| Hepatocellular carcinoma(HCC)is a highly malignant tumor with poor outcome worldwide.Early diagnosis is a crucial determinant for HCC prognosis.Detection of tumor biomarkers has played an increasingly important role in the early diagnosis and prognosis evaluation of liver cancer in recent years.Serum biomarkers,as the most direct,rapid and effective diagnostic method,are the main means of early tumor detection,early treatment and therapeutic effect monitoring.At present,alpha-fetoprotein(AFP)is the main HCC serum biomarker used in clinical.However,its diagnostic sensitivity and specificity are not optimal given that approximately 40% of early HCC patients are AFP negative.especially in the early stage of HCC.Therefore,the research and discovery of new serum biomarkers for liver cancer has become an important research focus.But it has never been easy to accurately detect these differentially expressed proteins in serum.In recent years,comparative proteomic methods are expected to identify serum protein biomarkers for early diagnosis and prognosis of HCC.Some potential serum protein markers for HCC have been identified based on serum proteomics studies.Nevertheless,many of these candidate biomarkers display limited sensitivity and specificity when used alone,especially for early stages of HCC.Furthermore,most studies are yet in the early phase of biomarker discovery,or performed with small sample size,and few of them are actually used in clinical practice.Therefore,in the present study,we performed comparative proteome analysis,i TRAQ labeling with LC-MALDI-MS/MS technology,to explore differentially expressed proteins(DEPs)in patients with HCC,chronic hepatitis B,liver fibrosis and healthy controls.With the help of the visual bioinformatics analysis platform,the candidate protein markers were further screened,and the selected candidate markers were verified by multi-level and large-sample experiments.Based on the clinical data of local HCC patients in Guangxi,the diagnostic model of liver cancer was established,and the application value of candidate protein marker RBP4 in the diagnosis of liver cancer was discussed and evaluated,so as to provide a useful theoretical basis for the prevention and control of liver cancer.Part One: Screening of potential serum protein markers for liver cancer,bioinformatics analysis and validation of candidate marker RBP4Objective: It is beneficial to combine i TRAQ marker with MALDI-MS /MS technology to study serum protein differential expression of liver cancer on the whole,screen serum biomarkers related to liver cancer,and verify candidate serum biomarkers at multiple levels.Methods:(1)The previously optimized i TRAQ-LC-MALDI-MS/MS technique of our research group was used to analyze and compare the differentially expressed proteins in serum of HCC,CHB,LF and health control groups,and the common differentially expressed proteins were obtained;(2)Through systematic bioinformatics analysis,the differentially expressed proteins were verified to further screen potential serum markers for liver cancer;(3)RBP4 levels in HCC cells,tissues and serum were verified by RT-PCR,immunohistochemistry,MRM and ELISA.Results :(1)The serum of HCC and healthy control,CHB and LF were detected by mass spectrometry,and 27,20 and 22 differential proteins were detected between HCC and healthy control group,CHB and LF population,respectively.Among them,there were 14 differential proteins in the intersection of HC,CHB and HCC,and 10 differential proteins in the intersection of HC,CHB,LF and HCC,among which SERPINA1,A2 M,Apo A4 and RBP4 were the four most significant differences;(2)The results of bioinformatics analysis showed that RBP4 expression was low in HCC,which was consistent with the results of mass spectrometry analysis,and the expression level was related to the progression of the disease.The poor prognosis of HCC patients with low RBP4 expression was negatively correlated,and the prognosis of HCC patients with low RBP4 expression was poor.Bioinformatics analysis results of the other three differentially expressed proteins were inconsistent with the results of previous mass spectrometry analysis;(3)RT-PCR results showed that the relative expression level of RBP4 gene in SMMC-7721 and Hep G2 cells was 0.344±0.024 times higher than that in HL-7702 normal liver cells,respectively and0.318±0.179 times,the expression of RBP4 was down-regulated in liver cancer cell lines.Immunohistochemical detection of RBP4 expression in hepatocellular carcinoma and paracancerous tissues of 80 patients showed that the distribution of RBP4 protein expression levels in HCC group and adjacent group was significantly different(P<0.0001).The expression level of RBP4 in liver cancer tissues was significantly lower than that in adjacent tissues.