Hepatocellular carcinoma is the major histologic type of malignant primary liver neoplasm. It is the fifth most common cancer and the third most common cause of death from cancer worldwide. The major risk factors for HCC are chronic ... Hepatocellular carcinoma is the major histologic type of malignant primary liver neoplasm. It is the fifth most common cancer and the third most common cause of death from cancer worldwide. The major risk factors for HCC are chronic hepatitis B virus (HBV) infection, chronic hepatitis C virus (HCV) infection, prolonged dietary aflatoxin exposure, alcoholic cirrhosis, and cirrhosis due to other causes. Hepatoblastomas comprise 1-2% of all malignant neoplasms of childhood, most often occurring in children under 3 years of age. Hepatoblastomas are thought to be derived from undifferentiated hepatocytes (Taniguchi et al., 2002).
Primary cancer of the liver in 3 brothers was described by Kaplan and Cole (1965) and by Hagstrom and Baker (1968). In these patients there was no recognized preexisting liver disease. Denison et al. (1971) described 2 adult ... Primary cancer of the liver in 3 brothers was described by Kaplan and Cole (1965) and by Hagstrom and Baker (1968). In these patients there was no recognized preexisting liver disease. Denison et al. (1971) described 2 adult brothers who died of primary hepatocellular carcinoma. Both had micronodular cirrhosis with features of subacute progressive viral hepatitis. Australia antigen was demonstrated in the brother in whom it was sought. Their father had died much earlier of hepatocellular carcinoma. Hepatoblastoma has been described in sibs (Fraumeni et al., 1969; Napoli and Campbell, 1977; Ito et al., 1987). See 231100 for description of liver cancer as a complication of giant cell hepatitis of infancy. Familial liver cell carcinoma might also have its explanation in alpha-1-antitrypsin deficiency (613490), hemochromatosis (235200), and tyrosinemia (276700).
Oda et al. (1996) observed loss of heterozygosity (LOH) at the APC and/or MCC (159350) loci in 4 (57%) of 7 informative hepatoblastoma tissues. Somatic mutations were detected in 8 (61.5%) of the ... - Somatic Mutations Oda et al. (1996) observed loss of heterozygosity (LOH) at the APC and/or MCC (159350) loci in 4 (57%) of 7 informative hepatoblastoma tissues. Somatic mutations were detected in 8 (61.5%) of the 13 total cases, with 9 cases (69%) showing genetic alterations in the APC gene as LOH or somatic mutations (see, e.g., 611731.0024). Double mutations were demonstrated in 2 cases. The nature of the somatic mutations observed in this study was unusual because 9 of the 10 mutations were missense, with only 1 case featuring a frameshift mutation due to an insertion. By contrast, more than 90% of mutations in the APC gene in colorectal tumors result in a truncated APC protein due to either frameshift or nonsense mutations. Thorgeirsson and Grisham (2002) reviewed the molecular pathogenesis of HCC. The malignant phenotype is heterogeneous, and is produced by the disruption of a number of genes that function in different regulatory pathways, producing several molecular variants of HCC. The authors diagrammed the chronologic sequence of cellular events starting with chronic hepatitis from hepatitis B virus (HBV), hepatitis C virus (HCV), and AFB1, which together are responsible for approximately 80% of all HCCS in humans, and proceeding through successive steps of dysplasia and neoplasia. They diagrammed 11 autosome arms in 9 different chromosomes that have been found to contain allelic deletions in more than 30% of reported HCC. Other autosome arms contain allelic deletions in more than 20% of HCC. Heterogeneity of genomic aberrations may reflect the actions of different causative agents. A notable example of an aberration in gene structure related to a specific cause of HCCs is high frequency of the mutation arg249 to ser (191170.0006) of p53 in the tumors of patients chronically exposed to AFB. Thorgeirsson and Grisham (2002) tabulated 14 genes affected by loss of heterozygosity (LOH), mutation, or both in more than 15% of HCCs. Activation of wingless (Wnt) signaling through mutations in beta-catenin (CTNNB1; 116806) contributes to the development of HCC and hepatoblastoma (review by Taniguchi et al., 2002). To explore the contribution of additional Wnt