Hepatocellular Carcinoma in Sub-Saharan Africa

By Michael Charles Kew

The book comprehensively covers all aspects of hepatocellular carcinoma (cancer) as it occurs in sub-Saharan black Africans and how the cancer differs from that in other parts of the world, in most of which it occurs infrequently.

The first chapter is introduced with a discussion of the difficulty in recording all cases of this cancer in sub Saharan Africa and then covers its epidemiology, emphasising the very high incidence of liver cancer in sub Saharan black Africans. The relatively young age at which the tumor occurs in comparison with the older age of the patients in resource-rich regions is stressed. The sex distribution of the cancer is also discussed.

The second chapter describes the ways in which hepatocellular cancer presents in black Africans, its symptoms and signs, and how its presentation differs from that in other parts of the world.

The third chapter summarises the grave outlook for patients with the tumor, the very short survival times of the patients, and the reasons for their deaths.

The fourth chapter discusses the diagnosis of the cancer and why this is far more difficult than it is in resource-rich countries.

The fifth chapter deals with the pathology of the tumorits similarities and differences from that in other parts of the world.

The sixth chapter is the longest in the book, and it deals comprehensively with the causes of liver cancer in sub-Saharan Africa. The major cause is chronic infection with the hepatitis B virus, an infection acquired in early childhood, and the possible mechanisms by which this may cause the cancer. Other less common but still important causes in the subcontinentsuch as exposure to the fungal carcinogen, aflatoxin, chronic hepatitis C virus infection, dietary iron overload among Africans, and membranous obstruction of the inferior vena cavaare also discussed.

The seventh chapter discusses the difficulty in treating hepatocellular cancer and the poor results obtained in sub-Saharan Africa in comparison with the results of treatment in industrialized countries. The penultimate chapter presents possible ways to prevent the cancer or, at least, to detect it at a far earlier time than is currently the case and when it may be more amenable to treatment. The current status of vaccination in preventing hepatitis B virus infection, and hence liver cancer, is discussed in detail.

The final chapter asks the question, How can we improve the diagnosis and treatment of this dreadful malignant disease?

• Language: English
• Publisher: Trafford Publishing
• Release date: May 16, 2012
• ISBN: 9781466918887

Michael Charles Kew

I am a qualified physician (a fellow of the College of Physicians of South Africa and of the Royal College of Physicians of London) and scientist (doctorate of philosophy, doctorate of medicine, and doctorate of science, all from the University of the Witwatersrand). I have practised as a physician/scientist as a member of staff of the Department of Medicine of the University of the Witwatersrand in Johannesburg, South Africa, for forty-five years and am now emeritus professor of medicine and honorary research professor of the University of the Witwatersrand and honorary research professor of medicine in the Department of Medicine in the University of Cape Town (for past four years). My clinical specialty is in the field of disease of the liver, and the main thrust of my research is the very high incidence of liver cancer (hepatocellular carcinoma) in the black population of sub-Saharan Africa. During my career, I have seen and cared for countless patients with this devastating form of cancer and have been involved in ascertaining its causes and how it can be treated or prevented. During my career, I have been the author of 451 scientific papers and 69 chapters in medical books, and the great majority of these have been concerned with hepatocellular carcinoma. I have presented my work on this disease at many international conferences.

Book preview

Contents

DEDICATION

ACKNOWLEDGEMENTS

FOREWORD

CHAPTER 1    EPIDEMIOLOGY

CHAPTER 2    CLINICAL PRESENTATION

CHAPTER 3    NATURAL HISTORY, PROGNOSIS, AND CAUSES OF DEATH

CHAPTER 4    DIAGNOSIS

CHAPTER 5    PATHOLOGY

CHAPTER 6    AETIOLOGY AND PATHOGENESIS

CHAPTER 7    TREATMENT

CHAPTER 8    PREVENTION, SCREENING AND SURVEILLANCE

CHAPTER 9    WITHER HEPATOCELLULAR CARCINOMA IN SUB-SAHARAN AFRICA?

REFERENCES

About the Author

DEDICATION

This book is dedicated to the countless Black African sufferers from hepatocellular carcinoma, from whom I have learnt so much about this dreadful tumor.

