Document oMeLxQE9kKRaRyVDKGkgGNb13

Original Paper Inidc^or-Bidlt Environment Indoor Built En' iron 19V' 6 264-276 V W.-.-*oJ KprU IS I90J . llgren* E. Chatfieid * Brvn Ma*vr. Pa.. USA. s Toronto. Canada Coalinga Fibre - A Short, Amphibole-Free Chrysotile Part t: Evidence for a Lack of Fibrogenic Activity PLAINTIFF'S EXHIBIT SA-474 Key Words Short fibre chrysotile Coalinga Fibrosis F344 rats Abstract Cont roversy continues to surround the biological activity ofshort fibre chryso tile. This is largely due to a lack of studies in which there has been `pure expo sure' to this material. Most of the exposures studied, whether in humans or animals, have been confounded by the additional presence of long fibre chry sotile and/or amphibole. This report presents the morphological and morpho metric findings of a lifetime inhalation study of F344 rats exposed to three types ofchrysotile. The first, from Coalinga. Calif., is comprised of fibres that are almost all less than 5 pm in length and is not contaminated with amphi bole. The other two. the Jeffrey fibre and the UICC/B standard, are both Canadian long fibre preparations with a minor degree of amphibole contami nation. Exposed animals displayed no fibrosis following exposure to Coalinga chrysotile but showed tlbrogenic responses with bath Canadian fibres. Introduction Two major theories are often cited to explain asbestosrelated disease. Of these, the `Stanton1 hypothesis (1. 2] proposes that short fibres are less harmful than long whilst the `amphibole' hypothesis [3. 4] claims that relatively .soluble fibres, like chrysotile, are less injurious than high ly durable amphiboles. If correct, these theories, taken together, would suggest that the least harmful type of asbestos would be uniformly short, amphibole-free, chry sotile. Short fibre chrysotile is the most common form of asbestos in air including the atmosphere of buildings (5 8], water, soil, and lung [9-11]. In fact, short fibre chryso tile'is so ubiquitous that it can confound even the most carefully done analyses [12. 13). Its targe atmospheric contribution comes from numerous, naturally occurring outcroppings as well as diverse commercial products such as brake linings, tiles, and cement in which it has been widely used [14). This study and its companion reports describe the unique analytical and biological properties of Coalinga fibre, a short, amphibole-free, chrysotile. and demon strate that it is largely devoid of biological activity. The proposal that this form ofasbestos should be regarded as a nuisance dust and `maybe innocuous in lung tissue' [5] is therefore put forth. This is not only consistent with the findings described herein but is also in keeping with the much larger body of scientific thought. Historically, sup portive conclusions reached almost 50 years ago by Vorwald et aL [ 15] are consistent with the later work of Davis et al. [16] and Berman et a1. [17]. Studies by all three groups concluded that the truly pathogenic fibres are KARGER l M 306 Cl i"9 '*.* *b v KuyerAG.Bod Dr. . I- flpm. MO. Dfol GjnuiKaei rtytiCMft. $*ae So JOJ I>3 Montycmcr- Kienue Tr. -*I0 5ZI -610 5101154 UCAREF00024350 those greater chan 20 jim in length. Recent panels (6. 7] distribution, and clearance of Coalinga chrysotile which have also endorsed the increased pathogenicity of long distinguish it from 'long' Canadian fibre. Taken together, fibres (i.e. > 5-8 pm in length). Although Lippmann (IS] the findings described in all these reports support the con claimed that the `critical fibre parameters for asbestosis' tention that short-fibred chrysotile is innocuous and were surface area and fibre lengths greater than 2 pm, this should be regarded and regulated as a nuisance dust. wjj based largely on Timbrell's (19-22J lung burden stud ies of relatively few amphibole workers. The inability to produce `pure*1 short fibre preparations Materials and Methods in large quantities, devoid of tong fibres, has been a chronic problem when comparing the biological potential of long versus short fibres. For example, Wagner et al. (23] assessed the biological activity of the shortest commercial sample, a grade 7 fibre from the Bells mine (Quebec), in relation to two experimental Canadian chrysotile prepa rations. Unfortunately, the Bells grade 7 material con tained more long (>5 pm) fibres than the other two (see fig. 1 in 23], and the resulting pathogenic responses were attributed to the high numbers of long fibres in each. More recently, Davis and Jones (24] administered `short' and 'long' chrysotile preparations to rats by inhalation and found that the `short fibre* produced significant pathogenic effects. This was again attributed to sample impurity due to long fibre contamination. Aside from the study by Davis and Jones [24], the most recent short fibre chrysotile inhalation investigation was conducted at NIOSH by Platek et al. [25]. These workers exposed rats and monkeys via inhalation to a highly puri fied short fibre chrysotile sample and found neither fibro sis nor tumours in either species.. Historically, the findings of inhalation studies aimed at assessing the role of fibre length have been supported and clarified by numerous instillation and injection experi Source ofMaterials The present study originated as part ofa large, joint investigation undertaken by the NIEHS in (he United States and (be Medical Research Council fMRO Pneumoconiosis Unit in (he United King dom to compare the results of similar inhalation studies earned out at two different locations under nearly identical conditions (31. 32|'. The comparative MRC - NIEHS study, as originally conceivea. con tained only one chrysotile exposure group, namely UICC/B: Two additional groups, tbe long' Jeffrey fibre. (Jeffrey Mine. Asbestos. Que.). and tbe 'short' Coalinga. were added to (he NIEHS protocol by (be NIEHS investigators-1. Node of the fibrosis or tumour pathology data have ever been published for Coalinga or Jeffrey fibre. Fibrosis score data have only been published for the untreated and UICC/B treated, interim sacri fice animals (31.32| but not for (he groups oa lifetime test. None of the Coalinga and UICC/B morphometry data have been published in scientific, peer-ceviewed journals though some have appeared in a thesis published by Pinkerton [33| and in several small abstracts by Crapo et al. (341, Pinkerton et al [35,36) and O'Neil et al. (37). How ever. most of (he morphometric data for the untreated control and the Jeffrey-treated animals have already been described [38--111. .All ofthe morphometric studies were performed by Professor Kent Pin kerton! ' The records, histology slides, paraffin blocks, and wet tissues of the animals on lifetime test were located in November 199! at the NTP archives by Drs Sits, Hamlin and Bridges as inventoried docu ments. All of these were reviewed solely by the senior author aside from 45 cases that were studied in conjunction with Dr J.C. Wagner. ments [see 5, 24 for excellent review and 26 tor critical discussion of method]. Injection studies using short fibre chrysotile preparations derived from ball milling [for cita tions see 24] have been consistent with those obtained with water sedimentation methods [27-30] in showing a general lack of biological effects with short fibres. The present report is based upon a lifetime study of F344 rats exposed by inhalation to either Coalinga. Jef frey or UICC/B chrysotile fibres that was performed at the National Institute of Environmental Health Sciences (NIEHS), Research Triangle Park, N.C., USA and the National Toxicology Program (NTP) between 1978 and 1980. Whilst this report is particularly concerned with fibrosis, the others in this series (see above) describe the implications of these findings for human fibrogenesis; the historical evidence in support of the greatly reduced pathogenicity of short fibre chrysotile; and the analytical properties, tumourigenic effects, differential deposition. ' The NIEHS investigation was coadticted it the NIEHS and funded by the NIH as in intramural project (NIEHS contract No. NOI-4-2I-0-I8CD and EPA Coop. Agreement No. CR8072SSI. Dr. Jidc Moore was the niperasor and Dr. Arnold Brody, the Project Officer. The NTP pathologists. Dr. Eugene Me Conodi and Dr. Guy Boorman, oversaw the pathofogical study ofthe am. mats. Mach at' the routine work, including monitoring of fibre levas and daily animal health, was initially performed under contract with Benoo Dickinson (8D> (vs '8CDT, by Dr. finis Csvendir and Dr. Coon* Stoee [cc BD Quar terly Reports: Efforts ofchronic exposure to autxxne environmental agents (I April-Juiy I97B: 1 Au* -it Oct 1971: and I Nov I97S-:-! Jan 1979)] and finalised by Nacthrup. Inc. by Dr. Bernard Atkins and Dr. Roden O'Connor. ! 'Union Internationale Contie te Oncer' funded the prepararara of 'stan dard'asbestos samples for researeh. UICC/B chrynctic was a mature otxained from eight different Canadian chrysotile mines blended in proportion to their output. 1 On 4 Sept 79, a Request for Proposal (RFPNIEHS-79-IJ) was sent out lor bid by Dr. Arnold Brody to have morphometry performed in additioa to bistopuhokagy. The RPP was awarded to Duke University under subcontract to Dr. James Cnpo (NIEHS No. NOl-ES-OaXXW). Most of the actual morphometry was done by Prof. Kent Pinkerton is Dr. Crapo's PhD student. Short Fibre Coalinga Chrysotile Lack of Fibrosis Indoor Built Environ 1997:6.764-176 765 UCAREF00024351 Tbla 1. Experimental design - NTP inhalation bioassay (197S- as NIEHS short range chrysotile and identified as COF-25 as 1980) Number of animals at each time point obtained from Union Carbide [46). The Coalinga chrysotile was sub jected to additional milling and separation but these processes did Treatment groups Time, months1 not destroy the cry stallme structure of the fibres (46], Jeffrey chryso- 03 12 24 tile (NIEHS intermediate range) was prepared by roller milling using standard industrial techniques, followed by three cycles of 'Hum- Controls, untreated I Coalinga 1 UICC/B Jeffrey f J 45m. 45f 36m. 36f 32m. 32f 28m. 2Sf cane' pulverisation to open further the fibre bundles. UiCC/B Chrysotile is a well described reference sample [47], Exposure Conditions Pmkcnon et al [39] and McConnell et al. [31.32] have described 1 At each time point. 4 animals were sacrificed for morphometneal analysis. At time zero. 5 animals were sacrificed for morphological analysis. in detatl the exposure conditions, including inhalation chambers and dust generation. Bnetlv, exposures were ofthe `occupational* type, 7 h/day. 5 days/week for 12 months, the fibres being aerosolised by a Timbrell generator. Total chamber dust miss concentrations were monitored daily with adjustments as required to maintain a constant dust concentration throughout the exposure period. The mean values (mg-m*J) far each chrysotile preparation (n = 240) were: 11.36 2.18 for Jeffrey, 10.99 t 2.11 for UICC/B, and 7.78 1.46 forCoa- The latter included most of the pulmonary neoplasms plus a random linga as described by Pmkcnon et al. [45], Methods used to deter selection of diverse, non-neoplastic lesions. Other records from the mine the mean chamber respirable dust concentrations and the com Becton Dickinson Research Facility from Dr Steven Bemadvne and plete results of these determinations, including PCM- and SEM- the NIEHS from Dr Elizabeth Ford and Mr Don Gouia were also denved fibre size determinations of all three fibre types, have also reviewed. been described by Pinkerton et aL (39,45], Direct and indirect TEM analyses ofCoalinga have been described by Chatfietd and tlgren [in Aiumais preparation). Four hundred 5-week-old, specific pathogen-free (SPF) male and female Fischer 344 rats (weighing 125-150 g), originally obtained Animal Sacrifice. Tissue Distribution, and Sample Preparation from a single source (Chas. River, Inc. Wilmington, Mass.), were Pinkerton et al. [39| and McConnell etaU [31.32] have described shipped to the NIEHS (Research Triangle Park, N.C., USA). Three the method of animal sacrifice, tissue distribution, and sample prep hundredandthirtyof these form the basis of this study. Upon receipt aration. All animals had a complete post-mortem examination. The at the NIEHS, 5 animals of each sex in each exposure group of the lungs of animals on lifetime test for morphological study were fixed 560 rats were randomly selected and killed for morphological study. ' in 10% neutral buffered forraol saline. Random sections of the left This included total body necropsies which found these animals to be lung; a slice from the right diaphragmatic lobe; and random sections free of infectious and non-infections disease ofall major organs. The through the rest of the right lung (apart from the cardiac and middle rats, thus found to be suitable for the inhalation study, were released lobes) were taken and stained for cells with haematoxylin and costa to the Becton Dickinson contractor at 6 weeks ofage and 40 animals (HAE). for connective tissue with Van Giesen's stain (HVG), and in ofeach sex were randomly assigned to either a control group orone of some cases, for mucin with Aldan blue. Lungs of interim sacrifice three exposure groups, at 7-8 weeks of age. A total of 96 rats out of animals to be studied for morphological evidence of fibrosis using the 320 were used for morphometric study: 4 miles and 4 females either the Wagner scoring method or a 'severity/extent' scoring sys were randomly selected from each group of 40 male rats and 40 tem (see below), were also fixed in formalin. Replicate sections ofthe female rats following 3 and i 2 months' exposure to either (1) Coalin- left lung were taken from the same area (mid-dorsal to ventral slices ga Mine chrysotile; (2) UICC/B chrysotile. or (3) Jeffrey chrysotile. through the hilar region) of each animal and again stained with H&E Four additional males and 4 females from each exposure group were and HVG. The methods used to fix and prepare (he lung tissues of randomly selected from the surviving animals l year following the interim sacrifice animals (four randomly selected sites from the ven end of 12 months' exposure to each type ofchrysotile. The remaining tral and dorsal portions of the cephalic and caudal regions of the left 224 animals not used for morphometry were kept on lifetime test as lung of each animal) for morphometric examination have been shown m table 1. described by Pinkerton et iL [391. Due to the paucity of normal life-span data for F344 rats. 1,200 5-week-old SPF male and female animals were set up as lifetime con Histopathologtcal Analysis trols to help interpret the NIEHS NTP study described in this report Animals on lifetime test were analysed bistopathologically as {sec 42 for results and also 43 analysis of a subset (3.548/ described by McConnell et al. ([31,32] (table 2) for the presence and 4,640) of the same 'historical' control group plus 44 for additional, degree of reaction to the fibres, using grades 1-8 as described by relevant, concurrent control data]. Wagner et al. (48). The grading severity is: 1 = normal; 2-3 = cellular change (no fibrosis); 4 = minimal fibrosis; 5 = mild fibresis;6 = mod Chrysotile Preparations erate fibrosis: 7-8 = severe fibrosis. Pinkerton et al. [45] have described (he origin and physical char The interim sacrifices were scored for severity and extent based acteristics of aerosols produced from the 3 chrysotile samples. Chat- upon the number of cells and amount of matrix present within the Held and tlgren (in preparation] have also provided further details of intetsiithun of each particular structure. Severity scores ranged from the physical and chemical properties ofthe Coalinga fibre, referred to 0 to 3+, 0 signifying no change and 3"** signifying the greatest change. 