pathway molecules to hepatocarcinogenesis, Taniguchi et al. (2002) examined CTNNB1 mutations and mutations in AXIN1 (603816) and AXIN2 (604025) in 73 HCCs and 27 hepatoblastomas. Beta-catenin mutations were detected in 19.2% (14 of 73) HCCs and 70.4% (19 of 27) hepatoblastomas. Beta-catenin mutations in HCCs were primarily point mutations, whereas more than half of the hepatoblastomas had deletions. AXIN1 mutations occurred in 7 (9.6%) HCCs and 2 (7.4%) hepatoblastomas. The AXIN1 mutations included 7 missense mutations, a 1-bp deletion, and a 12-bp insertion. The predominance of missense mutations found in the AXIN1 gene is different from the small deletions or nonsense mutations described previously. Loss of heterozygosity at the AXIN1 locus was present in 4 of 5 informative HCCs with AXIN1 mutations, suggesting a tumor suppressor function for AXIN1. AXIN2 mutations were found in 2 (2.7%) HCCs and in no hepatoblastomas. Two HCCs had both AXIN1 and beta-catenin mutations, and 1 HCC had both AXIN2 and beta-catenin mutations. About half of the HCCs with AXIN1 or AXIN2 mutations showed beta-catenin accumulation in the nucleus, cytoplasm, or membrane. Overall, the data indicated that besides the approximately 20% of HCCs and 80% of hepatoblastomas with beta-catenin mutations contributing to hepatocarcinogenesis, AXIN1 and AXIN2 mutations appear to be important in an additional 10% of HCCs and hepatoblastomas. Lee et al. (2005) detected somatic mutations in the PIK3CA gene (see, e.g., 171834.0007; 171834.0008) in 26 (35.6%) of 73 hepatocellular carcinomas. Li et al. (2011) performed exome sequencing of 10 HCV-associated HCCs and matched normal tissue from 10 patients and a subsequent evaluation of additional affected individuals, and discovered novel inactivating mutations of ARID2 (609539) in 4 major subtypes of HCC (HCV-associated HCC, hepatitis B virus (HBV)-associated HCC, alcohol-associated HCC, and HCC with no known etiology). Notably, 18.2% of individuals with HCV-associated HCC in the United States and Europe harbored ARID2 inactivating mutations, suggesting that ARID2 is a tumor suppressor gene that is relatively commonly mutated in this tumor subtype. Huang et al. (2012) used exome sequencing to identify somatic mutations in 10 hepatitis B virus-positive individuals with hepatocellular carcinoma with portal vein tumor thromboses (PVTTs), intrahepatic metastases. Both C:G-A:T and T:A-A:T transversions were frequently found among the 331 non-silent mutations. Notably, ARID1A (603024), which encodes a component of the SWI/SNF chromatin remodeling complex, was mutated in 14 of 110 (13%) HBV-associated HCC specimens. Huang et al. (2012) used RNA interference to assess the roles of 91 of the confirmed mutated genes in cellular survival. The results suggested that 7 of these genes, including VCAM1 (192225) and CDK14 (610679), may confer growth and infiltration capacity to HCC cells. - Gene Expression Studies Agarwal et al. (1998) reported a case of severe gynecomastia in a 17.5-year-old boy due to high levels of aromatase (CYP19A1; 107910) expression in a large fibrolamellar hepatocellular carcinoma, which caused extremely elevated serum levels of estrone (1200 pg/mL) and estradiol-17 (312 pg/mL) that suppressed follicle-stimulating hormone (FSH) and luteinizing hormone (LH) (1.3 and 2.8 IU/L, respectively) and consequently testosterone (1.53 ng/mL). After removal of the 1.5-kg tumor, gynecomastia partially regressed, and normal hormone levels were restored. By immunohistochemistry, diffuse intracytoplasmic aromatase expression was detected in the liver cancer cells. Northern blot analysis showed P450 aromatase transcripts in total RNA from the hepatocellular cancer but not in the adjacent liver nor in disease-free adult liver samples. Promoters I.3 and II were used for P450 aromatase transcription in the cancer. Schwienbacher et al. (2000) analyzed DNA and RNA from 52 human hepatocarcinoma samples and found abnormal imprinting of genes located at 11p15 in 51% of 37 informative samples. The most frequently detected abnormality was gain of imprinting, which led to loss of expression of genes present on the maternal chromosome. As compared with matched normal liver tissue, hepatocellular carcinoma showed extinction or significant reduction of expression of one of the alleles of the CDKN1C (600856), SLC22A1L (602631), and IGF2 (147470) genes. Loss of maternal-specific methylation of the KvDMR1 gene (607542) in hepatocarcinoma correlated with abnormal expression of CDKN1C and IGF2, suggesting a function for KvDMR1 as a long-range imprinting center active in adult tissues. These results pointed to the role of epigenetic mechanisms leading to loss of expression of imprinted genes at 11p15 in human tumors. Ye et al. (2003) analyzed the expression profiles of hepatocellular carcinoma samples without or with intrahepatic metastases. Using a supervised machine-learning algorithm, they generated a molecular signature that can classify metastatic HCC patients and identified genes that were relevant to metastasis and patient survival (which is related particularly to intrahepatic metastases). They found that the gene expression signature of primary HCCs with accompanying metastasis was very similar to that of their corresponding metastases, implying that genes favoring metastasis progression were initiated in the primary tumors. Osteopontin (OPN; 166490), which was identified as a lead gene in the signature, was overexpressed in metastatic HCC; an osteopontin-specific antibody effectively blocked HCC in vitro and inhibited pulmonary metastasis of HCC cells in nude mice. Thus, osteopontin acts as both a diagnostic marker and a potential therapeutic target for metastatic HCC. Tanabe et al. (2008) reported an association between a 61A-G SNP (dbSNP rs4444903) in the EGF gene (131530) and the development of hepatocellular carcinoma in patients with cirrhosis. Secretion of EGF was 2.3-fold higher in G/G hepatocellular carcinoma cell lines compared to A/A cell lines, and mRNA transcripts with the G allele showed a longer half-life and increased stability. Among 207 patients with cirrhosis, liver EGF levels were 2.4-fold higher in G/G patients compared to A/A patients. Fifty-nine of the 207 patients with cirrhosis also had hepatocellular carcinoma, and there was a 4-fold increased odds of hepatocellular carcinoma in G/G patients compared with A/A patients. The association was validated in a second cohort of 121 patients with alcoholic cirrhosis and hepatocellular carcinoma. Ji et al. (2009) analyzed hepatocellular carcinoma tissue derived from 455 patients from Shanghai and Hong Kong. Most of the patients were men (85.1%), were long-term carriers of the hepatitis B virus (HBV) (90.5%), had cirrhosis (88.0%), and an elevated serum level of alpha-fetoprotein (62.2%). The expression of MIRN26A1 (612151) and MIRN26B (612152) was decreased in tumor tissue compared to non-tumor tissue. Sex stratification showed that women had significantly higher expression of the MIRN26 genes in non-tumor liver tissue compared to men. Gene expression profiling using microarray showed distinct genetic patterns between tumors with low and high MIRN26 expression. The findings were consistent with the MIRN26 genes acting as tumor suppressors. Patients whose tumors showed low MIRN26 expression had shorter overall survival, but better response to interferon therapy compared to patients with higher expression of these MIRNs. Yong et al. (2013) screened specimens obtained from 179 patients with hepatocellular carcinoma from Singapore for the expression of SALL4 (607343). SALL4 is an oncofetal protein that is expressed in human fetal liver and silenced in the adult liver, but is reexpressed in a subgroup of patients that have hepatocellular carcinoma with unfavorable prognosis. Gene expression analysis showed the enrichment of progenitor-like gene signatures with overexpression of proliferative and metastatic genes in SALL4-positive hepatocellular carcinomas. Yong et al. (2013) found that in a multivariate Cox regression model, SALL4 is an independent prognostic factor for overall survival (hazard ratio for death, 2.87; 95% confidence interval, 1.09 to 7.52; p = 0.03) in the Singapore cohort and an independent predictor of both overall survival (hazard ratio for death, 1.52; 95% confidence interval, 1.00-2.32; p = 0.05) and early recurrence (hazard ratio, 1.67; 95% CI, 1.11-2.51; p = 0.01) in the Hong Kong cohort. Loss-of-function studies confirmed the critical role of SALL4 in cell survival and tumorigenicity. Blocking SALL4-corepressor interactions released suppression of PTEN (601728) and inhibited tumor formation in xenograft