Role of Estrogen in Liver Cancer

Abstract

Liver cancer is the fifth most common cancer worldwide and despite increasing implementation of ultrasonographic surveillance strategies, its incidence is rising, especially in western countries. A universal characteristic of hepatocellular carcinoma is the striking male prevalence that is found, with few exceptions, both in animals and in humans. Many different hypotheses have been put forward in an attempt to explain this finding, which is not a simple epidemiological oddity but could also have pathogenetic implications. An obvious trail to follow, as gender susceptibility is implicated, is the role played by sex hormones, namely estrogens. Estrogens are not simply involved in reproductive mechanisms; instead, it is increasingly evident that they have a role in such an enormous variety of cellular processes that their implication in liver carcinogenesis may be manifold. The purpose of this review is to provide an overview of the available data, with a special focus on the hormonal mechanisms potentially implicated in the development of liver cancer.

Keywords

chronic inflammation cirrhosis estrogens gender hepatocellular carcinoma menopause

Liver cancer remains, despite widespread implementation of screening programs [1,2], improvement of diagnostic tools and progress in therapeutic measures [3], the fifth most common cancer worldwide [4]. In addition, its incidence is rising, especially in countries with an intermediate/low rate of hepatocellular carcinoma (HCC), and despite the great attention to this problem, it is increasing most in highly developed countries [5].

Many different risk factors (e.g., hepatitis B virus [HBV], hepatitis C virus [HCV], afla-toxins, alcohol, diabetes and nonalcoholic fatty liver diseases) have been implicated in the pathogenesis of HCC. For many of them (e.g., HBV), a direct carcinogenic role has been often suggested (reviewed in [6]). However, common to all these conditions is the presence of underlying liver disease, characterized by prolonged periods of inflammation, necrosis and regeneration, conditions likely to be able to initiate and then maintain the carcinogenetic transformation in the liver. Indeed, while the risk of developing HCC per year in patients with cirrhosis ranges from 2 to 8%, depending on the different etiologies of cirrhosis, development of HCC in a healthy liver is exceedingly rare [7 –9].

The other epidemiologic characteristic constantly found worldwide in HCC is the striking prevalence in males, with a male:female ratio averaging 2:1 to 7:1, the latter proportion being more often found in HBV-positive cases [10,11]. The unbalanced relationship with gender is not limited to incidence; prognosis is also constantly better in females, both in term of spontaneous survival [12] and survival after resection [13,14]. Menopause attenuates these advantages [15].

The purpose of this review is to provide an overview of the available data on the role of gender dimorphism in the development of HCC, of the relationship with the different etiologic factors associated with the underlying liver disease and of the general and hormonal mechanisms potentially implicated in its development.

Epidemiology of liver cancer

In most countries, 75–90% of primary liver cancers are HCCs. Cholangiocarcinoma (CC), although rising in incidence in the last two decades, does not account for more than 5% of primary liver cancers [16]. Consequently, trends in liver cancer incidence and mortality tend to reflect trends in HCC incidence and mortality.

Worldwide distribution

The distribution throughout the world is uneven, approximately reflecting, at least for the high-incidence areas, the incidence of the etiologic factors of chronic liver disease. Overall, HBV and HCV infections are causally associated with HCC in more than 80% of cases [17]. High-rate areas for HCC are the countries in eastern Asia and sub-Saharan Africa (which are also hyperendemic for HBV infection). Northern America, northern Europe and Australia are low-rate areas, while central and southern Europe has intermediate HCC incidence.

A relevant finding in the last two decades is the opposite trend of incidence in countries having high or low HCC rate. High-rate areas are observing a substantial and sustained decline: for countries hyperendemic for HBV infection, such as China or Taiwan, implementation of nationwide campaigns for diet modification in order to avoid AFB1-contaminated food and widespread application of vaccination programs against HBV are likely explanations [18]. For countries such as Japan where HCV-related chronic liver disease (CLD) is most relevant, widespread use of effective antiviral therapies with known anti-inflammatory and suggested antitumoral action (IFN) may have been causal in determining decrease in incidence [5]. More surprising is the trend towards increase, which is observed in low-rate countries. In the UK and France, moderate increases in incidence were reported in the last decade [19,20]; however, the incidence in the USA, in the same period, has more than doubled, the age-adjusted incidence having risen from 1.4/100,000 in 1975–1977 to 3.0/100,000 in 1996–1998 [16].

Risk factors

Most of HCCs are associated with cirrhosis. The etiologic factors causing CLD (chronic infection with HBV or HCV, consumption of alcohol, hemochromatosis and nonalcoholic steatohepatitis [NASH]) may therefore be considered risk factors for HCC.

In high-rate HCC areas, HBV (and, possibly, associated AFB1 contamination of food) is the dominant factor, whereas HCV and alcohol are more important factors in low- to medium-rate areas (although NASH and associated conditions are continuously increasing in importance).

HBV

For HBV, a direct carcinogenic role has often been proposed on the basis of the interaction of HBV with the host genome; some of the proteins encoded by HBV (namely, preS/S and HBx) are thought to be able to determine critical modifications of the biochemical pathways that permit extended survival, resistance to immune-mediated apoptosis and growth advantage to modified cells [6]. Indeed, both case-control [21] and cohort studies [22] from Asia and Africa have shown a substantial increase in the relative risk for HCC, often further aggravated by the wide spread contamination of grains by aflatoxin. However, a long-term study of a cohort of HBsAg-positive blood donors followed for more than 30 years has shown that, in the absence of additional risk for chronic liver disease such as alcohol abuse [23], infection per se is not associated with an increased risk of HCC development, once more emphasizing the relevance of liver inflammation to favor carcinogenesis [24].

