Edited by J, EstEve, A. Kricker, J. Ferlay, D,M, Parkin
in collaboration with the Steering Committee of the European Network of Cancer Registries
This publication was supported by
the "Europe against Cancer" programme of the Commission of the European Communities,
Directorate General V of Employment, Industrial Relations and Social Affairs
International Agency for Research on Cancer Lyon, France
The "Europe against cancer" programme, which has been ongoing since l 986, is the response of the European Community to the fight against cancer, Initiatives have been taken, in collaboration with the various partners in the Member States, in cancer prevention, training of health personnel, information, health education and medical research.
The development of adequate statistics is an important adjunct to our knowledge, and is a major decisional aid to the various actions. Cancer registration is an essential component of such an approach it has proved its use in estimating the size of the public health problem posed by cancer, examining the causes of cancer and evaluating the benefits of cancer treatment.
For these reasons, the "Europe against cancer" programme has lent support to the "European Network of Cancer Registries". Collaboration at the European level in this field will help to improve effective collection of data and ensure their comparability.
This network is expected to provide regular information on cancer incidence and mortality in the EC. Facts and Figures of Cancer in the EC is an example of what can be done to inform , the public and shows that relatively simple, scientifically sound information can be made available to the interested layman. By such means this publication will help to combat cancer.
Continuous registration of new cases of. cancer began in Europe in 1929 when the first population-based cancer registry was set up in Hamburg, Germany, but it was not until 1942 that the first registry to cover a whole nation was begun in Denmark. Today, there are population-based cancer registries which cover all or part of the national populations in all countries of the European Community except Greece and Luxembourg.
Cancer registries gather information about the size, nature, and distribution of cancer in the community. While this information has many uses, the major concern of population-based cancer registries is to express the facts about cancer in a standardised way through the calculation of cancer incidence rates for their population. These rates are used to study risk of individual cancers in the registry area compared to elsewhere, or to compare different sub- groups of the population within the registry area itself. Cancer incidence rates, in addition, are used in planning priorities for cancer control measures such as prevention and early detection programmes, in developing an understanding of the causes and impact of cancer, and in evaluating the effects of control measures, The information on case numbers can also be used in the planning of cancer treatment and care facilities.
From the perspective of a cancer registry, the most relevant example of' basic research is epidemiological research into the causes of cancer. In the search for the causes of cancer researchers make use of the information gathered by the cancer registry in a number of different types of studies. Descriptive studies, for example, assess the size of the cancer problem different of the number of new cases and the incidence rate per unit of population at which new cases occur.
Descriptive studies are of various kinds. Comparisons of cancer occurrence in various populations may provide clues to aetiology. International variation in incidence and mortality, for example, contributes to understanding the environmental causes of. cancers. Analysis of patterns of incidence by age, sex, geographic region within countries, ethnic origin and Occupation, can both identify high risk groups for particular cancers and provide clues to causation and, ultimately, the development of. preventive measures. Changes in incidence of cancer in migrant populations show the relative importance of environmental and genetic factors in cancer aetiology, and permit inferences as to the periods of life in which environmental agents have their effect.
The description of time trends in cancer incidence can demonstrate patterns of. variation which may point to changes in exposure to carcinogenic agents or the effects of control measures. For example, the increasing incidence of malignant melanoma suggests an eff-ect of increasing exposure to the sun (Kricker A, Armstrong BK, Jones ME, Burton RC. Health, Solar. UV Radiation and Environmental Change, lARC Tech. Rep. No. l 3, Lyon, International
Agency for Research on Cancer, 1993) and the decrease in stomach cancer in many populations has been linked with changing diets (Jensen OM. T rends in the incidence of stomach cancer in the five Nordic countries. In: Magnus K (ed) Trends in Cancer. Incidence, New York, Hemisphere, pp. 127-142, l982). The flattening or downward trend in lung cancer incidence in males in a number of countries (Parkin DM. Trends in lung cancer incidence world-wide. Chest, 1989: 96; 5S-8S) points to the effectiveness of education and other measures to reduce smoking.
Cancer registries are an important resource for specific studies of the causes of cancer. Linkage of cancer registry data to records of populations with varying levels of. exposure to possibly carcinogenic agents as recorded in Occupational records or surveys of diet and other behaviours can provide strong evidence as to the cause of cancer. Similarly, cancer registration data are commonly used in studies in which exposure to Possible carcinogens is compared between patients with cancer and members of the general population. Cancer registry records are similarly used to examine the effects of changes in exposure to carcinogens, the effects of cancer screening, and the effects of changes in cancer treatment.
