Mesothelioma in Japan
Malignant Mesothelioma in Japan: Analysis of Registered Autopsy Cases.
In the Annual of the Pathological Autopsy Cases in Japan, issued by the Japanese Society of Pathology from 1958 to 1996, a total of 1,846 (0.17%) malignant mesothelioma cases (1,287 male, 558 female, I unknown) were registered among 1,056,259 autopsy cases. The frequency of mesothelioma (number of cases/total autopsy cases) was 0.10% (461/440,334) for the term 1958-1979, 0.18% (716/390,124) for 1980-1989, and 0.30% (669/225,801) for 1990-1996; the frequency of cases increased significantly over the time periods (p [is less than] .0001). Among 1,785 cases for which tumor sites were ascertained, there were 1,213 pleural mesothelioma (68.0%), 431 peritoneal (24.1%), 108 pericardial (6.1%), 6 tunica vaginalis testis (0.3%), and 28 "others" (1.6%). Histological cell type was noted in 598 cases; 245 (41.0%) were epithelial, 168 (28.1%) were biphasic, and 185 (30.9%) were sarcomatous. Seventy-three (0.007%) cases of malignant mesothelioma with asbestosis were found during the entire 39-y period. The frequency of those with asbestosis (number of cases/total autopsy cases) was 0.001% (5/440,334) for the term 1958-1979, 0.006% (27/390,124) for 1980-1989, and 0.018% (41/225,801) for 1990-1996; this increase over time was statistically significant (p [is less than] .0001). Researchers expect that cases of asbestos-related mesothelioma will increase in Japan in the future. Tumor sites and histological cell types of mesothelioma with asbestosis did not differ from those in individuals without asbestosis.
MALIGNANT MESOTHELIOMA originates from serosal cells of the pleural, peritoneal, and pericardial cavities, and from the tunica vaginalis testis. Since Wagner et al.[1] reported mesotheliomas from asbestos exposure in South Africa in 1960, many researchers have subsequently confirmed his conclusions. In Japan, consumption of asbestos has increased markedly since World War II, and it is one of the leading consumers in the world.
Since 1958, autopsy cases in Japan have been registered in the Annual of the Pathological Autopsy Cases in Japan (hereinafter Annual) by the Japanese Society of Pathology. Baba[2] surveyed the annuals from 1974 through 1980 and reported that, although there were no pleural malignant mesotheliomas associated with asbestosis until 1977, there was 1 death in 1978, 3 in 1979, and 2 in 1980. His report suggests that the number of occupational pleural malignant mesotheliomas has begun to increase in Japan. In this study, we sought to extend the survey of the Annuals and to elucidate the features of malignant mesothelioma with asbestosis.
Materials and Method
The Annuals of the Pathological Autopsy Cases in Japan (years 1958-1996) by the Japanese Society of Pathology were used. Until the annual for 1996 was issued, a total of 1,846 (1,287 male, 558 female, 1 unknown) autopsy cases with malignant mesothelioma had been recorded among 1,056,259 autopsy cases. This documentation is a passive system for the recording of autopsy reports received from individual pathologists; no confirmatory reviews are provided. Tissue slides are not sent in and archived. Some diagnoses, however, of mesothelioma have been confirmed by a panel or by consultation because mesothelioma is a difficult diagnosis.
The age, sex, clinical diagnosis, residence area, occupation, main pathological lesions, and some additional lesions are recorded in each Annual. Malignant mesotheliomas are regarded as the main pathological lesions, independent of tumor size. The space for the description of occupational history in the Annual is restricted in that only the main or most recent occupation is described. The life employment system has been relatively maintained in Japan.
The frequency of malignant mesotheliomas and those with asbestosis in autopsy cases in each of the three terms--1958-1979 (22 y), 1980-1989 (10 y), and 1990-1996 (7 y)--as well as the change over time, were examined. We used the Chi-square test for statistical evaluations.
Results
The frequency of malignant mesothelioma (number of cases/total number of autopsy cases) was 0.17% (1,846/1,056,259) for the entire 39-y period from 1958 to 1996 (Table 1). The comparison over the three periods revealed a significant increase in the frequency of mesothelioma, from 0.10% (461/440,334) in 1958-1979, to 0.18% (71 6/390,124) in 1980-1989, to 0.30% (669/225,801) in 1990-1996 ([chi square] = 316.751, p [is less than] .0001).