HCV

The same considerations hold true for HCV: meta-analysis of 21 case-control studies has shown that in HCV-infected patients, HCC risk is increased 17-fold compared with HCV-negative controls, but epidemiological data indicate that HCV increases HCC risk by promoting fibrosis and, eventually, cirrhosis. Once cirrhosis is established, HCC develops at an annual rate of 1–4%, although rates of up to 7% have been reported in Japan. Relevant support, although indirect, to the pre-eminent role of cirrhosis comes from a series of studies showing that antiviral therapy with IFN has been able to decrease the rate of development of HCC in sustained responders, not only when started in the hepatitis stage but also when cirrhosis was already established [25]. It is also noteworthy that rate of response to IFN and likelihood of developing more severe fibrosis are characterized by marked gender differences, with males and postmenopausal women being more often and more severely affected [26].

Alcohol

Heavy alcohol intake (i.e., more than 60–70 g/day) is associated with a high risk of development of cirrhosis and, eventually, HCC. There is strong indication of a synergistic role of heavy alcohol intake with HBV and HCV in increasing HCC risk [27]. It has long been well known that alcohol intake, even in quantities not exceeding 60 g, is able to determine significant liver damage in HBsAg-positive carriers otherwise lacking any sign of significant liver disease [23]. In addition, for anti-HCV-positive subjects, many studies have shown that a daily alcohol intake greater than 50 g is capable of determining more severe histological activity, increased rate of liver fibrosis progression and higher risk of liver cirrhosis and HCC (reviewed in [27]). However, no convincing evidence exists of a direct carcinogenic effect of alcohol.

Nonalcoholic fatty liver disease

In a substantial proportion of patients with HCC, approximately 30–40%, a definite etiologic factor is lacking (cryptogenic CLD). However, many of these cases fulfill the criteria for diagnosis of nonalcoholic fatty liver disease or for its more severe form (NASH). Prospective, unambiguous data exist regarding increased HCC risk in obesity and diabetes, two conditions that are strongly associated with NASH [28,29].

Gender differences

The unbalanced ratio between males and females in HCC is a constant finding, present in practically all reported series in the scientific literature. The male:female ratio usually ranges between 3:1 and 5:1, but in selected series [30] or in HBV patients, ratios between 7:1 and 9:1 are not unusual [10,30]. These ratios are not substantially different in different ethnic groups and are not influenced by emigration in other countries.

The marked difference between males and females is also maintained when considering death rates due to HCC [4]. However, Shimizu et al. have shown that after menopause, the male:female ratio becomes less unbalanced as a consequence of the increased number of HCC cases occurring in older females [31].

From an epidemiological point of view, a tentative explanation for these findings could be that sex-specific differences in exposure to risk factors such as alcohol intake, smoke and increased iron stores might be relevant. For example, for HBV infection, it has been suggested that response of parents to HBV infection may affect the sex ratio at birth of their offspring, resulting in a higher proportion of males among HBV carriers [32]. Similarly, local traditional practices may explain areas with very high malecarrier rate, as males become instrumental in the transmission of HBV [33].

However, other elements suggest that a broader approach than simple evaluation of exposure to risk factors is needed. During progression of liver disease of different etiologies, from chronic active hepatitis to cirrhosis and HCC, the male:female ratio becomes progressively more unbalanced, suggesting higher male susceptibility to liver damage [10,26]. During the natural history of HCV infection, males are also less likely to respond to IFN therapy, have a significantly higher likelihood to progress to significant liver fibrosis independently from age and alcohol intake (although alcohol further amplifies the risk), and have higher risk of early decompensation [26]. All of these clearly indicate that males have a greater risk of prolonged periods of hepatitic activity and have more rapid progression through all stages of chronic liver disease.

The pathophysiologic basis for these events will be analyzed later in this chapter.

Mechanisms of estrogens' action

General overview

Estrogens are steroid hormones that regulate growth, differentiation and function in several different target tissues in the human body. Classical target tissues for estrogens include breast, vaginal, vulvar, endometrial and uterine tissues; more recently, bone, brain and cardiovascular tissue have been reported to be target tissues for estrogen.

Estrogen-stimulated growth requires two estrogen receptors (ER), α and β, which are members of the superfamily of nuclear receptors. These receptors act as ligand-activated transcription factors. Target tissues with reproductive functions display the highest amount of ERs; in addition, nonclassical target tissues, among which is the liver (see next section), have a variable quantity of ERs. In the male organism, ERs can be detected mostly in the epididymis and the prostate.

The classical mechanism of ER action involves estrogen binding to receptors in the nucleus, after which the receptors dimerize and bind to specific response elements (estrogen response elements [EREs]) located in the promoters of target genes (Box 1) [34]. Regulation of ER expression is tissue specific and can be influenced by several factors, such as the type of coactivator or corepressor present in the tissue examined [35]. Binding of the hormone induces a conformational change within the ligand-binding domain (LBD) of the receptors, and this conformational change allows recruitment of coactivator proteins. The conformation of the ER differs with different ligands, such that with tamoxifen the interaction with some of these coactivators does not occur (leading to antagonism) but with others it persists (leading to agonism).

