Oncogene (2006) 25, 3771–3777
& 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00
www.nature.com/onc
Epidemiology of hepatocellular carcinoma in areas of low hepatitis B and
hepatitis C endemicity
LB Seeff and JH Hoofnagle
Liver Disease Research Branch, Division of Digestive Diseases and Nutrition, National Institute of Diabetes and Digestive and Kidney
Diseases, National Institutes of Health, Bethesda, MD, USA
Hepatocellular carcinoma (HCC) ranks among the 10
most common cancers worldwide. It evolves from several
chronic liver diseases, most of which culminate in
cirrhosis. As the most common causes, other than
alcoholic cirrhosis, are chronic hepatitis B and C
infections, its prevalence worldwide is linked to the
prevalence of these two viruses. Thus, the highest rates
are in southeast Asia and sub-Saharan Africa, the world’s
most populous nations, where hepatitis B virus infection
is endemic. In most western countries, hepatitis C virus
infection is the predominant cause, and hepatitis B-related
liver cancer occurs largely among immigrants from
countries of high hepatitis B endemicity. In most western
countries, the incidence and mortality from HCC is
increasing as a consequence of the chronic sequelae of the
‘epidemic’ of hepatitis C of the 1960–1980s. In the US,
modeling of this infection predicts a continued rise in liver
cancer over the next decade. Surveillance by the National
Cancer Institute and the Centers for Disease Control
confirms the increasing incidence of and mortality from
HCC to the year 2000, although subsequent analyses
suggest a slowing or possibly decline in the rate of
increase. Whether this trend will continue requires further
evaluation.
Oncogene (2006) 25, 3771–3777. doi:10.1038/sj.onc.1209560
Keywords: hepatocellular carcinoma; hepatitis B; hepatitis C
Introduction
Hepatocellular carcinoma (HCC) ranks in prevalence
and mortality among the top 10 cancers worldwide
(Bosch et al., 2004). Originally identified as a sequel to
alcoholic cirrhosis, hemochromatosis, some metabolic
disorders, certain types of drug injury (such as from oral
contraceptives), or exposure to environmental toxins
(such as aflatoxin), the cause of the bulk of cases
remained obscure for many years, although linked in
Correspondence: Dr LB Seeff, NIDDK, Liver Disease Research
Branch, Division of Digestive Diseases and Nutrition, National
Institute of Diabetes and Digestive and Kidney Diseases, National
Institutes of Health, 27A Center Drive, Room 9A27, Bethesda, MD
20892, USA. E-mail: [email protected]
most instances to cirrhosis of unknown origin. This
uncertainty held sway until the discovery of the hepatitis
B virus (HBV) in the mid 1960s (Blumberg et al., 1965),
and its subsequent association with HCC (Beasley et al.,
1981), and the recognition about 20 years later of the
hepatitis C virus (HCV) (Choo et al., 1989) with the
subsequent evidence that it, too, was linked to HCC
(Bruix et al., 1989). It, thus, became clear that the vast
majority of cases of HCC were in fact a consequence of
infection by one of these two viruses. That HCC was
among the most common of cancers worldwide could
then be accounted for by the fact that HBV infection,
the leading cause for HCC, is endemic in southeast Asia
(Beasley et al., 1981), the most populous area in the
world, as well as in sub-Saharan Africa. More recently,
it has become apparent that obesity, with its accompanying problems of diabetes mellitus and nonalcoholic
steatohepatitis (NASH), represents a growing addition
to the causes of HCC (Marrero et al., 2002; Caldwell
et al., 2004; El-Serag et al., 2004). Furthermore,
evidence is accumulating that both diabetes and chronic
alcoholism behave synergistically with HBV and HCV
infections in the induction of HCC (Hassan et al., 2002).
Indeed, even past heavy alcoholism can contribute to
liver disease progression in persons chronically infected
with hepatitis C and can hence increase the risk of
developing HCC (Delarocque-Asatagneau et al., 2005).
