Abstract
Objectives
To investigate the contribution of occupational exposure to asbestos and man‐made vitreous fibres (MMVF) to lung cancer in high‐risk populations in Europe.
Methods
A multicentre case‐control study was conducted in six Central and Eastern European countries and the UK, during the period 1998–2002. Comprehensive occupational and sociodemographic information was collected from 2205 newly diagnosed male lung cancer cases and 2305 frequency matched controls. Odds ratios (OR) of lung cancer were calculated after adjusting for other relevant occupational exposures and tobacco smoking.
Results
The OR for asbestos exposure was 0.92 (95% CI 0.73 to 1.15) in Central and Eastern Europe and 1.85 (95% CI 1.07 to 3.21) in the UK. Similar ORs were found for exposure to amphibole asbestos. The OR for MMVF exposure was 1.23 (95% CI 0.88 to 1.71) with no evidence of heterogeneity by country. No synergistic effect either between asbestos and MMVF or between any of them and smoking was found.
Conclusion
In this large community‐based study occupational exposure to asbestos and MMVF does not appear to contribute to the lung cancer burden in men in Central and Eastern Europe. In contrast, in the UK the authors found an increased risk of lung cancer following exposure to asbestos. Differences in fibre type and circumstances of exposure may explain these results.
Lung cancer is the most common tumour worldwide, apart from non‐melanoma skin cancer, and remains one of the most fatal cancer types.1 The causal relation between cigarette smoking and lung cancer is well established and in most populations over 90% of all male patients develop their disease following tobacco smoking.2,3,4 A change in smoking habits in recent decades has resulted in a decline in lung cancer mortality in men in most European countries including Central and Eastern Europe and the UK.5 Nevertheless, Hungary and Poland are among the countries with the highest lung cancer mortality in the world.6 Environmental and occupational lung carcinogens also contribute to the global burden of lung cancer, but the quantification of their role is complex.
Asbestos has been recognised as a human carcinogen for many years.7,8 Consequently, mining and use of asbestos have been greatly reduced since 1990 and banned in most European countries. Even so, high levels of exposure to asbestos are still found in Central Europe and the former Soviet Union.9 Due to their persistence in the environment, asbestos fibres also remain in many work sites as well as in neighbouring areas where asbestos was used.10,11 All commercial brands of asbestos, regardless of fibre type, are considered carcinogenic, but amphiboles (amosite, anthophyllite, crocidolite and tremolite) show greater carcinogenic potency than chrysotile.12,13 Chrysotile is by far the most common type worldwide and the only asbestos type mined in Europe today, mainly in Russia. In the 1970s amphibole asbestos from South Africa and other countries was used in over 50 countries, including UK, Italy, the Netherlands, France and the US.14
Asbestos has to some extent been replaced by man‐made vitreous fibres (MMVF) in the insulation and construction industry. Therefore, the carcinogenic role of MMVF in humans has been widely studied in recent years.15,16,17 In 2002 glass wool, rock wool and slag wool were categorised by the International Agency for Research on Cancer (IARC) in Group 3 (unclassifiable as carcinogens) and ceramic fibres were classified in Group 2B (possible human carcinogens) following evidence in experimental animals.18
Both community‐based and industry‐based studies can contribute to our understanding of occupational carcinogenesis, but both have limitations. Valid and precise exposure information is seldom available in community‐based studies while industry‐based studies frequently cannot take into account individual smoking patterns and complete work histories of the subjects. In addition, the exposure assessment in industry‐based studies is often based on job titles and employment time only, which limits the possibility to control for confounders.19,20,21
The current study provides an opportunity to overcome the above mentioned shortcomings. By analysing data from a large, multicentre lung cancer case‐control study in Europe, with extensive estimations of past occupational exposures and lifestyle characteristics, we aim to elucidate the extent to which lung cancer burden in men in this region is driven by asbestos and man‐made vitreous fibres while controlling for potential confounders.
Methods
This multicentre case‐control study on lung cancer was conducted during the period 1998–2002 in seven European countries. Sixteen centres were included: Borsod, Heves, Szabolcs, Szolnok, Budapest (Hungary), Lodz, Warsaw (Poland), Banska Bystrica, Bratislava, Nitra (Slovakia), Brno, Olomouc, Prague (Czech Republic), Bucharest (Romania), Moscow (Russia), and Liverpool (UK). Approvals were obtained from local as well as from the IARC ethical review committees. The study population comprised all incident lung cancer cases (age <75 years) diagnosed in the participating centres. The response rate in the overall study was 84% among cases and 85% among controls. A small proportion of women was ever exposed to asbestos (2.2%) or synthetic fibres (0.9%), and was not further analysed. Thus, 2205 male lung cancer cases and 2305 controls who were frequency matched to cases on age (±3 years) and sex were included in the analysis. Population controls were selected from the electronic register of residents in Warsaw and from the general practitioner registry in Liverpool. In the other centres, controls were selected among patients admitted to the same hospitals as the cases or from general hospitals serving the same population. Patients suffering from smoking‐related conditions were not considered eligible. No single disease represented more than 10% of the diagnoses of controls. Detailed occupational history was obtained from each participant by a face‐to‐face interview conducted by a trained interviewer. The questionnaire covered lifestyle factors, smoking habits and all jobs held for more than one year. For 18 specific occupations or industries, specialised questionnaires were designed and used. Detailed information on study design and methodology has been presented previously.22 Assessment of exposure to 70 agents including chrysotile and amphibole asbestos, glass fibres, mineral wool and ceramic fibres was conducted by a group of local trained experts who were blinded to the case‐referent status of the study subjects. A variable “asbestos” was created encompassing all fibre types. In addition, experts specified if subjects were exposed to amphiboles and/or chrysotile when there was enough confidence regarding the specific circumstances of exposure. For synthetic fibres the categories were glass fibres, mineral wool fibres, or ceramic fibres, whereupon we created the variable “MMVF” encompassing glass fibres and mineral wool fibres. Based on data in the general and specialised (when applicable) questionnaires, the experts assessed each job held by a concerned subject and indicated their confidence in the presence of specific exposures expressed in the following categories: possible, probable or certain. Exposures were further assessed by frequency and intensity. Frequency of exposure was defined as the percentage of working time the person was assumed to be exposed to the agent, with the categories 1–5%, 5–30% and over 30% of the working time. Each year was considered to represent 2000 working hours. Regarding intensity of exposure the cut‐points used for asbestos and MMVF were <0.1 fibres/ml for low, 0.1–1 fibres/ml for medium, and >1 fibres/ml for high exposure.