ERs can also regulate gene expression without directly binding to DNA (Box 1) [36]. This may occur through:

Protein-protein interactions with other DNA-binding transcription factors in the nucleus after ligand activation (gene regulation is affected by indirect DNA binding);

Membrane-associated ERs mediate nongenomic actions of estrogens, which can lead to both altered functions of proteins in the cytoplasm and altered gene expression. In this mechanism, the ligand activates a receptor, possibly associated with the membrane (either a classical ER or a distinct receptor), starting a signaling cascade via second messengers (SM) that affect ion channels or increase nitric oxide levels in the cytoplasm, leading to rapid physiological responses without involving gene regulation. Very recently, GPR30 (a member of the G-protein-coupled receptor family of the heptahelical transmembrane receptors) has been shown to be a distinct intracellular transmembrane ER, contributing to normal estrogen physiology as well as pathophysiology [37];

The ligand-independent pathway (nongenomic-to-genomic signaling) includes activation through other signaling pathways, such as growth factor signaling: activated kinases phosphorylate ERs and thereby activate them to dimerize, bind DNA and, finally, regulate genes.

Box 1.

Estrogen receptor signaling mechanisms.

Classical mechanism of estrogen receptor actions

Involves estrogen binding to receptors in the nucleus, after which the receptors dimerize and bind to specific response elements, known as estrogen response elements (EREs), located in the promoters of target genes.

Estrogen response element-independent genomic actions

Nuclear estradiol-estrogen receptor (ER) complexes associate through protein-protein interactions to a transcription-factor complex that contacts the target gene promoter.

Ligand-independent genomic actions

Growth factors activate protein-kinase cascades, leading to phosphorylation and activation of nuclear ERs at EREs.

Nongenomic actions

Membrane estradiol-ER complexes activate protein-kinase cascades, determining altered functions of proteins in the cytoplasm or regulation of gene expression through phosphorylation and activation of a transcription factor.

Estrogens, ERs & the liver

It has long been known that the liver is sensitive to the action of estrogens, to which it responds by increasing the synthesis and secretion of several glycoproteins, such as ceruloplasmin, corticosteroid-binding globulin, thyroid-binding globulin and testosterone-estradiol-binding globulin [38]. All types of cells in the liver respond to the action of estrogens (Figure 1).

Figure 1.

Estrogens influence all the different types of cells present in the liver (i.e. hepatocytes, endothelial cells, hepatic stellate cells and Kupffer cells).

ERα of normal mammalian liver has been extensively characterized: hepatic ERα is a high-affinity, low-capacity estrogen binder similar to those found in classical target tissues [39,40]. ERα content of normal liver is low (5–15 fmol/mg cytosol protein by ligand-binding assays) and this has prevented immunohistochemical demonstration by conventional techniques. With special fixation and embedding protocols and high anti-ER-antibody concentration, Ciocca et al. have been able to demonstrate that up to 40% of the hepatocytes in normal liver express ER [41].

Higher levels (although still below the levels found in target tissues) have been reported in male subjects abusing alcohol; this has been interpreted as a possible mechanism for the biochemical feminization observed in alcoholic chronic liver disease [42]. However, this finding has been disputed by other authors using a different method of detection of ERs and studying patients with far more advanced disease [43].

Regarding ERβ expression, contrasting results are reported in normal or neoplastic liver [44 –47]. Alvaro et al. did demonstrate ERβ in biliocytes in several pathological conditions characterized by reactive or neoplastic cholangiocyte proliferation, suggesting a role for estrogens and their receptors in modulating the proliferative activities of cholangiocytes [44]. However, they did not show evidence of ERβ in hepatocytes. Zhou et al., using improved methodology, did show ERβ both in hepatocytes and in hepatic stellate cells (HSC) [45], and suggested a possible inhibitory effect of ERβ on the proliferative activity of HSC in the presence of hepatic damage.

The recently described membrane-located G-coupled protein receptor 30 (GPR30), a non-classic ER involved in rapid nongenomic estrogen-mediated signaling through direct binding of 17β-estradiol [37], has been demonstrated to play a relevant role in mediating nongenomic effects of estrogens in reducing liver damage after trauma hemorrhage [48].

From a regulatory point of view, all molecular pathways described previously are also active in the liver [35 –37].

Gender dimorphism of hepatic metabolism

The liver responds to various types of stress in a gender-related manner [49,50]. Although the precise mechanism responsible for the gender dimorphism of hepatic metabolism remains to be determined, the level of several different circulating sex hormones, the differential gender-dependent expression of liver sex-hormone receptors and the pattern of growth-hormone secretion may be involved.

Although the most straightforward mechanism, it is unlikely that metabolic gender dimorphism depends on the different sex-hormone receptors, as no substantial difference has been demonstrated between male and female livers regarding levels of expression or localization of either ER receptor [46]. Therefore, although ERs are important in mediating estrogen action, gender dimorphism is probably influenced by other regulatory mechanisms [51,52]. With regards to this, there is increasing evidence for an extremely relevant role played by growth hormone (GH). In rodents and, to a lesser extent, in humans, males and females greatly differ in the temporal pattern of GH secretion: these different secretory patterns lead to marked sex differences in liver-gene expression [52 –54]. It has been experimentally shown that pulsatile GH infusion (characteristic of male secretory pattern) is sufficient to masculinize several biochemical pathways, while continuous infusion is sufficient to feminize them. Most striking are sex differences for genes coding for cytochrome P450 enzymes; although more relevant in rats and mice (in which up to 500-fold male/female differences are present), they are also seen to a smaller degree in humans, where they are an important determinant of the sex dependence of hepatic drug and steroid metabolism [55]. The possible implications for liver carcinogenesis will be discussed in the next section.