[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]
The age and sex distributions are shown in Table 2. Seventy-four percent of all cases were between 50 y and 79 y of age, and the highest frequency of malignant mesothelioma was found in both sexes in the 60-69 y age range. There were 17 (0.9%) cases under the age of 19 y.
Table 2.--Age and Sex Distribution of Malignant Mesothelioma Cases
Male Female Total
Age (y) n % n % n %
0-9 8 0.6 3 0.5 11 0.6
10-19 4 0.3 2 0.4 6 0.3
20-29 33 2.6 16 2.9 49 2.7
30-39 50 3.9 20 3.6 70 3.8
40-49 149 11.6 53 9.5 202 11.0
50-59 298 23.2 112 20.1 410 22.2
60-69 374 29.1 170 30.5 544 29.5
70-79 275 21.4 129 23.1 404 21.9
80-89 92 7.2 49 8.8 141 7.7
[is greater 2 0.2 4 0.7 6 0.3
than or equal
to] 90
Overall 1,285 100.0 558 100.0 1,843 100.0
Note: The ages of 2 cases and the sex of 1 case were not known.
Among the 1,846 mesothelioma cases, occupational histories were not described for 651 of them. Special attention was not given to asbestos exposure in all mesothelioma cases. The occupations of 1,285 cases were categorized into four groups, relative to asbestos exposure: (1) definite, (2) probable or possible, (3) uncertain, or (4) unlikely (Table 3). Twenty-six cases in the definite group were employed in asbestos-processing plants. Among the 226 cases in the probable or possible group, 96 had worked in the construction industry, 33 had been shipyard workers, 31 metal workers, 24 electricians, 22 railroad workers, 17 pipefitters, and 3 boilermen. Company workers, farmers, and factory workers were included in the uncertain group. Some farmers held side jobs during the "leisure season" for farmers in Japan. On the other hand, there were many cases that belonged to the unlikely group, and asbestos exposure could not be identified only by their occupations. Housewives, schoolteachers, medical doctors, nurses, and clergy were included.
Table 3.--Occupational Histories of Malignant Mesothelioma Cases
Occupational
exposure to No. of
asbestos cases Occupation
Definite 28 Worker in asbestos-processing
plant (n = 26), asbestos insula-
tion worker (n = 2).
Probable or 226 Worker in construction industry
possible (n = 96), shipyard worker (n =
33), metal worker (n = 31), elec-
trician (n = 24), railroad worker
(n = 22), pipefitter (n = 17), boil-
erman (n = 3)
Uncertain 455 Company worker (n = 201), farmer
(n = 79), factory worker (n = 46),
driver (n = 36), self-employed
shop (n = 24), miners (n = 9),
shipman (n = 7), laundryman (n
= 6), other (n = 47)
Unlikely 576 Housewife (n = 228), workers in
food and drink service industry
(n = 50), clerk (n = 47), school
teacher (n = 39), public servant
(n = 27), medical doctor (n =
11), worker in clothes-service
industry (n = 8), nurse (n = 6),
religionist (n = 6), student (n =
6), policeman (n = 6), other (n =
142)
Tumor sites of 1,785 cases were noted in the Annual (Table 4). There were 1,213 (68.0%) pleural mesotheliomas, 431 (24.1%) peritoneal mesotheliomas, and 108 (6.1%) pericardial mesotheliomas. Six (0.3%) mesotheliomas arose from the tunica vaginalis testis. Pleural mesotheliomas were predominant in both sexes; however, the frequency in males (72.7%) was higher than that in females (56.9%). On the contrary, the frequency of peritoneal mesothelioma in females (35.3%) was higher than that in males (19.3%). The male to female ratio for each tumor site was also calculated and appears in Table 4. The ratios (male/female) of the pleural and peritoneal mesotheliomas were 3.0 and 1.3, respectively.
Table 4.--Tumor Sites of Malignant Mesothelioma
Male Female Total
Site n % n % n % Male:
Female
Pleura 906 72.7 306 56.9 1,213(*) 68.0 3.0
Perito- 241 19.3 190 35.3 431 24.1 1.3
neum
Pericar- 76 6.1 32 5.9 108 6.1 2.4
dium
Tunica 6 0.5 -- 6 0.3 --
va-
ginalis
testis
Pleura 8 0.6 1 0.2 9 0.5 8.0
and
peri-
toneum
Pleura 5 0.4 2 0.4 7 0.4 2.5
and
peri-
cardium
Medi- 4 0.3 7 1.3 11 0.6 0.6
astinum
Dia- 1 0.1 -- 1 0.1 --
phragm
Overall 1,247 100.0 538 100.0 1,785 100.0 2.3
(*) The sex of 1 pleural case was not known.
