Malignant melanoma is rarer than non-melanocytic skin cancer. Apart from exposure to solar radiation, no other environmental factors show a consistent association with malignant melanoma of the skin. Associations with occupation, diet and hormonal factors are not firmly established (Koh et al. 1993).

Malignant melanoma is an aggressive skin cancer (ICD-9 172.0 to 173.9; ICD-10: C43). It arises from pigment-producing cells of the skin, usually in an existing naevus. The tumour is usually a few millimetres to several centimetres thick, brown or black in colour, that has grown in size, changed colour and may bleed or ulcerate (Balch et al. 1993).

Indicators of poor prognosis of malignant melanoma of the skin include nodular subtype, tumour thickness, multiple primary tumours, metastases, ulceration, bleeding, long tumour duration, body site and, for some tumour sites, male sex. A history of malignant melanoma of the skin increases the risk for a secondary melanoma. Five-year post-diagnosis survival rates in high incidence areas are 80 to 85%, but in low incidence areas the survival is poorer (Ellwood and Koh 1994; Stidham et al. 1994).

There are four histologic types of malignant melanoma of the skin. Superficial spreading melanomas (SSM) represent 60 to 70% of all melanomas in Whites and less in non-Whites. SSMs tend to progress slowly and are more common in women than in men. Nodular melanomas (NM) account for 15 to 30% of malignant melanomas of the skin. They are invasive, grow rapidly and are more frequent in men. Four to 10% of malignant melanomas of the skin are lentigo malignant melanomas (LMM) or Hutchinson’s melanotic freckles. LMMs grow slowly, occur frequently in the face of old persons and rarely metastasize. Acral lentiginous melanomas (ALM) represent 35 to 60% of all malignant melanomas of the skin in non-Whites and 2 to 8% in Whites. They occur frequently on the sole of the foot (Bijan 1993).

For the treatment of malignant melanomas of the skin, surgery, radiation therapy, chemotherapy and biologic therapy (interferon alpha or interleukin-2) may be applied singly or in combination.

During the 1980s, the reported age-standardized annual incidence rates of malignant melanoma of the skin varied per 100,000 from 0.1 in males in Khon Kaen, Thailand to around 30.9 in males and 28.5 in females in Queensland, Australia (IARC 1992b). Malignant melanomas of the skin represent less than 1% of all cancers in most populations. An annual increase of about 5% in melanoma incidence has been observed in most white populations from the early 1960s to about 1972. Melanoma mortality has increased in the last decades in most populations, but less rapidly than incidence, probably due to early diagnoses and awareness of the disease (IARC 1985b, 1992b). More recent data show different rates of change, some of them suggesting even downward trends.

Malignant melanomas of the skin are among the ten most frequent cancers in incidence statistics in Australia, Europe and North America, representing a lifetime risk of 1 to 5%. White-skinned populations are more susceptible than non-White populations. Melanoma risk in white-skinned populations increases with proximity to the equator.

The gender distribution of melanomas of the skin varies widely between populations (IARC 1992a). Women have lower incidence rates than men in most populations. There are gender differences in patterns of body distribution of the lesions: trunk and face dominate in men, extremities in women.

Malignant melanomas of the skin are more common in higher than in lower socio-economic groups (IARC 1992b).

Familial melanomas are uncommon, but have been well documented. with between 4% and 10% of patients describing a history of melanoma among their first degree relatives.

Solar UV-B irradiation is probably the major cause for the widespread increase in the incidence of melanomas of the skin (IARC 1993). It is not clear whether depletion of the stratospheric ozone layer and the consequent increase in UV irradiance has caused the increase in the incidence of malignant melanoma (IARC 1993, Kricker et al. 1993). The effect of UV irradiation depends on some characteristics, such as I or II phenotype and blue eyes. A role for UV radiation emanating from fluorescent lamps is suspected, but not conclusively established (Beral et al. 1982).

It has been estimated that reduction in recreational sun exposure and use of sun-screens could reduce the incidence of malignant melanomas in high risk populations by 40% (IARC 1990). Among outdoor workers, the application of sunscreens having a protective UV-B factor rating of at least 15 and UV-A sunscreen and the use of appropriate clothing are practical protective measures. Although a risk from outdoor occupations is plausible, given the increased exposure to solar radiation, results of studies on regular outdoor occupational exposure are inconsistent. This is probably explained by the epidemiological findings suggesting that it is not regular exposures but rather intermittent high doses of solar radiation that are associated with excess melanoma risk (IARC 1992b).

Therapeutic immunosuppression may result in increased risk of malignant melanoma of the skin. An increased risk with the use of oral contraceptives has been reported, but it seems unlikely to increase the risk of malignant melanoma of the skin (Hannaford et al. 1991). Melanomas can be produced by oestrogen in hamsters. There is no evidence of such an effect in humans.

In White adults, the majority of primary intraocular malignant tumours are melanomas, usually arising from uveal melanocytes. The estimated rates for these cancers do not show the geographic variations and increasing time trends observed for melanomas of the skin. The incidence and mortality of ocular melanomas are very low in Black and Asiatic populations (IARC 1990, Sahel et al. 1993) The causes of ocular melanoma are unknown (Higginson et al. 1992).

In epidemiological studies, excess risk for malignant melanoma has been observed in administrators and managers, airline pilots, chemical processing workers, clerks, electrical power workers, miners, physical scientists, policemen and guards, refinery workers and gasoline exposed workers, salesmen and warehouse clerks. Excess melanoma risks have been reported in industries such as cellulose fibre production, chemical products, clothing industry, electrical and electronics products, metal industry, non-metallic mineral products, petrochemical industry, printing industry and telecommunications. Many of these findings are, however, solitary and have not been replicated in other studies. A series of meta-analyses of cancer risks in farmers (Blair et al. 1992; Nelemans et al. 1993) indicated a slight, but significant excess (aggregated risk ratio of 1.15) of malignant melanoma of the skin in 11 epidemi-ological studies.

In a multi-site case-control study of occupational cancer in Montreal, Canada (Siemiatycki et al. 1991), the following occupational exposures were associated with a significant excess of malignant melanoma of the skin: chlorine, propane engine emissions, plastics pyrolysis products, fabric dust, wool fibres, acrylic fibres, synthetic adhesives, “other” paints, varnishes, chlorinated alkenes, trichloroethylene and bleaches. It was estimated that the population attributable risk due to occupational exposures based on the significant associations in the data of the same study was 11.1%.

Back

There are three histological types of non-melanocytic skin cancers (NMSC) (ICD-9: 173; ICD-10: C44): basal cell carcinoma, squamous cell carcinoma and rare soft tissue sarcomas involving the skin, subcutaneous tissue, sweat glands, sebaceous glands and hair follicles.

Basal cell carcinoma is the most common NMSC in white populations, representing 75 to 80% of them. It develops usually on the face, grows slowly and has little tendency to metastasize.

Squamous cell cancers account for 20 to 25% of reported NMSCs. They can occur on any part of the body, but especially on the hands and legs and can metastasize. In darkly pigmented populations squamous cell cancers are the most common NMSC.

Multiple primary NMSCs are common. The bulk of the NMSCs occur on the head and neck, in contrast with most of the melanomas which occur on the trunk and limbs. The localization of NMSCs reflects clothing patterns.

NMSCs are treated by various methods of excision, radiation and topical chemotherapy. They respond well to treatment and over 95% are cured by excision (IARC 1990).

The incidence of NMSCs is hard to estimate because of gross underreporting and since many cancer registries do not record these tumours. The number of new cases in the US was estimated at 900,000 to 1,200,000 in 1994, a frequency comparable to the total number of all non-cutaneous cancers (Miller & Weinstock 1994). The reported incidences vary widely and are increasing in a number of populations, e.g., in Switzerland and the US. The highest annual rates have been reported for Tasmania (167/100,000 in men and 89/100,000 in women) and the lowest for Asia and Africa (overall 1/100,000 in men and 5/100,000 in women). NMSC is the most common cancer in Caucasians. NMSC is about ten times as common in White as in non-White populations. The lethality is very low (Higginson et al. 1992).

Susceptibility to skin cancer is inversely related to the degree of melanin pigmentation, which is thought to protect by buffering against the carcinogenic action of solar ultraviolet (UV) radiation. Non-melanoma risk in white-skinned populations increases with the proximity to the equator.

In 1992, the International Agency for Research on Cancer (IARC 1992b) evaluated the carcinogenicity of solar radiation and concluded that there is sufficient evidence in humans for the carcinogenicity of solar radiation and that solar radiation causes cutaneous malignant melanoma and NMSC.

Reduction of exposure to sunlight would probably reduce the incidence of NMSCs. In Whites, 90 to 95% of NMSCs are attributable to solar radiation (IARC 1990).

NMSCs may develop in areas of chronic inflammation, irritation and scars from burns. Traumas and chronic ulcers of the skin are important risk factors for squamous cell skin cancers, particularly in Africa.

Radiation therapy, chemotherapy with nitrogen mustard, immunosuppressive therapy, psoralen treatment combined with UV-A radiation and coal tar preparations applied on skin lesions have been associated with an increased risk of NMSC. Environmental exposure to arsenic trivalent and arsenical compounds have been confirmed to be associated with skin cancer excess in humans (IARC 1987). Arsenicism can give rise to palmar or plantar arsenical keratoses, epidermoid carcinoma and superficial basal cell carcinoma.

Hereditary conditions such as lack of enzymes required to repair the DNA damaged by UV radiation may increase the risk of NMSC. Xeroderma pigmentosum represents such a hereditary condition.

A historical example of an occupational skin cancer is scrotal cancer that Sir Percival Pott described in chimney sweeps in 1775. The cause of these cancers was soot. In the early 1900s, scrotal cancers were observed in mulespinners in cotton textile factories where they were exposed to shale oil, which was used as a lubricant for cotton spindles. The scrotal cancers in both chimney sweeps and mulespinners were later associated with polycyclic aromatic hydrocarbons (PAHs), many of which are animal carcinogens, particularly some 3-, 4- and 5-ring PAHs such as benz(a)pyrene and dibenz(a,h)anthracene (IARC 1983, 1984a, 1984b, 1985a). In addition to mixtures that readily contain carcinogenic PAHs, carcinogenic compounds may be formed by cracking when organic compounds are heated.

Further occupations with which PAH-related excesses of NMSC have been associated include: aluminium reduction workers, coal gasification workers, coke oven workers, glass blowers, locomotive engineers, road pavers and highway maintenance workers, shale oil workers, tool fitters and tool setters (see table 1). Coal tars, coal-based pitches, other coal-derived products, anthracene oil, creosote oil, cutting oils and lubricating oils are some of the materials and mixtures that contain carcinogenic PAHs.

Table 1. Occupations at risk

Additional job titles that have been associated with increased NMSC risk include jute processors, outdoor workers, pharmacy technicians, sawmill workers, shale oil workers, sheep-dip workers, fishermen, tool setters, vineyard workers and watermen. The excess for watermen (who are primarily involved in traditional fishing tasks) was noticed in Maryland, USA and was confined to squamous cell cancers. Solar radiation probably explains fishermen’s, outdoor workers’, vineyard workers’ and watermen’s excess risks. Fishermen also may be exposed to oils and tar and inorganic arsenic from the consumed fish, which may contribute to the observed excess, which was threefold in a Swedish study, as compared with the county-specific rates (Hagmar et al. 1992). The excess in sheep dip workers may be explained by arsenical compounds, which induce skin cancers through ingestion rather than through skin contact. While farmers have slightly increased risk of melanoma, they do not appear to have increased risk of NMSC, based on epidemiological observations in Denmark, Sweden and the USA (Blair et al. 1992).

Ionizing radiation has caused skin cancer in early radiologists and workers who handled radium. In both situations, the exposures were long-lasting and massive. Occupational accidents involving skin lesions or long-term cutaneous irritation may increase the risk on NMSC.

Prevention (of Non-Melanocytic OccupationalSkin Cancer)

The use of appropriate clothing and a sunscreen having a protective UV-B factor of 15 or greater will help protect outdoor workers exposed to ultraviolet radiation. Further, the replacement of carcinogenic materials (such as feed stocks) by non-carcinogenic alternatives is another obvious protective measure which may, however, not always be possible. The degree of exposure to carcinogenic materials can be reduced by the use of protective shields on equipment, protective clothing and hygienic measures.

