- Email: [email protected]
Mortality from hepatocellular and biliary cancers: changing epidemiological trends
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2002 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.
Vol. 97, No. 1, 2002 ISSN 0002-9270/02/$22.00 PII S0002-9270(01)03994-6
Mortality From Hepatocellular and Biliary Cancers: Changing Epidemiological Trends Satheesh Nair, M.D., K. S. Shivakumar, M.D., and Paul J. Thuluvath, M.D., F.R.C.P. Department of Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland; and Mayo Clinic, Rochester, Minnesota
OBJECTIVES: The incidence of hepatocellular carcinoma may be rising in the United States. The aim of this study was to determine the epidemiological trends in mortality from hepatocellular carcinoma (HCC) and biliary cancers (BCs) in Maryland during the last 3 decades. METHODS: The number of deaths due to HCC and BCs from 1970 to 1997 were obtained from the Maryland State Department of Health & Hygiene vital statistics database. Malignant neoplasms of the gallbladder and intrahepatic and extrahepatic bile ducts were grouped together as biliary cancers. To determine the trend in mortality, the total time period was divided into seven 4-yr periods. RESULTS: Mortality from HCC increased from 0.94 to 1.84 per 100,000 population (rate ratio ⫽ 1.94, CI ⫽ 1.87–2.03) and that from BCs increased from 1.28 to 1.7 per 100,000 population (rate ratio ⫽ 1.31, CI ⫽ 1.26 –1.36) over the study period. Although mortality due to HCC doubled in men (1.34 to 2.7 per 100,000) during this period, only a modest increase was observed among women (0.59 to 1.06 per 100,000). Because of a marked increase in the number of deaths among white Americans, the difference in HCCrelated mortality between white Americans and African Americans decreased considerably during this period. Mean age at death increased steadily for BCs from 67 to 73 yr, whereas there was no real trend for HCC. Among African Americans, the death from HCC remained stable, but there was a 2-fold increase in BC-related death. CONCLUSIONS: There was a marked increase in deaths from HCC over the past 3 decades in Maryland. This increase was more evident among men and white Americans. Deaths due to BCs increased modestly during the same period of observation. The marked rise in BC-related deaths among African Americans remains unexplained. (Am J Gastroenterol 2002;97:167–171. © 2002 by Am. Coll. of Gastroenterology)
INTRODUCTION Although hepatocellular carcinoma (HCC) is common worldwide, it has been relatively uncommon in the West. A recent study reported an increase in the incidence of HCC in the United States from 1.4 to 2.4 per 100,000 population
over the past 2 decades, with a shift toward younger adults and a 41% increase in HCC-related mortality (1). Rising trends in HCC incidence and mortality have also been reported from the United Kingdom, Japan, and France (2– 4). The increase in mortality from HCC has been largely attributed to cirrhosis caused by the chronic hepatitis C virus (HCV) (1). Chronic HCV has been recognized as an important risk factor for HCC in many well-defined epidemiological studies (5, 6). Moreover, it has been suggested that deaths from HCC due to HCV in the United States are likely to increase significantly by 2010 (7). Biliary cancers (BCs), including those arising from the gallbladder and intra- and extrahepatic bile ducts, are less common than and differ in epidemiology from HCC. One report noted a stable national trend in age-adjusted incidence of intrahepatic cholangiocarcinoma from 1973 to 1987, with a small decrease in the incidence of gallbladder cancer over the same period (8). However, there are only limited recent data on mortality trends for these cancers. The objective of our study was to analyze the trends in mortality rates for both HCC and BCs in Maryland from 1970 to 1997.
MATERIALS AND METHODS Source of Data The number of deaths due to HCC and BCs from 1970 to 1997 were obtained from the Maryland State Department of Health vital statistics database. In this database, data on mortality were compiled from all death certificates issued in Maryland. Diagnostic Codes The deaths were coded according to International Classification of Diseases, eighth revision (ICD-8) until 1978 and ICD-9 from 1979 onwards. The following ICD-9 codes were used: 155.0 (HCC), 155.1 (malignant neoplasm of the bile duct, intrahepatic), 156.1 (malignant neoplasm of the bile duct, extrahepatic), and 156.0 (malignant neoplasm of the gallbladder). The following demographic variables were recorded for all cancers: age, gender, and race or ethnic status. Statistical Analysis To analyze time trends in mortality, the period 1970 –1997 was divided into seven 4-yr segments. The number of deaths
168
Nair et al.