May their suffering and poor prognosis not have been in vain.

ACKNOWLEDGEMENTS

Research funding for this work was provided by the South African Medical Research Council and the National Cancer Association of South Africa. Artwork for the book was provided by Michael Wyeth.

Thanks to my wife for her support and patience during the writing of the book and for always turning a blind eye on serum and cancer specimens stored in the domestic freezer.

FOREWORD

Liver cancer is the sixth most common cancer in the world, the vast majority being hepatocellular carcinomas. The International Agency for Research on Cancer (IARC) estimated around 750,000 new cases worldwide in 2008. Strikingly, there were also almost 700,000 deaths, making liver cancer the third most common cause of cancer death and illustrating the desperately poor prognosis associated with this malignancy.

This cancer is predominantly one of developing countries, where 85% of new cases occur: the highest incidence rates are seen in the regions of east and south-east Asia and sub-Saharan Africa. A further feature is the relatively early age of onset in developing countries, creating a particularly heavy social and economic toll on vulnerable families. This combination of circumstances makes the need for prevention of the disease all the more pressing.

Liver cancer is somewhat unusual in that the major causes of this cancer have been identified. Chronic infection with hepatitis B (HBV) and C (HCV) viruses are major risk factors in the high incidence regions, whilst chronic contamination of the diet by aflatoxins adds a further burden, particularly among chronic HBV carriers. In addition, in high income regions, the contribution of alcohol abuse is well-established. It is noteworthy in passing that there remain significant unexplored opportunities to understand more broadly the potential mechanisms of interaction of viruses and chemicals through study of the HBV/aflatoxin model and possible interactions with other risk factors, including dietary iron overload.

Much can be done in terms of prevention of liver cancer, but sadly the response of the international community historically has been slow, particularly in the regions where the disease is most common. However, there are hopeful signs. With the support of the Global Alliance on Vaccines and Immunization (GAVI), marked progress has been made in recent years in improving HBV vaccination coverage in the poorest populations of the world. Vaccine coverage needs to be expanded and sustained so that in 30 to 40 years a significant impact on liver cancer incidence is experienced. For HCV there is no vaccine, but initiatives based on avoidance of contaminated blood can make a major contribution, if implemented. For aflatoxins, the situation remains rather dire. Despite a wealth of knowledge on these potent dietary liver carcinogens and the identification of practical and feasible interventions, very little impact has been made to protect those people that are obliged to eat heavily contaminated staple foods on a daily basis. Surely this would be counted as a failure to translate clear scientific evidence into cancer prevention in the developing world.

There remain some 350 million chronic HBV carriers globally for whom the vaccine was made available too late. Similarly there are around 130-170 million chronic HCV carriers. What hope can be offered to these individuals? Reduction of exposure to co-factors, such as aflatoxins, is one step. Another is the exciting possibility that antiviral drugs, developed to treat human immunodeficiency virus (HIV) and whose price has therefore fallen, may offer an effective option for controlling hepatitis virus replication in chronic carriers. Treatment of hepatocellular carcinoma remains out of reach to those where the disease is most prevalent and the prognosis is poor. However, newer biomarkers for the early detection of the disease may offer hope for the future.

Liver cancer is one tumour for which there is an historic wealth of information on the occurrence, aetiology, pathology, prevention and treatment in low-income countries. In this volume, the collation of information specific to the people of sub-Saharan Africa provides an excellent platform for the future challenges of reducing the burden of this disease falling on some of the most economically poor people in the world. The book draws on many sources of information across several decades of investigations and is marked by its broad coverage of the topic, providing a rich source information. The hope must be that it provides a stimulus to the international community in tackling a largely preventable cancer in the developing world.

Christopher P. Wild,

Director, International Agency For Research on Cancer,

Lyon, France.

CHAPTER 1

EPIDEMIOLOGY

INTRODUCTION

Africa is the second largest of the world’s continents. It covers 30,300,000 square kilometers, approximately one-fifth of the global land area. The continent is sparsely inhabited, with a population of some 800 million people accounting for only 12% of the global population. Most of Africa’s population lives to the south of the Sahara desert, where the vast majority of the inhabitants are Blacks. North Africa, in climate, customs, and cultural background, is Mediterranean rather than African, and almost every distinctive feature of cancer in Africa belongs to sub-Saharan Africa [Parkin et al, 2003]. None more so than hepatocellular carcinoma.

