Download Sex Hormones and Risk of Liver Tumor

Sex Hormones and Risk of Liver Tumor
L. GIANNITRAPANI,a M. SORESI,a E. LA SPADA,a M. CERVELLO,b
N. D’ALESSANDRO,c AND G. MONTALTOa
a Dipartimento di Medicina Clinica e Patologie Emergenti, Università di
Palermo, Palermo, Italy
b Istituto
di Biomedicina e Immunologia Molecolare “A. Monroy,” C.N.R.,
Palermo, Italy
c Dipartimento
di Scienze Farmacologiche, Universita di Palermo, Palermo, Italy
ABSTRACT: The liver is morphologically and functionally modulated by
sex hormones. Long-term use of oral contraceptives (OCs) and anabolic
androgenic steroids (AASs) can induce both benign (hemangioma, adenoma, and focal nodular hyperplasia [FNH]) and malignant (hepatocellular carcinoma [HCC]) hepatocellular tumors. Hepatic adenomas
(HAs) are rare, benign neoplasms usually occurring in young women,
the development and the complications of which have been related to the
strength of OCs and the duration of their use. HA incidence has fallen
since the introduction of pills containing smaller amounts of estrogens.
FNH is a benign lesion, most commonly seen in young women, which is
thought to represent a local hyperplastic response of hepatocytes to a vascular abnormality. Because of the female predominance and the young
age at onset, a role of female hormones has been suggested. Furthermore,
a large proportion of women with FNH (50–75%) are OC users. Liver
hemangiomas (LHs) are the most common benign liver tumors and are
seen more commonly in young adult females. The female predilection
and clinical observations of LH growth under conditions of estrogenic
exposure suggest a possible role for estrogen in the pathogenesis of LHs.
HCC has become one of the most widespread tumors in the world in
recent years, representing the sixth leading cancer and the third most
common cause of death from cancer. Apart from liver cirrhosis, numerous other factors responsible for its onset have been proposed: hepatitis
infections from virus B (HBV) and C (HCV), alcohol, smoking, and aflatoxin. However, regardless of etiology, chronic liver diseases progress at
unequal rates in the two sexes, with the major sequelae, such as cirrhosis
and HCC, being more frequent in men than in women. These epidemiological data have prompted researchers to investigate the relationship
between sex hormones and liver tumors. The human liver expresses estrogen and androgen receptors and experimentally both androgens and
Address for correspondence: Prof. Giuseppe Montalto, Ordinario di Medicina Interna, Policlinico
Universitario di Palermo, via del Vespro, 141, Palermo, Italy. Voice: +39-0916552991; fax: +390916552847.
e-mail: [email protected]
C 2006 New York Academy of Sciences.
Ann. N.Y. Acad. Sci. 1089: 228–236 (2006). doi: 10.1196/annals.1386.044
228
GIANNITRAPANI et al.: SEX HORMONES AND RISK OF LIVER TUMOR
229
estrogens have been implicated in stimulating hepatocyte proliferation
and may act as liver tumor inducers or promoters.
KEYWORDS: benign
aromatase
liver
tumors;
HCC;
estrogens;
androgens;
INTRODUCTION
The liver is a hormone-sensitive organ, and in fact both normal liver and
hepatocellular carcinoma (HCC) tissues from male and female mammals have
been shown to express specific estrogen receptors (ERs). Experimentally, estrogens may act as liver tumor inducers or promoters in vivo,1,2 and are involved
in stimulating hepatocyte proliferation in vitro.3 Moreover, anti-estrogens like
tamoxifen have been shown to reduce levels of ERs and to inhibit hepatocyte
proliferation following partial hepatectomy.4
As regards the role of androgens, it has also been observed that androgen
receptors (ARs), specifically activated by testosterone, are present in normal
liver tissue from both males and females and that their expression is increased
in tumor tissue and in the surrounding liver of individuals with HCC.5
In addition, observations from clinical and epidemiological studies have
highlighted that the long-term use of OCs and anabolic androgenic steroids
(AASs) can induce benign and malignant hepatocellular tumors. Benign tumors of the liver are often discovered incidentally in asymptomatic individuals
during diagnostic imaging or exploratory laparotomy performed for other reasons. Hemangiomas are the most common benign liver tumors, followed in
prevalence by focal nodular hyperplasia (FNH) and the rarer condition of adenoma; their growth and development have been linked to hormonal stimulation.
