Cholangiocarcinoma: Epidemiology, Mechanisms of ...Vol. 2, 537-544, November/December 1993 Cancer...

Vol. 2, 537-544, November/December 1993 Cancer Epidemiology,Biomarkers & Prevention 537

Cholangiocarcinoma: Epidemiology, Mechanisms ofCarcinogenesis and Prevention

D. Maxwell Parkin,’ Hiroshi Ohshima,Petcharin Srivatanakul, and Vanchai Vatanasapt

International Agency for Research on Cancer, 1 50 coors Albert-Thomas,

69372 Lyon Cedex 08, France ID. M. P., H. 0.1; National Cancer

Institute, Rama VI Road, Bangkok 10400, Thailand lP. 5.1; and Cancer

Unit, Faculty d)f Medicine, Khon Kaen University, Khon

Kaen 40002, Thailand IV. V.1

Abstract

Cholangiocarcinoma is a relatively rare cancer;worldwide it accounts for an estimated 1 5 % of liver

cancers. In most areas, the etiology is rather obscure,and identified risk factors such as hepatolithiasis,inflammatory bowel disease, and exposure to Thorotrast

can account for only a small proportion of cases. Incertain areas of southeast and eastern Asia, however,incidence rates are very high, and here there is a strongassociation with infedion with the liver flukesClonorchis sinensis and Opisthorchis viverrini. Themechanisms of carcinogenesis in 0. viverrini infectionhave been the subject of considerable research; it seemsthat the presence of parasites induces DNA damageand mutations as a consequence of the formationof carcinogens/free radicals and of cellular proliferation

of the intrahepatic bile duct epithelium. Preventivestrategies in areas endemic for liver flukes appearstraightforward, but breaking the cycle of infection hasproved difficult in practice.

Introduction

CCA2 is one ofthe major histological types ofpnimary cancerof the liver, which usually occurs rather less frequently thanHCC. The tissue of origin is the epithelium ofthe bile ducts.

Strictly speaking, only tumors of the intrahepatic bile ductsare considered to be cholangiocarcinomas (1) although

clinically it is often difficultto distinguish between intra- and

extrahepatic cancers of the hilan region. Although cholan-giocancinoma is considered a specific morphological entityfor classification purposes (with a specific ICD-0 code,8160), it is histologically identical to adenocarcinoma, and

diagnoses such as duct carcinoma and adenocarcinoma are

considered to be synonymous (2).

Received 7/16/93; revised 7/16/93; accepted 7/16/93.1 To whom requests for reprints should he addressed.

2 The abbreviations used are: CCA, cholangiocarcinoma; HCC, hepatocel-

lular carcinoma; OV, Opisthorchis viverrini: NO, nitric oxide; ICD-O, inter-national c:lassification of diseases for oncology.

Epidemiology

Frequency

In published studies, the frequency ofCCA is almost alwaysreported from clinical series, as a percentage of all liver can-cers (on biopsy, or at autopsy). Frequencies reported varyfrom 5 to 30% of all liven cancers (3). The case-selectioninherent in clinical series can be overcome by studyingpopulation-based material (which includes all cases of livencancer arising in a defined population). Table i shows rela-tive frequency of intrahepatic carcinoma of glandular epi-thelium, stated to be primary, in i 5 large population-basedseries for time periods around 1985. The tumors include

adenocarcinomas (ICD-0 codes 81 40 + 81 41 ) and duct can-cinomas (ICD-0 8500) as well as cholangiocarcinomas, andthe denominator comprises all primary liver cancer ofknown histological type. Apart from the very high frequencyof cholangiocarcinoma in Khon Kaen, Thailand, the per-centage of such cancers lies between 5 and 30% in men andbetween 15 and 45% in women.

These figures have only limited utility. Apart from anybias which may anise from differential biopsy rates for he-patocellular and cholangiocarcinomas, the frequency ofcholangiocancinoma will vary inversely with that of hepa-tocellulan carcinoma, which is known to vary widely in dif-ferent parts of the world (4).

Incidence

Figs. i and 2 show incidence rates for liver cancer in 26populations around 1985, from data published in Ref. 5.Based upon the distribution ofthe different histological sub-types, within age groups (0-i4, 15-49, 50-59, 60-69, and70+ years), estimated incidence rates for the major histo-logical types of liver cancer have been prepared. Hepato-cellular carcinoma (ICD-0 8i 70) is distinguished from in-trahepatic bile duct carcinoma, on cholangiocancinoma(ICD-0 8140 + 8141, 8i60 + 816i, and 8500), hepato-blastoma (ICD-0 8970), and other types. Cases with un-known or unspecified histology were reallocated proratawithin each age/sex grouping to these four categories, beforecalculation of rates.

Despite a rather wide range in incidence of liver canceras a whole, the incidence of cholangiocarcinoma showsrather little variation, with rates in males mainly ranging 0.5-2.0 and a little lower in females. A very high incidence isestimated for Khon Kaen in northeast Thailand (84.6/i 0� inmen; 36.8/i05 in women), with moderately high rates inChiang Mai (northern Thailand) and Hong Kong. There is asuggestion that elsewhere in Asia (Japan, Philippines, andSingapore) the incidence is slightly higher than in Europe orthe Americas.

The sex ratio for CCA in the series in Table i is relativelyclose to unity (range, 0.7-1 .6). This is much lower than for

on August 18, 2020. © 1993 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

538 Cholangiocarcinoma

Table 1 Cholangio carcinoma as a percentage of all liver cancer”

Cancer regfstryTotal No

.(liver cancer)

of cholangiocarcinoma.-. . . - --

Male Female

Ratio:M:F

Thailand, Khon Kaen 1235 87 92 1.8

Thailand, Chiang Mai 633 29 43 1.2

Hong Kong 6765 13 32 1.6

lapan, Osaka 1 2260 6 1 8 1 .4

Philippines, Manila, 2610 5 22 0.7

and Rizal

Singapore: Chinese 1295 5 19 0.6

United States (SEER”): 2136 19 30 1.1

white

Canada 2749 16 30 1.1

Slovakia 585 26 40 1 .3

Denmark 1 337 30 50 0.9

Italy (six registries’) 1290 7 20 0.9

Spain (six registries”) 1049 9 23 0.8

United Kingdom, 847 24 39 1.1

Scotland

Australia (4 states”) 963 9 19 1.1

Puerto Rico 449 13 22 1.3

., Excluding cases of unknown or ill-defined histology. Calculated from data

published in Ref. 5.“ Registries of the Surveillance Epidemiology and End Results Programme.

