Document Qk62y3D8LBn296ego760zgwBo

Analysis of Asbestos Fiber Burden in Lung Tissue from Mesothelioma Patients Ronald F. Dodson, PhD Itfichael O'Sullivan, BS Carolyn J. Corn, BS Deoartment of Cell Biology and Environmental Sciences, The University of Texas Health Center at Tyler, Tyler, Texas, USA Jerry W. McLarty, PhD Department of Epidemiology and Biomathematics, The University of Texas Health Center at Tyler, Tyler, Texas, USA Samuel P. Hammar, MD Diagnostic Specialties Laboratory, Bremerton, Washington, USA Mesothelioma is a rare neoplasm that occurs most frequently in individuals with previous asbestos exposure. Differences for risk of development of asbestos-related mesothelioma and lung cancer have been attributed to the various types of as bestos, as well as to the dimension of the inhaled fibers. In the present study, 55 individuals with the pathological diagno sis of mesothelioma were evaluated as to ferruginous body and fiber content in lung tissue. The procedures used in the analysis included tissue digestion and analysis of the collected material for ferruginous bodies by light microscopy and for uncoated fibers by analytical transmission electron micros copy. Forty-six of the samples had ferruginous body concen trations of over 1000/per gram dry weight of lung tissue. The majority of the cores of these ferruginous bodies were amosite. Likewise, the most common uncoated asbestos fiber in the tissue was amosite. Only a small percentage of each type of asbestos would have been visible by light microscopy or even potentially by electron microscopy if the magnification was not sufficient to detect those with thin (<0.2 pm) diameters. The consistent finding in most of the cases was a consider able presence of asbestos, often of mixed types. Keywords asbestos, electron microscopy, ferruginous bod ies, mesothelioma The relationship between exposure to asbestos and the risk of developing mesothelioma was es tablished through the observations of Wagner in South Africa [1]. Additional reports involving other asbestos-exposed cohorts have further supported this observation. These diverse exposures occur during mining operations [2-4], to individuals working with asbestos-containing products, or with environmental exposures to asbestos-containing dust [2, 5-8], The most common site for develop ment of this rare tumor is in the pleural cavity, al though approximately 10% arise in the peritoneal cavity [9], In spite of greater worldwide awareness of risks from asbestos exposure, recent reviews pointed out that asbestos-associated cancers con tinue to increase in incidence [10, 11]. Chrysotile is the most widely used asbestos, ac counting for 97% of the world's production in 1976 [12], It is reasonable, therefore, that various expo sure levels from commercial applications should most often be to this type of asbestos. Even so, it Received 5 November 1996; accepted 6 February 1997. Address correspondence to Ronald F. Dodson, PhD, Professor and Chairman, Department of Cell Biology and Environmental Sci ences, The University of Texas Health Center at Tyler, P. 0. Box 2003, Tyler, TX 75/; 0. USA. has been suggested that in the risk for develop ment of mesothelioma, differences exist between exposures to various amphiboles and to chrysotile [2]. Crocidolite has been reported as having the most potential for causing tumors, followed by amosite, with chrysotile being the least tumorigenic [1, 2, 4]. An increased awareness of the presence of the amphibole tremolite as a component of some chrysotile veins, as well as other mineral de posits, has raised the issue that it may have a major role in the development of mesothelioma following exposure to products made from these minerals [13, 14]. These concepts have given rise to an "am phibole hypothesis" as a basis for explaining the risk of developing an asbestos-related malignancy [15]. In reality, many workplace exposures have been to a mixture of asbestiform minerals, as well as to other fibrous and nonfibrous dusts [16]. The potential for developing lung cancer from asbestos exposure is often confounded with envi ronmental exposures such as cigarette smoking. The risk of mesothelioma induction in humans and animals is more exclusively linked to the inhalation of asbestos fibers. Considerable data on the risks of developing mesothelioma from occupational set tings have come from historical reviews of various exposed cohorts. Likewise, tissue burden analysis Ultrastructural Pathology, 21:321-336, 1997 Copyright 1997 Taylor & Francis 0191-3123/97 $12.00 + .00 321 322 and a correlation of these findings have occurred with groups in which asbestos exposure was part of their daily work activities, such as miners, millers, or workers involved with the manufacture of asbes tos-containing products. Much less is known about the quantification of ferruginous bodies and un coated asbestos fibers in individuals with mesothe lioma where exposures were to asbestos-con taining end products in the workplace. The data from tissue analysis are further limited because few studies have been carried out by transmission electron microscopy (TEM) at a suffi ciently high magnification to permit a counting scheme that includes short (3=0.5 pm) as well as long, thin fibers. The inclusion of such fibers is nec essary for an accurate assessment of uncoated fi ber burden. Numbers of chrysotile fibrils, as well as thinner amphibole fibers, could also be ignored by TEM if the counting protocol includes only longer fibers (3=5 pm) and does not use a sufficiently high magnification to compensate for these thin "fibrils." This may be particularly important in me sothelioma, since these are the geometrically de fined fibers shown to reach extrapulmonary sites, including the lymph nodes and pleural plaques [17] . The primary purpose of the present study is to characterize the fiber burden in tissue in a series of mesothelioma cases. Unlike most previous studies, this group had diverse exposures ranging from shipyards, where some amphibole exposures would be expected, to vocations where the more likely exposure would be to chrysotile such as ce ment production and brake shoe factories [12]. With this mixed cohort, the questions to be an swered are (1) What relationship exists between the chrysotile and amphibole burden in individuals with mesothelioma? (2) Which amphibole was most commonly encountered? and (3) What are the length and width characteristics of the asbestos fi bers in these cases of mesothelioma? This final question is of particular interest, since a recent re port suggests there may be a relationship between low-aspect-ratio amosite fibers and mesothelioma [18] . This is in contrast with the theory that long, thin fibers [12, 19-23] are reported to be the most tumorigenically active of the inhaled fibers, and raises further issues as to the importance of the shorter fibers as contributors to such disease. The data collected in this study also provided an opportunity to evaluate other aspects of mesothe lioma, including latency period, length of survival, and influence of histologic type on survival time. In addition, each case was evaluated for whether asbestosis was present, and this was correlated with asbestos body and asbestos fiber concentrations in lung tissue. MATERIALS AND METHODS A cohort of 55 individuals with a pathological diag nosis of mesothelioma (made by one of the co R. F. Dodson et al. authors, SH) was selected for analysis and charac terization as to the types of asbestos fibers within their lung tissue. All of the cases included in this study either had complete or partial autopsies (47 cases) or had lungs and pleural tissue harvested (8 cases). The diagnosis of mesothelioma in each case was made by established criteria, including immunohistochemistry, and electron microscopy [24, 25]. His tologically, the mesotheliomas were grouped into four major categories: (1) epithelial, (2) biphasic, (3) sarcomatoid, and (4) desmoplastic, though other histologic subtypes of mesothelioma have been recognized [24, 25]. Lung tissue from patients was evaluated by one of the authors (SH) for pathologic asbestosis and graded according to the College of American Pa thologists and the National Institute for Occupa tional Safety and Health criteria [26]. An approxi mate 5-g sample from a site in each lobe was also collected for a digestion procedure in