266 Indoor Built Environ 1997;6:264-276 Ileren/Chaifield UCAREF00024352 TabU 2. Parenchymal fibrosis scores of rats on lifetime test treated with Coalinga. LTCC.'B. and Jeffrey chrysolite graded with the Wagner (severity) system Sex ill1 n2J Av. fibrosis score Comrol/NIEHS. 19965 Control/MRC. 1984* Coaling*/NIEHS. 1996 UICC/NIEHS. 1996 UICC/MRC, 1984 Jefirey/NIEHS. 1996 m 28 22 1.4 f 26 21 1.6 m. f 34 ns 1.0 m 27 21 1.9 r 24 17 2.0 m 30 25 5.4 i* 26 22 5.3 m. f 68 os 4.3 >-1.7* m , 26 25 5.6 f 26 24 5.9 1 n l = Number ofanimals at nsk at the statt of the study. : n2 = Number of animals that were/could be used for fibrosis scor ing. * N1EHS, 1996 present investigation. 1 MRC. 1984 = Wagner et al. (44]. ** UICC'B data for 500-749 days5 and 750+- days* given but not stratified for sex. Table 3. Parenchymal fibrosis (matm) and cellular changes in rats sacrificed at 3.13 and 24 months treated with Coalinga. Jeffrey, and U1CC B chrysotile graded semi-quantitativcly with a Seventy/ Extent sconng system n Cells and matrix Cells and matrix within the peribron around chial interjtttium alveolar ducts Controls 5 months 12 months 2-1 months Coalinga 5 months 12 months 24 months CICC/B 5 months 12 months 24 moaths JetTrey 5 moaths 12 months 24 months 8 0.00 0.00 8 0.03 003 8 0.00 0.00 8 0.34 0.24 3 0.16+0.05 S 0.28 0.16 8 0.53+0.23 8 1.75 +0.37* 8 1.03 0.25* 8 0.81 0.23 8 2.38 0.60* S l.750.I6* 0.00 0.00 000 + 0.00 0.03 0.03 0.S8+0.27 0.75 0.22 0.75 0.35 2.13 0.40* 2.25 0.16* 2.38 0.52* 2.63 0.60* 5_250.49*-b 3.33 0.42* Extent scores were based on involvement ofless than 1/3 ofthelung section. 1/3 to 2/3 of the lung section and greater than 3/3 ofthe lung section for scores of l. 2 and 3 respectively. The results were expressed as the product of the scores for severity and extent {331. Finally, although the difference in the percentage of Coalinl and Canadian-treated animals (22%) which could not be scored due to age-related lesions was probably treatment-related, there is no rea son to believe that the fibrosis scores ofthe Coalinga-ueated animals which could not be scored would be higher than those which could be graded. Thus, there is little reason to question their representative ness. ` Morphometric Analysts This was performed as described by Pinkertoa et *L [39). Statistical A nalyses Statistical analysis of survival was performed using onc-taikd tests for variance. Statistical analysis of morphometric findings was conducted using a two-way ANOVA test as described by Pinkerton et al- {39J. Results Morphological observations, based on histopathological study of animals on lifetime test treated with the three types of chrysotile. demonstrate that Coalinga chrysotfle * p < 0.05 for comparison ofthe asbestos-treated groups to the cor responding age-matched controls; - p < 0.05 for comparison ofthe group exposed to Jeffrey chrysotile to the corresponding age-matched groups exposed to U1COB and Coalinga chrysotile; also see Pinkerton [33]. . is not fibrogenic (table 2) in contrast to the Canadian fibres, both of which produced significant fibrosis. Using a semi-quantitative, histological scoring system to assess the extent and severity of fibrosis found around bronchiolar walls and alveolar ducts, Coalinga chrysotile failed to produce fibrosis at any of the three sacrifice intervals, the measured parameters not being significantly greater than controls at any time point. In contrast, both Canadian fibres induced significant fibrosis (table 3) that was also associated with adenomatoid change or bron chiolar metaplasia [15] characterised by air spaces filled with a mucin like material [see also 33] and alveoli lined with a cuboidal to columnar epithelium. The interstitial matrix underlying these cells was thickened. These changes were not seen with Coalinga chrysotile [see also 33]. Electron microscopically, the lungs ofCoaiinga-treated rats sacrificed at 3-, 12-, and 24-month intervals did not display a thickened basement membrane or other atten- Short Fibre Coalinga Chrysotile Lack of Fibrosis (ocour Built Environ I99".6--261 UCAREF00024353 MALES 3 MONTHS EXPOSURE SOO r FEMALES hVOTMfeum Endotfwlum CONT COM, UCC JEFF CONT COAL UfCC JEFF 12 MONTHS EXPOSURE 1000 Fig. 1. Changes in TITV in nus acriftced at 3. (2, and 24 months treated with Coalinga. ietfrey, and UICC/B chr> sottle. CONT = controls: COAL * Coalinga: L.1CC = UICC/B; JEFF Jeffrey. Each bar represents the mean of 4 animals and is sub* divided into the epitheliaL intemiiiaL and endothelial compartments; 'p < 0.05 for comparison with age-matched controls: see also Pinkerton [33). CONT COAL U(CC JEFF 12 MONTHS EXPOSURE * 12 MONTHS AIR 1000 i m CONT COM. U1CC JEFF dant changes. By contrast, prolonged exposure to UICC/B and Jeffrey chrysotilc clearly resulted in a generalised thickening of the basement membrane underlying much of the alveolar epithelium of the lung with notable mem branous deposits of an amorphous, electron-dense materi al and concentrically layered, calcium-phosphorus bear ing rings, along with marked collagenisation of the inter stitial matrix [see also 33]. Such qualitative EM findings were morphometrically confirmed using data parameters related to the whole interstitium: the non-cellular and cel lular components of the interstitium; and its individual cellular constituents. Changes in the thickness of the whole interstitial compartment are shown in figure 1. These are referred to as alterations in `total interstitial tis sue volume' or TITV. The TITV has two major compo nents: non-cellular (or `interstitial (matrix) volume' IMV) and cellular. Since the chief morphometric indica tor of fibrosis is an increase in the non-cellular compo nent, such IMV changes are of particular importance (ta ble 4). Changes in the total volume of the cellular compo nent of the interstitium are displayed in table 5 and are referred to here as alterations in `cellular interstitial vol ume (CIV). Data are also presented for changes in the number and volume of the individual interstitial cells within the CIV. The former are presented in table 6 as interstitial cell number (ICN) while the latter (fig. 2) are 268 indoor Bmh Environ 1997:6:26-4--276 Ugren/Chatfidd UCAREF00024354 Table 4. Changes in noncclluiar IMV in rats sacrificed ai 3. 