancement of care of the patients with gynecologic cancer. It is an international peer-reviewed periodical journal that is published bimonthly (January, March, May, July, September, and November). Supplement numbers are at times published. The journal publishes editorials, original and review articles, correspondence, book review, etc.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