Linkage of cancer registry records to registers of deaths permits the description of the pattern of survival following the treatment of cancer. This information is essential in advising patients and their families about the expected course of the disease and provides important information on the relative quality and effectiveness of cancer treatment services, Cancer registries are well suited to a role in evaluating the effectiveness of therapy and future progress since they already register the type of outcome to be measured, for example, the incidence of. advanced cancer, or death from cancer. Thus, the survival rate can be plotted over time to evaluate the progress made by different therapies, or compared between geographical areas or hospitals.
Finally, staff of cancer registries possess substantial expertise in disease registration, epidemiology, and cancer control. These skills are an important resource of' guidance and information for public health authorities and the general public.
A European Network of Cancer registries was launched in l 990 within the framework of. the EC 'Europe against Cancer' programme. The overall aim of the Network is "to improve comparability of cancer registry data and extend cancer registration within the European Community, with the aim of creating a solid basis for the continuous monitoring of the burden of cancer in the European Community and enhancing the use of cancer incidence information for research and planning". The Network was designed to provide regular information on the burden of cancer in European Community countries, on the patterns of cancer incidence and mortality by age, sex, and type of cancer, and on variations within and between EC member states.
The activities of the Network are co-ordinated through a steering committee, made up of members who are nominated by associations of cancer registries or directly elected by the membership. The International Agency for Research on Cancer provides a permanent secretariat.
The objectives of the project in the short term were to establish the administrative basis for the Network, to evaluate the data available, and to develop an incidence and mortality database on microcomputer for the EC countries. In the long-term the objectives include facilitating collaboration between cancer registries, the training of registry personnel, the regular publication of data on cancer in the EC, and the updating of the computerised incidence and mortality database.
Membership of the Network of European Cancer Registries is available to general population-based cancer registries in EC countries. The Steering Committee may recommend the admission of other cancer registries to either full or associate membership.
Applications for membership should be addressed to:
European Network of Cancer Registries International Agency for Research on Cancer Unit of Descriptive Epidemiology 150, Cours A1bert Thomas 69372 Lyon Cedex 08 France
Steering committee (July 1993):
Dr F. Berrino, Italy Dr R. Otter,the Netherlands Dr J. EstEve, IARC Dr D. Pheby, UK Dr F. Menegoz, France Dr H. Storm,Denmark Dr C. Navarro, Spain Dr H. Tulinius, Iceland Dr A. Vanvossel, CEC
The information on Facts and Figures of Cancer in Europe is mainly displayed graphically. Tables of the estimated numbers of cases and incidence and mortality rates are given in Appendix 2. For comparison purposes in the graphical presentation, incidence and mortality for the USA are given.
Population figures on l January l 990 were provided for each European Community country by the Statistical Office of the European Community, Luxembourg, and are given on page l. The population figures for East Germany were obtained from the 1990 Population Assessment (United Nations, World Population Prospects 1990, Population Studies 120, UN Department of International Economic and Social Affairs, New York, l 99 l ).
Mortality data are available for all EC countries but nation-wide cancer incidence rates are recorded only for Denmark, Scotland, and England and Wales. Although England and Wales have national coverage by cancer registries, the completeness and accuracy of the data vary substantially between regions of these countries. . lncidence data for 1983-87 were available in Canc.eI. lncidence in Five Continents Volume VI (Parkin DM, Muir CS, Whelan SL, Gao Y-T, Ferlay J, Powell J., IARC Sci. Pub. No. 120, Lyon, International Agency for Research on Cancer, 1992) for the whole of Denmark, and for parts of the Netherlands, France, Spain, Portugal, Italy, Germany, United Kingdom, and Ireland. Incidence for the USA in 1983-87 was available from the nine specialised US cancer registries (known as the SEER registries) which cover about 10% of the population. No reliable incidence data were available for Greece, Belgium, and Luxembourg.