Statistical analyses were done using SAS and STATA packages. Odds ratios (OR) of lung cancer and 95% confidence interval (CI) were calculated by applying unconditional logistic regression modelling, adjusted for age, centre, smoking (cumulative consumption), as well as for occupational exposures to silica, PAH, arsenic, cadmium and chromium. Odds ratios for asbestos were then also adjusted for MMVF and vice versa.
Odds ratios were calculated for ever exposure, defined as exposure for at least one year during the working life to the agent of interest. To model level of exposure to asbestos or MMVF, two indices were built: (1) duration of exposure and (2) cumulative exposure, based on the total number of hours effectively exposed, multiplied by the intensity level assessed for each exposed year. The frequency and intensity values were based on assigning mid‐interval weighting for each interval, while duration was expressed in years. Analysis were repeated with a 20‐ and a 30‐year lag, in which all jobs held within 20 (or 30) years before year of interview were considered not exposed. ISCO‐68 was used for the classification of occupations.23
Results
There were 2205 male cases and 2305 controls recruited from the seven participating countries. The mean age of the cases was 60.8 (SD 8.8) years and for the matched controls 60.6 (SD 8.5) years. Table 1 shows the distribution of the study group by country, age group and tobacco smoking. As expected, we found a marked difference between cases and controls with respect to smoking habits.
Table 1 Description of the study population, men only.
| Cases | Controls | |||
|---|---|---|---|---|
| n = 2205 | % | n = 2305 | % | |
| Country | ||||
| Czech Republic | 235 | 10.7 | 323 | 14.0 |
| Hungary | 315 | 14.3 | 266 | 11.5 |
| Poland | 557 | 25.3 | 581 | 25.2 |
| Romania | 141 | 6.4 | 169 | 7.3 |
| Russia | 521 | 23.6 | 521 | 22.6 |
| Slovakia | 288 | 13.1 | 294 | 12.8 |
| UK | 148 | 6.7 | 151 | 6.6 |
| Age (years) | ||||
| <45 | 94 | 4.3 | 119 | 5.2 |
| 45–54 | 455 | 20.6 | 510 | 22.1 |
| 55–64 | 829 | 37.6 | 777 | 33.7 |
| 65–74 | 762 | 34.6 | 799 | 34.7 |
| >75 | 65 | 3.0 | 100 | 4.3 |
| Tobacco consumption | ||||
| Never smoker | 48 | 2.2 | 534 | 23.2 |
| Ex‐smoker | 494 | 22.4 | 842 | 36.5 |
| Current smoker <20 pack‐years | 132 | 5.6 | 196 | 8.5 |
| Current smoker 20–39 pack‐years | 758 | 34.4 | 450 | 19.3 |
| Current smoker >40 pack‐years | 773 | 35.4 | 283 | 12.5 |
Among cases 14.4% were classified as ever exposed to asbestos and in the control group the prevalence of ever exposure to asbestos was 13.1%. For MMVF (glass and mineral wool), these proportions were 5.2% and 3.9%, respectively. The proportion of exposure to asbestos among controls was 8.1% in Poland, 8.3% in Romania, 9.4% Russia, 9.8% Hungary, 15.6 in Slovakia, 19.8% in the Czech Republic and 37.1% in the UK. The proportion of ever exposed and highly exposed to asbestos and synthetic fibres was considerably higher in the UK, while high frequency of exposure was more often reported among subjects in Central and Eastern Europe (see table 2).
