Machine learning prediction of early recurrence after surgery for gallbladder cancer.

IF 8.6 1区医学 Q1 SURGERY

British Journal of Surgery Pub Date : 2024-10-30 DOI:10.1093/bjs/znae297

Giovanni Catalano, Laura Alaimo, Odysseas P Chatzipanagiotou, Andrea Ruzzenente, Federico Aucejo, Hugo P Marques, Vincent Lam, Tom Hugh, Nazim Bhimani, Shishir K Maithel, Minoru Kitago, Itaru Endo, Timothy M Pawlik

{"title":"Machine learning prediction of early recurrence after surgery for gallbladder cancer.","authors":"Giovanni Catalano, Laura Alaimo, Odysseas P Chatzipanagiotou, Andrea Ruzzenente, Federico Aucejo, Hugo P Marques, Vincent Lam, Tom Hugh, Nazim Bhimani, Shishir K Maithel, Minoru Kitago, Itaru Endo, Timothy M Pawlik","doi":"10.1093/bjs/znae297","DOIUrl":null,"url":null,"abstract":"Background: Gallbladder cancer is often associated with poor prognosis, especially when patients experience early recurrence after surgery. Machine learning may improve prediction accuracy by analysing complex non-linear relationships. The aim of this study was to develop and evaluate a machine learning model to predict early recurrence risk after resection of gallbladder cancer.Methods: In this cross-sectional study, patients who underwent resection of gallbladder cancer with curative intent between 2001 and 2022 were identified using an international database. Patients were assigned randomly to a development and an evaluation cohort. Four machine learning models were trained to predict early recurrence (within 12 months) and compared using the area under the receiver operating curve (AUC).Results: Among 374 patients, 56 (15.0%) experienced early recurrence; most patients had T1 (51, 13.6%) or T2 (180, 48.1%) disease, and a subset had lymph node metastasis (120, 32.1%). In multivariable Cox analysis, resection margins (HR 2.34, 95% c.i. 1.55 to 3.80; P < 0.001), and greater AJCC T (HR 2.14, 1.41 to 3.25; P < 0.001) and N (HR 1.59, 1.05 to 2.42; P = 0.029) categories were independent predictors of early recurrence. The random forest model demonstrated the highest discrimination in the evaluation cohort (AUC 76.4, 95% c.i. 66.3 to 86.5), compared with XGBoost (AUC 74.4, 53.4 to 85.3), support vector machine (AUC 67.2, 54.4 to 80.0), and logistic regression (AUC 73.1, 60.6 to 85.7), as well as good accuracy after bootstrapping validation (AUC 75.3, 75.0 to 75.6). Patients classified as being at high versus low risk of early recurrence had much worse overall survival (36.1 versus 63.8% respectively; P < 0.001). An easy-to-use calculator was made available (https://catalano-giovanni.shinyapps.io/GallbladderER).Conclusion: Machine learning-based prediction of early recurrence after resection of gallbladder cancer may help stratify patients, as well as help inform postoperative adjuvant therapy and surveillance strategies.","PeriodicalId":136,"journal":{"name":"British Journal of Surgery","volume":"111 11","pages":""},"PeriodicalIF":8.6000,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"British Journal of Surgery","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1093/bjs/znae297","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"SURGERY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Gallbladder cancer is often associated with poor prognosis, especially when patients experience early recurrence after surgery. Machine learning may improve prediction accuracy by analysing complex non-linear relationships. The aim of this study was to develop and evaluate a machine learning model to predict early recurrence risk after resection of gallbladder cancer.

Methods: In this cross-sectional study, patients who underwent resection of gallbladder cancer with curative intent between 2001 and 2022 were identified using an international database. Patients were assigned randomly to a development and an evaluation cohort. Four machine learning models were trained to predict early recurrence (within 12 months) and compared using the area under the receiver operating curve (AUC).