However, although evidence from the literature concurs to a great extent on
the role of sex hormones in the development of benign liver tumors and in
particular liver adenoma, in the field of HCC this role is much more controversial. In fact, HCC usually occurs in individuals with chronic liver disease
with a clear disadvantage for the male sex, thus suggesting a possible causal
importance of androgens. Male cirrhotics who develop HCC, however, present
a characteristic imbalance with a relative hyperestrogenic state, so that a role
of estrogen in liver cancer has been hypothesized as well.
SEX HORMONES AND BENIGN LIVER TUMORS
Liver hemangiomas (LHs) are the most common benign liver neoplasms.
They are diagnosed more commonly in young adult females, with a female:male ratio of 5:1. In around 70% of patients they are multiple. Variants
include giant hemangiomas, which can occupy up to the entire hepatic lobe
and may expand the liver contour.
230
ANNALS NEW YORK ACADEMY OF SCIENCES
From an anatomic point of view LHs consist of large, well-defined bloodfilled spaces, lined by a single layer of endothelium and separated by fibrous
septae. Pathologically, they comprise vascular lakes and channels, some of
which can develop thrombosis and fibrosis.
Several case studies in the past have proposed that tumor growth may be
related to estrogens. This hypothesis comes from observations that estrogen
replacement therapy may play a role in the pathogenesis of recurrent LHs,6
that a prolonged administration of oral contraceptive (OCs) may facilitate the
growth of LHs,7 and that LHs can grow and become symptomatic during
pregnancy.8
However, despite these premises, in the last 2 years two studies have been
published yielding contradictory results.9,10 In a recent case–control study by
Gemer et al. the possible association of OCs with LHs was explored. Several
parameters, such as age, age at menarche, age at first pregnancy, number of
pregnancies, age at menopause and OC use, were compared in women with and
without LH and there was no significant difference between the two groups.
From these results the authors concluded that there was no association between
oral contraception, menstrual or reproductive history, and development of LH
and, as a consequence, that there were no indications for the withdrawal of oral
contraception in women with LH.9
On the contrary, in a study by Glinkova et al., the impact of female sex
hormones on the natural history of LH was prospectively evaluated, with the
conclusions that age at first period was inversely associated with the size of
LHs, age at menopause was positively correlated with the number of LHs, and
that hormone therapy increased the risk of LH enlargement.10
Other molecular studies have tended to explain the mechanisms by which
estrogens may regulate endothelial cell turnover, again with somewhat controversial results. In fact, it has been shown that estrogens can enhance endothelial cell proliferation, migration, and organization into capillary-like
structures in vitro and augment experimental angiogenesis in vivo.11 In
contrast, in vitro studies have suggested that certain steroids may inhibit
angiogenesis.12
FNH is a rare benign lesion which is seen more frequently in young adult
females, with a women:men ratio of 8:1. Although the risk factors for FNH are
largely unknown, a role for female hormones has been suggested in view of the
female predominance and the young age at onset. FNH forms as an unencapsulated mass, which consists of multiple pseudolobules around a central area of
fibrous tissue. The etiology of these lesions is unclear, but the histopathological
findings may be related to an underlying developmental abnormality with a hyperplastic response of the liver parenchyma and a disorganized growth pattern
of hepatocytes and ducts. They can be multiple; hemorrhage is exceedingly
rare; and they apparently have no malignant potential.