Table 2 Exposure to asbestos and MMVF among cases and controls in Central and Eastern Europe (CEE) and the UK respectively.
| Exposure status | Asbestos (all types) | Asbestos (chrysotile only) | Man‐made vitreous fibres (glass and mineral wool) | Ceramic fibres only* | ||||
|---|---|---|---|---|---|---|---|---|
| Cases (%) | Controls (%) | Cases (%) | Controls (%) | Cases (%) | Controls (%) | Cases (%) | Controls (%) | |
| Never exposed CEE† | 1824 (88.7) | 1908 (88.6) | 1824 (94.9) | 1908 (94.1) | 1954 (95.6) | 2070 (96.6) | 1954 (99.6) | 2070 (99.8) |
| Ever exposed CEE | 233 (11.3) | 246 (11.4) | 99 (5.1) | 120 (5.9) | 91 (4.4) | 73 (3.4) | 7 (0.4) | 5 (0.2) |
| Never exposed UK† | 64 (43.2) | 95 (62.9) | 64 (78.0) | 95 (88.8) | 124 (83.8) | 134 (89.3) | 124 (100) | 134 (100) |
| Ever exposed UK | 84 (56.8) | 56 (37.1) | 18 (22.0) | 12 (11.2) | 24 (16.2) | 16 (10.7) | 0 (0.0) | 0 (0.0) |
| Intensity of exposure CEE | ||||||||
| Low | 171 (73.4) | 194 (78.9) | 76 (76.8) | 98 (81.7) | 60 (65.9) | 54 (74.0) | 5 (71.4) | 3 (60.0) |
| Medium | 50 (21.5) | 42 (17.1) | 18 (18.2) | 15 (12.5) | 26 (28.6) | 15 (20.5) | 2 (28.6) | 2 (40.0) |
| High | 12 (5.2) | 10 (4.1) | 5 (5.1) | 7 (5.8) | 5 (5.5) | 4 (5.5) | ||
| Intensity of exposure UK | ||||||||
| Low | 49 (58.3) | 33 (58.9) | 12 (66.7) | 10 (83.3) | 13 (54.2) | 7 (43.8) | 0 (0.0) | 0 (0.0) |
| Medium | 22 (26.2) | 15 (26.8) | 4 (22.2) | 1 (8.3) | 8 (33.3) | 7 (43.8) | 0 (0.0) | 0 (0.0) |
| High | 13 (15.5) | 8 (14.3) | 2 (11.1) | 1 (8.3) | 3 (12.5) | 2 (12.5) | 0 (0.0) | 0 (0.0) |
| Frequency of exposure CEE | ||||||||
| 1–5% | 112 (48.1) | 124 (50.4) | 58 (58.6) | 64 (53.3) | 51 (56.0) | 42 (57.5) | 1 (14.3) | 1 (20.0) |
| 5–30% | 73 (31.3) | 74 (30.1) | 26 (26.3) | 31 (25.8) | 26 (28.6) | 20 (27.4) | 3 (42.9) | 3 (60.0) |
| >30% | 48 (20.6) | 48 (19.5) | 15 (15.2) | 25 (20.8) | 14 (15.4) | 11 (15.1) | 3 (42.9) | 1 (20.0) |
| Frequency of exposure UK | ||||||||
| 1–5% | 45 (53.6) | 41 (73.2) | 9 (50.0) | 10 (83.3) | 12 (50.0) | 8 (50.0) | 0 (0.0) | 0 (0.0) |
| 5–30% | 33 (39.3) | 12 (21.4) | 7 (38.9) | 2 (16.7) | 10 (41.7) | 7 (43.8) | 0 (0.0) | 0 (0.0) |
| >30% | 6 (7.1) | 3 (5.4) | 2 (11.1) | 0 (0.0) | 2 (8.3) | 1 (6.3) | 0 (0.0) | 0 (0.0) |
| Confidence in assessment CEE | ||||||||
| Possible | 30 (12.9) | 22 (8.9) | 11 (11.1) | 15 (12.5) | 8 (8.8) | 10 (13.7) | 0 (0.0) | 0 (0.0) |
| Probable | 80 (34.3) | 98 (39.8) | 37 (37.4) | 49 (40.8) | 29 (31.9) | 23 (31.5) | 1 (14.3) | 0 (0.0) |
| Certain | 123 (52.8) | 125 (50.8) | 50 (50.5) | 56 (46.7) | 54 (59.3) | 40 (54.8) | 6 (85.7) | 5 (100) |
| Confidence in assessment UK | ||||||||
| Possible | 0 (0.0) | 2 (3.6) | 0 (0.0) | 1 (8.3) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Probable | 24 (28.6) | 19 (33.9) | 5 (27.8) | 4 (33.3) | 5 (20.8) | 3 (18.8) | 0 (0.0) | 0 (0.0) |
| Certain | 60 (71.4) | 35 (62.5) | 13 (72.2) | 7 (58.3) | 19 (79.2) | 13 (81.3) | 0 (0.0) | 0 (0.0) |
*Excludes subjects exposed to mixed synthetic fibres.
†Never exposed to any type of asbestos or synthetic fibres, respectively.
Lung cancer risk following ever exposure to asbestos differed between countries (p = 0.01); the UK showed an elevated OR of 1.85 (95% CI 1.07 to 3.21) and Romania 2.08 (95% CI 0.82 to 5.27) while ORs for the rest of the countries ranged between 0.64–1.02. When testing the heterogeneity in Central and Eastern Europe there was no significant difference between the countries (p = 0.39). Thus, in the subsequent analysis on asbestos we decided to keep the countries from Central and Eastern Europe separated from the UK. With regard to MMVF, there was no significant difference between the countries, including the UK (p = 0.18).