‘ Florence; Genoa; Lombardy Region, Varese Province; Parma Province;Romagna; Trieste.,j Basque Country; Catalonia, Tarragona; Granada; Murcia; Navarra;

Zaragoza.(. New South Wales, South Australia, Tasmania, Victoria.

hepatocellular carcinoma, which ranges from 2-3 in Europeand North America to around 5 in the high-risk populationsof east Asia.

Figs. 3 and 4 show age-specific incidence rates for he-patocellulan carcinoma or cholangiocarcinoma from two ofthe population-based series in Figs. i and 2 (whites in theUnited States and Japanese in Osaka). They show that CCAoccurs at rather older ages than HCC; the median age isaround i (males) to 3 (females) years greater in the UnitedStates, and around 4 (females) to 5 (males) years greaten inJapan. These findings confirm the rather olden mean agefor CCA cases, when compared to HCC, in most clinicalseries (6).

Etiological Factors

There have been relatively few epidemiological studies ofcholangiocarcinoma. Most of the information on factorswhich are likely to be important derives from clinical series,in which association with other diseases on exposures ap-pears to be, a priori, higher than expected or higher thanobserved in the cases of HCC in the same clinical series.

Congenital Cystic and Dysplastic Lesions. These are asso-ciated with a proportion of cases (2).

Intrahepatic Calculi. Carcinoma of the gall bladder is wellknown to be a complication of cholelithiasis. By analogy,the presence of intnahepatic stones (hepatolithiasis) is ob-served in a fairly impressive proportion of CCA cases, from5.7 to i 7.5% of cases in four Japanese series (7). The bile

duct epithelium in hepatolithiasis shows chronic prolifera-tive cholangitis and epithelial hyperplasia (8).

Inflammatory Bowel Disease. Carcinoma of the bile duct

has been observed as a complication of ulcerative colitis,although it is much less frequent than cancers of the largebowel itself. Ritchie et a!. (9) report the frequency of occur-

nence as 1 in 246 ulcerative colitis patients, with one-fifth ofcases occurring in the intra-hepatic bile ducts. In a series ofcases from the Mayo Clinic, the average age of diagnosis ofbile duct cancer in ulcerative colitis patients was 38 years,after a mean follow-up period after ulcerative colitis diag-nosis of 1 9 years (1 0). Using these data, and the age-specificincidence rates of intrahepatic bile duct cancer in the UnitedStates estimated as described in “Incidence”, the expected

frequency of cancers between ages 20 and 39 years is 1 in95,000 (compared to about 1 in 1 ,250 observed), a relativerisk of about 75 in ulcerative colitis patients compared withthe general United States population.

Thorotrast. Thorotrast (a colloidal preparation of thoriumdioxide) is a radioactive a-particle emitter widely used as aradio-opaque contrast medium between 1 930 and 1955. Itwas administered by many different routes. When given iv.,70%ofthedose istaken up intheliven. Liver tumors developsome 1 0-1 2 years later, but the proportions of different his-tological types vary considerably between the different se-ries. This is probably related to the different baseline (non-Thonotrast-associated) risk of hepatocellulan canc i noma indifferent populations and to the fact that cases of CCA seemto occur rather earlier after Thorotrast than do cases ofangiosancoma or HCC (1 1).

In a cohort study of 241 males exposed to Thorotrastaround 1945 (12), follow-up between 1980 and 1985 re-vealed 51 deaths from liver cancer, a relative risk, comparedwith the general Japanese population, of 46.9. Based on au-topsy findings of 30 cases and the results of autopsies inJapan (1 980-i 984), relative risks for the different histologi-cal subtypes were estimated as follows: hepatoceliular car-cinoma, 21 ; cholangiocarcinoma, 303; and hemangiosar-coma, 3129.

In a later report on the same cohort, Kiyosawa et a!. (1 3)compared the findings in 36 liver cancer deaths (16 withcholangiocarcinoma) with 67 survivors without liven cancer.None of the cases on controls had evidence of previous in-fection with hepatitis B, and there was no association be-tween eithentype ofliver cancer and alcohol intake or smok-ing in this Thorotnast-exposed group.

Proliferative change and dysplasia of the bile duct epi-thelium are noted in the noncancerous areas of the liver inThorotrast-induced cases (14).

Parasites. The association between the occurrence of livercancer and the presence of liver flukes has been known foralmost a century. Hou (1 5) in Hong Kong was the first topoint out that, although 1 5% of liver cancers were associ-ated with flukes (in this case, C!onorchis sinensis), the can-cers concerned were mainly adenocarcinomas. Gibson (16),reporting on an autopsy series (1484 subjects) from HongKong, found that while 29% of 89 HCC cases had donor-chiasis (35.4#{176}/expected, based on age-sex-specific preva-lence of the whole series), 65% of 1 7 CCA cases (37.6%expected) were infected. Belmanic (1 7), in the same popu-lation, identified flukes in 1 8 of 1 9 autopsies of patients withcholangiocancinoma, but only one-third of ‘control’ patientsshowed cellular change suggestive of clonorchiasis. In ani-mals infected with C!onorchis, intrahepatic bile duct cancerswere observed (18).