the Bremer ton Laboratory [27] and determination of ferrugi nous bodies per gram wet lung. All cases in the cohort were from the Pacific Northwest and most had exposures associated with shipyard-related activities. The historical data listed in Table 1A include sex, age, smoking his tory, and occupation; Table IB has other medical history; and Table 1C lists dates of first and last exposure and dates of diagnosis and death. The lung tissue preparation for light and electron microscopy as carried out in the Tyler Laboratory consisted of the Williams' modified bleach proce dure [28], Multiple sites from tumor-free, formalinfixed lung parenchyma were sectioned from each lung for quantitative analysis. Two separate digest pools were done on each case, one from the right lung and one from the left. The right digest pool contained an average of 0.3478 g dry weight of lung tissue (range 0.8403 to 0.0970) and the left digest pool contained an average of 0.3276 g dry weight of lung tissue (range 0.6798 to 0.0801). These pools were processed through the modified bleach (9% sodium hypochlorite) digestion proce dure of Williams et al. [28]. Dry weights were cal culated based on a wet/dry ratio obtained for each digest pool. All reagents were prefiltered through 0.2-pm pored polycarbonate filters. Aliquots of the pools were collected on 0.2-pm pored polycarbon ate filters for analytical transmission electron mi croscopy (ATEM) and on 0.22-pm pored mixed cel lulose ester filters for quantitation of ferruginous bodies by light microscopy. Mixed cellulose ester filters were used to evalu ate the ferruginous body content. These filters con tained an average of 0.0329 g dry weight of lung tissue (range 0.0660 to 0.0068). One-fourth of the filter was collapsed in acetone vapor (making it transparent) and evaluated by light microscopy at xlOO, x200, and x400. The ferruginous bodies that were noted as being typical in morphology (beaded coat, reddish-brown color, and clear core) were Asbestos Burden in Cases of Mesothelioma 323 TABLE 1A Historical data for mesothelioma cases Assay number Sex Age3 Packs/yea rb Asbestosis/grade Work history3 L-1 L-2 L-3 L-4 L-5 L-6 L-7 L-8 L-9 L-10 L-11 L-12 L-13 L-14 L-15 L-16 L-17 L-18 L-19 L-20 L-21 L-22 L-23 L-2 4 L-25 L-26 L-27 L-2 8 L-29 L-30 L-31 L-32 L-33 L-34 L-35 L-36 L-37 L-38 L-3S L-40 L-41 L-42 L-43 L-44 L-45 L-46 L-47 L-48 L-49 L-60 L-51 L-5 2 L-53 L-54 L-55 M 64 M 69 M 58 M 82 M 73 M 62 M 85 M 63 M 57 M 60 MU M 75 M 71 M 67 M 73 MU M 76 M 60 M 72 MU M 71 M 76 M 70 M 51 M 72 M 62 M 74 M 74 M 83 M 65 M 71 F 67 M 70 M 69 M 65 M 67 M 73 M 55 M 75 M 72 M 43 M 62 M 82 M 74 M 79 M 65 M 87 M 82 M 66 M 77 M 60 M 75 M 66 M 74 F 74 NS U 32 50 NS NS 13 29 NS 2 37 38 50 43 4 NS 7 S(pipe) 20 10 30 S(pipe) 28 U NS 50 30-50 NS 16 16 100 23 35 35 U NS 74 U U 15 13 NS NS 4 52 60 NS 7-10 22 2 S 45 11 53 15 Y/1 N Y/1 Y/1 Y/1 N N Y/1 Y/1 N Y/1 Y/1 N Y/1 Y/1 N Y/1-3 N Y/1 N Y/1 N Y/1 Y/1 N N Y/1-3 Y/1 Y/1 N N N U N N N N N Y/1-2 Y/1 N Y/1 Y/1 Y/1-2 Y/1 N N Y/1-2 Y/1 Y/1 Y/3 N Y/1 N N Shipfitter, insulator asbestos exposed Unknown, asbestos exposed Shipyard insulator Electrical engineer Contractor (23), sheet metal (7), shipyard (1) Machinist (32), clerk (3), Navy (2) Shipyard painter (25), railroad brakeman (25) Carpenter (28), Navy (1), machinist (1) Electrical contractor (24), boilermaker (6), Army other (4) Drywall, janitor Asbestos pipe/brick (20), boilermaker (3) Painter Tool engineer/machinist/plant manager/mechanic (50), other (4) Shipyard (2) Electrician, laborer Civil and structural engineer Shipyard WWII, pipe fitter Contractor (28), seaman/master (9) Boilermaker Sheet metal/shipyard (5), sales Unknown Shipyard inspector, construction Railroad boilermaker Plumber/pipe fitter (13), construction (4) Ship refrigeration/Asbestos removal (28) Shipyard worker (2) Steel mill sales (15), boiiermaker/railroad (6) Merchant Marine Shipfitter ' Orchard work, boiler room welder (1.5) Cement mason (25) Electronic components assembler (11) Insulator (36) Shipyard, ship building and repair Unknown Shipyard pipe fitter Packing plant (used asbestos wool) Spraying asbestos Unknown Pipe fitter (1), machinist/shipyard (26) Laborer/construction Construction (14), concrete (6), asbestos tile (3) Sheet metal worker, furnace installer Shipyard insulator and pipe coverer Electrician (4) Shipyard machinist Shipyard rigger, crane operator Shipyard electrician (23) Pipefitter/shipfitter Shipyard painter Unknown Navy/shipyard (20) Electrician Unknown Brake shoe factory (2) Note. U and None, unknown or unavailable; Y, yes; N, no. 3Age at time of death. 3NS for nonsmoker, U for unknown, S for smoker listed as number of packs/year. cJob and (number of years worked). 324 R. F. Dodscri et al. TABLE IB Historical data for mesothelioma cases Assay Location of number mesothelioma Histological type3 Parietal Location of plaque metastases Medical history, other L-1 Left pleura Epith. L-2 Right pleura Mixed biph./sarco. L-3 Peritoneum Epith. L-4 Left pleura Mixed biph./sarco. L-5 Peritoneum Epith. L-6 Right pleura Biph. L-7 Right pleura Epith. Y Y Y Y Y N Y L-8 L-9 L-10 L-11 L-12 L-13 L-14 L-15 L-16 L-17 L-18 L-19 L-20 L-21 L-22 L-23 L-24 L-25 L-26 Left pleura Right pleura Right pleura Left pleura Left pleura Left pleura Peritoneum Right pleura Right pleura Right pleura Right pleura Right pleura Right pleura Left pleura Right pleura Left pleura Right pleura Peritoneum Right pleura Sarco. Epith. Epith. ' Sarco. Mixed desmo./sarco. Epith. Y Y N U Y Y Epith. Sarco./desmo. Y Y Poorly differentiated epith. U Mixed desmo./sarco. Y Sarco. Epith. Y Y Sarco. Biph. Y Y Mixed sarco./desmo U Variable differentiation Y Variable differentiation Mixed epith./sarco. Sarco. Y N Y L-27 L-2 8 L-29 L-30 L-31 L-32 L-33 L-34 L-35 L-36 L-37 L-38 L-39 Right pleura Left pleura Left pleura Right pleura Right pleura Right pleura Right pleura Left pleura Left pleura Right pleura Left pleura Peritoneum Right pleura Sarco. Epith. Mixed pleomorphic Epith. Epith. ' Pleomorphic Epith. Epith./ tubulopapillary Poorly differentiated Biph. Epith. Epith. Epith. Y U Y Y Y U Y Y U Y Y Y Y Peritoneum, peripancreatic, lymph node Mesentery Umbilicus Peritoneal Abdominal wall Liver, pericardium. Right lung Left visceral pleura. Left lung, lymph nodes None Peritoneal, periaortic Left lung, lymph nodes U Abdominal serosa U None U U None None Lymph nodes, serosa of stomach U None None Nodes, liver, spleen. Right Lung Right lung None Mesentery, adrenals. Right pleura Liver, peritoneum, spleen, node Right lung, lymph nodes Left lung, pericardium, liver, spleen Lymph nodes, Left chest wall Lymph nodes. Left pleura. Left Lung U Left visceral and parietal pleura Lymph nodes Right lung, epicardium Left lung, liver, adrenal, kidneys Hilar/bronchopulmonary None Left lung, pleura, liver, lymph nodes Cardiovascular, pneumonia Thymoma, prostate adenocarcinoma Cardiovascular None Hypertension, pericarditis Cardiovascular, diverticulosis Bronchitis, pneumonia Bronchitis, pneumonia Bronchitis, pneumonia Cardiovascular Chronic obstructive pulmonary disease Pulmonary thromboembolic Bronchitis, prostate adenocarcinoma Chronic obstructive pulmonary disease, diabetes Fibrosis, cardiovascular Bronchitis, pneumonia Bronchitis, pneumonia, alveolar damage Bronchitis, pneumonia, alveolar damage Emphysema, atherosclerosis None None Bronchitis, pneumonia, arteriosclerotic/ cardiovascular Bronchitis, pneumonia, emphysema Arteriosclerotic/ cardiovascular Arteriosclerotic/ cardiovascular Arteriosclerotic/ cardiovascular Arteriosclerotic/ cardiovascular None Arteriosclerotic/ cardiovascular Arteriosclerotic/ cardiovascular None None None None None Asbestos Burden in Cases of Mesothelioma TABLE IB (Continued) 325 Assay number L-40 L-41 L-42 L-43 L-44 L-45 L-46 L-47 L-48 L-49 L-50 L-51 L-52 L-53 L-54 L-55 Location of mesothelioma Right pleura Right pleura Right pleura Left pleura U Right pleura Left pleura Right pleura Right pleura Left pleura Right pleura Right pleura Right pleura Left pleura Left pleura Left pleura Histological type Mixed sarco./biphasic Sarco. Mixed sarco./desmo. Epith. Epith. Sarco. Biph. Biph. U Epith. Sarco. Epith. Epith. Epith. Mixed desmo./sarco. Epith./tubulopapillary Parietal plaque U N U Y U Y Y Y Y Y Y Y Probable Y Y N Location of metastases None U Epicardium Left and Right lung. abdominal cavity U Pancreas, liver. subcut. tissue Lymph