AJG – Vol. 97, No. 1, 2002
Table 1. Hepatocellular Carcinoma–Related Deaths in Different Time Periods
Period
No. of Deaths
Mean Age
Men (%)
White Americans (%)
African Americans (%)
1970–1973 1974–1977 1978–1981 1982–1985 1986–1989 1990–1993 1994–1997
151 146 174 242 242 322 371
60.4 62.3 63.2 64.3 62.5 64.1 63.1
68 73 69 72 68 67 72
68 60 63 64 62 60 62
30 40 37 30 31 31 28
Figure 1. Total number of deaths from HCC in Maryland, from 1970 –1973 to 1994 –1997.
was computed separately for each gender and ethnic group during the seven 4-yr segments. The mortality rates were calculated per 100,000 population, based on the population census in Maryland. The differences in death rates were compared using rate ratio and CIs. All analyses, except rate ratio and CIs, were performed using SPSS (SPSS, Chicago, IL).
RESULTS Hepatocellular Carcinoma The total number of deaths related to HCC in Maryland increased 2.45-fold, from 151 in the period 1970 –1973 to 371 in 1994 –1997 (Fig. 1). The increase in the number of deaths related to HCC was more obvious in the past 2 decades. The mortality rate from HCC doubled from 0.94 per 100,000 in the period 1970 –1973 to 1.84 per 100,000 in 1994 –1997 (rate ratio ⫽ 1.95 [CI ⫽ 1.87–2.03], p ⬍ 0.001) (Fig. 2). From the available data, we could not calculate the age-adjusted mortality rate. The mean age of the deceased remained more or less unchanged during the study period (Table 1). The mortality rate increased for both men and women; however, the increase was more obvious in men, in whom the mortality rate doubled from 1.34 in 1970 –1973 to 2.70 per 100,000 in 1994 –1997 (rate ratio ⫽ 2.07 [CI ⫽ 2.00 –2.14], p ⬍ 0.001) (Fig. 3). The mortality rate of women rose from 0.59 to 1.06 per 100,000 (rate ratio ⫽ 1.81 [CI ⫽ 1.72–1.90], p ⬍ 0.01). The mortality rate of men in
Figure 2. Mortality rate for HCC and BCs (per 100,000 persons) from 1970 –1973 to 1994 –1997.
1994 –1997 was almost twice that of women. The incidence of death from HCC in white Americans (rate ratio ⫽ 2.09 [CI ⫽ 2.00 –2.18], p ⬍ 0.001) increased to a much larger extent than in African Americans (rate ratio ⫽ 1.21 [CI ⫽ 1.17–1.25], p ⬍ 0.05) (Fig. 4). Although the mortality rate of white Americans more than doubled, from 0.71 in the period 1970 –1973 to 1.65 per 100,000 in 1994 –1997, only a modest increase was observed among African Americans over the same period (1.4 to 1.71 per 100,000). Although the death rate among African Americans due to HCC had historically been higher than white Americans, the racial difference in mortality had narrowed considerably over the past decade. Biliary Cancers There has been a steady increase in the total number of deaths from BCs over the past 3 decades. From the period 1970 –1973 to 1994 –1997, the total number of BC-related deaths increased from 207 to 341 (Fig. 5). The mortality rate was 1.28/100,000 in 1970 –1973 and 1.7/100,000 in 1994 – 1997 (rate ratio ⫽ 1.31 [CI ⫽ 1.26 –1.36], p ⬍ 0.05). The incidence of deaths from HCC exceeded the deaths due to BCs in Maryland toward the end of the observation period (Fig. 2). The mean age of those dying from BC increased steadily from 67 to 73 during the study period (p ⫽ 0.04) (Table 2). BC-related mortality increased in both men and women. Although the death rate in women increased modestly from 1.59/100,000 in 1970 –1973 to 1.88/100,000 in 1994 –1997 (rate ratio ⫽ 1.12 [CI ⫽ 1.09 –1.17]), the death rate in men increased from 0.89 to 1.47 per 100,000 (rate ratio ⫽ 1.66
Figure 3. Mortality rate for HCC (per 100,000 persons) in men and women from 1970 –1973 to 1994 –1997.
AJG – January, 2002
Mortality From Hepatocellular and Biliary Cancers
169
Table 2. Deaths Related to Biliary Cancers in Different Time Periods
Figure 4. Mortality rate for HCC (per 100,000 persons) in white Americans (WA) and African Americans (AA) from 1970 –1973 to 1994 –1997.
Period
No. of Deaths
Mean Age
Men (%)
White Americans (%)
African Americans (%)
1970–1973 1974–1977 1978–1981 1982–1985 1986–1989 1990–1993 1994–1997
207 212 254 261 274 313 341
67.4 66.4 69.2 69 71.4 71.4 73*
33 36 38 37 41 40 43
90 88 79 79 81 83 75
10 11 21 19 14 16 21
* p ⬍ 0.05 compared to 1970 –1973.
We found a 2.5-fold increase in mortality from HCC in Maryland over the past 3 decades. The increase was most pronounced among men and white Americans. During the study period, racial differences in HCC-related mortality narrowed considerably because of a significant increase in HCC-related deaths among white Americans. Although the death rate from HCC among African Americans was twice that of white Americans in the time period 1970 –1973, the rates were similar in the period 1994 –1997. A previous study had reported a progressive shift in the incidence of HCC in younger people between 1976 and 1995 (1). In our study, the mean age of patients remained more or less the same during the study period.