Of overriding importance is the education of the workforce about the nature of the hazard and the reasons for and value of the protective measures.

Finally, skin cancers usually take many years to develop and many of them pass through several premalignant stages before achieving their full malignant potential such as arsenic keratoses and actinic keratoses. These early stages are readily detectable by visual inspection. For this reason, skin cancers offer the real possibility that regular screening could reduce mortality among those known to have been exposed to any skin carcinogen.

Back

Lung Cancer

Lung cancer is the most common cancer worldwide. In 1985, it is estimated that worldwide 676,500 cases occurred in males and 219,300 cases in females, accounting for 11.8% of all new cancers, and this figure is increasing at a rate of about 0.5% per year (Parkin, Pisani and Ferlay 1993). About 60% of these cases occur in industrialized countries, in many of which lung cancer is the leading cancer cause of death among males. In both industrialized and developing countries, males have a higher incidence than females, the sex ratio ranging from two- to ten-fold. The international intergender variations in lung cancer incidence are largely explained by the variation in current and past smoking patterns.

A higher lung cancer risk has been consistently observed in urban compared to rural areas. In industrialized countries, a clear, inverse relationship is evident in males in lung cancer incidence and mortality by social class, while women show less clear and consistent patterns. Differences in social class in males principally reflect a different smoking pattern. In developing countries, however, there seems to be a higher risk in men from the upper social class than in other men: this pattern probably reflects the earlier adoption of Western habits by affluent groups in the population.

Incidence data from the United States National Cancer Institute’s SEER Program for the period 1980-86 indicate, similarly to previous years, that Black males have a higher incidence than White males, while incidence for females does not differ by race. These differences among ethnic groups in the United States can actually be attributed to socio-economic differences between Blacks and Whites (Baquet et al. 1991).

Lung cancer incidence increases almost linearly with age, when plotted in a log-log scale; only in the oldest age groups can a downward curve be observed. Lung cancer incidence and mortality have increased rapidly during this century, and continue to increase in most countries.

There are four principal histological types of lung cancer: squamous cell carcinoma, adenocarcinoma, large cell carcinoma and small cell carcinoma (SCLC). The first three are also referred to as non-small cell lung cancer (NSCLC). The proportions of each histological type change according to sex and age.

Squamous cell carcinoma is very strongly associated with smoking and represents the most common type of lung cancer in many populations. It arises most frequently in the proximal bronchi.

Adenocarcinoma is less strongly associated with smoking. This tumour is peripheral in origin and may present as a solitary peripheral nodule, a multifocal disease or a rapidly progressive pneumonic form, spreading from lobe to lobe.

Large cell carcinoma represents a smaller proportion of all lung cancers and has a similar behaviour as adenocarcinoma.

SCLC represents a small proportion (10 to 15%) of all lung cancers; it typically arises in the central endobronchial location and tends to develop early metastases.

The signs and symptoms of lung cancer depend on the location of the tumour, the spread and the effects of metastatic growth. Many patients present with an asymptomatic lesion discovered incidentally on x ray. Among NSCLC patients, fatigue, decreased activity, persistent cough, dyspnoea, decreased appetite and weight loss are common. Wheeze or stridor may also develop in advanced stages. Continuous growth may result in atelectasia, pneumonia and abscess formation. Clinical signs among SCLC patients are less pronounced than among those with NSCLC, and are usually related to the endobronchial location.

Lung cancer can metastasize to virtually any organ. The most common locations of metastatic lesions are pleura, lymph nodes, bone, brain, adrenals, pericardium and liver. At the moment of diagnosis, the majority of patients with lung cancer have metastases.

The prognosis varies with the stage of the disease. Overall five-year survival for lung cancer patients in Europe (in 1983-85) was between 7% and 9% (Berrino et al. 1995).

No population screening method is currently available for lung cancer.

Nasopharyngeal Cancer

Nasopharyngeal cancer is rare in most populations, but is frequent in both sexes in areas such as South-East Asia, Southern China and North Africa. Migrants from South China retain the high risk to a large extent, but second- and third-generation Chinese migrants to the United States have less than half the risk of first generation migrants.

Cancers of the nasopharynx are predominantly of squamous epithelial origin. According to WHO, these tumours are classified as: type 1, keratinizing squamous cell carcinoma; type 2, non-keratinizing carcinoma; and type 3, undifferentiated carcinoma, which is the most frequent histological type. Type 1 has an uncontrolled local growth, and metastatic spread is found in 60% of the patients. Types 2 and 3 have metastatic spread in 80 to 90% of the patients.

A mass in the neck is noticed in approximately 90% of nasopharyngeal carcinoma patients. Alterations in the hearing, serous otitis media, tinnitus, nasal obstruction, pain and symptoms related to the growth of the tumour into adjacent anatomical structures may be noticed.

The overall five-year survival for nasopharyngeal cancer patients in Europe between 1983 and 1985 was around 35%, varying according to the stage of the tumour and its location (Berrino et al. 1995).

Consumption of Chinese-style salted fish is a risk factor of nasopharyngeal cancer; the role of other nutritional factors and of viruses, in particular Epstein-Barr virus, although suspected, has not been confirmed. No occupational factors are known to cause nasopharyngeal cancer. No preventive measures are available at present (Higginson, Muir and Muñoz 1992).

Sinonasal Cancer

Neoplasms of the nose and nasal cavities are relatively rare. Together, cancer of the nose and nasal sinus—including maxillary, ethmoidal, sphenoid and frontal sinuses—account for less than 1% of all cancers. In most cases these tumours are classified as squamous carcinomas. In Western countries, cancers of the nose are more common than cancers of the nasal sinus (Higginson, Muir and Muñoz 1992).

They occur more frequently in men and among Black populations. The highest incidence is seen in Kuwait, Martinique and India. The peak of development of the disease occurs during the sixth decade of life. The major known cause of sinonasal cancer is exposure to wood dust, in particular from hardwood species. Tobacco smoking does not seem to be associated with this type of cancer.

Most tumours of the nasal cavity and para-nasal sinus are well differentiated and slow growing. Symptoms may include non-healing ulcer, bleeding, nasal obstruction and symptoms related to the growth into the oral cavity, orbit and pterygoid fossa. The disease is usually advanced at the time of diagnosis.

Overall five-year survival for nose and nasal sinus cancer patients in Europe between 1983 and 1985 was around 35%, varying according to the size of the lesion at diagnosis (Berrino et al. 1995).

Laryngeal Cancer

The highest incidence of laryngeal cancer is reported in Sao Paolo (Brazil), Navarra (Spain) and Varese (Italy). High mortality has also been reported in France, Uruguay, Hungary, Yugoslavia, Cuba, the Middle East and North Africa. Laryngeal cancer is predominantly a male cancer: an estimated 120,500 cases among males and 20,700 cases among females occurred in 1985 (Parkin, Pisani and Ferlay 1993). In general, incidence is higher among Black populations as compared to Whites, and in urban areas compared to rural.

Almost all cancers of the larynx are squamous carcinomas. The majority are located in the glottis, but they may also develop in the supraglottis or, rarely, in the subglottis.

Symptoms may not occur or be very subtle. Pain, a scratchy sensation, alteration of tolerance for hot or cold foods, a tendency to aspirate liquids, airway alteration, a slight change in the voice during several weeks and cervical adenopathy may be present, according to the location and stage of the lesion.

Most larynx cancers are visible with laryngeal inspection or endoscopy. Pre-neoplastic lesions can be identified in the larynx of smokers (Higginson, Muir and Muñoz 1992).

The overall five-year survival for laryngeal cancer patients in Europe between 1983 and 1985 was around 55% (Berrino et al. 1995).

Pleural Mesothelioma

Mesotheliomas may arise from the pleura, peritoneum and pericardium. Malignant mesothelioma represents the most important pleural tumour; it occurs mainly between the fifth and seventh decade of life.

Pleural mesothelioma was once a rare tumour and remains so in most female populations, while in men in industrialized countries it has increased by 5 to 10% per year during the last decades. In general, men are affected five times as much as women. Precise estimates of incidence and mortality are problematic because of difficulties in the histological diagnosis and changes in the International Classification of Diseases (ICD) (Higginson, Muir and Muñoz 1992). However, incidence rates seem to present very important local variations: they are very high in areas where asbestos mining is present (e.g., North West Cape Province of South Africa), in major naval dockyard cities, and in regions with environmental fibre contamination, such as certain areas of central Turkey.

Patients may be asymptomatic and have their disease diagnosed incidentally on chest radiographs, or they may have dyspnoea and chest pain.

Mesotheliomas tend to be invasive. The median survival is 4 to 18 months in various series.

Occupational Risk Factors of Respiratory Cancer

Apart from tobacco smoke, a causal association with respiratory cancer has been demonstrated according to the International Agency for Research on Cancer (IARC) for 13 agents or mixtures and nine exposure circumstances (see table 1). Furthermore, there are eight agents, mixtures or exposure circumstances which according to IARC are probably carcinogenic to one or more organs in the respiratory tract (table 2). All but azathioprine, an immunosuppressant drug, are primarily occupational exposures (IARC 1971-94).

Table 1. Established human respiratory carcinogens according to IARC

 Source: IARC, 1971-1994.

Table 2. Probable human respiratory carcinogens according to IARC

Source: IARC, 1971-1994.

Occupational groups demonstrating an increased risk of lung cancer following exposure to arsenic compounds include non-ferrous smeltery workers, fur handlers, manufacturers of sheep-dip compounds and vineyard workers (IARC 1987).

A large number of epidemiological studies have been carried out on the association between chromium (VI) compounds and the occurrence of lung and nasal cancer in the chromate, chromate pigment and chromium plating industries (IARC 1990a). The consistency of findings and the magnitude of the excesses have demonstrated the carcinogenic potential of chromium (VI) compounds.

Nickel refinery workers from many countries have shown substantial increased risks of lung and nasal cancers; other occupational groups exposed to nickel among which an increased risk of lung cancer has been detected include sulphide nickel ore miners and high nickel alloy manufacture workers (IARC 1990b).

Workers exposed to beryllium are at elevated risk of lung cancer (IARC 1994a). The most informative data are those derived from the US Beryllium Case Registry, in which cases of beryllium-related lung diseases were collected from different industries.

An increase in lung cancer occurrence has been found in cohorts of cadmium smelters and nickel-cadmium battery workers (IARC 1994b). Concurrent exposure to arsenic among smelters and to nickel among battery workers, cannot explain such an increase.

Asbestos is an important occupational carcinogen. Lung cancer and mesothelioma are the major asbestos-related neoplasms, but cancers at other sites, such as the gastro-intestinal tract, larynx and kidney, have been reported in asbestos workers. All forms of asbestos have been causally related to lung cancer and mesothelioma. In addition, talc-containing asbestiform fibres have been shown to be carcinogenic to the human lung (IARC 1987). A distinctive characteristic of asbestos-induced lung cancer is its synergistic relationship with cigarette smoking.

A number of studies among miners, quarry workers, foundry workers, ceramic workers, granite workers and stone cutters have shown that individuals diagnosed as having silicosis after exposure to dust containing crystalline silica have an increased risk of lung cancer (IARC 1987).

Polynuclear aromatic hydrocarbons (PAHs) are formed mainly as a result of pyrolytic processes, especially the incomplete combustion of organic materials. However, humans are exposed exclusively to mixtures of PAHs, such as soots, coal-tars and coal-tar pitches. Cohort studies of mortality among chimney-sweeps have shown an increased risk of lung cancer, which has been attributed to soot exposure. Several epidemiological studies have shown excesses of respiratory cancer among workers exposed to pitch fumes in aluminium production, calcium carbide production and roofing. In these industries, exposure to tar, and particularly coal tar, does also occur. Other industries in which an excess of respiratory cancer is due to exposure to coal-tar fumes are coal gasification and coke production (IARC 1987). An increased risk of respiratory (mainly lung) cancer was found in some, but not all the studies tried to analyse diesel engine exhaust exposure separately from other combustion products; the occupational groups which were studied include railroad workers, dockers, bus garage workers, bus company employees and professional lorry drivers (IARC 1989a). Other mixtures of PAHs that have been studied for their carcinogenicity to humans include carbon blacks, gasoline engine exhaust, mineral oils, shale oils, and bitumens. Shale oils and untreated and mildly treated mineral oils are carcinogenic to humans, whereas gasoline engine exhaust is possibly carcinogenic and highly refined mineral oils, bitumens and carbon blacks are not classifiable as to their carcinogenicity to humans (IARC 1987, 1989a). Although these mixtures do contain PAHs, a carcinogenic effect on the human lung has not been demonstrated for any of them, and the evidence of carcinogenicity for untreated and mildly treated mineral oils and for shale oils is based on increased risk of cancers from sites other than respiratory organs (mainly skin and scrotum) among exposed workers.