models in vivo. To help avoid additional risks associated with tumor biopsy in patients with SALL4-negative tumors, Hopkins et al. (2013) requested that Yong et al. (2013) determine the sensitivity and specificity of the serum alpha-fetoprotein level in predicting the SALL4 status of their hepatocellular carcinoma patients. Yong et al. (2013) replied that in the Hong Kong cohort of patients with hepatocellular carcinoma they observed a significant correlation between SALL4 mRNA expression and the serum alpha-fetoprotein level (p less than 0.001). The sensitivity and specificity of serum alpha-fetoprotein levels in predicting SALL4 expression in this cohort of 228 patients were calculated to be 66.7% and 68.4%, respectively. The data suggested that with a cutoff value of 100 ng per milliliter, a serum alpha-fetoprotein level of 100 ng per milliliter or more can identify 66.7% of patients with SALL4 mRNA expression. However, 31.6% of patients without SALL4 expression would be falsely identified as being positive for SALL4 expression. Yong et al. (2013) suggested that, to assess SALL4 expression without the need for patient biopsy specimens, a noninvasive assay to test serum SALL4 expression be developed. Suzuki et al. (2013) asked of Yong et al. (2013) whether the unfavorable clinical outcome in high-SALL4 hepatocellular carcinoma is attributable to features of cholangiocarcinoma. Masuda et al. (2013) inquired whether reprogramming factors such as KLF5 or TBX3 could be involved in the pathogenesis of SALL4-related hepatocellular carcinoma, and if reexpression of SALL4 through hepatitis B infection has a role in upregulation of SALL4. Yong et al. (2013) replied that more studies would be necessary to address these queries. - Hepatitis B Infection Integration of HBV into cellular DNA occurs during long-term persistent infection in man. Hepatocellular carcinomas isolated from carriers of virus often contain clonally propagated viral DNA. Shen et al. (1991) presented evidence for the interaction of inherited susceptibility and hepatitis B viral infection in cases of primary hepatocellular carcinoma in eastern China. Complex segregation analysis of 490 extended families supported the existence of a recessive allele with population frequency approximately 0.25, which results in a lifetime risk of HCC in the presence of both HBV infection and genetic susceptibility, of 0.84 for males and 0.46 for females. The model further predicted that, in the absence of genetic susceptibility, lifetime risk of HCC is 0.09 for HBV-infected males and 0.01 for HBV-infected females and that regardless of genotype the risk is virtually zero for uninfected persons. The finding of small deletions in retinoblastoma and Wilms tumor prompted Rogler et al. (1985) to look for the same in association with HBV integration in hepatocellular carcinoma. They demonstrated a deletion of at least 13.5 kb of cellular sequences in a liver cancer. The HBV integration and the deletion occurred on the short arm of chromosome 11 at location 11p14-p13. The deleted sequences were lost in tumor cells leaving only a single copy. Clones of the DNA flanking the deleted segment were used for the mapping of the deletion in somatic cell hybrids and by in situ hybridization. Cellular sequences homologous to the deleted region were cloned and used to exclude the possibility that this DNA had been moved to other positions in the genome. Fisher et al. (1987) extended the observations of Rogler et al. (1985). Using somatic cell hybrids that contained defined 11p deletions, 2 cloned DNA sequences that flank the deletion generated by a hepatocellular carcinoma (as a consequence of hepatitis B virus integration) were mapped to 11p13. Wilms tumor (194070) and the tumors of Beckwith-Wiedemann syndrome (130650) are also determined by changes on 11p. Wang and Rogler (1988) found loss of heterozygosity in 11p and 13q. Integration of DNA from the hepatitis B virus has been shown to occur frequently in human hepatocellular carcinomas. Recombinant DNA probes have been isolated from such a tumor. These probes detected rearrangements of the corresponding DNA domain in 10% of liver tumors regardless of whether they were HBV-related or not (Pasquinelli et al., 1988). Blanquet et al. (1987, 1988) cloned the normal allele and used it for mapping an HCC locus by