The prognosis of HCC is very poor, and there are no reliable treatment options. The median survival of patients with HCC increased only slightly from 0.57 yr during 1977– 1981 to 0.64 yr in 1992–1996 (9). It is therefore likely that the mortality rate reflects the incidence rate of the disease. Although it is inappropriate to attribute causality from interpretation of time trends in mortality, a significant upward trend observed in our study is most likely due to the epidemic of HCV. Other risk factors for HCC in the United States include hepatitis B virus (HBV), alcoholic cirrhosis, and hemochromatosis (10). In addition, cirrhosis of any cause is a risk factor for the development of HCC. Epidemiological data suggest that the mortality related to cirrhosis has steadily declined since the 1970s, after showing an increase between 1950 and 1970 (11). The incidence of HBV in the United States has also been declining, after reaching a peak of 11.5/100,000 in 1985 (12). Similarly, mortality from alcoholic cirrhosis has declined over the past 2 decades (11). However, it is still possible that alcohol may have played an additive role by causing disease progression and development of HCC in HCV- and HBV-infected persons (13–15). It is likely that the increase in HCC in the past 3 decades is related to HCV infection. There is both circumstantial and experimental evidence to suggest that HCV plays a major role in the pathogenesis of HCC (6, 14 –18). In patients with HCV infection, HCC typically develops after a latent period of 2–3 decades and usually in the background of underlying cirrhosis (17). It is therefore likely that most patients devel-
Figure 5. Total number of deaths from BCs in Maryland from 1970 –1973 to 1994 –1997.
Figure 6. Mortality rate for BCs (per 100,000 persons) in men and women from 1970 –1973 to 1994 –1997.
[CI ⫽ 1.60 –1.73], p ⬍ 0.05) (Fig. 6). Men accounted for 43% of those dying from BCs in 1994 –1997, a 10% increase from 1970 –1973. In African Americans the mortality rate increased from 0.64/100,000 in 1970 –1973 to 1.21/ 100,000 in 1994 –1997 (rate ratio ⫽ 1.90 [CI ⫽ 1.81–2.00], p ⬍ 0.01), and in white Americans the rate increased from 1.43 to 1.83 per 100,000 (1.28 [CI ⫽ 1.23–1.32], p ⬍ 0.05) during the same period (Fig. 7). Although the death rate due to BCs remained higher among white Americans, the racial difference in the mortality rate had decreased. The proportion of African Americans dying from BCs increased from 10% in 1970 –1973 to 21% in 1994 –1997 (Table 2).
DISCUSSION
170
Nair et al.
Figure 7. Mortality rate for biliary cancers (per 100,000 persons) in white Americans (WA) and African Americans (AA) from 1970 – 1973 to 1994 –1997.
oping HCC in the 1980s or 1990s were perhaps infected in the 1960s and 1970s. The annual incidence of HCV has declined from an estimated 230,000 in the 1980s to about 28 –36,000 currently, because of improvements in screening of blood and blood products and changes in the behavior of i.v. drug users (19). The mortality from HCC is likely to peak in 2020 –2030 before we see a significant decline. Currently, the majority of subjects with HCV remain undiagnosed, and of those who come to medical attention, a significant number do not receive appropriate antiviral treatment. Of those who receive treatment, only 30 – 40% have sustained clearance of virus. It is therefore unlikely that the treatment of HCV may make a major impact on the incidence and mortality of HCC in the near future. The relative increase in HCC in white Americans remains poorly explained. HCV infection is commonly seen in both white Americans and African Americans, and this may partly explain the increase in HCC-related mortality in white Americans. In our study, a significant rise in the incidence of death from BCs was noted in men and in African Americans. The reasons for the sharp increase in BC-related deaths in the period 1978 –1981 among African Americans are unclear. It is possible that advances in health care and improved certification of deaths in this group may partly explain the rise in BC-related mortality. Advances in diagnostic modalities may explain the overall increase in BC-related death, but they are unlikely to be an explanation for the disproportionate increase in mortality in men and African Americans. Etiological factors in the development of cancer of the bile ducts include cystic dilation of bile ducts, clonorchiasis, sclerosing cholangitis, and the radiocontrast agent thorotrast. Because there has been no significant variation in any of the known etiological factors for BCs in the United States over the study period, the reasons for the rise in mortality rates, especially among men and African Americans, remain unexplained. As BCs are rare in the United States, there are only limited data on the epidemiology of these cancers. Our report provides important information on time trends in mortality for these cancers. The mortality statistics must be interpreted with caution, in view of the possible improvements in death certification
AJG – Vol. 97, No. 1, 2002
procedures over time and advances in diagnostic techniques. However, the consistent upward trend in HCC-related mortality during the entire study period among all groups including African Americans, white Americans, men, and women argues against improvements in death certification alone accounting for the findings observed. Death certificate errors are common, but these errors are more common with nonspecific diagnoses and not with defined diagnoses such as liver cancer. Moreover, in general, cancers are more likely to be underreported (20). In one study, the quality of death certificate reporting was found to be constant in the study period 1961–1987 (20). Although we cannot extrapolate these findings to our study, on the basis of anecdotal clinical experience in our own institution, we believe that the increase we have shown is more likely to be real. Some of the rise could be explained by improved diagnostic modalities for detection of HCC in cirrhosis. However, sonography has been in use for over the past 2 decades, whereas mortality has continued to rise over the same period. In summary, there has been an increase in mortality from HCC in Maryland, most notably among men and white Americans. Further epidemiological studies are needed to better understand the role of HCV in the increasing incidence of mortality from HCC. Deaths due to BCs have also increased steadily, and the marked increase in mortality from BCs among African Americans remains unexplained.
