Validation And Refining Of Prognosis Predict Signatures In Hepatocellular Carcinoma Patients After Curative Resection And Clinical Transformation

Hepatocellular carcinoma (HCC) is one of the most common and aggressive human malignancies worldwide and also the second most common cause of cancer-related death, in China. It has an extremely poor prognosis, mainly attributed to the high frequency of intrahepatic metastasis, the recurrence rate still is up to60-70%in5years.It is a challenge to identify patients who are at a greater risk for tumor recurrence after curative treatment for HCC. Base on traditional clinical characteristic and staging system, the patients were hard to predict very accurately. Recently, identification of molecular markers could provide supplemental and useful information for predicting clinical outcome in patients with a given stage of disease and improve the selection of patients for personal therapies after resection.The metastasis of HCC also depends on multiple interactions of metastatic cells with favorable host microenviroment. Investigating the tumor itself as well as the favorable tumor microenviroment may be useful for understanding the mechanisms of HCC metastasis and prognosis prediction. Although many molecular markers have identified and shown potential prognostic values for HCC patients, so far, no any biomarker has been used to clinical routine practice.In our previous studies, by comparing the gene expression profiles using c DNA microarray, we established153genes’ signature from tumor tissue and17immune cytokines’ signature from noncancerous tissue, which could successfully predict the metastasis of HCC after curative resection. In this study, we focused on the two factors to further validate and refine the signatures which can predict prognosis of HCC after curative treatment, and carried on the clinical transforming research for the prediction kit. PARTIValidation and refining of prognosis predict signatures in hepatocellular carcinoma patients after curative resection and clinical transformation Part APurpose:A153genes expression signature was recently found to predict HCC metastases by our previous study. We examined whether a more refined5genes expression profile from liver cancer tissue could predict recurrence of HCC after curative hepatectomy.Methods:We analyzed a26gene expression profile in liver cancer tissue from361patients of HCC using quantitative RT-PCR. Using methods of bioinformatics, we built a predict model, and refined it to explore optimal factor. Identification of a5immune-gene signature derived from the training cohort and validated in an independent cohort. We validated the important genes on protein levels with liver cancer cell lines.Results:Log-rank test analyzed the potency on recurrence SPP2(P<0.01), ENO2(P<0.01), ASPH(P<0.01), FKBP10(P<0.01) and SAFB2(P<0.01) were ranked in the first five sites. Using bioinformatics methods, a5immune-gene signature could predict the recurrence accurately (P<0.001). Take the gene expression in the median number of expression for critical value, validation group by threshold value divided into high-risk groups and low risk groups, Kaplan-Meier analysis showed that the high risk group time to recurrence significantly short in low risk group (P<0.001), high risk group overall survival period is short in low risk group (P<0.001).Conclusions:The5factors the expression profile from liver tumor tissue predicted recurrence of hepatocellular carcinoma has reached high accuracy by the large samples validation.Part BPurpose:A17genes expression signature in non-cancerous hepatic tissue from patients with HCC was recently found to predict HCC recurrence. We examined whether a more refined5genes expression profile could predict prognosis of HCC after curative hepatectomy.Methods:Using quantitative PCR, the expression of10immune genes in resected tumour tissues from180patients was analyzed. The bioinformatics prediction method was used to derive and test a prognostic signature from this training cohort. The signature was then validated in an independent cohort of200patients, validating protein levels with immunohistochemistry.Results:Log-rank test analyzed the potency on prognosis IL-6, CSF-1, SPP1, TNF and HLADRA, were ranked in the first five sites. Using bioinformatics methods, a5immune-gene signature could predict the prognosis accurately (P<0.001). Take the gene expression in the median number of expression for critical value, validation group by threshold value divided into high-risk groups and low risk groups, Kaplan-Meier analysis showed that the high risk group time to recurrence significantly short in low risk group (P<0.001), high risk group overall survival period is short in low risk group (P<0.001).Conclusions:The "immune microenvironment signature" is reproducible and powerful strategy for prospective patient prediction.Part CPurpose:Integration and verification the prognosis prediction signatures from tumor and noncancerous tissue in prospective study, to promote the research of the clinical transformation.Methods:By analyzing the gene expression profiles of cancerous and noncancerous hepatic tissue from patients with hepatocellular carcinoma, we validated the refined signatures in prospective study. Using the Concordance Index (C-Index) andKaplan-Meier method, to compare and integrate the two signatures, to manufacture the RT-PCR array prediction kit for postoperative recurrence and metastasis of HCC.Results:Comparing with low-risk group (control group), the recurrence risk of high-risk group (exposed group) was significantly increased, and survival time was significantly decreased (p<0.001). The refined predict signatures were further validated in prospective study. Integration of Clinical stage and molecular model will further enhance the accuracy of the prediction of HCC prognosis. It has been applied for the patent RT-PCR array prediction kit for postoperative prognosis of hepatocellular carcinoma.Conclusions:The refined signatures of cancerous and noncancerous hepatic tissue are reproducible and powerful strategy in prospective study. PART IIPostoperative Serum Osteopontin Level is a Novel Monitor forTreatment Response and Tumor Recurrence after Resection of Hepatitis B-related Hepatocellular CarcinomaPurpose:Presurgery serum osteopontin (OPN) level has been demonstrated to correlate to tumor recurrence and survival of HCC patients. This study was to investigate the dynamic changes of serum OPN level after operation and its clinical significance in HCC patients.Methods:Presurgery serum OPN levels were measured by enzyme-linked immunosorbent assay in cohort A of179patients with HCC and were compared with multiple controls including different kinds of liver diseases and healthy individuals. And in another cohort B of110patients with resectable HCC, besides preoperative assays, serum OPN was monitored at1week,1month and≥2months after operation.Results:The baseline presurgery serum OPN (164.3±18.6ng/mL) was significantly higher than that of patients with chronic liver diseases (CLD)(p<0.0001) and with metastatic cancer of liver (p<0.0001) and with benign liver lesions (p<0.0001). The postsurgery OPN were significantly elevated within1week (214.4±14.1ng/mL) after operations, then decreased back to47.6±5.4ng/mL within1month, and reached the nadir (40.2±5.9ng/mL) later than2months after operations. OPN levels increased again at the time of tumor recurrence; and declined after the second removal of recurrent HCCs. Moreover, postoperative OPN in AFP-negative and-positive HCC patients had same changing pattern. Postsurgery OPN was only significantly associated with liver function and C-reactive protein levels.Conclusions:There is a transient fluctuation in serum OPN which might be related to acute inflammatory response after operation. Serum OPN levels significantly decrease after resection of HCCs. It could serve as a surrogate serologic biomarker for monitoring treatment response and tumor recurrence after HCC resection, including those AFP-negative ones. Novelties1. For the first time, we refined and integrated the prognosis predict signatures in hepatocellular carcinoma patients after curative resection from tumor and nontumor tissue.2. We invented and assembled RT-PCR array prediction kit for postoperative recurrence and metastasis of hepatocellular carcinoma.3. We firstly confirmed that Serum OPN levels could serve as a surrogate serologic biomarker for monitoring treatment response and tumor recurrence after HCC resection, including those AFP-negative ones.Potential application of this studyWe validated the more refined five genes expression signatures from tumor tissue and cytokine gene expression profile of noncancerous liver tissue could predict prognosis of HCC after curative hepatectomy. it promoted "translation" of basic research into clinical application. It has been applied for the patent (" RT-PCR array prediction kit for postoperative recurrence and metastasis of hepatocellular carcinoma"; Chinese patent applying number is201110091076.6) and (" RT-PCR array prediction kit for postoperative survival of hepatocellular carcinoma"; Chinese patent applying number is201110091077.0). In addition, this study supported that serum OPN could be used to monitor treatment response and tumor recurrence after HCC resection.