Histological cell types of 598 mesotheliomas were ascertained in the Annual, and there were 245 (41.0%) epithelial, 168 (28.1%) biphasic, and 185 (30.9%) sarcomatous types (Table 5). The frequency of epithelial cell type in peritoneal mesothelioma (52.4%) was higher than those in the other tumor sites.
Table 5.--Histological Cell Type of Malignant Mesotheliomas
Sarcoma-
Epithelial Biphasic tous Total
Site n % n % n % n %
Pleura 155 37.4 115 27.8 144 34.8 414 100.0
Peritoneum 66 52.4 37 29.4 23 18.3 126 100.0
Pericardium 18 48.6 10 27.0 9 24.3 37 100.0
Other 6 28.6 6 28.6 9 42.9 21 100.0
Overall 245 41.0 168 28.1 185 30.9 598 100.0
Of all 1,846 mesothelioma cases, 73 cases were accompanied with asbestosis (Table 6). The prevalence of asbestosis was higher among mesothelioma cases (3.95%; 73/1846) than in the overall data system (0.05%; 525/1056,259). The first peritoneal and pleural mesotheliomas with asbestosis were found in 1976 and in 1978, respectively. The frequency of malignant mesotheliomas with asbestosis (number of cases/total autopsy cases) was 0.001% (5/440,334) for the term 1958-1979, 0.006% (27/390,124) for 1980-1989, and 0.018% (41/225,801) for 1990-1996. There was a significant increase during all three periods ([chi square] = 62.582, p [is less than] .0001). The tumor sites were ascertained in 71 cases, and there were 50 (70.4%) pleural and 18 (25.4%) peritoneal mesotheliomas. Histological cell types of mesotheliomas with asbestosis were obtained in 40 cases. The epithelial cell type was predominant (18 cases [45.0%]); the next most-frequent cell type was biphasic (12 cases [25.0%]), followed by sarcomatous (10 cases [30.0%]).
Table 6.--Description of Malignant Mesotheliomas Accompanied by Asbestosis
No. of cases/
total cases
Year autopsied Percentage
Frequency
1958-1979 5/ 440,334 0.001(*)
1980-1989 27/ 390,124 0.007(*)
1990-1996 41/ 225,801 0.018(*)
1958-1996 73/1,056,259 0.007
Site No. of cases Percentage
Tumor site
Pleura 50 70.4
Peritoneum 18 25.4
Pleura and
peritoneum 2 2.8
Mediastinum 1 1.4
Overall 71 100.0
Cell type No. of cases Percentage
Histological
cell type
Epithelial 18 45.0
Biphasic 12 25.0
Sarcomatous 10 30.0
Overall 40 100.0
(*) [Chi.sup.2] = 62.582, p [is less than] .0001.
Discussion
Takahashi et al.[3] reported that recent mesothelioma incidence among the 10 Western countries and Japan ranged between 5 (Japan) and 25 (Australia), with a median of 15 per one million per year (United States). They speculated that the asbestos consumption curve for Japan in past years lagged behind that for Western countries, and that the cumulative exposure effect has not yet reached the level that can be expected from other countries. In autopsy cases, the frequency (0.17%) of mesothelioma in this survey (1958-1996) was also lower than that of Western countries.[4,5] Hartmann and Schutze[4] reported 106 cases of pleural mesothelioma among 33,500 autopsy cases (0.32%) from 1957 to 1987 in Germany. Giarelli et al.[5] reported 170 cases of malignant pleural mesothelioma among 40,826 autopsy cases (0.42%) during the period 1968-1987 in Italy. Compared with Western countries, there would be a time lag in the increase of mesotheliomas in Japan. The frequency for the term 1990-1996 (i.e., 0.3%) from Japanese autopsies was three times higher than that for 1958-1979 (i.e., 0.1%). Present consumption of asbestos in Japan remains high, and we expect that asbestos-related mesotheliomas will increase in the future.