The global cancer burden has been, and will continue to be, increasingly influenced by the growth and aging of the world’s population. In the year 2000, 10.1 million new cases of cancer were recorded worldwide, and this figure is expected to increase to 15 million by the year 2020 [Stewart and Kleihues, 2003]. Furthermore, by the year 2030 an annual total of 27 million new cases of cancer, 17 million deaths attributed to cancer, and 75 million people living with cancer could be expected [Boyle, 2006]. The impact of this increase will be felt in all geographical regions, but especially in the economically-constrained (low income; developing) regions, including sub-Saharan Africa.

To date, information on the occurrence of hepatocellular carcinoma in sub-Saharan Africa, in common with that on all cancers in the sub-continent, has been neither complete nor always accurate, as anyone who has ever tried to obtain accurate estimates of cancer incidences in the sub-continent will attest [Berman, 1951; Oettlé, 1964; Cook and Burkitt, 1971; Parkin et al, 2003]. As a result, the true incidence of hepatocellular carcinoma in sub-Saharan Africa has been significantly understated. Knowledge of cancer incidence and patterns in the African sub-continent has, until quite recently, been based largely on the work of a few pioneering clinicians and pathologists, who described series of cancer patients seen in their practices in terms of age, sex, cancer site, and histology [Berman, 1951; Oettlé, 1964; Cook and Burkitt, 1971]. Comparisons based on relative frequencies of different cancers in such case series may be misleading [Parkin et al, 2003]. Because biopsies and necropsies were, and to a lesser extent still are, seldom performed in Black African patients thought to be suffering from hepatocellular carcinoma, estimates of the occurrence of the tumor in the sub-continent, but particularly in rural areas where 80% of the Black population lives, under-estimate the true incidence of the tumor. Medical facilities in these rural areas are few and far between, and it has long been known that attendance at the local hospital decreases with its distance from the patient’s home [Davies et al, 1965]. Furthermore, in the past, hepatocellular carcinoma has been, and even now is, often diagnosed solely on the basis of compatible clinical features, supported in some patients by either a raised serum α-fetoprotein concentration or the finding of a mass-lesion in the liver on first-line ultrasonographic imaging of the liver [Kew, 2007]. The more sophisticated and expensive dynamic hepatic imaging techniques that today enable a firm diagnosis of hepatocellular carcinoma to be made on the basis of the early hypervascularity of the tumor followed by rapid wash-out of the indicator from the tumor are available in a few medical centers only in the large cities in sub-Saharan Africa, and certainly not in the rural areas.

Reports on hospital-based series of patients with hepatocellular carcinoma in the African sub-continent have been, and remain, biased by the interests and competence of the local clinicians, the diagnostic facilities and therapeutic options available, and the accuracy and completeness of record keeping [Oettlé, 1964; Parkin et al, 2003]. These short-comings are compounded by the nihilistic attitude of many doctors in sub-Saharan Africa to making a definitive diagnosis of hepatocellular carcinoma, which is engendered by the extremely poor results of treatment of this tumor and the dismal prognosis of the patients [Kew, 2007].

The appropriate statistic for comparing cancer risks between populations is the incidence rate [Parkin et al, 2003]. Incidence rates should be derived from population-based cancer registries, which aim to record information on all new cases of cancer that occur in a defined population [Parkin et al, 2003]. Accurate registration depends upon identifying every case of the cancer, which can only be achieved if patients have ready access to appropriate health services. When the necessary resources are inadequate, the number of patients with access to these services is limited and meaningful cancer registration is unlikely to be achieved [Oettlé, 1964; Parkin et al, 2003].

