| Background:Although in the past decade, there have been improvements in surgical and medical treatments, the outcome of patients with hepatocellular carcinoma(HCC) remains unsatisfactory. Most patients with HCC are still diagnosed at intermediate or advanced disease stages, where curative approaches are often not feasible. Even patients in early stage still recurrence after get curative treatment. However, there is no effective treatment for these advanced stage or recurrence HCC patients. Its great important to explore more effective treatment diagrams for a longer life for these patients.Molecular targeted agents(MTAs) for the treatment of cancer including HCC are a very promising approach, demonstrated by the results of the multikinase inhibitor sorafenib. This indicates molecular therapy plays an important role in the treatment of advanced HCC and recurrence HCC. The regulatory approval of sorafenib paved the way for testing a wide range of molecular therapies. Unfortunately, none of these therapies, including sunitinib, brivanib, erlotinib, linifanib, everolimus and ramucirumab, have demonstrated survival benefits in patients with HCC. The reasons for the disappointing clinical trial results partly associated with no molecular subtype conducted for selecting sensitive patients. There are only about 50% patients showing response to sorafenib in clinical trails, however, we cannot select those patients because of lack of biomarkers. It is urgent to explore more MTAs with effect and also the molecular biomarkers to select sensitive patients and predict patient clinical outcome.Patient-derived xenograft models(PDXs) generated directly from fresh human tumors is considered to be the mostly closed to clinical patients. PDX models mostly retain the principal histologic and genetic characteristics of their donor tumor and remain stable across passages. These models have been shown to be predictive of clinical outcomes and are being used for preclinical drug evaluation, biomarker identification, biologic studies, and personalized medicine strategies. However, there are fewer reports to establish and descript PDXs of liver cancer.Sorafenib is currently the standard treatment of hepatocelluar carcinoma, with modest effective. About one half of population showed resistant to sorafenib. The underling mechanism of sorafenib resistance has reported in many studies. However, because of limitation of the preclinical models used in these studies, the mechanism of resistance in HCC to sorafenib is still unknown and sorafenib resistant biomarkers is still lacking. Methods:We established a large PDXs cohort of liver cancer through xenografting fresh tumor tissue into nude mouse. The concordance between tumors and the models derived from them was evaluated by comparing the principal histologic and genetic characteristics. Whole exome sequence, SNP microarray and gene expression microarray were used to descript the genetic information of those models, and compared with HCC patients. To explore biomarkers associated with sorafenib sensitivity, we compared the gene expression date of PDX models with different sensitivity to sorafenib. Results:Two hundred and forty two PDXs of HCC and 71 PDXs of ICC were established with the success xenograft rate 30.3% and 33.8% respectively. The average time for HCC PDXs tumors grown to 1000 mm3 is 90(65,130) days, while 110(80, 150) for ICC PDXs, which is shorter compared with HCC patients recurrence-free survival.PDX tumor retained principle histologic and genetic characteristics and some biomarkers, such as AFP, were concordant with derived primary tumors. For HCC patients, the successful xenograft in mouse is associated with age<50(P=0.045), tumor diameter >5cm(P<0.001), poor differentiation(P<0.001), microvascular invasion(P<0.001), macrovascular invasion(P<0.001), advanced stage(P<0.001), and early recurrence(P<0.001). The 1,3 years disease-free survival(DFS) and overall survival(OS) were poorer in successful engraftment(P<0.001), and successful engraftment ability could serve as an independent prognostic factor for RFS and OS(1.394(1.124-1.729), P=0.002). Similar with HCCs, patients with successful engraftment in ICCs showed shorter RFS and OS(P<0.001).Most of genome variation, which is frequently occurred in HCC tumors, can be detected in PDXs tumors. We detected gene mutation of TP53(67.6%), TTN(93.3%), CTNNB1(11.4%), JAK1(6.7%). Copy number gains were detected in ARNT(20.5%) MTDH(16.2%), CCND1(12.4%), FGF19(11.4%) and MET(16.1%). For example, copy number losses of RB1(11.4%), SMAD4(18.1%), WRN(15.2%), ARID1A(7.6%), CDKN2A(12.4%), PTEN(5.7%) genes were identified.About 40% HCC PDXs responsed to sorafenib, that is concordant with clinical trials. Compared with sensitive PDXs, some cancer related signal pathway is disregulated in resistant ones, such as cell cycle, WNT pathway and INF related genes. Interestingly, combination of inhibitor of these pathways can enhance the anti-cancer activity of sorafenib. Among all differently expressed genes, DKK1 is the mostly up regulated gene, and the high level of DKK1 is associated with shorter progress-free survival in PDX models. Knocking down the expression of DKK1 enhanced sorafenib activity in vivo and in vitro. Serum DKK1 of patient death within 1 year was higher than those with still survival.Conclusion: 1. PDXs of HCC remained the biological and genetic characteristics. Genome variations frequently happened in HCC can be detected in PDX. Patients with successful xenograft in mouse indicate poor prognosis and need interventions.2. Cell cycle, WNT pathway and INF related genes might involve in sorafenib resistance and inhibitors of these pathways enhance the activity of sorafenib. DKK1 is a potential biomarker of sorafenib sensitivity. |