Estrogens, ERs & hepatic carcinogenesis

Experimental liver carcinogenesis

Early experimental evidence in hamster, rat and mouse models indicates that synthetic estrogens (such as estradiol benzoate or 17α-ethinyestradiol [EE₂]) may promote liver carcinogenesis after initiation with diethylnitrosamine [56 –58]. The same synthetic substances or the natural estrogens, when administered for long periods of time in the absence of an initiating agent, fail to increase tumor formation, thus suggesting that estrogens, in these experimental models, act primarily as promoters, possibly through stimulation of receptor-mediated cell-proliferative responses, although they are also able to favor the formation of free radical-mediated DNA and RNA adducts that are potentially mutagenic [57]. An increasing ER concentration was shown to be associated with tumor promotion, suggesting that the clonal expansion of preneoplastic liver cells requires ERs. Furthermore, EE₂ was shown to increase EGFR number and binding of EGF (known to be a potent mitogen for liver), which suggests that this is one of the mechanisms by which EE₂ may contribute to clonal expansion of initiated cells [59 –61].

A finding in experimental carcinogenesis in mice, which is also of extreme interest for human carcinogenesis, is the higher susceptibility of almost all strains of male mice both to spontaneous or carcinogen-induced carcinogenesis [62 –66]. This has been attributed to opposite effects of sex hormones, testosterone increasing the risk of tumor development while estrogens have the opposite action [67,68].

Later studies have shown that modulation of susceptibility to develop HCC involves a complex net of players, of which estrogens are one. GH is certainly one of them: several studies have demonstrated that GH overexpression favors development of liver tumor [69], while the absence of GH eliminates the modifications of susceptibility determined by sex hormones and/or genetic background [70]. Similar effect has been obtained by ablating IL-6 [71]. In this paper, it has been shown that diethylnitrosamine administration was able to cause greater increases in serum IL-6 concentration in males than in females and that ablation of IL-6 abolished the gender differences in hepatocarcinogenesis in mice. Of notable interest is the fact that the marked decrease of HCC incidence obtained by IL-6 ablation was seen only in IL-6 knockout (^-/-) males but not in IL-6^−/− females. The authors therefore suggested this to be a relevant mechanism to better explain the constant male prevalence of HCC [71]. However, the authors themselves demonstrated that estrogen inhibited secretion of IL-6 from Kupffer cells exposed to necrotic hepatocytes and reduced circulating concentrations of IL-6 in diethylnitrosamine-treated male mice. Therefore, it could be hypothesized that the relevant events in liver carcinogenesis are, indeed, directly dependent on the action of estrogens.

Human liver carcinogenesis

Benign liver tumors

In humans, the chronic use of estrogens has been associated in the past with increased risk of developing liver tumors such as benign nodular hyperplasia and hepatocellular adenoma (HCA). HCA is characterized by benign proliferation of hepatocytes, which may be steatotic or show glycogen storage, in an otherwise normal liver. Mitosis or cholangiolar proliferation are rare. In HCA, at least three different molecular pathways (HNF1α inactivation, β-catenin and inflammatory activation) have been recognized. HCAs with inflammatory characteristics are found mostly in patients with a high BMI and excessive alcohol consumption, suggesting that the latter may have a direct initiating role. In women with HCA associated with oral contraceptives use, all three molecular subtypes are equally represented [72]. While a few case reports of malignant transformation of HCA in HCC have been described [73], a recent metaanalysis of observational epidemiological studies, which examined the association between oral contraceptives and HCC, showed no conclusive relationship between oral contraceptives and risk of HCC [74].

Hepatocellular carcinoma

Far more complex is the situation regarding spontaneous (or disease-related) modifications of sex-hormone balance and development of HCC. It has long been known that patients (especially males) with chronic liver diseases have marked alterations of the sex-hormone balance, resulting in a ‘feminized’ phenotypic appearance [75 –77]. Serum estradiol:testosterone ratio is higher in individuals with HCC and cirrhosis than in normal individuals or individuals with cirrhosis alone [42,75]. The mechanisms underlying these alterations are diverse and can be attributed to altered hormone metabolism due to chronic liver disease, failure of the hypothalamus/pituitary/gonadal axis and direct effect on gonads by toxic agents (i.e., alcohol).

ERs (especially ERα but, to a lesser extent, also ERβ) have also been extensively studied in CLD. Early results for ERα were extremely variable, depending on the heterogeneity of the methods used and of the samples tested. When studying CLD during its progression, a trend towards a slight increase in ERα capacity was demonstrated [42,47]. Investigation of ERβ in CLD and in HCC has shown the presence of all possible combinations of ERα and ERβ without a clearcut association of type of association and clinical or pathological characteristics of the tumor [47].

Although ERs in HCC are evidently functional, their specific role in promoting or maintaining carcinogenesis is still not clear, as a simple increase in their levels may not be sufficient to determine relevant biologic consequences. The scarce results of classical antiestrogenic drugs such as tamoxifen in the palliative treatment of inoperable HCC, reviewed in a recent meta-analysis [78], raise further uncertainties regarding the significance of hepatic ERs. Even the evaluation in the liver of ERs by immunohistochemistry as in the study by Liu et al. [79], was not useful for the interpretation of results. The presence of hepatic ER receptors did not relate with treatment failure, course of disease or survival, perhaps as the method used was too insensitive to pick up low-level hepatic ERs.