somatic cell hybrid studies and by in situ hybridization. These experiments showed that the locus is located in the area 4q32.1. Buetow et al. (1989) found that 7 of 11 primary liver tumors tested against a panel of RFLPs demonstrated loss of constitutional heterozygosity for markers on chromosome 4, particularly 4q. Buetow et al. (1989) suggested that chronic hepatitis B virus infection and other environmental agents may operate through genetic events leading to loss of a tumor suppressor locus (anti-oncogene) on chromosome 4. The findings were thought to be consistent with those of Pasquinelli et al. (1988), which placed the critical region in the vicinity of 4q32. Smith et al. (1989) gave evidence for microdeletions of chromosome 4q involving the alcohol dehydrogenase isoenzyme gene ADH3 (ADH1C; 103730) and hepatomas from 3 of 5 individuals heterozygous for an XbaI RFLP detectable by the ADH probe. Two of 7 individuals heterozygous for an epidermal growth factor RFLP had lost 1 EGF allele in their hepatoma tissue. Henderson et al. (1988) demonstrated that the integration of HBV DNA can result in, or be accompanied by, interchromosomal exchange of genomic material containing the integrated DNA. Using in situ hybridization, they found that unique cellular DNA to the left of an HBV DNA integration site, cloned from a primary tumor, mapped to chromosome 18q (18q11.1-q11.2); right-hand flanking DNA mapped to chromosome 17 (17q22-q25). In a hepatoma specimen from Shanghai, Zhou et al. (1988) identified integration of hepatitis B virus into 17p12-p11.2, which is near the human protooncogene p53 (191170). Furthermore, the sequence of flanking cellular DNA showed highly significant homology with a conserved region of a number of functional mammalian DNAs, including the human autonomously replicated sequence-1 (ARS1; 109110). ARS1 is a sequence of human DNA that allows replication of Saccharomyces cerevisiae integrative plasmids as autonomously replicating elements in S. cerevisiae cells. Since integration of viral DNA is not a required step in the replicative cycle of the hepatitis virus, the presence of integrated HBV sequences in many human hepatocellular carcinomas suggests a causal relationship. Since any 1 of several integration sites may lead to the same result, the crucial cellular targets involved in triggering liver cell malignant transformation may differ from tumor to tumor. Primary hepatocellular carcinoma occurs at high frequencies in east Asia and sub-Saharan Africa. In these areas of the world, chronic infection with the hepatitis B virus is the best documented risk factor; however, only 20 to 25% of HBV carriers develop HCC. Exposure to the fungal toxin aflatoxin B1 (AFB1) has been suggested to increase HCC risk, in part because in vitro experiments demonstrated that AFB1 mutagenic metabolites bind to DNA and are capable of inducing G-to-T transversions. In certain areas of the HCC endemic regions, a mutation hotspot has been reported in the p53 tumor suppressor gene (TP53; 191170): an AGG-to-AGT transversion (arginine to serine) of codon 249 in exon 7 (191170.0006). Microsomal epoxide hydrolase (EPHX; 132810) and glutathione-S-transferase M1 (GSTM1; 138350) are both involved in AFB1 detoxification in hepatocytes. Polymorphism of both genes has been identified. In Ghana and China, McGlynn et al. (1995) conducted studies to determine whether mutant alleles at one or both of these loci are associated with increased levels of serum AFB1-albumin adducts, with HCC, and with mutations at codon 249 of p53. In a cross-sectional study, they found that mutant alleles at both loci were significantly overrepresented in individuals with serum AFB1 albumin adducts. Additionally, in a case-control study, mutant alleles of EPHX were significantly overrepresented in persons with HCC. The relationship of EPHX to HCC varied by hepatitis B surface antigen status, indicating that a synergistic effect may exist. Mutations at codon 249 of p53 were observed only among HCC patients with one or both high-risk genotypes. These findings by McGlynn et al. (1995) supported the existence of genetic susceptibility in humans to the environmental carcinogen AFB1 and indicated that there is a synergistic increase in risk of HCC with the combination of hepatitis B virus infection and susceptible genotype. Chiu et