Bis(b-chloroethyl)sulphide, known as mustard gas, was widely used during the First World War, and the studies of soldiers exposed to mustard gas as well as of workers employed in its manufacture have revealed a subsequent development of lung and nasal cancer (IARC 1987).

Numerous epidemiological studies have demonstrated that workers exposed to chloromethyl methyl ether and/or bis(chloromethyl)-ether have an increased risk of lung cancer, primarily of SCLC (IARC 1987).

Workers exposed to acrylonitrile have been found to be at higher risk of lung cancer in some but not all studies which have been conducted among workers in textile fibre manufacture, acrylonitrile polymerization and the rubber industry (IARC 1987).

Excess occurrence has been reported for workers exposed to formaldehyde, including chemical workers, wood workers, and producers and users of formaldehyde (IARC 1987). The evidence is strongest for nasal and nasopharyngeal cancer: the occurrence of these cancers showed a dose-response gradient in more than one study, although the number of exposed cases was often small. Other neoplasms at possible increased risk are lung and brain cancer and leukaemia.

An increased risk of laryngeal cancer has been found in several studies of workers exposed to mists and vapours of sulphuric and other strong inorganic acids, such as workers in steel pickling operations, and in soap manufacture and petrochemical workers (IARC 1992). Lung cancer risk was also increased in some, but not all, of these studies. Furthermore, an excess of sinonasal cancer was found in a cohort of workers in isopropanol manufacture using the strong-acid process.

Woodworkers are at increased risk of nasal cancer, in particular adenocarcinoma (IARC 1987). The risk is confirmed for furniture and cabinet-makers; studies on workers in carpentry and joinery suggested a similar excess risk, but some studies produced negative results. Other wood industries, such as sawmills and pulp and paper manufacture, were not classifiable as to their carcinogenic risk. Although carcinogenicity of wood dust was not evaluated by IARC, it is plausible to consider that wood dust is responsible for at least part of the increased risk of nasal adenocarcinoma among woodworkers. Woodworkers do not seem to be at increased risk of cancer in other respiratory organs.

Nasal adenocarcinoma has been caused also by employment in boot and shoe manufacture and repair (IARC 1987). No clear evidence is available, on the other hand, that workers employed in the manufacture of leather products and in leather tanning and processing are at increased risk of respiratory cancer. It is not known at present whether the excess of nasal adenocarcinoma in the boot and shoe industry is due to leather dust or to other exposures. Carcinogenicity of leather dust has not been evaluated by IARC.

Lung cancer has been common among uranium miners, underground hematite miners and several other groups of metal miners (IARC 1988; BEIR IV Committee on the Biological Effects of Ionizing Radiation 1988). A common factor among each of these occupational groups is exposure to a-radiation emitted by inhaled radon particles. The main source of data on cancer following exposure to ionizing radiation is derived from the follow-up of atomic bomb survivors (Preston et al. 1986; Shimizu et al. 1987). The risk of lung cancer is elevated among the atomic bomb survivors as well as among people who have received radiation therapy (Smith and Doll 1982). No convincing evidence, however, is currently available on the existence of an elevated lung cancer risk among workers exposed to low-level ionizing radiation, such as those occurring in the nuclear industry (Beral et al. 1987; BEIR V, Committee on the Biological Effects of Ionizing Radiation 1990). Carcinogenicity of ionizing radiation has not been evaluated by IARC.

An elevated risk of lung cancer among painters was found in three large cohort studies and in eight small cohort and census-based studies, as well as eleven case-control studies from various countries. On the other hand, little evidence of an increase in lung cancer risk was found among workers involved in the manufacture of paint (IARC 1989b).

A number of other chemicals, mixtures, occupations and industries which have been evaluated by IARC to be carcinogenic to humans (IARC Group 1) do not have the lung as the primary target organ. Nonetheless, the possibility of an increased risk of lung cancer has been raised for some of these chemicals, such as vinyl chloride (IARC 1987), and occupations, such as spraying and application of insecticides (IARC 1991a), but the evidence is not consistent.

Furthermore, several agents which have the lung as one of the main targets, have been considered to be possible human carcinogens (IARC Group 2B), on the basis of carcinogenic activity in experimental animals and/or limited epidemiological evidence. They include inorganic lead compounds (IARC 1987), cobalt (IARC 1991b), man-made vitreous fibres (rockwool, slagwool and glasswool) (IARC 1988b), and welding fumes (IARC 1990c).

Back

Kidney Cancer

Epidemiology

Historically, kidney cancer has been used to mean either all malignancies of the renal system (renal cell carcinoma (RCC), ICD-9 189.0; renal pelvis, ICD-9 189.1; and ureter, ICD-9 189.2) or RCC only. This categorization has led to some confusion in epidemiological studies, resulting in a need to scrutinize previously reported data. RCC comprises 75 to 80% of the total, with the remainder being primarily transitional cell carcinomas of the renal pelvis and ureter. Separation of these two cancer types is appropriate since the pathogenesis of RCC and of transitional cell carcinoma is quite different, and epidemiological risk factors are distinct as are the signs and symptoms of the two diseases. This section focuses on RCC.

The major identified risk factor for kidney cancer is tobacco smoking, followed by suspected but poorly defined occupational and environmental risk factors. It is estimated that the elimination of tobacco smoking would decrease the incidence of kidney cancer by 30 to 40% in industrialized countries, but occupational determinants of RCC are not well established. The population attributable risk due to occupational exposures has been estimated to be between zero, based on recognized carcinogenesis, and 21%, based on a multicentric multisite case-control study in the Montreal area of Canada. Early biomarkers of effect in association with biomarkers of exposure should assist in clarifying important risk factors. Several occupations and industries have been found in epidemiological studies to entail an increased risk of renal cancer. However, with the possible exception of agents used in dry cleaning and exposures in petroleum refining, the available evidence is not consistent. Statistical analysis of epidemiological exposure data in relation to biomarkers of susceptibility and effect will clarify additional aetiological causes.

Several epidemiological studies have associated specific industries, occupations and occupational exposures with increased risks of renal cell carcinoma. The pattern that emerges from these studies is not fully consistent. Oil refining, printing, dry cleaning and truck driving are examples of jobs associated with excess risk of kidney cancer. Farmers usually display decreased risk of RCC, but a Danish study linked long-term exposure to insecticides and herbicides with an almost fourfold excess of RCC risk. This finding requires confirmation in independent data, including specification of the possible causal nature of the association. Other products suspected of being associated with RCC include: various hydrocarbon derivatives and solvents; products of oil refining; petroleum, tar and pitch products; gasoline exhaust; jet fuel; jet and diesel engine emissions; arsenic compounds; cadmium; chromium (VI) compounds; inorganic lead compounds; and asbestos. Epidemiological studies have associated occupational gasoline vapour exposure with kidney cancer risk, some in a dose-response fashion, a phenomenon observed in the male rat for unleaded gasoline vapour exposure. These findings gain some potential weight, given the widespread human exposure to gasoline vapours in retail service stations and the recent increase in kidney cancer incidence. Gasoline is a complex mixture of hydrocarbons and additives, including benzene, which is a known human carcinogen.

The risk of kidney cancer is not consistently linked with social class, although increased risk has occasionally been associated with higher socio-economic status. However, in some populations a reverse gradient was observed, and in yet others, no clear pattern emerged. Possibly these variations may be related to lifestyle. Studies with migrant people show modification in RCC risk towards the level of the host country population, suggesting that environmental factors are important in the development of this malignancy.

Except for nephroblastoma (Wilms’ tumour), which is a childhood cancer, kidney cancer usually occurs after 40 years of age. An estimated 127,000 new cases of kidney cancer (including RCC and transitional cell carcinoma (TCC) of the renal pelvis and ureter), corresponding to 1.7% of the world total cancer incidence, occurred globally in 1985. The incidence of kidney cancer varies among populations. High rates have been reported for both men and women in North America, Europe, Australia and New Zealand; low rates in Melanesia, middle and eastern Africa and southeastern and eastern Asia. The incidence of kidney cancer has been increasing in most western countries, but stagnated in a few. Age-standardized incidence of kidney cancer in 1985 was highest in North America and western, northern and eastern Europe, and lowest in Africa, Asia (except in Japanese men) and the Pacific. Kidney cancer is more frequent in men than in women and ranks among the ten most frequent cancers in a number of countries.

Transitional cell carcinoma (TCC) of the renal pelvis is associated with similar aetiological agents as bladder cancer, including chronic infection, stones and phenacetin-containing analgesics. Balkan nephropathy, a slowly progressive, chronic and fatal nephropathy prevalent in the Balkan countries, is associated with high rates of tumours of the renal pelvis and ureter. The causes of Balkan nephropathy are unknown. Excessive exposure to ochratoxin A, which is considered possibly carcinogenic to humans, has been associated with the development of Balkan nephropathy, but the role of other nephrotoxic agents cannot be excluded. Ochratoxin A is a toxin produced by fungi which can be found in many food stuffs, particularly cereals and pork products.

Screening and diagnosis of kidney cancer

The sign and symptom pattern of RCC varies among patients, even up to the stage when metastasis appears. Because of the location of the kidneys and the mobility of contiguous organs to the expanding mass, these tumours are frequently very large at the  time  of  clinical  detection.  Although  haematuria  is  the primary symptom of RCC, bleeding occurs late compared to transitional cell tumours because of the intra-renal location of RCC. RCC has been considered the “medical doctor’s dream” but the “surgeon’s curse” because of the interesting constellation of symptoms related to paraneoplastic syndromes. Substances that increase the red blood cell count, calcium and factors which mimic abnormal adrenal gland function have been reported, and abdominal mass, weight loss, fatigue, pain, anaemia, abnormal liver function and hypertension have all been observed. Computerized axial tomography (CAT scan) of the abdomen and ultrasound are being ordered by physicians with increased frequency so, consequently, it is estimated that 20% of RCCs are diagnosed serendipitously as a result of evaluation for other medical problems.

Clinical evaluation of an RCC case consists of a physical examination to identify a flank mass, which occurs in 10% of patients. A kidney x ray with contrast may delineate a renal mass and the solid or cystic nature is usually clarified by ultrasound or CAT scan. The tumours are highly vascular and have a characteristic appearance when the artery is injected with radio-opaque contrast material. Arteriography is performed to embolize the tumour if it is very large or to define the arterial blood supply if a partial nephrectomy is anticipated. Fine-needle aspiration may be used to sample suspect RCC.

Localized RCC tumours are surgically removed with regional lymph nodes and, operatively, early ligation of the artery and vein is important. Symptomatically, the patient may be improved by removing large or bleeding tumours that have metastasized, but this does not improve survival. For metastatic tumours, localized pain control may be achieved with radiation therapy but the treatment of choice for metastatic disease is biological response modifiers (Interleukin-2 or α-interferon), although chemotherapy is occasionally used alone or in combination with other therapies.

Markers such as the cancer gene on chromosome 3 observed in cancer families and in von Hippel-Lindau disease may serve as biomarkers of susceptibility. Although tumour marker antigens have been reported for RCC, there is currently no way to detect these reliably in the urine or blood with adequate sensitivity and specificity. The low prevalence of this disease in the general population requires a high specificity and sensitivity test for early disease detection. Occupational cohorts at risk could potentially be screened with ultrasound. Evaluation of this tumour remains a challenge  to  the  basic  scientist,  molecular  epidemiologist  and clinician alike.

Bladder Cancer

Epidemiology

More than 90% of bladder cancers in Europe and North America are transitional cell carcinomas (TCC). Squamous cell carcinoma and adenocarcinoma account for 5 and 1%, respectively, of bladder cancer in these regions. The distribution of histopathological types in bladder cancer is strikingly different in regions such as the Middle East and Africa where bladder cancer is associated with schistosomal infection. For instance, in Egypt, where schistosomiasis is endemic and bladder cancer is the major oncogenic problem, the most common type is squamous cell carcinoma, but the incidence of TCC is increasing with the rising prevalence of cigarette smoking. The discussion which follows focuses on TCC.