node, pericardium. myocardium Hilar, lymph nodes. Right lung U Left Lung Adrenal gland, heart. mesentery U U Hilar, lymph nodes. mesentery Omentum, myocardium, lymph nodes Left lung. vertebra, psoas muscle Medical history, other None None None None None None None None None None None ' None None None Emphysema None Note. Y, yes; N, no; U, unknown or unavailable. Biph., biphasic; Sarco., sarcomatoid; Desmo., desmoplastic; Epith., Epithelial. considered asbestos bodies and thus a component of the asbestos burden. For each case, results from right and left sides were averaged and reported as ferruginous bodies per gram dry weight of lung tissue. The polycarbonate filters contained an average of 0.00338 gram dry weight of lung tissue (range 0.0132 to 0.0019). They were examined by light mi croscopy at xlOO, x200, and x400, and ferruginous bodies were evaluated. Filters were prepared via direct method for ATEM. By this method, the filters were carbon coated, cut into 3-mm2 pieces, and mounted on 100-mesh copper grids. A modified Jaffe-Wick method was used to remove the filter matrix, leaving a carbon film containing the en trapped fibers, ferruginous bodies, and other par ticulates [29]. Fiber quantitation and ferruginous body core analysis were done by ATEM. Qualitycontrol procedures, in addition to the prefiltering of reagents, included analysis of selected filters from the lot used in the procedures, as well as reagent/ filter blanks. The data from these analyses were used to establish background for comparative pur poses. The grids were evaluated for continuity and overall quality of the films at x20,000 or at a lower magnification of x2,500 in a JEOL 1200EX, which was interfaced with either a TN-5500 X-ray analyzer or an EDAX NX-2 X-ray analyzer. Uncoated fibers ?=0.5 pm long having parallel sides for a majority of their length and with a length/width ratio greater than 3:1 were counted. Either the first 100 fibers in a completed opening or all the fibers in 10 open ings were counted from each of 3 grids, which is equivalent to 1.32 mm2. In addition, the remainder of the grids were scanned at the lower magnifica tion for ferruginous bodies. When ferruginous bod ies were found, the cores were analyzed by ATEM. These were characterized by X-ray energy disper sive spectrometry (XEDS) and selected area elec tron diffraction (SAED). Length/width, composition, and various ratios were noted for the asbestos fi bers. Results were determined as fibers per gram dry weight of lung tissue for total asbestos, as well as for individual fiber types. To provide further compensation for random sampling, the data are reported as a weighted average obtained from the two sides (Table 2). Fiber concentrations were found to have highly skewed distributions, as illustrated in Figure 1, for all fiber types; therefore, medians were used as the usual summary measures of central ten dency. Fiber lengths, widths, and aspect ratios were all summarized in terms of geometric means. The nonnormal distributions of the data dictated that nonparametric statistical methods, such as Fisher's exact for 2 x 2 tables, and the Wilcoxon and Mann-Whitney tests for comparisons of fiber concentrations, lengths, widths, and aspect ratios, be used. 326 TABLE 1C Historical data for mesothelioma cases Assay number Year of first exposure Year of last exposure Date of diagnosis L-1 L-2 L-3 L-4 L-5 L-6 L-7 L-8 L-9 L-10 L-11 L-12 L-13 L-H L-15 L-16 L-17 L-18 L-19 L-20 L-21 L-22 L-23 L-24 L-25 L-2 6 L-27 L-28 L-29 L-30 L-31 L-32 L-33 L-34 L-35 L-36 L-37 L-38 L-39 L-40 L-41 L-42 L-43 L-44 L-45 L-46 L-47 L-48 L-49 L-50 L-51 L-52 L-53 L-54 L-55 1956 U 1954 1942 1937 1945 1921 1945 1960 1959 1942 1941 1939 1951 1940 1955 1941 1938 1939 1942 1941 1941 1940 1961 . 1946 1946 1939 1943 1941 1942 1945 1951-1959 1947 1942 U 1945 1937 1948 U 1941 1966 1961 1916 1942 1941 1954 1940 1942 1943 1941 U 1945 1942 1935 1929 1975 U 1962 1973 1946 1984 1971 1976 1990 1960 1958 , 1976 1978 1953 1970 1992 1945 1988 1972 1947 1946 1945 1954 1974 1982 1965 1980 1951 1949 1975 1981 1970 1983 1974 U 1980 1938 1948(3 Mon) U 1974 1980 1987 1971 1949 1945 1980 1941' 1955 1971 1970 U 1965 1944 1946 1931 07-91 11-91 07-92 01-92 07-91 10-90 07-91 10-91 12-90 01-90 08-92 04-91 01-92 01-93 09-92 03-92 01-93 07-92 01-93 01-93 10-86 08-87 04-90 11-90 11-90 03-91 04-90 10-90 08-90 05-88 09-85 12-87 05-87 12-85 U 02-87 09-87 07-87 U 03-93 09-91 12-87 05-87 10-87 08-86 03-89 05-89 03-90 10-88 03-90 02-87 07-87 02-87 01-89 03-86 Note. U, unavailable. 8Month-year. bLatency period = date of diagnosis minus date of first exposure. Latency period (years)b 35 U 38 50 54 45 70 46 30 31 50 50 53 42 30 37 52 54 54 51 45 46 50 29 44 45 51 47 49 46 40 36 40 43 U 42 50 41 U 52 25 46 71 45 45 35 49 48 45 39 U 42 U 54 57 R, F. Dodson et al. Date of death 11-92 12-92 11-92 05-92 10-91 04-91 11-91 10-91 10-91 04-91 01-93 04-91 05-93 03-93 01-93 12-92 04-93 04-93 07-93 05-93 10-87 10-87 11-90 08-91 02-91 05-91 09-90 10-91 08-91 01-90 11-87 02-89 12-87 02-88 07-84 11-87 01-88 01-88 05-88 09-93 10-92 04-88 07-88 03-88 07-87 06-89 09-89 02-91 06-90 04-90 10-87 08-87 10-87 06-89 06-87 Survival from date of diagnosis (months) 16 13 4 4 3 7 4 4 10 14 5 1 wk 16 2 4 9 3 9 6 4 12 2 7 8 2Vi 2 5Vz 12 12 19 26 12 7 25 U 9 5 6 U 5 12 4 14 5 12 3 4 10 20 1 8 2 8 5 15 Asbestos Burden in Cases of Mesothelioma TABLE 2 Total uncoated asbestos fibers and ferruginous bodies per case Total fibers/ g dry FB/g drya FB/g weta Fiber/ FB's ratiob Assay number 68,950,000 63,417,431 32,812,500 13,625,806 6,875,000 5,701,613 4,136,364 3,999,548 3,989,464 2,925,000 2,722,222 2,581,967 2,230,392 2,119,141 1,894,587 1,682,692 1,563,333 1,439,394 1,422,764 1,045,300 1,014,493 928,030 923,261 899,306 855,263 758,808 740,385 698,357 694,444 665,205 606,436 588,235 559,764 505,556 490,924 446,697 433,559 391,791 352,011 331,439 318,627 291,667 239,726 222,458 217,662 153,091 150,538 131,579 112,179 72,016 32,500 43,532 43,532 22,096 0 1,122,333 1,248,686 683,269 1,008,031 227,787 150,670 170,082 85,783 34,456 38,930 45,114 33,704 83,524 97,618 181,712 21,077 21,278 43,627 38,447 2,686 4,635 17,984 8,851 3,620 81 3,571 4,304 107,211 3,089 9,564 134,634 268 7,849 17,891 13,425 2,867 4,115 36,266 1,938 8,591 7,718 16,609 3,993 0 1,764 2,082 1,240 965 347 3,762 9,699 281 60 365 0 201,416 170,985 114,788 195,419 31,739 29,077 27,780 13,114 4,903 6,832 7,219 5,234 14,116 15,523 26,464 4,100 3,325 8,391 3,952 458 528 2,968 1,536 546 13 547 699 18,968 488 2,114 17,677 40 971 3,338 1,631 400 759 6,074 279 1,409 1,148 2,542 737 0 299 325 240 160 60 756 1,737 46 10 60 0 61 51 48 14 30 38 24 47 116 75 60 77 27 22 10 80 73 33 37 389 219 52 104 248 10,559 212 172 7 225 70 5 2,195 71 28 37 156 105 11 182 39 41 18 60 0 123 76 121 136 323 19 6 155 726 61 0 L-33 L-38 L-44 L-3 L-48 L-9 L-11 L-19 L-14 L-52 L-17 L-22 L-53 L-36 L-15 L-43 L-23 L-7 L-34 L-40 L-42 L-51 L-37 L-20 L-35 L-28 L-41 L-1 L-55 L-21 L-50 L-26 L-24 L-5 L-39 L-16 L-47 L-49 L-30 L-46 L-29 L-12 L-4 L-18 L-45 L-32 L-31 L-54 L-27 L-8 L-2 L-6 L-10 L-25 L-13 aF3/g determined on digestion pool from multiple sites of right and left lung by Tyier laboratory. b3ased on dry weight per gram. 327 FIG. 1 Tremolite fiber loads: the number of par tients having fiber toads (fibers per g) of tremolite for a range of fiber loads. RESULTS Because of the extensiveness and complexity of the data, the sections are discussed as Medical Evaluation (Tables 1A-C), Ferruginous Body (FB) Data (Table 2), and Uncoated Fiber Data (Table 3). When the term "ferruginous body" is used in the text to describe structures as seen by light micros copy, it implies a morphology consistent with an asbestos body. Medical Evaluation Table 1A There were 53 males and 2 females in the study and the mean age,at time of death was 69.5 years (range 43 to 87). Thirty-eight were smokers, 11 were nonsmokers, and smoking history was un available for 6. The majority of the cases had occu pational exposures related to shipyard activities. The specific type of job and number of years at that job are listed on Table 1A. Table IB The majority of the cases (49) had pleural mesothe lioma; (30 were right pleura and 19 were left pleura), and there were 5 cases of peritoneal me sothelioma. Location of tumor was unavailable on 1 of the cases. Pleural plaques were found in 39 of 44 cases. This information was unavailable in 11 cases. The