Mesothelioma incidence seems to have leveled off in Sweden

Pleural mesothelioma is an asbestos-related disease that has been increasing constantly in all industrialized countries despite abolished or restricted use.1, 2 Reasons for the increase are the delay between exposure and disease, which is thought to be 30 or more years, and indirect exposure.3, 4, 5 The model-based predictions on future incidence in mesothelioma have estimated maximal incidence between years 2015 and 2030 in Western Europe but sooner in USA where the use ceased earlier.1, 2, 3, 6 Consistent with these models, the incidence of pleural mesothelioma has been increasing in the countries for which data are available, including Sweden.2, 7, 8, 9 Asbestos imports to Sweden were drastically reduced in 1976, which time has coincided with the maximal use of asbestos in many industrialized countries. The difference in the phasing of asbestosis, however, was that in Sweden the use was almost abolished completely whereas in many Western European countries the use has tailed to the 1990s.3, 6, 8 Sweden would be expected to be among the first countries to show the leveling off of the incidence trend for pleural mesothelioma. Mesothelioma risks are thought to be higher from amphibole types of asbestos, crocidolite and amosite, than from chrysotile.10 Chrysotile has been the most common type of asbestos in Sweden, but according to a study on Swedish insulation workers, about 50% had also used amosite and crocidolite.11 Crocidolite was the first type of asbestos that was prohibited in 1975. We collected data on pleural mesothelioma from the Swedish Cancer Registry, covering years 1961–2000. The International Classification of Diseases-7 code was 162.2, malignant mesothelioma in the pleura, according to the Cancer Registry.12 The histology code is malignant mesothelioma in 97% of male and 96% of female cases during 1958–93. The specific histology was not available to us after 1993, however, and we used only the code 162.2 to define the cases. We calculated incidence rates for men and women in 5-year periods (Fig. 1). Age-standardized rates were calculated in 5-year age bands and they were adjusted to the European standard population. For the first time throughout the observation period, the incidence rate for of pleural tumors appeared to decrease for men and women. The incidence for men was 1.82/100,000 (95% CI = 1.66–1.99/100,000) in 1991–95, compared to 1.79/100,000 (1.63–1.95/100,000) in 1996–2000. For women the rates were 0.32 (0.26–0.38/100,000) and 0.23/100,000 (0.18–0.28/100,000), respectively. The rates between the periods did not differ significantly from each other (95% CI overlapped) indicating that the increase in incidence has leveled off. The difference in linear regression coefficients for annual incidence rates between 1966–95 and 1996–2000 was highly significant for men (p = 0.0006). Moreover, the similar trend in men and women supports the notion that indeed a maximum has been reached in years 1991–95. In Figure 2 we plotted the incidence rates by diagnostic age in men and women. The incidence was decreasing in the oldest age group (>75 years) for men and women, and it was quite stable in the youngest age group (45–59 years). Among 60- to 74-year-old men, however, some increase appeared to take place toward the last observation period; in this group of women the incidence rates fluctuated. Cases of mesothelioma have been reported in Turkish immigrants to Sweden, because of exposure to erionite in Turkey, and it is thus likely that some cases are due to this source.4 Age-adjusted incidence of pleural mesothelioma for men and women in the Swedish Cancer Registry between 1961–2000. Numbers on top of the curves show the number of cases in 5-year periods and the vertical bars are confidence intervals. Incidence of pleural mesothelioma in male and female age groups. To our knowledge this is the first report on the leveling off of the incidence for pleural mesothelioma. We have shown elsewhere that peritoneal mesothelioma has not increased in Swedish men since 1985.13 There have been no diagnostic changes in the past decades that would invalidate the comparison of time trends since the 1970s. It is assumed that before that time some underreporting was likely,4 and as pointed out above, some cases in the past 2 decades can be ascribed to immigrants. We want to emphasize a cautious interpretation, however, and a need to observe future incidence data. The maximum in years 1991–95 would appear early, considering the long latency for the disease and the UK parallels between asbestos imports and mesothelioma mortality with a lag time of some 50 years.14 One possibility is that we are observing a transient leveling off, due to a major restructuring of asbestos-related industries, such as shipbuilding that collapsed in Sweden in the 1970s to 1980s. If indeed the maximum was in the period 1991–95 (1993 as the central year), however, it occurred 17 years after the decimation of asbestos imports to Sweden in 1976.8 This would be good news to other Western European countries where maxima have been predicted between 2015 and 2030. The data on exposure to asbestos, and to different fiber types, are quite difficult to compare anywhere, but if we assume that a country achieved the Swedish level of asbestos exposure 10 or 20 years later, incidence maxima would then be expected for 2003 and 2013, respectively. The future rates will tell the truth. The asbestos toll has been predicted at 250,000 male deaths from mesothelioma in 6 Western European countries counting cases for the next 30 years.3 Based on the historic trends in the incidence it has been estimated that the background rate without asbestos was at 1–2 per million, close to the Swedish female rates of 1 per million in the early period in our study.15 This background would correspond to 20 male cases in a 5-year period, and 160 cases during the 8 5-year periods that our study covered. As the total number of male mesothelioma cases was 2190, 2,030 could be ascribed to asbestos up to year 2000. The female mesothelioma toll is less accurate, but applying the same background rate as was used for men, the asbestos related excess would be 350 (504–160). Even if mesothelioma incidence had culminated in Sweden, asbestos-related deaths will continue to accumulate, and we have not included the contribution of asbestos to the lung cancer burden. Yours sincerely, The cancer data was provided by M. Talbäck from the Swedish Cancer Registry. Kari Hemminki* , Xinjun Li*, * Department of Biosciences at Novum, Karolinska Institute, Huddinge, Sweden, Division of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany.