微信公众号

官网二维码
中国癌症防治杂志

中国癌症防治杂志 ›› 2020, Vol. 12 ›› Issue (1): 63-69.doi: 10.3969/j.issn.1674-5671.2020.01.11

• 临床研究 • 上一篇    下一篇

原发性肝癌射频消融术后复发预测模型的建立及评价

  

  1. 广西医科大学附属肿瘤医院肝胆外科；广西肝癌诊疗工程术研究中心；区域性高发肿瘤早期防治研究教育部重点实验室
  • 出版日期:2020-02-25 发布日期:2020-03-10
  • 通讯作者: 白涛 E-mail：25859373@qq.com
  • 基金资助:
    国家自然科学基金项目（81860502）；广西科技计划重点研发项目（桂科AB16380242）；广西自然科学基金项目（2017GXNSFBA198234）；区域性高发肿瘤早期防治研究教育部重点实验室（GK201801）

Development and evaluation of a predictive model for primary liver cancer recurrence after radiofrequency ablation

  • Online:2020-02-25 Published:2020-03-10

PDF (PC)

186

赞

摘要： 目的 构建和评价用于预测原发性肝癌（primary liver cancer，PLC）患者射频消融（radiofrequency ablation，RFA） 术后无瘤生存率的列线图模型。 方法 回顾性分析2009年6月至2017年5月于广西医科大学附属肿瘤医院接受射频消融治疗的213例PLC患者的临床资料。ＰＬＣ患者被随机分为训练组（n=133）和验证组（n=80）。采用Cox回归模型分析射频消融术后复发的因素，并建立复发的列线图模型。通过校准曲线评估模型的预测符合度，Kaplan-Meier 曲线评估模型的实用性，一致性指数（C-index）评估模型的准确度。结果 训练组1年、3年、5年无瘤生存率分别为65.25%、40.91%、26.99%，验证组分别为66.29%、48.10%、24.59%，两组生存曲线比较差异无统计学意义（P=0.785）。Cox回归分析结果显示，肿瘤数目（HR=1.921， 95%CI：1.136~3.251）、丙肝抗体阳性（HR=4.545，95%CI：1.700~12.149）、HBV-DNA≥102 IU/mL（HR=1.993，95%CI：1.209~3.284）及血清前白蛋白（HR=0.996，95%CI：0.993~0.999）为无瘤生存率的影响因素。基于肿瘤数目、HBV-DNA和血清前白蛋白等因素建立列线图模型，训练组和验证组的 C-index 分别为 0.649（95%CI：0.588~0.710）、0.641（95%CI：0.556~0.724），校准图形中标准曲线与预测校准曲线贴合良好。采用列线图将患者分为高风险组和低风险组，高风险组无瘤生存率低于低风险组（P<0.05）。结论 基于肿瘤数目、HBV-DNA和血清前白蛋白等因素建立的列线图测模型可预测PLC射频消融术后的无瘤生存率，对患者辅助治疗具有一定指导价值。

关键词: 原发性肝癌, 射频消融, 复发, 列线图

Abstract: Objective To develop and evaluate a nomogram model for predicting the disease-free survival rate of primary liver cancer（PLC） patients after radiofrequency ablation（RFA）. Methods The clinical data of 213 PLC patients receiving RFA in Guangxi Medical University Cancer Hospital from June 2009 to May 2017 were analyzed retrospectively. The PLC patients were randomly divided into a training group（n=133） and a validation group（n=80）. The Cox regression model was used to analyze the factors of recurrence after RFA，and a nomogram model was developed. The model agreement was evaluated by the calibration curve，the applicability of the model was validated by the Kaplan-Meier curve，and the accuracy of the model prediction was evaluated by the C-index. Results The 1-year，3-year，and 5-year disease-free survival rates in the training group were 65.25%，40.91%，and 26.99%，respectively，and those in the validation group were 66.29%，48.10%，and 24.59%，respectively. No significant difference in survival curves was found between the two groups（P=0.785）. The Cox regression analysis showed that the key influential factors of disease-free survival rate included the tumor number（HR=1.921，95%CI：1.136-3.251），hepatitis C antibody positive（HR=4.545，95%CI：1.700-12.149），HBV-DNA≥102 IU/mL（HR=1.993，95%CI：1.209-3.284），and serum prealbumin（HR=0.996，95%CI：0.993-0.999）. Based on the tumor number，HBV-DNA，serum prealbumin，and so on，a nomogram model was developed. The C-indexes of the training group and the validation group were 0.649（95%CI：0.588-0.710） and 0.641（95%CI：0.556-0.724），respectively，and the calibration curve fitted well with the predicted calibration curve in the calibration graph. The patients were divided into the high-risk group and low-risk group by the nomogram，the disease-free survival rate in the high-risk group was lower than that in the low-risk group（P<0.05）. Conclusions The nomogram model based on the tumor number，HBV-DNA，and serum prealbumin can predict well the disease-free survival rate of liver cancer patients after RFA，providing guidance for patient adjuvant treatment.

Key words: Primary liver cancer, Radiofrequency ablation, Recurrence, Nomogram

中图分类号: 

  • R735.7