mesothelioma had metastasized in 35 cases. No metastases were found in 9 cases and data were unavailable in 11 cases. The histological type of the tumor is listed in Table IB. 328 Table 1C Latency period for the purposes of this study was defined as the difference between date of diagnosis and date of first exposure. While the latency period ranged from 25 to 71 years, the mean was 45 years, with 45 and 50 years being most frequent. Latency information was not available in 6 cases. Survival from date of diagnosis ranged from 1 week to 26 months and had a mean of 8.3 months. Survival information was unavailable in 2 cases. Survival was also determined according to his tologic type of mesothelioma. Twenty-five epithe lial mesotheliomas, which included four peritoneal mesotheliomas, had a mean survival of 10.64 months, with a range of 2-26 months. When ex cluding the four peritoneal mesotheliomas, the mean survival was 11.95 months. Eleven sarcoma toid mesotheliomas had a mean survival of 6.14 months, with a range of 1-12 months. Eight biphasic mesotheliomas had a mean survival of 6.81 months, with a range of 2.5-12 months. Four me sotheliomas exhibited variable differentiation, showing more than four histologic patterns. These four mesotheliomas had a mean survival of 7.38 months, with a range of 2.5-12 months. One of the mesotheliomas showing variable differentiation was a peritoneal mesothelioma. When that case is excluded, the mean survival was 9 months, with a range of 7-12 months. Six desmoplastic mesothe liomas had a mean survival of 3.04 months, with a range of 1 week to 5 months. Five peritoneal me sotheliomas (four epithelial, one showing variable differentiation) had a mean survival of 3.5 months, with a range of 2-6 months. Ferruginous Body Data From Tyler Laboratory As described in methodology, aliquots from the right and left lung were evaluated for ferruginous bodies. To compensate for random sampling, the .data represent an average between the 2 aliquots from each patient (Table 2). Forty-six of the 55 patients (83.6%) were found to have an average ferruginous body burden of greater than 1,000 fer ruginous bodies per gram of dry weight of lung tissue, a concentration often quoted as represent ing occupational levels of tissue burden [30]. This is clearly above the <20 ferruginous bodies per gram of wet lung tissue reported for the general popula tion in studies by Roggli et al. [31, 32], Breedin and Buos [33], and Dodson et al. [34], The highest con centration was 1,248,686 ferruginous bodies per gram dry of lung tissue (case L-38), while the mean in this group of 46 cases was 125,567 ferruginous bodies per gram dry of lung tissue. A subpopula tion of the group (4 out of 55, 7.3%) was found to have ferruginous body levels within the range re ported in general populations (<20 ferruginous bodies per gram wet weight of lung tissue) [33, 34], The tissue from 6 patients contained a range of R. F. Dodson et al. from 60 to 965 ferruginous bodies per gram dry weight of lung tissue. Tissue from 2 individuals (L13, L-18) did not have ferruginous bodies as de fined within the limit of detection inherent in the methods used in the light microscopy analysis (10 35 ferruginous bodies per gram of wet lung tissue and 85 to 240 ferruginous bodies per gram of dry lung tissue). Core analysis of the 841 ferruginous bodies found by ATEM indicated that 781 (92.9%) were formed on amosite, 24 (2.9%) on crocidolite, 8 (1.0%) on tremolite, 3 (0.4%) on anthophyllite, 3 (0.4%) on actinolite, and 1 (0.1%) formed on chrysotile. In addition, 11 (1.3%) were formed on nonas bestos fibers, and 10 (1.2%) were either totally coated or successful analysis of the core material could not be achieved. There was considerable variation in the ratio of uncoated fiber to ferrugi nous body burden. These ranged from 5:1 to as high as 10,559:1 (Table 2). Uncoated Fiber Data The uncoated fiber data were based on analysis by XEDS, SAED, as well as measurement of length and width of 3,244 asbestos fibers. The total un coated asbestos burden in the patients ranged from a high of 68,950,000 fibers per gram dry weight of lung tissue to a low of 22,096 fibers per gram dry weight of lung tissue. The latter was rep resented by an amosite burden. While the numbers of uncoated fibers can vary with the techniques used in the counting scheme [35], a synopsis of the data on the subject is included in a recent review [36]. Within the limits of detection used in the study (17,100 to 350,000 fibers per gram dry weight of lung tissue), no asbestos fibers were found in samples from one patient, case L-13 (Table 3). There were no significant right/left side differences as to concentration or prevalence of fiber types. The same analysis indicated no statistical differ ences for fiber types per left versus right side for lengths, widths, and aspect ratios. Peritoneal tu mors had significantly thinner amosite fibers than did pleural tumors (p < .04). As shown in Figure 2, the most common com mercial amphibole was amosite in that 53 of the cases (96.4%) were found to have detectable levels. The highest concentration of amosite was 67,550,000 fibers per gram dry weight of lung tis sue with 18 of the patients (32.7%) having over one million amosite fibers per gram dry weight of lung tissue. A total of 39 patients (70.9%) had amosite levels greater than 200,000 fibers per gram dry weight of lung tissue. Uncoated amosite was found in lengths up to 137 pm; however, the majority was less than 13 pm in length. The geometric mean of the length of all amosite fibers measured was 12.9 pm and the geometric mean of the width of the measured amosite fibers was 0.244 pm (Table 4). The geometric mean of the aspect ratios for the f, fj ' f\ VT < / O V. 4b K\ \ 4 A- Asbestos Burden in Cases of Mesothelioma TABLE 3 Uncoated fiber data per case (fibers/g dry) Total asbestos 68,950,000 63,417,431 32,812,500 13,625,806 6,875,000 5,701,613 4,136,364 3,999,548 3,989,464 2,925,000 2,722,222 2,581,967 2,230,392 2,119,141 1,894,587 1,682,692 1,563,333 1,439,394 1,422,764 1,045,300 1,014,493 928,030 923,261 899,306 855,263 758,808 740,385 698,357 694,444 665,205 606,436 588,235 559,764 505,556 490,924 466,697 433,559 391,791 352,011 331,439 318,627 291,667 239,723 222,458 217,662 159,091 150,538 131,579 112,179 72,016 62,500 43,532 43,532 22,069 0 Amosite 67,550,000 29,434,251 31,506,530 13,412,903 6,593,750 4,770,161 4,003,788 3,651,762 3,453,065 2,800,000 2,527,778 2,175,546 2,144,608 1,936,849 1,346,154 1,383,547 1,143,333 1,060,606 853,659 613,546 855,978 839,646 776,379 753,472 614,035 23,713 673,077 492,958 300,926 614,035 462,046 392,157 382,997 427,778 404,290 262,763 374,437 348,259 176,006 287,247 220,588 72,917 199,772 49,435 195,896 159,091 37,634 109,649 56,090 36,008 62,500 21,766 0 22,069 0 Crocidolite 1,400,000 28,899,083 326,493 0 156,250 28,226 0 223,577 134,100 50,000 0 23,907 0 22,786 0 56,090 0 0 0 45,448 0 0 62,950 97,222 0 23,713 0 41,080 0 34,113 0 0 29,461 0 0 0 39,414 43,532 0 0 0 0 0 0 0 0 37,634 0 28,045 0 0 0 0 0 0 Tremolite 0 0 163,246 0 31,250 169,355 0 0 134,100 75,000 21,605 47,814 64,338 45,573 448,718 37,393 46,667 303,030 106,707 204,515 63,406 22,096 0 0 65,789 284,553 0 0 46,296 0 28,878 73,529 29,461 0 57,756 52,553 0 0 125,718 0 73,529 97,222 19,977 0 21,766 0 18,817 21,930 28,045 0 0 0 0 0 0 Actinolite 0 0 0 0 0 56,452 0 0 234,674 0 21,605 47,814 0 45,573 0 130,876 0 18,939 71,138 68,172 31,703 66,288 41,966 24,306 87,719 165,989 22,436 61,260 23,148 0 0 0 0 0 0 0 0 0 0 0 0 0 19,977 0 0 0 56,452 0 0 0 0 0 21,766 0 0 Anthophyllite Chrysotile 0 0 326,493 0 0 56,452 00 0 0 21,605 23,907 0 0 49,858 0 46,667 37,879 177,846 22,724 0 0 0 0 87,719 260,840 0 20,540 0 17,057 0 0 117,845 0 28,878 52,553 19,707 0 0 44,192 24,510 121,528 0 74,153 0 0 0 0 0 0 0 0 0 0 0 0 0 489,739 212,903 93,750 620,968 132,576 124,210 33,525 0 129,630 262,978 21,446 68,359 49,858 74,786 326,667 18,939 213,415 90,896 63,406 0 41,966 0 0 0 44,872 82,160 324,074 0 115,512 122,549 0 77,778 0 78,829 0 0 50,287 0 ,0 0 0 98,870 0 0 0 0 0 36,008 0 21,766 21,766 0 0 Nonasbestos 2,100,000 0 2,122,201 1,064,516 500,000 1,552,419 2,280,303 596,206 804,598 475,000 1,123,457 2,486,339 150,123 250,651 698,006 616,987 233,333 1,193,182 960,366 2,863,214 475,543 66,288 440,647 388,889 1,162,281 1,067,073 291,667 246,479 509,259 447,583 0 588,235 353,535 194,444 