Correspondence
The Rising Incidence of Hepatocellular Carcinoma
N Engl J Med 1999; 341:451-452August 5, 1999
- Article
To the Editor:
El-Serag and Mason (March 11 issue)1 report an increase in the incidence of hepatocellular carcinoma in the United States and suggest that the hepatitis B virus (HBV) and hepatitis C virus are the cause. Similar trends in the incidence of hepatocellular carcinoma have been reported worldwide.2 We have also found an appreciable increase in the age-standardized incidence of hepatocellular carcinoma in Göteborg, Sweden, from 2.3 cases per 100,000 population to 3.6 cases during a 22-year period.3 However, when we used a Poisson model,4 we found no significant increase in incidence after adjusting for age, either in men or in women.5
It is well known that the risk of most tumors is age-dependent. The changing age structure of a population thus may strongly influence the incidence of cancer. To lessen this effect on studies of time trends within a population, an age-standardized incidence method has been used. With the use of this method, many studies have shown an increased incidence of hepatocellular carcinoma in both high- and low-incidence areas,1-3 suggesting an increased exposure to possible carcinogenic factors.
In our study, the generally accepted age-standardized incidence method gave the impression of a substantially increased incidence over time, a result that could not be confirmed by the Poisson model. The reason for these seemingly contradictory results seems to be that the age-standardized incidence method does not fully account for the aging of a population, and hepatocellular carcinoma is an extremely age-dependent cancer.5 Thus, one may question whether the incidence of hepatocellular carcinoma is really increasing. We suggest that in studies of time trends of the incidence of hepatocellular carcinoma, the Poisson model is more informative than the age-standardized incidence method.
5 ReferencesJerzy Kaczynski, M.D., Ph.D.
Sahlgrenska–Östra University Hospital, S-416 85 Göteborg, SwedenAnders Odén, Ph.D.
Valler 190, 442 92 Romelanda, Sweden1
El-Serag HB ,Mason AC . Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999;340:745-750
Full Text | Web of Science | Medline2
Saracci R ,Repetto F . Time trends of primary liver cancer: indication of increased incidence in selected cancer registry populations. J Natl Cancer Inst 1980;65:241-247
Web of Science | Medline3
Kaczynski J ,Hansson G ,Wallerstedt S . Incidence of primary liver cancer and aetiological aspects: a study of a defined population from a low-endemicity area. Br J Cancer 1996;73:128-132
CrossRef | Web of Science | Medline4
Breslow NE, Day NE. Statistical methods in cancer research. Vol. 2. The design and analysis of cohort studies. Lyon, France: International Agency for Research on Cancer, 1987. (IARC scientific publications no. 82.)
5
Kaczynski J ,Oden A . Increasing incidence of primary liver cancer -- matter of ageing? GI Cancer 1999;3:67-71
Web of Science
To the Editor:
In their otherwise illuminating editorial on the increasing incidence of hepatocellular carcinoma, Ince and Wands1 make a serious misstatement regarding infection with HBV. Their assertion that “among those who become infected with HBV at birth, men have an estimated lifetime risk of hepatocellular carcinoma of 50 percent, and women have a risk of 20 percent” is incorrect. The reference provided in support of this statement is a 1981 study of 22,707 Taiwanese men,2 3454 of whom were positive for the hepatitis B surface antigen (HBsAg). In that study, there were 40 deaths from primary hepatocellular carcinoma among affected men during a mean follow-up period of 3.3 years. The mean annual rates of death from hepatocellular carcinoma are shown according to age at diagnosis in Table 1Table 1
Mean Annual Rates of Death from Hepatocellular Carcinoma..The ability to draw any conclusions about the lifetime risk of hepatocellular carcinoma is severely limited by the absence of data on the life expectancy of the population, the age at onset of infection, and the time to occurrence of hepatocellular carcinoma. Even so, for a hypothetical life expectancy of 72 years, the data support a probability of hepatocellular carcinoma of no more than 21.5 percent for a male life-long carrier. The true death rate — and by extension, the risk of hepatocellular carcinoma — is likely to be substantially lower than that, given the strong indication of selection bias (as evidenced by the decreasing annual rates of death from hepatocellular carcinoma over the first three years of the study, from 0.43 percent to 0.21 percent); given that over half of all projected excess mortality is observed among patients 60 years of age or older (whose numbers are so small as to render the data statistically unreliable); and given the failure to control for any other types of chronic liver disease and alcohol use. With regard to the alleged 20 percent lifetime risk of hepatocellular carcinoma in HBsAg-positive women, the study presents no data whatsoever.
Although Ince and Wands are right to call attention to the mortality associated with chronic HBV infection, their unsupportable assertion that, among persons infected at birth, there is a 50 percent likelihood of a cancer that predestines literally hundreds of millions of people to premature death has an incalculable potential for engendering harmful psychosocial consequences on a global scale. It is extremely important for physicians and their HBsAg-positive patients to know the reality — that a large majority of people with HBsAg infections acquired at birth can expect to live full lives and that the risk of hepatocellular carcinoma increases toward the end of life.