Another difficulty is to correctly identify the place of residence of cancer patients, so that when incidence rates are calculated, only those cases belonging to the population at risk are included [Flegg Mitchell, 1966; Parkin et al, 2003]. Unfortunately, place of residence is not an obvious concept in some sub-Saharan African regions. This is especially true in eastern and southern Africa, where Blacks living and working in urban areas often retain a foothold in their home village, to which they return intermittently on vacation. They may also return home to seek medical advice or care from the local traditional healers (witch-doctors; ‘nyangas’), or, when seriously ill or very old, to die [Flegg Mitchell, 1966]. Consequently, they may not be included in either an urban-based or a rural-based cancer registry. Moreover, many Black Africans living permanently in rural areas rely on the diagnostic and treatment services of traditional healers rather than conventional doctors [Flegg Mitchell, 1966] and they too would not be included in rural-based cancer registries. On the other hand, some Black men and women living in rural areas travel to towns or cities to stay with a relative when they are ill and need to visit a hospital. The address given to the hospital will be that of the urban relative. The few cancer registries in sub-Saharan Africa are almost all based in urban centers where diagnostic and treatment facilities are available. In compiling these cancer registries accurately the main technical problem is to avoid registration of these ‘temporary’ residents migrating into the urban area for short-term treatment or other reasons [Flegg Mitchell, 1966].

Compounding these difficulties, may be a low standard of literacy of the rural African population [Flegg Mitchell, 1966].

In order to be able to study time trends in the occurrence of a tumor it is important that the degree of completeness of registration of incident cancer cases should be constant throughout the period under consideration [Wabinga et al, 2000]. This has not always been the case in the few such studies performed in sub-Saharan Africa.

Without a comprehensive and accurate population survey, it is not possible to ascertain the true incidence of hepatocellular carcinoma in a rural Black African population, and one can only guess at how well the profile obtained from urban areas reflects that of the population as a whole [Friis and Storm, 1993].

A further difficulty in recording accurate information about the incidence of hepatocellular carcinoma in sub-Saharan Africa is the lack of appropriately trained personnel [Parkin et al, 2003]. There has been and continues to be a shortage, not only of epidemiologists and statisticians to plan and execute the collection of data in a processable form, but also of suitably trained personnel to man the registry infrastructure. Moreover, health ministries in most sub-Saharan African countries, when apportioning their limited financial and other resources, have regarded cancer as a lower public health priority than infectious and nutritional diseases [Parkin et al, 2003]. As a result, adequate governmental funds have not been, and continue not to be, made available for the institution and running of cancer registries.

With regard to mortality data, the information available in sub-Saharan Africa is even more questionable than that for incidence. Death registration and certification of cause of death are neither feasible nor a statutory requirement in virtually all countries in the sub-continent [Parkin et al, 2003].

As a result of these hurdles and shortcomings, the history of cancer registration in sub-Saharan Africa from the time that the earliest registries were instituted in the 1950s has been disappointing. Even today, a minority of countries have reliable cancer registries and, with few exceptions, these registries are based in urban settings. The first country-wide cancer registry with substantial rural coverage was established as recently as 1988 [Bah et al; 1990]. Moreover, these deficiencies have been compounded by the infrequency and unreliability of population censuses in the African subcontinent [Parkin et al, 2003].

Accurate knowledge of the epidemiological data concerning hepatocellular carcinoma will be required to play a role in guiding future treatment and ultimately strategies for the prevention of the tumor in sub-Saharan countries.

Despite the under-diagnosis and under-reporting of hepatocellular carcinoma in the African sub-continent, it is manifestly evident from the information available that the tumor occurs commonly and is a major cause of cancer deaths in Black African peoples living in sub-Saharan Africa.

INCIDENCE OF HEPATOCELLULAR CARCINOMA

Allowing for the limitations in ascertaining the true incidence of hepatocellular carcinoma in the indigenous populations of sub-Saharan Africa, the information available should be considered in perspective with the global incidence of the tumor.

Global incidence

The International Statistical Classification of Diseases and Related Health Problems Code (ICD) in current use does not distinguish between hepatocellular carcinoma and the other primary cancers of the liver. In regions with a high incidence of hepatocellular carcinoma, including sub-Saharan Africa, this tumor accounts for 90 to 95% of all primary malignant tumors of the liver, whereas in regions with a low or intermediate incidence it accounts for 70 to 85% of these tumors [Berman, 1951; Parkin et al,1997; Kew, 2003b; Ahmed et al, 2008]. The only exceptions to this pattern of occurrence of primary malignant tumors of the liver are in north-eastern Thailand and parts of the Philippines, where the frequent occurrence of cholangiocarcinoma as a result of endemic infestation of the liver with the flukes, Clonorchis sinensis or Opisthorchis viverini, is responsible for this tumor being the most common hepatic malignancy.

