Molecular epidemiology in HCV-related hepatocellular carcinoma: First steps

COMMENTARY ON: Variation in the DEPDC5 locus is associated with progression to hepatocellular carcinoma in chronic hepatitis C virus carriers. Miki D, Ochi H, Hayes CN, Abe H, Yoshima T, Aikata H, Ikeda K, Kumada H, Toyota J, Morizono T, Tsunoda T, Kubo M, Nakamura Y, Kamatani N, Chayama K. Nat Genet 2011 July 3;43(8):797–800. Copyright (2011). Abstract reprinted with permission from McMillan Publishers Ltd. http://www.ncbi.nlm.nih.gov/pubmed/21725309 Abstract: Chronic viral hepatitis is the most important risk factor for progression to hepatocellular carcinoma (HCC). To identify genetic risk factors for progression to HCC in individuals with chronic hepatitis C virus (HCV), we analyzed 467,538 SNPs in 212 Japanese individuals with chronic HCV with HCC and 765 individuals with chronic HCV without HCC. We identified one intronic SNP in the DEPDC5 locus on chromosome 22 associated with HCC risk and confirmed the association using an independent case–control population (710 cases and 1625 controls). The association was highly significant when we analyzed the stages separately as well as together (rs1012068, P(combined)=1.27×10−13, odds ratio=1.75). The significance level of the association further increased after adjustment for gender, age, and platelet count (P=1.35×10−14, odds ratio=1.96). Our findings suggest that common variants within the DEPDC5 locus affect susceptibility to HCC in Japanese individuals with chronic HCV infection. The global incidence of hepatocellular carcinoma (HCC) is increasing, and it remains among the top three cancer killers worldwide. Chronic liver disease due to viral hepatitis B and C infection underlies HCC in more than 70% of cases. In cirrhotic patients with HCV infection, the annual HCC development rates range between 2% and 7%. Considering that the worldwide prevalence of HCV infection is 2%, and despite recent successes from first generation direct acting antivirals, it is likely that the HCV-related HCC medical and economic burden will continue to increase during the current decade. Surveillance programs for early detection of HCC have a positive impact on patient outcomes, and thus have been endorsed by clinical practice guidelines. Different studies have identified the key factors defining at-risk populations, such as male gender, bilirubin and AFP levels, portal hypertension indicators (e.g. HVPG measurements, platelet count, presence of varices), and viral genotypes [[1]Villanueva A. Minguez B. Forner A. Reig M. Llovet J.M. Hepatocellular carcinoma: novel molecular approaches for diagnosis, prognosis, and therapy.Annu Rev Med. 2010; 61: 317-328Crossref PubMed Scopus (207) Google Scholar]. However, none is routinely used to stratify risk within cirrhotic patients. Some cohort studies have underlined differences in HCC incidence according to patient's ethnicity, which can be about 2-fold higher in Japan than in the Western countries (5-year cumulative incidence of 30% and 17%, respectively) [[2]Altekruse S.F. Mcglynn K.A. Reichman M.E. Hepatocellular carcinoma incidence, mortality, and survival trends in the United States from 1975 to 2005.J Clin Oncol. 2009; 27: 1485-1491Crossref PubMed Scopus (1438) Google Scholar]. This fact suggests that inherited polymorphic traits could favor HCC development in certain populations. In this sense, previous data suggest that ethnic background seems to influence disease outcome and response to therapies in liver diseases. For example, a genetic polymorphism in IL28B gene defines a better response to conventional HCV therapy in European populations as opposed to African-American populations [[3]Ge D. Fellay J. Thompson A.J. Simon J.S. Shianna K.V. Urban T.J. et al.Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance.Nature. 2009; 461: 399-401Crossref PubMed Scopus (3148) Google Scholar]. Thus, the consequences of an accurate stratification of risk based on clinical or genomic parameters might have direct clinical implications. First, it would facilitate the conduction of chemoprevention trials by optimizing the selection of the target population. In principle, selection of high-risk patients would significantly reduce sample size and duration of this type of trials [[4]Llovet J.M. Di Bisceglie A.M. Bruix J. Kramer B.S. Lencioni R. Zhu A.X. et al.Design and endpoints of clinical trials in hepatocellular carcinoma.J Natl Cancer Inst. 2008; 100: 698-711Crossref PubMed Scopus (1478) Google Scholar]. Notably, not a single drug has been approved yet by regulatory agencies to prevent HCC development in cirrhotic patients of any etiology. Second, once the risk factors are thoroughly validated, it might refine the surveillance policies in specific populations. Genome-wide association studies (GWAS) were recently developed as an unbiased high-throughput genotyping tool to discover common genetic traits associated with human diseases. Between 2005 and June 2011, a total of 951 of such studies were deposited at the National Institutes of Health (www.genome.gov/gwastudies). Their hypothesis-free nature (i.e. unsupervised analysis of hundreds of thousands of single-nucleotide polymorphisms (SNP) in large cohorts of individuals) favors both false positive and negative results. The massive amount of statistical tests performed requires stringent corrections for multiple comparisons and robust validation in independent cohorts. Other potential sources of error are related to selection of study participants and genotyping inaccuracies [[5]Pearson T.A. Manolio T.A. How to interpret a genome-wide association study.JAMA. 2008; 299: 1335-1344Crossref PubMed Scopus (708) Google Scholar]. Despite their wide application in different liver diseases, until 2010 few data was available on molecular epidemiology in HCC [[6]Karlsen T.H. Melum E. Franke A. The utility of genome-wide association studies in hepatology.Hepatology. 