In our study, the most common site of mesothelioma was pleura, followed by peritoneum, and pericardium. Mesothelioma of the tunica vaginalis testis was very rare. Hillerdal[6] reviewed 4,710 cases of mesothelioma, and 88.8% of the cases were pleural in origin, 9.6% were peritoneal, 0.6% were pleural and peritoneal, 0.7% were pericardial, 0.2% were in the tunica vaginalis testis, and 0.1% occurred at an undetermined site. The frequency (68.0%) of pleural mesothelioma in Japan is relatively low. Perhaps asbestos-related mesothelioma may occur most frequently in pleura; therefore, it may reflect the different cumulative exposure effect between Japan and Western countries. This speculation is supported by the fact that pleural mesothelioma is more common in males than females. Males have more opportunities to be exposed to asbestos by virtue of their occupations. However, in the cases of malignant mesothelioma with asbestosis, the frequency of pleural mesothelioma is not so different from those without asbestosis. The frequency of peritoneal cases in asbestos-related mesotheliomas may be higher in those with asbestosis (i.e., highly exposed asbestos cases) than in the cases with lower asbestos exposure, because a certain quantity of asbestos fibers may need to reach the peritoneum.
Mesotheliomas are commonly diagnosed between the ages of 50 y and 79 y. In our study, however, there were a few cases under 19 y of age. Special attention should be given to childhood mesothelioma, whether or not it is diagnosed correctly. In the 22 childhood cases Fraire et al.[7] reviewed, 45% of them were accepted as mesothelioma, and 14% were of "doubtful or uncertain nature." In the remaining 41%, the diagnosis of mesothelioma was rejected. Given that the diagnosis of mesothelioma in recent years has become more accurate as a result of the development of immunohistochemical staining, investigators must accumulate more data concerning mesothelioma and its relationship with asbestos exposure. A few childhood mesothelioma patients were reportedly exposed to asbestos; however, most cases had no such exposure.[7] It is said that asbestos-related mesothelioma needs a long latency period from the beginning of the asbestos exposure to the time the tumor is manifested. Therefore, the existence of childhood mesothelioma may be one reason why researchers do not attribute all mesotheliomas to asbestos exposure.
The most common etiologic factor of mesothelioma is asbestos. Other factors reported are erionite,[8] radiation,[9] chronic inflammation,[10] among others.[11] Investigators confirmed by animal experiment that SV40 virus caused mesothelioma in the hamster.[12] McDonald and McDonald[13] reported that in approximately one-half of the male mesothelioma cases, no potential for occupational asbestos exposure could be identified. Other researchers have reported various percentages (i.e., 0-87%) of cases for which asbestos exposure was not documented.[14] In our earlier report,[15] we evaluated asbestos body counts in the lungs of 27 malignant mesothelioma cases, and approximately one-half of mesothelioma cases had counts of more than 151 asbestos bodies per 5 gram wet lung tissue, thus indicating occupational asbestos exposure. In the current study, the frequency of the patients who were clearly identified as having asbestos exposure was not very high. However, we note that all occupational records related to asbestos exposure have not been described in the Annuals. In addition, the absence of proof of asbestos exposure does not constitute a lack of probability that the epidemic of mesothelioma in Japan has resulted from asbestos. One must consider that asbestos-induced malignant mesotheliomas may result from nonoccupational exposure to asbestos. It is reported that bystander and consumer exposures are the "third wave" of asbestos disease playing out now.[16]
Three main histological types are recognized in mesothelioma: (1) epithelial, (2) sarcomatous, and (3) biphasic. Hillerdal[6] reported that the epithelial type (50%) was the most common in 829 pleural mesothelioma cases; the next most-frequent type was biphasic (34%), followed by sarcomatous (16%). Roggli et al.[17] reported the following histological cell types from 200 cases of malignant pleural mesothelioma: epithelial (40%), biphasic (37%), and sarcomatous (23%). In our survey, the histological type that occurred most frequently (41.0%) was epithelial; however, the frequency of biphasic (28.1%) and sarcomatous (30.9%) cell types was the opposite of that reported by Roggli et al.[17]
Whether specific histological cell types of mesotheliomas are related to asbestos exposure remains uncertain. Magner and McDonald[18] suggested that mixed mesothelial tumors were specifically associated with asbestos exposure. Leigh et al.[19] reported there was a significant correlation in lung fiber content with cell type from epithelial (i.e., low fiber content) through mixed to sarcomatous (i.e., high fiber content). However, Hirsh et al.[20] reported that the proportion of epithelial mesothelioma type in the group with definite past asbestos exposure was higher than that in the group that was unexposed to asbestos. On the other hand, Wright and Sherwin[21] concluded that their data did not support the hypothesis that any specific histological type of mesothelioma is especially related to asbestos exposure. Although only mesotheliomas with asbestosis were selected in this study, our data support the conclusion posited by Wright and Sherwin.[21]
The authors thank all of those who contributed to the Annual of the Pathological Autopsy Cases in Japan.