In breast cancer, the identification of ER-/PgR+ tumors suggested the existence of ERs that are defective in estrogen binding but still functional in stimulating estrogen-responsive genes such as those coding for progesterone. This variant ERα (ERα δ5) is derived from an exon 5-deleted transcript giving rise to a protein lacking the hormone-binding domain but able to constitutively activate transcription of a normally estrogen-dependent gene construct in yeast cells [80]. This variant has been associated, in breast, with the development of an aggressive pathological phenotype and with the progression from hormone dependence to hormone independence [80 –82]. Its presence in the liver has been demonstrated both in cirrhotic and tumoral tissue; in some tumors, it predominates and sometimes becomes the only form expressed, especially in males [83]. In subjects with ERα δ5, growth rate of HCC and occurrence of multinodular HCC is significantly higher [47], and spontaneous survival significantly worse, than in patients with tumors expressing the wild-type form (wtERα) [84]. It is of special interest that ERα δ5 is often predominant in HBV-infected patients [83]. Starting from this finding, a recent study by Han et al. has thoroughly investigated the interaction between ERα, ERα δ5 and HBV (namely HBx, its putative oncogenic protein) [85]. ERα δ5 has been shown to exert a dominant negative activity in hepatoma cells when expressed together with wtERα. Furthermore, HBx decreases ERα transcriptional activity and this inhibition is specific, as HBx deletion mutant that lacks the ERα-binding site abolishes HBx repression of ERα. On the whole, HBx and ERαδ5 have an additive effect on inhibition of ERα signaling, possibly through recruitment of histone deacetylase 1 (HDAC1). The exact understanding of these mechanisms could also have therapeutic implications as HDAC inhibitors such as trichostatin have potent antitumor activity in vitro and in vivo [86].

The presence of ERα δ5 also has prognostic [87] and therapeutic implications [88]; certainly, its presence identifies a subgroup of HCC with extremely rapid progression of disease.

Conclusion

It is not easy to reconcile the different aspects of experimental and human carcinogenesis with the role of estrogens in a coordinate hypothesis: there are, in fact, elements that seem to support a role in favoring carcinogenesis and others that seem to suggest the opposite.

In the experimental setting, estrogens, when administered prior to the initiating event, are mostly protective against HCC development. To acquire promoting properties, they must be administered after initiation. In humans, it is certainly more difficult to evaluate their role and to demonstrate that they can be protective against HCC development, as it is almost impossible to define the timing of the initiating event, as it is usually performed in the experimental setting. However, epidemiological data in humans constantly point towards the favorable role played by female sex: lower HCC incidence, better response to treatment and better survival, at least during fertile age.

Recent experimental data have suggested different gender-related explanations for the most puzzling characteristic of HCC (i.e., male susceptibility). These studies suggested that male susceptibility may be attributed to molecular pathways (such as IL-6/Myd88 or GH), which are regulated according to a strict gender dimorphism [70,71]. However, going into the details of these studies, it is clear that the same results, obtained by knocking down these pathways, are obtained by eliminating or administrating estrogens. Therefore, the critical factor in defining gender susceptibility to HCC seems to be gender.

To be born a female seems to be the key to protection: females live between the ages of 10 and 50 years in an estrogen-permeated environment. Therefore, it is likely that when they come into contact with the possible initiating factor, they are in a less-susceptible condition than males. It is not the case that HCC incidence in females increases sharply some time after menopause. If these considerations are proved true, it will be more difficult to use estrogens as chemopreventive agents in males, as it will be extremely difficult to identify the exact timing of administration, as the encounter with the possible risk factor will be difficult to guess.

Future perspective

It is increasingly clear that inflammation plays a relevant role in the pathogenesis of liver cancer. The main risk factors for CLD (HBV, HCV and alcohol), giving rise to a state of continuous inflammation, play a relevant role in increasing HCC risk. Therefore, it will be extremely relevant to understand therapies that, like IFN (which controls the inflammatory state) or nucleoside analogues (which suppress viral replication), will be able to substantially decrease the risk of HCC development by interfering with the hepatitic process, possibly through an estrogen-mediated downregulation of the highly activated IL-6/Stat3 pathway.

Deeper investigation of the relationship between HBV and ER signaling is desirable: the demonstrated interaction between HBx, ERα and ERα δ5 suggests the opportunity of exploring ERα agonists (also as substances present in the diet, such as the phytoestrogens) as chemopreventive or therapeutic agents. Restoration of the transcriptional activity of ERα, inhibited by HBx, and trichostatin A, a specific inhibitor of HDAC enzyme, suggests additional therapeutic opportunities, as HDAC inhibitors have already been used to inhibit cancer cell growth in vitro and in vivo.

Executive summary

Epidemiology

Liver cancer is the fifth most common cancer worldwide and despite ultrasonographic surveillance strategies, its incidence is rising, especially in western countries.

Males are predominantly affected worldwide: they also have worse response to surgery and lower survival.

Cirrhosis (mostly due to hepatitis viruses and/or alcohol) is the common pathological background.

Experimental studies

Almost all strains of male mice display higher susceptibility to spontaneous or carcinogen-induced carcinogenesis.

Estrogen administration may promote carcinogenesis but only after previous initiation; without initiation, they fail to increase tumor formation.

Human studies

Sex-hormone balance is altered in patients with chronic liver diseases (especially in males).

Estrogens receptors (ER; both α and β) show slight modifications during progression of chronic liver disease, but it is not easy to relate their status to risk of development of hepatocellular carcinoma (HCC).

Presence of ERα δ5 characterizes a subgroup of HCC with extremely rapid progression of neoplastic disease. ERα δ5 interacts with the putative oncogenic protein of HBV.

Conclusion

Overall, experimental and human data indicate that estrogens may play a relevant role in protecting females from developing HCC.

Footnotes

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

References

Papers of special note have been highlighted as either of interest (•) or of considerable interest (••) to readers.

Sherman

: Screening for hepatocellular carcinoma. Baillieres Best Pract. Res. Clin. Gastroenterol. 13, 623–635 (1999).

Chalasani

Horlander

Sr Said

: Screening for hepatocellular carcinoma in patients with advanced cirrhosis. Am. J. Gastroenterol. 94, 2988–2993 (1999).

Liu

Chen

Asch

: Surgery for hepatocellular carcinoma: does it improve survival? Ann. Surg. Oncol. 11, 298–303 (2004).

Parkin

: Global cancer statistics in the year 2000. Lancet Oncol. 2, 533–543 (2000).

