Predictive model for severe adverse reaction associated with bevacizumab based on the global trigger tool and machine learning

FU Yongfei; LONG Xin; XU Hongzhen; TANG Jian; LI Xiangqing; LONG Yucheng; QIN Dong

doi:10.6039/j.issn.1001-0408.2026.04.15

您当前的位置：

首页 >

文章列表页 >

Predictive model for severe adverse reaction associated with bevacizumab based on the global trigger tool and machine learning

更新时间：2026-02-27

- Predictive model for severe adverse reaction associated with bevacizumab based on the global trigger tool and machine learning
- ZHONGGUO YAOFANG Vol. 37, Issue 4, Pages: 497-503(2026)
- 作者机构：
  
  1.桂林医科大学药学院，广西桂林　541199
  2.桂林市人民医院药学部，广西桂林　541002
  3.桂林市人民医院数据中心，广西桂林　541002
- 作者简介：
- 基金信息：
- DOI：10.6039/j.issn.1001-0408.2026.04.15
  CLC： R979.1;R969.3;TP181
- Received：05 August 2025，
  
  Revised：2026-01-05，
  
  Accepted：07 January 2026，
  
  Published：28 February 2026
- 稿件说明：
移动端阅览
符永妃,龙芯,徐宏珍,等.基于全面触发工具与机器学习的贝伐珠单抗严重不良反应预测模型研究[J].中国药房,2026,37(04):497-503.

FU Yongfei,LONG Xin,XU Hongzhen,et al.Predictive model for severe adverse reaction associated with bevacizumab based on the global trigger tool and machine learning[J].ZHONGGUO YAOFANG,2026,37(04):497-503.
符永妃,龙芯,徐宏珍,等.基于全面触发工具与机器学习的贝伐珠单抗严重不良反应预测模型研究[J].中国药房,2026,37(04):497-503. DOI： 10.6039/j.issn.1001-0408.2026.04.15.

FU Yongfei,LONG Xin,XU Hongzhen,et al.Predictive model for severe adverse reaction associated with bevacizumab based on the global trigger tool and machine learning[J].ZHONGGUO YAOFANG,2026,37(04):497-503. DOI： 10.6039/j.issn.1001-0408.2026.04.15.

摘要

目的

确定贝伐珠单抗相关不良反应（ADR）的触发器条目，判定并分析相关ADR发生情况，构建该药致严重不良反应（SAR）的预测模型。

方法

基于全面触发工具（GTT）理念，参考GTT白皮书、药品说明书及相关文献，结合单轮德尔菲法确定贝伐珠单抗相关ADR触发器条目；利用所建条目，基于中国医院药物警戒系统对桂林市人民医院2020年1月至2024年9月相关患者的电子病历进行主动筛查，由药师判定并统计贝伐珠单抗致ADR的发生情况。以所得触发器条目阳性患者的资料为对象（按7∶3划分训练集和测试集），通过Boruta算法从训练集相关39项变量中筛选候选特征变量，以是否发生SAR为因变量进行多因素Logistic回归分析；以上述候选特征变量为基础，构建Logistic回归、极端梯度提升、轻量级梯度提升机、随机森林、类别特征提升模型，通过受试者操作特征曲线的曲线下面积（AUC）、召回率等指标评估模型性能，应用Shapley加性解释（SHAP）法分析解释各变量的贡献，并基于最优模型构建列线图。

结果

最终确定用于贝伐珠单抗相关ADR主动监测的触发器条目38项，含检验指标17项、临床表现13项、干预措施8项。共纳入触发器条目阳性患者483例，其中发生ADR者318例，发生SAR者83例；触发器条目和病例阳性预测值分别为43.57%（708/1 625）和63.84%（318/483）。贝伐珠单抗致ADR涉及7个系统/器官，以血液系统受累最为常见（64.15%）。经Boruta算法共筛选到血钾、红细胞压积、白/球蛋白比值、前白蛋白、既往高血压史、年龄、红细胞计数7个候选变量。多因素Logistic回归分析显示，血钾水平升高（OR＝0.234，

＝0.002）与贝伐珠单抗致SAR风险降低相关，既往高血压史（OR＝2.642，

＝0.006）和年龄增加（OR＝1.040，

＝0.025）与贝伐珠单抗致SAR风险升高有关。Logistic回归模型的AUC、F1值、召回率（0.761、0.447、0.607）均高于其他模型；SHAP评估结果显示，血钾、红细胞压积、年龄等变量的重要性位居前列。

结论

成功确定38项触发器条目用于贝伐珠单抗相关ADR的主动筛查。血钾水平升高是贝伐珠单抗致SAR的保护因素，而既往高血压史、年龄增加则是危险因素；Logistic回归模型为贝伐珠单抗致SAR的最优预测模型。

Abstract

OBJECTIVE

To confirm trigger items for adverse drug reaction （ADR） induced by bevacizumab， to identify and analyze the occurrence of related ADR， and to establish a predictive model for severe adverse reaction （SAR） caused by this drug.