ACKNOWLEDGMENT We sincerely thank Joseph Eustace, M.D., M.S.H., for statistical assistance. Reprint requests and correspondence: Paul J. Thuluvath, M.D., F.R.C.P., The Johns Hopkins Hospital, Room 1429, 1830 East Monument Street, Baltimore, MD 21205. Received Apr. 26, 2001; accepted Aug. 14, 2001.
REFERENCES 1. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999;340:745– 50. 2. Taylor-Robinson SD, Foster GR, Arora S, et al. Increase in primary liver cancer in the UK 1979-94. Lancet 1997;350: 1142–3 (letter). 3. Okuda K, Fujimoto I, Hanai A, Urano Y. Changing incidence of hepatocellular carcinoma in Japan. Cancer Res 1987;47: 4967–72. 4. Deuffic S, Poynard T, Buffat L, Valleron AJ. Trends in primary liver cancer. Lancet 1998;351:214 –5. 5. Haydon GH, Jarvis LM, Simmonds P, et al. Association between chronic hepatitis C infection and hepatocellular carcinoma in a Scottish population. Gut 1997;40:128 –32. 6. Blum HE. Does hepatitis C virus cause hepatocellular carcinoma? Hepatology 1994;19:251–5. 7. Wong JB, McQuillan GM, McHutchison JG, Poynard T. Estimating future hepatitis C morbidity, mortality and costs in the United States. Am J Public Health 2000;90:1562–9.
AJG – January, 2002
8. Carriaga MT, Henson DE. Liver, gall bladder extrahepatic bile ducts and pancreas. Cancer 1995;75:171–90. 9. El-Serag HB, Mason AC, Key C. Trends in survival of patients with hepatocellular carcinoma between 1977 and 1996 in the United States. Hepatology 2001;33:62–5. 10. Colombo M. Hepatocellular carcinoma. J Hepatol 1992;15: 225–36. 11. La Vecchia C, Levi F, Lucchini F, et al. Worldwide patterns and trends in mortality from liver cirrhosis, 1995 to 1960. Ann Epidemiol 1994;4:480 – 6. 12. McQuillan GM, Coleman PJ, Kruszon-Moran D, et al. Prevalence of hepatitis B virus infection in the United States: The National Health and Nutrition Examination Surveys, 1976 through 1994. Am J Public Health 1999;89:14 – 8. 13. Nalpas B, Feitelson M, Brechot C, Rubin E. Alcohol, hepatotropic viruses and hepatocellular carcinoma. Alcohol Clin Exp Res 1995;19:1089 –95. 14. Tanaka T, Yabusako T, Yamashita T, et al. Contribution of hepatitis C virus to the progression of alcoholic liver disease. Alcohol Clin Exp Res 2000;24:112S– 6S.
Mortality From Hepatocellular and Biliary Cancers
171
15. Donato F, Tagger A, Chiesa R, et al. Hepatitis B and C virus infection, alcohol drinking, and hepatocellular carcinoma: A case control study in Italy. Brescia HCC study. Hepatology 1997;26:579 – 84. 16. El-Serag HB, Mason AC. Risk factors for the rising rates of primary liver cancer in the United States. Arch Intern Med 2000;160:3227–30. 17. Castells L, Vargas V, Gonzalez A, et al. Long interval between HCV infection and development of hepatocellular carcinoma. Liver 1995;15:159 – 63. 18. Moriya K, Fujie H, Shintani Y, et al. The core protein of hepatitis C virus induces hepatocellular carcinoma in transgenic mice. Nat Med 1998;4:1065–7. 19. Center for Disease Control and Prevention. Recommendations for prevention and control of hepatitis C virus (HCV) infection and HCV-related chronic disease. MMWR Morb Mortal Wkly Rep 1998;47:1–39. 20. Hoel DG, Ron E, Carter R, Mabuchi K. Influence of death certificate errors on cancer mortality trends. J Natl Cancer Inst 1993;85:1063– 8.