12. and 2-1 months treated with CoaJinga. Jeffrey, and UICC/B chrysotile Table 5. Changes in CIV in rats sacrificed at j. 12. and 24 months treated with Coalmga. Jeffrey, and LICCB chrvsotile Time. Sex months Total volume. mmVboih lungs Control Coalinga UICC/B Jeffrey Time. Sex months Total volume. mm*/both lungs Control Coalinga UICC/B Jeffrey 3 12 12+ 12 air m r m r m c 126=16 30*10 149= 13 I09!l 178 = 20* 153=3* 195=22* 82*5 177*3* 134=11* 163=13 98=5 239*15* 230=25* 362=27**-' 138=9 146=10 265=21**-' 287=38 168 = 15 439 5*-* 422=80* 211=8*-* 282=-22*-3 J 12 12+ 12 air m r m r m f 67 = 5 59 = 7 57 = 8 49*4 57 = 7 56=6 124*2* 97 = 9 77*3 47*6* 31 21 53*9 104=5* 72=10 80=16 62 = 10 I2" = 2* 4<} = 4 131 = 15* 140=24* 145=1* 138*11* 168*54* 107= 16*5 All data mean t SEM. n <= 4 for each group and time period: I p < 0.05 when comparing the age-matched control values using Duncan's multiple comparison test after first testing for significance using a one-way ANOVA: II p < 0.05 when comparing groups exposed to Jeffrey chrysolite with all other asbestos-exposed groups ofthe tame time period: ' p < 0.05 when comparing one treatment group to its correspond ing treatment group ofthe preceding time point; * p < 0.05 when comparing 24 month values to (hose at 5 months. See text for explanation. .All data SEM. n = 4 for each group and time period. * p < 0.05 when comparing the age-matched control values using Duncan's multiple comparison test after first testing lor significance using a one-way ANOVA: * p < 0.0S when comparing groups exposed to Jeffrey chcysotile with all other asbestos-exposed groups ofthe same time period: 0 p < 0.05 when comparing one treatment group to its correspond ing treatment group ofthe preceding time point. Table 6. Changes in the ICN (x 10*) in rats sacrificed at 3,12. and 24 months treated with Coalinga. Jeffrey, and UICC/B chrvsotile Time, months Sex Control 3 161=9 f 131 = 16 12 tn 131=9 r 118=4 24 m 130=15 r 136=3 Ref.1 38.39 38,39 38.39 38.39 38.39 38.39 Coalinga UICC/B Jeffrey Ret'.1 251 = 12* 170=4* 160=12 131 =65* 140=9 132*13 197=9 172=4* 274 = 39* 39 187 = 10* 39 163=125* 342 = 14*-9 39 118*6 233 =22*~* 39 200=13 199=13* 317 = 61*-* 39 191=4*-* 39 See text tor explanation. All data mean SEM; n > 4 for each group and time period; 1 p < 0.05 when comparing to the age-matched; control values using Duncan's multiple comparison test after lim testing fot significance using a one-way ANOVA; * p < 0.05 when comparing the group exposed to Jeffrey chrysotile with all other treatment groups; e p < 0.05 when comparing one treatment group to its corresponding treatment group ofthe proceeding time point; 1 References in which these data have previously been presented in pan or whole replicated here for the sake ofcomparison: Pinkerton <t at [38. table 7.J; Pinkerton et at. [39. table 31. referred to as interstitial cell volume (ICV). Finally, nu merical changes in the seven interstitial cell types namely fibroblasts, monocytes-macrophages, lymphocytes, neu trophils. mast cells, plasma cells, and pericytes axe dis played in table 7. The time course of events within the interstitium and its compartments during and following exposure to each of the three ebrysotiles is classified here as acute, subchronic, and chronic for alterations taking place after 3, 12, and 24 months' exposure, respectively [see also 33]. Shon Fibre Coalinga Chrysotile Lack of Fibrosis Indoor Built Environ 199*:=264-27fi 269 UCAREF00024355 Ftg. 2. Changes in ICV m rats sacrificed at 3,12. and 24 months treated with Coalinga, Jeffrey, and UICC/B chrysolite. Each point represents the pooled mean of 4 males and 4 females: see also Pinkerton [331. Over the first 3 months of exposure, acute changes were generally noted within the non-cellular and cellular. interstitium in response to all three types of chrysotile. After 12 months' exposure, Coalinga failed to significant ly increase the interstitial matrix volume over the IMV values noted after 3 months' exposure (table 4) even though these values were nonetheless significantly in creased above controls for males, but not for females (ta ble 4). By contrast, the IMV values noted in ail Jeffreytreated animals were significantly and dramatically in creased above those seen at 3 months, over age-matched controls, and above the other two treatment groups as well. Subacute IMV changes in UICC/B-treated animals resembled those seen with Coalinga. The attendant cellu lar responses (see above) for the 3 fibre types at 12 months appeared to be less dramatic than those found in the noncellular components. Still, the cellular changes (e.g. CIV and ICN) in the Coalinga-treated females, despite contin uous exposure from 3 to 12 months, nonetheless fell very significantly during this time period. In contrast, highly significant and dramatic subacute cellular increases were noted following exposure to Jeffrey fibre though again most of the non-cellular and cellular interstitial responses elicited by UICC/B did not differ significantly from those produced by Coalinga. By 24 months, Coalinga was the only fibre that failed to produce a persistent, significant increase in. IMV over age-matched untreated controls. Similarly, the Coalinga associated cellular responses (see above) were also not significantly different from controls. In marked contrast to Coalinga. both Canadian fibretreated groups displayed very marked, non-cellular and cellular interstitial changes by 24 months. Moreover, a unique change was observed in the IMV of the UICC/Btreated males dunng the post-exposure period with values 4-5 times greater than those seen in either untreated con trols or in animals exposed to Coalinga or Jeffrey chryso tile. In a manner similar to IMV, the cellular responses (e.g. CIV. CIN and ICV) and the TH'V remained signifi cantly above control values in UICC/B- and Jeffreytreated groups in almost every case. Finally, the quantita tive changes in the individual, cellular components of the interstitium in rats sacrificed at 3-, 12-, and 24-month intervals assessed morphometrically (UICC/B data not available), generally paralleled those seen in the tissue components just described. Coalinga again failed to pro duce a significant increase in the number of any intersti tial cell type by the end of the study whilst Jeffrey did so dramatically in virtually every case being significant for fibroblasts, monocytes-macrophages, lymphocytes, and mast cells but not for neutrophils and pericytes. 270 Indoor 8uilt Envtroa 1997:6:264-276 Dgrcn/Chaifirid UCAREF00024356 Table 7. Changes in number of the individual cellular compo nents of the alveolar imerstitium (x 10*) m rats sacrificed at 3. 12. and 24 months treated with Coalinga. Jeffrey, and UtCC/B chrysoule " ' Survival Rates ofAnimals on Lifetime Test Controls. Four (2 male and 2 female) out of 53 concur rent controls died spontaneously by 24 months on test, a mortality level consistent with that (10-15%) noted by Control1 Coalinga Jeffrey Solleveld etal. [42|. The mean survival for untreated, con Fibroblasts 3 months 12 months 24 months * 94 8 86*9 82 9 Monocyte-macrophages 3 months 21 4 12 months 202 24 months !93 Lymphocytes 3 months 12 months 24 months 82 2l 94 Neutrophils 3 months 12 months 24 months ND ND ND 133*7* 939 86 + 9 71 4*-" 39 3* 24 I02 52 93 5*1* 5*2* 2*2 108*9 146*16*-* 13719" 106*27* 132* I2*-C 89*19*-' 5t 10*4 61*30*-* 2*1 82*- 3 1* current controls dying after 24 months in this studv was 809 days (604-967) for males (n = 28), 712 day's (482 967) for females (n = 25), and 767 days for both sexes combined. The difference in length of survival between the concurrent males and females just failed to attain sta tistical significance fp = 0.07). Treatment Groups. Eleven treated animals (Coalinga. I male and 2 females; LTICC/B. 3 females; and Jeffrey. 2 males and 3 females) died spontaneously by 24 months for reasons unrelated to treatment The difference in sur vival between males and females just failed to reach statis tical significance (p = 0.06) for the Jeffrey-treated animals whereas no sex-related differences were noted in length of survival in the Coalinga- and the UICC/B-treated ani mals. Coalinga failed to reduce survival duration com pared with controls whilst Jeffrey nearfy did (p ** 0.06). Mast cells 3 months 12 months 24 months Plasma cells 3 months 12 months 24 months Pericytes 3 months 12 months 24 months ND ND ND ll 11 11 20+4 123 13*4 ND ND !*l 5*2* lit 2l 18*5 7*2 7*2 ND 10*1*-* ;*!