Results: Among 374 patients, 56 (15.0%) experienced early recurrence; most patients had T1 (51, 13.6%) or T2 (180, 48.1%) disease, and a subset had lymph node metastasis (120, 32.1%). In multivariable Cox analysis, resection margins (HR 2.34, 95% c.i. 1.55 to 3.80; P < 0.001), and greater AJCC T (HR 2.14, 1.41 to 3.25; P < 0.001) and N (HR 1.59, 1.05 to 2.42; P = 0.029) categories were independent predictors of early recurrence. The random forest model demonstrated the highest discrimination in the evaluation cohort (AUC 76.4, 95% c.i. 66.3 to 86.5), compared with XGBoost (AUC 74.4, 53.4 to 85.3), support vector machine (AUC 67.2, 54.4 to 80.0), and logistic regression (AUC 73.1, 60.6 to 85.7), as well as good accuracy after bootstrapping validation (AUC 75.3, 75.0 to 75.6). Patients classified as being at high versus low risk of early recurrence had much worse overall survival (36.1 versus 63.8% respectively; P < 0.001). An easy-to-use calculator was made available (https://catalano-giovanni.shinyapps.io/GallbladderER).

Conclusion: Machine learning-based prediction of early recurrence after resection of gallbladder cancer may help stratify patients, as well as help inform postoperative adjuvant therapy and surveillance strategies.

查看原文本刊更多论文

胆囊癌术后早期复发的机器学习预测

背景：胆囊癌通常预后较差，尤其是当患者术后出现早期复发时。机器学习可通过分析复杂的非线性关系来提高预测的准确性。本研究旨在开发和评估一种机器学习模型，用于预测胆囊癌切除术后的早期复发风险：在这项横断面研究中，利用国际数据库对 2001 年至 2022 年间接受胆囊癌根治性切除术的患者进行了识别。患者被随机分配到开发队列和评估队列。对四种机器学习模型进行了训练，以预测早期复发（12个月内），并使用接收器工作曲线下面积（AUC）进行比较：在 374 名患者中，有 56 人（15.0%）出现早期复发；大多数患者的病情为 T1（51 人，13.6%）或 T2（180 人，48.1%），还有一部分患者出现淋巴结转移（120 人，32.1%）。在多变量考克斯分析中，切除边缘（HR 2.34，95% 置信区间为 1.55 至 3.80；P < 0.001）、更大的 AJCC T（HR 2.14，1.41 至 3.25；P < 0.001）和 N（HR 1.59，1.05 至 2.42；P = 0.029）类别是早期复发的独立预测因素。与 XGBoost（AUC 74.4，53.4 至 85.3）、支持向量机（AUC 67.2，54.4 至 80.0）和逻辑回归（AUC 73.1，60.6 至 85.7）相比，随机森林模型在评估队列中表现出最高的区分度（AUC 76.4，95% c.i.66.3 至 86.5），并且在引导验证后具有良好的准确性（AUC 75.3，75.0 至 75.6）。被归类为早期复发高风险和低风险的患者的总生存率要差得多（分别为36.1%和63.8%；P < 0.001）。我们还提供了一个简单易用的计算器（https://catalano-giovanni.shinyapps.io/GallbladderER）：基于机器学习的胆囊癌切除术后早期复发预测有助于对患者进行分层，并为术后辅助治疗和监测策略提供依据。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

British Journal of Surgery 医学-外科

CiteScore

12.70

自引率

7.30%

发文量

1102

审稿时长

1.5 months

期刊介绍： The British Journal of Surgery (BJS), incorporating the European Journal of Surgery, stands as Europe's leading peer-reviewed surgical journal. It serves as an invaluable platform for presenting high-quality clinical and laboratory-based research across a wide range of surgical topics. In addition to providing a comprehensive coverage of traditional surgical practices, BJS also showcases emerging areas in the field, such as minimally invasive therapy and interventional radiology. While the journal appeals to general surgeons, it also holds relevance for specialty surgeons and professionals working in closely related fields. By presenting cutting-edge research and advancements, BJS aims to revolutionize the way surgical knowledge is shared and contribute to the ongoing progress of the surgical community.