Incidence rates from the Danish Cancer Registry were used for Denmark while for the UK, incidence was derived from the registry data for England, Wales and Scotland, and rates were estimated for Ireland according to the method described below. National incidence rates of cancer in 1990 in each of the other 10 EC countries have been estimated from mortality by use of the method of Parkin et al (Parkin DM, Pisani P, Ferlay J. Estimates of the world-wide incidence of eighteen major cancers in 1985. lnt J CanceI. 1993; 34: 594-606). Incidence and mortality data in 1983-87 were used to express the logarithm of incidence as a linear function of the logarithm of mortality with allowance for the effects of age, sex, and geographic region. Three geographic regions were used, northern (the UK, Ireland, Denmark), southern (Greece, Italy, Spain, Portugal), and western Europe (Germany, the Netherlands, Belgium, Luxembourg, France) in developing the cancer incidence estimation models. Incidence in 1990 was estimated by use of these models from mortality data in each country in the most recent period available: 1988-90 in France, West Germany (the former Federal Republic of Germany), Luxembourg, Netherlands, Portugal, and the UK, 1987-89 in East Germany (the former German Democratic Republic), Greece, Ireland, Italy and the USA, 1986-88 in Spain and 1984-86 in Belgium. For the USA, the incidence from the SEER registries was used and multiplied by the ratio of mortality in the whole USA to mortality in the SEER registry areas. For all sites of cancer (excluding non-melanoma skin cancer) the estimated number of new cases was based on the incidence-mortality ratio for all sites.
Since the prediction of incidence is based on mortality, and the quality of death certification is known to vary between EC countries (Jensen OM, Esteve J, Moller H, Renard H. Cancer in the European Community and its member states. Eur J Cancer 1990 , . 26 : 1167 - l 256), variations in the certification and coding of cause of death will influence the incidence estimates and in some instances may give rise to mortality rates that are apparently higher than incidence rates. Cancers of the liver and uterus provide examples of two additional specific problems. In some countries, secondary cancers of the liver are not well distinguished from primary cancer on death certificates and may give rise to spuriously high mortality rates, and therefore high estimated incidence rates, from primary cancer of the liver. For cancer of the body of the uterus, on the other hank, a high proportion of cancers is often found in the category "cancer of the uterus, not otherwise specified" (NOS) and the rates for the specific cancers - cancer of the cervix uteri and cancer of the corpus uteri (body of the uterus) are spuriously low. A correction has been make in the models for this latter problem so that for these two sites, the figures represent the best estimates of the true mortality.
Differences in the age structure of each population were accounted for by age- standardisation. This procedure uses a standard set of population figures to calculate a weighted average of age-specific rates within each population, thus giving each country a single summary rate that reflects the incidence and mortality that would have been expected if each country being compared had an identical age structure. The standard world population was used for age- standardisation .
The estimated number of new cancer cases in childhood are based on the data in International Incidence of Childhood Cancer (Parkin DM, Stiller CA, Draper GJ, Bieber CA, Terracini B, Young JL, IARC Sci. Pub. No. 87, Lyon, International Agency for Research on Cancer, 1988). Three geographic regions, as described above, were used to pool the data and calculate crude rates of childhood cancers within each country in 1990.
Enquiries about specific cancer incidence data can be addressed to individual cancer registries. A full list of addresses in EC countries for members of the European Network of Cancer Registries is given in Appendix 1. Mortality data are available from the Global Health Situation Assessment and Projections Unit, World Health ORGANIZATION, Geneva.
This section identifies the leading sites of cancer in 1990 at all ages, and in children from birth to 14 years of age.
These charts are based on the estimated numbers of new cancers and deaths in 1990. They allow the reader to compare the total number of new cases and the deaths in each sex in the 12 EC countries combined.
The age-standardised incidence and mortality rates by site are illustrated by bar-charts and arranged in order of increasing incidence or mortality in each country.
Bar charts are provided for each major site of cancer (Mouth and pharynx, Oesophagus, Stomach, Colon and rectum, Liver, Pancreas, Larynx, Lung, Melanoma, Bladder, Kidney, Non- Hodgkin Lymphoma, Hodgkin's disease, Leukaemia, Breast, Cervix, Body of the uterus, Ovary, Prostate, Testis, and all sites excluding non-melanocytic skin cancer). For each individual site or grouping of cancer, comments are given on the main risk factors, trends in incidence and mortality in the period 1965-1985, and survival.
Each chart contains two bars for each country the top bar illustrates the incidence rate and the lower one is the mortality rate.
The bar charts of age-standardised mortality and incidence of cancer at each major site have been produced for each sex and each EC country, thus allowing valid comparisons between countries. Corresponding results are also shown for all EC countries combined and, as an international comparison, the USA.
Comments are provided on the main risk factors for each major cancer site. A major reference source for these comments has been CanceI.. . Causes, Occurrence and Control (Tomatis L, Aitio A, Day NE, Heseltine E, Kaldor J, Miller AB, Parkin DM, Riboli E eds., IARC Sci. Pub. No. 100, Lyon, International Agency for Research on Cancer, 1990).