Global prevalence and mortality from hepatocellular
carcinoma
In the year 1990, there were an estimated 437, 408 cases
of HCC worldwide, among whom approximately threequarters were men (Parkin et al., 1999; Bosch et al.,
2004) (Table 1). At that time, HCC ranked fifth among
cancers in men, well behind cancers of the lung and
stomach, and also a little behind colon/rectum and
prostate cancers. Among women, HCC ranked eighth.
By far, the highest numbers of cases were reported from
China (178 100 in men and 57 500 in women) followed in
frequency by the numbers reported from other south
eastern Asian countries, Japan and western Africa. The
gender difference in rates of HCC was less evident in
North America, where 5400 men and 3000 women were
reported to have this tumor.
Liver cancer in areas of low hepatitis frequency
LB Seeff and JH Hoofnagle
3772
Table 1
Global frequency of new cases of hepatocellular carcinoma
Year (reference)
1990 (Parkin et al. (1999);
Bosch et al. (2004))
2000 (Parkin et al. (2001))
2002 (The World health
report (2003))
Total number
Males
Females
Table 2 Estimated age-adjusted incidence rates of hepatocellular
carcinoma in men per 100 000 population worldwide, 2000 (Ferlay
et al., (2001))
437 408
316 300
121 100
Geographic area
Region
564 300
714 600
398 364
504 600
165 972
210 000
Asia
Eastern
South-Eastern
Western
35.5
18.3
5.6
Africa
Middle
Eastern
Western
Southern
Northern
24.2
14.4
13.5
6.2
Europe
Southern
Eastern/western
Northern
By 2000, the estimated number of new cases of HCC
had risen to 564 300, with 70 percent of cases occurring
in men (Parkin et al., 2001) (Table 1). In the same year,
548 600 people with HCC died, representing 97.2 percent
of those with this diagnosis. Not unexpectedly, 77
percent of the cases occurred in Asia and 7.4 percent in
Africa (Ferlay et al., 2001; Bosch et al., 2004). In
contrast, the 12 543 cases of HCC in North America
constituted 2.2 percent of the total number of cases of
HCC recorded. When viewed as estimated age-adjusted
incidence rates of liver cancer per 100 000 men (Table 2),
the figures: in Asia ranged as follows, from 35.5 in
eastern Asia, 18.3 in southeastern Asia, to 5.6 in western
Asia; in Africa, from 24.2 in Middle Africa, 14.4
in eastern Africa, 13.5 in western Africa, 6.2 in southern
Africa, to 4.9 in Northern Africa; in Europe, from 9.8 in
southern Europe, 5.8 in eastern and western Europe, to
2.6 in Northern Europe; and to values of 4.8 in south
America; 4.1 in North America; 3.6 in Australia/New
Zealand and, finally, 2.1 in central America. In all
regions, the rates recorded were two to three times
higher in men than in women. The lethality of this
cancer is evident by the fact that, in the above databases,
mortality from HCC was virtually equivalent to the
incidence figures in the same year.
The most recent worldwide data on HCC come from
the World Health Report, 2003 of the World Health
Organization (WHO) that provides estimates of death
by cause and sex in WHO regions for the year 2002 (The
World health report, 2003). The estimated burden of
disease by cause indicated that there were a total of
714 600 new cases of HCC (71 percent among men)
(Table 1). Liver cancer ranked fourth in mortality due to
cancer, after cancers of the trachea/bronchus/lung,
stomach, and colon/rectum. For male subjects, liver
cancer ranked third; whereas for women, it ranked fifth.
Geographically, there were 45 000 liver cancer deaths in
Africa, 37 000 in the Americas, 15 000 in the eastern
Mediterranean, 67 000 in Europe, 61 000 in southeast
Asia, and 394 000 in the western Pacific, which includes
China and Japan. In the same year, death from cirrhosis
was reported in 783 000 persons, consisting of 501 000
men and 282 000 women.