The prevalence of asbestos exposure among all subjects in Central and Eastern Europe was 11.4%, while in the UK the corresponding proportion was 46.8%. In Central and Eastern Europe 2.9% had been exposed to amphiboles (along with chrysotile), 5.2% to chrysotile but not amphiboles and 3.3% to unknown type of asbestos. In the UK 36.8% had experienced exposure to amphiboles (along with chrysotile), and 10% to chrysotile but not amphiboles. No subjects were assigned “unknown” fibre type in the UK. No increased risk of lung cancer following exposure to asbestos was observed in Central and Eastern Europe (see table 3), nor was there evidence of a trend with intensity of exposure (p = 0.46), cumulative exposure (p = 0.60) or year of first exposure (p = 0.35). Similarly, analyses based on a lag of 20 and 30 years or restricted to individuals with the highest confidence level did not show any association.
Table 3 Asbestos exposure and lung cancer risk in Central and Eastern Europe and the UK respectively.
| Type of exposure | Exposure category | Central and Eastern Europe | UK | ||
|---|---|---|---|---|---|
| Cases/controls | OR* (95% CI) | Cases/controls | OR* (95% CI) | ||
| Asbestos | Never | 1824/1908 | 1.00 (reference) | 64/95 | 1.00 (reference) |
| Ever | 233/246 | 0.92 (0.73 to 1.15) | 84/56 | 1.85 (1.07 to 3.21) | |
| Amphiboles | 58/63 | 0.93 (0.60 to 1.45) | 66/44 | 1.91 (1.06 to 3.45) | |
| Chrysotile | 99/120† | 0.77 (0.56 to 1.05) | 18/12 | 1.53 (0.61 to 3.86) | |
| Asbestos intensity | Low | 171/194 | 0.88 (0.68 to 1.13) | 49/33 | 2.01 (1.08 to 3.76) |
| Medium | 50/42 | 1.08 (0.67 to 1.73) | 22/15 | 1.66 (0.71 to 3.85) | |
| High | 12/10 | 0.95 (0.36 to 2.47) | 13/8 | 1.53 (0.51 to 4.54) | |
| Test for linear trend, p value | 0.46 | 0.62 | |||
| Cumulative asbestosexposure (fibres/ml*h) | <41.99 | 64/62 | 1.05 (0.70 to 1.59) | 18/14 | 1.66 (0.72 to 3.85) |
| <125.98 | 55/63 | 0.87 (0.57 to 1.31) | 19/15 | 2.03 (0.87 to 4.76) | |
| <699.95 | 49/68 | 0.70 (0.45 to 1.09) | 20/7 | 2.95 (1.05 to 8.26) | |
| ⩾699.95 | 65/53 | 1.07 (0.70 to 1.63) | 27/20 | 1.49 (0.67 to 3.33) | |
| Test for linear trend, p value | 0.60 | 0.95 | |||
| Year of first asbestosexposure | –1960 | 99/89 | 1.05 (0.74 to 1.50) | 57/41 | 1.77 (0.97 to 3.24) |
| 1961–70 | 74/86 | 0.90 (0.62 to 1.31) | 18/13 | 1.48 (0.56 to 3.91) | |
| 1971–80 | 36/45 | 0.72 (0.44 to 1.18) | 7/2 | 4.51 (0.72 to 28.24) | |
| 1981–90 | 16/19 | 0.70 (0.33 to 1.47) | 2/0 | ||
| 1991– | 8/7 | 1.47 (0.47 to 4.60) | 0/0 | ||
| Test for linear trend, p value | 0.35 | 0.20 | |||
*Odds ratio adjusted for centre, age, tobacco smoking, silica, PAHs, MMVF, arsenic, chromium and cadmium.
†Excludes subjects exposed to unknown type of asbestos or combined exposure to amphiboles.
In contrast, an increased risk of lung cancer was found in the UK (OR 1.85, 95% CI 1.07 to 3.21), in particular in workers exposed to amphiboles (OR 1.91, 95% CI 1.06 to 3.45). However, we found no trend of increased risk with rising cumulative exposure (p = 0.95) and no trend with regard to latency (p = 0.20). Analysis including only subjects with the highest confidence level or applying a lag of 20 and 30 years gave similar results.
There was no detectable modification of the effect of asbestos between smokers and non‐smokers, in either Central and Eastern Europe (p = 0.37) or in the UK (p = 0.62).
Table 4 describes lung cancer risk following exposure to MMVF and ceramic fibres. Forty nine per cent of exposed subjects were exposed to only glass fibres, 24% to only mineral wool fibres and 27% to both glass and mineral wool fibres. Eight of the 115 exposed cases and 6 of the 89 exposed controls had experienced high levels of MMVF exposure. Twelve cases and 12 controls had been exposed to ceramic fibres, 12 of them (7 cases, 5 controls) to ceramic fibres but not MMVF and the remaining 12 (5 cases, 7 controls) to ceramic fibres and MMVF. None of the subjects was exposed to high levels of ceramic fibres.