al. (2007) examined the roles of androgen receptor (AR; 313700) and the HBV nonstructural protein HBx in hepatocellular carcinoma, a disease that predominantly affects males. HBx increased the anchorage-independent colony formation potency of AR in a nontransformed mouse hepatocyte cell line. AR-mediated transcriptional activity was enhanced by HBx in an androgen concentration-dependent manner. Mutation analysis showed that HBx-enhanced AR gene transcriptional activity required intact HBx and the hinge region of AR. Immunoprecipitation and cell fractionation analyses revealed that HBx-AR interactions occurred mainly in the cytosol. HBx-enhanced AR activation involved SRC (190090) activity. Chiu et al. (2007) concluded that HBx is a noncellular positive coregulator of AR. - Glycogen Storage Disease Type Ia Hepatocellular adenoma (HCA) is a frequent long-term complication of glycogen storage disease type I (GSD I; 232200), and malignant transformation to hepatocellular carcinoma (HCC) occurs in some cases. Kishnani et al. (2009) performed genomewide SNP analysis and mutation detection of target genes in 10 GSD Ia-associated HCA and 7 general population HCA cases. Chromosomal aberrations were detected in 60% of the GSD Ia HCA and 57% of general population HCA. Coincident gain of chromosome 6p and loss of 6q were seen only in GSD Ia HCA (3 cases) with 1 additional GSD I patient showing submicroscopic 6q14.1 deletion. The sizes of GSD Ia adenomas with chromosome 6 aberrations were larger than the sizes of adenomas without the changes (P = 0.012). Expression of IGF2R (FCGR2A; 146790) and LATS1 (603473) candidate tumor suppressor genes at 6q was reduced in more than 50% of 7 GSD Ia HCA examined. None of the GSD Ia HCA had biallelic mutations in the HNF1A (142410) gene. The authors suggested that chromosome 6 alterations could be an early event in the liver tumorigenesis in GSD I, and possibly in general population. - Associations with HCC in Chronic HBV Carriers Pending Confirmation By means of genetic association analysis, Shin et al. (2003) showed that the interleukin-10 (IL10; 124092) haplotype IL10-ht2 was strongly associated with hepatocellular carcinoma in a well-characterized HBV cohort. The frequency of susceptible IL10-ht2 was much higher in HCC patients and significantly increased in order of susceptibility to HBV progression from chronic hepatitis to liver cirrhosis and HCC among hepatitis B patients. In addition, survival analysis showed that the onset age of HCC was also accelerated among chronic hepatitis B patients who were carrying IL10-ht2. Shin et al. (2003) suggested that increased IL10 production mediated by IL10-ht2 accelerates progression of chronic HBV infection, especially to HCC development. In a genomewide association study of 355 chronic HBV carriers with HCC and 360 chronic HBV carriers without HCC, all of Chinese ancestry, Zhang et al. (2010) found an association between HBV-related HCC and a SNP (dbSNP rs17401966) in an intron of the KIF1B gene (605995) on chromosome 1p36.22. The association was confirmed in 5 additional independent Chinese samples, consisting of 1,962 individuals with HCC, 1,430 control subjects, and 159 family trios. Across all 6 studies, the combined p value for the protective G allele of dbSNP rs17401966 was 1.7 x 10(-18), with an odds ratio of 0.61. - Associations with HCC in Chronic HCV Carriers Pending Confirmation Kumar et al. (2011) conducted a genomewide association study using 432,703 autosomal SNPs in 721 individuals with HCV-induced HCC and 2,890 HCV-negative controls of Japanese origin. Eight SNPs that showed possible association in the genomewide association study were further genotyped in 673 cases and 2,596 controls. Kumar et al. (2011) found a previously unidentified locus in the 5-prime flanking region of MICA (600169) on 6p21.33 (dbSNP rs2596542, combined p = 4.21 x 10(-13), odds ratio = 1.39) to be strongly associated with HCV-induced HCC. Subsequent analyses using individuals with chronic hepatitis C (CHC) indicated that this SNP is not associated with CHC susceptibility but is significantly associated with progression from CHC to HCC (p = 3.13 x 10(-8)). Kumar et al. (2011) also found that the risk allele of dbSNP rs2596542 was associated with lower soluble MICA protein levels in individuals with HCV-induced HCC (p = 1.38 x 10(-13)).