Based on the number of new cases of cancer in humans each year, hepatocellular carcinoma is the sixth most common cancer worldwide [Ferlay et al, 2010], the fifth in males and the seventh in females [Jemal et al, 2011]. During the recent time that the number of new cases of hepatocellular carcinoma worldwide each year has been recorded, the number has increased year by year. In the most recently reported year, 748,000 new cases of the tumor were recorded, constituting 9.2% of all new cancers [Ferlay et al, 2010]. Of these, 522,000 were in men and 226,000 in women, giving a sex ratio of 2.3:1.0 [Jemal et al, 2011]. The total annual death rate from hepatocellular carcinoma in that year was 695,900: 478,300 among men (the second most common cause of death from cancer) and 217,600 among women (the sixth most common cause of death from cancer) [Jemal et al, 2011]. Of the patients developing hepatocellular carcinoma during that year, 93% died from the effects of the tumor within 12 months of the onset of symptoms. This annual fatality ratio (0.93) is the highest of any human tumor.

Hepatocellular carcinoma does not have a uniform world-wide distribution. Rather, of all the new cases of cancer recorded during recent years, the vast malority have occurred in resource-constrained regions.

Incidence in resource-constrained regions

During the year in question, 626,700 new cases of hepatocellular carcinoma were recorded in resource-constrained regions: 440,700 in men (third in cancer incidence) and 186,000 in women (sixth in cancer incidence) (male:female ratio 2.4:1.0) [Jemal, 2011]. The number of deaths during the year was 580,600, and the 12-month fatality ratio 0.93: 402,900 deaths in men (the second most common cause of cancer deaths) and 177,700 in women (the fifth most common cause of cancer deaths) [Jemal et al, 2011].

As disturbing as these numbers undoubtedly are, they under-estimate the true incidence of, and the death rate from, hepatocellular carcinoma in resource – constrained regions. As discussed in the INTRODUCTION to this chapter, the reasons for this are that the tumor is both under-diagnosed and under-reported in these regions, particularly in the rural areas of these regions. Nevertheless, based on published incidences, it is evident that 84% of all global cases of the tumor occur in economically-constrained regions, with 55% occurring in China alone [Bosch et al, 2004; Ferlay et al, 2010]. Resource-rich regions have a low incidence of hepatocellular carcinoma. Exceptions to this pattern of incidence are Japan, a resource-rich country which has a high incidence of the cancer, and southern Europe, a resource rich region with an intermediate incidence [Parkin et al, 2005: Gomaa et al, 2008]. The incidence of hepatocellular carcinoma varies by as much as 100-fold between high-risk and low-risk regions, a ratio that is among the highest for any of the major cancers. Differences in the frequency of occurrence of the principal causes of hepatocellular carcinoma are responsible for the regional differences in incidence.

An age-standardized incidence of greater than 15 per 100,000 of the population per year is regarded as a high incidence of hepatocellular carcinoma. The age-standardized incidence of this tumor in China is 52.1 per 100,000 persons/year, in Melanesia 25.0 per 100,000/ persons/year, and in sub-Saharan Africa, 41.2/100,000/persons/ year in middle Africa, 29.7/100,000 persons/year in eastern Africa, and 20.9 /100,000 persons/year in western Africa [Parkin et al, 2003]. By comparison, the incidence of hepatocellular carcinoma is less than 5/100,000 persons/year in almost all resource-rich regions, and the tumor accounts for less than one % of all malignant diseases.

In countries with a high incidence of hepatocellular carcinoma, the tumor is not necessarily uniformly common throughout the country. For example, in China hepatocellular carcinoma is most common in the north-eastern province of Jilin, along the south-eastern seaboard, and in the Guanxi Autonomous Region [Chinese Academy of Medical Sciences, 1981], and in Mozambique the highest incidence is in the Inhambane province [Harington et al, 1985].