To obtain more information about the association of FNH and OC use a case–
control study was recently conducted by Scalori et al. in an area of northern
GIANNITRAPANI et al.: SEX HORMONES AND RISK OF LIVER TUMOR
231
Italy.13 In this study the distribution of cases and controls and the corresponding
OR according to OC use, duration of use, age at starting use, and time since
stopping use, were evaluated. This study provided definite and quantitative
evidence that OC use was significantly, although modestly associated with
FNH. The time–risk relation gave convincing support to the existence of a real
association, given that there was a direct trend in risk with duration and an
inverse trend with age at first use.13
However, in a previous study a few years ago, Mathieu et al. studied the
relationship between the number and size of FNH lesions in women divided
into five groups: no OC use; high-dose OC use; low-dose OC use; low-dose
and high-dose OC use; and pure progestagen use.14 They found no differences
between these groups as regards the number and size of the lesions. These data
showed that neither the size nor the number of FNH lesions were influenced
by OC use and that size changes during follow-up were rare and did not seem
to depend on OC use, so they concluded that low-dose OC can be maintained
in young women with FNH.14
Finally, hepatic adenomas (HAs) are uncommon benign neoplasms usually
occurring in young women. They are considered noncancerous lesions with
little clinical significance; however, even though rarely, they can become cancerous. Moreover, they can become large-sized during pregnancy, presumably
as a result of estrogen stimulation, and under these circumstances they can
rupture, resulting in acute bleeding and peritonitis. Liver cell adenomas are
usually well demarcated, but the capsule may not be clearly obvious. They are
large when detected (25–30 cm in diameter). The histological appearance is of
benign-looking hepatocytes often arranged in cords.
HA has been strongly associated with the use of OCs; in fact, it has been
calculated that about 320 new cases are diagnosed each year, mostly attributable
to OC use. This association was first suggested by Baum et al.,15 who in 1973
reported seven cases of HA, all related to OC, and has been supported by
many other subsequent publications.16,17 In women who have never used OCs
or who have used them for less than 24 months, HA develops at an annual rate
ranging from 1 to 1.3 per million in the age ranges of 16–30 and 31–44 years,
respectively.17 Its incidence, however, has fallen since the introduction of pills
containing smaller amounts of estrogens.
Consequently, in contrast with what happens for LH and FNH, at least for
HA there is an agreement among authors about the fact that the association
between OCs and HA is strong and depends on the duration of use. Furthermore, unresected lesions may decrease in size in young women once they stop
OC use. All these data taken together suggest that the association between HA
and OC use is one of cause and effect.
In recent times AASs have also been proven to be involved in the development of HA.18 Apparently, androgen-induced HAs are relatively rare. However, the possibility that an oral AAS can induce liver cell proliferation must
be taken into account and sportsmen taking AASs over a long period should
232
ANNALS NEW YORK ACADEMY OF SCIENCES
be considered a group at risk for developing hepatic sex hormone–related
tumors.
SEX HORMONES AND HCC
HCC has become one of the leading causes of death for cancer worldwide,
being the fifth most frequent neoplasia in the world, with 564,300 new cases in
2000 (5.6%) and more than 12,000 new cases per year in Italy.19,20 Its incidence
is now increasing all over the world with a variable geographical distribution
according to the spread of the main risk factors, namely hepatitis B and C
viruses (HBV, HCV) and liver cirrhosis (LC) whatever the cause. In fact, in
areas like India, Southeast Asia, and the developing countries in general, where
the prevalence of HBV infection is still high, the incidence of HCC is more
than 20 per 100,000 inhabitants.20 On the contrary, HCV predominates as a
cause of HCC in the developed countries, with an intermediate (Italy, Spain) to
low (USA) incidence of tumor.21,22 Other risk factors (Aflatoxin B1, alcohol,
hereditary diseases) may play a significant role in specific contexts. Finally,
among the other factors associated with an increased risk of development of
HCC regardless of the geographical setting, there is male gender. In most published series a striking predominance of the tumor in males has been described,
with a male-to-female ratio ranging from 2 to 11:1. Moreover, the prognosis
seems to be more benign in females than in males because women have a better
survival rate and a reduced recurrence of the disease after treatment.23
However, it has to be considered that in Western countries 80–90% of HCCs
develop in a liver with an underlying cirrhosis and this complicates the understanding of the role of sex hormones in liver carcinogenesis, especially because
male cirrhotics present a so-called “feminization” of their phenotype due to
a relatively hyperestrogenic condition. These premises make it necessary to
distinguish between HCC with and HCC without underlying LC to try to eliminate this confounding factor. In women without underlying LC there is a great
deal of evidence for the responsibility of OC use in the development of HCC.