2010; 51: 1833-1842Crossref PubMed Scopus (48) Google Scholar]. The study by Miki et al. [[7]Miki D. Ochi H. Hayes C.N. Abe H. Yoshima T. Aikata H. et al.Variation in the DEPDC5 locus is associated with progression to hepatocellular carcinoma in chronic hepatitis C virus carriers.Nat Genet. 2011; 43: 797-800Crossref PubMed Scopus (150) Google Scholar] is the last of a recent series of GWAS focused on HCC (Table 1 [7Miki D. Ochi H. Hayes C.N. Abe H. Yoshima T. Aikata H. et al.Variation in the DEPDC5 locus is associated with progression to hepatocellular carcinoma in chronic hepatitis C virus carriers.Nat Genet. 2011; 43: 797-800Crossref PubMed Scopus (150) Google Scholar, 8Zhang H. Zhai Y. Hu Z. Wu C. Qian J. Jia W. et al.Genome-wide association study identifies 1p36.22 as a new susceptibility locus for hepatocellular carcinoma in chronic hepatitis B virus carriers.Nat Genet. 2010; 42: 767-770Google Scholar, 9Kumar V. Kato N. Urabe Y. Takahashi A. Muroyama R. Hosono N. et al.Genome-wide association study identifies a susceptibility locus for HCV-induced hepatocellular carcinoma.Nat Genet. 2011; 43: 455-458Crossref PubMed Scopus (318) Google Scholar, 10Clifford R.J. Zhang J. Meerzaman D.M. Lyu M.-S. Hu Y. Cultraro C.M. et al.Genetic variations at loci involved in the immune response are risk factors for hepatocellular carcinoma.Hepatology. 2010; 52: 2034-2043Crossref PubMed Scopus (121) Google Scholar]). Authors analyzed a large set of HCV Japanese carriers (n = 3312) using a case–control design, and a two-step approach. In the training set (272 cases and 765 controls), they interrogated 467,538 germ-line SNPs and identified one, rs1012068, significantly associated with the risk of developing HCC, with an odds ratio of 2.2. The SNP retained significant predictive power upon logistic regression modeling including age, gender, and platelet count. This was further confirmed in a validation set including 710 cases and 1625 controls, using a different genotyping technique. The SNP is located on chromosome 22, and by using fine mapping studies, the authors identified DEPDC5 as the target gene harboring the different genotypes. Despite the fact that the function of this gene is unknown, there is some evidence of aberrations affecting its locus in human cancer (e.g. glioblastoma). The study has several limitations, noteworthy the lack of information regarding the stage of fibrosis and/or cirrhosis of patients and controls, which is critical to adjust for risk development. Only adjustment for platelet count, a suboptimal surrogate of portal hypertension, was provided. It is also concerning that such a well-powered study was unable to identify previously reported SNPs in other candidate genes (i.e. MICA) already documented as defining high-risk populations in Japanese patients [[9]Kumar V. Kato N. Urabe Y. Takahashi A. Muroyama R. Hosono N. et al.Genome-wide association study identifies a susceptibility locus for HCV-induced hepatocellular carcinoma.Nat Genet. 2011; 43: 455-458Crossref PubMed Scopus (318) Google Scholar]. Unfortunately, failure to reproduce the associations between SNPs and disorders is not uncommon in GWAS [[5]Pearson T.A. Manolio T.A. How to interpret a genome-wide association study.JAMA. 2008; 299: 1335-1344Crossref PubMed Scopus (708) Google Scholar], which emphasizes the importance of study design in these initiatives. In the case of HCV–HCC studies, selection of controls was heterogeneous among studies, one including HCV infected individuals [[7]Miki D. Ochi H. Hayes C.N. Abe H. Yoshima T. Aikata H. et al.Variation in the DEPDC5 locus is associated with progression to hepatocellular carcinoma in chronic hepatitis C virus carriers.Nat Genet. 2011; 43: 797-800Crossref PubMed Scopus (150) Google Scholar], while the other also included healthy controls [[9]Kumar V. Kato N. Urabe Y. Takahashi A. Muroyama R. Hosono N. et al.Genome-wide association study identifies a susceptibility locus for HCV-induced hepatocellular carcinoma.Nat Genet. 2011; 43: 455-458Crossref PubMed Scopus (318) Google Scholar], which could partially justify results' discrepancy.Table 1Published GWAS in hepatocellular carcinoma. Open table in a new tab At last, the concept of defining high-risk populations based on germ-line SNPs has entered the HCC field, following what has been reported in several solid tumors. The lack of reproducibility among the SNPs found in HCC reports (Table 1) reflects both the high molecular heterogeneity linked to the initiation and progression of this neoplasm, along with some methodological issues in study design that need to be overcome in future investigations. Further studies in Asian and Western HCV-related HCC patients should confirm the relevance of the SNPs identified in these seminal investigations. Ideally, these findings will ultimately refine surveillance interventions and clear the path for trial enrichment testing chemoprevention strategies. The authors declare that they have nothing to disclose regarding funding of conflict of interest with respect to this manuscript.