EPIDEMIOLOGY OF HEPATOCELLULAR CARCINOMA IN SUB-SAHARAN AFRICA

In common with the resource-constrained regions of south-eastern Asia and Melanesia, sub-Saharan Africa has a high incidence of hepatocellular carcinoma. Given the rates at which many of the Far Eastern countries have been developing during recent years, sub-Saharan Africa may, in the not too distant future, become the only resource-constrained region with a high incidence of this tumor.

Hepatocellular carcinoma has been reported to occur in sub-Saharan Africa since 1879 [Bergeret and Roulet, 1947], although the first report of the high incidence of the tumor in Black Africans living in the subcontinent was published only in 1921 [Pirie, 1921]. Since those early reports, a great deal has been learnt about the occurrence of hepatocellular carcinoma in sub-Saharan Africa, and it is now known that all Black ethnic groups throughout the sub-continent are affected by the tumor, although not to the same extent.

Published incidences of hepatocellular carcinoma in sub-Saharan Africa underestimate the true incidence of the tumor because of the many instances in which the tumor is either not definitively diagnosed or is not recorded in a cancer registry. In recent times, 46,000 new cases of hepatocellular carcinoma have been recorded as being diagnosed in sub-Saharan Africa each year [Parkin et al, 2003]. Because the recorded annual occurrence rate of the tumor in sub-Saharan Black Africans is almost the same as its annual fatality rate, the occurrence figure at any one time is virtually the same as the total incidence of the tumor in the population at that time. Age-standardized incidence rates and relative (proportional) frequencies of the tumor in specific countries in sub-Saharan Africa, documented in the International Agency for Research In Cancer (IARC) registry in 2003 [Parkin et al, 2003], are listed in Tables 1 and 2. Of the six West African countries for which incidences are recorded in this analysis, four have high incidences and two intermediate incidences of hepatocellular carcinoma: the one Central African country included has a high incidence; three of the five southern African countries* have high incidences, one has an intermediate incidence, and one a low incidence; and the two East African countries have intermediate incidences [Parkin et al, 2005]. Mozambique and many other countries in the sub-continent are not included in the tables because recent cancer registry figures are not available for those countries.

High incidences of the tumor had earlier been recorded in a number of sub-Saharan countries, with age-standardized incidence rates of between 19.2 and 28.4 per 100,000 persons/year and with the tumor accounting for about 20% of all malignant diseases [Berman, 1951; Steiner, 1961; Davies, 1961; Higginson, 1963; Cook and Burkitt, 1971]. In other reports, the frequency of hepatocellular carcinoma was expressed only in the latter way: the tumor accounted for five % of all malignant diseases in Uganda [Alpert et al, 1969], 11.2% of those in Ethiopia [Pavlica and Samuel, 1970], and 42% of those in Senegal [Payet et al, 1967]. Mozambique, the world’s poorest country in 1986, had the highest recorded incidence of hepatocellular carcinoma among sub-Saharan countries (101.7 per 100,000 persons/year in males and 31.4 per 100,000 persons/year in females) in an International Agency for Research in Cancer (IARC) registry covering the years 1956 to 1960 [Doll et al, 1966]. At that time hepatocellular carcinoma accounted for 65.5% of all malignant diseases in males and 31% of those in females, an incidence 58 times higher than that in United States Caucasians [Higginson, 1963; Prates and Torres, 1965].

As is true in south-eastern Asia, hepatocellular carcinoma is not uniformly common within sub-Saharan African countries, although the information in this regard is limited by the small number of reliable cancer registries. This phenomenon is most obvious in Mozambique, where hepatocellular carcinoma occurs appreciably more often in the eastern regions around Inhambane, Panda, Inharrime, and Morrumbene than in other regions [Prates and Torres, 1965]. Differences in frequency rates have also been reported between high-and low-lying geographical regions in Swaziland [Peers et al, 1976] and between tribes in Uganda [Alpert et al, 1969]. In South Africa hepatocellular carcinoma is more common in rural than in urban Blacks (ratio 4.4:1.0) [Kew et al, 1986c]. These differences, at both a sub-continental and a more local level, are in keeping with environmental factors playing an important role, but one which varies geographically, in the etiology of hepatocellular carcinoma in sub-Saharan Africa.