231,023 367,868 256,194 195,896 678,879 44,192 171,569 145,833 739,155 568,503 544,154 70,707 188,172 87,719 84,135 252,058 395,833 108,831 261,194 22,096 66,794 32S Assay number L-33 L-38 L-44 L-3 L-48 L-9 L-11 L-19 L-14 L-52 L-17 L-22 L-53 L-36 L-15 L-43 L-23 L-7 L-34 L-40 L-42 L-51 L-37 L-20 L-35 L-28 L-41 L-1 L-55 L-21 L-50 L-26 L-24 L-5 L-39 L-16 L-47 L-49 L-30 L-46 L-29 L-12 L-4 L-18 L-45 L-32 L-31 L-54 L-27 L-8 L-2 L-6 L-10 L-25 L-13 33C R. F. Dodson et al. V s FIG. 2 Percentage of cases with fibers present, by fiber type. amosite was 43.6. A comparison of geometric means is shown by the boxplots in Figure 3. Crocidoiite fibers were found in 22 (40.0%) of the cases, with none detected in the remaining 33 (60.0%) cases. The concentration of crocidoiite was similar to that of amosite in 3 of the cases (5.5%) (Table 3). In all 22 cases, where both fiber types of commercial amphiboles were found, amosite was generally the predominant type by ra tios varying from 1:1 to as high as 169:1. There were 31 additional cases (56.4%) where amosite was present with no crocidoiite detected. In those cases, the amosite ranged from 21,770 fibers to over 13,413,000 fibers per gram dry weight of lung tissue. The length as determined by geometric mean of the crocidoiite fibers was 8.46 pm, while the width was 0.194 pm. The crocidoiite fibers were, therefore, shorter and thinner when com pared to the same parameters for amosite. A fur ther comparison as shown in Figure 3 shows that crocidoiite had a slightly higher aspect ratio. Case L-38, which had the highest concentration of un coated crocidoiite, and an approximately equal amount of amosite, was found to have ferruginous bodies formed on both amphiboles. The average length of ferruginous bodies formed on crocidoiite was 28.6 pm, while that for amosite was 37.4 pm. TABLE 4 Fiber characteristics: geometric means Fiber type Length (pm) Width (pm) Aspect ratio Amosite Crocidoiite Chrysotile Tremolite Actinolite Anthophyllite 12.90 8.46 6.36 6.25 8.71 8.51 0.244 0.194 0.076 0.285 0.328 0.564 43.6 51.6 103.6 27.6 31.4 23.1 spond to the 75th percentile (top quartile), 25th percentile (bottom quartile), and 50th percentile (median) respectively. The whiskers extend from the 10th per centile (bottom) and the 90th percentile (top). Chrysotile was either found in equal concentra tion with amphiboles or as the majority asbestiform mineral in only three of the patients (cases L-6, L-8, and L-10). These individuals had an overall low asbestos burden. No chrysotile was detected in 24 (43.6%) cases. The chrysotile concentration ranged from 18,900 to 621,000 fibers per gram dry weight of lung tissue. As has been reported in pre vious studies, chrysotile fibers were generally short, with the length (based on geometric means) being 6.36 pm and a very thin width of 0.076 pm. The most common noncommercial amphibole was tremolite, which was found in 33 cases (60.0%). Tremolite fibers were similar in length (geometric mean of 6.25 pm) to chrysotile, but in diameter (geometric mean of 0.285 pm) were more similar to amosite. The average aspect ratio for tremolite based on geometric mean was 27.6. The longest tremolite fiber observed was 26 pm. Sev eral studies have suggested that tremolite, as a contaminant of chrysotile, is important in the risk of developing mesothelioma among chrysotile work ers [37], as well as in miners and millers [38, 39]. In the present study, 11 of the patients (20.0%) with chrysotile did not have detectable tremolite. Con versely, 13 patients (23.6%) with detectable levels of tremolite had no chrysotile detected. Thus, the authors do not believe the present data could sup port an argument that the tremolite exposure oc curred from tremolite contamination in chrysotile. <u , i fa Asbestos Burden in Cases of Mesothelioma Even though there is a mild correlation between chrysotile fibers per gram versus tremolite fibers per gram (0.23 = p value of <.05), there is an even stronger correlation between amosite and chryso tile (p = -45, p < .0001). If one chooses to make the argument that tremolite presence was attributable to chrysotile, then a stronger argument could be made that chrysotile accounts for the presence of amosite. The latter point has not been suggested in the literature as a naturally occurring link. Lung tissue from 21 (38.2%) cases had detect able limits of actinolite. The length (geometric mean) of actinolite fibers was 8.71 pm, with the aspect ratio (geometric mean) being 31.4. The other noncommercial amphibole, anthophyllite, was found in 21 (38.2%) cases. Anthophyllite fibers, as with the actinolite findings, tend to occur more frequently in the individuals with the overall higher fiber burdens. The anthophyllite fibers were similar in length (geometric mean, 8.51 pm) to the actino lite fibers. The aspect ratio of the anthophyllite fi bers was 23.1. While no ferruginous bodies were found in pa tient L-18, the individual did have uncoated asbes tos fibers (222,500 fibers per gram dry weight of lung tissue). The number of uncoated chrysotile fi bers in this patient, when compared with those of the other individuals, indicated there were only 12 patients (21.8%) with higher chrysotile burden. Pa tient L-18 had the sixth lowest amosite burden and the seventh highest concentration of anthophyllite. Patient L-13, reported as having no ferruginous bodies based on detectable limits used in the study, had no uncoated asbestos fibers, but did have nonasbestos fibers in the tissue. There have been several reports in the literature that longer fibers are more tumorigenic than shorter fibers. While it should be appreciated that considerable variation occurred in total numbers of the types of uncoated fibers, nevertheless, the question was asked as to what percentage of the fiber population from each type could be consid ered "long" as per Stanton dimensions. Thus, a comparison was made of the fibers in each type that conform to the Stanton hypothesis [19, 20, 40], that is, those fibers being 5=8 pm long and <0.25 in diameter (Table 5A). Using this Stanton definition, 53 of the 55 (96.4%) patients had long, uncoated asbestos fibers. Fifty-two of 55 patients (94.5%) had TABLE 5A Fibers fitting the Stanton hypothesis Fiber type Percentage Amosite Anthophyllite Crocidolite Chrysotile Actinolite Tremolite 48 46 41 39 34 16 Note. Length a8.0 urn and diameter <0.25 urn. 331 long amosite fibers, while 21 of 55 patients (38.2%) had long chrysotile fibers. Long fibers of crocidolite were present in 16 of 55 (29.1%) patients. The non commercial amphiboles (tremolite, actinolite, and anthophyllite) were found as long fibers in 17, 13, and 18 patients, respectively. Since environmental air monitoring under the NIOSH-582 (National Institute for Occupational Safety and Health) counting scheme is based on fibers longer than 5.0 pm, a determination of this "definition" of short and long (or regulated) fibers was made using this standard. The percentage of the fibers in each type of asbestos that could have been potentially counted as "long" fibers (5=5.0 pm, per NIOSH counting criteria) are shown in Table 5B. The NIOSH counting standard is based on the use of phase-contrast light microscppy; thus, the fibers shorter than 5pm and those longer than 5pm, but thinner than the functional detect able limit of the light microscope (<0.25 pm) would have appreciably changed the fibers per category. Under these conditions, far fewer fibers would be counted (see Table 5B). The geometric mean concentration of fibers s=5.0 pm is, in decreasing order; amosite > crocid olite > chrysotile > anthophyllite > actinolite > trem olite. For fibers >8.0 pm (per Stanton hypothesis), the order is amosite > crocidolite > chrysotile > tremolite > actinolite > anthophyllite. The total number of "short" fibers per asbestiform was also compared. The decreasing concentrations for those <5.0 pm or <8.0 pm were amosite > crocido lite > chrysotile > tremolite > actinolite > anthophyl lite (Table 6). Comparisons: Pathologic Asbestosis With FB/Asbestos Fiber Data Cases with pathologic asbestosis and those with out pathologic asbestosis were compared with re spect to asbestos bodies per gram of wet lung tissue (Bremerton Laboratory [BL]), as well as a multisite quantitative analysis of ferruginous bod ies per gram of wet lung tissue (Tyler Laboratory TABLE 5B Fibers meeting and countable by NIOSH criteria Fiber type Percentage >5 pm Percentage detectable by light microscopy3 Amosite Crocidolite Anthophyllite Actinolite Chrysotile Tremolite 70 67 67 60 55 36 33.0 16.0 53.7 37.9 1.4 27.6 '>5 urn length and a0.25 urn diameter. 