References
[1.] Wagner JC, Sleggs CA, Marchand P. Diffuse pleural mesothelioma and asbestos exposure in the North Western Cape Province. Br J Ind Med 1960; 17:260-71.
[2.] Baba K. Indications of an increase of occupational pleural mesothelioma in Japan. J Univ Occup Environ Health 1983; 5:3-15.
[3.] Takahashi K, Huuskonen MS, Tossavainen A, et al. Ecological relationship between mesothelioma incidence/mortality and asbestos consumption in ten Western countries and Japan. J Occup Health 1999; 41:8-11.
[4.] Hartmann CA, Schutze H. Mesothelioma-like tumors of the pleura: a review of 72 autopsy cases. J Cancer Res Clin Oncol 1994; 120:331-47.
[5.] Giarelli L, Bianchi C, Grandi G. Malignant mesothelioma of the pleura in Trieste, Italy. Am J Ind Med 1992; 22:521-30.
[6.] Hillerdal G. Malignant mesothelioma 1982: review of 4710 published cases. Br J Dis Chest 1983; 77:321-43.
[7.] Fraire AE, Cooper S, Greengerg SD, et al. Mesothelioma of childhood. Cancer 1988; 62:838-47.
[8.] Selcuk ZT, Coplu L, Kalyoncu AF, et al. Malignant pleural mesothelioma due to environmental mineral fibre exposure in Turkey: analysis of 135 cases. Chest 1992; 102:790-96.
[9.] Shannon VR, Nesbitt JC, Libshitz HI. Malignant pleural mesothelioma after radiation therapy for breast cancer: a report of two additional patients. Cancer 1995; 76:437-41.
[10.] Hillerdal G, Berg J. Malignant mesothelioma secondary to chronic inflammation and old scars: two new cases and review of the literature. Cancer 1985; 55:1968-72.
[11.] Attanoos RL, Gibbs AR. Pathology of malignant mesothelioma. Histopathology 1997; 30:403-18.
[12.] Cicala C, Pompetti F, Carbone M. SV40 induces mesotheliomas in hamsters. Am J Pathol 1993; 142:1524-33.
[13.] McDonald AD, McDonald JC. Malignant mesothelioma in North America. Cancer 1980; 46:1650-56.
[14.] Peterson JT, Greenberg SD, Buffler PA. Non-asbestos-related malignant mesothelioma: a review. Cancer 1984; 54:951-60.
[15.] Murai Y, Kitagawa M. Asbestos fiber analysis in 27 malignant mesothelioma cases. Am J Ind Med 1992; 22:193-207.
[16.] Landrigan PJ, Kazemi H (Eds). The Third Wave of Asbestos Disease: Exposure to Asbestos in Place: Public Health Control. Annuals of the New York Academy of Sciences. New York: The New York Academy of Sciences, 1991; pp 1-628, vol 643.
[17.] Roggli VL, Sanfilippo F, Schelburne JD. Mesothelioma. In: Roggli VL, Greenberg SD, Pratt PC (Eds). Pathology of Asbestos-Associated Diseases. Boston/Toronto/London: Little, Brown and Co., 1992; pp 109-64.
[18.] Magner D, McDonald AD, Malignant mesothelial tumors--histologic type and asbestos exposure. N Eng J Med 1972; 287: 570-71.
[19.] Leigh J, Rogers AJ, Ferguson DA, et al. Lung asbestos fiber content and mesothelioma cell type, site, and survival. 1991; 68:135-41.
[20.] Hirsch A, Brochard P, De Cremoux H, et al. Features of asbestos-exposed and unexposed mesothelioma. Am J Ind Med 1982; 3:413-22.
[21.] Wright WE, Sherwin RP. Histological types of malignant mesothelioma and asbestos exposure. Br J Ind Med 1984; 41:514-17.
YOSHIHIRO MURAI Department of Pathology Faculty of Medicine Toyama Medical and Pharmaceutical University Toyama, Japan
Submitted for publication July 21, 1999; revised; accepted for publication March 29, 2000.
Requests for reprints should be sent to Dr. Yoshihiro Murai, Department of Pathology, Faculty of Medicine, Toyama Medical and Pharmaceutical University, 2630 Sugitani, Toyama, 930-0194 Japan.