* 8*3*' 2l 4*3 12*4 U1 10*2 Parenchymal Fibrosis andSurvival The survival times for control and Coalinga. male and female animals on lifetime test with combined scores of 1 and 2, were compared with U1CC/B- and Jeffrey-treat ed rats with scores of 6 and 7. Cases with pulmo nary tumours, severe teukaemic infiltration, or marked uraemic involvement of the lung were excluded from this analysis. The severe fibrosis due to the Canadian fibres thus significantly reduced survival (p = 0.04) whilst the minimal cellular changes associated with Coalinga expo sure did not. See test for explanation. All data are mean * SEM; n 4. ND = not detected within the tntetstitiuin; 1 p < 0.05 for comparisons to the age-matched controls; * p < 0.05 for within group comparisons to the previous time point; ` p < 0.05 for comparisons to animals exposed to Coalinga Mine chrysotile. 1 Data previously presented in Pinkertcu et aL (33, table 8] in part or whole. Pleural Disease Treatment-related pleural lesions, Le. pleural plaques (6. 49J, diffuse (visceral) fibrous thickening (6, 50], and benign pleural effusion / pleuritis [51] were not found in this study. Age-Related Lesions Uraemic pneumonitis and its attendant oedema, cellularicy, and membrane formation were sufficiently severe in 1.8% (8/208) ofcases so as to make reading of the slides impossible. Leukaemia, autolysis and congestion were also sufficiently severe enough to make scoring of the slides impossible in 6.0% (13/208), 2.0% (5/208), and 0.4% (1/208) of the animals respectively. The percentage of age-related lesions (leukaemia, uraemia, and conges tion) revere enough to make reading of the slides impossi ble amongst the four groups was as follows; 22% (11/50) for Coalinga, 14% (7/53) for untreated controls, 8% (4/55) for UICG'B; and 0% for Jeffrey. There was no difference between the sexes. A small percentage ofslides were miss ing 2.0% (5/208). Short Fibre Coalinga Chrysolite -- Lack of Fibrosis UuSoor Budi Environ 199t'.6'-754--276 271 UCAREF00024357 Tabl 8. Summary of selected inhalation bioassays using `non-Coalinga' and Coaimga chrysotlies Treatment Average cone.1 rag-m-' Fibre, % >10 pm Fibre-tnl'1 Fibre-mi"1 Interstitial (>5 pm) per average fibrosis cone.' (`extent' score)2 Tumour response %n ShortLongChry. yam2 WDC yam2 Fac WDC2 Exp WDC2 Exp WDC (IL rev.)2 Coalinga RG* Coalinga COF256 10 10 4.3 4.6 48 57 5.6 5.9 3.3 27 35 40 50 35 90 90 (20)5 (23)* 1.170 5.510 428 679 468 108 III 131 100 117 551 100 148 98 19 20 2^ 30 2.4 (0.3-6 7) 12.6(6.4-24) 8.8(3.1-17) 12 3(2.2-41) 12.1(0 3-35) 9 6(0.5-22) 10.8(0.6-22) 0 0 20 53 40 44 51 44 44 0 1.8 3/40 22/40 16/40 18/40 21/40 18/40 18/40 0/50 2/90 1 Average chamber dust mass concentration. 2 Davis and Jones [24]. 1 Davis et al. [52]; WDC is wet dispersed chrysotile: chrysotile treated with amoaic surfactant to break down fibrils. Fac WDC was dust collected from the air m a WDC yarn factory. Etp WDC was an experimental sample not prepared for commercial use: (It. rev.) in this experiment the animals were light reversed. . * Muhle et ai. [511; RG resin grade. 2 The Coalinga samples contained long fibres artificially produced by a spinning process. These aggregates broke down when wetted to material containing 0% fibres >10 pm. 4 The present study. COF25 cyclone overflow (25 mm final port): see Pinkerton et al. [45]. Note: RG and COF25 are dimensionally similar materials. Discussion In conclusion, our findings demonstrate that long term, high-dose exposure to Coalinga fibre, a short, amphibole-free, chrysotile, does not induce parenchymal fibrosis in contrast to two long Canadian samples which are both clearly fibrogenic. These results were based upon our own analysis of animals on lifetime test plus careful review of interim sacrifice data provided by Dr. Kent Pin kerton. The latter had been analysed both morphological ly using light microscopy and morphometrically with elec tron microscopic methods. The Findings of previous stud ies. described bdow, support these conclusions. Other Studies with Coalinga Chrysotile The investigation by Muhle et aL [51 ] is the only other inhalation study of Coalinga chrysotile and it too failed to find fibrosis. In this investigation, fibrosis was assessed as `septal thickening', the level (42%, 21/50) in the Coalingatreated animals resembling that seen in the combined controls (36% i.e. 12% (6/50) for sham- and 24% (12/50) for untreated controls). The interpretation of these find ings is made somewhat difficult by the fact that the authors failed to indicate whether cellularity or fibrosis made up the greater part of the observed septal thicken ing. The starting materials used in this study (COF25 fibre, [45]) and those used by Muhle et al. [51] (RG144 preparation) were not identical. However, this is not con founding since extensive analyses of COF25 and RG144 have not demonstrated significant differences in their fibre size distributions. [Chatfield and Ilgren, in prepara tion]. The observed lack of fibrosis was not simply due to low [ca 131 fibrcs-ml*1 >5 pm) concentrations of long fibre since very comparable fibre concentrations [111 fibres-tnl"1 >5 pm] of other chrysotile preparations [52], administered over identical exposure durations, have been very fibrogenic (table 8). There are only two other investigations to assess specif ically the fibrogenic potential of Coalinga chrysotile ([53, 54] for Union Carbide and (55] for Johns Manville). Both used injection methods and fibrosis was induced in each instance. However, the observed fibrotic lesions were clearly non-specific, being called `artifactual (due largely to high doses and also) attributable to the technique* by Gross and de Treville [55]. Further confounding was created by inadequate controls; contaminated starting materials; pulmonary infection; a paucity of animals in each treatment group; the failure to study histologically all ZTZ indoor Built Environ 1997:6:264-276 ngrcn/Chaifieitl UCAREF00024358 but a few animals; and various non-specific histological reactions e.g. 'foreign body reactions' wrongly attributed to Coalmga fibre [53. 54|. Other Comparable UICC/B and Jeffrey Inhalation studies Although McConnell et al. (31. 32) examined animals from the same UICC/B-treated and untreated control groups as we did for this report, their findings for animals kept on lifetime test were never published, only those for the interim sacrifices. The 24-month fibrosis score (3.6) reported by McConnell et al. [31. 32) for UICC/B animals is thus not strictly comparable to the lifetime scores (5.3 5.4) we report here (see table 2). However, McConnell et al. [31. 32] did describe progression in their investigation (see below), thereby making it most likely that their 3.6 score would have increased further by the end of life. Although the extent to which this would occur is obvious ly uncertain, it could very well come to within one grade or less of that found by us in this report. This is suggested by the fact that the lifetime fibrosis score reported by Wagner et al. [44) for identically treated, UICC/Bexposed animals was 4.7, which is less than one grade low er than those noted by us. This level of variation is not uncommon when individual, age - matched cases, are compared [e.g. see 23, 31. 32. 44; also noted in MRC Expt 121. 3 Oct. 'SO). Observed differences in fibrosis score are probably not due to dose and fibre size but may be related to other factors such as interreader variation, the choice of fixative and/or the mode of fixation (31). UICC/B-treated animals, 12 months after a 12-month exposure, had a mean fibrosis score of 5.0. There appears to be only one other inhalation rat study of Jeffrey fibre [56]. In this study. McConnell et al. exposed F344 rats for the same duration and intensity as the present investigation. These workers stated that grade 4 lesions typified the fibrotic changes [see 56. pp. 329. 