Biologic Basis of Liver Cancer From Chronic Hepatitis B

Overview

The focus of the study is to identify viral factors and host immune responses that differentiate HBV-related HCC patients from HBV patients who have not progressed to HCC. To that end, the investigators will compare gene expression levels between HCC patients and non-HCC patients categorized into high and low risk profiles. The investigators will perform ANOVA to compare three groups (HCC, high risk, low risk). Multiple comparison corrections will be performed using Benjamini and Hochberg False Discovery Rate (FDR) with a 90% confidence that the discovery lists will contain no more than 5% false positives (FDR\<0.05) (PMID: 12584122, 11682119). A p-value \<0.05 is considered statistically significant using this multiple comparison correction approach. Post-hoc Student-Newman-Keuls or Tukey tests will be used following ANOVA for comparisons of HCC patients with high risk and low risk. If data are not normally distributed when log-transformed, then Kruskall-Wallis tests will be used. ANCOVA will be used to adjust for the effects of covariates, such as age, gender, and HBV genotype (B or C). Further, the investigators often use an additional 2-fold change criterion for significance because the investigators consider a fold change of this magnitude to be biologically significant. Hierarchical clustering analyses and principal component analyses will be used to visualize how well the genes separate the groups, or to discover new subgroups. For the analysis of SNVs, the exact binomial test will be performed and p-values will be adjusted by the Benjamini-Hochberg correction.