Bladder cancer continues to be a disease of significant importance. It accounted for about 3.5% of all malignancies in the world in 1980. In 1985, bladder cancer was estimated to be 11th in frequency on a global scale, being the eighth most frequent cancer among men, with an expected total of 243,000 new cases. There is a peak incidence in the seventh decade of life, and worldwide the male to female ratio is around three to one. Incidence has been increasing in almost all populations in Europe, particularly in men. In Denmark, where annual incidence rates are among the highest in the world, at 45 per 100,000 in men and 12 per 100,000 in women, the recent trend has been a further rise of 8 to 9% every 5 years. In Asia, the very high rates among the Chinese in Hong Kong have declined steadily, but in both sexes bladder cancer incidence is still much higher than elsewhere in Asia and more than twice as high as that among the Chinese in Shanghai or Singapore. Bladder cancer rates among the Chinese in Hawaii are also high.

Cigarette smoking is the single most important aetiological factor in bladder cancer, and occupational exposures rank second. It has been estimated that tobacco is responsible for one-third of all bladder cancer cases outside of regions where schistosomal infection is prevalent. The number of bladder cancer cases attributed in 1985 to tobacco smoking has been estimated at more than 75,000 worldwide, and may account for 50% of bladder cancer in western populations. The fact that all individuals who smoke similar amounts do not develop bladder cancer at the same rate suggests genetic factors are important in controlling the susceptibility. Two aromatic amines, 4-aminobiphenyl and 2-naphthylamine, are carcinogens associated with cigarette smoking; these are found in higher concentrations in “black tobacco” (air-cured) than in “blend tobacco” (flue-cured). Passive smoke increases the adducts in the blood and a dose-response of adduct formation has been correlated with increased risk of bladder cancer. Higher levels of adduct formation have been observed in cigarette smokers who are slow acetylators compared to fast acetylators, which suggests that genetically inherited acetylation status may be an important biomarker of susceptibility. The lower incidence of bladder cancer in Black compared to White races may be attributed to conjugation of carcinogenic metabolic intermediates by sulphotransferases that produce electrophiles. Detoxified phenolic sulphates may protect the urothelium. Liver sulphotransferase activity for N-hydroxyarylamines has been reported to be higher in Blacks than Whites. This may result in a decrease in the amount of free N-hydroxymetabolites to function as carcinogens.

Occupational bladder cancer is one of the earliest known and best  documented  occupational  cancers.  The  first  identified case of occupational bladder cancer appeared some 20 years after the inception of the synthetic dye industry in Germany. Numerous  other  occupations  have  been  identified  in  the  last  25 years as occupational bladder cancer risks. Occupational exposures may contribute to up to 20% of bladder cancers. Workers occupationally exposed include those working with coal-tar pitches, coal gasification and production of rubber, aluminium, auramine and magenta, as well as those working as hairdressers and barbers. Aromatic amines have been shown to cause bladder cancer in workers in many countries. Notable among this class of chemicals are 2-naphthylamine, benzidine, 4-nitrobiphenyl and 3,3r´-dichlorobenzidine. Two other aromatic amines, 4,4´-methylene dianiline (MDA) and 4,4´-methylene-bis-2-chloroaniline (MOCA) are among the most widely used of the suspected bladder carcinogens. Other carcinogens associated with industrial exposures are largely undetermined; however, aromatic amines are frequently present in the workplace.

Screening and diagnosis of bladder cancer

Screening for bladder cancer continues to receive attention in the quest to diagnose bladder cancer before it becomes symptomatic and, presumably, less amenable to curative treatment. Voided urine cytology and urinalysis for haematuria have been considered candidate screening tests. A pivotal question for screening is how to identify high-risk groups and then individuals within these groups. Epidemiological studies identify groups at risk while biomarkers potentially identify individuals within groups. In general, occupational screening for bladder cancer with haematuria testing and Papanicolaou cytology has been ineffective.

Improved detection of bladder cancer may be possible using the 14-day hemastick testing described by Messing and co-workers. A positive test was observed at least once in 84% of 31 patients with bladder cancer at least 2 months prior to the cystoscopic diagnosis of disease. This test suffers from a false-positive rate of 16 to 20% with half of these patients having no urological disease. The low cost may make this a useful test in a two-tier screen in combination with biomarkers and cytology (Waples and Messing 1992).

In a recent study, the DD23 monoclonal antibody using quantitative fluorescence image analysis detected bladder cancer in exfoliated uroepithelial cells. A sensitivity of 85% and specificity of 95% were achieved in a mixture of low- and high-grade transitional cell carcinomas including TaT1 tumours. The M344 tumour-associated antigen in conjunction with DNA ploidy had a sensitivity approaching 90%.

Recent studies indicate combining biomarkers with haematuria testing may be the best approach. A list of the applications of quantitative fluorescence urinary cytology in combination with biomarkers is summarized in Table 1. Genetic, biochemical and morphological early cell changes associated with premalignant conditions support the concept that individuals at risk can be identified years in advance of the development of overt malignancy. Biomarkers of susceptibility in combination with biomarkers of effect promise to detect individuals at risk with an even higher precision. These advances are made possible by new technologies capable of quantitating phenotypic and genotypic molecular changes at the single cell level thus identifying individuals at risk. Individual risk assessment facilitates stratified, cost-effective monitoring of selected groups for targeted chemoprevention.

Table 1. Applications of urinary cytology

Detection of CIS¹ and bladder cancer

Monitoring surgical therapy:

Monitoring bladder following TURBT²
Monitoring upper urinary tract
Monitoring urethral remnant
Monitoring urinary diversion

Monitoring intravesical therapy

Selecting intravesical therapy

Monitoring effect of laser therapy

Evaluation of patients with haematuria

Establishing need for cystoscopy

Screening high-risk populations:
Occupational exposure groups
Drug abuse groups at risk for bladder cancer

Decision criteria for:
Cystectomy
Segmental ureteral resection versus nephroureterectomy

Other indications:
Detecting vesicoenteric fistula
Extraurological tumours invading the urinary tract
Defining effective chemopreventive agents
Monitoring effective chemotherapy

¹ CIS, carcinoma in situ.

² TURBT, transurethral resection for bladder tumour.
Source: Hemstreet et al. 1996.

Signs and symptoms of bladder cancer are similar to those of urinary tract infection and may include pain on urination, frequent voiding and blood and pus cells in the urine. Because symptoms of a urinary tract infection may herald a bladder tumour particularly when associated with gross haematuria in older patients, confirmation of the presence of bacteria and a keen awareness by the examining physician is needed. Any patient treated for a urinary tract infection which does not resolve immediately should be referred to a urology specialist for further evaluation.

Diagnostic evaluation of bladder cancer first requires an intravenous pyelogram (IVP) to exclude upper tract disease in the renal pelvis or ureters. Confirmation of bladder cancer requires looking in the bladder with a light (cystoscope) with multiple biopsies performed with a lighted instrument through the urethra to determine if the tumour is non-invasive (i.e., papillary or CIS) or invasive. Random biopsies of the bladder and prostatic urethra help to define field cancerization and field effect changes. Patients with  non-invasive  disease  require  close  monitoring,  as  they are at risk of subsequent recurrences, although stage and grade progression are uncommon. Patients who present with bladder cancer that is already high-grade or invasive into the lamina propria are at equally high risk of recurrence but stage progression is much more likely. Thus, they usually receive intravesical instillation of immuno- or chemotherapeutic agents following transurethral resection. Patients with tumours invading the muscularis propria or beyond are much more likely to have metastasis already and can rarely be managed by conservative means. However, even when treated by total cystectomy (the standard therapy for muscle-invading bladder cancer), 20 to 60% eventually succumb to their disease, almost always due to metastasis. When regional or distal metastasis is present at diagnosis, the 5-year survival rates drop to 35 and 9%, respectively, despite aggressive treatment. Systemic chemotherapy for metastatic bladder cancer is improving with complete response rates reported at 30%. Recent studies suggest chemotherapy prior to cystectomy may improve survival in selected patients.

Bladder cancer staging is predictive of the biological potential for progression, metastasis, or recurrence in 70% of the cases. Staging of bladder cancer usually requires CAT scan to rule out liver metastasis, radioisotope bone scan to exclude spread to the bone, and chest x ray or CAT scan to exclude lung metastasis. A search continues for biomarkers in the tumour and the bladder cancer field that will predict which tumours will metastasize or recur. The accessibility of exfoliated bladder cells in voided specimens shows promise for using biomarkers for monitoring recurrence and for cancer prevention.

Back

Pancreatic cancer (ICD-9 157; ICD-10 C25), a highly fatal malignancy, ranks amongst the 15 most common cancers globally but belongs to the ten most common cancers in the populations of developed countries, accounting for 2 to 3% of all new cases of cancer (IARC 1993). An estimated 185,000 new cases of pancreatic cancer occurred globally in 1985 (Parkin, Pisani and Ferlay 1993). The incidence rates of pancreatic cancer have been increasing in developed countries. In Europe, the increase has levelled off, except in the UK and some Nordic countries (Fernandez et al. 1994). The incidence and mortality rates rise steeply with advancing age between 30 and 70 years. The age-adjusted male/female ratio of new cases of pancreatic cancer is 1.6/1 in developed countries but only 1.1/1 in developing countries.

High annual incidence rates of pancreatic cancer (up to 30/100,000 for men; 20/100,000 for women) in the period 1960-85, have been recorded for New Zealand Maoris, Hawaiians, and in Black populations in the US. Regionally, the highest age-adjusted rates in 1985 (over 7/100,000 for men and 4/100,000 in women) were reported for both genders in Japan, North America, Australia, New Zealand, and Northern, Western and Eastern Europe. The lowest rates (up to 2/100,000 for both men and women) were reported in the regions of West and Middle Africa, South-eastern Asia, Melanesia, and in temperate South America (IARC 1992; Parkin, Pisani and Ferlay 1993).

Comparisons between populations in time and space are subject to several cautions and interpretation difficulties because of variations in diagnostic conventions and technologies (Mack 1982).

The vast majority of pancreatic cancers occur in the exocrine pancreas. The major symptoms are abdominal and back pain and weight loss. Further symptoms include anorexia, diabetes and obstructive jaundice. Symptomatic patients are subjected to procedures such as a series of blood and urine tests, ultrasound, computerized tomography, cytological examination and pancreatoscopy. Most patients have metastases at diagnosis, which makes their prognosis bleak.

Only 15% of patients with pancreatic cancer are operable. Local recurrence and distant metastases occur frequently after surgery. Irradiation therapy or chemotherapy do not bring about significant improvements in survival except when combined with surgery on localized carcinomas. Palliative procedures provide little benefit. Despite some diagnostic improvements, survival remains poor. During the period 1983-85, the five-year average survival in 11 European populations was 3% for men and 4% for women (IARC 1995). Very early detection and diagnosis or identification of high-risk individuals may improve the success of surgery. The efficacy of screening for pancreatic cancer has not been determined.

Mortality and incidence of pancreatic cancer do not reveal a consistent global pattern across socio-economic categories.

The dismal picture offered by diagnostic problems and treatment inefficacy is completed by the fact that the causes of pancreatic cancer are largely unknown, which effectively hampers the prevention of this fatal disease. The unique established cause of pancreatic cancer is tobacco smoking, which explains about 20-50% of the cases, depending on the smoking patterns of the population. It has been estimated that elimination of tobacco smoking would decrease the incidence of pancreatic cancer by about 30% worldwide (IARC 1990). Alcohol consumption and coffee drinking have been suspected as increasing the risk of pancreatic cancer. On closer scrutiny of the epidemiological data, however, coffee consumption appears unlikely to be causally connected to pancreatic cancer. For alcoholic beverages, the only causal link with pancreatic cancer is probably pancreatitis, a condition associated with heavy alcohol consumption. Pancreatitis is a rare but potent risk factor of pancreatic cancer. It is possible that some as yet unidentified dietary factors might account for a part of the aetiology of pancreatic cancer.

