| Hepatocellular carcinoma (HCC) is a kind of malignant tumor with poor prognosis. HCC tumors progress rapidly, and most likely to end up with metastasis. Among all the cancers, HCC is the third leading cause of cancer related death worldwide, and the second one in China. Therefore, it is important to improve the diagnosis and treatment of HCC in early stage. Currently, alpha fetoprotein (AFP) is the most widely used serum biomarkers for HCC diagnosis. However, about40%of the HCC patients are AFP negative, and AFP may also remarkably elevates in hepatitis, urogenital cancer, gastrointestinal cancer and pregnant women. On the other way, imaging exams are difficult to be used as survey methods for early HCC diagnosis in large population, because they usually cannot detect small tumors with diameters less than0.5-1cm. So, the rates of HCC misdiagnosis and missed diagnosis are unsatisfactory using AFP as biomarker solely. Cumulatively, integrating multiple factors for HCC prognosis is an inevitable trend to improve the accuracy of early HCC diagnosis.The development of proteomic technologies makes it possible to screen and discover novel serum biomarkers for cancers. Scientists have already found some differential proteins which are potential cancer biomarkers, and these proteins have close correlations with cancer initiation and progression. Besides, antibody microarray, a new proteomic technology, has been established for cancer serum biomarker discovery. Antibody microarray is a high-throughput method with high sensitivity and specificity, and integrates multiple detections into a mini-system. Multiple antibodies are immobilized on the surface of chips, and then this chip can be subjected to detect levels of various proteins in a sample simultaneously. This novel technology solves the problem of traditional method that in each experiment, researchers can only measure the level of one kind of protein in one time. Some researches of antibody microarray aim at clinical applications, such as antibody microarray of cancer biomarkers. In summary, antibody microarray technology is a valuable method to discover novel serum biomarkers of cancer initiation and progression, which will improve the research of this field and facilitate the development of diagnosis and treatment of early HCC. Part OneSelection of serum protein biomarkers for early diagnosis of HCCDiscovering biomarkers of early hepatocellular carcinoma (HCC) is popular field of proteomic research of liver related diseases. Most of HCC patients have backgrounds of chronic infection of Hepatitis B virus (HBV) and Hepatitis C virus (HCV).In recent years, because of the development and application of genomic, proteomic and metabolic technologies, the research of HCC serum protein markers has made some progresses, such as the discovery of lots of potential serum protein biomarkers for early diagnosis. In the process of tumor progression, stimulation of chronic inflammation leads to release of variant autocrine and paracrine growth factors from tumor cells, which produce an inflammatory microenvironment consisting of tumor cells, host cells, extracellular matrix, secretary factors et al. This microenvironment contains a complicated interaction network, and plays an important role of regulation of tumor progression. After the infection of chronic virus, a regulation network of inflammation is constructed involving IL-6, IL-10, TNF et al. secreted from affect T-cells and Th cells. These cytokines effect regulation of target cells though receptor binding, and result in pathological damage of tissues and organs. Therefore, detection of serum cytokines of HCC patients may reflect the tumor progression and help prognosis to some extent. Some reports show that several chemotactic factors and growth factors, such as IL-8, HGF, TGF-β1, VEGF et al. released from tumor cells and stroma cells play important roles in tumor angiogenesis and infiltration. These factors closely correlate to the initiation, progression, biological behaviors and prognosis of HCC. However, all the serum markers are unspecific tumor-related antigens, so their specificities and sensitivities as solo makers are unsatisfactory. Therefore, we should not only try to discover novel serum biomarkers of early HCC, but also optimize and integrate currently used biomarkers to construct a multi-factor model for early diagnosis of HCC.First, we analyze15potential serum protein biomarkers for early diagnosis of HCC in a cohort of50HCC patients and32liver cirrhosis patients using enzyme-linked immune sorbent assay (ELISA). These15factors are HGF, IGF, IL-6, IL-8, IL-10, TGF-β1, VEGF, HCCR, SCCA, GPC3, GP73, DCP, AFP-L3, AFU and HSP27. Based on serum concentration and differential expression of these factors, together with the functional linkage of inflammatory factors and HCC imitation and progression, we select8serum biomarkers, AFP, HGF, IGF, IL-6, IL-8, IL-10, TGF-β1and VEGF for later antibody microarray establishment.Part TwoEstablishment of antibody microarray detection system using multiple tumor serum protein biomarkersThe traditional detection methods of tumor biomarkers, ELISA, immunohistochemistry (IHC) and radio immune assay (RIA), can only detect single biomarkers in each assay. So, the antibody microarray technology which can analyze the whole protein profile of samples attracts more attention in the post-genomic era. Antibody microarray is a kind of protein array. This high throughput and time-saving method can analyze multiple samples parallely using very small amount of samples, and achieves signals with high ratio of signal to background. Antibody microarray is a new method of detecting protein expression profile of clinical samples, and becomes an important platform of identifying and quantifying protein expressions parallely. It can reflect change of protein expression during the progression of diseases, especially in cancer biomarker researches. Tumor biomarkers can be screened and identified using antibody microarray for early diagnosis and follow-up study of cancer patients. Currently, multiple tumor markers are detected in clinical applications in order to improve the accuracy of early diagnosis of HCC. So, based on the result of part1, purified proteins, monoclonal antibodies and chip related materials of8selected tumor biomarkers were purchased. Double antibody sandwich method was used to optimize the antibody concentration, exclude the unspecific cross-binding, establish the antibody immobilization and signal detection methods. Detection window of each biomarker, limit of detection, accuracy and stability of the whole system were also measured using our antibody microarray. After the optimization of the antibody microarray system, we established a quantification method to detect multiple tumor markers in serum samples, and analyzed the correlation of data from this method and data from traditional ELISA. The results show that data from antibody microarray and data from ELISA had strong correlation to each other in a particular range of sample concentration. Then, the antibody microarray was subjected to preliminary analysis of clinical samples.Part ThreeClinical application of antibody microarray detection system using multiple tumor serum biomarkersSerum samples from160HCC patients [average age:46.5(21-71),128male and32female] and58LC patients [average age:46.7(23-68),46male and12female] were analyzed using our antibody microarray system. Serum levels of AFP and another7factors were measured, and were subjected to retrospective research together with clinical information of these patients. Logistic regression analysis was applied using SPSS to establish a diagnostic model containing8factors in a training set. Diagnostic value of each factor was evaluated using receiver operating characteristic curve (ROC), and areas under the curve (AUC) and cutoff values were calculated. A testing set was applied to validate the diagnostic model. The result showed that when combining all the8factors using logistic regression, the AUC reached0.940(cutoff0.31), while the AUC of AFP as single factor was only0.665(cutoff20ng/ml). The difference of diagnostic capacity of these two models was significant with a P<0.001. In the training set, the sensitivity of the8factor model was0.933, the specificity was0.833and the accuracy was0.909. The sensitivity, specificity and accuracy in testing set were0.890,0.773and0.860, respectively. The sensitivity, specificity and accuracy of AFP model in testing set were only0.700,0.590and0.640, respectively. In conclusion, the8factor model has higher sensitivity, specificity and accuracy than AFP model in early diagnosis of HCC. This multi-factor diagnostic system may have potential value for clinical applications. |
PDF Full Text Request