Viewing these three time periods, it is apparent that
the numerical frequency of HCC has increased with each
report, and it is probable these data represent valid
tracking of the trend. However, strict comparisons
cannot be made because of the use of differing ICD
coding data sources over the course of these time
periods, differences in diagnostic criteria, and uncertainty of reporting. Suffice it to say, HCC is a common
Oncogene
Rate/100 000
South America
North America
Australia/New Zealand
Central America
4.9
9.8
5.8
2.6
4.8
4.1
3.6
2.1
cancer worldwide, whose frequency seems to be
increasing, tied largely to infections with hepatitis B
and C and possibly to the newly emerging epidemic of
obesity and its associated fatty liver disease, the latter
more common in western than in eastern countries,
although obesity is increasing even in some eastern
countries. The relative contributions of hepatitis B and
hepatitis C to the development of HCC clearly differ
among the various countries of the world.
Geographic distribution of hepatitis B and C viral
infections
Hepatitis B is the most common cause of HCC, the virus
being endemic to Asia and Africa but occurring in far
lower frequency in western countries. As both chronic
hepatitis B and C play a dominant role in the etiology of
HCC, it is worth reviewing the worldwide prevalence of
these two viruses to determine their role as risk factors
for HCC in different geographic areas of the world.
Data from WHO indicate that there are an estimated
2 billion people who have been infected with HBV
worldwide, among whom 350 million have chronic HBV
infection (Lavanchy, 2004) (Table 3). There is marked
geographic variability in the HBV prevalence, with the
chronic infection involving fewer than 2 percent of the
populations of North America and western and northern European countries; between 2 and 7 percent of
peoples in the southern parts of eastern and central
Europe, the Amazon basin, the Middle East, and the
Indian subcontinent; and greater than 8 percent of
persons in Asia, sub-Saharan Africa and the Pacific.
With regard to hepatitis C, it is believed that there are
170 million infected persons globally, and although
there are some geographic differences in prevalence, the
variability is far less than exists with hepatitis B (http://
www.who.int/mediacentre/factsheets/fs164/en/print.html)
Liver cancer in areas of low hepatitis frequency
LB Seeff and JH Hoofnagle
3773
Table 3
Prévalence (%)
Worldwide Distribution of Hepatitis B and C
Hepatitis C (carriers, 170 million)a
Hepatitis B (Lavanchy, 2004) (carriers, 350 million)
Geographic location
Prevalence (%)
Geographic location
>8
Asia, sub-Saharan Africa, Pacific
5.3
4.6
3.9
Africa
E. Mediterranean
W. Pacific
2–7
Eastern/central Europe, Amazon basin, Middle East, Indian sub-continent
2.2
1.7
1.0
S-E Asia
Americas
Europe
o2
North America, western/northern Europe
a
http://www.who.int/mediacentre/factsheets/fs164/en/print.html
(Table 3). Data from WHO estimate the worldwide
prevalence of hepatitis C to be 3.1 percent, ranging from
5.3 percent in Africa, to 4.6 percent in the eastern
Mediterranean, 3.9 percent in the western Pacific, 2.2
percent in southeast Asia, 1.7 percent in the Americas,
and 1.0 percent in Europe.
Incidence and mortality rates of hepatocellular carcinoma
in the US
Information on cancers in the US is tracked and
recorded by the Surveillance, Epidemiology, and End
Results (SEER) program of the National Cancer
Institute (NCI) and National Program of Cancer
Registries (NPCR) maintained by the Centers for
Disease Control and Prevention (CDC) (Hankey et al.,
1999; Hutton et al., 2001; National Cancer Institute,
2005; National Center for Chronic Disease Prevention
and Health Promotion, 2005). In 1999, data from the
SEER program indicated that for liver and intrahepatic
bile duct cancers combined, the estimated total number
of new cases in the US was 14 500 (9600 men; 4900
women) with an estimated number of deaths of 13 600
(8400 men; 5200 women).
Incidence rates of HCC, expressed as a number per
100 000, age-adjusted to the 1970 US standard population over the period 1992–1996, are shown in Table 4a
(National Cancer Institute, 2005). Thus, the total
incidence for all races and both sexes combined was
3.1, the frequency among men being three times that of
women. Moreover, the frequency was higher among
blacks than among whites, the highest figure being that
of 7.6 among black males. Analysis for secular trends in
the incidence of liver cancer from 1973 to 1996
(Table 4b) demonstrated a general increase of 59.8
percent, an estimated annual percent change (APC) over
that time interval of 2.2 percent, placing combined liver
and intrahepatic bile duct cancer fifth among all those
cancers showing an increase in frequency (following
melanoma, lung cancer in female subjects, prostate
carcinoma and non-Hodgkins lymphoma) (Hutton
et al., 2001). The trends in incidence between 1973 and
1996 according to race are also shown in Table 4b.