Table 4 MMVF exposure and lung cancer risk in Central/Eastern Europe and the UK.
| Type of exposure | Exposure category | Central/Eastern Europe and the UK | |||
|---|---|---|---|---|---|
| Cases/controls | OR* | 95% CI | |||
| MMVF (GF, MF) | Never† | 2078/2204 | 1.00 | ||
| Ever | 115/89 | 1.23 | 0.88 to 1.71 | ||
| Ceramic fibres only‡ | 7/5 | 1.25 | 0.33 to 4.74 | ||
| Exposure intensity to MMVF | Low | 73/61 | 1.23 | 0.82 to 1.84 | |
| Medium | 34/22 | 1.28 | 0.70 to 2.34 | ||
| High | 8/6 | 1.02 | 0.31 to 3.33 | ||
| Test for linear trend, p value | 0.87 | ||||
| Duration of MMVF exposure (years) | <2 | 7/2 | 3.81 | 0.59 to 24.52 | |
| 2–5 | 29/24 | 1.02 | 0.55 to 1.87 | ||
| 6–10 | 17/9 | 1.80 | 0.72 to 4.52 | ||
| 11–20 | 26/16 | 2.02 | 0.98 to 4.15 | ||
| 21–30 | 21/20 | 0.93 | 0.47 to 1.84 | ||
| 30+ | 15/18 | 0.77 | 0.35 to 1.68 | ||
| Test for linear trend, p value | 0.10 | ||||
| Cumulative MMVF exposure (fibres/ml*h) | <56.99 | 34/23 | 1.60 | 0.85 to 3.00 | |
| <180.99 | 25/25 | 0.97 | 0.52 to 1.80 | ||
| <839.97 | 27/23 | 1.17 | 0.63 to 2.19 | ||
| ⩾839.97 | 29/18 | 1.26 | 0.64 to 2.47 | ||
| Test for linear trend, p value | 0.62 | ||||
| Year of first MMVFexposure | –1960 | 45/31 | 1.36 | 0.80 to 2.31 | |
| 1961–70 | 30/29 | 1.09 | 0.60 to 1.97 | ||
| 1971–80 | 25/21 | 1.05 | 0.54 to 2.03 | ||
| 1981–90 | 11/6 | 1.64 | 0.54 to 4.96 | ||
| 1991– | 4/2 | 1.57 | 0.25 to 9.88 | ||
| Test for linear trend, p value | 0.72 | ||||
*Odds ratio adjusted for centre, age, tobacco smoking, silica, PAHs, asbestos, arsenic, chromium and cadmium.
†Never exposed to any kind of synthetic fibres.
‡Excludes 12 subjects exposed to mixed synthetic fibres.
The OR for lung cancer among men ever exposed to MMVF was 1.23 (95% CI 0.88 to 1.71). The OR for lung cancer following exposure to ceramic fibres but not MMVF was similar 1.25 (95% CI 0.33 to 4.74) but the number of exposed subjects was very small. With regard to intensity, latency, duration and cumulative exposure to MMVF, some ORs were elevated but there was no consistent pattern, the p values for trend were 0.87, 0.72, 0.10 and 0.62 respectively. In none of the countries was there a significantly increased OR for MMVF exposure. Similarly, there was no significant modification of the effect of MMVF exposure in stratified analysis for smoking or occupational exposure to asbestos.
Discussion
Asbestos and MMVF did not appear to exert an effect on lung cancer risk in our study population in Central and Eastern Europe, while in the UK we found an increased lung cancer risk following occupational exposure to asbestos. These results may be due to differences in fibre type or circumstances of exposure or could be the result of a difference in degree of misclassification in the UK versus Central and Eastern Europe—that is, if experts systematically overestimated asbestos exposure in the UK while experts in Central and Eastern Europe underestimated exposure. However, simulations assuming that the exposure assessment method has a sensitivity of 0.4 and a specificity of 0.9 in our study have suggested that differences in exposure misclassification are not likely to explain such a big difference in observed ORs: exposure prevalence has a strong effect on OR attenuation in the presence of misclassification. The lower prevalence of exposure to asbestos in Central and Eastern European countries as compared to UK might therefore explain a stronger attenuation in the OR in the former countries.22
The prevalence of occupational asbestos exposure in the present study, 11.4% in Central and Eastern Europe is comparable to what could be expected in a community‐based study in Europe but was lower than that of lung cancer patients classified as exposed to five or more fibre‐years in a series of 297 lung cancer cases from Hungary (18.7%).24
The prevalence of occupational asbestos exposure among controls from the UK appears to be high (37.1%) compared to previous population‐ or hospital‐based case‐control studies in this country (11–26%).9 Part of the explanation could be that the UK part of the study represents only the Liverpool area, which is a highly industrialised city including shipyards, and therefore may show a higher prevalence of exposure than the country overall. Exposure to asbestos clusters in certain occupations—for example, workers at extraction facilities, asbestos‐cement workers, insulation workers and ship‐yard workers.25 In our data few of the exposed men (0.2%) in Central and Eastern Europe had worked as dockers and freight handlers compared to 13% in the UK. A majority of UK dockworkers reported handling asbestos in loose bags, sacks or boxes up to the early 1970s. Such exposure circumstances were also similar there for insulators. In Central and Eastern Europe exposed men were more often employed as motor vehicle mechanics, where asbestos exposure may originate from the brakes and clutches and usually occurs at low levels. Other relevant differences in occupations held by exposed men in the UK and Central and Eastern Europe are listed in table 5.