In a recent meta-analysis of eight studies, which confirms previous data,24,25
it was stated that OC use increases the risk of HCC with an overall OR of 2.5
in ever- versus never-users of OCs and an overall OR of 5.8 for the longest
duration of use.26
Evidence for the role of sex hormones in the development of LC-correlated
HCC is much more controversial. LC determines an alteration in sex hormone
balance, which is more evident in males as the activity of ERs is increased
in the liver with an enhancement of its response to estrogens.27 Our group
has published a study in which it was observed that the serum estradiol-totestosterone ratio was higher in individuals with HCC and LC than in normal
individuals or individuals with LC alone.28 In contrast, two recent studies
from Korea and Japan suggested that elevated serum testosterone levels or an
GIANNITRAPANI et al.: SEX HORMONES AND RISK OF LIVER TUMOR
233
imbalanced testosterone–estradiol rate is associated with an increased risk of
HCC,29,30 but trials using different antiandrogenic compounds to treat or reduce
the progression of liver cancer have shown quite disappointing results, with
an almost complete lack of effect for this therapeutic approach.31,32 However,
studies also using antiestrogen drugs have yielded controversial results. In fact,
most of the studies published in the early 1990s reported reduced tumor growth
rates and prolonged survival in subjects treated with tamoxifen compared to
untreated controls, but they were based on small numbers of patients. Two
recent papers, a multicentric trial including 496 patients with HCC at any
stage and a trial with 119 patients with unresectable HCC, in which the patients
were randomized to receive tamoxifen or placebo, concluded that tamoxifen
was not effective in HCC treatment.33,34 A possible explanation for the failure
of the therapeutic approach with antiestrogen drugs can be found in some
experimental and clinical observations of the presence of a variant form of ER
(vER), deriving from an exon 5-deleted transcript, which lacks the hormonebinding domain, also in the liver.35 It has been observed that in HCC vER
largely predominates and sometimes becomes the only form expressed and
that it is also expressed by the peritumoral cirrhotic tissues of patients with
HCC, especially males.36 Moreover, the growth rate of HCC in subjects with
vER is significantly higher than in patients with tumors expressing the wildtype form (wtER), and the spontaneous survival in patients with wtER is
exceedingly better than in patients with HCC characterized by vER.36 Another
explanation could be that estrogen has been described as exerting a possible
role in the growth regulation of both normal and cancer human liver cells by
alternative, nonreceptorial mechanisms.37 Finally, our group recently published
a study in which the activity and expression of aromatase enzyme (the one that
converts androgens into estrogens) was investigated in nontumoral, cirrhotic,
and malignant human liver tissues and cells. Our observations were that human
HCC tissues showed elevated aromatase activity, with consequently higher
estrogen formation rates than in nontumoral liver tissues. If it can be assumed
that estrogen plays a role in hepatoma cell growth via nonreceptor pathways, a
strategy reducing estrogen concentration in the tumor with the use of aromatase
inhibitors could be attempted.38
CONCLUSIONS
Benign liver neoplasms, such as LH and FNH, whose growth and development have been variably linked to sex hormones, have a benign course and can
be managed conservatively; only HA must be followed up more carefully on
account of its potential malignancy.
As regards the role of estrogens in HCC, it seems that in the physiological
status of premenopausal women, in the absence of other risk factors for liver
disease, they have a somewhat protective role against the development of HCC.
234
ANNALS NEW YORK ACADEMY OF SCIENCES
On the contrary, the hyperestrogenic status of the cirrhotic male or the high
concentrations of estrogens in the old formulations of OCs, together with the
presence of vER in the liver and/or other risk factors, may increase the risk of
developing HCC.
Finally, if the evidence of a local estrogen formation from androgens is
confirmed to have a role in the development and progression of human HCC,
this may provide a basis to improve endocrine treatment of HCC patients using
antiaromatase drugs.