Workplace exposures may be causally associated with pancreatic cancer. Results of several epidemiological studies that have linked industries and jobs with an excess of pancreatic cancer are heterogeneous and inconsistent, and exposures shared by alleged high-risk jobs are hard to identify. The population aetiologic fraction for pancreatic cancer from occupational exposures in Montreal, Canada, has been estimated to lie between 0% (based on recognized carcinogens) and 26% (based on a multi-site case-control study in the Montreal area, Canada) (Siemiatycki et al. 1991).

No single occupational exposure has been confirmed to increase the risk of pancreatic cancer. Most of the occupational chemical agents that have been associated with an excess risk in epidemiological studies emerged in one study only, suggesting that many of the associations may be artefacts from confounding or chance. If no additional information, e.g., from animal bio-assays, is available, the distinction between spurious and causal associations presents formidable difficulties, given the general uncertainty about the causative agents involved in the development of pancreatic cancer. Agents associated with increased risk include aluminium, aromatic amines, asbestos, ashes and soot, brass dust, chromates, combustion products of coal, natural gas and wood, copper fumes, cotton dust, cleaning agents, grain dust, hydrogen fluoride, inorganic insulation dust, ionizing radiation, lead fumes, nickel compounds, nitrogen oxides, organic solvents and paint thinners, paints, pesticides, phenol-formaldehyde, plastic dust, polycyclic aromatic hydrocarbons, rayon fibres, stainless steel dust, sulphuric acid, synthetic adhesives, tin compounds and fumes, waxes and polishes, and zinc fumes (Kauppinen et al. 1995). Among these agents, only aluminium, ionizing radiation and unspecified pesticides have been associated with excess risk in more than one study.

Back

The predominant type of malignant tumour of the liver (ICD-9 155) is hepatocellular carcinoma (hepatoma; HCC), i.e., a malignant tumour of the liver cells. Cholangiocarcinomas are tumours of the intrahepatic bile ducts. They represent some 10% of liver cancers in the US but may account for up to 60% elsewhere, such as in north-eastern Thai populations (IARC 1990). Angiosarcomas of the liver are very rare and very aggressive tumours, occurring mostly in men. Hepatoblastomas, a rare embryonal cancer, occur in early life, and have little geographic or ethnic variation.

The prognosis for HCC depends on the size of the tumour and on the extent of cirrhosis, metastases, lymph node involvement, vascular invasion and presence/absence of a capsule. They tend to relapse after resection. Small HCCs are resectable, with a five-year survival of 40-70%. Liver transplantation results in about 20% survival after two years for patients with advanced HCC. For patients with less advanced HCC, the prognosis after transplantation is better. For hepatoblastomas, complete resection is possible in 50-70% of the children. Cure rates after resection range from 30-70%. Chemotherapy can be used both pre- and postoperatively. Liver transplantation may be indicated for unresectable hepatoblastomas.

Cholangiocarcinomas are multifocal in more than 40% of the patients at the time of diagnosis. Lymph node metastases occur in 30-50% of these cases. The response rates to chemotherapy vary widely, but usually are less than 20% successful. Surgical resection is possible in only a few patients. Radiation therapy has been used as the primary treatment or adjuvant therapy, and may improve survival in patients who have not undergone a complete resection. Five-year survival rates are less than 20%. Angiosarcoma patients usually present distant metastases. Resection, radiation therapy, chemotherapy and liver transplantation are, in most cases, unsuccessful. Most patients die within six months of diagnosis (Lotze, Flickinger and Carr 1993).

An estimated 315,000 new cases of liver cancer occurred globally in 1985, with a clear absolute and relative preponderance in populations of developing countries, except in Latin America (IARC 1994a; Parkin, Pisani and Ferlay 1993). The average annual incidence of liver cancer shows considerable variation across cancer registries worldwide. During the 1980s, average annual incidence ranged from 0.8 in men and 0.2 in women in Maastricht, The Netherlands, to 90.0 in men and 38.3 in women in Khon Kaen, Thailand, per 100,000 of population, standardized to the standard world population. China, Japan, East Asia, and Africa represented high rates, while Latin and North American, European, and Oceanian rates were lower, except for New Zealand Maoris (IARC 1992). The geographic distribution of liver cancer is correlated with the distribution of the prevalence of chronic carriers of hepatitis B surface antigen and also with the distribution of local levels of aflatoxin contamination of foodstuffs (IARC 1990). Male-to-female ratios in incidence are usually between 1 and 3, but may be higher in high-risk populations.

Statistics on the mortality and incidence of liver cancer by social class indicate a tendency of excess risk to concentrate in the lower socio-economic strata, but this gradient is not observed in all populations.

The established risk factors for primary liver cancer in humans include aflatoxin-contaminated food, chronic infection with hepatitis B virus (IARC 1994b), chronic infection with hepatitis C virus (IARC 1994b), and heavy consumption of alcoholic beverages (IARC 1988). HBV is responsible for an estimated 50-90% of hepatocellular carcinoma incidence in high-risk populations, and for 1-10% in low-risk populations. Oral contraceptives are a further suspected factor. The evidence implicating tobacco smoking in the aetiology of liver cancer is insufficient (Higginson, Muir and Munoz 1992).

The substantial geographical variation in the incidence of liver cancer suggests that a high proportion of liver cancers might be preventable. The preventive measures include HBV vaccination (estimated potential theoretical reduction in incidence is roughly 70% in endemic areas), reduction of contamination of food by mycotoxins (40% reduction in endemic areas), improved methods of harvesting, dry storing of crops, and reduction of consumption of alcoholic beverages (15% reduction in Western countries; IARC 1990).

Liver cancer excesses have been reported in a number of occupational and industrial groups in different countries. Some of the positive associations are readily explained by workplace exposures such as the increased risk of liver angiosarcoma in vinyl chloride workers (see below). For other high-risk jobs, such as metal work, construction painting, and animal feed processing, the connection with workplace exposures is not firmly established and is not found in all studies, but could well exist. For others, such as service workers, police officers, guards, and governmental workers, direct workplace carcinogens may not explain the excess. Cancer data for farmers do not provide many clues for occupational aetiologies in liver cancer. In a review of 13 studies involving 510 cases or deaths of liver cancer among farmers (Blair et al. 1992), a slight deficit (aggregated risk ratio 0.89; 95% confidence interval 0.81-0.97) was observed.

Some of the clues provided by industry- or job-specific epidemiological studies do suggest that occupational exposures may have a role in the induction of liver cancer. Minimization of certain occupational exposures therefore would be instrumental in the prevention of liver cancer in occupationally exposed populations. As a classical example, occupational exposure to vinyl chloride has been shown to cause angiosarcoma of the liver, a rare form of liver cancer (IARC 1987). As a result, vinyl chloride exposure has been regulated in a large number of countries. There is increasing evidence that chlorinated hydrocarbon solvents may cause liver cancer. Aflatoxins, chlorophenols, ethylene glycol, tin compounds, insecticides and some other agents have been associated with the risk of liver cancer in epidemiological studies. Numerous chemical agents occurring in occupational settings have caused liver cancer in animals and may therefore be suspected of being liver carcinogens in humans. Such agents include aflatoxins, aromatic amines, azo dyes, benzidine-based dyes, 1,2-dibromoethane, butadiene, carbon tetrachloride, chlorobenzenes, chloroform, chlorophenols, diethylhexyl phthalate, 1,2-dichloroethane, hydrazine, methylene chloride, N-nitrosoamines, a number of organochlorine pesticides, perchloroethylene, polychlorinated biphenyls and toxaphene.

Back

Leukaemias

Leukaemias constitute 3% of all cancers worldwide (Linet 1985). They are a group of malignancies of blood precursor cells, classified according to cell type of origin, degree of cellular differentiation, and clinical and epidemiological behaviour. The four common types are acute lymphocytic leukaemia (ALL), chronic lymphocytic leukaemia (CLL), acute myelocytic leukaemia (AML) and chronic myelocytic leukaemia (CML). ALL develops rapidly, is the most common form of leukaemia in childhood and originates in the white blood corpuscles in the lymph nodes. CLL arises in bone marrow lymphocytes, develops very slowly and is more common in aged persons. AML is the common form of acute leukaemia in adults. Rare types of acute leukaemia include monocytic, basophilic, eosinophilic, plasma-, erythro- and hairy-cell leukaemias. These rarer forms of acute leukaemia are sometimes lumped together under the heading acute non-lymphocytic leukaemia (ANLL), due in part to the belief that they arise from a common stem cell. Most cases of CML are characterized by a specific chromosomal abnormality, the Philadelphia chromosome. The eventual outcome of CML often is leukaemic transformation to AML. Transformation to AML also can occur in polycythaemia vera and essential thrombocythaemia, neoplastic disorders with elevated red cell or platelet levels, as well as myelofibrosis and myeloid dysplasia. This has led to characterizing these disorders as related myeloproliferative diseases.

The clinical picture varies according to the type of leukaemia. Most patients suffer from fatigue and malaise. Haematological count anomalies and atypical cells are suggestive of leukaemia and indicate a bone marrow examination. Anaemia, thrombocytopenia, neutropenia, elevated leucocyte count and elevated number of blast cells are typical signs of acute leukaemia.

Incidence: The annual overall age-adjusted incidence of leukaemias varies between 2 and 12 per 100,000 in men and between 1 and 11 per 100,000 in women in different populations. High figures are encountered in North American, western European and Israeli populations, while low ones are reported for Asian and African populations. The incidence varies according to age and to type of leukaemia. There is a marked increase in the incidence of leukaemia with age, and there is also a childhood peak which occurs around two to four years of age. Different leukaemia subgroups display different age patterns. CLL is about twice as frequent in men as in women. Incidence and mortality figures of adult leukaemias have tended to stay relatively stable over the past few decades.

Risk factors: Familial factors in the development of leukaemia have been suggested, but the evidence for this is inconclusive. Certain immunological conditions, some of which are hereditary, appear to predispose to leukaemia. Down’s syndrome is predictive of acute leukaemia. Two oncogenic retroviruses (human T-cell leukaemia virus-I, human T-lymphotropic virus-II) have been identified as being related to the development of leukaemias. These viruses are thought to be early-stage carcinogens and as such are insufficient causes of leukaemia (Keating, Estey and Kantarjian 1993).

Ionizing radiation and benzene exposure are established environmental and occupational causes of leukaemias. The incidence of CLL, however, has not been associated with exposure to radiation. Radiation and benzene-induced leukaemias are recognized as occupational diseases in a number of countries.

Much less consistently, leukaemia excesses have been reported for the following groups of workers: drivers; electricians; telephone linepersons and electronic engineers; farmers; flour millers; gardeners; mechanics, welders and metal workers; textile workers; paper-mill workers; and workers in the petroleum industry and distribution of petroleum products. Some particular agents in the occupational environment have been consistently associated with increased risk of leukaemia. These agents include butadiene, electromagnetic fields, engine exhaust, ethylene oxide, insecticides and herbicides, machining fluids, organic solvents, petroleum products (including gasoline), styrene and unidentified viruses. Paternal and maternal exposures to these agents prior to conception have been suggested to increase the leukaemia risk in the offspring, but the evidence at this time is insufficient to establish such exposure as causative.

Treatment and prevention: Up to 75% of male cases of leukaemia may be preventable (International Agency for Research on Cancer 1990). Avoidance of exposure to radiation and benzene will reduce the risk of leukaemias, but the potential reduction worldwide has not been estimated. Treatments of leukaemias include chemotherapy (single agents or combinations), bone marrow transplant and interferons. Bone marrow transplant in both ALL and AML is associated with a disease-free survival between 25 and 60%. The prognosis is poor for patients who do not achieve remission or who relapse. Of those who relapse, about 30% achieve a second remission. The major cause of failure to achieve remission is death from infection and haemorrhage. The survival of untreated acute leukaemia is 10% within 1 year of diagnosis. The median survival of patients with CLL before the initiation of treatment is 6 years. The length of survival depends on the stage of the disease when the diagnosis is initially made.

Leukaemias may occur following medical treatment with radiation and certain chemotherapeutic agents of another malignancy, such as Hodgkin’s disease, lymphomas, myelomas, and ovarian and breast carcinomas. Most of these secondary cases of leukaemia are acute non-lymphocytic leukaemias or myelodysplastic syndrome, which is a preleukaemic condition. Chromosomal abnormalities appear to be more readily observed in both treatment-related leukaemias and in leukaemias associated with radiation and benzene exposure. These acute leukaemias also share a tendency to resist therapy. Activation of the ras oncogene has been reported to occur more frequently in patients with AML who worked in professions deemed to be at high risk of exposure to leukaemogens (Taylor et al. 1992).