331, and 333 as well as fig. 5j thus produced. However, these were by no means characteristic of the changes noted in the Jeffrey-treated animals reviewed in the present study which were clearly grade 5 and grade 6 lesions. Although it is presently not possible to reconcile the observed differences, the high proportion of minimal ly fibrotic lesions is surprising given the rate of neoplasia (40%) reported by McConnell et al. [56] and the recog nised association between the degree of asbestosis and extent of pulmonary neoplasia [57]. Whole body exposure was used in the experiments since the nose only technique was probably not in use at the time the study was being designed (ca 1976) and because the NIEHS study was to be modelled after the MRC unit which only used whole body exposures. .Al though whole body exposure has been criticized on sever al grounds including adventitious exposure, avoidance of exposure and ingestion of sample through grooming, the necessary comparative experiments have not yet been done to demonstrate whether this is actually the case. [Da vis, personal eontmun.] Moreover, the daily restraint associated with nose only exposure may, in itself, signifi cantly influence toxicity outcome [58], Overall Level ofParenchymal Fibrosis In a manner similar to this study, a substantial percent age of the rats exposed to UICC/B chrysotile by McCon nell et al. [31, 32) and Wagner et al. [44] also had grade 5 fibrosis scores (NIEHS-MRC raw data, not shown). Thus, 33% (5/15) of the UICC/B-treated males and 55% (5/9) of the UICC/B-treated females reported by McConnell et al. [31, 32] had grade 5 lesions (raw data, not shown). Simi larly, 28% (7/24) of the UICC/B-treated males and 32% (8/24) of the UICC/B-treated females reported by Wagner et al. [44] had grade 5 scores, with one grade 6 lesion noted in each sex. Wagner et al. [23,48] are possibly the only other group to administer UICC/B chrysotile to rats via inhalation in approximately the same duration and intensity as was used in the present study. Wagner et al. [48] noted that both sacrificed and surviving UICC/B-treated rats ulti mately developed fibrosis scores that averaged between grade 5.1 and 6.0 whilst Wagner et al. [23] found that Progression ofParenchymal Fibrosis Comparison ofour lifetime fibrosis scores for UICC/B treatment (table 2) with those noted at 12 and 24 months by McConnell et al. [31, 32] is consistent with minimal to moderate progression. This is in keeping with McConnell et aL [31. 32], who also observed a minimal amount of progression in the UICC/B-treated animals between 12 and 24 months. In contrast, progression was not reported during the post-exposure year in the UICC'B-treated ani mals studied at the MRC (grade 4.0 at 12 months: 4.1 at 24 months) by Wagner et al. [44], Wagner et al. [23, 48] did note an approximate one grade increase in animals exposed for 12 months [0.8 increase i.e. 4.3 at 12 months and 5.1 in survivors, tables 5,6 in 48]; [1.3 grade increase i.e. 3.7 at 12 months and 5.0 in the 24-month sacrifices, table 4 in 23]. The failure of Wagner et aL [44] to note progression might be related to other factors [e.g. inter reader variation, McConnell et aL. 31, 32], although the 2.5 grade increment found in those exposed only for short Fibre Coalinga Chrysotile Lacic of Fibrosis IndoorBiuh Environ 1997;6:264-276 273 UCAREF00024359 6 months (3.0 in the 6-month sacrifice and 5.3 in the those sacrificed 6 months after 18 months non-exposure [23J) tends to go against this idea. clear contrast to the Canadian fibres. Crapo et al. [34] morphometrically analysed a subset of the same tissues we assessed in this study but concluded that Coaiinga Sample variation within the UICC/B stock might ex could be `injurious', thus inferring that it was also fibro- plain some of these inconsistencies. Thus, Himes [58J genic. However, this claim was based solely on 12-month noted that since the UICC/B sample 'was the most popu data neither stratified for non-cellular interstitial changes lar (research material, the) original stock ran out and was nor correlated with the histological absence of fibrosis. replaced by a new stock, which turned out to differ yet By 24 months, the volume of non-cellular interstitial again from the original standards in the number and the matrix, the chief morphometrical indicator of fibrosis, in distribution of fibres'. Others have found the UICC stan the Coalinga-treated animals, fell to levels that no longer dard chrysotile samples to be less active than expected differed from (hose of the age-matched controls (table 6). probably due to 'overmilling' making them `too short to The statistically significant, morphometrically detectable, produce a maximal tissue response' [52]. Perhaps the increases in non-cellular 1MV seen in the Coaiinga treated more recent LfiCC/B samples were even more extensively males at 3 and 12 months were thus only transient in milled than the earlier ones thus further reducing the nature. These were therefore, most likely, non-specific, number of long fibres and their attendant fibrogenic `nuisance* dust type reactions [see 59]. potency. The morphometric data almost certainly underesti Pinkerton [33] proposed, on morphometrical grounds mate the observed differences between the Coaiinga- and (see table 4), that progression only occurred in the UICC/ Canadian-treated tissues because of sampling bias due to B-treated animals possibly due to `the presence of fewer exclusion ofareas ofbronchiolar metaplasia seen solely in long fibres in the lungs of animals exposed to UICC/B the tissues treated with Canadian fibre. Selective sam chrysotile (that) resulted in a slower rate of lung fibrosis pling could therefore have seriously underestimated the which continued to progress during the post-exposure actual amount of morphometrically detectable `fibrosis' period'. This led Pinkerton to surmise that `The greater in the Canadian-treated groups since the excluded meta exposure to very long fibres (>50 pm) in the animals plastic regions `were probably the areas of most intense exposed to Jeffrey mine chrysotile stimulated a much injury' [39]. more intense initial infiammatory reaction which encased The observed differential response in the individual many of these fibres in layers of non-cellutar, collagenous interstitial ceil types was yet another important indicator interstitial matrix thereby reducing their inflammation of the difference in fibrogenic potential between the inducing potential. This would have allowed the inflam Canadian and the Coaiinga fibres. The failure ofCoaiinga matory reaction in these animals to have decreased in fibre to elicit a mast cell reaction is particularly relevant in intensity during the post-exposure year at an initially this regard since it has been said to be of importance in the slower but steady pace.' production of pulmonary fibrosis both in animals and In contrast to the UICC/B-treated groups, Pinkerton humans [60]. [33] (see above) and-Pinkerton et al. [391 failed to find progression with the Jeffrey fibre stating `that the intersti Survival. Parenchymal Fibrosis, and Competing tial fibrotic reaction (with the Jeffrey fibre) did not signifi Causes ofDeath cantly progress during the post exposure period'. We do Fibrosis with the Canadian fibres significantly reduced