Document rBGOdMa8gd0jmGvdge4g89M7J

Volume 159 Number 2 January 15, 2004 SPACIAC ARTI CLE American Journal of EPIDEMIOLOGY Copyright 2004 by The Johns Hopkins Bloomberg School of Public Health Sponsored by the Society for Epidemiologic Research Published by Oxford University Press Mesothelioma Trends in the United States: An Update Based on Surveillance, Epidemiology, and End Results Program Data for 1973 through 2003 Bertram Price and Adam Ware From Price Associates, Inc., White Plains, NY. Received for publication June 3, 2003; accepted for publication July 25, 2003. Using 1973-2000 mesothelioma incidence data released by the Surveillance, Epidemiology, and End Results Program in April 2003, the authors estimated the parameters of a birth-cohort and age model to determine whether previously reported patterns of mesothelioma in the United States have changed. Compared with analyses based on data through 1992, a slower decline was found in male cases immediately after a peak in 2000-2004, but no other notable changes in the time pattern were detected. Analysis confirmed that the annual number of male mesothelioma cases, which increased steeply from the 1970s through the mid-1990s, has leveled off in terms of both the age-adjusted rate and the absolute numbers of cases. After a peak of approximately 2,000 cases, a return to background levels is expected by 2055. The total projected number of male mesothelioma cases in 2003-2054 is approximately 71,000. The maximum lifetime risk for males, which occurs for the 1925-1929 birth cohort, is 1.8 x 10-3. The age-adjusted rate for females is constant, as are the female lifetime mesothelioma risk across birth cohorts (3.6 x 10-4) and the annual risk (3.9 x 10-6). The time pattern of cases for females supports the existence of a threshold exposure for mesothelioma and a quantifiable background rate. asbestos; incidence; mesothelioma Abbreviation: SEER, Surveillance, Epidemiology, and End Results. Mesothelioma incidence often is interpreted as an indi cator of exposure to asbestos. Projections of the number of mesothelioma cases over time may be used to evaluate various asbestos health management interventions including exposure limits and products bans. Mesothelioma projec tions also provide a foundation for estimating the number of potential lawsuits from persons claiming occupational expo sure to asbestos or exposure resulting from use of previously manufactured asbestos-containing products. Noting recent publications discussing mesothelioma incidence in North America and that projected numbers of future mesothelioma cases in the Netherlands, Sweden, and the United Kingdom have been adjusted based on new data (1-4), we analyzed current Surveillance, Epidemiology, and End Results (SEER) Program data (5) to update mesothelioma trends in the United States. Reprint requests to Dr. Bertram Price, Price Associates, Inc., One North Broadway, Suite 406, White Plains, NY 10601 (e-mail: bprice@ priceassociatesinc.com). 107 Am J Epidemiol 2004;159:107-112 108 Price and Ware FIGURE 1. Age-adjusted incidence rates of mesothelioma (pleural + peritoneal) in the United States based on Surveillance, Epidemiology, and End Results Program data released in April 2003. MATERIALS AND METHODS We applied maximum likelihood estimation (6) to SEER data (5) to estimate the parameters of a birth-cohort and age model (7). The SEER database contains information on cancer incidence in the United States by year of diagnosis and age from nine cancer registries for 1973 through 1991 and 11 registries for 1992 through 2000. The SEER registries represent 14 percent of the US population, are similar to the United States regarding measures of poverty and education, and tend to be more urban than the general US population (8-10). For males and females, we extracted counts of mesothelioma cases and corresponding population counts for each year and each 5-year age interval from 0-4 through 85 years or older. These data were realigned into 5-year birth-cohort categories (1885-1889 through 1970-1974). Incidence estimates calculated from the model were combined with mortality rates for all causes in a survival analysis to calculate, for each birth cohort, the probability of mesothelioma occurring in each 5-year age interval, condi tioned on cancer-free survival to that age interval (11). These probabilities were used to calculate the lifetime risk of mesothelioma for each birth cohort. For all-cause mortality, we created a table of cohort mortality rates by 5-year