摘要
目的构建胃癌发生的影响因素模型,验证其预测胃癌发生风险效能。方法选取胃癌患者105例和健康人群210例,按照8∶2的比例将315例受试者分为训练集和验证集,分析训练集中胃癌患者和健康人群的临床资料,采用多因素logistic回归分析胃癌发生的独立危险因素,构建列线图模型预测胃癌发生风险并进行验证。结果训练集中胃癌患者与健康人群胃癌家族史、粪便隐血试验、三酰甘油-葡萄糖(TyG)指数、TG、FBG、癌胚抗原(CEA)、糖类抗原(CA)19-9、CA125、CA153、中性粒细胞计数、中性粒细胞/淋巴细胞比值(NLR)比较均有统计学差异(P<0.05)。CEA、CA19-9、CA125、CA153、粪便隐血试验及TyG指数是胃癌发生的独立影响因素(P<0.05)。训练集和验证集中列线图模型预测胃癌发生的AUC分别为0.820[95%CI(0.763~0.877)]和0.774[95%CI(0.631~0.917)]。校准曲线显示,列线图模型在训练集与验证集中的预测结果与实际观测结果一致性较高,模型拟合良好。临床决策曲线显示,在临床工作中应用列线图模型预测胃癌发生风险可以获得正向净收益。结论CEA、CA19-9、CA125、CA153、粪便隐血试验和TyG指数是胃癌发生的影响因素。基于多因素分析构建的模型可用于预测胃癌的发生风险。
Objective To investigate the the factors influencing the occurrence of gastric cancer and construct the model to predict the risk of gastric cancer.Methods A total of 105patients with gastric cancer and 210healthy people were selected,and 315subjects were divided into training set and validation set according to the ratio of 8∶2.The clinical data of the patients with gastric cancer and the healthy people in the training set were analyzed.Multivariate logistic regression was used to analyze the independent risk factors for gastric cancer.The nomogram model was built to predict the efficiency for the risk of gastric cancer.Results There were significant differences in family history of gastric cancer,fecal occult blood test,triglyceride and glucose(TyG)index,TG,FBG,carcinoembryonic antigen(CEA),carbohydrate antigen(CA)19-9,CA125,CA153,neutrophil count,ratio of neutrophil to lymphocyte(NLR)between gastric cancer patients and healthy people in training set(P<0.05).CEA,CA19-9,CA125,CA153,fecal occult-blood test and TyG index were the independent influencing factors for gastric cancer(P<0.05).The AUC of nomogram model for predicting gastric cancer in training set and validation set were 0.820[95%CI(0.763-0.877)]and 0.774[95%CI(0.631-0.917)],respectively.The calibration curve showed that the predicting results of the nomogram model in training set and validation set were in good agreement with the actual observation results,and the model fitted well.The clinical decision curve showed that the application of nomogram model to predict the risk of gastric cancer in clinical could obtain positive net benefit.Conclusion CEA,CA19-9,CA125,CA153,fecal occult blood test and TyG index are the influencing factors for gastric cancer.The model based on multivariate analysis can be used to predict the risk for gastric cancer.
作者
陈虎
刘世育
陈光侠
CHEN Hu;LIU Shiyu;CHEN Guangxia(First Clinical College,Xuzhou Medical University,Xuzhou221002,CHINA)
出处
《江苏医药》
CAS
2024年第4期401-408,共8页
Jiangsu Medical Journal
基金
徐州市医学领军人才培养项目(XWRCHT20210025)
徐州市重点研发计划(社会发展)项目(KC22095)。
关键词
胃癌
列线图
模型构建
Gastric cancer
Nomogram
Model construction