2 ReferencesRobert W. Watson, M.D.
Allianz Life Insurance Company of North America, Minneapolis, MN 55403-21951
Ince N ,Wands JR . The increasing incidence of hepatocellular carcinoma. N Engl J Med 1999;340:798-799
Full Text | Web of Science | Medline2
Beasley RP ,Hwang LY ,Lin CC ,Chien CS . Hepatocellular carcinoma and hepatitis B virus: a prospective study of 22 707 men in Taiwan. Lancet 1981;2:1129-1133
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: There are three possible explanations for our findings of temporal increases in the incidence of hepatocellular carcinoma in the United States and the mortality and hospitalization rates associated with it: an age effect, a period effect, and a cohort effect.1
Kaczynski and Odén propose an age effect, which refers to the increase in disease rates that is related to aging populations. A period effect refers to a change in the rate of a disease when exposure affects a population over a specified period, as is the case with an epidemic. A cohort effect may be present when there are changes in exposure over generations. There is no model that incorporates all three effects.2 However, a birth-cohort analysis can be particularly useful in clarifying the underlying cause of time trends.2 A birth-cohort analysis entails plotting the birth dates of affected persons against mortality and incidence rates. In our study, the birth-cohort analysis showed progressively increasing rates of hepatocellular carcinoma in successive birth cohorts. The incidence rates increased throughout all age groups in the cohorts born between 1928 and 1958, indicating a risk exposure that has been retained throughout most of the lifetime of the people in these cohorts. If the overall increase in the incidence of hepatocellular carcinoma was the reflection of an aging population, one would see an increase in incidence that is either restricted to or out of proportion among persons in older age groups, irrespective of when they were born.
We also found shifts in the age-specific incidence rates toward younger ages. The relative increase in the incidence rate has mostly affected those 45 to 70 years of age, whereas the incidence among those 70 to 85 years old has decreased slightly.
It is legitimate, on the basis of the birth-cohort analysis, age standardization, and the trends in age-specific incidence rates, to deduce that a true increase in the incidence of hepatocellular carcinoma has occurred. This cannot be explained solely by the overall aging of the population.
2 ReferencesHashem B. El-Serag, M.D., M.P.H.
Andrew C. Mason, M.D.
Veterans Affairs Medical Center, Albuquerque, NM 87108Author/Editor Response
We thank Dr. Watson for his comments regarding the development of hepatocellular carcinoma in carriers of HBsAg when infection was acquired at birth. Because of the limited number of references that could be cited in our editorial, we elected to quote the initial paper published by Beasley et al.1 On the basis of these data, they subsequently estimated that the lifetime risk of death from hepatocellular carcinoma or cirrhosis in a male Chinese carrier was 40 to 50 percent,2 as we stated in our editorial. Additional documentation on the high mortality rate in patients with chronic HBV infection may be found elsewhere.3
3 ReferencesNedim Ince, M.D.
Jack R. Wands, M.D.
Rhode Island Hospital, Providence, RI 029031
Beasley RP ,Hwang LY ,Lin CC ,Chien CS . Hepatocellular carcinoma and hepatitis B virus: a prospective study of 22 707 men in Taiwan. Lancet 1981;2:1129-1133
CrossRef | Web of Science | Medline2
Beasley RP . Hepatitis B virus as the etiologic agent in hepatocellular carcinoma: epidemiologic considerations. Hepatology 1982;2:21S-26S
Web of Science3
Chen DS . From hepatitis to hepatoma: lessons from type B viral hepatitis. Science 1993;262:369-370
CrossRef | Web of Science | Medline
- Citing Articles (7)
Citing Articles
1
H.-H. Thein, S. R. Walter, H. F. Gidding, J. Amin, M. G. Law, J. George, G. J. Dore. (2011) Trends in incidence of hepatocellular carcinoma after diagnosis of hepatitis B or C infection: a population-based cohort study, 1992-2007. Journal of Viral Hepatitis 18:7, e232-e241
CrossRef2
Aekarach Ariyachaipanich, Tatek Takele, Paul McFarlane, Charles H. Berkelhammer, Allan B. Zelinger. (2009) The paradox of severe hypercholesterolemia and cachexia as a paraneoplastic manifestation of hepatocellular carcinoma. Journal of Clinical Lipidology 3:6, 398-400
CrossRef3
Fumio Imazeki, Osamu Yokosuka, Kenichi Fukai, Shigenobu Kawai, Tatsuo Kanda, Hiroshige Kojima, Hiromitsu Saisho. (2005) Lower incidence of hepatic failure than hepatocellular carcinoma in Japanese patients with chronic Hepatitis C.. Liver International 25:4, 772-778
CrossRef4
Samila Siavoshian, Jean Daniel Abraham, Christine Thumann, Marie Paule Kieny, Catherine Schuster. (2005) Hepatitis C virus core, NS3, NS5A, NS5B proteins induce apoptosis in mature dendritic cells. Journal of Medical Virology 75:3, 402-411
CrossRef5
P P Anthony. (2001) Hepatocellular carcinoma: an overview. Histopathology 39:2, 109-118
CrossRef6
Masao Honda, Shuichi Kaneko, Hiroshi Kawai, Yukihiro Shirota, Kenichi Kobayashi. (2001) Differential gene expression between chronic hepatitis B and C hepatic lesion. Gastroenterology 120:4, 955-966
CrossRef7
John T. Edwards, Graeme A. Macdonald. (2000) Hepatocellular carcinoma. Current Opinion in Gastroenterology 16:3, 275-281
CrossRef
