肝脏 ›› 2020, Vol. 25 ›› Issue (12): 1330-1333.

• 其他肝病 • 上一篇    下一篇

小儿肝母细胞瘤预后风险模型的建立及其与PRETEXT分期系统的比较

贾益, 应海燕, 李少芳   

  1. 723000 陕西 汉中市3201医院新生儿科
  • 收稿日期:2020-06-05 出版日期:2020-12-31 发布日期:2021-02-26
  • 通讯作者: 贾益,Email:jen4722@163.com
  • 基金资助:
    陕西省自然科学基础研究计划项目任务(2016JM149)

The establishment of prognostic risk model of hepatoblastoma in children and the comparison with PRETEXT staging system

JIA Yi, YING Hai-yan, LI Shao-fang   

  1. Department of Neonatology, 3201 hospital, Shanxi 723000, China
  • Received:2020-06-05 Online:2020-12-31 Published:2021-02-26
  • Contact: JIA Yi,Email:jen4722@163.com

摘要: 目的 探讨肝母细胞瘤患儿预后风险模型的建立及其与PRETEXT分期系统的比较。方法 2015年2月至2017年2月我院收治的肝母细胞瘤患儿114例,回顾性分析患儿临床资料,记录患儿3年总生存率。采用Cox风险模型分析患儿3年总生存率的影响因素,并建立预后风险模型。采用受试者操作特征曲线(ROC)分析预后风险模型与PRETEXT分期系统,判断3年生存率的价值。结果 114例患儿3年存活72例,死亡42例,3年总生存率为63.2%。Cox回归模型分析显示AFP半衰期(95%CI=1.262~5.291,HR=2.584)、ApoA-I(95%CI=0.327~0.833,HR=0.522)、PRETEXT分期(95%CI=1.146~3.596,HR=2.030)、病理分型(95%CI=1.213~2.803,HR=1.844)及远处转移(95%CI=2.074~5.096,HR=3.251)是影响肝母细胞瘤患儿3年生存率的独立影响因素(P<0.05)。根据Cox多因素回归分析结果获得预后风险模型:风险系数(R)=0.949(AFP半衰期)-0.650(ApoA-I)+0.708(PRETEXT分期)+0.612(病理分型)+1.179(远处转移)。受试者工作曲线分析结果显示R与PRETEXT分期对判断3年生存率均具有较高价值(P<0.05),R判断3年生存率的曲线下面积(AUC)为0.806,高于PRETEXT分期的0.663。结论 基于Cox多因素回归模型分析建立的预后风险模型与PRETEXT分期均有助于判断肝母细胞瘤患儿3年生存率,而预后风险模型准确性更高。

关键词: 肝母细胞瘤, 预后风险模型, PRETEXT分期系统, 预测价值

Abstract: Objective To study the establishment of prognostic risk model of hepatoblastoma in children and the comparison with PRETEXT staging system.Methods One hundred and fourteen Shanxi children with hepatoblastoma between February 2015 and February 2017 were admitted to our hospital,the clinical data of children were analyzed retrospectively. The 3-year overall survival rate was recorded.The Cox risk model was used to analyze the influencing factors of 3-year overall survival rate of children, and the prognosis prediction model was established.The value of prognostic risk model and PRETEXT staging system to judge the 3-year survival rate were analyzed by using receiver operating characteristic curve (ROC) model.Results There were 72 survived in 3 years and 42 died among the 114 children,the 3-year overall survival rate was 63.2%.The Cox regression model analysis showed that the AFP half-life (95%CI=1.262~5.291,HR=2.584),ApoA-I (95%CI=0.327~0.833,HR=0.522), PRETEXT stage (95%CI=1.146~3.596, HR=2.030), pathological classification (95% CI=1.213~2.803,HR=1.844) and distant metastasis (95%CI=2.074~5.096, HR=3.251) were independent factors which affected the 3-year survival rate of children with hepatoblastoma (P<0.05).The risk model of prognosis was obtained,which accorded to the results of Cox multivariate regression analysis.Risk factor (R)=0.949(half-life of AFP)-0.650(ApoA-I)+0.708(PRETEXT stage)+0.612(pathological classification) +1.179(metastasis).The results of ROC analysis showed that the R and PRETEXT stages were of high value in judging the 3-year survival rate (P<0.05).The area under curve (AUC) of R to judge 3-year survival rate was 0.806,which was higher than 0.663 of PRETEXT stage.Conclusion The prognosis risk model based on Cox multiple factor regression and the PRETEXT stage can help to judge the 3-year survival rate of children with hepatoblastoma, and the accuracy of prognosis risk model is higher.

Key words: Hepatoblastoma, Prognostic risk model, PRETEXT staging system, Predictive value