Table 4 (a) Age-Adjusted SEER Incidence and Mortality Rates of
Hepatocellular Carcinoma in the US 1992–1996 (National Cancer
Institute (2005)) (b) Trends in Incidence of Hepatocellular Carcinoma
in the US 1973–1996 (National Cancer Institute (2005))
(a)
Incidencea
Mortalitya
All races
Total
Males
Females
3.1
4.9
1.6
2.8
4.2
1.7
Whites
Total
Males
Females
2.4
3.7
1.3
2.5
3.7
1.5
Blacks
Total
Males
Females
4.7
7.6
2.4
4.2
6.6
2.5
Percent changeb
Annual percent changec (APC)
All races
Total
Males
Females
59.8
72.9
31.3
2.2
2.5
1.4
Whites
Total
Males
Females
37.8
55.3
6.7
1.6
2.0
0.6
Blacks
Total
Males
Females
53.2
42.3
78.5
1.9
2.2
1.4
Group
(b)
Group
Rates per 100 000 age-adjusted to the 1970 U.S. standard population.
a
Incidence and mortality rates are per 100 000 age-adjusted to 1970 US
standard population. bPercent change over the time interval. cEstimated annual percent change over the time interval.
In the same SEER program analysis, the trend in the
incidence and mortality of the top 20 cancers was
ranked over the period 1973–1996, linking together
cancer of the liver and intrahepatic bile ducts. Thus, the
incidence rates among male subjects placed liver and
Oncogene
Liver cancer in areas of low hepatitis frequency
LB Seeff and JH Hoofnagle
3774
intrahepatic bile duct cancer in fourth place, after
melanoma of the skin, prostate carcinoma and nonHodgkins lymphoma, whereas the mortality rates
ranked liver and intrahepatic bile duct cancer in first
place. For female subjects, liver and intrahepatic bile
duct carcinoma was placed fourth, after lung cancer,
melanoma, and non-Hodgkins lymphoma, whereas
mortality was ranked in fifth place.
The most recent available data on the incidence and
mortality of HCC in the US cover the period up to 2002.
There are two sources of this information.
The first is the SEER Cancer Statistics Review, 1975–
2002, assembled by the NCI (Reis et al., 2005). Focusing
again on liver and intrahepatic bile duct cancer
combined, they report the SEER incidence and US
mortality using the joinpoint regression program (http://
srab.cancer.gov/joinpoint/; Kim et al., 2000). This
program is statistical software for the analysis of trends
using models where several different lines are connected
together at the ‘joinpoint.’ This enables the user to test
that an apparent change in the trend is statistically
significant at a given time point. The incidence of liver
and intrahepatic bile duct cancer for what is referred to
as joinpoint segment 1, covering the period 1975–1984,
showed an APC for all races and both sexes combined of
1.4 percent, increasing after 1984 to 4.6 percent,
statistically different from the earlier period, and then
declining after 1999 to 2.4 percent; an impressive,
although, nonsignificant decrease (Table 5). A similar
trend was noted for mortality, the most recent period
(1995–2002) showing a slight but significant slowing of
the rate of increase.
