基于HALP评分的列线图模型对肝细胞癌患者肝切除术后预后的预测价值

Value of the nomogram based on HALP score in predicting the prognosis of patients with hepatocellular carcinoma after hepatectomy

目的 探讨HALP 评分在肝细胞癌(HCC)患者肝切除术后预后评估中的应用价值,分析基于HALP评分的列线图模型能否有效预测患者术后生存情况。方法 回顾性分析2013年7月-2020年3月在西南医科大学附属医院肝胆外科手术治疗的253例HCC患者临床资料。通过绘制ROC曲线,计算出 HALP 和其他有关指标的最佳截断值。采用χ^(2)检验分析HALP与临床病理特征之间的关系。使用Kaplan-Meier方法绘制生存曲线,并采用Log-rank检验进行比较。进行单因素分析和多因素Cox回归模型,分析HALP及其他临床参数与患者预后的关系。通过 R 3.6 软件构建列线图,使用C指数及校准图评价列线图的预测能力,通过净重新分类指数(NRI)、综合判别改善指数(IDI)比较列线图模型和传统模型的预测能力。结果 Kaplan-Meier分析显示高HALP组患者OS、RFS优于低HALP组,差异均有统计学意义(P值均<0.001)。单因素Cox回归分析显示WBC、GGT、ALP、AFP、手术方式、微血管侵犯、TNM分期、分化程度、HALP、AST/ALT、NLR、MLR均与总生存期(OS)显著相关(P值均<0.05),将单因素Cox回归分析中有统计学意义的变量纳入多因素Cox回归分析,结果显示HALP、AST/ALT、ALP、AFP、分化程度、TNM分期是HCC患者术后OS的独立影响因素(P值均<0.05)。单因素Cox回归分析显示GGT、ALP、AFP、微血管侵犯、TNM分期、分化程度、HALP、AST/ALT、NLR、MLR均与无复发生存期(RFS)显著相关(P值均<0.05);多因素Cox回归分析,结果显示HALP、AST/ALT、NLR、ALP、AFP、TNM分期是HCC患者术后RFS的独立影响因素(P值均<0.05)。根据多因素分析结果分别构建HCC患者OS、RFS的列线图,计算预测OS列线图的C指数为0.732(0.691~0.774),预测1、3、5年生存率的AUC分别为0.795、0.791、0.775。预测RFS列线图的C指数为0.677(0.637~0.717),预测1、3、5年生存率的AUC分别为0.742、0.733、0.716;并且1、3、5年OS、RFS校准图表现出了较好的拟合度。结论 术前低水平HALP是接受手术治疗的HCC患者长期预后不佳的预测因素,基于HALP评分的列线图模型优于BCLC分期模型,可以更好地预测HCC的预后情况。

Objective To investigate the value of HALP score in evaluating the prognosis of patients with hepatocellular carcinoma(HCC)after hepatectomy and whether the nomogram based on HALP score could effectively predict the postoperative survival of patients.Methods A retrospective study was performed for the clinical data of 253 HCC patients who underwent surgical treatment in Department of Hepatobiliary Surgery,The Affiliated Hospital of Southwest Medical University,from July 2013 to March 2020.The receiver operating characteristic(ROC)curve was plotted to calculate the optimal cut-off values of HALP score and other related indicators;the chi-square test was used to investigate the association between HALP score and clinicopathological features;the Kaplan-Meier method was used to plot survival curves,and the Log-rank test method was used for comparison.The univariate and multivariate Cox regression analyses were used to investigate the association of HALP score and other clinical parameters with the prognosis of patients.R3.6 was used to establish a nomogram;C-index and calibration curve were used to evaluate the predictive ability of the nomogram,and net reclassification index(NRI)and integrated discrimination improvement(IDI)were used to compare predictive ability between the nomogram model and the conventional model.Results The Kaplan-Meier analysis showed that the high HALP group had significantly better overall survival(OS)and recurrence-free survival(RFS)than the low HALP group(P<0.001).The univariate Cox regression analysis showed that white blood cell count,gamma-glutamyl transpeptidase(GGT),alkaline phosphatase(ALP),alpha-fetoprotein(AFP),surgical approach,microvascular invasion,TNM stage,degree of tumor differentiation,HALP,aspartate aminotransferase(AST)/alanine aminotransferase(ALT)ratio,neutrophil-to-lymphocyte ratio(NLR),and monocyte-to-lymphocyte ratio(MLR)were significantly associated with OS(all P<0.05).The variables with statistical significance in the univariate Cox regression analysis were included in the multivariate Cox regression analysis,and the results showed that ALP,AST/ALT ratio,ALP,AFP,degree of tumor differentiation,and TNM stage were independent influencing factors for OS after surgery in HCC patients(all P<0.05).The univariate Cox regression analysis showed that GGT,ALP,AFP,microvascular invasion,TNM stage,degree of tumor differentiation,HALP,AST/ALT ratio,NLR,and MLR were significantly associated with RFS(all P<0.05),and the multivariate Cox regression analysis showed that HALP,AST/ALT ratio,NLR,ALP,AFP,and TNM stage were independent influencing factors for RFS after surgery in HCC patients(all P<0.05).The nomograms for OS and RFS of HCC patients were established based on the multivariate analysis.The nomogram for OS had a C-index of 0.732(95%confidence interval[CI]:0.691-0.774)and an area under the ROC curve of 0.795,0.791,and 0.775,respectively,in predicting 1-,3-,and 5-year survival rates,and the nomogram for RFS had a C-index of 0.677(95%CI:0.637-0.717)and an area under the ROC curve of 0.742,0.733,and 0.716,respectively,in predicting 1-,3-,and 5-year survival rates.The calibration curves of 1-,3-,and 5-year OS were well fitted to those of 1-,3-,and 5-year RFS.Conclusion A low level of HALP before surgery is a predictive factor for poor long-term prognosis in HCC patients undergoing surgical treatment,and the nomogram model based on HALP score is superior to the BCLC staging model and can better predict the prognosis of HCC.