2\) Objectives Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related death worldwide (Jemel et al CA Cancer J Clin 2011). Asian and Pacific Islander (API) Americans are perennially the most affected racial group with regard to HCC incidence and HCC-related mortality due to the high prevalence of chronic hepatitis B (CHB). While APIs carry the greatest burden of CHB-related HCC, differences in outcome exist amongst the different Asian ethnic groups. The investigators have previously analyzed ethnically disaggregated data from 6,068 Californians of Asian ancestry with HCC diagnosed between 1998 and 2007 (Ryerson et al Cancer 2016). Cause-specific mortality was significantly higher among Laotian/Hmong (hazard ratio, 2.08; 95% CI, 1.78-2.44) and Cambodian patients (hazard ratio 1.26; 95% CI, 1.06-1.51). Notably, Laotian/Hmong patients presented at an early age and with more advanced stages of HCC suggesting that these differences were not related to delays in diagnosis or presentation, but rather biologic differences in the behavior of the HCC. In fact, both viral factors and host immune responses have been associated with the prognosis of CHB-related HCC. Currently, the investigators are investigating these potential mechanisms in Hmong/Laotian, Chinese, and Vietnamese individuals infected with the hepatitis B virus (HBV). In contrast to evaluating mechanisms associated with poor outcomes once HCC has developed, the current proposal will build upon the data the investigators have collected with the goal to identify markers of HCC development. Central hypothesis is that prior to the development of HBV-related HCC, there are specific biologic responses reflected in the gene expression profile of whole blood that can differentiate CHB patients with and without HCC. In order to test this hypothesis the investigators will perform the following specific aims: Part 1 (Aim #1): Associate viral sequences and host gene expression signatures with established HCC risk factors in Asian Americans. In order to meet this aim, blood samples will be obtained from 10 consented subjects. The DNA extracted from these samples will be used as part of an ongoing study comparing viral sequences and gene expression profiles across Asian ethnic groups, the investigators will analyze the same data comparing individuals with high and low HCC risk factors (i.e. the presence and absence of cirrhosis, high HBV viral load, men who have sex with men). Part 2 (Aim #2): Investigate HBV-related HCC tissue for HCC-associated viral sequences and human proteins. HCC tissue sections on slides with no personal health information will be obtained from UC Davis Cancer Center Biorepository (CCB) and other biobanks including, but not limited to, Cooperative Human Tissue Network. From previously banked HCC tissue, the investigators will 1) extract and sequence HBV DNA from tumor tissue and compare it to our collection of non-HCC HBV sequences and 2) stain the tissues for proteins corresponding to genes identified in whole blood gene expression analysis. 3\) Background According to an analysis of the Surveillance Epidemiology and End Results (SEER) database, API had the highest age-adjusted HCC incidence rates from 2006-2010 at 4.7 per 100,000 followed by Hispanics (3.2), Blacks (2.5), and whites (1.4). The highest age-specific HCC incidence rate found was 54.7 per 100,000 in API greater than 65 years of age. In this analysis, HCC-related mortality among API was the highest compared to other racial groups (2.8 per 100,000 compared to 0.9 per 100,000 for whites) (Altekruse et al. AJG 2014). This increased risk for HCC and its complications among API has been attributed to chronic hepatitis B (CHB) (El Sarag et al. Gastroenterology 2012). For persons with CHB, there is a strong association between increasing hepatitis B virus (HBV) DNA levels and increasing incidence of HCC and cirrhosis (Chen et al. JAMA 2006). The most common routes of HBV transmission are vertical from infected mothers to infants and horizontal in early childhood among those living in areas of high HBV endemicity (Nguyen et al. J Cancer Educ 2007) Despite the availability of a preventative vaccine, global vaccine coverage is only 75% and does not address the disease burden in the estimated 350 million people infected with HBV worldwide (Meireles et al. World J Hepatol 2015). Viral factors include viral load, genotype, and specific mutations. Notably, genotype B is associated with the early onset of HCC and poor prognosis (Yin et al. Carcinogenesis 2008) Viral mutants including the A1762T/G1764A basal core promoter mutations and Pre-S region deletions are also associated with poor HCC outcomes and may be the result of selection by host immune responses and activation of endogenous cytidine deaminase by pro-inflammatory cytokines (e.g. TNF-alpha) induced by NFkappa-B leading to G-to-A hypermutation (Chen et al. Eur J Cancer 2012). Other immune response pathways have been associated with outcomes from HBV-related HCC including the presence of natural killer cells, CD8+ T cells, regulatory T cells as well as IL-2 and IL-15 levels in the tumor microenvironment (Bowlus CL, Tumor Immunology 2007) Applicability of NGS to sequencing of HBV: NGS will allow the investigation of viral factors at multiple levels, including the frequency of genotypes and mutations, viral diversity, and mutation rates as a possible indication of cytidine deaminase activity. In addition, this technology will be used to explore the immune responses based upon transcript profiling. Further, analysis can also be done on T cell receptor usage and HLA alleles, which have been associated with HBV-related outcomes as well as association analysis of SNV within transcription profiles. Preliminary data: Under the R21, the investigators are utilizing NGS on blood samples from treatment naïve Chinese, Vietnamese, and Laotian/Hmong patients to identify viral sequences and expression profiles from whole blood that may segregate with Laotian/Hmong origins and thus be candidates that explain the poor HCC outcomes in this population. In addition, this data can be used to identify viral sequences and expression patterns which associate with known HCC-risk factors and may then be developed as potential biomarkers for the HCC surveillance among HBV infected people. The first step in this biomarker development process will need testing in a cohort of patients with HBV-related HCC and is the thrust of this protocol . Enrollment for this study has been completed (n = 99) and blood samples are currently being analyzed. The investigators conducted a pilot study using NGS for transcriptome profiling of whole blood samples to demonstrate the power of blood genomics for identifying molecular profiles associated with specific disease states and providing ultra-sensitive detection of molecular pathways associated with HBV pathogenesis (Caruccio et al. Methods Mol Bol 2011). RNA sequencing (RNA-Seq) was performed on total RNA isolated from whole blood samples obtained from a cohort of 10 Asian American individuals, composed of 3 healthy individuals (2 Chinese, 1 Filipino), 6 HBsAg-positive subjects without HCC (5 Chinese, 1 Vietnamese), and 1 HBsAg-positive Laotian subject with HCC. The HCC patient possessed a distinctive expression signature defined by 634 differentially expressed genes. Meta-analysis of the 173-gene overexpression cluster with immune cell transcriptome data (Li et al. Bioinformatics 2009) revealed that this "HBV-related HCC signature" contained prominent neutrophil- and macrophage-driven components, as well as up-regulated expression of regulatory T cell (TReg)-related genes, which is associated with worse prognosis for HCC (Tepper et al ICG Americas 2012 International Conference). Gene ontology analysis indicated over-represented processes consistent with HCC pathogenesis, including inflammation. The investigators plan to validate this "HBV-related HCC signature" in a larger cohort of patients with HBV-related HCC. Additional information about preliminary data can be found on the grant attached with this submission.