332 TABLE 6 Geometric mean of the concentrations of asbestos >5 pm >8 pm <5 pm <8 pm Amosite > Amosite > Amosite > Amosite > Crocidolite > Crocidolite > Crocidolite > Crocidolite > Chrysotile > Chrysotile > Chrysotile > Chrysotile > Anthophyllite > Tremolite > Tremolite > Tremolite > R. F. Dodson e; Actinolite > Actinolite > Actinolite > Actinolite > Tremolite > Anthophyllite > Anthophyllite > Anthophyllite > [TL]), used for comparison with total asbestos fi bers per gram of dry lung tissue (Tables 7A and 7B). As stated under the Materials and Methods, the numbers listed as "ferruginous bodies per gram wet lung--BL" were determined by routine light microscopic examination at the Bremerton Laboratory of 5 g of digested peripheral lung tis sue. The numbers listed under the designation "ferruginous bodies per gram wet lung--TL" rep resented a combined digest pool from multiple sites of the right and left lungs that were obtained as part of the quantitative analysis of bodies and fibers at the Tyler facility. Over 97% of the ferrugi nous bodies analyzed by ATEM were asbestos bod ies. Twenty-nine cases showed pathologic asbesto sis, and were graded according to the criteria of the College of American Pathologists and the National Institute for Occupational Safety and Health [261. As shown in Table 7A, 22 of 29 patients showed pathologic grade 1 asbestosis which was usually focal (extent A). The mean asbestos body concen tration per gram of wet lung tissue in patients with pathologic asbestosis as determined by routine di gestion and light microscopic evaluation in the Bremerton analysis was 6,122, with a range be tween 422 and 25,000 and a median of 3,440. In contrast, mean ferruginous body concentration per gram of wet lung tissue in the 26 cases that did not show pathologic asbestosis was 1,569, with a TABLE 7A Cases with pathologic asbestosis Assay number L-1 L-3 L-4 L-8 L-9 L-11 L-12 L-14 L-15 L-17 L-19 L-21 L-23 L-24 L-27 L-28 L-29 L-33 L-39 L-40 L-42 L-43 L-44 L-45 L-48 L-49 L-50 L-51 L-53 Pathologic grade of ' asbestosis 1 1 1 1 1 1 1 1 1 1-3 1 1 1 1 1-3 1 1 1-2 1-2 1 1 1 1-2 1 1-2 1 1 3 1 Extent of asbestosis A A A A B B A A A C B A A A B A A B B -A A A B A B A A C A FBs/g weta 12,140 25,000 712 422 1,852 12,771 1,320 20,100 11,200 5,620 25,000 1,980 3,440 1,260 1,560 850 4,220 5,700 813 1,088 694 643 ND 2,975 13,000 11,600 2,450 496 2,521 FBs/g wetb 18,968 195,419 737 756 29,077 27,780 2,542 4,903 26,464 7,219 13,114 2,114 3,325 971 60 547 1,148 201,416 1,631 458 528 4,100 114,788 299 31,739 6,074 17,677 2,968 14,116 Total asbestos fibers/g dry 107,211 13,625,806 239,726 72,016 5,701,613 4,136,364 291,667 3,989,464 1,894,587 2,722,222 3,999,548 665,205 1,563,333 559,764 112,179 758,808 331,439 68,950,000 490,924 1,045,300 1,014,493 1,682,692 32,625,692 217,662 6,875,000 391,791 134,634 928,030 2,230,392 aFerruginous bodies/g wet lung as determined by light microscopy of 5 g of digested lung from site where blocks taken for grading by light microscopy (Bremerton Laboratory). bFerruginous bodies/g wet lung determined on digestion pool from multiple sites of right and left lung as used in quantitative analysis by light and electron microscopy (Tyler Laboratory). .Asbestos Burden in Cases of Mesothelioma TABLE 7B Cases without pathologic asbestosis Assay number L-2 L-5 L-6 L-7 L-10 L-13 L-16 L-.18 L-20 L-22 L-2 5 L-26 L-30 L-31 L-32 L-34 L-35 L-36 L-37 L-38 L-41 L-46 L-47 L-52 L-54 L-55 Ferruginous bodies/g wet'1 1,572 5,190 220 2,390 117 117 382 144 108 1,541 840 540 1,580 57 327 1,221 0 2,382 678 13,109 360 968 4,420 929 59 183 Ferruginous bodies/g wetfc' 1,737 3,338 46 8,391 10 0 400 0 546 5,234 60 40 279 240 325 3,952 13 15,523 1,536 170,985 699 1,409 759 6,832 160 488 Total asbestos fibers/g dry 62,500 505,556 43,532 1,439,394 43,532 0 446,697 222,458 899,306 2,581,967 22,096 588,235 352,011 150,538 159,091 1,422,764 855,263 2,119,141 923,261 63,417,431 740,385 331,439 433,559 2,925,000 131,579 694,444 aFerruginous bodies/g wet lung as determined by light mi croscopy of 5 g of digested lung from site where blocks taken for grading by light microscopy (Bremerton Laboratory). bFerruginous bodies/g wet lung determined on digestion pool from multiple sites of right and left lung as used in quantitative analysis by light and electron micrsocopy (Tyler Laboratory). range between 0 and 13,109 and a median of 540 (Table 7B). Ferruginous bodies per gram of wet lung tissue in the 29 cases with pathologic asbestosis as deter mined from the Tyler analysis ranged from 60 to 201,416, with a mean of 25,505 and a median of 4,100. In contrast, the 26 cases without pathologic asbestosis had a mean ferruginous body count per gram of wet lung tissue of 8,918, with a range of 0 to 170,985, and a median of 546 (Table 7B). Of the 29 cases with pathologic asbestosis, mean total asbestos fibers per gram of dry lung tissue was 5,426,123, with a range of 72,016 to 68,950,000, and a median of 1,045,300 (Table 7A). In contrast, the 26 cases without pathologic asbes tosis had a mean asbestos fiber concentration per gram of dry lung tissue of 3,258,705 with a range of 0 to 63,417,431 and a median of 505,556 (Table 7B). discussion Medical Mesothelioma, a rare neoplasm, occurs most fre quently in individuals with previous asbestos ex 333 posure. The life expectancy of patients with differ ent histologic types of mesothelioma reported herein is similar to what has been reported in the literature [24]. Patients with epithelial mesothelio ma have a longer mean survival than those with sarcomatoid, desmoplastic, and biphasic mesothe liomas and mesotheliomas showing variable differ entiation. The reason for this difference in survival is not clear. The five peritoneal mesotheliomas re ported in this study had a mean survival of 3.5 months, which is consistent with what we have found in a much larger group (n = 70) of patients with peritoneal mesothelioma (average survival = 4.2 months; range <1 month-18 months; unre ported data collected by SH). In this study, we found that those patients whose lung tissue showed pathologic asbestosis had higher mean asbestos body and asbestos fiber con centrations than those whose lung tissue showed no asbestosis. However, our data have shown there was an extensive range in asbestos body and fiber concentration in patients with and without pathologic asbestosis, making it impossible to pre dict the presence of pathologic asbestosis based on lung asbestos body/fiber concentration. Our obser vations are similar to those of Warnock and Isenberg [41], who demonstrated that the lungs from some patients with lung cancer had a heavy burden of asbestos without showing fibrosis, whereas some with the same concentration of asbestos did have asbestosis. Warnock and Isenberg suggested that fibrosis may depend on factors besides total fiber burden. Leigh et al. [42] found a statistically significant trend between lung fiber content and mesothelio ma cell type from epithelial (low fiber content) through mixed to sarcomatous (high fiber content). This trend was stated to be most apparent for total uncoated fibers and crocidolite. A higher lung fiber content was found in peritoneal mesotheliomas versus pleural mesotheliomas. In our study, pleural epithelial mesotheliomas had a mean total asbes tos fiber concentration per gram of dry lung tissue of 6,275,651 (range 0-68,950,000). The biphasic me sotheliomas (n = 7) had a mean total asbestos fiber concentration per gram of dry lung tissue of 532,485 (range 43,532-2,119,141). The sarcomatoid (n = 10) had mean total asbestos fiber concentra tion per gram of dry lung tissue of 133,625 (range 134,634-4,136,364). The mesotheliomas that had a predominantly desmoplastic histologic appearance (n = 5) had a mean asbestos fiber concentration per gram of dry lung tissue of 1,524,404 (range 131,579-2,722,222). The mesotheliomas showing variable differentiation (n = 5) had a mean asbestos fiber count per gram of dry lung tissue of 693,778 (range 159,091-1,563,333). The peritoneal meso theliomas (n = 5) had a mean asbestos fiber con centration per gram of dry lung tissue of 16,312,071 (range 22,096-63,417,431). Our