METHODS

Based on the global trigger tool （GTT） theory， and referencing the GTT White Paper， drug package inserts and relevant literature， trigger items for bevacizumab-related ADR were confirmed using a single-round Delphi method. Utilizing these established items， electronic medical records of relevant patients at Guilin People’s Hospital from January 2020 to September 2024 were actively screened via the China Hospital Pharmacovigilance System. Pharmacists then identified and tallied the occurrence of bevacizumab-induced ADR. Data from patients with any positive trigger item served as the study subjects （divided into training and test sets at a ratio of 7∶3）， candidate feature variables were selected from 39 related variables using the Boruta algorithm， and the multivariable Logistic regression analysis was performed with the occurrence of SAR as the dependent variable. Based on these candidate features， Logistic Regression， Extreme Gradient Boosting， Light Gradient Boosting Machine， Random Forest， and Categorical Boosting models were constructed. Model performance was evaluated using metrics including the area under the curve （AUC） of receiver operating characteristic curve and recall rate. The Shapley Additive exPlanations （SHAP） method was applied to analyze and interpret the contribution of each variable. A nomogram was constructed based on the optimal model.

RESULTS

A total of 38 trigger items for active monitoring of bevacizumab-related ADR were determined， comprising 17 laboratory indicators， 13 clinical manifestations， and 8 intervention measures. In total， 483 patients with positive trigger items were included， and 318 patients with bevacizumab-induced ADR were identified， including 83 SARs. The positive predictive values for the trigger items and cases were 43.57% （708/1 625） and 63.84% （318/483）， respectively. Bevacizumab-induced ADR involved 7 systems/organs， with the hematological system being the most frequently involved （64.15%）. The Boruta algorithm selected 7 vari

ables： serum potassium， hematocrit， albumin-to-globulin ratio， prealbumin， hypertension history， age and red blood cell count. Multivariable Logistic regression showed that elevated serum potassium levels were associated with a decreased risk of bevacizumab-induced SAR （OR＝0.234，

＝0.002）， while a history of hypertension （OR＝2.642，

＝0.006） and increased age （OR＝1.040，

＝0.025） were associated with an increased risk. The Logistic Regression model demonstrated superior performance with higher AUC， F1 score and recall rate （0.761， 0.447， 0.607）， compared to other models. SHAP evaluation results indicated that variables such as serum potassium， hematocrit， and age ranked highest in importance.

CONCLUSIONS

Totally 38 trigger entries have been successfully identified for active screening of bevacizumab-related ADR. Elevated serum potassium levels are a protective factor against bevacizumab-induced SAR， whereas the hypertension history and increased age are risk factors. The Logistic Regression model is the optimal predictive model.

关键词

Keywords

references

BRAY F ， LAVERSANNE M ， SUNG H ， et al . Global cancer statistics 2022：GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries [J ] . CA Cancer J Clin ， 2024 ， 74 （ 3 ）： 229 - 263 .

CHEN S M ， CAO Z ， PRETTNER K ， et al . Estimates and projections of the global economic cost of 29 cancers in 204 countries and territories from 2020 to 2050 [J ] . JAMA Oncol ， 2023 ， 9 （ 4 ）： 465 - 472 .

ZHANG M Y ， ZHU J ， BAO Y ， et al . Bevacizumab in ovarian cancer therapy：current advances，clinical challenges，and emerging strategies [J ] . Front Bioeng Biotechnol ， 2025 ， 13 ： 1589841 .

MAO C L ， SEOW K M ， CHEN K H . The utilization of bevacizumab in patients with advanced ovarian cancer：a systematic review of the mechanisms and effects [J ] . Int J Mol Sci ， 2022 ， 23 （ 13 ）： 6911 .

MENG X ， WU Y Z ， LIU Z X ， et al . Active monitoring of antifungal adverse events in hospitalized patients based on global trigger tool method [J ] . Front Pharmacol ， 2024 ， 15 ： 1322587 .

WU S ， YIN Q N ， WU L Y ， et al . Establishing a trigger tool based on global trigger tools to identify adverse drug events in obstetric inpatients in China [J ] . BMC Health Serv Res ， 2024 ， 24 （ 1 ）： 72 .

LIU Y ， LIU X J ， XIA B B ， et al . The application of global trigger tool in monitoring antineoplastic adverse drug events：a retrospective study [J ] . Front Oncol ， 2024 ， 14 ： 1230514 .

GARCIA J ， HURWITZ H I ， SANDLER A B ， et al . Bevacizumab（avastin ® ）in cancer treatment：a review of 15 years of clinical experience and future outlook [J ] . Cancer Treat Rev ， 2020 ， 86 ： 102017 .

PATEL J N ， JIANG C ， OWZAR K ， et al . Pharmacogenetic and clinical risk factors for bevacizumab-related gastrointestinal hemorrhage in prostate cancer patients treated on CALGB 9040：alliance [J ] . Pharmacogenomics J ， 2024 ， 24 （ 2 ）： 6 .