Malignant Lymphomas and Multiple Myeloma

Malignant lymphomas constitute a heterogeneous group of neoplasms primarily affecting lymphoid tissues and organs. Malignant lymphomas are divided into two major cellular types: Hodgkin’s disease (HD) (International Classification of Disease, ICD-9 201) and non-Hodgkin lymphomas (NHL) (ICD-9 200, 202). Multiple myeloma (MM) (ICD-9 203) represents a malignancy of plasma cells within the bone marrow and accounts usually for less than 1% of all malignancies (International Agency for Research on Cancer 1993). In 1985, malignant lymphomas and multiple myelomas ranked seventh among all cancers worldwide. They represented 4.2% of all estimated new cancer cases and amounted to 316,000 new cases (Parkin, Pisani and Ferlay 1993).

Mortality and incidence of malignant lymphomas do not reveal a consistent pattern across socio-economic categories worldwide. Children’s HD has a tendency to be more common in less developed nations, while relatively high rates have been observed in young adults in countries in more developed regions. In some countries, NHL seems to be in excess among people in higher socio-economic groups, while in other countries no such clear gradient has been observed.

Occupational exposures may increase the risk of malignant lymphomas, but the epidemiological evidence is still inconclusive. Asbestos, benzene, ionizing radiation, chlorinated hydrocarbon solvents, wood dust and chemicals in leather and rubber-tire manufacturing are examples of agents that have been associated with the risk of unspecified malignant lymphomas. NHL is more common among farmers. Further suspect occupational agents for HD, NHL and MM are mentioned below.

Hodgkin’s disease

Hodgkin’s disease is a malignant lymphoma characterized by the presence of multinucleated giant (Reed-Sternberg) cells. Lymph nodes in the mediastinum and neck are involved in about 90% of the cases, but the disease may occur in other sites as well. Histological subtypes of HD differ in their clinical and epidemiological behaviour. The Rye classification system includes four subtypes of HD: lymphocytic predominance, nodular sclerosis, mixed cellularity and lymphocytic depletion. The diagnosis of HD is made by biopsy and treatment is radiation therapy alone or in combination with chemotherapy.

The prognosis of HD patients depends on the stage of the disease at diagnosis. About 85 to 100% of patients without massive mediastinal involvement survive for about 8 years from the start of the treatment without further relapse. When there is massive mediastinal involvement, about 50% of the cases suffer a relapse. Radiation therapy and chemotherapy may involve various side effects, such as secondary acute myelocytic leukaemia discussed earlier.

The incidence of HD has not undergone major changes over time but for a few exceptions, such as the populations of the Nordic countries, in which the rates have declined (International Agency for Research on Cancer 1993).

Available data show that in the 1980s the populations of Costa Rica, Denmark and Finland had median annual incidence rates of HD of 2.5 per 100,000 in men and 1.5 per 100,000 in women (standardized to world population); these figures yielded a sex ratio of 1.7. The highest rates in males were recorded for populations in Italy, the United States, Switzerland and Ireland, while the highest female rates were in the United States and Cuba. Low incidence rates have been reported for Japan and China (International Agency for Research on Cancer 1992).

Viral infection has been suspected as involved in the aetiology of HD. Infectious mononucleosis, which is induced by the Epstein-Barr virus, a herpes virus, has been shown to be associated with increased risk of HD. Hodgkin’s disease may also cluster in families, and other time-space constellations of cases have been observed, but the evidence that there are common aetiological factors behind such clusters is weak.

The extent to which occupational factors can lead to increased risk for HD has not been established. There are three predominant suspect agents—organic solvents, phenoxy herbicides and wood dust—but the epidemiological evidence is limited and controversial.

Non-Hodgkin lymphoma

About 98% of the NHLs are lymphocytic lymphomas. At least four different classifications of lymphocytic lymphomas have been commonly used (Longo et al. 1993). In addition, an endemic malignancy, Burkitt’s lymphoma, is endemic in certain areas of tropical Africa and New Guinea.

Thirty to fifty per cent of NHLs are curable with chemotherapy and/or radiotherapy. Bone marrow transplants may be necessary.

Incidence: High annual incidences of NHL (over 12 per 100,000, standardized to world standard population) have been reported during the 1980s for the White population in the United States, particularly San Francisco and New York City, as well as in some Swiss cantons, in Canada, in Trieste (Italy) and Porto Alegre (Brazil, in men). The incidence of NHL is usually higher in men than in women, with the typical excess in men being 50 to 100% greater than in women. In Cuba, and in the White population of Bermuda, however, the incidence is slightly higher in women (International Agency for Research on Cancer 1992).

NHL incidence and mortality rates have been rising in a number of countries worldwide (International Agency for Research on Cancer 1993). By 1988, the average annual incidence in US White men increased by 152%. Some of the increase is due to changes in diagnostic practices of physicians and part due to an increase in immunosuppressive conditions which are induced by the human immunodeficiency virus (HIV, associated with AIDS), other viruses and immunosuppressive chemotherapy. These factors do not explain the entire increase, and a considerable proportion of residual increase may be explained by dietary habits, environmental exposures such as hair dyes, and possibly familial tendencies, as well as some rare factors (Hartge and Devesa 1992).

Occupational determinants have been suspected to play a role in the development of NHL. It is currently estimated that 10% of NHLs are thought to be related to occupational exposures in the United States (Hartge and Devesa 1992), but this percentage varies by time period and location. The occupational causes are not well established. Excess risk of NHL has been associated with electric power plant jobs, farming, grain handling, metal working, petroleum refining and woodworking, and has been found among chemists. Occupational exposures that have been associated with an increased NHL risk include ethylene oxide, chlorophenols, fertilizers, herbicides, insecticides, hair dyes, organic solvents and ionizing radiation. A number of positive findings for phenoxyacetic acid herbicide exposure have been reported (Morrison et al. 1992). Some of the herbicides involved were contaminated with 2,3,7,8-tetrachlorodibenzo-para-dioxin (TCDD). The epidemiological evidence for occupational aetiologies of NHL is still limited, however.

Multiple myeloma

Multiple myeloma (MM) involves predominantly bone (especially the skull), bone marrow and kidney. It represents malignant proliferation of B-lymphocyte-derived cells that synthesize and secrete immunoglobulins. The diagnosis is made using radiology, a test for the MM-specific Bence-Jones proteinuria, determination of abnormal plasma cells in the bone marrow, and immunoelectrophoresis. MM is treated with bone marrow transplantation, radiation therapy, conventional chemotherapy or polychemotherapy, and immunological therapy. Treated MM patients survive 28 to 43 months on the average (Ludwig and Kuhrer 1994).

The incidence of MM increases sharply with increasing age. High age-standardized annual incidence rates (5 to 10 per 100,000 in men and 4 to 6 per 100,000 in women) have been encountered in the United States Black populations, in Martinique and among the Maoris in New Zealand. Many Chinese, Indian, Japanese and Filipino populations have low rates (less than 10 per 100,000 person-years in men and less than 0.3 per 100,000 person-years in women) (International Agency for Research on Cancer 1992). The rate of multiple myeloma has been on the increase in Europe, Asia, Oceania and in both the Black and White United States populations since the 1960s, but the increase has tended to level off in a number of European populations (International Agency for Research on Cancer 1993).

Throughout the world there is an almost consistent excess among males in the incidence of MM. This excess is typically of the order of 30 to 80%.

Familial and other case clusterings of MM have been reported, but the evidence is inconclusive as to the causes of such clusterings. The excess incidence among the United States Black population as contrasted with the White population points towards the possibility of differential host susceptibility among population groups, which may be genetic. Chronic immunological disorders have on occasion been associated with the risk of MM. The data on social class distribution of MM are limited and unreliable for conclusions on any gradients.

Occupational factors: Epidemiological evidence of an elevated risk of MM in gasoline-exposed workers and refinery workers suggests a benzene aetiology (Infante 1993). An excess of multiple myeloma has repeatedly been observed in farmers and farm workers. Pesticides represent a suspect group of agents. The evidence for carcinogenicity is, however, insufficient for phenoxyacetic acid herbicides (Morrison et al. 1992). Dioxins are sometimes impurities in some phenoxyacetic acid herbicides. There is a reported significant excess of MM in women residing in a zone contaminated with 2,3,7,8-tetrachlorodibenzo-para-dioxin after an accident in a plant near Seveso, Italy (Bertazzi et al. 1993). The Seveso results were based on two cases which occurred during ten years of follow-up, and further observation is needed to confirm the association. Another possible explanation for the increased risk in farmers and farm workers is exposure to some viruses (Priester and Mason 1974).

Further suspect occupations and occupational agents that have been associated with increased risk of MM include painters, truck drivers, asbestos, engine exhaust, hair-colouring products, radiation, styrene, vinyl chloride and wood dust. The evidence for these occupations and agents remains inconclusive.

Back

Cancer is a common disease in all countries of the world. The probability that a person will develop cancer by the age of 70 years, given survival to that age, varies between about 10 and 40% in both sexes. On average, in developed countries, about one person in five will die from cancer. This proportion is about one in 15 in developing countries. In this article, environmental cancer is defined as cancer caused (or prevented) by non-genetic factors, including human behaviour, habits, lifestyle and external factors over which the individual has no control. A stricter definition of environmental cancer is sometimes used, comprising only the effect of factors such as air and water pollution, and industrial waste.

Geographical Variation

Variation between geographical areas in the rates of particular types of cancer can be much greater than that for cancer as a whole. Known variation in the incidence of the more common cancers is summarized in table 1. The incidence of nasopharyngeal carcinoma, for example, varies some 500-fold between South East Asia and Europe. This wide variation in frequency of the various cancers has led to the view that much of human cancer is caused by factors in the environment. In particular, it has been argued that the lowest rate of a cancer observed in any population is indicative of the minimum, possibly spontaneous, rate occurring in the absence of causative factors. Thus the difference between the rate of a cancer in a given population and the minimum rate observed in any population is an estimate of the rate of the cancer in the first population which is attributable to environmental factors. On this basis it has been estimated, very approximately, that some 80 to 90% of all human cancers are environmentally determined (International Agency for Research on Cancer 1990).

Table 1.  Variation between populations covered by cancer registration in the incidence of common cancers.¹

¹ Data from cancer registries included in IARC 1992. Only cancer sites with cumulative rate larger or equal to 2% in the high-incidence area are included. Rates refer to males except for breast, cervix uteri, corpus uteri and ovary cancers.
² Cumulative rate % between 0 and 74 years of age.
Source: International Agency for Research on Cancer 1992.

There are, of course, other explanations for geographical variation in cancer rates. Under-registration of cancer in some populations may exaggerate the range of variation, but certainly cannot explain differences of the size shown in table 1. Genetic factors also may be important. It has been observed, however, that when populations migrate along a gradient of cancer incidence they often acquire a rate of cancer which is intermediate between that of their home country and that of the host country. This suggests that a change in environment, without genetic change, has changed the cancer incidence. For example, when Japanese migrate to the United States their rates of colon and breast cancer, which are low in Japan, rise, and their rate of stomach cancer, which is high in Japan, falls, both tending more closely towards United States’ rates. These changes may be delayed until the first post-migration generation but they still occur without genetic change. For some cancers, change with migration does not occur. For example, the Southern Chinese retain their high rate of cancer of the nasopharynx wherever they live, thus suggesting that genetic factors, or some cultural habit which changes little with migration, are responsible for this disease.

Time Trends

Further evidence of the role of environmental factors in cancer incidence has come from the observation of time trends. The most dramatic and well-known change has been the rise in lung cancer rates in males and females in parallel with but occurring some 20 to 30 years after the adoption of cigarette use, which has been seen in many regions of the world; more recently in a few countries, such as the United States, there has been the suggestion of a fall in rates among males following a reduction in tobacco smoking. Less well understood are the substantial falls in incidence of cancers including those of the stomach, oesophagus and cervix which have paralleled economic development in many countries. It would be difficult to explain these falls, however, except in terms of reduction in exposure to causal factors in the environment or, perhaps, increasing exposure to protective factors—again environmental.