Changes in incidence of hepatocellular carcinoma in sub-Saharan Africa over time

Few studies have been reported on changes over time in the incidence of hepatocellular carcinoma in sub-Saharan African countries, and the results have been conflicting. Early reports from Uganda showed incidences to be unchanged between 1954-1960 and 1993-1997 [Muir et al, 1976; Wabinga et al, 2000]. By contrast, the occurrence of the tumor has recently been reported to be increasing in The Gambia [Ocama et al, 2009; Umoh et al, 2011]. In a study of changes over time in the incidence of hepatocellular carcinoma in a large cohort of Black Africans recruited from a number of southern African countries to work on a contract basis in the South African gold mines, McGlashan et al [2003] recorded a decrease in the incidence of hepatocellular carcinoma of 32% between 1964 and 1994. But there had been, over this time, changes in the relative frequency of recruits from the various countries of origin having different incidences of hepatocellular carcinoma, and this may have influenced the findings. In particular, the number of recruits from Mozambique, which early on supplied approximately 75% of the mine labor force, decreased significantly over this time [McGlashan et al, 2003].

In the few published studies in Asian countries with a high incidence of hepatocellular carcinoma, such as Hong Kong, Singapore, and Korea, the incidence of the tumor is static or decreasing [Yuen et al, 2009].

Changes in incidence of hepatocellular carcinoma over time in resource-rich regions

The incidence of hepatocellular carcinoma has been increasing, and is expected to continue to do so, in a number of resource-rich regions with a low or intermediate incidence of the tumor [Llovet, 2005]. In Japan, a resource-rich country with a high incidence of hepatocellular carcinoma, the incidence of the tumor increased by 15 to 20% during the three decades before 1990 [Okuda et al, 1997], and has continued to increase since that time. This rise was attributed to the increasing prevalence, during the preceding years, of chronic hepatitis C virus infection (and, to a far lesser extent, of alcohol abuse) in the population [Okuda et al, 1997]. The tumor is also steadily increasing in frequency in North America, with a two-fold rise in overall incidence between 1980 and 1998, as well as a 37% increase in tumor mortality between 1997 and 2001 [El-Serag, 2004], and an increase from 1.5 to 4.9 per 100,000 of the population over the past 30 years [Altekruse et al, 2009]. Here too, an increasing incidence of chronic hepatitis C virus infection was mainly responsible. Lesser increases in occurrence were reported in France, Italy, Spain, Britain, Israel, India, Australia, and Canada [Taylor-Robinson et al, 1997; Deuffic et al, 1998; Bosch et al, 1999]. In Italy, the rising incidence was attributed mainly to an increase in chronic hepatitis C virus infection [Stroffolini et al, 1998].

On the other hand, in some resource-rich countries with a low incidence of hepatocellular carcinoma, for example, Denmark, data are beginning to accumulate that the incidence of the tumor may be decreasing, or is projected to begin decreasing in the near future [Jepsen et al, 2007].

Effects of migration on the occurrence of hepatocellular carcinoma

Caucasians, and their offspring, migrating from a country with a low incidence of hepatocellular carcinoma to a high incidence resource-poor country, retain the low incidence of the tumor of their country of origin [Dunham and Bailar, 1968; Higginson, 1970; Waterhouse et al, 1977; Parkin et al, 2003]. These individuals almost always retain their original behavioral patterns in the new environment, and generally enjoy a higher standard of living than do the natives (Bosch, 1997]. One exception to this general rule are Indians, and their offspring, who settle in Hong Kong or Singapore. They, in time, acquire incidence rates similar to the native population and appreciably higher than those in their home countries [Bosch, 1997]. Little information is available on the incidence of hepatocellular carcinoma in the relatively small number of Indians now living in sub-Saharan Africa. But in a study of Indians resident in the Natal province of South Africa, hepatocellular carcinoma was shown to be rare [Akoojee et al, 1990].

Individuals, and their offspring, migrating from high incidence regions of hepatitis B virus-induced hepatocellular carcinoma to low incidence regions, initially retain the high incidence of their country of origin. The rate at which the incidence declines thereafter is determined by the dominant mode of acquisition of the chronic viral infection in the country