A second report combined data gathered by the
American Cancer Society (ACS), CDC, NCI and North
American Association of Central Cancer Registries
(NAACCR) (Edwards et al., 2005). This report focused
on the 15 most common cancers in the US, incidence
figures being derived from the NCI, CDC and
NAACCR, and mortality data from the CDC. Reported
as incidence and death rates age-adjusted in the year
2000 US standard populations, the methods employed
included linear regression and joinpoint regression
analysis. In general, the incidence and mortality figures
report differed only slightly from those described above
Table 5 Trends in SEER Incidence and Mortality of Hepatocellular
and Intrahepatic Bile Duct Carcinoma in the US, 1975–2002 (Reis
et al., 2005)
Group
Period
Annual percent changea (APC)
Incidence
All races (both genders) 1975–1984
1984–1999
1999–2002
1.4
4.6*
2.4
Mortality
All races (both genders) 1975–1978
1978–1987
1987–1995
1995–2002
0.8
1.7*
4.0*
1.4*
a
Age-adjusted to the 2000 U.S. standard population. *Significantly
different from zero (Po0.05) using the joinpoint regression program.
Oncogene
but showed identical trends. Thus, the APC for the
incidence rate trends of liver and intrahepatic bile duct
cancer for the period 1975–1984 was 1.7, for 1984–1999
was 4.5, and for 1999–2002 was 2.1 (results not
shown). Ranking the incidence rates of the 15 most
common cancers for the years 1992–2002 showed that
among male subjects of all races, liver and intrahepatic
bile duct cancer combined ranked twelfth with a rate of
8.6 per 100 000 persons and an APC of 3.0 percent
(Table 6a). Combined liver and intrahepatic bile duct
cancer ranked eighteenth among women, with a rate of
3.3 per 100 000 and an APC of 3.0 percent. Differences
were found, however, among the races and by gender
(Table 6a).
Mortality rates placed liver and intrahepatic bile duct
cancers higher, ranking 10th for men and 13th for
women overall. Death rates according to race followed
the same pattern as the incidence rates (Table 6b).
Mortality rates were highest for Asian/Pacific Islanders
and for Hispanics/Latinos.
These data indicate that there was a rapid increase in
the incidence and mortality rate of liver cancer in the
mid-1980s through to the late 1990s, after which the
rates of both appear to have leveled off or have even
begun to show a slight decline. This is surprising as a
number of epidemiologic modeling studies have concluded that the incidence of HCC will continue to rise in
the US over the next decade as a consequence of the
‘epidemic’ of acute hepatitis C that occurred in the 1960s
to the early 1980s, and only then it will begin to decline
(Armstrong et al., 2000; Wong et al., 2000). There is no
ready explanation for the apparent premature decline in
the frequency of and reduction in mortality from HCC.
This decline may be a statistical quirk, so the next survey
will once again confirm that the HCC incidence and
mortality continue to climb. Perhaps the relatively short
duration of observation during the last period (a total of
4 years) as compared to the earlier observation periods
of 10–15 years has distorted the true frequency. Also, as
the data reviewed include liver and intrahepatic bile duct
cancers together, it is possible that an increase in the
frequency of one of the components, namely HCC, is
obscured by an even greater decrease in the incidence of
the other cancer, namely of the bile ducts.
There is currently a widespread consensus view, based
on broad clinical experience, that the number of cases of
HCC being seen in medical clinics in the US continues to
increase, even to the present. Moreover, important
epidemiologic surveys have clearly demonstrated that
HCC alone has significantly increased in the US over the
past decade (El-Serag and Mason, 1999; El-Serag et al.,
2003). What follows is a review of these reports.