C. sinensis was originally endemic in Korea, Japan,China, and Vietnam. However, it is much less prevalent thanit was, and cholangiocarcinoma from this cause appears to

be relatively infrequent in recent years.This is not the case for OV, the fluke found in northeast

Thailand. Recent population surveys suggest (3 continuing



k A Aqe-stand. ratevia e �per 100,000)

0 10 20 30 40 85 95 �, I I I jj I

. �“//////A I 84.6 6.8 94.2

�‘/////////////////////////////////////////iI I ‘ 2.8 37.9 41.6

S Cholangiocarcinorna

� Hepatocellular carcinoma

D Other

Cancer Epidemiology, Biomarkers & Prevention 539

LIVER CANCER: Hepatocellular & Cholangiocarcinoma

Thailand, Khon Kaen

Japan. OsakaHong Kong

Japan, HiroshimaSingapore, Chinese

Philippines, Manila

Thailand, Chiang Mal

Singapore, Malay

Italy, VareseFrance, Bas-Rhin

Switzerland, Zurich

Poland (Cracow+N.S.)

Spain (6 registries)

Finland

USA, SEER, Black

Denmark

Costa Rica

Puerto Rico

Slovenia

Slovakia

Scotland

Israel, Jews

Canada

USA, SEER, White

New Zealand, non MaoriAustralia (4 states)

W///////////,I I

�////////A I

21

5.4 27.9 39.2

1 .0 24.6 28.5

1.0 24.2 27.2

1.4 18.2 23.5

6.1 11.3 19.7

1.9 11.5 14.3

0.9 8.6 10.6

1.0 7.8 9.1

0.4 5.1 6.0

0.6 1.0 5.9

0.5 4.1 5.8

1.2 2.9 4.7

0.6 3.3 4.2

1.1 2.1 4.0

0.3 2.8 4.0

0.4 2.6 3.3

0.7 1.5 3.3

0.8 1 .9 2.9

0.7 1.7 2.9

0.3 2.2 2.8

0.4 1.8 2.5

0.4 1.6 2.4

0.4 1.3 2.1

0.2 1 .7 2.0

Fig. 1. Age-standardized incidence of liver cancer, per 100,000, in males (5). Rates for cholangiocarcinoma and hepatocellular carcinoma are estimates (seetext). Poland: registries of dracow and Nowy Sacz. Spain: registries of Basque Country, Tarragona, Granada, Murcia, Navarra, and Zaragoza. Australia: registries

of New South Wales, South Australia, Tasmania, and Victoria.

high prevalence of infection. The evidence for the role of OVin induction of CCA is compelling. The first reports wereessentially case series, drawing attention to the coincidenceof two diseases, normally rare, in the same geographic areaand in the same individuals. Thus, cholangiocarcinomawas observed to comprise a high percentage of biopsiedliven cancers in northeast Thailand (i 9, 20) and prevalenceof OV infection is higher in northeast Thailand than else-where (21, 22).

More formal correlations have been performed. The in-cidence of cholangiocancinoma in the five regions of Thai-land varies at least 1 2-fold and correlates strongly withprevalence of OV infection, as measured by anti-OV anti-body titer in the general population (23). HCC shows nosuch relationship. The association with fecal egg count wasless strong, presumably because this may be more affectedby recent therapy and a poorer indicator of duration/intensity of infection. A similar association between the in-tensity of OV infection and the risk of CCA has been ob-served at district level within northeast Thailand (24).Similarly, surveys of villagers in northeast Thailand show astrong association between the intensity ofOV infection andabnormalities of the biliary tract, presumably themselvesassociated with CCA (25, 26).

Only one adequate case-control study of CCA has beenperformed (27). One hundred three cases that were inhab-itants of northeast Thailand were compared with a similarnumber of age- and sex-matched controls. Infection with OV(past on present) was estimated in terms of an increase in titer

of anti-OV antibodies in serum, since this is known to con-relate with intensity of infection (28, 29) while the count ofOV eggs in feces may be low or zero in cases of CCA withbiliary obstruction. A relative risk of 5.0 for an antibody titergreaten than 1 :40 was found; this implies that at least two-thirds ofCCA cases were attributable to OV. These estimatesare conservative, since the antibody test used a crude extractof parasite as antigen and is known to be rather nonspecific(30) with some consequent misclassification ofexposure sta-tus. The study suggested that the risk associated with OVinfection was higher in males than in females, a finding con-sistent with the higher incidence of CCA in men in OV-endemic areas in the face of a similar prevalence of infectionin the two sexes, but if it is true, the responsible mechanismis obscure.

Other Factors. The case-control study in Thailand (27)found no association with chronic carriage of hepatitis B, nonwith recent aflatoxin intake. No dietary constituents in-creased on decreased risk and there was no association withtobacco use; there was, however, a strong association withthe regular use of betel nut (odds ratio, 6.4).

In a case-control study of primary liver cancer in Swe-den (31 ), 1 5 cases ofcholangiocarcinoma were included; nosignificant association was found with exposure to organicsolvents on with alcohol intake (the odds ratio for one-halfof a bottle of spirits per week was 3.3 (not significant) com-pared with teetotallers).

A case-control study of liver cancer in women aged20-44 years in England (32) which included 1 1 cases of CCA



0 5 10

Female

�1

-�t’///A I

V/Z//A I

t”//A I

vz�

U Cholangiocarcinoma

0 Hepatocellular carcinoma

0 Other


LIVER CANCER: Hepafocellular & Cholangiocarcinoma

Thailand, Khon Kaen

Thailand, Chiang Mai

Hong Kong

Japan, Osaka

Philippines, ManilaJapan, Hiroshima

Singapore, ChinesePoland (Cracow+N.S.)