(5)A total of 204 serum samples from HCC(46 cases),LC(50 cases),CHB(54 cases)and healthy controls(54 cases)were detected by ELISA.The results showed that the serum level of RBP4 protein in HCC group(9.84±6.70 g/m L)and LC group(8.09±3.89g/m L)was significantly lower than that in HC group(32.62±13.58 g/m L)and CHB group(17.95±8.69 g/m L).There was no difference in serum RBP4 levels between the HCC group and the LC group(P>0.05).(6)Results of serum MRM showed that the expression of RBP4 protein in HCC group(24 cases)were significantly different from that in HC group(13 cases)and CHB group(10 cases)(P<0.0001),serum RBP4 protein expression level of HCC group was 0.45±0.17 folds of HC group(P<0.0001),there was no difference between AFP negative HCC group and AFP positive HCC group(P>0.05).The level of serum RBP4 in LC group(10 cases)was lower than that in HC group and CHB group(P<0.0001),but no difference compared with HCC group(P>0.05).Conclusions: Four significantly differentially expressed proteins were screened by i TRAQ marker combined with LC-MALDI-MS/MS technology,and bioinformatics analysis and experimental verification indicated that RBP4 may be a diagnostic marker for liver cancer.Further studies are needed on the exact mechanism of RBP4 in tumor genesis and its clinical application value as a marker for liver cancer.Part two: The correlation analysis and diagnostic value of RBP4 and liver cancer in GuangxiObjective: Combined with the clinical data of primary liver cancer patients in Guangxi,to screen the influential factors for the diagnosis of liver cancer,to explore the significance of RBP4 in the occurrence and development of liver cancer,and to evaluate the application value of RBP4 in the diagnosis of liver cancer.In addition,patients with HCC were followed up to obtain their survival data and relevant clinical information,and univariate survival analysis and multivariate Cox regression analysis were performed for survival time to establish the prognostic model of serum combination of RBP4 and other clinical indicators,and to evaluate its application value as a biomarker for prognosis of HCC.Methods:(1)Serum and related clinical information of liver cancer,hepatitis and healthy controls from Guangxi were collected.Serum RBP4 levels in the three groups were detected by ELISA,and the relationship between serum RBP4 levels and relevant clinical information was analyzed.Serum RBP4 levels in different liver cancer subgroups were compared.(2)Univariate Logistic regression analysis was performed on the related factors affecting the diagnosis of HCC to screen the influencing factors for the diagnosis of HCC;(3)By drawing ROC curve,the diagnostic value of RBP4 and AFP as markers of liver cancer was analyzed and compared;(4)Life table method,Kaplan-Meier method,Cox proportional risk model regression method and other methods were used to analyze the survival time of follow-up patients with HCC,screen the influencing factors of prognosis of HCC,and analyze the correlation between prognosis of HCC patients and serum RBP4 expression level.(5)The combined prognosis diagnosis model of liver cancer was established,and the ROC curve was drawn to evaluate the diagnostic value of the combined model.Results:(1)Serum RBP4 levels of 761 samples were detected by ELISA,including 290 cases of HCCP,202 cases of CHB and 269 cases of HCB.Serum RBP4 level in HCC patients(10.64±6.56μg/m L)was significantly lower than that in CHB patients(20.71±11.95μg/m L,P<0.0001)and HC group(32.61±17.32μg/m L,P<0.0001),the levels of RBP4 in AFP negative and AFP positive HCC groups were 10.8±7.39μg/m L and 10.47±5.15μg/m L,respectively,and there was no significant difference between the two groups(P<0.05);(2)The clinical characteristics of 202 CHB and 290 HCC patients were analyzed by unifactor Logistic