Vol. 97, No. 1, 2002 ISSN 0002-9270/02/$22.00 PII S0002-9270(01)03994-6
Mortality From Hepatocellular and Biliary Cancers: Changing Epidemiological Trends Satheesh Nair, M.D., K. S. Shivakumar, M.D., and Paul J. Thuluvath, M.D., F.R.C.P. Department of Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland; and Mayo Clinic, Rochester, Minnesota
OBJECTIVES: The incidence of hepatocellular carcinoma may be rising in the United States. The aim of this study was to determine the epidemiological trends in mortality from hepatocellular carcinoma (HCC) and biliary cancers (BCs) in Maryland during the last 3 decades. METHODS: The number of deaths due to HCC and BCs from 1970 to 1997 were obtained from the Maryland State Department of Health & Hygiene vital statistics database. Malignant neoplasms of the gallbladder and intrahepatic and extrahepatic bile ducts were grouped together as biliary cancers. To determine the trend in mortality, the total time period was divided into seven 4-yr periods. RESULTS: Mortality from HCC increased from 0.94 to 1.84 per 100,000 population (rate ratio ⫽ 1.94, CI ⫽ 1.87–2.03) and that from BCs increased from 1.28 to 1.7 per 100,000 population (rate ratio ⫽ 1.31, CI ⫽ 1.26 –1.36) over the study period. Although mortality due to HCC doubled in men (1.34 to 2.7 per 100,000) during this period, only a modest increase was observed among women (0.59 to 1.06 per 100,000). Because of a marked increase in the number of deaths among white Americans, the difference in HCCrelated mortality between white Americans and African Americans decreased considerably during this period. Mean age at death increased steadily for BCs from 67 to 73 yr, whereas there was no real trend for HCC. Among African Americans, the death from HCC remained stable, but there was a 2-fold increase in BC-related death. CONCLUSIONS: There was a marked increase in deaths from HCC over the past 3 decades in Maryland. This increase was more evident among men and white Americans. Deaths due to BCs increased modestly during the same period of observation. The marked rise in BC-related deaths among African Americans remains unexplained. (Am J Gastroenterol 2002;97:167–171. © 2002 by Am. Coll. of Gastroenterology)
INTRODUCTION Although hepatocellular carcinoma (HCC) is common worldwide, it has been relatively uncommon in the West. A recent study reported an increase in the incidence of HCC in the United States from 1.4 to 2.4 per 100,000 population
over the past 2 decades, with a shift toward younger adults and a 41% increase in HCC-related mortality (1). Rising trends in HCC incidence and mortality have also been reported from the United Kingdom, Japan, and France (2– 4). The increase in mortality from HCC has been largely attributed to cirrhosis caused by the chronic hepatitis C virus (HCV) (1). Chronic HCV has been recognized as an important risk factor for HCC in many well-defined epidemiological studies (5, 6). Moreover, it has been suggested that deaths from HCC due to HCV in the United States are likely to increase significantly by 2010 (7). Biliary cancers (BCs), including those arising from the gallbladder and intra- and extrahepatic bile ducts, are less common than and differ in epidemiology from HCC. One report noted a stable national trend in age-adjusted incidence of intrahepatic cholangiocarcinoma from 1973 to 1987, with a small decrease in the incidence of gallbladder cancer over the same period (8). However, there are only limited recent data on mortality trends for these cancers. The objective of our study was to analyze the trends in mortality rates for both HCC and BCs in Maryland from 1970 to 1997.
MATERIALS AND METHODS Source of Data The number of deaths due to HCC and BCs from 1970 to 1997 were obtained from the Maryland State Department of Health vital statistics database. In this database, data on mortality were compiled from all death certificates issued in Maryland. Diagnostic Codes The deaths were coded according to International Classification of Diseases, eighth revision (ICD-8) until 1978 and ICD-9 from 1979 onwards. The following ICD-9 codes were used: 155.0 (HCC), 155.1 (malignant neoplasm of the bile duct, intrahepatic), 156.1 (malignant neoplasm of the bile duct, extrahepatic), and 156.0 (malignant neoplasm of the gallbladder). The following demographic variables were recorded for all cancers: age, gender, and race or ethnic status. Statistical Analysis To analyze time trends in mortality, the period 1970 –1997 was divided into seven 4-yr segments. The number of deaths
168
Nair et al.
AJG – Vol. 97, No. 1, 2002
Table 1. Hepatocellular Carcinoma–Related Deaths in Different Time Periods
Period
No. of Deaths
Mean Age
Men (%)
White Americans (%)
African Americans (%)
1970–1973 1974–1977 1978–1981 1982–1985 1986–1989 1990–1993 1994–1997
151 146 174 242 242 322 371
60.4 62.3 63.2 64.3 62.5 64.1 63.1
68 73 69 72 68 67 72
68 60 63 64 62 60 62
30 40 37 30 31 31 28
Figure 1. Total number of deaths from HCC in Maryland, from 1970 –1973 to 1994 –1997.
was computed separately for each gender and ethnic group during the seven 4-yr segments. The mortality rates were calculated per 100,000 population, based on the population census in Maryland. The differences in death rates were compared using rate ratio and CIs. All analyses, except rate ratio and CIs, were performed using SPSS (SPSS, Chicago, IL).