Main Environmental Carcinogenic Agents

The importance of environmental factors as causes of human cancer has been further demonstrated by epidemiological studies relating particular agents to particular cancers. The main agents which have been identified are summarized in table 10. This table does not contain the drugs for which a causal link with human cancer has been established (such as diethylstilboestrol and several alkylating agents) or suspected (such as cyclophosphamide) (see also Table 9). In the case of these agents, the risk of cancer has to be balanced with the benefits of the treatment. Similarly, Table 10 does not contain agents that occur primarily in the occupational setting, such as chromium, nickel and aromatic amines. For a detailed discussion of these agents see the previous article “Occupational Carcinogens.” The relative importance of the agents listed in table 8 varies widely, depending on the potency of the agent and the number of people involved. The evidence of carcinogenicity of several environmental agents has been evaluated within the IARC Monographs programme (International Agency for Research on Cancer 1995) (see again “Occupational Carcinogens” for a discussion of the Monographs programme); table 10 is based mainly on the IARC Monograph evaluations. The most important agents among those listed in table 10 are those to which a substantial proportion of the population is exposed in relatively large amounts. They include particularly: ultraviolet (solar) radiation; tobacco smoking; alcohol drinking; betel quid chewing; hepatitis B; hepatitis C and human papilloma viruses; aflatoxins; possibly dietary fat, and dietary fiber and vitamin A and C deficiency; reproductive delay; and asbestos.

Attempts have been made to estimate numerically the relative contributions of these factors to the 80 or 90% of cancers which might be attributed to environmental factors. The pattern varies, of course, from population to population according to differences in exposures and possibly in the genetic susceptibility to various cancers. In many industrialized countries, however, tobacco smoking and dietary factors are likely to be responsible each for roughly one-third of environmentally determined cancers (Doll and Peto 1981); while in developing countries the role of biological agents is likely to be large and that of tobacco relatively small (but increasing, following the recent increase in the consumption of tobacco in these populations).

Interactions between Carcinogens

An additional aspect to consider is the presence of interactions between carcinogens. Thus for example, in the case of alcohol and tobacco, and cancer of the oesophagus, it has been shown that an increasing consumption of alcohol multiplies manyfold the rate of cancer produced by a given level of tobacco consumption. Alcohol by itself may facilitate transport of tobacco carcinogens, or others, into the cells of susceptible tissues. Multiplicative interaction may also be seen between initiating carcinogens, as between radon and its decay products and tobacco smoking in miners of uranium. Some environmental agents may act by promoting cancers which have been initiated by another agent—this is the most likely mechanism for an effect of dietary fat on the development of breast cancer (probably through increased production of the hormones which stimulate the breast). The reverse may also occur, as, for example, in the case of vitamin A, which probably has an anti-promoting effect on lung and possibly other cancers initiated by tobacco. Similar interactions may also occur between environmental and constitutional factors. In particular, genetic polymorphism to enzymes implicated in the metabolism of carcinogenic agents or DNA repair is probably an important requirement of individual susceptibility to the effect of environmental carcinogens.

The significance of interactions between carcinogens, from the point of view of cancer control, is that withdrawal of exposure to one of two (or more) interacting factors may give rise to a greater reduction in cancer incidence than would be predicted from consideration of the effect of the agent when acting alone. Thus, for example, withdrawal of cigarettes may eliminate almost entirely the excess rate of lung cancer in asbestos workers (although rates of mesothelioma would be unaffected).

Implications for Prevention

The realization that environmental factors are responsible for a large proportion of human cancers has laid the foundation for primary prevention of cancer by modification of exposure to the factors identified. Such modification may comprise: removal of a single major carcinogen; reduction, as discussed above, in exposure to one of several interacting carcinogens; increasing exposure to protective agents; or combinations of these approaches. While some of this may be achieved by community-wide regulation of the environment through, for example, environmental legislation, the apparent importance of lifestyle factors suggests that much of primary prevention will remain the responsibility of individuals. Governments, however, may still create a climate in which individuals find it easier to take the right decision.

Back

The control of occupational carcinogens is based on the critical review of scientific investigations both in humans and in experimental systems. There are several review programmes being undertaken in different countries aimed at controlling occupational exposures which could be carcinogenic to humans. The criteria used in different programmes are not entirely consistent, leading occasionally to differences in the control of agents in different countries. For example, 4,4-methylene-bis-2-chloroaniline (MOCA) was classified as an occupational carcinogen in Denmark in 1976 and in the Netherlands in 1988, but only in 1992 has a notation “suspected human carcinogen” been introduced by the American Conference of Governmental Industrial Hygienists in the United States.

The International Agency for Research on Cancer (IARC) has established, within the framework of its Monographs programme, a set of criteria to evaluate the evidence of the carcinogenicity of specific agents. The IARC Monographs programme represents one of the most comprehensive efforts to review systematically and consistently cancer data, is highly regarded in the scientific community and serves as the basis for the information in this article. It also has an important impact on national and international occupational cancer control activities. The evaluation scheme is given in table 1.

Table 1.  Evaluation of evidence of carcinogenicity in the IARC Monographs programme.

1. The evidence for the induction of cancer in humans, which obviously plays an important role in the identification of human carcinogens is considered. Three types of epidemiological studies contribute to an assessment of carcinogenicity in humans: cohort studies, case-control studies and correlation (or ecological) studies. Case reports of cancer in humans may also be reviewed. The evidence relevant to carcinogenicity from studies in humans is classified into one of the following categories:

• Sufficient evidence of carcinogenicity: A causal relationship has been established between exposure to the agent, mixture or exposure circumstance and human cancer. That is, a positive relationship has been observed between the exposure and cancer in studies in which chance, bias and confounding could be ruled out with reasonable confidence.
• Limited evidence of carcinogenicity: A positive association has been observed between exposure to the agent, mixture or exposure circumstance and cancer for which a causal interpretation is considered to be credible, but chance, bias or confounding could not be ruled out with reasonable confidence.
• Inadequate evidence of carcinogenicity: The available studies are of insufficient quality, consistency or statistical power to permit a conclusion regarding the presence or absence of a causal association, or no data on cancer in humans are available.
• Evidence suggesting lack of carcinogenicity: There are several adequate studies covering the full range of levels of exposure that human beings are known to encounter, which are mutually consistent in not showing a positive association between exposure to the agent and the studied cancer at any observed level of exposure.

2. Studies in which experimental animals (mainly rodents) are exposed chronically to potential carcinogens and examined for evidence of cancer are reviewed and the degree of evidence of carcinogenicity is then classified into categories similar to those used for human data.

3. Data on biological effects in humans and experimental animals that are of particular relevance are reviewed. These may include toxicological, kinetic and metabolic considerations and evidence of DNA binding, persistence of DNA lesions or genetic damage in exposed humans. Toxicological information, such as that on cytotoxicity and regeneration, receptor binding and hormonal and immunological effects, and data on structure-activity relationship are used when considered relevant to the possible mechanism of the carcinogenic action of the agent.

4. The body of evidence is considered as a whole, in order to reach an overall evaluation of the carcinogenicity to humans of an agent, mixture or circumstance of exposure (see table 2).

Agents, mixtures and exposure circumstances are evaluated within the IARC Monographs if there is evidence of human exposure and data on carcinogenicity (either in humans or in experimental animals) (for IARC classification groups, see table 2).

Table 2.  IARC Monograph programme classification groups.

The agent, mixture or exposure circumstance is described according to the wording of one of the following categories:

Known and Suspected Occupational Carcinogens

At present, there are 22 chemicals, groups of chemicals or mixtures for which exposures are mostly occupational, without considering pesticides and drugs, which are established human carcinogens (table 3). While some agents such as asbestos, benzene and heavy metals are currently widely used in many countries, other agents have mainly an historical interest (e.g., mustard gas and 2-naphthylamine).

Table 3. Chemicals, groups of chemicals or mixtures for which exposures are mostly occupational (excluding pesticides and drugs).
Group 1-Chemicals carcinogenic to humans¹

¹ Evaluated in the IARC Monographs, Volumes 1-63 (1972-1995) (excluding pesticides and drugs).
² CAS Registry Nos. appear between parentheses.
³ This evaluation applies to the group of chemicals as a whole and not necessarily to all individual chemicals within the group.

An additional 20 agents are classified as probably carcinogenic to humans (Group 2A); they are listed in table 4, and include exposures that are currently prevalent in many countries, such as crystalline silica, formaldehyde and 1,3-butadiene. A large number of agents are classified as possible human carcinogens (Group 2B, table 5) - for example, acetaldehyde, dichloromethane and inorganic lead compounds. For the majority of these chemicals the evidence of carcinogenicity comes from studies in experimental animals.

Table 4. Chemicals, groups of chemicals or mixtures for which exposures are mostly occupational (excluding pesticides and drugs).
Group 2A—Probably carcinogenic to humans¹

¹ Evaluated in the IARC Monographs, Volumes 1-63 (1972-1995) (excluding pesticides and drugs).
² CAS Registry Nos. appear between parentheses.

Table 5. Chemicals, groups of chemicals or mixtures for which exposures are mostly occupational (excluding pesticides and drugs).
Group 2B—Possibly carcinogenic to humans¹

¹ Evaluated in the IARC Monographs, Volumes 1-63 (1972-1995) (excluding pesticides and drugs).
² CAS Registry Nos. appear between parentheses.

Occupational exposures may also occur during the production and use of some pesticides and drugs. Table 6 presents an evaluation of the carcinogenicity of pesticides; two of them, captafol and ethylene dibromide, are classified as probable human carcinogens, while a total of 20 others, including DDT, atrazine and chlorophenols, are classified as possible human carcinogens.

Table 6. Pesticides evaluated in IARC Monographs, Volumes 1-63(1972-1995)

¹ CAS Registry Nos. appear between parentheses.

Several drugs are human carcinogens (table 9): they are mainly alkylating agents and hormones; 12 more drugs, including chloramphenicol, cisplatine and phenacetin, are classified as probable human carcinogens (Group 2A). Occupational exposure to these known or suspected carcinogens, used mainly in chemotherapy, can occur in pharmacies and during their administration by nursing staff.

Table 7. Drugs evaluated in IARC Monographs, Volumes 1-63 (1972-1995).

¹ CAS Registry Nos. appear between parentheses.
² Suspected target organs are given in parentheses.

Several environmental agents are known or suspected causes of cancer in humans and are listed in table 8; although exposure to such agents is not primarily occupational, there are groups of individuals exposed to them because of their work: examples are uranium miners exposed to radon decay products, hospital workers exposed to hepatitis B virus, food processors exposed to aflatoxins from contaminated foods, outdoor workers exposed to ultraviolet radiation or diesel engine exhaust, and bar staff or waiters exposed to environmental tobacco smoke.

The IARC Monograph programme has covered most of the known or suspected causes of cancer; there are, however, some important groups of agents that have not been evaluated by IARC—namely, ionizing radiation and electrical and magnetic fields.

Table 8. Environmental agents/exposures known or suspected to cause cancer in humans.¹

¹ Agents and exposures, as well as medicines, occurring mainly in the occupational setting are excluded.

² Suspected target organs are given in parentheses.

³ IARC Monograph evaluation reported wherever available (1: human carcinogen; 2A: probable human carcinogen; 2B: possible human carcinogen); otherwise E: established carcinogen; S: suspected carcinogen.

⁴ Human exposure to polycyclic aromatic hydrocarbons occurs in mixtures, such as engine emissions, combustion fumes and soots. Several mixtures and individual hydrocarbons have been evaluated by IARC.

Industries and Occupations

Current understanding of the relationship between occupational exposures and cancer is far from complete; in fact, only 22 individual agents are established occupational carcinogens (table 9), and for many more experimental carcinogens no definitive evidence is available based on exposed workers. In many cases, there is considerable evidence of increased risks associated with particular industries and occupations, although no specific agents can be identified as aetiological factors. Table 10 present lists of industries and occupations associated with excess carcinogenic risks, together with the relevant cancer sites and the known (or suspected) causative agent(s).

Table 9. Industries, occupations and exposures recognized as presenting a carcinogenic risk.

Table 10.  Industries, occupations and exposures reported to present a cancer excess but for which the assessment of the carcinogenic risk is not definitive.

¹ PAH, polycyclic aromatic hydrocarbon.

² MOCA, 4,4’-methylene-bis-2-chloroaniline.

³ PCBs, polychlorinated biphenyls.