regression analysis.RBP4,TP,GLO and AST were included in the regression equation,but AFP,ALB and AST/ALT were not included in the regression equation;(3)With HC as control,the AUC of ROC analysis of RBP4 for the diagnosis of HCC(cut-off=14.06μg/m L)was 0.931(95%CI: 0.911～0.950).The AUC of ROC analysis for the diagnosis of AFP negative HCC was 0.923(95%CI: 0.897～0.948)and 0.954(95%CI: 0.939～0.970)for the combination of AFP and RBP4.With HC and CHB as control,the AUC of RBP4 for the diagnosis of HCC was 0.879(95%CI: 0.854～0.903),the AUC of AFP negative HCC was0.875(95%CI: 0.843～0.906),and the AUC of AFP combined with RBP4 was0.919(95%CI: 0.898～0.940).The sensitivity and specificity of combined diagnosis were 87.9% and 80.0%,respectively.When CHB was used as control,the AUC of RBP4 alone in the diagnosis of HCC was 0.809(95%CI: 0.771～0.846),that of AFP alone(cut-off=25ng/m L)was 0.689(95%CI: 0.643～0.735),and that of AFP combined with RBP4 was 0.872(95%CI: 0.840～0.903).The AUC of RBP4 was significantly higher than that of AFP(Z=-4.901,P<0.0001),the AUC of AFP combined with RBP4 was higher than that of RBP4 alone(Z=-5.01,P<0.0001);(4)The sensitivity and specificity of RBP4 were 80.3% and 80.9%,respectively,which were much higher than that of AFP(40.7%)in HCC screening in CHB and HC control groups.Youden index of RBP4 was 61.2%,which was also higher than that of AFP(39.4%).The sensitivity and specificity of RBP4 and AFP were 87.9% and 80.0%,respectively.When used for screening high-risk population(CHB as control),the sensitivity of RBP4 was 80.3%,much higher than that of AFP(40.8%),and the specificity of RBP4 was 68.3%,lower than that of AFP,but the Youden index of RBP4 in the diagnosis of HCC was 48.6,higher than that of AFP(37.9%).The sensitivity and specificity of combined diagnosis of RBP4 and AFP were 81.0% and 82.7%,respectively,and Youden index was63.6%,which were higher than those of AFP and RBP4 alone.(5)Survival analysis of 202 HCC patients with follow-up showed that the one-year cumulative survival rates of HCC patients with low and high RBP4 expression were 78% and87%,the three-year cumulative survival rates were 50% and 75%,and the fiveyear cumulative survival rates were 43% and 70%,respectively.The survival rate of patients with low RBP4 expression was significantly lower than those with high RBP4 expression.(6)Unified Logistic regression analysis of HCC patients with follow-up showed that AFP,AST/ALT and metastasis were independent risk factors,while normal A/G and RBP4 level were protective factors for the prognosis of patients with HCC.The Exp(B)for RBP4 was 0.191(95% confidence interval 0.063～0.586,P=0.004).The Exp(B)of AFP was 29.418(95% confidence interval 10.894～79.442,P<0.0001).Gender,age,HBV or HCV infection,TP,ADA,CEA,CA724,CA19-9,CA153,CA125 and other indicators were not included in the regression equation,and had no correlation with the overall survival of HCC patients(P<0.05);(7)Cox multivariate regression analysis of HCC patients with follow-up showed that serum RBP4 level,AFP level,AST/ALT ratio,metastasis,and treatment were the main factors affecting the prognosis of patients,among which RBP4 level was negatively correlated with overall survival time,and the relative risk of RBP4 was 0.424(95%CI:0.245～0.725,P=0.002).Builting joint diagnosis model(with RBP4,AFP,AST/ALT,ALB/GLO metastasis,recurrence,treatment)and drawing ROC curve to assess its predictive value.The results showed that the AUC of the joint diagnosis model was 0.926(95%CI: 0.888～0.964),sensitivity and speciality rate were 89.2% and 87.4% respectively,and the AUC of the diagnosis model was higher(Z=5.288,P<0.0001)than that of AFP alone(AUC=0.809).Conclusions:(1)A multi-level case-control study of clinical HCC patients,hepatitis and healthy controls in Guangxi region showed that RBP4 may be a good biomarker for HCC effect.The combination of RBP4 and AFP may improve the detection efficiency of HCC in high-risk groups of HCC;(2)RPB4 can be a promising serum biomarker for hepatocellular carcinoma and a reliable complementary diagnostic biomarker for AFP negative HCC;(3)Low expression of RBP4 can be used as a prognostic indicator of HCC,but further studies are needed to confirm its specific function in the molecular pathway and oncogenic process of HCC. |
PDF Full Text Request