REFERENCES
1. REZNIK-SCHULLER, H. 1979. Carcinogenic effects of diethylstilbestrol in male Syrian golden hamsters and European hamsters. J. Natl. Cancer Inst. 62: 1083–1088.
2. LI, J.J. & S.A. LI. 1984. High incidence of hepatocellular carcinomas after synthetic
estrogen administration in Syrian golden hamsters fed alpha-naphthoflavone: a
new tumor model. J. Natl. Cancer Inst. 73: 543–547.
3. FRANCAVILLA, A., L. POLIMENO, M. BARONE, et al. 1993. Hepatic regeneration and
growth factors. J. Surg. Oncol. Suppl. 3: 1–7.
4. FRANCAVILLA, A., L. POLIMENO, A. DI LEO, et al. 1989. The effect of estrogen
and tamoxifen on hepatocyte proliferation in vivo and in vitro. Hepatology 9:
614–620.
5. OHNISHI, S., T. MURAKAMI, T. MORIYAMA, et al. 1986. Androgen and estrogen
receptors in hepatocellular carcinoma and in the surrounding noncancerous liver
tissue. Hepatology 6: 440–443.
6. CONTER, R.L. & W.P. LONGMIRE, JR. 1988. Recurrent hepatic hemangiomas. Possible association with estrogen therapy. Ann. Surg. 207: 115–119.
7. MATHIEU, D., E.S. ZAFRANI, M.C. ANGLADE, et al. 1989. Association of focal
nodular hyperplasia and hepatic hemangioma. Gastroenterology 97: 154–157.
8. SAEGUSA, T., K. ITO, N. OBA, et al. 1995. Enlargement of multiple cavernous
hemangioma of the liver in association with pregnancy. Intern. Med. 34: 207–
211.
9. GEMER, O., O. MOSCOVICI, C.L. BEN-HORIN, et al. 2004. Oral contraceptives and
liver hemangioma: a case-control study. Acta Obstet. Gynecol. Scand. 83: 1199–
1201.
10. GLINKOVA, V., O. SHEVAH, M. BOAZ, et al. 2004. Hepatic haemangiomas: possible
association with female sex hormones. Gut 53: 1352–1355.
11. SCHNAPER, H.W., K.A. MCGOWAN, S. KIM-SCHULZE, et al. 1996. Oestrogen and
endothelial cell angiogenic activity. Clin. Exp. Pharmacol. Physiol. 23: 247–250.
12. JAGGERS, D.C., W.P. COLLINS & S.R. MILLIGAN. 1996. Potent inhibitory effects of
steroids in an in vitro model of angiogenesis. J. Endocrinol. 150: 457–464.
13. SCALORI, A., A. TAVANI, S. GALLUS, et al. 2002. Oral contraceptives and the risk
of focal nodular hyperplasia of the liver: a case-control study. Am. J. Obstet.
Gynecol. 186: 195–197.
14. MATHIEU, D., H. KOBEITER, D. CHERQUI, et al. 1998. Oral contraceptive intake in
women with focal nodular hyperplasia of the liver. Lancet 352: 1679–1680.
15. BAUM, J.K., F. HOLTZ, J.J. BOOKSTEIN, et al. 1976. Possible association between
benign hepatomas and oral contraceptives. Lancet 2: 926–929.
16. EDMONSON, H.A., B. HENDERSON & B. BENTON. 1976. Liver-cell adenomas association with use of oral contraceptives. N. Engl. J. Med. 294: 470–472.
GIANNITRAPANI et al.: SEX HORMONES AND RISK OF LIVER TUMOR
235
17. ROOKS, J.B., H.W. ORY, K.G. ISHAK, et al. 1979. Epidemiology of hepatocellular
adenoma. The role of oral contraceptive use. JAMA 242: 644–648.
18. NAKAO, A., K. SAKAGAMI, Y. NAKATA, et al. 2000. Multiple hepatic adenomas
caused by long-term administration of androgenic steroids for aplastic anemia in
association with familial adenomatous polyposis. J. Gastroenterol. 35: 557–562.
19. LANDIS, S.H., T. MURRAY, S. BOLDEN, et al. 1998. Cancer statistics, 1998. CA
Cancer J. Clin. 48: 6–29.