Table 9 presents industries, occupations and exposures in which the presence of a carcinogenic risk is considered to be established, whereas Table 10 shows industrial processes, occupations and exposures for which an excess cancer risk has been reported but evidence is not considered to be definitive. Also included in table 10 are some occupations and industries already listed in table 9, for which there is inconclusive evidence of association with cancers other than those mentioned in table 9. For example, the asbestos production industry is included in table 9 in relation to lung cancer and pleural and peritoneal mesothelioma, whereas the same industry is included in table 10 in relation to gastrointestinal neoplasms. A number of industries and occupations listed intables 9 and 10 have also been evaluated under the IARC Monographs programme. For example, “occupational exposure to strong inorganic acid mist containing sulphuric acid” was classified in Group 1 (carcinogenic to humans).

Constructing and interpreting such lists of chemical or physical carcinogenic agents and associating them with specific occupations and industries is complicated by a number of factors: (1) information on industrial processes and exposures is frequently poor, not allowing a complete evaluation of the importance of specific carcinogenic exposures in different occupations or industries; (2) exposures to well-known carcinogenic exposures, such as vinyl chloride and benzene, occur at different intensities in different occupational situations; (3) changes in exposure occur over time in a given occupational situation, either because identified carcinogenic agents are substituted by other agents or (more frequently) because new industrial processes or materials are introduced; (4) any list of occupational exposures can refer only to the relatively small number of chemical exposures which have been investigated with respect to the presence of a carcinogenic risk.

All of the above issues emphasize the most critical limitation of a classification of this type, and in particular its generalization to all areas of the world: the presence of a carcinogen in an occupational situation does not necessarily mean that workers are exposed to it and, in contrast, the absence of identified carcinogens does not exclude the presence of yet unidentified causes of cancer.

A particular problem in developing countries is that much of the industrial activity is fragmented and takes place in local settings. These small industries are often characterized by old machinery, unsafe buildings, employees with limited training and education, and employers with limited financial resources. Protective clothing, respirators, gloves and other safety equipment are seldom available or used. The small companies tend to be geographically scattered and inaccessible to inspections by health and safety enforcement agencies.

Back

Magnitude of the Problem

The first clear-cut evidence of cancer causation involved an occupational carcinogen (Checkoway, Pearce and Crawford-Brown 1989). Pott (1775) identified soot as the cause of scrotal cancer in London chimney-sweeps, and graphically described the abysmal working conditions, which involved children climbing up narrow chimneys that were still hot. Despite this evidence, reports of the need to prevent fires in chimneys were used to delay legislation on child labour in this industry until 1840 (Waldron 1983). An experimental model of soot carcinogenesis was first demonstrated in the 1920s (Decoufle 1982), 150 years after the original epidemiological observation.

In subsequent years, a number of other occupational causes of cancer have been demonstrated through epidemiological studies (although the association with cancer has usually first been noted by occupational physicians or by workers). These include arsenic, asbestos, benzene, cadmium, chromium, nickel and vinyl chloride. Such occupational carcinogens are very important in public health terms because of the potential for prevention through regulation and improvements in industrial hygiene practices (Pearce and Matos 1994). In most instances, these are hazards which markedly increase the relative risk of a particular type or types of cancer. It is possible that other occupational carcinogens remain undetected because they involve only a small increase in risk or because they simply have not been studied (Doll and Peto 1981). Some key facts about occupational cancer are given in table 1.

Table 1. Occupational cancer: Key facts.

• Some 20 agents and mixtures are established occupational carcinogens; a similar number of chemicals are highly suspected occupational carcinogens.
• In industrialized countries, occupation is causally linked to 2 to 8% of all cancers; among exposed workers, however, this proportion is higher.
• No reliable estimates are available on either the burden of occupational cancer or the extent of workplace exposure to carcinogens in developing countries.
• The relatively low overall burden of occupational cancer in industrialized countries is the result of strict regulations on several known carcinogens; exposure to other known or highly suspected agents, however, is still allowed.
• Although several occupational cancers are listed as occupational diseases in many countries, a very small fraction of cases is actually recognized and compensated.
• Occupational cancer is-to a very large extent-a preventable disease.

Occupational causes of cancer have received considerable emphasis in epidemiological studies in the past. However, there has been much controversy regarding the proportion of cancers which are attributable to occupational exposures, with estimates ranging from 4 to 40% (Higginson 1969; Higginson and Muir 1976; Wynder and Gori 1977; Higginson and Muir 1979; Doll and Peto 1981; Hogan and Hoel 1981; Vineis and Simonato 1991; Aitio and Kauppinen 1991). The attributable cancer risk is the total cancer experience in a population that would not have occurred if the effects associated with the occupational exposures of concern were absent. It may be estimated for the exposed population, as well as for a broader population. A summary of existing estimates is shown in table 2. Universal application of the International Classification of Diseases is what makes such tabulations possible (see box).

Table 2.  Estimated proportions of cancer (PAR) attributable to occupations in selected studies.

The International Classification of Diseases

Human diseases are classified according to the International Classification of Diseases (ICD), a system that was started in 1893 and is regularly updated under the coordination of the World Health Organization. The ICD is used in almost all countries for tasks such as death certification, cancer registration and hospital discharge diagnosis. The Tenth Revision (ICD-10), which was approved in 1989 (World Health Organization 1992), differs considerably from the previous three revisions, which are similar to each other and have been in use since the 1950s. It is therefore likely that the Ninth Revision (ICD-9, World Health Organization 1978), or even earlier revisions, will still be used in many countries during the coming years.

The large variability in the estimates arises from the differences in the data sets used and the assumptions applied. Most of the published estimates on the fraction of cancers attributed to occupational risk factors are based on rather simplified assumptions. Furthermore, although cancer is relatively less common in developing countries due to the younger age structure (Pisani and Parkin 1994), the proportion of cancers due to occupation may be higher in developing countries due to the relatively high exposures which are encountered (Kogevinas, Boffetta and Pearce 1994).

The most generally accepted estimates of cancers attributable to occupations are those presented in a detailed review on the causes of cancer in the population of the United States in 1980 (Doll and Peto 1981). Doll and Peto concluded that about 4% of all the deaths due to cancer may be caused by occupational carcinogens within “acceptable limits” (i.e., still plausible in view of all the evidence at hand) of 2 and 8%. These estimates being proportions, they are dependent on how causes other than occupational exposures contribute to produce cancers. For example, the proportion would be higher in a population of lifetime non-smokers (such as the Seventh-Day Adventists) and lower in a population in which, say, 90% are smokers. Also the estimates do not apply uniformly to both sexes or to different social classes. Furthermore, if one considers not the whole population (to which the estimates refer), but the segments of the adult population in which exposure to occupational carcinogens almost exclusively occurs (manual workers in mining, agriculture and industry, broadly taken, who in the United States numbered 31 million out of a population, aged 20 and over, of 158 million in the late 1980s), the proportion of 4% in the overall population would increase to about 20% among those exposed.

Vineis and Simonato (1991) provided estimates on the number of cases of lung and bladder cancer attributable to occupation. Their estimates were derived from a detailed review of case-control studies, and demonstrate that in specific populations located in industrial areas, the proportion of lung cancer or bladder cancer from occupational exposures may be as high as 40% (these estimates being dependent not only on the local prevailing exposures, but also to some extent on the method of defining and assessing exposure).

Mechanisms and Theories of Carcinogenesis

Studies of occupational cancer are complicated because there are no “complete” carcinogens; that is, occupational exposures increase the risk of developing cancer, but this future development of cancer is by no means certain. Furthermore, it may take 20 to 30 years (and at least five years) between an occupational exposure and the subsequent induction of cancer; it may also take several more years for cancer to become clinically detectable and for death to occur (Moolgavkar et al. 1993). This situation, which also applies to non-occupational carcinogens, is consistent with current theories of cancer causation.

Several mathematical models of cancer causation have been proposed (e.g., Armitage and Doll 1961), but the model which is simplest and most consistent with current biological knowledge is that of Moolgavkar (1978). This assumes that a healthy stem cell occasionally mutates (initiation); if a particular exposure encourages the proliferation of intermediate cells (promotion) then it becomes more likely that at least one cell will undergo one or more further mutations producing a malignant cancer (progression) (Ennever 1993).

Thus, occupational exposures can increase the risk of developing cancer either by causing mutations in DNA or by various “epigenetic” mechanisms of promotion (those not involving damage to DNA), including increased cell proliferation. Most occupational carcinogens which have been discovered to date are mutagens, and therefore appear to be cancer initiators. This explains the long “latency” period which is required for further mutations to occur; in many instances the necessary further mutations may never occur, and cancer may never develop.

In recent years, there has been increasing interest in occupational exposures (e.g., benzene, arsenic, phenoxy herbicides) which do not appear to be mutagens, but which may act as promoters. Promotion may occur relatively late in the carcinogenic process, and the latency period for promoters may therefore be shorter than for initiators. However, the epidemiological evidence for cancer promotion remains very limited at this time (Frumkin and Levy 1988).

Transfer of Hazards

A major concern in recent decades has been the problem of the transfer of hazardous industries to the developing world (Jeyaratnam 1994). Such transfers have occurred in part due to the stringent regulation of carcinogens and increasing labour costs in the industrialized world, and in part from low wages, unemployment and the push for industrialization in the developing world. For example, Canada now exports about half of its asbestos to the developing world, and a number of asbestos-based industries have been transferred to developing countries such as Brazil, India, Pakistan, Indonesia and South Korea (Jeyaratnam 1994). These problems are further compounded by the magnitude of the informal sector, the large numbers of workers who have little support from unions and other worker organizations, the insecure status of workers, the lack of legislative protection and/or the poor enforcement of such protection, the decreasing national control over resources, and the impact of the third world debt and associated structural adjustment programmes (Pearce et al. 1994).

As a result, it cannot be said that the problem of occupational cancer has been reduced in recent years, since in many instances the exposure has simply been transferred from the industrialized to the developing world. In some instances, the total occupational exposure has increased. Nevertheless, the recent history of occupational cancer prevention in industrialized countries has shown that it is possible to use substitutes for carcinogenic compounds in industrial processes without leading industry to ruin, and similar successes would be possible in developing countries if adequate regulation and control of occupational carcinogens were in place.

Prevention of Occupational Cancer

Swerdlow (1990) outlined a series of options for the prevention of exposure to occupational causes of cancer. The most successful form of prevention is to avoid the use of recognized human carcinogens in the workplace. This has rarely been an option in industrialized countries, since most occupational carcinogens have been identified by epidemiological studies of populations that were already occupationally exposed. However, at least in theory, developing countries could learn from the experience of industrialized countries and prevent the introduction of chemicals and production processes that have been found to be hazardous to the health of workers.

The next best option for avoiding exposure to established carcinogens is their removal once their carcinogenicity has been established or suspected. Examples include the closure of plants making the bladder carcinogens 2-naphthylamine and benzidine in the United Kingdom (Anon 1965), termination of British gas manufacture involving coal carbonization, closure of Japanese and British mustard gas factories after the end of the Second World War (Swerdlow 1990) and gradual elimination of the use of benzene in the shoe industry in Istanbul (Aksoy 1985).

In many instances, however, complete removal of a carcinogen (without closing down the industry) is either not possible (because alternative agents are not available) or is judged politically or economically unacceptable. Exposure levels must therefore be reduced by changing production processes and through industrial hygiene practices. For example, exposures to recognized carcinogens such as asbestos, nickel, arsenic, benzene, pesticides and ionizing radiation have been progressively reduced in industrialized countries in recent years (Pearce and Matos 1994).

A related approach is to reduce or eliminate the activities that involve the heaviest exposures. For example, after an 1840 act was passed in England and Wales prohibiting chimney-sweeps from being sent up chimneys, the number of cases of scrotal cancer decreased (Waldron 1983). Exposure also can be minimized through the use of protective equipment, such as masks and protective clothing, or by imposing more stringent industrial hygiene measures.

An effective overall strategy in the control and prevention of exposure to occupational carcinogens generally involves a combination of approaches. One successful example is a Finnish registry which has as its objectives to increase awareness about carcinogens, to evaluate exposure at individual workplaces and to stimulate preventive measures (Kerva and Partanen 1981). It contains information on both workplaces and exposed workers, and all employers are required to maintain and update their files and to supply information to the registry. The system appears to have been at least partially successful in decreasing carcinogenic exposures in the workplace (Ahlo, Kauppinen and Sundquist 1988).

Back