age interval for each birth cohort. The cohort mortality rates were developed from two sources: 1) cross-sectional mortality rates for 5-year age intervals from the SEER system, years 1969-2000 (12); and 2) cross sectional mortality rates by 10-year age intervals for indi vidual years 1900 through 1968 (13). We factored the 10year age intervals into two 5-year intervals and then realigned the results to form a table of cohort mortality rates for birth cohorts 1900-1904 through 1995-1999. To project the annual number of future mesothelioma cases, we required natality data (14), mesothelioma proba bilities for each age interval, and all-cause mortality rates for future birth cohorts. The number of births for all future birth cohorts was set equal to the number of births reported for the most recent birth cohort, 1995-1999. Future birth-cohort mesothelioma probabilities for females, starting with the 1965-1969 cohort, were set to the average of past probabilities for females, which have been relatively constant since the 1900-1904 birth cohort. Mesothelioma probabilities for males have been declining toward those for females. Therefore, probabilities for males in future birth cohorts starting with 1965-1969 were set equal to the averages for females, which we interpreted as background rates for mesothelioma. All-cause mortality rates for the year 2000 were used for all future birth cohorts. RESULTS Trend in age-adjusted rates The age-adjusted incidence (figure 1) indicates a consis tently higher rate for males versus females across all years, Am J Epidemiol 2004;159:107-112 Current Mesothelioma Trends in the United States 109 FIGURE 2. Asbestos use (consumption) in the United States and projected numbers of male and female mesothelioma cases based on a birthcohort and age model estimated from Surveillance, Epidemiology, and End Results (SEER) Program data for two periods, 1973-1992 and 1973 2000. an increasing trend over the years for males, and a virtually constant rate for females. Accounting for a latency period of 20-40 years for mesothelioma, the increased incidence for males reflects the increased use of asbestos (15) in the United States (figure 2) and the high levels of occupational exposure prior to the late 1960s, when formal workplace exposure limits were established. The age-adjusted rate for males was flat from 1990 through 2000. Projections of future male mesothelioma cases, discussed later, indicate that the number of male cases per year has peaked and is declining. Comparison of trends: data for 1973-1992 versus 1973-2000 A direct comparison of mesothelioma trends based on current data with trends published by Price (7) would not be meaningful because the SEER data have changed since 1992. Two registries were added starting with 1992 data, and mesothelioma counts for the years 1973 through 1991 have been corrected. Therefore, our modeling is based on the current SEER database, first using cases diagnosed from 1973 through 1992 and then using all cases, 1973 through 2000. Am J Epidemiol 2004; 159:107-112 Model fits For each data set (males, females; two time intervals), the predicted values tracked the observed values reasonably well, although there were a few data points with large deviances. The 1973-1992 data sets had 80 data points; the 1973-2000 data sets had 101 data points. When the large deviances were excluded, the statistical fits were acceptable (p > 0.05 based on total deviance) for each model. The numbers of data points remaining were, for 1973-1992, 79 for males and 80 for females and, for 1973-2000, 92 for males and 99 for females. Lifetime risk of mesothelioma For females, the lifetime risk of mesothelioma is approxi mately constant at 3.6 x 10-4 (figure 3). The lifetime risk for males (1973-1992 data, not shown) is a maximum of 2.1 x 10-3 for the 1925-1929 birth cohort and then declines sharply for subsequent cohorts, except for an isolated increase for the 1940-1944 cohort. Regarding the 1973 2000 data, the maximum risk is 1.8 x 10-3 (figure 3). The trend is relatively flat before starting downward after the 1940-1944 cohort. 110 Price and Ware Qj-------- ,---------------------- !