Italy, Varese

Costa Rica

Spain (6 registries)

Finland

Denmark

Switzerland, ZurichSlovenia

USA, SEER, BlackFrance, Bas-Rhin

Puerto Rico

Israel, Jews

Slovakia

Scotland

USA, SEER, White

Canada

New Zealand, non Maori

Australia (4 states)

Age-stand. rate

(per 100,000)

35 4oL�L�I1�1‘//Z��I� ‘ 36.8 1.6 39.1

4.8 4.5 10.4

3.1 4.1 9.7

1.7 7.4 9.6

1.5 5.0 7.9

1.2 5.2 7.3

1.1 5.1 6.9

0.8 0.4 3.2

0.6 1.4 2.7

0.3 1.7 2.6

0.5 1.2 2.2

0.9 0.9 2.1

0.9 0.8 2.0

0.5 1.1 1.7

0.5 0.5 1.5

0.3 0.9 1.4

0.3 0.9 1.3

0.3 0.7 1.3

0.3 0.8 1.2

0.4 0.6 1.1

0.4 0.5 1.1

0.3 0.5 1.1

0.3 0.5 1.0

0.4 0.3 0.9

0.1 0.4 0.6

Fig. 2. Age-standardized incidence of liver cancer, per 100,000, in females 15). Rates for cholangiocarcinoma and hepatocellular carcinoma are estimates (see

text). Poland: registries ofCracow and Nowy Sacz. Spain: registries of Basque Country, Tarragona, Granada, Murcia, Navarra, and Zaragoza. Australia: registries

of New South Wales, South Australia, Tasmania, and Victoria.

found no association with past use of oral contraceptives, asrecorded in general practioner notes. A similar study in theUnited States included 22 deaths from cancer of the intra-hepatic bile ducts in women aged 25-49 years; no excessrisk of past oral contraceptive use, ascertained from ques-

tionnaires to relatives, was found (33).

Mechanisms of Liver Fluke-induced Carcinogenesis

Human Studies. Snianujata et a!. (34) observed that humansinfested with liven fluke had higher urinary excretion of ni-trate and N-nitrosoproline than the uninfected. A study infive different areas ofThailand (35)found excretion of nitrateand nitnosamine, and endogenous nitrosation potential didnot correlate with the incidence of CCA. However, withinthe two high-risk areas of northeast Thailand, subjects whowere positive for OV (by antibody level) showed consider-ably enhanced endogenous nitnosation, and this was ne-duced to the same level as control (noninfected) subjects byadministration of vitamin C, which inhibits this process.

These results suggest that, in common with other situ-ations where cancer is a sequel of long-term infectious pro-cesses, there is a role of nitrosation/nitnosamines consequentupon fluke infestation in the etiology of CCA.

The possibility that immunological reactions are in-volved in pathogenesis has been postulated, based on a con-relation between OV-specific lgG antibody and ultrasound-diagnosed changes in the biliany tract in an endemic area(36). It seems equally likely that this association is indirect;

the antibody levels simply reflect duration/intensity or someother relevant parameter of OV infection.

There is currently a great interest in the relationshipbetween specific mutation spectra in protooncogenes (such

as ras) and tumor suppressor genes (such as pS3) and ex-posunes to different etiological factors (37). For example, he-

patocellulan carcinomas in subjects probably exposed to di-

etary aflatoxin B contain frequent G to T transVersionsoccurring at codon 249 of p53 gene (38, 39). This type ofmutation, however, is seldom observed in the tumors of sub-jects from areas in which aflatoxin B, exposure is low. Re-cently, Tada et a!. (40) observed that 9 of 1 8 intrahepaticcholangiocarcinomas from Japan had point mutations in therasgene (6 ofwhich were in K-rascodon 1 2), a feature neverobserved in hepatocellular carcinoma. The mutation spec-

trum was very similar to that of colon cancer, suggesting anetiologicallinkage;thehigh niskofbothcancers in ulcerativecolitis patients, mentioned earlier, might be relevant. Tsudaet a!. (41) compared point mutations of the c-Ki-ras pro-tooncogene in cholangiocarcinomas from Japan and Thai-

land. They found that 5 (56%) of 9 cholangiocarcinomasamples from Japan had point mutations at codon 1 2 (2 GGT

�* GCT; 2 GGT -� GAT; 1 GGT -� GTT), but none of the1 2 tumors from northeast Thailand contained these niuta-

tions (41 ). Although the pathogenetic significance of pointmutation ofthec-Ki-rasprotooncogene in human cancer hasnot been well established, these results suggest that pointmutation ofc-Ki-ras is not involved in Opisthorchis viverrini-



., .

.. . �1..

H

a,

a. .

: .

A.

00- 25- 50- 75-

AQe


Fig. 3. Age-specific incidence rates of liver can-cer: whites in the United States (5). Median agesof patients with hepatocellular carcinoma were

67.9 (males) and 69.9 (females) years. Median

ages of patients with cholangiocarcinoma were68.7 (male) and 72.9 (females) years.

100�

10�

0.1

0.01

USA: White

Hepatocellular & Cholarigiocarcinoma

IncIdence (Log)

Hepatocellular carcinoma

. . .. .. Male � � A� ‘. Female

associated carcinogenesis, although other genetic alter-ations resulting from infection and mechanical irritation bythe flukes could be important. It would be interesting to com-pane mutation spectra in the p53 tumor suppressor gene incholangiocarcinomas from 0. viverrini-endemic and non-endemic areas.

Animal Studies. Studies on experimental OV infestations inSyrian golden hamsters have demonstrated that liver flukeinfestation alone rarely induces cholangiocarcinoma, butif infested hamsters are treated with hepatocancinogenssuch as N-nitnosodimethylamine (NDMA) and N-nitroso-di-iso-propanolamine, they can develop bile duct tumorsresembling those seen in humans (42, 43). The effect ofOV infestation and NDMA dose on the development ofcholangiocarcinoma was synergistic (44). These resultssuggest that the development of cholangiocarcinoma inthe OV-infested host is a multifactonial process.

There are several possible mechanisms underlying theenhancement of neoplasia by opisthorchiasis to be consid-ered. First, the presence of the parasite in the bile duct

Cholanglocarclnoma

-.-- Male -a.-- Female

mechanically damages tissues, resulting in increased cellproliferation, which converts and fixes DNA-carcinogenadducts to mutations, and also results in an elevated

“spontaneous” mutation at a later modulation stage of car-cinogenesis (45, 46). Repeated treatment of the infestedhamsters with Praziquantel at levels sufficient for removalof parasite infestation resulted in reduction in chronic pro-liferative cholangitis, considered to be a precancerousstage of cholangiocarcinoma (47).

Second, reactive oxygen species released by activatedmacrophages and neutrophils induce DNA damage andsubsequent mutations leading to cancer (48). A third pos-sibility is that presence of parasites may stimulate releaseby macrophages and other cell types of cytokines such asinterleukin 1 and tumor necrosis factor, which may beinvolved in tumor enhancement by stimulating angio-genesis (49).