McGlynn

London

: Epidemiology and natural history of hepatocellular carcinoma. Best Pract. Res. Clin. Gastroent. 19, 3–23 (2005).

Cougot

Neuveut

Buendia

: HBV induced carcinogenesis. J. Clin. Virol. 34(Suppl. 1), S75–S78 (2005).

Concise but comprehensive review of hepatitis B virus (HBV)-related carcinogenesis.

International Agency for Research on Cancer: Cancer Incidence in Five Continents (Volume VII). IARC, Lyon, France (1997).

10.

Ikeda

Saitoh

Koida

: A multivariate analysis of risk factors for hepatocellular carcinogenesis: a prospective observation of 795 patients with viral and alcoholic cirrhosis. Hepatology 18, 47–53 (1993).

11.

El-Serag

Rudolph

: Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology 132, 2557–2576 (2007).

12.

Villa

Baldini

Pasquinelli

: Risk factors for hepatocellular carcinoma in Italy. Male sex, hepatitis B virus, non-A non-B infection, and alcohol. Cancer 62, 611–615 (1988).

13.

El-Serag

: Hepatocellular carcinoma and hepatitis C in the United States. Hepatology 36(Suppl. 1), S74–S83 (2002).

14.

Tangkijvanich

Mahachai

Suwangool

Poovorawan

: Gender difference in clinicopathologic features and survival of patients with hepatocellular carcinoma. World J. Gastroenterol. 10, 1547–1550 (2004).

15.

Fan

: Better survival in women with resected hepatocellular carcinoma is not related to tumor proliferation or expression of hormone receptors. Am. J. Gastroenterol. 92, 1355–1358 (1997).

16.

Fukuda

Itamoto

Amano

: Clinicopathologic features of hepatocellular carcinoma patients with compensated cirrhosis surviving more than 10 years after curative hepatectomy. World J. Surg. 31, 345–352 (2007).

17.

Shimizu

Ito

: Protection of estrogens against the progression of chronic liver disease. Hepatol. Res. 37, 239–247 (2007).

18.

El Serag

Davila

Petersen

: The continuing increase in the incidence of hepatocellular carcinoma in the United States: an update. Ann. Intern. Med. 139, 817–823 (2003).

19.

Pisani

Parkin

Munoz

Ferlay

: Cancer and infection: estimates of attributable fraction in 1990. Cancer Epidemiol. Biomarkers Prev. 6, 387–400 (1997).

20.

Chang

Chen

Lai

: Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. N. Engl. J. Med. 336, 1855–1859 (1997).

21.

Taylor-Robinson

Foster

Arora

Hargreaves

Thomas

: Increase in primary liver cancer in the UK, 1979–94. Lancet 350, 1142–1143 (1997).

22.

Deuffic

Poynard

Buffat

Valleron

: Trends in primary liver cancer. Lancet 351, 214–215 (1998).

23.

Zaman

Melia

Johnson

Portmann

Johnson

Williams

: Risk factors in development of hepatocellular carcinoma in cirrhosis: prospective study of 613 patients. Lancet 15, 1357–1360 (1985).

24.

Relevant study, among the first to show the relationship between chronic liver disease and development of hepatocellular carcinoma (HCC).

25.

Beasley

Hwang

Lin

Chien

: Hepatocellular carcinoma and hepatitis B virus. A prospective study of 22 707 men in Taiwan. Lancet 1129–1133 (1981).

26.

Seminal epidemiological study on the HBV/HCC relationship.

27.

Villa

Rubbiani

Barchi

: Susceptibility of chronic symptomless HBsAg carriers to ethanol-induced hepatic damage. Lancet 2, 1243–1244 (1982).

28.

First study to show the sinergistic effect between alcohol abuse and HBV infection.

29.

Manno

Camma

Schepis

: Natural history of chronic HBV carriers in Northern Italy: morbidity and mortality after 30 years. Gastroenterology 127, 756–763 (2004).

30.

Camma

Giunta

Andreone

Craxi

: Interferon and prevention of hepatocellular carcinoma in viral cirrhosis: an evidence-based approach. J. Hepatol. 34, 593–602 (2001).

31.

Poynard

Bedossa

Opolon

: Natural history of liver fibrosis progression in patients with chronic hepatitis C. Lancet 349, 825–832 (1997).

32.

Regev

Jeffers

: Hepatitis C and alcohol. Alcohol Clin. Exp. Res. 23, 1543–1551 (1999).

33.

Calle

Rodriguez

Walker-Thurmond

Thun

: Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N. Engl. J. Med. 348, 1625–1638. (2003).

34.

El Serag

Tran

Everhart

: Diabetes increases the risk of chronic liver disease and hepatocellular carcinoma. Gastroenterology 126, 460–468 (2004).

35.

Lerose

Molinari

Rocchi

Manenti

Villa

: Prognostic features and survival of hepatocellular carcinoma in Italy: impact of stage of disease. Eur. J. Cancer 37, 239–245 (2000).

36.

Shimizu

Inoue

Yano

: Estrogen receptor levels and lipid peroxidation in hepatocellular carcinoma with hepatitis C virus infection. Liver 21, 342–349 (2001).

37.

Blumberg

: Sex differences in response to hepatitis B virus. I. History. Arthritis Rheum. 22, 1261–1266 (1979).

38.

Langendorfer

Davenport

London

Blumberg

Mazzur

: Sex-related differences in transmission of hepatitis B infection in a Melanesian population. Am. J. Phys. Anthropol. 64, 243–254 (1984).

39.

Nilsson

Mäkelä

Treuter

: Mechanisms of estrogen action. Physiol. Rev. 81, 1535–1565 (2000).