Surveys to determine the incidence and etiology of
hepatocellular carcinoma in the US
To study the frequency, mortality, and also the etiology
of HCC in the US, El-Serag and co-workers reviewed
the SEER databases that had been linked to Medicare
Liver cancer in areas of low hepatitis frequency
LB Seeff and JH Hoofnagle
3775
claims from the Centers for Medicare and Medicaid
Services (Warren et al., 2002) in order to establish risk
factors for HCC (Davila et al., 2004). The study
population were all persons 65 years of age or older
with a diagnosis in the SEER registries of HCC who had
enrolled in Medicare between 1993 and 1999. Among
4015 patients identified as having HCC in these
databases, 2584 satisfied the inclusion criteria for
the study. Their mean age was 74, two-thirds were male
subjects, 67 percent were white, 9 percent were black, 4
percent were Hispanic, 12 percent were Asian and 9
percent were of other race. The age-adjusted incidence
of HCC overall was 16.2 and, when examined over time,
increased from 14.2 per 100 000 in 1993 to 18.1 per
100 000 in 1999. Regarding etiology (Table 7), however,
no cause could be identified in 41.3 percent of cases,
excess alcohol was believed to be responsible in 21.2
percent, nonspecific cirrhosis in 17.1 percent, HCV
infection in 16.3 percent, HBV infection in 8.8 percent
and nonspecific hepatitis in 4.0 percent of the cases. In
some instances, these etiologic factors overlapped. In
comparing the contributing etiologic factors between
two time periods, 1993–June 1996 and July 1996–1999,
the association with HCV increased from 11.0 percent in
the first period to 21.2 percent in the second, HBV
increased from 6.4 percent in the first period to 11.0
percent in the second, alcohol-induced disease showed
little change (20.6 percent to 22.0 percent, respectively),
whereas the contribution of the other factors fell
(Table 7). Using multiple logistic regression analyses,
the adjusted odds ratio, associating the time period of
HCC diagnosis with the risk factors, was 2.26 for HCV,
1.67 for HBV, 1.16 for alcoholic liver disease, 0.84 for
nonspecific cirrhosis and 0.83 for the idiopathic group.
These authors also compiled data showing that there
has been an increase in age-adjusted incidence rates of
HCC computed at 3-year intervals for the period
between 1975 and 2002 (El-Serag, 2004). Thus the ageadjusted incidence rate per 100 000 that was 1.0 among
whites between 1976 and 1978 was found to have risen
slowly to reach 2.5 in the period 2000–2002. Similarly,
the rate for blacks, starting at 2.5 between 1976 and
1978, reached 5.0 for the period of 2000–2002. Finally,
the rate among other racial groups (predominantly
Asians) started at 6.0 between 1976 and 1978 and
subsequently rose but appeared to have reached a
Table 6 (a) Ranking, incidence rates, and annual percent changes of
hepatocellular and intrahepatic bile duct carcinomas in the US, 1992–
2002 (Edwards et al., 2005) (b) Ranking, mortality rates, and annual
percent changes of hepatocellular and intrahepatic bile duct carcinomas in the US, 1992–2002 (Edwards et al., 2005)
Group
Rank
Rate/100 000
(a)
All races
Males
Females
12
18
8.6
3.3
3.0
3.0
Whites
Males
Female
15
18
6.8
2.7
2.9
3.7
Blacks
Males
Females
14
17
10.8
3.6
4.5
1.4
Asian/Pacific Islanders
Males
5
Females
11
20.9
7.9
1.0
0.2
American Indian/Alaskan natives
Males
8
Females
12
9.0
5.6
Annual percent
change (APC)
—
—
Hispanics/Latinos
Males
Females
8
13
13.4
5.4
2.2
5.0
(b)
Whites
Male
Female
12
14
5.9
2.7
2.1
1.1
Blacks
Male
Female
8
12
9.2
3.7
1.3
0.6
Asian/Pacific Islanders
Male
3
Female
6
15.9
6.5
0.6
0.7
7.6
4.1
1.6
1.7
10.3
4.8
1.6
2.1
American Indian/Alaskan natives
Male
4
Female
7
Hispanics/Latinos
Male
Female
4
8
Table 7
Etiologies of Hepatocellular Carcinoma in the US As Reported in Two Studies
El-Serag (Davila et al., 2004)
Unknown (%)
Alcohol (%)
NSP* cirrhosis. (%)
Hepatitis C (%)
HCV+Alcohol
Hepatitis B (%)
NSP* hepatitis (%)
Marrero (Marrero et al., 2002)
1993–June 1996
July 1996–1999
Total
Total
43.5
20.6
18.5
11.0
—
6.4
4.2
38.5
22.0
15.7
21.2
—
11.0
3.7
41.3
21.2
17.1
16.3
—
8.8
4.0
4.0
10.0
—
39.0
12.0
6.0
29.0
NSP ¼ Nonspecific.
