基于T淋巴细胞亚群构建原发性肝癌患者氩氦刀冷冻消融术后残癌风险预测模型  

T-lymphocyte-subsets-based construction of a residual cancer risk prediction model for primary liver cancer patients after argon-helium cryoablation

作　　者：黄晓路 罗剑 何远春 HUANG Xiaolu;LUO Jian;HE Yuanchun(Department of Oncology,Lu'an Hospital Affiliated to Anhui Medical University,Lu'an 237000,China)

机构地区：[1]安徽医科大学附属六安医院肿瘤内科,安徽六安237000

出　　处：《中国肿瘤外科杂志》2024年第2期148-152,共5页Chinese Journal of Surgical Oncology

摘　　要：目的探讨基于T淋巴细胞亚群构建原发性肝癌患者氩氦刀冷冻消融术后残癌风险的预测模型。方法回顾性分析2018年8月至2023年8月安徽医科大学附属六安医院医院收治的80例行氩氦刀冷冻消融术治疗的原发性肝癌患者的临床资料,根据术后3个月CT或MRI检查是否出现残留癌组织,将患者分为残癌组(35例)和非残癌组(45例),收集患者的临床资料、手术情况及T淋巴细胞亚群(CD3+、CD4+、CD8+、CD4+/CD8+)水平,经Logistic回归模型分析原发性肝癌患者术后发生残癌风险的影响因素,并建立Logistic回归方程及术后残癌风险预测模型,并利用Calibration曲线评价模型的校准能力;经受试者工作特征曲线(ROC)及曲线下面积(AUC)验证模型预测效能。结果残癌组CD3+、CD4+、CD4+/CD8+水平低于非残癌组;肿瘤数目、肿瘤直径大小、CD8+水平高于非残癌组(P<0.05);经Logistic回归分析结果显示,肿瘤数目、肿瘤直径大小、CD3+、CD4+、CD8+、CD4+/CD8+均是原发性肝癌患者术后发生残癌风险的危险因素(P<0.05);Calibration曲线、Hosmer-Lemeshow检验拟合优度检验结果显示χ^(2)=1.538,df=7,P=0.964;ROC曲线预测原发性肝癌患者术后发生残癌的AUC为0.873(95%CI:0.815~0.942),灵敏度为90.48%,特异度为75.18%;Bootstrap内部验证法验证结果显示,C-index为0.824。结论肿瘤数目、肿瘤直径大小、T淋巴细胞亚群是原发性肝癌患者术后发生残癌风险的危险因素,且构建的列线图预测模型具有较高区分度、精准度,临床可以通过检测T淋巴细胞亚群水平判断患者术后是否存在残留病灶。Objective To explore the prediction model of residual cancer risk in patients with primary liver cancer after argon-helium cryoablation based on T lymphocyte subsets.Methods The clinical data of 80 patients with primary liver cancer treated by argon-helium cryoablation in Lu'an Hospital Affiliated to Anhui Medical University from August 2018 to August 2023 were retrospectively analyzed.According to whether there were residual cancer tissues in MRI or CT examination three months after operation,the patients were divided into a residual cancer group(35 cases)and a non-residual cancer group(45 cases),and the clinical data,operation conditions and T lymphocyte subsets(CD3+,CD4+,CD8+)were collected.Logistic regression model was used to analyze the influencing factors of postoperative residual cancer risk in patients with primary liver cancer,and Logistic regression equation and postoperative residual cancer risk prediction model were established,and Calibration curve was used to evaluate the calibration ability of the model.The performance characteristic curve(ROC)and the area under the curve(AUC)of the subjects verify the prediction efficiency of the model.Results The levels of CD3+,CD4+and CD4+/CD8+in residual cancer group were lower than those in non-residual cancer group.The number of tumors,tumor diameters,and CD8+level was higher than in the residual cancer group those in the non-residual cancer group(P<0.05).Logistic regression analysis showed that the number of tumors,tumor diameter,CD3+,CD4+,CD8+and CD4+/CD8+were all risk factors for postoperative residual cancer in patients with primary liver cancer(P<0.05).Calibration curve and Hosmer-Lemeshow test showed thatχ^(2)=1.538,df=7,P=0.964.ROC curve prediction was 0.873(95%CI:0.815-0.942),the sensitivity was 90.48%,and the specificity was 75.18%.Bootstrap internal verification results show that the C-index was 0.824.Conclusions The number of tumors,tumor diameter,and T-lymphocyte subsets are the risk factors of residual cancer in patients with primary liver canc

关 键 词：T淋巴细胞亚群 原发性肝癌 氩氦刀冷冻消融术 影响因素 风险预测 

分 类 号：R735.7[医药卫生—肿瘤]