The IARC publication Trends in CanceI. lncidence and Mortality (Coleman M, EstEve J, Damiecki P, Arslan A, Renark H, IARC Sci. Pub. No. 121, Lyon, International Agency for Research on Cancer, 1993) was used as the basis for comments on time trends in each country in the period 1965- 1985, although for some cancers data were available up to 1987.
Population-based survival statistics are based on estimates from the EUROCARE project (Berrino F, Sant M, Verdecchia A, Capocaccia R, Hakulinen T, EstEve J, CanceI. Survival in Europe. IARC Sci. Pub. No. 132, Lyon, International Agency for Research on Cancer, in preparation). The five-year relative survival proportions that are given indicate the percentage of patients who survive five years.
Summary tables of incidence and mortality by sex and major site for each country and the EC as a whole are given in Appendix 2.
CANCER AT ALL AGES
|Numbers of new cancer cases in the European Community in 1990 - MALES|
|Numbers of new cancer cases in the European Community in 1990 - FEMALES|
|Numbers of cancer deaths in the European Community in 1990 - MALES|
|Numbers of cancer deaths in the European Community in 1990 - FEMALES|
CANCER IN CHILDREN
|Numbers of new cancer cases in the European Community in 1990 - BOYS|
|Numbers of new cancer cases in the European Community in 1990 - GIRLS|
|Numbers of cancer deaths in the European Community in 1990 - BOYS|
|Numbers of cancer deaths in the European Community in 1990 - GIRLS|
This group excludes non-melanocytic skin cancers because of the large number of these cancers and the difficulty in accurately estimating their numbers. The highest cancer incidence rates in males were in Luxembourg, Belgium and France and, in females, in Denmark and Ireland, while the lowest incidence rates for both sexes were in Spain, Greece and Portugal. The annual number of cancer deaths in the EC is about 837 000 and the estimated number of new cancer cases is 1 300 000. Both incidence and mortality rates were higher in males than females in all countries with striding sex differences in some countries. For example, males in France experienced a rate of cancer that was 50% higher than in females; by comparison, in Denmark the excess in males was only about 7%. The higher incidence and mortality rates experienced by males is largely due to their greater use of tobacco and alcohol, although other factors such as occupational exposures to carcinogens and biological differences between the sexes also contribute. Cancer incidence in males in the USA is about 10% higher than in males in EC countries with the highest rates and the difference is slightly more than this in females. In contrast cancer mortality in US males is low, about the same as in Spain, and slightly higher than in Greece and Portugal. Cancer mortality in US females is approximately midway between the highest and the lowest EC rates.
France and Luxembourg have the highest incidence rates of cancer of. the mouth and pharynx in men. Incidence rates were much lower in women and did not vary greatly between countries. France is one of two countries of' the world (the other is India) with a very high incidence of cancers of the mouth and pharynx in men. Both cancers are caused mainly why the consumption of alcohol and the smoking or chewing of tobacco. Poor oral hygiene may contribute to cancers of the mouth. Incidence of cancer of the mouth and pharynx in men in the EC is similar to the USA, although mortality rates are rather higher. ln women, incidence in the EC is lower than the USA, while mortality rates are much the same.
Trends: lncidence and mortality rates have generally increased in men except in France, which has also seen a decrease in alcohol consumption in the last 20 years. The increase in cancer rates has been especially steep in Germany. In women the risk of cancer of. the mouth has decreased in the UK in contrast to an increase in continental Europe except Greece and Portugal.
Survival: Survival differs according to anatomic sub-site. The five-year relative survival rates are better in women than men, and overall range from 35% to 45%.
Incidence and mortality rates of cancer of the oesophagus are nearly identical reflecting the high fatality of oesophageal cancer. The rates in men in France were about double or more those in other EC countries except the United Kingdom and Ireland. In contrast rates in women were low in France. In the United Kingdom and Ireland, women had rates that were four-folk greater than the low rates in Spain and Italy. Greece had very much the lowest rates in both sexes. In developed countries, cancer of the oesophagus is caused mainly by drinking alcohol and smoking tobacco. A poor diet, particularly a low intake of meat, dairy products and fruit and
Vegetables, may also make a contribution to causation. Incikence and mortality of this cancer in most EC countries is similar to that in the USA except for France, the United Kingdom and Ireland which have higher rates, and Greece, where rates are lower.
Trends: Cancer of the oesophagus in men has, in general, increased in the EC, although rates have been stable in Greece and the increase has been less steep in Italy ank Portugal. This cancer has decreased in France. In women, while there have been increases in Denmark, Germany and the Netherlanks, little change has been seen elsewhere.