Oncogene
Liver cancer in areas of low hepatitis frequency
LB Seeff and JH Hoofnagle
3776
plateau at 8.4 in 1994–1996 and then decreased slightly
thereafter. These data, therefore, indicate forcefully that
the incidence of HCC has been increasing in the US, but
the data reviewed reached just to the year 2000. The
apparent decline in the incidence and mortality of liver
cancer and intrahepatic bile duct cancer identified above
in SEER and CDC databases seemed to have emerged
after the year 2000. Whether or not this represents a real
trend remains to be determined by future surveillance.
In yet another effort to establish etiology for HCC in
the US, Marrero and co-workers evaluated 105 cases
with this diagnosis seen at a single academic medical
between 2000 and 2002 for potential risk factors
(Marrero et al., 2002) (Table 7). HCV was the underlying cause for the liver disease in 51 percent (39 percent
had only HCV as a risk factor and another 12 percent
also had a history of chronic alcoholism), 29 percent had
cryptogenic liver disease, 10 percent had only heavy
alcoholism as a potential etiology, 6 percent were
infected with HBV and 4 percent were associated with
other conditions. Of much interest was that, among
those with cryptogenic cirrhosis, one-half had either
biopsy evidence of NASH or clinical evidence compatible with nonalcoholic fatty liver disease. Thus, in both
studies, while the responsible cause for HCC could not
be determined in a third to almost one-half of the cases,
infection with HCV, and to a lesser extent, HBV,
accounted for the bulk of cases. But clearly, fatty liver
disease, especially NASH, appears to be an important
and probably growing contributor to the etiology of
HCC in the US.
Epidemiology of hepatocellular carcinoma in Canada
Like the US, its northern neighbor, Canada, is also
experiencing a rising incidence of HCC. In a recent
report from Canada, the age-adjusted incidence of HCC
increased in male subjects from 4.0 per 100 000 in 1984
to 5.5 per 100 000 in the year 2000, and in female
subjects, from 1.6 to 2.2 per 100 000 (Dyer et al., 2005).
It was also reported that there has been an increase in
mortality from HCC of 48 percent in male subjects and
of 39 percent in female subjects. The primary etiologies
were considered to be HCV infection and the obesity/
diabetes problem, both of which have shown a
sharp increase in frequency in Canada during the past
decade.
Epidemiology of hepatocellular carcinoma in developed
countries
The highest rates of HCC occur in countries where HBV
predominates. However, it appears that the incidence in
some of these countries is beginning to fall (Deuffic
et al., 1999; McGlynn et al., 2001; El-Serag, 2002). In
contrast, the incidence of HCC is increasing in France,
Italy, Spain, Switzerland, Australia, New Zealand and,
especially, Japan (El-Serag, 2002). In these countries,
HCV infection is the primary source of chronic liver
disease and, ultimately, of HCC.
Summary
There has been a slow increase in the frequency of and
mortality from HCC in North America over the past
10–15 years, attributable in large part to infection with
the HCV and possibly to the problem of obesity and the
accompanying nonalcoholic steatohepatitis. Such increases have also occurred in parts of Europe, Australia,
New Zealand and Canada. There is now some evidence
that the increase in the incidence and mortality from
HCC has slowed down in the US, but it is unclear
whether this is a transient phenomenon or whether this
represents a definite trend, and must await confirmation
through further epidemiologic surveys. Presently, it is
believed that the incidence of HCC will continue to
climb in these western countries, where HBV infection is
not endemic for at least another decade, after which it
should begin to decline because of the decreasing
number of new cases of hepatitis C. What role the
‘epidemic’ of obesity and diabetes will play in the future
needs further exploration and close surveillance.
References
Armstrong GL, Alter MJ, McQuillan GM, Margolis HS.
(2000). Hepatology 31: 777–782.
Beasley RP, Hwang LY, Lin CC, Chien CS. (1981). Lancet 2:
1129–1133.
Blumberg BS, Alter HJ, Visnich S. (1965). JAMA 191:
541–546.
Bosch FX, Ribes J, Diaz M, Cleries R. (2004). Gastroenterology
127: S5–S16.
Bruix J, Barrera JM, Calver X, Ercilla G, Costa J, SanchezTapias JM et al. (1989). Lancet 2: 1004–1006.
