动态监测血GP73和自噬相关蛋白对HBV相关ACLF短期预后的临床价值

Clinical value of dynamic monitoring of blood GP73 and autophagy-related proteins on short-term prognosis of HBV-related ACLF

目的 动态监测乙型肝炎病毒(HBV)相关慢加急性肝衰竭(ACLF)患者高尔基体蛋白73(GP73)、自促泛素结合衔接子(p62)、自噬相关蛋白(Beclin1)、微管相关蛋白1轻链3-Ⅱ(LC3-Ⅱ)水平变化,并探讨其短期预后价值。方法 选取2016年1月-2019年12月于中国人民解放军联勤保障部队第九一〇医院收治的120例HBV-ACLF患者(研究组)和同期50名健康体检者(对照组)作为研究对象,比较两组患者血GP73、p62、Beclin1和LC3-Ⅱ差异,并实时监测研究组入院不同时间段(入院第0、1和2月)血GP73、p62、Beclin1和LC3-Ⅱ变化,并分析其与HBVDNA病毒载量、终末期肝病模型(MELD)评分的关系;根据入院后3个月的短期预后(临床结局)将研究组分为生存组(n=71)和死亡组(n=49),比较生存组和死亡组不同时间段血GP73、p62、Beclin1和LC3-Ⅱ差异,用Cox回归模型分析其对HBV-ACLF临床结局的影响,同时绘制受试者工作特征(ROC)曲线分析其对临床结局的预测价值。结果 与对照组比较,研究组不同时间段(入院第0、1和2月)外周血GP73、Beclin1和LC3-Ⅱ均升高,p62降低,差异均有统计学意义(F=29.485、12.485、21.371、10.873,P均<0.05)。生存组和死亡组不同时间段外周血GP73、p62、Beclin1、LC3-Ⅱ、HBV-DNA载量和MELD评分比较,差异均有统计学意义(F=37.982、13.256、18.373、30.275、12.784、20.381,P均<0.05)。经Pearson相关分析显示,不同时间段(入院第0、1和2月),HBV-ACLF患者HBV-DNA载量与GP73(r=0.721、0.734、0.715)、Beclin1(r=0.683、0.726、0.717)和LC3-Ⅱ(r=0.634、0.672、0.659)均成正相关,与p62成负相关(r=-0.694、-0.703、-0.711);HBV-ACLF患者MELD评分与GP73(r=0.784、0.752、0.769)、Beclin1(r=0.815、0.794、0.836)和LC3-Ⅱ(r=0.859、0.873、0.848)均成正相关,与p62成负相关(r=-0.749、-0.802、-0.786);以上差异均有统计学意义(P均<0.05)。Cox回归分析得知,不同时间段(入院第0、1和2月)GP73、p62、Beclin1和LC3-Ⅱ均可影响HBV-ACLF临床结局(P均<0.05),其中入院第0月GP73、p62、Beclin1和LC3-Ⅱ对HBV-ACLF临床结局影响最大,HR绝对值分别为2.673、2.132、2.498和2.995。选取入院第0月GP73、p62、Beclin1和LC3-Ⅱ结果进行分析,绘制ROC曲线显示,GP73、p62、Beclin1和LC3-Ⅱ预测HBV-ACLF临床结局的曲线下面积(AUC)分别为0.588、0.531、0.582和0.549,四者联合效果最佳(AUC=0.871)(P<0.05)。结论 不同时间点GP73、p62、Beclin1和LC3-Ⅱ在HBVACLF中均异常表达,与HBV-DNA载量和MELD评分相关,且均可影响HBV-ACLF临床结局,其中初入院时影响最大,可预测HBV-ACLF临床结局,四者联合检测效果最佳。

Objective To dynamically monitor Golgi protein 73(GP73), p62, autophagy-related proteins(Beclin1),microtubule-associated protein 1 light chain 3-Ⅱ(LC3-Ⅱ) in patients with hepatitis B virus(HBV)-related acute-onchronic liver failure(ACLF), and explore its short-term prognostic value. Methods A total of 120 HBV-ACLF patients(study group) and 50 healthy subjects(control group) during the same period were selected From January 2016 to December 2019 in No. 910 Hospital, Joint Logistics Support Force of the Chinese People′s Liberation Army as the research subjects, and the differences in blood GP73, p62, Beclin1 and LC3-Ⅱ between the two groups of patients were compared.Real-time monitoring of blood GP73, p62, Beclin1 and LC3-II changes in the study group at different time periods since admission(on 0, 1 and 2 months of admission), and analyzing their correlation with HBV-DNA viral load and model for end-stage liver disease(MELD) score. The study group was divided into survival group(n=71) and death group(n=49) based on the short-term prognosis(clinical outcome) 3 months after admission. Compare the differences in blood GP73, p62, Beclin1 and LC3-Ⅱ in different time periods between the survival group and the death group. The Cox regression model was used to analyze its impact on the clinical outcomes of HBV-ACLF, and a receiver operating characteristic(ROC) curve was drawn to analyze its predictive value for clinical outcomes. Results Compared with the control group, peripheral blood GP73, Beclin1 and LC3-Ⅱ were all increased and p62 was decreased in the study group at different time periods(on 0, 1 and 2 months of admission), and the differences were statistically significant(F=29.485,12.485,21.371,10.873;all P<0.05). Comparing the peripheral blood GP73, p62, Beclin1, LC3-Ⅱ, HBV-DNA load and MELD score at different time periods between the survival group and the death group, the differences were statistically significant(F=37.982, 13.256, 18.373, 30.275, 12.784, 20.381,all P<0.05). Pearson correlation analysis showed that at different time periods(on 0, 1and 2 months of admission), the HBV-DNA load of HBV-ACLF patients were positively correlated with GP73(r=0.721, 0.734, 0.715), Beclin1(r=0.683, 0.726, 0.717) and LC3-Ⅱ (r=0.634, 0.672, 0.659), and negatively correlated with p62(r=-0.694,-0.703,-0.711). The MELD score of HBV-ACLF patients was positively correlated with GP73(r=0.784, 0.752, 0.769), Beclin1(r=0.815, 0.794, 0.836) and LC3-Ⅱ (r=0.859, 0.873, 0.848), and negatively correlated with p62(r=-0.749,-0.802,-0.786);the above differences were all statistically significant(all P<0.05). Cox regression analysis revealed that GP73, p62, Beclin1 and LC3-II could affect the clinical outcome of HBV-ACLF at different time periods(on 0, 1 and 2 months of admission)(all P<0.05). Among them, GP73, p62, Beclin1 and LC3-II had the greatest impact on the clinical outcome of HBV-ACLF in the 0th month of admission, with absolute HR values of 2.673, 2.132, 2.498 and 2.995, respectively. The results of GP73, p62, Beclin1 and LC3-II in the 0th month of admission were selected for analysis, and the ROC curve was drawn to show that the areas under the curve(AUC) of GP73, p62, Beclin1 and LC3-II in predicting the clinical outcome of HBV-ACLF were 0.588, 0.531, 0.582 and 0.549, respectively;the combination of the four had the best predicting value(AUC=0.871)(P<0.05). Conclusions GP73, p62, Beclin1 and LC3-II were all abnormally expressed in HBV-ACLF at different time points, which were related to HBV-DNA load and MELD score, and could affect the clinical outcome of HBV-ACLF, among which the greatest impact was on initial admission. GP73, p62, Beclin1 and LC3-II could be used to predict the clinical outcome of HBV-ACLF, and the combination of the four had the best effect.