Survival: Survival is very poor with five-year relative survival rates which are uniforrnly lower than 10%.
The rates of stomach cancer in males are about double those in females. There is quite a wide range of risk within the EC with rates in the southern countries two-folk or more those in the north. The geographic pattern is similar in the two sexes, however, in contrast to cancer of the oesophagus. Mortality, while lower than incidence, is not suhstantially so, reflecting the rather poor survival for this cancer. Stomach cancer has decreased in incikence in most Western countries over the past 50 years although the causes remain essentially unknown. Dietary factors are regarded as probably most important in causing the disease. The decreasing kependence on starchy, smoked, salted and fried foods, and an increasing consumption of green leafy vegetables and citrus fruits may be responsible for the kecline of stomach cancer in Western populations. Rates are considerably higher than in the USA; for example, incikence in southern Europe is hetween 2-4-fold higher, and mortality between 3-5-fold, than the US rates.
Trends: Stomach cancer in both sexes has been decreasing at more than 5% annually for the last 25 years. The kecline started later in Portugal and Spain, and has been less steep in Portugal, Greece and Italy than in other countries.
Survival: The poor survival for persons diagnosed with cancer of the stomach is reflected in the five-year relative survival rates which are between 10% and 20%.
Cancers of the colon and rectum taken together are, after lung cancer., the most commonly occurring cancers in the EC. lncidence and mortality are high in the northern and western parts of the EC. There is little apparent variation in rates. While there is a certain geographical similarity in rates for males and females, it is not as striking as for stomach cancer. The lowest rates are in the southern EC countries, that is, Italy Spain, Portugal, and Greece. These cancers are associated with the modern , western diet which is high in fat and protein, factors that may increase the risk, and low in vegetahles and fruit and fibre which are thought to be protective. Certain occupatlonal exposures, some aspects of reproductive history, reduced physical activity and a history of cholecystectomy, although their likely impact is thought to be quite small, have all been suggested as potentially causing some of' these cancers.
Incidence rates are lower than in the USA, but mortality rates in all except the southern countries are higher.
Trends: Incidence rates have been stable or increasing in all EC countries, apart from a slight decrease in younger people in Denmark and the UK. In contrast, mortality rates have decreased, most steeply in women.
Survival: Survival has improved in recent years and the five-year felative survival fates vafy from 30% to 45%,
Incidence and mortality of' liver cancer is highest in the three southernmost EC countries. The variation in rates is considerable: Incidence in the south is six times greater than in the north in males and five times in females. The apparently higher rate of mortality than incidence is the result of recording deaths with metastatic disease as due to liver cancer. Liver cancer is uncommon in western countries compared with the very high fates in southern Africa and south- east Asia. A strong link has been established with infection with both hepatitis B and C viruses. The occurrence of' these viruses in Europe is much lower than in African and Asian populations. It has been suggested that the high rates of' liver cancer in Greece may be associated with chronic hepatitis B infection. Consumption of aflatoxins and alcohol, especially chronic alcohol abuse, are other known risk factors, and in addition, associations with cirrhosis from causes other than alcohol, and tobacco smoking are suspected. The rates in most of the EC countries are higher than in the USA, while only the UK and the Netherlands have rather lower incidence and mortality,
Trends: The evaluation of' time trends in liver cancer is not possible given the poor quality of mortality data, in particular, over the last 25 years, as mentioned above. That is, diagnostic difficulties have led to the recording of metastatic tumours as primary liver cancers.
Survival: Survival from cancer of the liver is extremely poor and the five year relative survival rates are about 5% or less.
Incidence and mortality of this cancer are high in the north, and rather lower in Greece, Portugal, and Spain. Italy is the exception with slightly higher rates than the other southern countries. The geographic pattern is similar in males and females. Rates are about 20% to 45% lower in females than males. Cancer of the pancreas is almost uniformly fatal. Smoking increases the risk of this cancer, and kietary factors may also play a role. In EC countries the incidence of pancreas cancer is generally much lower than that of stomach cancer and mortality is lower too. The high-risk EC countries have incidence and mortality rates similar to those observed in the USA.
Trends: Incidence and mortality have been increasing in recent times in southern countries to a level of risk closer to that seen in the north. In consequence, geographic variation in risk of this cancer has been diminishing.
Survival: Persons with cancer of the pancreas have very poor survival with five-year relative survival rates below 5%.