A fourth possibility is involvement of NO synthase. Ourrecent studies show that NO synthase is induced in mac-nophages, eosinophils, and mast cells which are present in



, . .. . � . a. . �. . #{149}..

U

.,. .A,AAA

. AU

A

PA.

U

‘

U

�‘A .�

. A’� �“A U

00- 25- 50- 75-

Age


Fig. 4. Age-specific incidence rates of liver cancer:

Osaka (5). Median ages of hepatocellular carcinomawere 60.4 (male) and 67.9 (female) years. Median

ages of patients with cholangiocarcinoma were 65.2

(males) and 71 .8 (female) years.

1000

100

10

0.1

0.01

OSAKA

Hepatocellular & Cholangiocarcinoma

IncIdence (Log)

Hepatocelluiar carcinoma. . .. .- Male � � A’ ‘- Female

the inflamed zone surrounding the parasite-containing bileduct.3 If present in excess, NO exerts cytotoxic and muta-genic effects and also immunosuppressive activity (50, 51).Increased endogenous formation of nitrosamines could oc-cur locally in the inflamed bile ducts if nitrosatable aminesare present (52)� OV-infested hamsters administered ami-nopynine and nitrite, precursors of NDMA, showed an in-creased yield of cholangiocarcinoma, suggesting that en-dogenous nitrosation could be an important factor (53).

Finally, it was found recently in this laboratory that ac-tivation of nitrosamines and afiatoxin B1 to DNA-bindingmetabolites is increased in OV-positive hamsters comparedto controls.4 The increased metabolism is related to certainP-450 isozymes highly expressed in regions of liver facing or

close to the inflammation.5

‘ H. Ohshima, T.Y. Bandaletova, H. Bartsch, I. Brouet, I. Kirby, F. Ogunhiyi,

V. Vatanasapt, and d.P. Wild, submitted for publication, 1993.4 G. Kirby, manuscript in preparation.S p, Pelkonen, G. Kirby, C, Wild, H. Bartsch, and M. Lang, manuscript inpreparation, 1 993.

Cholangiocarcinoma

-.--- Male -h-- Female

in summary, it seems that the presence of parasitescould induce DNA damage and mutations as a consequenceofthe formation ofcarcinogens/free radicals and of cell pro-lifenation in the intrahepatic bile ducts, which may play acrucial role in the development of cholangiocarcinoma.

Prevention

Early detection of liver cancer, essentially hepatocellularcarcinoma, has been attempted in an effort to improve prog-nosis and reduce mortality, relying upon testing fora-fetoprotein on ultrasound examination of high-risk groups(54, 55). The prognosis of cholangiocarcinoma is equallypoor (56) and the observation of better survival for small (lessthan 4 cm) lesions has induced similar attempts at early di-

agnosis. One possibility is the use oftumon markers such asCA19-9 and CEA to identify early cancers. Used in combi-nation, they have high sensitivity for detecting clinical cases[97.5% according to Sripa et a!. (57)1, although experiencewith gallbladder cancer (58) suggests that this could only beachieved with such a low specificity that it would be uselessfor the purpose ofscreening. The feasibility of using anti-OVIgG antibodies to identify high-risk individuals who could be




4. Mu#{241}oz,N., and Bosch, X. Epidemiology of hepatocellular carcinoma. in:

followed up by ultrasound is also under study (59). As forhepatoceliular carcinoma, however, it is unlikely thatscreening will yield benefits commensurate with the costsinvolved, and a more national approach to control is throughprimary prevention.

Primary prevention ofchoiangiocancinoma in the high-risk areas where the tumors are associated with liver flukeinfestation is apparently straightforward; control ofthe pana-site should result in a fall in incidence of cholangiocanci-noma to that observed in nonendemic areas. Effective treat-ment for opisthorchiasis is available by use of the drugPraziquantel which, administered in a single dose, can suc-cessfully eliminate parasites from infested individuals (60,6i). Unfortunately, studies in Thailand have shown that re-infection occurs rapidly after successful treatment, panticu-lanly in individuals with high pretreatment intensities of in-

fection (62). This is presumably related to persistence of theparasite in the environment (encysted in the fish interme-diate hosts) and the lifestyle habits associated with infection(eating these raw fish). Successful control will require ne-peated treatment, concentrating upon individuals at highestrisk, coupled with attempts to change traditional dietary pat-terns, although the latter have proved rather refractory toeducational programs in the past. It is possible that, if en-dogenous nitnosation is truly important in the etiological pro-cess, administration of vitamin C may be effective in reduc-ing the risk of cholangiocarcinoma in infected subjects.

It is uniikelythatthe effectiveness ofpnogramsto reduceparasitic infection in reducing incidence of cholangiocan-cinoma will ever be demonstrable through a controlled trial.Leaving aside the ethical difficulties of maintaining an un-treated control group, a trial to demonstrate an effect oncancer incidence would need to be very long and, in the faceof noncompliance and reinfection, extremely large (63). Anopisthorchiasis control program has been introduced by theHealth Ministry in the 1 7 provinces of northeast Thailand,and it will be importantto establish careful monitoring of theprevalence of infection via population surveys, and the in-cidence of cholangiocarcinoma via cancer registration.

North and northeast Thailand and Laos constitute areasof particularly high risk for cholangiocarcinoma, and it canbe estimated from the incidence rates of liver cancer andproportion of cases which are cholangiocarcinoma (64) thatcurrently some 9,000-1 0,000 new cases occur there eachyear. There are, however, some 3i 5,000 new cases of livencancer in the world each year (65). Using the appropriateproportions of cholangiocarcinomas from Table 1 (7.5% in

men and 20% in women in high risk areas ofAfnica and Asia,and 20% of cases in men and 33% in women elsewhere),about 1 5% of liver cancer (46,000 cases a year) can be es-timated to be cholangiocarcinoma. At least 80% of thisworld total is unrelated to Opisthorchis, therefore, and,

based on current knowledge of the epidemiology, there areno clues to appropriate preventive measures.