not have the morphological data to confirm or refute this. survival but only in those animals with the highest fibrosis However, it.is consistent with the failure by McConnell et scores. The present report may actually be the first to al. [56] to demonstrate progression with this fibre type at demonstrate such a fibrosis-related reduction in survival. least for severity since they did not assess extent. Thus, Wagner et al. [23, 44], McConnell et al. [31, 32], Davis and Jones [24], and Davis et aL [16, 52, 61,62] do Morphological and Morphometrical Correlation of not discuss fibrosis and survival whilst Wagner et al. [48] Parenchymal Fibrosis state that `the effect was very slight'. The natural inci This is the first standard, inhalation asbestos bioassay dence of pulmonary parenchymal fibrosis in untreated to be rigorously assessed both morphologically and mor F344 rats is very low. No spontaneously occurring cases phometrically. The findings thus derived are, in turn, were noted in this study, in the investigations of Wagner both complimentary and consistent, fully supporting the et al. [44] and McConnell et aL [31,32], or by others [42, proposal that Coaiinga chrysotile is. not fibrogenic, in 43,63,64]. Taken together, the above clearly suggests that :74 Indoor Built Haviroa 1997;6:264-276 ngren/Chatfieid UCAREF00024360 fibrosis, in itself, rarely, if ever, causes death in expenmental animals exposed to asbestos. This is not too sur prising, since it is not typically a cause ofdeath in humans either, many of whom are also smokers. Progressive dif fuse fibrosis, as produced by exceptionally high doses such as those used in the very early studies (65), could produce death directly by cor pulmonale. An indirect reduction in survival also might occur either by infection or an enhanced likelihood of tumour formation [57] though the lethality of such neoplasms has been ques tioned [31]. Acknowledgements We are very grateful to Dr. Kevin Browne. MO; Professor John Craighead. MD: Dr. John Davis, DSc; Professor Edward Gaensfer, MD: Professor Ulrich Gruber. MD; Dr. Arthur Morgan. DSc Profes sor Brooke Mossman, PhD; and Dr. William Weiss, MD for kindly providing critical comments on this manuscript. Dr. Morgan also reviewed many ofthese data at an earlier time m great detail when he and the Senior author were kindly afforded the opportunity to dis cuss the morphometric findings with Professor Pinkerton in person at the University of California at Davis m June, 1995. We are also grateful to Prolessor Kent Pinkenon and his stafffor their hospitality during this visit. We would like to thank Dr. Gene McConnell for providing earlier versions of (he pathology data and for discussing these with the Senior author. Dr. Gary Boorman, who with Dr. McConnell reviewed the earlier pathological analysts of many of these materials, also kindly discussed his earlier work with us. We would like to thank (he NTP archive scientists including Dr. Mdvm Hamlin. Mr. Davtd Bridge, and Dr. Robert Sits and (heir staff who very graciously assisted m locating the lifetime test materials and allowed the Senior author to study these with theirassistance, helpful comments and support. We would also like to thank Dr. Jerry Hardisty at the EPL laboratories. Dr. Benue Atkins and Dr. Dan Mclaunn of Bccton Dickinson. Dr. Jack Moore of the Institute of Environmental Health. Mr. Dennis Mundav of the MRC external unit. Dr. FinasCavendar, and Dr. Elizabeth Ford and .Mr. Don Goulaofthe NIEHS procurement division for their assistance in attempt ing to locate ail ofthe original data underlying this study and its colla borative. comparative investigation. Finally, we would also like to thank Dr. J.C. Wagner lor his very kind assistance in reviewing a number of the slides of (he animals on lifetime test, lor discussions about the study and its origins, and tor insightful comments concern ing the scientific bases of the observations made herein. Again, we are most grateful to all ofthose just mentioned for their help but it is we, not they, who will take responsibility for any errors or defects in the work. Similarly, acknowledgement of assistance docs not neces sarily imply agreement with the conclusions expressed. References 1 Stanton Vt. Wrench F: Mechanisms of meso thelioma induction with asbestos and fibrous pais. J Natl Cancer Inst 1972:48:797-821. 2 Sttmoo M. Layird M.Teteris A. Miller E, May M. Morgan E Smith A: Relation of parade dimenlion to carcinogenicity m amphibole as bestos and other fibrous minerals. 1 Nad Can cer (nu 1981.67-965-975. 3 Pott F- Asbestos use and careioogeneity in Ger mans and a comparison with animal studies. Ann Occup Hyt 1994:38J89-600. 4 Mossman B. Bignon J, Corn M. Seaton A. Gee JBLc Vsbesros: Scientific developments and implications for public policy. Science 1990; 24:294-301. 3 Davis J: Mineral fibre caremogenesis: Experi mental data relating to ihe importance of fibre type. size, deposition, dissolution and migra tion: in Bignon J, Peto J. Sarracci R (erisk NooOccupational Exposure u> Mineral Fines. Lyon. lot. Agency for Research oc Cancer. t989.ix)l90. pp 33-43. 6 HEI (Health Effects Institute - Asbestos Re search) Asbestos in Public and Commercial Buildings: A Literature Review and Synthesis of Current Knowledge. 1991. 7 HEI (Health Effects Institute - Asbestos Re>earchr Asbestos in Public and Commercial Buildings: Supplementary Analyses ofSelected Data Previously Considered by the Literature Review Panel. 1992. 8 Corn M: Airboroe concentration* ofasbestos in aan-occupattonal environments. Ann Occup Hyg 1994;4:495-502. 9 Schrier H: Asbestos to the natural environ ment; m: Studies in Environmental Science. No. 37. Amsterdam. Elsevier, 1989. 10 Langer AM. WoUTMERohl A. SdOrntri: Vari ation of properties of chrysolite asbestos sub jected to milling. J To* Environ Health 1978:4: 173-188. 11 Churf A: .Asbestos hint burden and disease pat terns in man; in Ribbe P fed): Health Effects of Mineral Dusts. Reviews in Mineralogy. Wash ington. Mincrakifical Society of .America. Boakcrafters lnc_ 1993. rot 28. chap 13. pp 410-431. 12 Case B; Biological indicators of chrysotde ex posure. Ann Occup Hyg 1994:38:4:503-518. 13 Lee Rt .Asbestos contamination in paraffin tis sue blocks. .Arch Pathol Lab Med 1995:10:528 532. 14 PiggRtUsesofduysotile asbestos. Ann Occup Hyg 1994:38:4:453-458. 15 Vorwald Ai. Durkin TM, Pratt PC: Experi mental studies of ubatoat. AMA .Arch Ind Hyg Occup Med 1951:3:1-43. 16 Diva J. Beckeu S. Bolton R_ Codings P. Midd leton A: Mast and number of fibres in the pathogenesis ofasbestos related king disease m rats. Br 1 Cancer 197107:673-688. 17 Berman W, Crump K-Chatfirid E. Davis J. Jooes A; The sues, shapes and mineralogy of asbestos structures (hat induce lung tumours or mesotheliomas m AF/HAN rats foBowing in halation. Risk .Analysis 1995:15:181-195. IS Lippmann M: Asbestos exposure indices. Re view. Environ Res 1988:46:86-106. 19 TtmbreQ V: Deposition and retention ofasbes tos fibres in die human lung. Ann Occup Hyg 1982:26:347-369. 20 TimbreUV: Fibres and carcinogenesis. J Occup Health Soe Austral 19830:3-21. 21 Timbretl V: Pulmooary deposition and retenuon of South African amphibolc fibres: Identi fication of asbestosis-related measure of fibre concentration. Proc 7tb bat Pneumoconiosis Conf. Bochum. 1983 vol 2. pp 998-1008. 22 Timbreli V, Ashcroft T, Gotdneia B. Heyworth F. Mcurmsn LO. RendtU REG. Reynolds 3A. Shilkin KB, Whitaker O: Relationships be tween retained amphibote (Bites and fibrosis to human half specimens; in Walton W fed): inhaled Parades VL Oxford. Pergamoa. 1987. 