Cytokine polymorphisms and hepatocellular carcinoma

To B or not to B


  • Raymond T. Chung M.D.

    Corresponding author
    1. Gastrointestinal Unit, Massachusetts General Hospital, Boston, Massachusetts
    • Gastrointestinal Unit, GRJ 825, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114
    Search for more papers by this author
    • See referenced original article on pages 740–8, this issue.

  • Fax: (617) 726-5895


Regardless of whether monitoring practices change as a result of genotyping, the mandate that replicative hepatitis B virus disease must be treated because of the long-term risk for necroinflammatory disease, fibrosis, cirrhosis, and hepatocellular carcinoma remains firmly in place.

See also pages 740–8.

Hepatocellular carcinoma (HCC), one of the world's most common malignancies, is responsible for over 1 million deaths annually. The incidence of HCC is rising steadily, especially in countries in which chronic viral hepatitis is common. Worldwide, chronic infection with hepatitis B virus (HBV) represents the most important cause of HCC.1

HBV, which is a member of the hepadnaviruses, is a partially double-stranded DNA virus with a genome that contains four partially overlapping open reading frames. These open reading frames encode the pre-S1-S2-S, precore-core, polymerase, and X proteins. The pre-S1-S2-S gene product is processed to become hepatitis B surface antigen (HBsAg), and the secretion of HBsAg is emblematic of chronic infection. In most patients, the precore-core gene product yields secreted HBeAg, a marker of active replication, which also can be confirmed by molecular tests for circulating HBV DNA. HBV contributes to hepatocarcinogenesis through several mechanisms. HBV, as a DNA virus, is capable of integration into host chromosomal DNA, in which random insertion adjacent to protooncogenes or tumor suppressor genes could activate proliferative pathways. In addition, the HBV X protein itself may be oncogenic.2 Most commonly, active replication is associated with a necroinflammatory response, which increases hepatocyte necrosis, regeneration, and fibrosis and is associated with increased risk for the accumulation of mutations that contribute to malignant transformation.3 Finally, the development of cirrhosis, which is the culmination of fibrosis accumulation, itself also appears to be associated with an increased risk for HCC. A recent large study from Taiwan found that the HCC risk was increased 10-fold among HBsAg-reactive men who lacked HBeAg, that the HCC risk was increased 60-fold among men who were reactive for both HBsAg and HBeAg, and that this risk increased with age.4

The risk of developing persistent HBV infection is related directly to the patient's age at the time of exposure. Neonatal infection is associated with chronicity rates exceeding 90%, whereas individuals who are exposed in adulthood experience virologic clearance rates > 95%. Among individuals from areas that are highly endemic for HBV, perinatal transmission accounts for the vast majority of infections. Accordingly, the risk for HCC in these areas is the highest in the world. However, even in those regions, only a minority of chronic HBV carriers will develop HCC, suggesting that host factors or other environmental factors are important determinants of the risk for HCC.

The host immune response to HBV infection has been characterized relatively well.5 CD4 cells recognize viral peptides in conjunction with major histocompatability complex Class II molecules at the hepatocyte cell surface and release T-helper cell Th1 and Th2 cytokines. Th1 cytokines include interleukin 2 (IL-2), interferon γ (IFN-γ), tumor necrosis factor α (TNF-α), IL-12, and IL-18. These cytokines, in turn, lead to the activation of virus-specific CD8-positive cytolytic T lymphocytes, which effect clearance of virally infected cells. Th2 cytokines include IL-4 and IL-10, which regulate Th1 cytokine production and lead to activation and differentiation of antibody-producing B cells. A reductionist model proposes that excess Th1 cytokine production leads to a net antiviral state, whereas excess Th2 production tends to counteract the Th1 effect toward a less antiviral state.

Against this backdrop, the production of these cytokines, of course, is governed not only by the engagement of CD4 cells by HBV-infected hepatocytes but also by the intrinsic level of secretion of these cytokines, which is determined in large measure at the level of individual gene expression. Thus, it is plausible that, among chronically HBV-infected patients who have high rates of HCC development, cytokine production is altered fundamentally compared with cytokine production among individuals do not develop HCC. Given such a scenario, an imbalance of cytokines toward a less antiviral state may increase viral replication and may lead to an increased frequency of events that produce malignant transformation.