References

1 . Gibson, I. B., and Sobin, L. H. Histological typing of tumours ofthe liver,

biliary tract and pancreas. Geneva: WHO, 1978.

2. Craig, I. R., Peters, R. L., and Edmondson, H. A. Tumors of the liver andintrahepatic bile ducts. Washington, DC: Armed Forces Institute of Pathology(Atlas of Tumor Pathology, Second Series; Fasc. 26.) 1989.

3. Mori, W., and Nagasako, K. Cholangiocarcinoma and related lesions. In:K. Okuda and R. L. Peters (eds.), Hepatocellular Carcinoma, pp. 227-246.New York: John Wiley, 1976.

K. Okuda and K. G. Ishak (eds.), Neoplasms of the Liver, pp. 3-19. Tokyo:

Springer Verlag, 1 987.

5. Parkin, D. M., Muir, C. S., Whelan, S. L., Gao, Y-T., Ferlay, j., and Powell,

I., (eds.). Cancer Incidence in Five Continents, Vol. VI (IARC Scientific Pub.No. 1 20). Lyon, France: International Agency for Research on Cancer, 1992.

6. Okuda, K., and the Liver Cancer Study Group ofjapan. Primary liver can-

cers in Japan. Cancer (Phila.), 45: 2663-2669, 1980.

7. Sugihara, S., and Kojiro, M. Pathology of cholangiocarcinoma. In: K.

Okuda and K. G. Ishak (eds.), Neoplasms ofthe Liver. Tokyo: Springer Verlag,1987.

8. Nakanuma, Y., Terada, T., Tanaka, Y., and Ohta, G. Are hepatolithiasisand cholangioma aetiologically related? A morphological study of 1 2 casesof hepatolithiasis associated with cholangiocarcinoma. Virchows Arch. Abt.

A Pathol. Anat., 406: 45-58, 1985.

9. Ritchie, j. K., Allan, R. N., Macartney, I., Thompson, H., Hawley, P. R., and

Cooke, W. T. Biliary tract carcinoma associated with ulcerative colitis. j.Med., 43:263-279, 1974.

10. Akwari, 0. E., van Heerden, j. A., Foulk, W. T., and Baggenstoss, A. H.

Cancer of the bile ducts associated with ulcerative colitis. Ann. Surg., 181:

303-309, 1975.

1 1 . Yamada, S., Hosoda, S., Tateno, H., Kido, C., and Takahashi, S. Survey

of Thorotrast-associated liver cancers in Japan. j. NatI. Cancer Inst., 70: 31-

35, 1983.

1 2. Kato, I., and Kido, C. Increased risk of death in Thorotrast-exposed pa-tients during the late follow-up period. jpn. J. Cancer Res., 78: 1 1 87-1 192,1987.

1 3. Kiyosawa, K., Imai, H., Sodeyama, T., Franca, S. T. M., Yousuf, M., Fu-ruta, S., Fujisawa, K., and Kido, C. Comparison ofanamnestic history, alcoholintake and smoking, nutritional status, and liver dysfunction between thoro-

trast patients who developed primary liver cancer and those who did not.

Environ. Res., 49: 166-172, 1989.

1 4. Rubel, L. R., and Ishak, K. G. Thorotrast associated cholangiocarcinoma:

an epidemiologic and clinicopathologic study. Cancer (Phila.), 50: 1408-

1415, 1982.

1 5. Hou, P. C. Relationship between primary carcinoma of the liver andinfestation with Clonorchis sinensis. J. Pathol. Bacteriol., 72: 239-246, 19�6.

1 6. Gibson, j. B. Parasites, liver disease and liver cancer. In: Liver Cancer

(IARC Scientific Pub. No. 1 1. Lyon, France: International Agency for Researchon Cancer, 1971.

1 7. Belmaric, j. Intrahepatic bile duct carcinoma and C. sinensis infection in

Hong Kong. Cancer (Phila.), 31:468-473, 1973.

1 8. Hou, P. C. Primary carcinoma ofthe bile duct ofthe liver ofthe cat (Fells

Catus) infested with Clonorchis sinensis. j. Pathol. Bacteriol., 87: 239-243,1964.

19. Bhamarapravati, N., and Viranuvatti, V. Liver diseases in Thailand: an

analysis of liver biopsies. Am. J. Gastroenterol., 45: 267-275, 1966.

20. Bunyaratvej, S., Meenakanit, V., Tantachamrun, T., Srinawat, P., Susi-lavorn, R., and Chongchitnan, N. Nationwide survey of major liver diseasesin Thailand: analysis of 3305 biopsies as to year-end 1978. j. Med. Assoc.

Thail., 64:432-438, 1981.

21 . Harinasuta, C., and Vajasthira, S. Opisthorchiasis in Thailand. Ann. Trop.

Med. Parasitol., 54: 100-105, 1960.

22. Sadun, E. H. Studies on Opisthorchis viverrini in Thailand. Am. J. Hyg.,

62:81-115, 1955.

23. Srivatanakul, P., Parkin, D. M., Jiang, Y-Z., Khlat, M., Kao-Ian, U-T.,

Sontipong, S., and Wild, C. The role of infection by Opisthorchis viverrini,hepatitis B virus and aflatoxin exposure in the etiology of liver cancer inThailand. A correlation study. Cancer IPhila.), 68: 2411-2417, 1991.

24. Vatanasapt, V., Tangvoraphonkchai, V., Titapant, V., Pipitgool, V.,Viriyapap, D., and Sriamporn, S. A high incidence of liver cancer in Khon

Kaen province, Thailand. Southeast Asian j. Trop. Med. Public Health, 21:

489-494, 1990.

25. Elkins, D. B., Haswell-Elkins, M. R., Mairiang, E., Mairiang, P., Sithitha-worn, P., Kaewkes, S., Bhudhisawasdi, V., and Uttaravichien, T. A high fre-

quency of hepatobiliary disease and suspected cholangiocarcinoma associ-ated with heavy Opisthorchis viverrini infection in a small community innortheast Thailand. Trans. R. Soc. Trop. Med. Hyg., 84: 715-719, 1990.

