摘要
目的:构建基于超声造影及血清学指标的无创模型(HCC-ER)用于预测肝癌术后早期复发(术后2年内),并对模型进行内部与外部验证。方法:纳入2017年1月—2020年1月于泉州市第一医院及福建孟超肝胆医院实行了肝切除术的患者,收集患者的一般信息、超声信息和血清学信息。超声信息包括二维超声特征和超声造影LI-RADS分类等;血清学包括AFP、肝功能等。以福建孟超肝胆医院70%的患者数据作为建模数据,余30%作为内部验证组数据,泉州市第一医院数据作为外部验证组数据。构建的模型命名为HCC-ER预测模型。比较HCC-ER与其他经典肿瘤分级系统(TNM,BCLC和CNLC)的预测效率。结果:建模组、内部验证组、外部验证组的基线是均衡的。LI-RADS分类、AFP、肿瘤最大径和总胆红素(TBIL)是预测早期复发的独立风险因素,HR值分别为1.505,1.286,1.115,1.020。依据预测模型,计算所有患者风险评分,以100作为最佳截断值,在建模组、内部验证组和外部验证组,都能很好的将复发高危组与复发低危组进行区分。结论:笔者所构建的HCC-ER模型对于肝癌术后的早期复发预测具有较高的准确性,可为临床制定精准治疗方案提供重要价值,值得在临床中进一步推广。
Objective:To construct a non-invasive prediction model based on ultrasonography and serological indicators for early recurrence(within 2 years after surgery)of hepatocellular carcinoma,and to validate the model(HCC-ER).Methods:Hepatectomies performed in Quanzhou First Hospital and Fujian Mengchao Hepatobiliary Hospital from January 2017 to January 2020 were included.General information,ultrasound information(two-dimensional features)and LI-RADS classification,serological indexes(AFP and liver function)were collected.The data from Mengchao Hospital were used as the modeling group and internal validation group data in the ratio of 7:3,and the data from Quanzhou First Hospital were used as the external validation group data.The constructed model was named as HCC-ER predict model.The prediction efficiency of HCC-ER was compared with other classical models.Results:Baseline was balanced in the modeling group,internal validation group,and external validation group.LI-RADS classification,AFP,maximum tumor diameter,and total bilirubin were independent risk factors for predicting early recurrence with HR values of 1.505,1.286,1.115,1.020,respectively.Based on the prediction model,risk scores were calculated for all patients,and 100 was used as the best cutoff value to distinguish the high-risk recurrence group from the low-risk recurrence group well in the model building group,the internal validation group,and the external validation group.The correction curve of the prediction model slightly overestimated the risk of recurrence.Conclusion:The HCC-ER model constructed by the author has high accuracy for early recurrence prediction after hepatocellular carcinoma surgery,which can provide important value for the clinical formulation of accurate treatment plans and is worthy of further promotion in clinical practice.
作者
余卫峰
涂海斌
曾志雄
林少坤
YU Wei-feng;TU Hai-bin;ZENG Zhi-xiong;LIN Shao-kun(Department of Ultrasound,Quanzhou First Hospital Affiliated to Fujian Medical University,Quanzhou Fujian 362000,China;Department of Ultrasound,Mengchao Hepatobiliary Hospital of Fujian Medical University,Fuzhou 350000,China)
出处
《中国临床医学影像杂志》
CAS
CSCD
2023年第7期490-495,共6页
Journal of China Clinic Medical Imaging
基金
福建医科大学启航基金资助:2019QH1302。
关键词
肝肿瘤
胆管造影术
超声检查
Liver Neoplasms
Cholangiography
Ultrasonography