Caldwell SH, Crespo DM, Kang HS, Al-Osaimi AMS. (2004).
Gastroenterology 127: S97–S103.
Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW,
Houghton M. (1989). Science 244: 359–362.
Data available at:http://www.who.int/mediacentre/factsheets/
fs164/en/print.html.
Oncogene
Data available at:http://srab.cancer.gov/joinpoint/.
Davila JA, Morgan RO, Shaib Y, McGlynn A, El-Serag HB.
(2004). Gastroenterology 127: 1372–1380.
Delarocque-Asatagneau E, Roudot-Thoraval F, Campese C,
Desenclos J-C, the Hepatitis C Surveillance System Steering
Committee. (2005). Ann Epidemiol 15: 551–557.
Deuffic S, Poynard T, Valleron J. (1999). J Viral Hepat 6: 411–413.
Dyer Z, Peltakian K, Veldhuyzen van Zanten S. (2005).
Aliment Pharmacol Therap 22: 17–22.
Edwards BK, Brown ML, Wingo PA, Howe HL, Ward E,
Ries LAF et al. (2005). J Natl Cancer Inst 97: 1407–1427.
El-Serag HB. (2002). J Clin Gastroenterol 35(Suppl 2):
S72–S78.
El-Serag HB. (2004). Gastroenterology 127: S27–S34.
El-Serag HB, Davila J, Petersen N, McGlynn K. (2003).
Ann Intern Med 139: 817–823.
Liver cancer in areas of low hepatitis frequency
LB Seeff and JH Hoofnagle
3777
El-Serag HB, Mason AC. (1999). N Engl J Med 340:
745–750.
El-Serag HB, Tran T, Everhart JE. (2004). Gastroenterology
126: 460–468.
Ferlay J, Bray F, Pisani P, Parkin DM. (2001). GLOBOCAN
2000: Cancer incidence, mortality and prevalence worldwide,
Version 1.0. Lyon: IARC Press.
Hankey BF, Ries LA, Edwards BK. (1999). Cancer Epidemiol
Biomarkers Prev 8: 1117–1121.
Hassan MM, Hwang LY, Hatten CJ, Swaim M, Li D,
Abbruzzese JL et al. (2002). Hepatology 36: 1206–1213.
Hutton MD, Simpson DL, Miller DS, Weir HK, McDavid K,
Hall HI. (2001). J Registry Manage 28: 113–120.
Kim HJ, Fay MP, Feuer EJ, Midthune DN. (2000). Stat Med
19: 335–351 (correction: 2001;20:655).
Lavanchy D. (2004). J Viral Hepatitis 11: 97–107.
Marrero JA, Fontana RJ, Su GL, Conveejaram HS, Emick
DM, Lok AS. (2002). Hepatology 36: 1349–1354.
McGlynn KA, Tsao L, Hsing AW et al. (2001). Int J Cancer
94: 290–296.
National Cancer Institute (2005) About SEER. Available
athttp://seer.cancer.gov/about/.
National Center for Chronic Disease Prevention and Health
Promotion (2005) National Program of Cancer RegistriesCancer Surveillance System. June 20, 2005. Available
at:http://www.cdc.gov/cancer/npcr/css.htm.
Parkin DM, Bray F, Ferlay J, Pisani P. (2001). Int J Cancer 94:
153–156.
Parkin DM, Pisani P, Ferlay J. (1999). Int J Cancer 80:
827–841.
Reis LAG, Eisner MP, Kosary CL, Hankey BF, Miller BA,
Clegg L et al. (eds) (2005). SEER Cancer Statistics Review,
1975–2002. National Cancer Institute, Bethesda, MD.
http://seer.cancer.gov/csr/1975_2002/.
The World health report (2003): shaping the future. World
Health Organization, 2003.
Warren JL, Klabunde CN, Schrag D, Bach PB, Riley GF.
(2002). Med Care 40(suppl 8): IV3–IV18.
Wong JB, McQuillan GM, McHutchison JG, Poynard T.
(2000). Am J Public Health 20: 1562–1569.
Oncogene