findings are similar to those reported by 334 Leigh et al. [42], with respect to peritoneal meso theliomas having a higher lung fiber burden than pleural mesotheliomas. Our findings were different than those reported by Leigh et al. [42], with re spect to mean fiber concentration and histologic type of pleural mesothelioma. In decreasing order of total mean asbestos fiber burden, the histologic types of pleural mesothelioma were ranked as fol lows: (1) epithelial, (2) desmoplastic, (3) variable differentiation, (4) biphasic, and (5) sarcomatoid. The total number of cases is small and with a larger sample, the results could obviously be different. Analytical While the amphibole crocidolite has been linked with the development of mesothelioma [12, 43] in the studies of Sluis-Cremer et al. [4], amosite has also been suggested to be important in the produc tion of this tumor [18, 44, 45], In the present study, amosite is the most commonly found amphibole, while only 40.0% of the patients had crocidolite. This prevalence of amosite in the present study (96.4%) compares favorably with the observation by Churg and Vedal [18], where amosite was found in all lungs of shipyard workers and insulators from the Pacific Northwest [18]. The occurrence of amos ite in the present study of mesothelioma is higher than reported by Roggli et al. [45] where 81% of the cases contained amosite. Roggli et al. reported 58% of all fibers >5 pm were amosite, while amosite accounted for 64.3% of such fibers in the present study. The length/width ratios of amosite burden in work of Churg and Vedal [18] led to the suggestion that an association existed between low-aspectratio amosite fibers and mesothelioma. This con trasts with the present data in which the amosite population is represented by fibers with a higher aspect ratio (as based on uncoated fibers data). This shift would be further toward higher aspect , ratios if the lengths/widths of amosite fibers that were cores of the ferruginous bodies were included in the overall amosite data. In our study, the only asbestiform mineral found (based on geometric mean of the length) to have the majority of fibers less than 5 pm was tremolite (64% < 5.0 pm). Since the work histories suggest chrysotile ex posure, its absence in tissue digests requires an explanation. For example, chrysotile is often in haled as a shorter fiber than amphiboles. In 16 pa tients, there were longer chrysotile fibers (3=5.0 pm) in the autopsy tissue, while all patients had popu lations of short amphiboles. The absence of chrysotile in some samples thus could have re sulted from (1) a greater preferential clearance of short chrysotile while sparing the equivalent length amphiboles, or (2) an initial disproportionate inha lation of amphiboles had occurred. Tremolite has been suggested as an important component of chrysotile dust and a contributor to R. F. Dodscn et al. the development of mesothelioma among miners [13, 14]. In the present study, if one chooses to em phasize the slight correlation between the levels of chrysotile and tremolite and conclude that an ex posure to chrysotile involves a resultant exposure to tremolite, the stronger correlation between chrysotile and amosite must also be appreciated. We suggest the exposure was to mixed asbesti form minerals and the combination in the tissue most likely reflects such exposures to the various asbestiform minerals rather than a contaminant of one asbestos containing product with another type of asbestos--i.e., tremolite in chrysotile. The absence of tremolite fibers in 22 of the 55 patients (40.0%) is in contrast with the suggestion of Srebro and Roggli [13] that this asbestiform is "nearly ubiquitous and represents the most com mon amphibole fiber in the lungs of urbanites." While the data in the present study cannot be used to speculate where tremolite exposures may have occurred, there is no disagreement with Srebro and Roggli's conclusion [13] that "moderate exposures may produce malignant pleural mesothelioma in some patients." The same argument, however, can be made from the present study for each asbesti form at lower exposures as it may impact on indi vidual susceptibility for developing mesothelioma. Ferruginous bodies are a marker of past asbes tos exposure. Among this group of individuals with mesothelioma, 83.6% of the patients would have been classified as having had occupational expo sures based on the concentrations of ferruginous bodies (>1000 ferruginous bodies per gram of dry weight of lung tissue). Two of the patients (3.6%) were found to have no ferruginous bodies (L-13, L-18). These individuals could, therefore, be con sidered to have "control" levels by this standard. However, by electron microscopy, the tissue from patient L-18 was found to have uncoated amosite, anthophyllite, and chrysotile fibers. The critical question in such cases is that if their work history clearly indicated repetitive exposures to asbestos, why were their tissue burdens within control levels for FBs? Rates of clearance and toxicity of different dusts vary as per individual susceptibility. Likewise, the development of ferruginous bodies in individuals vary and some are poor "coaters" even if appro priate asbestos fibers are present [34, 46]. Con versely, some individuals, such as L-2, appear to be very efficient "coaters" since there are occupa tional levels of ferruginous bodies (>1000 FBs/g dry) and reasonably low numbers of uncoated fi bers. The impact of clearance could be offered to ex plain the lack of chrysotile burden in numbers of the patients, even though logic would dictate their workplace exposures (defined by the levels of am phiboles) surely should have included appreciable chrysotile. It should be noted that 31 of the patients (56.4%) did have chrysotile burdens, with portions Asbestos Burden in Cases of Mesothelioma of that burden (55%} being made up of fibers longer than 5 pm. An explanation for the development of mesothe lioma in individuals with low fiber burden could always be offered that disease had spontaneously developed, even though there was a history of a known past exposure, suggesting asbestos as a contributor. Another explanation is based on Churg's suggestion that mesothelioma appeared at a much lower amosite burden than that required for asbestosis, which could apply to all types of asbestos [18]. Thus, combining this concept with the impact of clearance efficiency in some individu als, a result could be the reduced or very low tissue burden at the time of death. Exposure to the noncommercial asbestiform anthophyllite has been reported to have a high po tency for causing lung cancer, but suggested as a low potency in the tumorigenicity of mesothelioma when compared with other types of asbestos [47]. In part, these differences may have some basis in that a much greater emphasis has been placed on evaluating human exposures to commercial types of asbestos. However, one of the few studies of cohorts with appreciable exposure to the noncom mercial asbestos anthophyllite reported the occur rence of mesotheliomas [48]. Concern for exposure and resultant risks from noncommercial amphiboles is not a trivial issue since, in the present study, the noncommercial types of asbestos are found in some patients at higher levels than the combina tion of the commercial types. For example, in cases such as L-12 and L-18, anthophyllite is the predomi nant amphibole. When all noncommercial forms (tremolite, actinolite, and anthophyllite) are com bined, the majority of all asbestos fibers in L-12 and L-28 were noncommercial types. The concentration of fibers, per asbestos type, is dependent on the quality of exposure for each co hort. Thus, while tremolite was in greater concen tration than other asbestiforms (3=5.0 pm) in asbes tos miners [39], our observations indicate amosite as being the prevalent fiber in these cases. How ever, the underlying conclusion in the present study, as well as that of Dufresne et al. [39], is that individuals with mesothelioma have in most cases a confirmed asbestos burden and that most of these burdens are above general population levels for either ferruginous body and/or uncoated fiber burden. The risk for mesothelioma appears to be clearly tied in most cases to the exposure to asbestos. The relationship of longer amphibole fibers to the risk of developing mesothelioma may, as suggested by Churg and Vedal [13], simply constitute the major component at time of autopsy due in part to its reduced rate of clearance. While chrysotile burden among the patients in the present study was over shadowed by the amphibole burdens, 55% of the chrysotile burden is made up of fibers longer than 5 pm. Once again,, either indicating a selective in- 335 halation of longer chrysotile fibers to the exclusion of shorter fibers or more likely the resultant impact of clearance on the short fiber burden. This obser vation should not be used to exclude the composite importance of shorter fibers as contributors to the disease, nor, from the present data, can arguments be offered in support of varying levels of risk for mesothelioma from exposure to the various types of asbestos. In fact, the present data offer the op posite argument. The common link in these meso thelioma patients appears to be the presence of asbestos of various types and variable lengths. REFERENCES 1. Wagner JC. Mesothelioma and mineral fibers. Cancer. 