CAJ | 학술논문

目的：探讨血清细胞外蛋白激酶 A（ECPKA）在肝细胞癌（HCC）诊断中的价值。方法采用病例对照研究方法，收集研究对象的环境因素、疾病史及健康状况信息；检测血清中 ECPKA 和 AFP 活性。利用 Logistic 回归模型分析血清 ECPKA 与 HCC 的关系，对 ECPKA 进行 Box-Cox 正态转换后，计算 ROC 曲线下面积及其95％可信区间。将ECPKA 和 AFP 转化为有序数据后，拟合 Logistic 回归模型，计算 ECPKA、AFP 及二者同时存在时的 ROC 曲线下面积及其95％可信区间，并进行两两比较。根据 Youden 指数最大原则，并结合 Logistic 回归分析结果及 ROC 曲线，确定最佳临界值。结果 HCC 患者血清 ECPKA 表达水平为（15．579±7．441）U／mL，健康对照为（5．277±3．190）U／mL，两组间差异有统计学意义（P ＜0．05）。单因素和多因素 Logistic 回归结果显示，ECPKA 粗 OR 值及调整年龄和性别后 OR 值分别为1．48（1．37，1．59）和1．49（1．38，1．61）。ROC 曲线下面积为0．908（0．870，0．946）。将 ECPKA 和 AFP 转换为有序数据后，ECPKA＋AFP、ECPKA 和 AFP ROC 曲线下面积分别为0．931（0．897，0．967）、0．895（0．858，0．932）和0．789（0．721，0．838）。ECPKA 和 ECPKA＋AFP 的 ROC 曲线下面积均高于 AFP，u 值分别为3．38和4．70（均 P ＜0．01）；ECPKA 与ECPKA＋AFP 之间的差异无统计学意义（u＝1．41，P ＝0．079）。根据约登指数最大原则，结合 Logistic 回归结果及 ROC曲线，选择10 U／mL 为 ECPKA 的最佳临界值，此时，灵敏度为75．25％，特异度为91．87％，阳性预测值为69．72％，阴性预测值为93．72％。结论 HCC 患者血清中 ECPKA 表达显著上调，ECPKA 可能是一个新的、有价值的 HCC 诊断的替代指标。
목적：탐토혈청세포외단백격매 A（ECPKA）재간세포암（HCC）진단중적개치。방법채용병례대조연구방법，수집연구대상적배경인소、질병사급건강상황신식；검측혈청중 ECPKA 화 AFP 활성。이용 Logistic 회귀모형분석혈청 ECPKA 여 HCC 적관계，대 ECPKA 진행 Box-Cox 정태전환후，계산 ROC 곡선하면적급기95％가신구간。장ECPKA 화 AFP 전화위유서수거후，의합 Logistic 회귀모형，계산 ECPKA、AFP 급이자동시존재시적 ROC 곡선하면적급기95％가신구간，병진행량량비교。근거 Youden 지수최대원칙，병결합 Logistic 회귀분석결과급 ROC 곡선，학정최가림계치。결과 HCC 환자혈청 ECPKA 표체수평위（15．579±7．441）U／mL，건강대조위（5．277±3．190）U／mL，량조간차이유통계학의의（P ＜0．05）。단인소화다인소 Logistic 회귀결과현시，ECPKA 조 OR 치급조정년령화성별후 OR 치분별위1．48（1．37，1．59）화1．49（1．38，1．61）。ROC 곡선하면적위0．908（0．870，0．946）。장 ECPKA 화 AFP 전환위유서수거후，ECPKA＋AFP、ECPKA 화 AFP ROC 곡선하면적분별위0．931（0．897，0．967）、0．895（0．858，0．932）화0．789（0．721，0．838）。ECPKA 화 ECPKA＋AFP 적 ROC 곡선하면적균고우 AFP，u 치분별위3．38화4．70（균 P ＜0．01）；ECPKA 여ECPKA＋AFP 지간적차이무통계학의의（u＝1．41，P ＝0．079）。근거약등지수최대원칙，결합 Logistic 회귀결과급 ROC곡선，선택10 U／mL 위 ECPKA 적최가림계치，차시，령민도위75．25％，특이도위91．87％，양성예측치위69．72％，음성예측치위93．72％。결론 HCC 환자혈청중 ECPKA 표체현저상조，ECPKA 가능시일개신적、유개치적 HCC 진단적체대지표。
Objective To investigate the diagnostic value of sera extracellular cAMP-dependent protein kinase (ECPKA)level in hepatocellular carcinoma (HCC)patients.Methods Case-control study design were performed to estimate the relationship between ECPKA level and HCC with logistic regression model.After Box-Cox transformation, receiver operating characteristic (ROC)curves were drawn.To compare areas under the ROC curve (AUCs)between AFP and ECPKA,the data of AFP and ECPKA were transformed into order data.AUCs and their 95%CI of AFP,ECPKA and AFP+ ECPKA were calculated according to logistic regression model.The best cut-off values were obtained based on Youden's index,the results of logistic regression analysis and the ROC curves.Results The ECPKA levels were (15.579± 7.441)U/mL in the HCC patients,and (5.277 ± 3.190 )U/mL in the control population,which showed significant difference between the two groups (P <0.05 ).The results of univariate logistic regression analysis and multiple logistic regression analysis showed that the crude odds ratio (OR)and the adjusted OR were 1 .48(1 .37,1 .59)and 1 .49(1 .38, 1 .61),respectively.The AUC for ECPKA was 0.908 (0.870,0.946).Compared with 0.5,the difference was statistically significant (P <0.001).The AUCs for ECPKA+AFP,ECPKA and AFP were 0.931 (0.897,0.967),0.895 (0.858, 0.932)and 0.789(0.721 ,0.838),respectively.The AUC for ECPKA (u=3.38,P <0.001 ),as well as for ECPKA+AFP (u=4.70,P <0.001),was higher than that for AFP.Statistical difference had not been found between the AUC for ECPKA and that for ECPKA+AFP (u =1 .41 ,P =0.079).Best cut-off value was 10 U/mL based on Youden's index,results of logistic regression analysis and the ROC curves.The sensitivity (Se),specificity (Sp),positive predictive value (PPV)and negative predictive value (NPV)were 75.25%,91 .87%,69.72% and 93.72%,respectively.Conclusion The expression of ECPKA in sera of HCC patients shows significant up-regulation.ECPKA might be a novel,promising biomarker for diagnosis of HCC.