RESULTS Hepatocellular Carcinoma The total number of deaths related to HCC in Maryland increased 2.45-fold, from 151 in the period 1970 –1973 to 371 in 1994 –1997 (Fig. 1). The increase in the number of deaths related to HCC was more obvious in the past 2 decades. The mortality rate from HCC doubled from 0.94 per 100,000 in the period 1970 –1973 to 1.84 per 100,000 in 1994 –1997 (rate ratio ⫽ 1.95 [CI ⫽ 1.87–2.03], p ⬍ 0.001) (Fig. 2). From the available data, we could not calculate the age-adjusted mortality rate. The mean age of the deceased remained more or less unchanged during the study period (Table 1). The mortality rate increased for both men and women; however, the increase was more obvious in men, in whom the mortality rate doubled from 1.34 in 1970 –1973 to 2.70 per 100,000 in 1994 –1997 (rate ratio ⫽ 2.07 [CI ⫽ 2.00 –2.14], p ⬍ 0.001) (Fig. 3). The mortality rate of women rose from 0.59 to 1.06 per 100,000 (rate ratio ⫽ 1.81 [CI ⫽ 1.72–1.90], p ⬍ 0.01). The mortality rate of men in
Figure 2. Mortality rate for HCC and BCs (per 100,000 persons) from 1970 –1973 to 1994 –1997.
1994 –1997 was almost twice that of women. The incidence of death from HCC in white Americans (rate ratio ⫽ 2.09 [CI ⫽ 2.00 –2.18], p ⬍ 0.001) increased to a much larger extent than in African Americans (rate ratio ⫽ 1.21 [CI ⫽ 1.17–1.25], p ⬍ 0.05) (Fig. 4). Although the mortality rate of white Americans more than doubled, from 0.71 in the period 1970 –1973 to 1.65 per 100,000 in 1994 –1997, only a modest increase was observed among African Americans over the same period (1.4 to 1.71 per 100,000). Although the death rate among African Americans due to HCC had historically been higher than white Americans, the racial difference in mortality had narrowed considerably over the past decade. Biliary Cancers There has been a steady increase in the total number of deaths from BCs over the past 3 decades. From the period 1970 –1973 to 1994 –1997, the total number of BC-related deaths increased from 207 to 341 (Fig. 5). The mortality rate was 1.28/100,000 in 1970 –1973 and 1.7/100,000 in 1994 – 1997 (rate ratio ⫽ 1.31 [CI ⫽ 1.26 –1.36], p ⬍ 0.05). The incidence of deaths from HCC exceeded the deaths due to BCs in Maryland toward the end of the observation period (Fig. 2). The mean age of those dying from BC increased steadily from 67 to 73 during the study period (p ⫽ 0.04) (Table 2). BC-related mortality increased in both men and women. Although the death rate in women increased modestly from 1.59/100,000 in 1970 –1973 to 1.88/100,000 in 1994 –1997 (rate ratio ⫽ 1.12 [CI ⫽ 1.09 –1.17]), the death rate in men increased from 0.89 to 1.47 per 100,000 (rate ratio ⫽ 1.66
Figure 3. Mortality rate for HCC (per 100,000 persons) in men and women from 1970 –1973 to 1994 –1997.
AJG – January, 2002
Mortality From Hepatocellular and Biliary Cancers
169
Table 2. Deaths Related to Biliary Cancers in Different Time Periods
Figure 4. Mortality rate for HCC (per 100,000 persons) in white Americans (WA) and African Americans (AA) from 1970 –1973 to 1994 –1997.
Period
No. of Deaths
Mean Age
Men (%)
White Americans (%)
African Americans (%)
1970–1973 1974–1977 1978–1981 1982–1985 1986–1989 1990–1993 1994–1997
207 212 254 261 274 313 341
67.4 66.4 69.2 69 71.4 71.4 73*
33 36 38 37 41 40 43
90 88 79 79 81 83 75
10 11 21 19 14 16 21
* p ⬍ 0.05 compared to 1970 –1973.
We found a 2.5-fold increase in mortality from HCC in Maryland over the past 3 decades. The increase was most pronounced among men and white Americans. During the study period, racial differences in HCC-related mortality narrowed considerably because of a significant increase in HCC-related deaths among white Americans. Although the death rate from HCC among African Americans was twice that of white Americans in the time period 1970 –1973, the rates were similar in the period 1994 –1997. A previous study had reported a progressive shift in the incidence of HCC in younger people between 1976 and 1995 (1). In our study, the mean age of patients remained more or less the same during the study period.