1986;57:1905-1911. 2. McDonald AD, McDonald JC. Malignant mesothelioma in North America. Cancer. 1980;46:1650-1656. 3. Botha JL, Irwig LM, Strebel PM. Excess mortality from stomach cancer, lung cancer, and asbestosis and/or meso thelioma in crocidolite mining districts in South Africa. Am J Epidemiol. 1986;123:30-40. 4. Sluis-Cremer GK, Liddell FDK, Logan WPD, Bezuidenhout BN. The mortality of amphibole miners in South Africa, 1946-80. BrJInd Med. 1992;49:566-575. 5. Hansen J, de Klerk NH, Eccles JL, Musk W, Hobbs MST. Malignant mesothelioma after environmental exposure to blue asbestos. Int J Cancer. 1993;54:578-581. 6. McConnochie K, Simonato L, Mavrides P, Christofides P, Pooley FD, Wagner JC. Mesothelioma in Cyprus: the role of tremolite. Thorax. 1987;42:342-347. 7. Mancuso TF. Relative risk of mesothelioma among rail road machinists exposed to chrysotile. Am J Ind Med. 1988;13:639-657. 8. Spirtas R, Heineman EF, Bernstein L, et al. Malignant me sothelioma: attributable, risk of asbestos exposure. Occup Environ Med. 1994;51:804-811. 9. Gelder TV, Hoogsteden HC, Versnel MA, de Beer PH, Vandenbroucke JP, Planteydt HT. Malignant peritoneal meso thelioma: a series of 19 cases. Digestion. 1989;43:222-227. 10. Fitzpatrick DR, Peroni DJ, Bielefeldt-Ohmann H. The role of growth factors and cytokines in the tumorigenesis and immunobiology of malignant mesothelioma. Am J Respir Cell Mol Biol. 1995;12:455-460. 11. Peto J, Hodgson JT, Matthews FE, Jones JR. Continuing increase in mesothelioma mortality in Britain. Lancet. 1995;345:535-539. (Abstract). 12. Wagner JC, Pooley FD. Mineral fibres and mesothelioma. Thorax. 1986;41:161-166. 13. Srebro SH, Roggli VL. Asbestos-related disease associated with exposure to asbestiform tremolite. Am J Ind Med. 1994;26:809-819. 14. Lippmann M. Deposition and retention of inhaled fibres: effects on incidence of lung cancer and mesothelioma. Occup Environ Med. 1994;51:793-798. 15. Mossman BT, Bignon J, Corn M, Seaton A, Gee JBL. As bestos: scientific developments and implications for public policy. Science. 1990;247:294-301. 16. Becklake MR, Churg A, Rosenstock L, Weill H. Mesothelio ma update for clinicians. Am Rev Respir Dis. 1986;134: 1042-1045. 17. Dodson RF, Williams MG, Corn CJ, Brollo A, Bianchi C. A comparison of asbestos burden in lung parenchyma, lymph nodes, and plaques. Ann NY Acad Sci. 1991;643: 53-60. 18. Churg A, Vedal S. Fiber burden and patterns of asbestosrelated disease in workers with heavy mixed amosite and chrysotile exposure. Am J Respir Crit Care Med. 1994;150: 663-669. 19. Stanton MF, Layard M, Tegeris E, et al. Relation of particle 336 dimension to carcinogenicity in amphibole asbestoses and other fibrous minerals. J Natl Cancer Inst. 1981;67:965. 20. Stanton MF, Layard M, Tegeris A, Miller E, May M, Kent E. Carcinogenicity of fibrous glass: pleural response in the rat in relation to fiber dimension. J Natl Cancer Inst. 1977; 58:587-603. 21. Timbrell V, Skidmore JW. Significance of fiber length in experimental asbestosis. In: Holstein E, Anspach M,, eds. Internationale Konferenz der biologische Wirkungen des Asbestes, Dresden. Berlin: Dtsch Zentralinst Abeitsmed; 1968:52-56. 22. Lippmann M. Asbestos exposure indices. Environ Res. 1988;46:86-106. 23. Lippmann M. Effects of fiber characteristics on lung depo sition, retention, and disease. Environ Health Perspect. 1990;88:311-317. 24. Hammar SP. Pleural diseases. In: Dail DH, Hammar SP, eds. Pulmonary Pathology. New York: Springer; 1994: 1463-1580. 25. Henderson DW, Shilkin KB, Whitaker D, et al. The pathol ogy of malignant mesothelioma, including immunohistology and ultrastructure. In: Henderson DW, Shilkin KB, Langlois SLP, Whitaker D, eds. Malignant Mesothelioma. New York: Hemisphere; 1992:69-139. 26. Craighead JE, Abraham JL, Churg A, et al. The pathology of asbestos-associated diseases of the lungs and pleural cavities: diagnostic criteria and proposed grading schema. Arch Pathol Lab Med. 1982;106:544-596. 27. Smith MJ, Naylor B. A method of extracting ferruginous bodies from sputum and pulmonary tissue. Am J Clin Pathol. 1972;58:250-254. 28. Williams MG, Dodson RF, Corn C, Hurst GA. A procedure for the isolation of amosite asbestos and ferruginous bod ies from lung tissue and sputum. J Toxicol Environ Health. 1982;10:627-638. 29. Jaffe MS. Handling and washing fragile replicas. J Appl Phys. 1948;19:1187 (Abstract). 30. Churg A, Warnock ML. Correlation of quantitative asbes tos body counts and occupation in urban patients. Arch Pathol Lab Med. 1977;101:629-634. 31. Roggli VL, Pratt PC, Brody AR. Asbestos content of lung tissue in asbestos associated diseases: a study of 110 cases. Br J Ind Med. 1986;43:18-28. 32. Roggli VL, Benning TL. Asbestos bodies in pulmonary hi lar lymph nodes. Mod Pathol. 1990;3:513-517. 33. Breedin PH, Buss DH. Ferruginous (asbestos) bodies in the lungs of rural dwellers, urban dwellers and patients with pulmonary neoplasms. South Med J. 1976;69:401-404. 34. Dodson RF, Greenberg SD, Williams MG, Corn CJ, E~ F. Dodson et al. O'Sullivan MF, Hurst GA. Asbestos content in lungs of oc cupationally and nonoccupationalh, txposed individuals JAMA. 1984;252:68-71. ' 35. Dodson RF, O'Sullivan MF, Corn CJ. Technique dependent variations in asbestos burden as illustrated in a case of nonoccupational exposed mesothelioma. Am J Ind Med 1993;24:235-240. ' 36. Hammar SP, Dodson RF. Asbestos. In: Dail DH, Hammar SP, eds. Pulmonary Pathology. New York: Springer; 1994 901-983. ` 37. Churg A, Wiggs B, Depaoli L, Kampe B, Stevens B. Lung asbestos content in chrysotile workers with mesothelioma. Am Rev Respir Dis. 1984;130:1042-1045. 38. Churg A, Wright JL, Vedal S. Fiber burden and patterns of asbestos-related disease in chrysotile miners and millers. Am Rev Respir Dis. 1993;148:25-31 (Abstract). 39. Dufresne A, Harrigan M, Masse S, Begin R. Fibers in lung tissues of mesothelioma cases among miners and millers of the township of Asbestos, Quebec Am J Ind Med. 1995;27:581-592 (Abstract). 40. Stanton MF, Wrench C. Mechanisms of mesothelioma in duction with asbestos and fibrous glass. J Natl Cancer Inst. 1972;48:797-821. 41. Warnock ML, Isenberg W. Asbestos burden and the pa thology of lung cancer. Chest. 1986;89:20-26. 42. Leigh J, Rogers AJ, Ferguson DA, Mulder HB, Ackad M, Thompson R. Lung asbestos fiber content and mesothelio ma cell type, site, and survival. Cancer. 1991;68:135-141 (Abstract). 43. Wagner JC. The discovery of the association between blue asbestos and mesotheliomas and the aftermath. Br J Ind Med. 1991;48:399-403. ~ 44. Selikoff IJ, Churg J. Carcinogenicity of amosite asbestos. Arch Environ Health. 1972;25:183-186 (Abstract). 45. Roggli VL, Pratt PC, Brody AR. Asbestos fiber type in ma lignant mesothelioma: an analytical scanning electron mi croscopic study of 94 cases. Br J Ind Med. 1993;23:605- 614 (Abstract). 46. Dodson RF, Williams MG, O'Sullivan MF, Corn CJ, Green berg SD, Hurst GA. A comparison of the ferruginous body and uncoated fiber content in the lungs of former asbes tos workers. Am Rev Respir Dis. 1985;132:143-147. 47. Meurman LO, Pukkala E, Hakama M. Incidence of cancer among anthophyllite asbestos miners in Finland. Occup Environ Med. 1994;51:421-425 (Abstract). 48. Karjalainen A, Meurman LO, Pukkala E. Four cases of me sothelioma among Finnish anthophyllite miners. Occup Environ Med. 1994;51:212-215 (Abstract).