40.

Diel

: Tissue-specific estrogenic response and molecular mechanisms. Toxicol. Lett. 127, 217–224 (2002).

41.

Bjornstrom

Sjoberg

: Mechanisms of estrogen receptor signaling: convergence of genomic and nongenomic actions on target genes. Mol. Endocrinol. 19, 833–842 (2005).

42.

Revankar

Cimino

Sklar

Arterburn

Prossnitz

: A transmembrane intracellular estrogen receptor mediates rapid cell signaling. Science 307, 1625–1630 (2005).

43.

Eagon

Porter

Francavilla

DiLeo

van Thiel

: Estrogen and androgen receptors in liver: their role in liver disease and regeneration. Semin. Liver Dis. 5, 59–69 (1985).

44.

Porter

Elm

Van Thiel

Eagon

: Characterization and quantitation of human hepatic estrogen receptor. Gastroenterology 84, 704–712 (1983).

45.

Rossini

Baldini

Villa

Manenti

: Characterization of estrogen receptor from human liver. Gastroenterology 96, 1102–1109 (1989).

46.

Ciocca

Jorge

: Estrogen receptors, progesterone receptors and heat-shock 27KD protein in liver biopsy specimens from patients with hepatitis B virus infection. Hepatology 13, 838–844 (1991).

47.

Villa

Baldini

Rossini

: Ethanol-induced increase in cytosolic estrogen receptors in human male liver: a possible explanation for biochemical feminization in chronic liver disease due to alcohol. Hepatology 8, 1610–1614 (1988).

48.

Becker

Andersen

Poulsen

Horn

: Variation in hepatic estrogen receptor concentrations in patients with liver disease. A multivariate analysis. Scand. J. Gastroenterol. 27, 355–361 (1992).

49.

Alvaro

Mancino

Onori

: Estrogens and the pathophysiology of the biliary tree. World J. Gastroenterol. 12, 3537–3545 (2006).

50.

Zhou

Shimizu

: Hepatic stellate cells contain the functional estrogen receptor beta but not the estrogen receptor alpha in male and female rats. Biochem. Biophys. Res. Commun. 286, 1059–1065 (2000).

51.

Wang

Lee

Kim

: The expression of estrogen receptors in hepatocellular carcinoma in Korean patients. Yonsei Med. J. 47, 811–816 (2000).

52.

Iavarone

Lampertico

Seletti

: The clinical and pathogenetic significance of estrogen receptor-β expression in chronic liver diseases and liver carcinoma. Cancer 98, 529–534 (2003).

53.

Hsieh

Frink

: G protein-coupled receptor 30-dependent protein kinase A pathway is critical in nongenomic effects of estrogen in attenuating liver injury after trauma-hemorrhage. Am. J. Pathol. 170, 1210–1218 (2007).

54.

Harada

Pavlick

Hines

: Selected contribution: effects of gender on reduced-size liver ischemia and reperfusion injury. J. Appl. Physiol. 91, 2816–2822 (2000).

55.

Colantoni

Idilman

de Maria

: Hepatic apoptosis and proliferation in male and female rats fed alcohol: role of cytokines. Alcohol Clin. Exp. Res. 27, 1184–1189 (2003).

56.

Yokoyama

Nimura

Nagino

Bland

Chaudry

: Current understanding of gender dimorphism in hepatic pathophysiology. J. Surg. Res. 128, 147–156 (2005).

57.

Mode

Gustafsson

Jansson

Eden

Isaksson

: Association between plasma level of growth hormone and sex differentiation of hepatic steroid metabolism in the rat. Endocrinology 111, 1692–1697 (1982).

58.

Wong

Kawajiri

Negishi

: Gene family of male-specific testosterone 16 α-hydroxylase (C-P-450(16) α) in mouse liver: cDNA sequences, neonatal imprinting, and reversible regulation by androgen. Biochemistry 26, 8683–8690 (1987).

59.

Noshiro

Negishi

: Pretranslational regulation of sex-dependent testosterone hydroxylases by growth hormone in mouse liver. J. Biol. Chem. 261, 15923–15927 (1986).

60.

Waxman

O'Connor

: Growth hormone regulation of sex-dependent liver gene expression. Mol. Endocrinol. 20, 2613–2629 (2006).

61.

Comprehensive and informative review of growth hormone (GH) signaling.

62.

Yager

Jr Yager

: Oral contraceptive steroids as promoters of hepatocarcinogenesis in female Sprague-Dawley rats. Cancer Res. 40, 3680–3685 (1980).

63.

Cameron

Imaida

Ito

: Promotive effects of ethinyl estradiol in hepato-carcinogenesis initiated by diethylnitrosamine in male rats. Gann 72, 339–340 (1981).

64.

Wanless

Medline

: Role of estrogens as promoters of hepatic neoplasia. Lab Invest. 46, 313–320 (1982).

65.

St Hilaire

Jones

: Epidermal growth factors: its biologic and metabolic effects with emphasis on the hepatocyte. Hepatology 2, 601–613 (1982).

66.

Eckl

Meyer

Whitcombe

Jirtle

: Phenobarbital reduces EGF receptors and the ability of physiological concentrations of calcium to suppress hepatocyte proliferation. Carcinogenesis (Lond.) 9, 479–483 (1988).

67.

Aem Nelson

Vickers K

McCoy

Lucier

: Changes in estrogen receptor, DNA ploidy, and estrogen metabolism in rat hepatocytes during a two-stage model for hepatocarcinogenesis using 17 α-ethinylestradiol as the promoting agent. Cancer Res. 49, 6512–6520 (1989).

68.