26. Mairiang, E., Elkins, D. B., Mairiang, P., Chaiyakum, j., Chamadal, N.,Loapaiboon, V., Posri, S., Sithithaworn, P., and Haswell-Elkins, M. Relation-

ship between intensity of Opisthorchis viverrini infection and hepatobiliarydisease detected by ultrasonography. I. Gastroenterol. Hepatol., 7: 17-21,1992.

27. Parkin, D. M., Srivatanakul, P., Khlat, M., Chenvidhya, D., Chotiwan, P.,Insiripong, S., L’Abb#{233}, K. A., and Wild, C. P. liver cancer in Thailand: a

case-control study ofcholangiocarcinoma. nt . Cancer, 48: 323-328, 1991.




28. Srivatanakul, P., Viyanant, V., Kurathong, S., and Tiwawech, D. Enzyme-linked immunosorbent assay for detection of Opisthorchis viverrini infection.

Southeast Asian I. Trop. Med. Public Health, 16: 234-239, 1985.

29. Sirisinha, S. Immunodiagnosis of human liver fluke infections. Asian Pac.I. Allergy Immunol., 4:81-88, 1986.

30. Sirisinha, S., dhawengkirttikul, R., Haswell-Elkins, M. R., and Elkins, D.M. Immunodiagnosis of a Liver Fluke Infection Caused by Opisthorchis viver-rini (Abstract), XIII International Congress for Tropical Medicine and Malaria,Vol. 1, p. 263. Pattaya, Thailand: lomtien, 1992.

31. Hardell, L., Bengtsson, N. 0., jonsson, U., Eriksson, S., and Larsson, L.

G. Aetiological aspects on primary liver cancer with special regard to alcohol,organic solvents and acute intermittent porphyria: an epidemiological inves-tigation. Br. I. Cancer, 50: 389-397, 1984.

32. Forman, D., Vincent, T. J., and Doll, R. dancer of the liver and the useof oral contraceptives. Br. Med. I., 292: 1 357-1 361 , 1986.

33. Hsing, A. W., Hoover, R. N., McLaughlin, I. K., do-Chien, H. T.,Wacholder, S., Blot, W. I., and Fraumeni, I. F. Oral contraceptives and pri-mary liver cancer among young women. Cancer Causes Control, 3: 43-48,1992.

34. Srianujata, S., Tonbuth, S., Bunyaratvej, S., Valyasevi, A., Promvanit, N.,and dhaivatsagul, W. High urinary excretion of nitrate and N-nitrosoprolinein opisthochiasis subjects. In: H. Bartsch, I. K. O’Neill, and R. Schulte-

Hermann (eds.), Relevance of N-Nitroso Compounds to Human Cancer: Ex-posures and Mechanisms (lARd Scientific Pub. No. 84) pp. 544-546. Lyon,France: International Agency for Research on Cancer, 1987.

35. Srivatanakul, P., Ohshima, H., Khlat, M., Parkin, M., Sukaryodhin, S.,Brouet, I., and Bartsch, H. Opisthorchis viverrini infestation and endogenousnitrosamines as riskfactorsforcholangiocarcinoma in Thailand. tnt. I. Cancer,48:821-825, 1991b.

36. Haswell-Elkins, M. R., Sithithaworn, P., Mairiang, E., Elkins, D. B., Won-gratanacheewin, S., Kaewkes, P., and Mairiang, P. Immune responsiveness

and parasite specific antibody levels in human hepatobiliary disease asso-

ciated with Opisthorchis viverrini infection. Clin. Exp. Immunol., 84: 213-218, 1991.

37. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. C. p53 mu-tations in human cancers. Science (Washington DC), 253:49-53, 1991.

38. Hsu, I. C., Metcalf, R. A., Sun, T., Welsh, j. A., Wang, N. J., and Harris,C. C. Mutational hotspot in the p53 gene in human hepatocellular carcino-mas. Nature (Lond.), 350: 427-428, 1 991.

39. Bressac, B., Kew, M., Wands, I. , and Ozturk, M. Selective G to T mutationofp53 gene in hepatocellular carcinoma from southern Africa. Nature (Lond.),350:429-431, 1991.

40. Tada, M., Omata, M., and Ohto, M. High incidence of ras gene mutation

in intrahepatic cholangiocarcinoma. Cancer (Phila.), 69: 1 1 1 5-1 1 1 8, 1 992.

41 . Tsuda, H., Satarug, S., Bhudhisawasdi, V., Kihana, T., Sugimura, T., and

Hirohashi, S. Cholangiocarcinomas in Japanese and Thai patients: difference

in etiology and incidence of point mutation of the c-Ki-ras proto-oncogene.Mol. Carcinog., 6: 266-269, 1992.

42. Thamavit, W., Bhamarapravati, N., Sahaphong, S., Vajrasthira, S., and

Angsubhakorn, S. Effects of dimethylnitrosamine on induction of cholangio-carcinoma in Opisthorchis viverrini-infected Syrian golden hamsters. Cancer

Res., 38: 4634-4639, 1978.

43. FlaveIl, D. J., and Lucas, S. B. Potentiation by the human liver fluke,

Opisthorchis viverrini, ofthe carcinogenic action of N-nitrosodimethylamine

upon the biliary epithelium of the hamster. Br. I. dancer, 46: 985-989, 1983.

44. Thamavit, W., Kongkanuntn, R., Tiwawech, D., and Moore, M. A. Levelof Opisthorchis infestation and carcinogen dose-dependence of cholangio-carcinoma induction in Syrian golden hamsters. Virchows Arch. Abt. B Zell-

pathol., 54:52-58, 1987.

45. Ames, B. N., and Gold, L. S. Chemical carcinogenesis: too many rodent

carcinogens. Proc. NatI. Acad. Sci. USA, 87: 7772-7776, 1990.

46. Cohen, S. M., and Ellwein, L. B. Cell proliferation in carcinogenesis.

Science (Washington DC), 249: 1 007-1 01 1 , 1990.