Table 5 Proportion of exposed men ever employed in jobs entailing exposure to asbestos in the UK and the Central/Eastern European countries respectively.
| ISCO‐68 | Occupation | UK (%) | Central/Eastern Europe (%) |
|---|---|---|---|
| 581 | Firemen | 6 | 0.4 |
| 700 | Production supervisors and general foremen | 9 | 3 |
| 839 | Blacksmiths, toolmakers and machine‐tool operators not elsewhere classified | 1 | 4 |
| 843 | Motor vehicle mechanics | 9 | 22 |
| 871 | Plumbers and pipe fitters | 4 | 7 |
| 872 | Welders and flame‐cutters | 5 | 6 |
| 943 | Non‐metallic mineral product maker (asbestos cement) | 0.7 | 2 |
| 951 | Bricklayers, stonemasons and tile setters | 0.7 | 4 |
| 953 | Roofers | 5 | 2 |
| 954 | Carpenters, joiners and parquetry workers | 6 | 0.6 |
| 956 | Insulators | 3 | 1 |
| 959 | Construction workers not elsewhere classified | 2 | 5 |
| 971 | Dockers and freight handlers | 13 | 0.2 |
| 985 | Motor vehicle drivers | 4 | 9 |
The fibre lung burden analysis of 25 Hungarian lung cancer cases belonging to the series mentioned above and compared to 66 cases from Germany showed comparable level of chrysotile fibres but lower level of amphibole fibres.26 Our study showed similar results where 2.9% of the subjects in Central and Eastern Europe had been exposed to amphibole fibres and chrysotile, 5.2% to only chrysotile and 3.3% to unknown type of asbestos.
Stayner et al argue that the potency of amphiboles with respect to lung cancer is not different from that of chrysotile and that variations in risk depend more on fibre characteristics such as length and diameter than on type.27 Others state that chrysotile is less carcinogenic.12,13 Asbestos production in the former Soviet Union was estimated to 2 500 000 tons in 1986, which represented more than half of the world production at the time.9 Most of the asbestos used in Central and Eastern Europe was chrysotile imported from Russia, of which there are few epidemiological studies available. The largest amount of data on carcinogenicity concerning chrysotile is based on minerals from Canada, the US and South Africa.28,29 A study on fibres obtained from the lung tissue of workers and residents in a Russian chrysotile industry area concluded that the mean size of chrysotile fibres in Russia was about the same as in Canada, but that mean concentration of tremolite in the Russian samples were at least one order of magnitude lower compared to Canada. Moreover, no amosite or crocidolite fibres could be detected in the Russian samples, while low levels were detected in the Canadian samples.30 However, experimental data have shown that Russian chrysotile has a mutagenic effect on human lymphocytes and that intratracheal injection into hamsters resulted in tumours in 63% of the animals, of which 41% were malignant lung tumours.31,32 Pylev et al have also shown that the carcinogenicity of chrysotile varies between deposits within Russia.33
Several epidemiological studies involving small portions of amphibole asbestos, sometimes only in terms of a tremolite contamination, have shown an increased risk of lung cancer,21,34,35,36 whereas studies based on amphibole‐free exposure circumstances have not produced consistent evidence of an effect.37 A few epidemiological studies have been conducted on lung cancer risk among asbestos‐exposed workers in Central and Eastern Europe. Smailyte et al studied morbidity in two Lithuanian asbestos‐cement producing plants using only chrysotile asbestos and showed no increased risk of lung cancer (SIR 0.90, 95% CI 0.70 to 1.30).19 Likewise, a cohort from two asbestos cement plants in Poland using only chrysotile until 1985 failed to detect an increased risk of lung cancer in men (SMR 0.84, 95% CI 0.56 to 1.21).20 In the present study the OR for lung cancer after exposure to chrysotile only was lower than the OR for combined exposure to chrysotile and amphiboles in the UK (see table 3). Thus, our results are in line with previous studies. We did not detect an increased risk of lung cancer following exposure to amphiboles in Central and Eastern Europe. Possible reasons are low exposure levels, misclassification of exposure and different characteristics of fibres used in Central and Eastern Europe as compared to those used in the UK and elsewhere. It is also worth considering that asbestos consumption in the UK peaked early (in the 1960s) and declined significantly thereafter.38 In Central and Eastern Europe asbestos consumption peaked later (in the 1970s and 1980s) and still continues, so lung cancer burden from asbestos in Central and Eastern Europe may not yet have reached a sufficient level to show in our study.9
Recall bias must be considered as cases suffering from a chronic disease may ponder the possible causes of their disease, and therefore be more likely to recall some past exposures than would healthy controls. Nevertheless, our study investigated a wide range of occupational, environmental and lifestyle factors so the subjects could not reveal any particular hypothesis with regard to specific exposures. The subjects were also not confronted with questions about agents but were asked about job titles and job activities, and thereafter local experts designated corresponding exposures and levels.
Another concern is the possibility of misclassification of exposure. In our study the experts (for example, industrial hygienists) were trained together but worked in respective countries where they were familiar with the materials and processes in the different time periods. A validation study on the occupational exposure assessment showed that inter‐team agreement was poor with regard to asbestos and man‐made fibres and this was partly due to true differences in technology and use of materials between the countries.22 The same study suggested that misclassification was non‐differential between cases and controls and that the likely effects of misclassification on risk estimates were an attenuation towards the null, in particular for agents with poor agreement and with low prevalence in the study population. High exposure prevalence (such as observed in the UK) makes the attenuation effect of non‐differential exposure misclassification less severe. Thus, high prevalence of exposure alone makes it more likely to detect an association in the UK population and observe a higher OR, even if the true ORs would be comparable in both populations.