The prognosis of HCC is very poor, and there are no reliable treatment options. The median survival of patients with HCC increased only slightly from 0.57 yr during 1977– 1981 to 0.64 yr in 1992–1996 (9). It is therefore likely that the mortality rate reflects the incidence rate of the disease. Although it is inappropriate to attribute causality from interpretation of time trends in mortality, a significant upward trend observed in our study is most likely due to the epidemic of HCV. Other risk factors for HCC in the United States include hepatitis B virus (HBV), alcoholic cirrhosis, and hemochromatosis (10). In addition, cirrhosis of any cause is a risk factor for the development of HCC. Epidemiological data suggest that the mortality related to cirrhosis has steadily declined since the 1970s, after showing an increase between 1950 and 1970 (11). The incidence of HBV in the United States has also been declining, after reaching a peak of 11.5/100,000 in 1985 (12). Similarly, mortality from alcoholic cirrhosis has declined over the past 2 decades (11). However, it is still possible that alcohol may have played an additive role by causing disease progression and development of HCC in HCV- and HBV-infected persons (13–15). It is likely that the increase in HCC in the past 3 decades is related to HCV infection. There is both circumstantial and experimental evidence to suggest that HCV plays a major role in the pathogenesis of HCC (6, 14 –18). In patients with HCV infection, HCC typically develops after a latent period of 2–3 decades and usually in the background of underlying cirrhosis (17). It is therefore likely that most patients devel-
Figure 5. Total number of deaths from BCs in Maryland from 1970 –1973 to 1994 –1997.
Figure 6. Mortality rate for BCs (per 100,000 persons) in men and women from 1970 –1973 to 1994 –1997.
[CI ⫽ 1.60 –1.73], p ⬍ 0.05) (Fig. 6). Men accounted for 43% of those dying from BCs in 1994 –1997, a 10% increase from 1970 –1973. In African Americans the mortality rate increased from 0.64/100,000 in 1970 –1973 to 1.21/ 100,000 in 1994 –1997 (rate ratio ⫽ 1.90 [CI ⫽ 1.81–2.00], p ⬍ 0.01), and in white Americans the rate increased from 1.43 to 1.83 per 100,000 (1.28 [CI ⫽ 1.23–1.32], p ⬍ 0.05) during the same period (Fig. 7). Although the death rate due to BCs remained higher among white Americans, the racial difference in the mortality rate had decreased. The proportion of African Americans dying from BCs increased from 10% in 1970 –1973 to 21% in 1994 –1997 (Table 2).
DISCUSSION
170
Nair et al.
Figure 7. Mortality rate for biliary cancers (per 100,000 persons) in white Americans (WA) and African Americans (AA) from 1970 – 1973 to 1994 –1997.
oping HCC in the 1980s or 1990s were perhaps infected in the 1960s and 1970s. The annual incidence of HCV has declined from an estimated 230,000 in the 1980s to about 28 –36,000 currently, because of improvements in screening of blood and blood products and changes in the behavior of i.v. drug users (19). The mortality from HCC is likely to peak in 2020 –2030 before we see a significant decline. Currently, the majority of subjects with HCV remain undiagnosed, and of those who come to medical attention, a significant number do not receive appropriate antiviral treatment. Of those who receive treatment, only 30 – 40% have sustained clearance of virus. It is therefore unlikely that the treatment of HCV may make a major impact on the incidence and mortality of HCC in the near future. The relative increase in HCC in white Americans remains poorly explained. HCV infection is commonly seen in both white Americans and African Americans, and this may partly explain the increase in HCC-related mortality in white Americans. In our study, a significant rise in the incidence of death from BCs was noted in men and in African Americans. The reasons for the sharp increase in BC-related deaths in the period 1978 –1981 among African Americans are unclear. It is possible that advances in health care and improved certification of deaths in this group may partly explain the rise in BC-related mortality. Advances in diagnostic modalities may explain the overall increase in BC-related death, but they are unlikely to be an explanation for the disproportionate increase in mortality in men and African Americans. Etiological factors in the development of cancer of the bile ducts include cystic dilation of bile ducts, clonorchiasis, sclerosing cholangitis, and the radiocontrast agent thorotrast. Because there has been no significant variation in any of the known etiological factors for BCs in the United States over the study period, the reasons for the rise in mortality rates, especially among men and African Americans, remain unexplained. As BCs are rare in the United States, there are only limited data on the epidemiology of these cancers. Our report provides important information on time trends in mortality for these cancers. The mortality statistics must be interpreted with caution, in view of the possible improvements in death certification
AJG – Vol. 97, No. 1, 2002
procedures over time and advances in diagnostic techniques. However, the consistent upward trend in HCC-related mortality during the entire study period among all groups including African Americans, white Americans, men, and women argues against improvements in death certification alone accounting for the findings observed. Death certificate errors are common, but these errors are more common with nonspecific diagnoses and not with defined diagnoses such as liver cancer. Moreover, in general, cancers are more likely to be underreported (20). In one study, the quality of death certificate reporting was found to be constant in the study period 1961–1987 (20). Although we cannot extrapolate these findings to our study, on the basis of anecdotal clinical experience in our own institution, we believe that the increase we have shown is more likely to be real. Some of the rise could be explained by improved diagnostic modalities for detection of HCC in cirrhosis. However, sonography has been in use for over the past 2 decades, whereas mortality has continued to rise over the same period. In summary, there has been an increase in mortality from HCC in Maryland, most notably among men and white Americans. Further epidemiological studies are needed to better understand the role of HCV in the increasing incidence of mortality from HCC. Deaths due to BCs have also increased steadily, and the marked increase in mortality from BCs among African Americans remains unexplained.