47. Thamavit, W., Moore, M. A., Ruchirawat, S., and Ito, N. Repeated ex-

posure to Opisthorchis viverrini and treatment with the antihelniinthic Prazi-

quantel lacks carcinogenic potential. Carcinogenesis (Lund.), /3: 309-311,1992.

48. Weitzman, S. A., and Gordon, L. I. Inflammation and cancer: role of

phagocyte-generated oxidants in carcinogenesis. Blood, 76: 655-663, 1 990.

49. Leihovich, S. I.. Polverini, P. 1.. Sheparcl, H. M., Wiseman, D. M., Shively,V., and Nuseir, N. Macrophage-inclucecl angiogenesis is mediated by tumour

necrosis factor-n. Nature (Lond.), 329: 630-632, 1 987.

50. Mathan, C. Nitric oxide as a secretory proclict of mammalian cells.FASEB l.� 6: 3051-3064, 1992.

51 . Schmidt, H. H. H. W., Warner, T. D., and Muracl, F. Double-eclged role

of endogenous nitric oxide Iletterl. Lancet, 339: 986, 1992.

52. Ohshima, H., Tsuda, M., Adachi, H., Ogura, T., Sugimura, T., and Esunii,H. -Arginine-dependent formation of N-nitrosomaines by the cytosol of mac-

rophages activated with Iipopolysaccharicle and interteron-y. Carcinogenesis,

(Lond.), 12:1217-1220, 1991.

53. Thamavit, W., Moore, M. A., Hiasa, Y., and Ito, N. Generation of highyields of Syrian hamster cholangiocellular carcinomas and hepatocellular

nodules by combined nitrite and aminopyrine administration and Opisthor-chis viverrini infection. lpn. I. Cancer. Res., 79: 909-916, 1988.

54. Okuda, K., and Kojiro, M. Small hepatocellular carcinoma. In: K. Okuda

and K. C. Ishak (eds.), Neoplasms ofthe Liver, pp. 21 5-226. Tokyo: Springer-Verlag, 1987.

55. Sun, T. T., Yu, H. T., Hsia, C. C., Wang, N. I., and Huang, X. Y. Evaluation

of sero-survey trials for the early detection of hepatocellular carcinoma in

areas of high prevalence. In:A. B. Miller and I. Chamberlain )ecls.), Screeningfor Gastro-intestinal Cancer, pp. 81-86. Toronto: Hans Huber, 1988.

56. Kamiyana, Y., and Tohe, T. Treatment of primary liver cancer in lapan:

a national study. In: K. Okuda and K. G. Ishak (ecls.), Neoplasnis of the Liver,pp. 375-380. Tokyo: Springer Verlag, 1987.

57. Sripa, B., Pairojkul, C., Bhudhisawasdi, V., Uttaravichien, T., Kular-bkaew, I.. Saew, 0., and Satarug, S. Comparison ofcarcinoenihryonic antigen

(CEA) and CAl 9-9 tumor markers in patients with cholangiocarcinoma. Pro-

ceedings of the First Cancer Research Symposium, pp. 54-62. Khon Kaen:Faculty of Medicine, Khon Kaen University, 1991.

58. Strom, B. L., Maislin, G., West, S. L., Atkinson, B., Herlyn, M., Saul, S.,Rodriguez-Martinez, H. A., Rios-Dalenz, I.. Iliopoulos, D., and Soloway, R.

D. Serum CEA and CA19-9: potential future diagnostic or screening tests forgallbladder cancer? tnt. I. Cancer, 45: 821-824, 1990.

59. Vatanasapt, V., Sriamporn, S., Mairiang, E., Chaiyakhani, I., Haswell-Elkins, M. R., Chamadol, N., Srinagarindthra, I.. Laopaiboon, V., Kanpittaya,

I., Nitinawakarn, B., and Pipitgool, V. Epicleniiologic study of liver cancerusing a population-based cancer registry as a guide in Khon Kaen, Thailand.

Health Rep., 5: 51-58, 1993.

60. Bunnag, D., and Harinasuta, T. Studies on the chemotherapy of human

opisthorchiasis in Thailand, I: Clinical trial of Praziquantel. Southeast Asian

I. Trop. Med. Public Health, 1 1: 528-531, 1980.

61 . Bunnag, D., and Harinasuta, T. Studies on the chemotherapy of humanopisthorchiasis Ill: Minimum effective dose of Praziquantel. Southeast Asian

I. Trop. Med. Public Health, /2:413-417, 1981.

62. Upatham, E. S., Viyanant, V., Brockelman, W. Y., Kurathong, S., Lee, P.,and Kaengraeng, R. Rate of reinfection by Opisthorchis vit’errini in an en-demic northeast Thai community after cheniotherapy. tnt. I. Parasitol., 18:643-649, 1988.

63. Zelen, M. Are primary cancer prevention trials feasible? I. NatI. Cancer

Inst., 80: 1442-1444, 1988.

64. Vatanasapt, V., Martin, N., Sriplung, H., Chindavijak, K., Sontipong, S.,

Sriamporn, S., Parkin, D. M., and Ferlay, I. (eds.). Cancer in Thailand, 1988-1 991 . ARC Technical Report No. 1 6. Lyon, France: International Agency for

Research on Cancer, 1993.

65. Parkin, D. M., Pisani, P., and Ferlay, I. Estimates of the worldwide in-

ciclence ofeighteen major cancers in 1 985. Int. I. Cancer, 54: 594-606,1 993.



1993;2:537-544. Cancer Epidemiol Biomarkers Prev D M Parkin, H Ohshima, P Srivatanakul, et al. carcinogenesis and prevention.Cholangiocarcinoma: epidemiology, mechanisms of

Updated version

http://cebp.aacrjournals.org/content/2/6/537

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cebp.aacrjournals.org/content/2/6/537To request permission to re-use all or part of this article, use this link



http://cebp.aacrjournals.org/cgi/alerts

mailto:[email protected]



Cholangiocarcinoma: Epidemiology, Mechanisms of ...Vol. 2, 537-544, November/December 1993 Cancer...

Documents

Transcript of Cholangiocarcinoma: Epidemiology, Mechanisms of ...Vol. 2, 537-544, November/December 1993 Cancer...