Incidence rates of laryngeal cancer in men are approximately 30% to 60% higher than mortality. Overall, the highest rates of this cancer are in males in southern EC countries. In contrast to the pattern in men, nearly all rates in women are negligible, with very high sex ratios in the countries of southern Europe. The major risk factors for laryngeal cancer are alcohol and tobacco consumption. In men, incidence and mortality rates in most EC countries are higher than in the USA. In women, incidence is lower than in the USA, but mortality is not.
Trends: Trends in men have been variable, with decreases in France and Italy, increases in Denmark and Germany but little change in the other EC countries. Trends have not been examined in women because there are very few cases.
Survival: The five-year relative survival rates for this cancer differ by anatomic sub site but, overall, are approximately 60%.
Considering both males and females together, cancer of the lung is the commonest cancer in the EC countries. The highest fates in females are in the UK, Denmark, and Ireland, and in males in Belgium and the Netherlands. Among men in EC countries, 29% of all cancer deaths are caused by lung cancer, and 22% of all new cancer cases. The corresponding figures in women are 9% and 6% respectively . Tobacco smoking, particularly cigarette smoking, is by far the most important cause of cancer of the lung, and explains the dramatic increase in its frequency over the past 50 years. Some 83% of the lung cancer cases in men and 34% in women are estimated to be due to tobacco smoking. Certain occupational exposures including asbestos, coal-tars, and arsenic, have also been associated with this cancer. Incidence in males in the USA is at the level of the highest rates in the EC but the mortality rates are lower than in these countries. Both incidence and mortality in US females are higher than in EC countries although mortality in Denmark and the UK approaches the US rate.
Trends: In men, incidence rates appear to have stopped increasing in a number of high risk EC countries such as Belgium and the Netherlands and in the UK rates have been decreasing in successive generations for some time. In contrast, the incidence is increasing steeply in men and women in France, Spain, and Portugal. Mortality rates in men are now falling in the UK, and are stable in several countries although still increasing in the south. ln women, rapid increases in incidence and mortality rates are occurring in the south. There is evidence that a decrease in incidence has started in women in the UK.
Survival: Outcome after diagnosis of lung cancer is poor with five-year relative survival rates of less than l0%.
Incidence of melanoma varies fourfold in males in the EC countries, and five-fold in females, with the lowest incidence in countries closer to the equator such as Spain, Italy and Portugal, and a high incidence in more northern countries such as Denmark. While mortality tends to be much lower than incidence, a similar strong north-south gradient is seen in both incidence and mortality and in both sexes, Unlike other cancers, there is a female excess in incidence in all countries. While there is evidence that sun exposure increases the risk of malignant melanoma, the relationship is obviously not straightforward. The inconsistency is thought to be due to differing patterns of sun exposure, particularly recreational and vacation sun exposure, and a gradient of risk for melanoma which declines from the north to the south of Europe and matches the increasing protective natural pigmentation of the skin. Incidence rates in males in Europe are lower than in the USA, while the differential in mortality is less marked. In women, incidence in most EC countries (the exception is Denmark and Luxembourg) is lower than in the US, while mortality rates are rather similar.
Trends: Incidence rates of melanoma have increased steeply by about 7% annually in most EC countries. Mortality has increased in all EC countries, most steeply in France, Italy, and Spain.
Survival: ln persons diagnosed with melanoma, the five-year relative survival rates are greater than 80%.
Breast cancer is the most common cancer of females in the EC and represents 19% of cancer deaths and 24% of all cancer cases. Incidence and mortality are higher in northern EC countries than in the south, and the geographical variation is somewhat less than two-fold. The following risk factors are associated with an increased risk of breast cancer: having a family history of the disease, increased body size, a past history of benign breast disease, reproductive factors including late age at first birth, nulliparity, early menarche and late age at menopause, the use of oral contraceptives at young ages and unopposed oestrogens for the relief of menopausal symptoms, and exposure to ionizing radiation. Diet, especially one rich in fat and protein of animal origin, has been suggested to have a possible causal role, and several studies have suggested a positive association between increased alcohol consumption and this cancer. Diet during childhood may be especially important. Incidence rates in the EC are lower than in the USA but US mortality rates are similar to the EC average, and only Spain, Portugal and Greece have appreciably lower mortality than the USA.
Trends: Incidence rates have increased in all EC countries, while mortality has started to decrease recently in younger women in Denmark, the Netherlands, and the UK.
Survival: Five-year relative survival rates vary between 60% and 70%.