徐宏珍，陈嘉，廖倩倩，等 . 基于CHPS系统主动监测对贝伐珠单抗不良反应发生特点及相关性回顾性分析 [J ] . 中文科技期刊数据库（引文版）医药卫生， 2024 （ 5 ）： 65 - 68 .

丁年羊，李莉，方攀奇，等 . 我院贝伐珠单抗生物类似药与原研药相关不良反应回顾性分析 [J ] . 中国药房， 2024 ， 35 （ 4 ）： 472 - 475 .

杨江，王明航，李建生，等 . 基于改良德尔菲法的支气管哮喘急性发作风险预警因素调查表及条目筛选研究 [J ] . 中国全科医学， 2022 ， 25 （ 35 ）： 4425 - 4432 .

SUN J ， DENG X Y ， CHEN X P ， et al . Incidence of adverse drug reactions in COVID-19 patients in China：an active monitoring study by hospital pharmacovigilance system [J ] . Clin Pharmacol Ther ， 2020 ， 108 （ 4 ）： 791 - 797 .

ZHANG Y F ， ZHANG H Y ， XU H Q ， et al . Efficacy and safety of hepatic arterial infusion chemotherapy combined with lenvatinib and PD-1 inhibitors for advanced hepatocellular carcinoma with macrovascular invasion [J ] . World J Surg Oncol ， 2024 ， 22 （ 1 ）： 122 .

卫生部 . 药品不良反应报告和监测管理办法 [EB/OL ] .（ 2011-05-04 ）[ 2025-07-31 ] . http：//www.gov.cn/zhengce/2021-06/29/content_5723552.htm http://www.gov.cn/zhengce/2021-06/29/content_5723552.htm .

MOONS K G M ， ALTMAN D G ， REITSMA J B ， et al . Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis（TRIPOD）：explanation and elaboration [J ] . Ann Intern Med ， 2015 ， 162 （ 1 ）： W1 - W73 .

ZHANG N ， PAN L Y ， CHEN W Y ， et al . A risk-factor model for antineoplastic drug-induced serious adverse events in cancer inpatients：a retrospective study based on the global trigger tool and machine learning [J ] . Front Pharmacol ， 2022 ， 13 ： 896104 .

LUNDBERG S M ， LEE S I . A unified approach to interpreting model predictions [J ] . NeurIPS ， 2017 ， 30 ： 4765 - 4774 .

LIMA S I V C ， MARTINS R R ， SALDANHA V ， et al . Development and validation of a clinical instrument to predict risk of an adverse drug reactions in hospitalized patients [J ] . PLoS One ， 2020 ， 15 （ 12 ）： e0243714 .

ZHENG Z L ， ZHAO Y H ， XIE J W ， et al . Clinical risk factors of bevacizumab-related hypertension in patients with metastatic colorectal cancer：a retrospective study [J ] . Front Pharmacol ， 2024 ， 15 ： 1463026 .

ROUSSEAU A ， GÉRAUD A ， GEISS R ， et al . Safety of solid oncology drugs in older patients：a narrative review [J ] . ESMO Open ， 2024 ， 9 （ 11 ）： 103965 .

MOHAMMED T ， SINGH M ， TIU J G ， et al . Etiology and management of hypertension in patients with cancer [J ] . Cardiooncology ， 2021 ， 7 （ 1 ）： 14 .

KIM J M ， JUNG H ， KWON H E ， et al . Predicting prognostic factors in kidney transplantation using a machine learning approach to enhance outcome predictions：a retrospective cohort study [J ] . Int J Surg ， 2024 ， 110 （ 11 ）： 7159 - 7168 .

GENG Z ， YANG C Y ， ZHAO Z Y ， et al . Development and validation of a machine learning-based predictive model for assessing the 90-day prognostic outcome of patients with spontaneous intracerebral hemorrhage [J ] . J Transl Med ， 2024 ， 22 （ 1 ）： 236 .

LIU X Z ， XIE Z L ， ZHANG Y ， et al . Machine learning for predicting in-hospital mortality in elderly patients with heart failure combined with hypertension：a multicenter retrospective study [J ] . Cardiovasc Diabetol ， 2024 ， 23 （ 1 ）： 407 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Construction and validation of a medication deviation prediction model for hospital-to-home transition period in coronary heart disease patients with initial treatment

Advances in the application of machine learning in the prediction of adverse drug reactions

我院317 例药品不良反应报告分析

我院抗菌药物致不良反应233例分析

化疗联合贝伐珠单抗治疗结直肠癌肝转移的临床观察

Related Author

LI Yushuang

LI Shu

ZHANG Qianying

HUANG Yan

LIU Kun

GU Xiulin

JIANG Huanhuan

XU Mengjia

Related Institution

Dept. of Pharmacy， the Affiliated Hospital of North China University of Science and Technology

School of Pharmacy， North China University of Science and Technology

Dept. of Pharmacy， the Second Affiliated Hospital of Hebei Northern University

Dept. of Pharmacy， the First Affiliated Hospital of Soochow University

College of Pharmaceutical Sciences， Soochow University

⁰