ACKNOWLEDGMENT We sincerely thank Joseph Eustace, M.D., M.S.H., for statistical assistance. Reprint requests and correspondence: Paul J. Thuluvath, M.D., F.R.C.P., The Johns Hopkins Hospital, Room 1429, 1830 East Monument Street, Baltimore, MD 21205. Received Apr. 26, 2001; accepted Aug. 14, 2001.
REFERENCES 1. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999;340:745– 50. 2. Taylor-Robinson SD, Foster GR, Arora S, et al. Increase in primary liver cancer in the UK 1979-94. Lancet 1997;350: 1142–3 (letter). 3. Okuda K, Fujimoto I, Hanai A, Urano Y. Changing incidence of hepatocellular carcinoma in Japan. Cancer Res 1987;47: 4967–72. 4. Deuffic S, Poynard T, Buffat L, Valleron AJ. Trends in primary liver cancer. Lancet 1998;351:214 –5. 5. Haydon GH, Jarvis LM, Simmonds P, et al. Association between chronic hepatitis C infection and hepatocellular carcinoma in a Scottish population. Gut 1997;40:128 –32. 6. Blum HE. Does hepatitis C virus cause hepatocellular carcinoma? Hepatology 1994;19:251–5. 7. Wong JB, McQuillan GM, McHutchison JG, Poynard T. Estimating future hepatitis C morbidity, mortality and costs in the United States. Am J Public Health 2000;90:1562–9.
AJG – January, 2002
8. Carriaga MT, Henson DE. Liver, gall bladder extrahepatic bile ducts and pancreas. Cancer 1995;75:171–90. 9. El-Serag HB, Mason AC, Key C. Trends in survival of patients with hepatocellular carcinoma between 1977 and 1996 in the United States. Hepatology 2001;33:62–5. 10. Colombo M. Hepatocellular carcinoma. J Hepatol 1992;15: 225–36. 11. La Vecchia C, Levi F, Lucchini F, et al. Worldwide patterns and trends in mortality from liver cirrhosis, 1995 to 1960. Ann Epidemiol 1994;4:480 – 6. 12. McQuillan GM, Coleman PJ, Kruszon-Moran D, et al. Prevalence of hepatitis B virus infection in the United States: The National Health and Nutrition Examination Surveys, 1976 through 1994. Am J Public Health 1999;89:14 – 8. 13. Nalpas B, Feitelson M, Brechot C, Rubin E. Alcohol, hepatotropic viruses and hepatocellular carcinoma. Alcohol Clin Exp Res 1995;19:1089 –95. 14. Tanaka T, Yabusako T, Yamashita T, et al. Contribution of hepatitis C virus to the progression of alcoholic liver disease. Alcohol Clin Exp Res 2000;24:112S– 6S.
Mortality From Hepatocellular and Biliary Cancers
171
15. Donato F, Tagger A, Chiesa R, et al. Hepatitis B and C virus infection, alcohol drinking, and hepatocellular carcinoma: A case control study in Italy. Brescia HCC study. Hepatology 1997;26:579 – 84. 16. El-Serag HB, Mason AC. Risk factors for the rising rates of primary liver cancer in the United States. Arch Intern Med 2000;160:3227–30. 17. Castells L, Vargas V, Gonzalez A, et al. Long interval between HCV infection and development of hepatocellular carcinoma. Liver 1995;15:159 – 63. 18. Moriya K, Fujie H, Shintani Y, et al. The core protein of hepatitis C virus induces hepatocellular carcinoma in transgenic mice. Nat Med 1998;4:1065–7. 19. Center for Disease Control and Prevention. Recommendations for prevention and control of hepatitis C virus (HCV) infection and HCV-related chronic disease. MMWR Morb Mortal Wkly Rep 1998;47:1–39. 20. Hoel DG, Ron E, Carter R, Mabuchi K. Influence of death certificate errors on cancer mortality trends. J Natl Cancer Inst 1993;85:1063– 8.