In this issue of Cancer, Nieters et al. tested this hypothesis.6 They performed a prospective case–control study examining Th1 and Th2 genotypes among chronic HBV carriers in a region of China known to have one of the highest prevalence rates of HCC in the world. From 1995 to 1998, 250 incident hospitalized patients with of HCC in southern Guangxi, China, were paired with contemporaneously admitted age-, gender-, and ethnicity-matched controls without HCC or cirrhosis. The great majority of patients, as expected, were HBsAg-positive, whereas the great majority of controls were HBsAg-negative. Cytokine genotypes were then assessed and classified as high-activity or low-activity, depending on associations described in the literature.

Nieters et al. found that the presence of each predicted low-activity Th1 genotype of IFN-γ, IL-12, and IL-18 was associated with a nonsignificant risk for HCC; however, when the genotypes were considered in composite fashion, the risk for HCC rose with increasing numbers of low-activity genotypes. It should be noted that this finding did not hold true for the major Th1 genes IL-2 and TNF-α, which were distributed comparably between cases and controls. An analysis of the major Th2 genes IL-4 and IL-10 also revealed a significant trend for low-activity genotypes and an association with decreased risk for HCC. Taken together, these findings suggest that individuals who had the maximum number of low-activity Th1 genes and the minimum number of low-activity Th2 genes had the highest (20-fold) relative risk for HCC compared with those who had no low-activity Th1 genes and had at least 1 low-activity Th2 gene. Subgroup comparisons of cases and controls demonstrated that these cytokine-genotype associations were even stronger among individuals who harbored chronic HBV infection compared with individuals who were HBsAg-negative. Taken together, those findings implicate host genetic predispositions in the development of HBV-related HCC. Specifically, the finding of host genotypes collectively predicted to be associated with an attenuated cellular immune response appears to be associated with higher rates of HCC. If these findings can be confirmed, then they open up a potentially new area of risk stratification for HCC.

However, these provocative findings beg several additional questions. First and most important, there is the issue of genotype-phenotype correlation. The prediction of a low-activity Th1 gene is not necessarily borne out in direct testing of gene activity; because, potentially, there are many modifiers of gene expression that cannot be predicted by gene testing, such as the presence of other unrecognized gene polymorphisms that compensate for the predicted activity. Thus, it would be important to correlate diminished IFN-γ or IL-4 release from lymphocytes in individuals who have low-activity polymorphisms using functional assays. Such a correlation would strengthen the mechanistic underpinning of the association.

Perhaps even more important, if, in fact, the presence of multiple low-activity Th1 genotypes and few low-activity Th2 genotypes is associated with diminished host-immune control of virus, then this should be reflected in measurements of viral control. Therefore, it would have been of interest to know the replicative status of HBV among the patients surveyed. If high HBV DNA levels and HBeAg status in fact do correlate with the genotypes associated with diminished immune control, then the prediction may be validated. The polymorphisms would then help to explain who among chronic carriers would be prone to HBV replication. Indeed, other studies have correlated low-activity IFN-γ polymorphisms with risk of chronic HBV infection.7 Thus, HBV replication itself may be the dominant factor driving hepatocarcinogenesis, in that a persistent inflammatory state (albeit blunted) in the face of ongoing replication is a greater risk for transformation than the more robust immune response that succeeds in keeping HBV in an inactive carrier state. This prediction squares with the observed finding of higher HCC risk in HBeAg-positive individuals.4 The utility of gene testing for these polymorphisms, then, may lie more in identifying those who are at risk for replicative disease than those who are at risk for HCC per se.

Although it may appear attractive to stratify patients who have perinatal HBV for subsequent risk of replication and HCC risk, it remains doubtful that any single genotype or combination of genotypes will do so with absolute sensitivity and specificity. Thus, the recommendation to monitor HBsAg-positive individuals for the development of HBV replication, which can occur spontaneously at any time during the life of a carrier, would appear unlikely to change. The intensity of that monitoring, however, may be influenced subtly if the findings from Nieters et al.6 can be confirmed. For instance, a high-activity Th1 and low-activity Th2 genotype may require even less intensive monitoring than that already practiced for HBsAg-positive, inactive chronic carriers. Whether monitoring practices change or not as a result of genotyping, the mandate that replicative HBV disease must be treated because of the long-term risk for necroinflammatory disease, fibrosis, cirrhosis, and HCC remains firmly in place. Ultimately, it will be the introduction of ever-improving antiviral chemotherapy against HBV that represents our best weapon against hepatocarcinogenesis.