---------------------- 1---------------------- 1---------------------- 1---------------------- 1---------------------- I---------------------- 1---------------------- 1---------------------- 1---------------------- 1---------------------- 1------- 1 1907 1912 1917 1922 1927 1932 1937 1942 1947 1952 1957 1962 Birth Cohort Mid-Interval Year FIGURE 3. Lifetime probability (risk) of mesothelioma (pleural + peritoneal) and 95% confidence intervals (vertical bars) based on a birthcohort and age model estimated from 2003 Surveillance, Epidemiology, and End Results Program data covering 1973-2000. Projected number of mesothelioma cases For the two data sets, mesothelioma projections for females (figure 2) are virtually identical. Although the ageadjusted rate for females is constant, the number of mesothe lioma cases increases slightly over time as a function of population size and shifting age distribution. Currently (2003), the annual estimate of female mesothelioma cases is approximately 560. The trend for males in both data sets peaks in the 2000 2004 time frame at slightly more than 2,000 mesothelioma cases (figure 2). The 1973-2000 model shows fewer mesothelioma cases than the 1973-1992 model during the peak years, but not overall. The total numbers of male mesothelioma cases projected for 2003 through 2054 are 70,990 for the 1973-2000 data and 73,892 for the 1973 1992 data. DISCUSSION Overall, there have been no substantive changes in the time pattern of mesothelioma incidence in the United States since 1992. This conclusion would not be apparent if the current results using SEER data for 1973-2000 were compared with the results published by Price (7) that were based on SEER data for 1973 through 1992. That estimated model (7) showed a sharper decline in the number of annual male mesothelioma cases, projecting approximately 7,000 fewer cases between 2003 and 2054, than the model based on the most current data. However, a comparison to the Price results (7) would not be meaningful because of changes in SEER data and application of the cohort mortality rates used here. Therefore, it was necessary to reestimate the model using revised data through 1992 for comparison to the model based on current data through 2000. This comparison indi cated no substantive change in the time pattern or number of future cases. In addition to describing the time pattern of mesothelioma incidence, our results contain information about the potential effects of environmental exposures to asbestos and two related issues: 1) the existence of an exposure threshold and 2) the existence and magnitude of a background rate (i.e., mesotheliomas not associated with asbestos exposure). (In this discussion, threshold does not mean an exposure below which the risk of mesothelioma is zero; the threshold is an asbestos exposure level associated with a negligible incre mental risk of mesothelioma.) The constancy alone of the mesothelioma risk for females over time supports the existence of a threshold exposure. The effects, if any, of environmental exposures for the US popu lation as a whole cannot be observed for males; as a group, males experienced high occupational exposures that masked low-exposure effects. In contrast, female exposures to asbestos have been primarily environmental. In the 1930s through the 1960s, women generally did not work in indus tries in which men experienced high levels of exposure to asbestos. For some women, exposures have been domestic (i.e., they live in households with men who were exposed Am J Epidemiol 2004;159:107-112 Current Mesothelioma Trends in the United States 111 occupationally), but the number of women with high levels of domestic exposure, although not zero, would have been relatively small. However, all women were exposed to asbestos in the environment, an exposure that would have increased since the 1930s, especially the dramatic increase during the 40-year period from 1930 to 1970 in the amount of asbestos used in Us products (figure 2). As use increased, higher environmental exposures would have resulted from asbestos mining, manufacturing, and product use (e.g., vehicle brake systems, construction materials, and other asbestos-containing products) during this period and beyond for as long as the asbestos-containing products were in use. Nevertheless, the mesothelioma risk for women has not increased. The age-adjusted mesothelioma rate for females was constant at an average of approximately 0.30 per 100,000 between 1973 and 1982, when it showed a one-time increase to 0.40 per 100,000 (figure 1). One might be tempted to inter pret this change as a response to increasing environmental exposure. However, the female rate, which should have continued to increase after 1982 if the environmental exposure explanation is correct, remained constant after 1982. There fore, the one-time mesothelioma increase for females