We did not find a carcinogenic effect of exposure to MMVF or for ceramic fibres. Also, no synergistic effect—that is departure from a multiplicative joint effect—of simultaneous exposure to asbestos and MMVF could be shown. The number of exposed subjects in our study was limited (5.2% to MMVF, even lower to ceramic fibres) and most exposed subjects were exposed to low levels, which reduced the power to detect an excess risk.
Conclusion
According to our data, asbestos does not contribute to the high burden of lung cancer in Central and Eastern Europe, but because of increasing trends in asbestos consumption, the full extent of lung cancer burden in Central and Eastern European countries may not yet be realised. In the UK men exposed to asbestos, in particular to amphiboles, experienced an increased risk of lung cancer. It is not clear to what extent this result is due to differences in fibre type, exposure conditions or degree of exposure misclassification but the most plausible explanation appears to be a true difference in exposure prevalence.
With regard to man‐made vitreous fibres and ceramic fibres, we did not observe an overall significant increased risk of lung cancer, but the power of our study to detect a small increased risk was low.
Main messages
This study did not detect an association between occupational exposure to asbestos and lung cancer risk in Central and Eastern Europe.
Low exposure levels in combination with misclassification of exposure may explain this result.
The epidemic of asbestos related lung cancer in Central and Eastern Europe may not yet have reached its maturity as the consumption of asbestos peaked relatively late, in the 1970s and 1980s.
Acknowledgements
We acknowledge the work of the coders and collaborators in each centre. We acknowledge also the contribution of Dr L Nadon and J Siemiatycki in the design of the study and G Ferro and N Travier in the data management and analysis. This work was supported by a grant from the European Commission's INCO‐COPERNICUS program (Contract No IC15‐CT96‐0313). In Warsaw, the study was supported by a grant from the Polish State Committee for Scientific Research grant#: SPUB‐M‐COPERNICUS/P‐05/DZ‐30/99/2000. In Liverpool, the work was funded by the Roy Castle Foundation as part of the Liverpool Lung Project.
Abbreviations
IARC - International Agency for Research on Cancer
ISCO - International Standard Classification of Occupations
MMVF - man‐made vitreous fibres
PAHs - polycyclic aromatic hydrocarbons
SIR - standardised incidence ratios
Footnotes
Competing interests: None.
References
- 1.Stewart B W, Kleihues P. eds. World cancer report. Lyon: IARC Press, 2003182–187.
- 2.Tobacco smoke and involuntary smoking/IARC working group on the evaluation of carcinogenic risks to humans IARC monographs on the evaluation of carcinogenic risks to humans; 83. Lyon: France, 2004 [PMC free article] [PubMed]
- 3.McDuffie H H, Klaassen D J, Dosman J A. Men, women and primary lung cancer—a Saskatchewan personal interview study. J Clin Epidemiol 199144537–544. [DOI] [PubMed] [Google Scholar]
- 4.Sridhar K S, Raub W, Jr, Duncan R C.et al Lung carcinoma in 1,336 patients. Am J Clin Oncol 199114496–508. [DOI] [PubMed] [Google Scholar]
- 5.Didkowska J, Manczuk M, McNeill A.et al Lung cancer mortality at ages 35–54 in the European Union: ecological study of evolving tobacco epidemics. BMJ 2005331189–191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ferlay J, Bray F, Pisani P.et al Globocan 2002: Cancer incidence, mortality and prevalence worldwide. IARC cancerbase No 5. Version 2.0. Lyon: IARC Press, 2004
- 7.Asbestos/IARC working group on the evaluation of carcinogenic risks to humans IARC monographs on the evaluation of carcinogenic risks to humans; 14. Lyon: France, 1977
- 8. Overall evaluation of carcinogenicity: an updating of IARC Monographs, Volumes 1 to 42. In: IARC monographs on the evaluation of carcinogenic risks to humans. Lyon: France, 1987(Suppl 7)P106–P116. [PubMed]
- 9.Albin M, Magnani C, Krstev S.et al Asbestos and cancer: An overview of current trends in Europe. Environ Health Perspect 1999107(Suppl 2)289–298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Karjalainen A, Anttila S, Vanhala E.et al Asbestos exposure and the risk of lung cancer in a general urban population. Scand J Work Environ Health 199420243–250. [DOI] [PubMed] [Google Scholar]
- 11.Yeung P, Rogers A. An occupation‐industry matrix analysis of mesothelioma cases in Australia 1980–1985. Appl Occup Environ Hyg 20011640–44. [DOI] [PubMed] [Google Scholar]
- 12.Hughes J, Weill H. Asbestos and man‐made fibres. In: Samet JM, ed. Epidemiology of lung cancer. NY, Marcel Dekker Inc 1994185–194.
- 13.Sluis‐Cremer G K, Liddell F D, Logan W P.et al The mortality of amphibole miners in South Africa, 1946–80. Br J Ind Med 199249566–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McCulloch J. Asbestos mining in Southern Africa, 1893–2002. Int J Occup Environ Health 20039230–235. [DOI] [PubMed] [Google Scholar]
- 15.Kjaerheim K, Boffetta P, Hansen J.et al Lung cancer among rock and slag wool production workers. Epidemiology 200213445–453. [DOI] [PubMed] [Google Scholar]
- 16.Plato N, Westerholm P, Gustavsson P.et al Cancer incidence, mortality and exposure‐response among Swedish man‐made vitreous fiber production workers. Scand J Work Environ Health 199521353–361. [DOI] [PubMed] [Google Scholar]
- 17.Pohlabeln H, Jockel K H, Bruske‐Hohlfeld I.et al Lung cancer and exposure to man‐made vitreous fibers: results from a pooled case‐control study in Germany. Am J Ind Med 200037469–477. [DOI] [PubMed] [Google Scholar]
- 18.Man‐made Vitreous Fibres/IARC working group on the evaluation of carcinogenic risks to humans IARC monographs on the evaluation of carcinogenic risks to humans; 81. Lyon: France, 2002 [PMC free article] [PubMed]
- 19.Smailyte G, Kurtinaitis J, Andersen A. Cancer mortality and morbidity among Lithuanian asbestos‐cement producing workers. Scand J Work Environ Health 20043064–70. [DOI] [PubMed] [Google Scholar]
- 20.Szeszenia‐Dabrowska N, Wilczynska U, Szymczak W. Mortality of workers at two asbestos‐cement plants in Poland. Int J Occup Med Environ Health 200013121–130. [PubMed] [Google Scholar]
- 21.Ulvestad B, Kjaerheim K, Martinsen J I.et al Cancer incidence among workers in the asbestos‐cement producing industry in Norway. Scand J Work Environ Health 200228411–417. [DOI] [PubMed] [Google Scholar]
- 22.Mannetje A, Fevotte J, Fletcher T.et al Assessing exposure misclassification by expert assessment in multicenter occupational studies. Epidemiology 200314585–592. [DOI] [PubMed] [Google Scholar]
- 23.ILO International Standard Classification of Occupations, second edition. Geneva: Switzerland: International Labour Office 1968
- 24.Mandi A, Posgay M, Vadasz P.et al Role of occupational asbestos exposure in Hungarian lung cancer patients. Int Arch Occup Environ Health 200073555–560. [DOI] [PubMed] [Google Scholar]
- 25.Niklinski J, Niklinska W, Chyczewska E.et al The epidemiology of asbestos‐related diseases. Lung Cancer 200445(Suppl 1)S7–S15. [DOI] [PubMed] [Google Scholar]
- 26.Rodelsperger K, Mandi A, Tossavainen A.et al Inorganic fibres in the lung tissue of Hungarian and German lung cancer patients. Int Arch Occup Environ Health 200174133–138. [DOI] [PubMed] [Google Scholar]
- 27.Stayner L T, Dankovic D A, Lemen R A. Occupational exposure to chrysotile asbestos and cancer risk: a review of the amphibole hypothesis. Am J Public Health 199686179–186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hodgson J T, Darnton A. The quantitative risks of mesothelioma and lung cancer in relation to asbestos exposure. Ann Occup Hyg 200044565–601. [PubMed] [Google Scholar]
- 29.Steenland K, Stayner L. Silica, asbestos, man‐made mineral fibers, and cancer. Cancer Causes Control 19978491–503. [DOI] [PubMed] [Google Scholar]
- 30.Tossavainen A, Kovalevsky E, Vanhala E.et al Pulmonary mineral fibers after occupational and environmental exposure to asbestos in the Russian chrysotile industry. Am J Ind Med 200037327–333. [DOI] [PubMed] [Google Scholar]
- 31.Korkina L G, Durnev A D, Suslova T B.et al Oxygen radical‐mediated mutagenic effect of asbestos on human lymphocytes: suppression by oxygen radical scavengers. Mutat Res 1992265245–253. [DOI] [PubMed] [Google Scholar]
- 32.Pylev L N. Pretumorous lesions and lung and pleural tumours induced by asbestos in rats, Syrian golden hamsters and Macaca mulatta (rhesus) monkeys. IARC Sci Publ 198030343–355. [PubMed] [Google Scholar]
- 33.Pylev L N, Vasil'eva L A, Krinari G A. [Insignificant carcinogenicity of chrysotile asbestos of the Ak‐Dovurak deposits in Russia and its possible links]. Gig Sanit 200472–73. [PubMed]
- 34.Albin M, Jakobsson K, Attewell R.et al Mortality and cancer morbidity in cohorts of asbestos cement workers and referents. Br J Ind Med 199047602–610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Liddell F D, McDonald A D, McDonald J C. Dust exposure and lung cancer in Quebec chrysotile miners and millers. Ann Occup Hyg 1998427–20. [DOI] [PubMed] [Google Scholar]
- 36.Raffn E, Villadsen E, Engholm G.et al Lung cancer in asbestos cement workers in Denmark. Occup Environ Med 199653399–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Piolatto G, Negri E, La Vecchia C.et al An update of cancer mortality among chrysotile asbestos miners in Balangero, northern Italy. Br J Ind Med 199047810–814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Tweedale G.Magic mineral to killer dust: Turner and Newall and the asbestos hazard. Oxford, UK: Oxford University Press, 2000