Cervical cancer is the only cancer of women which does not show a clear north-south geographical gradient of risk. Incidence and mortality vary nearly two-fold throughout the EC countries, with the highest incidence in Denmark and high mortality also in the UK, Portugal and Germany. The southern countries, Luxembourg and the Netherlands have the lowest rates in the EC, The main cause of cancer of the cervix is infection of the cervix with human papilloma virus (HPV) particularly types 16 and 18, This infection is acquired most commonly by sexual intercourse, thus explaining the association which has been observed with number of sexual partners and age at first intercourse. In the presence of HPV infection, oral contraceptives may further increase the risk of cancer of the cervix. Incidence in the EC countries is rather lower than in the USA (although rates in the UK and Denmark are higher). In contrast, mortality in the USA is lower than in almost all EC countries.
Trends: Incidence has generally decreased except in younger women in the UK, who have experienced a relatively large increase. The overall decrease is partly attributable to the success of screening programmes. Trends in mortality are difficult to interpret because of the problem that many death certificates have insufficient detail other than "cancer of the uterus not otherwise specified". Nonetheless, these trends are consistent with the incidence trends.
Survival: Five-year relative survival rates are between 50% and 60%.
Cancers of the body of the uterus are mainly endometrial cancers. The low rate of incidence and mortality in the southern most EC countries resembles the pattern of breast cancer rates. Mortality, although low, has a nearly four-fold geographical variation which is the other cancers of the female reproductive system. Endometrial cancers came to prominence in the 1970s in the USA with an increasing incidence due to the use of oestrogens for the relief of menopausal symptoms. Other characteristics of women with endrometrial carcinoma are obesity, low parity, and a late age of menopause, Use of combination oral contraceptives is associated with a consistent reduction in the risk of developing endometrial cancer. There is some suggestion that dietary factors such as high fat intake may be responsible for a proportion of the cases. Incidence rates in Europe are considerably lower than in the USA, while mortality rates are higher (except in Greece).
Trends: Incidence rates have decreased in younger women. A decrease in Mortality rates has also been seen in several of the EC countries.
Survival: The five-year relative survival rates are between 65% and 75%.
Rates of ovarian cancer are higher in northern Europe than in the south, and resemble the pattern of breast cancer, with a high incidence and mortality in the UK and Ireland. There is quite a large variation in the risk between countries with, for example, incidence in Denmark twice that in Greece, and mortality in Denmark some three times higher. While the numbers of cases (and rates) of ovarian, cervical and endometrial cancer in the EC are rather similar, the mortality from ovarian cancer is substantially greater (almost as many deaths from ovarian cancer as from the two others combined). Like breast and endometrial cancer, ovarian cancer tends to be associated with nulliparity, and use of oral contraceptives appears to lead to almost a halving of risk in women over 40 years of' age. An association with dietary fat levels has also been suggested. lncidence and mortality in the EC is rather similar to that in the USA.
Trends: Incidence has decreased in the younger women of the EC countries apart from Spain and Greece, and mortality rates have declined also, especially in Denmark and the Netherlands.
Survival: Survival is poor with five year relative survival rates of' about 25% to 30%.
Prostate cancer accounts for an estimated l 2% of' all new male cancer cases each year in the EC and 10% of all cancer deaths. It is the second most frequent cancer of' men. lncidence and mortality are higher in the northern countries than in the south but within these broad geographic divisions the variation is slight. The aetiology of prostate cancer remains obscure. Sexual f-actors have been proposed as important. Dietary factors may also be associated with increased risk, especially high fat and protein consumption. lncidence in the EC is much lower (about one half) than in the USA, although there is very little difference in the rates of' mortality, This may be partly due in the USA to careful search for signs of cancer in tissue removed at operations for prostatic enlargement, so that more asymptomatic "cancers" are detected.
Trends: Rates of this cancer have been increasing in all EC countries although these rates may reflect, to an unknown extent, the introduction of diagnostic practices which have increased the availability of tissue specimens for examination.
Survival: The five-year relative survival rates are slightly greater than 50%.
Testicular cancers are relatively uncommon. lncidence of' this cancer is highest in Denmark and other northern countries, and low in the south. The variation is extraordinary, for example, Danish incidence is five-fold higher than that of Spain and Portugal. The Luxembourg rates are based upon very few cases or deaths, and are hence unreliable. Testicular cancer is primarily a disease of' young adults. Undescended testicles are a predisposing condition, although little is known of the other causes of this cancer. Within the EC, the US incidence rates are exceeded only by Denmark and Germany; US mortality, however, is low.
Trends: lncidence of testicular cancer has increased in all EC countries except Spain, and most steeply in Germany. ln contrast, mortality rates have been decreasing annually by about S C/c for the last 20 years.
Survival: Survival is good with five-year relative survival rates greater than 80%.