does not indicate increasing exposure to asbestos in the environment but more likely a diagnostic effect. Environmental exposure levels, although increasing, have not triggered a risk response in women. Therefore, those exposure levels must have been below a threshold for mesothelioma. A threshold higher than typical environmental asbestos exposures, as argued above, implies the existence of back ground mesotheliomas caused by agents other than asbestos that could account for most, if not all, female cases of mesothelioma. A number of studies provide evidence of background mesotheliomas (16-28). Simian virus 40 (SV40) is one of the nonasbestos agents that has been identi fied as a possible cause of mesothelioma (29, 30). Simian virus 40 was a contaminant in polio vaccines administered to 10-30 million people in the United States, mostly children, between 1955 and 1963 (30). It has not been determined whether simian virus 40 acting alone can cause mesothe lioma (30). Our analysis does not indicate an increased risk in the 1955-1959 or 1960-1964 birth cohorts, those most likely to have received the contaminated vaccine (refer also to Strickler et al. (31)). If all female cases of mesothelioma were unrelated to asbestos exposure, our analysis indicates that the lifetime background risk would be 3.6 x 10-4 (figure 3) and the current annual risk would be approximately 4 per million (over 5 per million if the population at risk is aged >20 years). These background risk levels would be upper bounds if a portion of female cases of mesothelioma were due to occupational, domestic, or unique high environmental exposures. The background rates for females may not apply directly to males because the percentages of pleural and peritoneal mesotheliomas are different for men and women. In the SEER database, peritoneal mesotheliomas account for 7 percent of all mesotheliomas in males but 17 percent of all mesotheliomas in females. It is possible that peritoneal mesotheliomas in women are of two types but cannot be distinguished histologically: 1) those that are similar to mesotheliomas in males and 2) those that are unique, as Am J Epidemiol 2004;159:107-112 evidenced in part by longer survival times of the cases (32; Andrew Churg, University of British Columbia, personal communication, 2002). On the basis of the differences between cases of mesothelioma in males and females, an adjustment to the background rates suggested above for females would be required if they were to be applied for males. For example, if all peritoneal mesotheliomas in females were excluded, the annual background rate for males would be 83 percent of the background rate for females. The mesothelioma epidemic in the United States, which is beginning to recede, has been a consequence of ongoing high-level occupational exposure to asbestos from the 1930s through the 1960s in workplaces populated almost exclu sively by males. Asbestos regulations promulgated by the US Occupational Safety and Health Administration (OSHA) in the early 1970s have led to dramatic reductions in expo sure. These reductions are reflected in the pattern of mesothelioma incidence in the SEER data and were magni fied through our modeling and analysis. It is unlikely that low environmental exposure to asbestos, or any other type of low-level asbestos exposure, is associated with more than a negligible risk of mesothelioma. REFERENCES 1. Greenberg AK, Lee TC, Ross WN. The North American expe rience with mesothelioma incidence. In: Robinson BWS, Chahinian AP, eds. Mesothelioma. London, United Kingdom: Martin Dunitz Ltd, 2002. 2. Segura O, Burdorf A, Looman C. Update of predictions of mor tality from pleural mesothelioma in the Netherlands. Occup Environ Med 2003;60:50-5. 3. Hemminki K, Li X. Mesothelioma incidence seems to have lev eled off in Sweden. Int J Cancer 2003;103:145-6. 4. White C. Annual deaths from mesothelioma in Britain to reach 2000 by 2010. BMJ 2003;326:1417. 5. SEER 1973-2000 public-use data. Bethesda, MD: National Cancer Institute, Division of Cancer Control & Population Sci ences, Surveillance Research Program, Cancer Statistics Branch, released April 2003, based on the November 2002 sub mission. (www.seer.cancer.gov/publicdata). 6. Aptech Systems, Inc. GAUSS mathematical and statistical sys tem, version 4.0. Maple Valley, WA: Aptech Systems, Inc. 7. Price B. Analysis of current trends in United States mesothe lioma incidence. Am J Epidemiol 1997;145:211-18. 