20. MONTALTO, G., M. CERVELLO, L. GIANNITRAPANI, et al. 2002. Epidemiology, risk
factors, and natural history of hepatocellular carcinoma. Ann. N.Y. Acad. Sci.
963: 13–20.
21. STROFFOLINI, T., P. ANDREONE, A. ANDRIULLI, et al. 1998. Characteristics of hepatocellular carcinoma in Italy. J. Hepatol. 29: 944–952.
22. EL-SERAG, H.B. & A.C. MASON. 1999. Rising incidence of hepatocellular carcinoma in the United States. N. Engl. J. Med. 340: 745–750.
23. EL-SERAG, H.B. 2001. Epidemiology of hepatocellular carcinoma. Clin. Liver Dis.
5: 87–107.
24. NEUBERGER, J., D. FORMAN, R. DOLL, et al. 1986. Oral contraceptives and hepatocellular carcinoma. Br. Med. J. (Clin. Res. Ed.). 292: 1355–1357.
25. FORMAN, D., T.J. VINCENT & R. DOLL. 1986. Cancer of the liver and the use of oral
contraceptives. Br. Med. J. (Clin. Res. Ed.). 292: 1357–1361.
26. YU, M.C. & J.M. YUAN. 2004. Environmental factors and risk for hepatocellular
carcinoma. Gastroenterology 127: S72–S78.
27. ROSSINI, G.P., G.M. BALDINI, E. VILLA, et al. 1989. Characterization of estrogen
receptors from human liver. Gastroenterology 96: 1102–1109.
28. MONTALTO, G., M.D. MICELI, M. SORESI, et al. 1997. Sex hormones in patients
with liver cirrhosis and hepatocellular carcinoma. Oncol Rep. 4: 1–4.
29. YU, M.W., S.W. CHENG, M.W. LIN, et al. 2000. Androgen-receptor gene CAG
repeats, plasma testosterone levels, and risk of hepatitis B-related hepatocellular
carcinoma. J. Natl. Cancer Inst. 92: 2023–2028.
30. TANAKA, K., H. SAKAI, M. HASHIZUME, et al. 2000. Serum testosterone:estradiol
ratio and the development of hepatocellular carcinoma among male cirrhotic
patients. Cancer Res. 60: 5106–5110.
31. CHAO, Y., W.K. CHAN, Y.S. HUANG, et al. 1996. Phase II study of flutamide in the
treatment of hepatocellular carcinoma. Cancer 77: 635–639.
32. GRIMALDI, C., H. BLEIBERG, F. GAY, et al. 1998. Evaluation of antiandrogen therapy
in unresectable hepatocellular carcinoma: results of a European Organization for
Research and Treatment of Cancer multicentric double-blind trial. J. Clin. Oncol.
16: 411–417.
33. CLIP GROUP (CANCER OF THE LIVER ITALIAN PROGRAMME). 1998. Tamoxifen in
treatment of hepatocellular carcinoma: a randomised controlled trial. Lancet
352: 17–20.
34. LIU, C.L., S.T. FAN, I.O. NG, et al. 2000. 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.
35. VILLA, E., L. CAMELLINI, A. DUGANI, et al. 1995. Variant estrogen receptor messenger RNA species detected in human primary hepatocellular carcinoma. Cancer
Res. 55: 498–500.
36. VILLA, E., A. GROTTOLA, A. COLANTONI, et al. 2002. Hepatocellular carcinoma:
role of estrogen receptors in the liver. Ann. N. Y. Acad. Sci. 963: 37–45.
236
ANNALS NEW YORK ACADEMY OF SCIENCES
37. JIANG, S.Y., R.Y. SHYU, M.Y. YEH, et al. 1995. Tamoxifen inhibits hepatoma cell
growth through an estrogen receptor independent mechanism. J. Hepatol. 23:
712–719.
38. CASTAGNETTA, L.A., B. AGOSTARA, G. MONTALTO, et al. 2003. Local estrogen
formation by nontumoral, cirrhotic, and malignant human liver tissues and cells.
Cancer Res. 63: 5041–5045.