Bladder cancer is between three and five times more common in men than in women. The variation in incidence and mortality between countries is about l.5-2 fold, with the highest rates observed in Denmark. The geographic pattern is not as consistent for bladder cancer as for some other sites. This inconsistency may, in part, be due to differences in cancer registration practices: some registries record benign and non invasive tumours of the bladder and others do not. Bladder cancer is associated with smoking although a variety of' occupational exposures such as aromatic amines have also been identified as risk f-actors. lncidence rates in the EC are lower than in the USA (with the exception of Denmark). Mortality rates, however, are considerably higher.
Trends: Incidence trends in bladder cancer are difficult to interpret because of the difference between cancer registries in recording benign and non invasive tumours. Mortality fates in men have decreased in the UK, Denmark, the Netherlands, Belgium, and Germany, while the increases which have occurred in the southern countries have been relatively small. Mortality rates in women have remained stable.
Survival: Five year relative survival rates are slightly less than 50%.
Considerable variation in incidence and mortality of kidney cancer exists in EC countries with a 2-3-fold difference in both sexes between low rates in Portugal and high rates in Germany and Denmark. Overall the highest rates are in the northern countries. These patterns are the same in males and females. Cigarette smoking is the principal risk factor and may account for between 17% and 45% of kidney cancer. European incidence rates are lower than in the USA (except for females in Denmark). In contrast, mortality is higher in most EC countries except the southernmost than in the USA.
Trends: Incidence and mortality rates are increasing in both sexes in the EC countries, especially in the UK.
Survival: For persons diagnosed with cancer of' the kidney, the five year relative survival rates vary between 30% and 45%.
There is little variation in the rates of' non-Hodgkin lymphoma among the EC countries. Incidence and mortality vary by no more than about 20% to 25% from north to south, and the
rates are about 40% to 50% lower in females than males. Internationally, however, much higher rates are seen in developed countries than in developing countries, This discrepancy, and the changes in incidence and mortality over time suggest that environmental factors make an important contribution to the cause of this cancer. Viral infections, particularly with the Epstein- Barr virus, may account for a proportion of cases, and the human immunodeficiency (AIDS) virus considerably enhances the risk. A small proportion may be due to Occupational
industrial exposures to herbicides. lncidence and mortality are substantially lower in both sexes in the EC than in the USA.
Trends: lncidence rates have increased steeply and mortality less steeply, in both sexes in all
EC countries over the last 25 years. The increase began before the AIDS epidemic.
Survival: The five year relative survival rates are between 30% and 40%.
Hodgkin's disease shows no marked variation in rates within the EC, and there is no evident latitude gradient. The rates are slightly higher for males than females. The international variation in incidence rates of this disease suggests that, like non Hodgkin lymphoma, environmental factors are involved in its causation. It has been suggested that, of the possible environmental factors, those of' infectious origin are the most important and probably operate in childhood or in the young adult. The Epstein-Barr virus is suspected as a possible candidate. lncidence rates in the EC are a little lower than in the USA, but mortality rates in both sexes are at the same level or higher.
Trends: lncidence rates have remained stable in the EC while mortality has declined, less markedly in Belgium and Ireland than elsewhere.
Survival: The five year relative survival rates vary between 60% and 65%.
ln the EC countries, male rates are up to 60% higher than in females, but there is little geographic variation. The modest variation in rates suggests that the effect of environmental factors is weak. Exposures known to cause leukaemia include ionizing radiation, drugs used to treat cancer, some chemicals, and the human T-cell leukaemia virus. Population exposure to these agents, however, is usually too low to account for more than a small percentage of cases. lncidence in the EC is lower than in the USA (except Denmark) although EC mortality is similar to that of the USA.
Trends: lncidence rates have been mainly stable although increases in both sexes have occurred in Denmark and Spain. While mortality rates have decreased in several EC countries, Spain has experienced an increase.
Survival: Survival after diagnosis of leukaemia is poor with five-year relative survival rates between 20% and 40%.
More detailed information on cancer incidence and on trends in both incidence and mortality in E.C. countries and elsewhere is available in the following IARC publications :
These and other IARC Scientific Publications are distributed through Oxford University Press, Walton Street, Oxford OX2 6QS. ln case of difficulty, or for further information on lARC publications, write to : Editorial and Publications Service, International Agency for Research on Cancer, 150 cours Albert Thomas, 69372 Lyon Cedex 08, France.
This information is provided by the International Agency for Research
and has been made available for TeleSCAN by the Netherlands Cancer Institute.
Date last modified: 12-1-1996. Copyright Reserved.