8. Ries LAG, Eisner MP, Kosary CL, et al, eds. SEER cancer sta tistics review, 1975-2000. Bethesda, MD: National Cancer Institute. (http://seer.cancer.gov/csr/1975_2000/, 2003). 9. About SEER. Background discussion. Bethesda, MD: National Cancer Institute, Division of Cancer Control & Population Sci ences, Surveillance Research Program. (http://seer.cancer.gov/ about/). 10. SEER registries. Registries discussion. Bethesda, MD: National Cancer Institute, Division of Cancer Control & Population Sci ences, Surveillance Research Program. (http://seer.cancer.gov/ registries/). 11. Wun LM, Merrill RM, Feuer EJ. Estimating lifetime and ageconditional probabilities of developing cancer. Lifetime Data Anal 1998;4:169-86. 12. SEER*Stat database: mortality--all COD, public-use with 112 Price and Ware state, total U.S. (1969-2000). Bethesda, MD: National Cancer Institute, Division of Cancer Control & Population Sciences, Surveillance Research Program, Cancer Statistics Branch, released April 2003. (http://seer.cancer.gov). (Underlying mor tality data provided by the National Center for Health Statistics (Hyattsville, MD) (www.cdc.gov/nchs)). 13. National Center for Health Statistics. Death rates for selected causes by 10-year age groups, race, and sex: death registration states, 1900-32, and United States, 1933-98. (http:// www.cdc.gov/nchs/datawh/statab/unpubd/mortabs/ hist290.htm). 14. Ventura SJ, Hamilton BE, Sutton PD. Revised birth and fertility rates for the United States, 2000 and 2001. Natl Vital Stat Rep 2003;51:1-18. 15. Selikoff IJ, Lee DH. Asbestos and disease. New York, NY: Academic Press, 1978. 16. Peto J, Henderson BE, Pike MC. Trends in mesothelioma inci dence in the United States and the forecast epidemic due to asbestos exposure during World War II. Banbury report 9: quantification of occupational cancer. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1981:51-69. 17. Spirtas R, Heineman EF, Bernstein L, et al. Malignant mesothe lioma: attributable risk of asbestos exposure. Occup Environ Med 1994;51:804-11. 18. Talerman A, Montero JR, Chilcote RR, et al. Diffuse malignant peritoneal mesothelioma in a 13-year-old girl. Report of a case and review of the literature. Am J Surg Pathol 1985;9:73-80. 19. Nascimento AG, Keeney GL, Fletcher CD. Deciduoid perito neal mesothelioma. An unusual phenotype affecting young females. Am J Surg Pathol 1994;18:439-45. 20. Orosz Z, Nagy P, Szentirmay Z, et al. Epithelial mesothelioma with deciduoid features. Virchows Arch 1999;434:263-6. 21. Shanks JH, Harris M, Banerjee SS, et al. Mesotheliomas with deciduoid morphology: a morphologic spectrum and a variant not confined to young females. Am J Surg Pathol 2000;24:28594. 22. Ordonez NG. Epithelial mesothelioma with deciduoid features: report of four cases. Am J Surg Pathol 2000;24:816-23. 23. Gloeckner-Hofmann K, Zhu XZ, Bartels H, et al. Deciduoid pleural mesothelioma affecting a young female without prior asbestos exposure. Respiration 2000;67:456-8. 24. Puttagunta L, Vriend RA, Nguyen GK. Deciduoid epithelial mesothelioma of the pleura with focal rhabdoid change. Am J Surg Pathol 2000;24:1440-3. 25. Henley JD, Loehrer PJ Sr, Ulbright TM. Deciduoid mesothe lioma of the pleura after radiation therapy for Hodgkin's dis ease presenting as a mediastinal mass. Am J Surg Pathol 2001; 25:547-8. 26. Reis-Filho JS, Pope LZ, Milanezi F, et al. Primary epithelial malignant mesothelioma of the pericardium with deciduoid fea tures: cytohistologic and immunohistochemical study. Diagn Cytopathol 2002;26:117-22. 27. Serio G, Scattone A, Pennella A, et al. Malignant deciduoid mesothelioma of the pleura: report of two cases with long sur vival. Histopathology 2002;40:348-52. 28. Peterson JT, Greenberg DS, Buffler PA. Non-asbestos-related malignant mesothelioma: a review. Cancer 1984;54:951-60. 29. Carbone M, Fisher S, Powers A, et al. New molecular and epi demiological issues in mesothelioma: role of SV40. J Cell Physiol 1999;180:167-72. 30. Stratton K, Almario DA, McCormick MC, eds. Immunization safety review: SV40 contamination of polio vaccine and can cer. Immunization Safety Review Committee. Washington, DC: National Academies Press, 2002. 31. Strickler HD, Goedert JJ, Devesa SS, et al. Trends in U.S. pleu ral mesothelioma incidence rates following simian virus 40 contamination of early poliovirus vaccines. J Natl Cancer Inst 2003;95:38-45. 32. Kerrigan SA, Turnnir RT, Clement PB, et al. Diffuse malignant epithelial mesotheliomas of the peritoneum in women: a clinicopathologic study of 25 patients. Cancer 2002;4:378-85. Am J Epidemiol 2004;159:107-112