Andervont

: Studies on the occurrence of spontaneous hepatomas in mice of strains C3H and CBA. J. Natl Cancer Inst. 11, 581–592 (1950).

69.

Vesselinovitch

: The sex-dependent difference in the development of liver tumors in mice administered dimethylnitrosamine. Cancer Res. 29, 1024–1027 (1969).

70.

Kemp

Drinkwater

: Genetic variation in liver tumor susceptibility plasma testosterone levels, and androgen receptor binding in six inbred strains of mice. Cancer Res. 49, 5044–5047 (1989).

71.

Poole

Drinkwater

: Strain dependent effects of sex hormones on hepatocarcinogenesis in mice. Carcinogenesis 17, 191–196 (1996).

72.

Toh

: Effect of neonatal castration on liver tumor induction by N-2-fluorenylacetamide in suckling BALB/c mice. Carcinogenesis 2, 1219–1221 (1981).

73.

Yamamoto

Iishi

Tatsuta

Tsuji

Terada

: Roles of ovaries and testes in hepatocellular tumorigenesis induced in mice by 3′-methyl-4-dimethylaminoazobenzene. Int. J. Cancer 49, 83–88 (1991).

74.

Orian

Tamakoshi

Mackay

Brandon

: New murine model for hepatocellular carcinoma, transgenic mice expressing metallothionein-ovine growth hormone fusion gene. J. Natl Cancer Inst. 82, 393–398 (1990).

75.

Messing

Chen

Palmiter

Brinster

: Peripheral neuropathies, hepatocellular carcinomas and islet cell adenomas in transgenic mice. Nature 316, 461–463 (1985).

76.

Bugni

Poole

Drinkwater

: The little mutation suppresses DEN-induced hepatocarcinogenesis in mice and abrogates genetic and hormonal modulation of susceptibility. Carcinogenesis 22, 1853–1862 (2000).

77.

Naugler

Sakurai

Kim

: Gender disparity in liver cancer due to sex differences in MyD88-dependent IL-6 production. Science 317, 121–124 (2007).

78.

Rebouissou

Bioulac-Sage

Zucman-Rossi

: Molecular pathogenesis of focal nodular hyperplasia and hepatocellular adenoma. J. Hepatol. (2007) doi:10.1016/j.jhep. 2007.10.003 (Epub ahead of print).

79.

Ito

Sasaki

Wen

: Liver cell adenoma with malignant transformation: a case report. World J. Gastroenterol. 9, 2379–2381 (2003).

80.

Maheshwari

Sarraj

Kramer

El-Serag

: Oral contraception and the risk of hepatocellular carcinoma. J. Hepatol. 47, 506–513 (2007).

81.

Farinati

de Maria

Marafin

: Hepatocellular carcinoma in alcoholic cirrhosis: is sex hormone imbalance a pathogenetic factor. Eur. J. Gastroenterol. Hepatol. 7, 145–150 (1995).

82.

Nagasue

Ogawa

Yukaya

Ohta

Ito

: Serum levels of estrogens and testosterone in cirrhotic men with and without hepatocellular carcinoma. Gastroenterology 88, 768–772 (1985).

83.

Guechot

Peigney

Ballet

Vaubourdolle

Giboudeau

Poupon

: Sex hormone imbalance in male alcoholic cirrhotic patients with and without hepatocellular carcinoma. Cancer 62, 760–762 (1985).

84.

Nowak

Stockler

Chow

Findlay

: Use of tamoxifen in advanced-stage hepatocellular carcinoma. A systematic review. Cancer 103, 1408–1414 (2005).

85.

Accurate meta-analysis of largest trials using tamoxifen for palliative treatment of inoperable HCC.

86.

Liu

Fan

Poon

Wong

: Treatment of advanced hepatocellular carcinoma with tamoxifen and the correlation with expression of hormone receptors: a prospective randomized study. Am. J. Gastroenterol. 95, 218–222 (2000).

87.

Fuqua

Fitzgerald

Chamness

: Variant human breast tumor estrogen receptor with constitutive transcriptional activity Cancer Res. 51, 105–109 (1991).

88.

Villa

Dugani

Fantoni

: Type of estrogen receptor determines response to antiestrogen therapy. Cancer Res. 56, 3883–3885 (1996).

89.

Villa

Camellini

Dugani

: Variant estrogen receptor messenger RNA species detected in human primary hepatocellular carcinoma. Cancer Res. 55, 498–500 (1995).

90.

Villa

Dugani

Moles

Camellini

: Variant liver estrogen receptor transcripts already occur at an early stage of chronic liver disease. Hepatology 27, 983–988 (1988).

91.

Villa

Moles

Ferretti

: Natural history of inoperable hepatocellular carcinoma: estrogen receptors' status in the tumor is the strongest prognostic factor for survival. Hepatology 32, 233–238 (2000).

92.

Han

Ding

Yuan

: Hepatitis B virus X protein and the estrogen receptor variant lacking exon 5 inhibit estrogen receptor signaling in hepatoma cells. Nucleic Acids Res. 34, 3095–3106 (2006).

93.

Extremely interesting as it links estrogen signaling with HBV infections.

94.

Ganslmayer

Ocker

Zopf

: A quadruple therapy synergistically blocks proliferation and promotes apoptosis of hepatoma cells. Oncol. Rep. 11, 943–950 (2004).

95.

Villa

Colantoni

Camma

: Estrogen receptor classification for hepatocellular carcinoma: comparison with clinical staging systems. J. Clin. Oncol. 21, 441–446 (2003).

96.

Villa

Ferretti

Grottola

: Hormonal therapy with megestrol in inoperable hepatocellular carcinoma characterized by variant oestrogen receptors. Br. J. Cancer. 84, 881–885 (2001).