23 Wagner JC, Berry G. Skidmore JW, Poofey FD: The comparative effects of ibree dtrysotiles by injection and inhalation in rat Biotofical Effects of Mineral Fibres. 1ARC Sd Puhl 1980:363-372. 24 Davis JM. Jones A: Comparisons of the patho genicity of loog and sbon fibres of cbrysotile asbestos in rats. 8rJ Exp Pathol 1988:69:717 737. Short Fibre Coalinga Chrysolite Lack of Fibrosis Indoor Built Environ 1997:6--64-276 275 UCAREF00024361 15 Plaice S Groih D UInch C. Slcnlcr L. Finncll M. Sioil M Chronic innaiaiion of short asbeslos fibres rand Appl Toxicol 1985:5 327>40 -6 Ilgren E. 5VagnCr J. The xalucot expenmenial animals in analyzing the hazards ofexposure 10 mineral fibres II. Consideration of natural background incidence of mesotheliomas and non-physiologicai imraperiioneal technique Submission to ihe German MAK Commission on Mineral Fibre Safety. 1989. 3? Lemaire I Characienzanon of ihe bronchoalcoiar cellular response in expenmentai asbestosis Am Rev Respir Dis 1985:131:144-149 38 Lemaire I: an assessment of Ihe tibrofemc potential of very short 4-T-30 chrysottle by inirairacheal instillation ut rats. Environ Res 1985:36:314-326 39 Lemaire I. Selective differences in macrophage populations and monokine production in re solving pulmonary granuloma and fibtosu. Am J Pathol 1991:138:487-495. 30 Lemaire I. Dionne P. Nadeau D. Dunmgan J Rat lung reactivity ip natural and man-made fibrous silicates following short-term exposure. Environ Res 1989:48:193-210. 31 McConnell EE. Wagner JC. Skidmore 5W. Moore JA. A comparative study of (he fibragenic and carcinogenic effects of UICC Cana dian chrysolite B asbestos and glass microlibre (7M 100). m. Biological Elfects of Man-Made Mineral Fibres (Repots on a WHO/tARC meeting). Copenhagen. 20-22 .April 1982. (An nex 1984:45:234-242). 32 McConnell E. Wagner J. Skidmore J. Moore): Two inhalation studies: A comparison, in: Bio logical Effects of Man-Made Mineral Fibres (Report on a WHO/tARC meeting), Copenha gen. 20-22 Apn) 1982. (Annex 198428:118 120). 33 Ptnkenon 1C Lung reaction lo chronic inhalal ion of three types ofchrysolite asbestos during ihe lifespan of the Fischer 344 rat; PhD thesis Duke University, 1912. 34 Crapo J. Barry B, Brody A. O'Neil J: Morpho logical. morphometric and X-ray microaniiytical studies on tuag tissue of rats exposed to chrysolite asbestat in inhalation chambers. I ARC Sci Mono. Bid Effects Miner Fibers 19803:273-283. 35 Pinkerton K. Pratt P. Brody A. Crapo 3: Lung injury patterns as a result of exposure to differ ent chrysolite asbestos fibres (abstract). Am Rev Resp Dis 1981:123:21. 36 Pinkerton K. Piatt P. Crapo J: Morphologic lung changes following exposure to chrysolite (abstract). Am Rev Resp Dis 1981:121251. .7 O'Nttl J. Pinkerton K. Crapo C: Lang vdume changes in rats exposed lo chrysolite asbestos (abstract). .Am Rev Resp Dis 1981:123:146. >S Pinkerton K. Barry B. O'Neil J. Raub J. Pratt P. Crapo i: Morphologic changes in ibe lung Junng the lifespan of Fischer 344 rats Am J Anal 1982.164 155-174 39 Pinkerton K. Pratt C. Brody AR. Crapo IO Fiber localization and its relationship io lung reaction in rats afterchronic inhalation ofchry solite asbestos Am 1 Pathol 1984:117 484 498 40 Pinkerton K. Plopper CG. Meroer RR. Roggli V[_ Patra AL. Brody AR. Crapo ID: Airway brandling patterns influence asbestos fiber lo cal ion and Ihe extent oilissue injury in the pulmonarv puenchvma. Lab Invest 1986.55:688 695. 41 Pinkerton K. Young S. Fram E. Crapo J- Alveo lar it pe II ceil responses to chronic inhalation of chrxsotite asbestos in rats Am i Respir Cell Mol Bid 1990:3:543-552. 42 Solleidd H. Haseman J. McConnell E: Natural history of body weight gam. sumvaL and neo plasia at the F344 rat. J Natl Cancer Inst 1984; 72:939-940. 43 Goodman DG. Ward JM. Squire R. Chu K. Lmban M: Neoplastic and non-ncopiastic te nons in ageing F344 rate. Toxical Appl Phar macol 1979:48237-248. 44 Wagner JC. Betty GB. HtU Rf. Mundiy DE. Skidmore JW- .Animal experiments with MM.M(V)F - elfects of tnhalanoo and intra pleural inoculation in rats; in: 8iotogicat Ef fects ofMan-Made Mineral Fibres (Report on a WHO/IARC meeting). Copenhagen. 20-22 April 1982. (Annex 1914:42210-2353. 45 Pinkerton K. Brody A. McLaunnD. Adkins B. O'Connor R. Pratt P, Crapo: Characterization of three types of chrysonte asbestos after aerosoluanon. Environ Res 1983:3122-53: ' 46 Campbell W, Huggins C, Wybe A: Chemical and physical characterixaiioo of amosite. chrysoute and non-fibrous tremoMc for oral inges tion studies by ibe National Institales of Envi ronmental Health Sciences. U S Bureau Mines Rep invert No. 8452.1980. 47 Timbred V: Characteristics ofthe (nlematioaal Union against Cancer standard reference sam ples of asbestos; in Shapiro H (edk Pneumo coniosis Proc Int Conf, Johannesburg. London. Oxfotd Universily Press, 1970. p 28. 48 Wagner JC The effects of the inhalation of asbestos in rats. Br J Cancer 197429:252-269. 49 Gaeastef EA: Asbestos related pleural plaques: Much ado about very little. DR1 Asbestos Med icine Seminar. Scottsdale, 1991. section H. 50 Gibbs A. Stephens M. Griffiths D. Blight B. Pooley ^ Fibre distribution in the lungs and pleura af subjects of asbestos related diffuse pleunl fibrosis. Br J Ind Med 1991:48:762 770. 51 McLoudTC. Woods BO.CarringtonCB.Epter GR. Gacnsler EA Diffuse pleural thickening in an asbestos-exposed population: Prevalence and cuotogies Am J Roentgenol 1985:144 9 18 52 Muhlc H. Pott F. Bellman B Takenalu S. Zicm U' Inhalation and injection experiments m rats io lest the eareinogcmcity of MMMF Ann OcoipHyg 1987J3I-755-764 53 Davis J. Addtsoo J. Bolton R. Donaldson. K. tones. A. Inhalation and injection studies in rats using dust samples from chrysolite asbes tos prepared by a wet dispersion process. Br J Exp Pathol 1986.67:113-129. 34 Mellon Institute: Report No. 39-55 produced for the Union Carbide company 1966 faccessed from Union Carbide on request) 55 Mellon Insulate: Report No. 34-70 produced far ihe Union Carbidecompany 1971 (accessed from Union Carbide on request). 56 Gross P, de TreviUe U The Pulmonary Re sponse io Coalings Asbestos Dust: A Prdntu- nary Investigation (Final Report). (Produced far the JM company) lad Hvg Foundation of America. 7-1968. , $7 McCooaefl E Adkms B: Studies on thechronic toxicity (inhalation) of woliastonitc in Fischer 344 tats. Inha) Toxicol 1991.3223-337. 58 Davis J. Cowie B: The retanooshtp between fibrosa usd cancer in experimental animals exposed to asbestos and other fibres. Em iron Health Perapcct 1990:88205-309. 59 Elmes P: Conflicts in evidence on the health effectsof mmesai fibres. Chapter 23; in Liddd! D, Miller Kfedsk Mineral Fibres in Health and Disease. Boca Raton. CRC Press. 1991. pp 321-337. . 60 Oberdonter G: Evaluation and use of animal models to assess mechanisms of fibre carcino genicity. [ARC Sci Pub! 1996;140:107-127. 61 Wagner M. Wagner J, Griffiths M: Mast cells and inhalation ofasbestos to rats. Thorax 1984; 39239-544. 62 Davis JM: The effects of mtenzuttenc high asbestos exposure (peak dose levels) on the lungs of tats. Br J Exp Pathol 1980:61:272 - 279. 63 Davis JM: Experimental lesions tn rail corre sponding to advanced human isbesosis. Exp MotPithol 1986:4207-221. 64 Sass B. Rabnetn. Madison R, Nims R. Peters R. Kelloff G: Incidence of spontaneous neo plasms in F344 rats throughout the natural life span. J Natl Cancer lost 1975:54:1449-1453. 65 Coleman G, Barthold S. Oshildision G. Foster S. Jooas A: Pathological changes during ageing ra Barrier-reared Fischer 344 mate rats. J Ge rontol 197722258-278. 66 Wagner, J: Asbestosis in experimental animals. BrJ Ind Med 196320:1-12. 276 Indoor Built Environ 1997:6264--276 Ugrcn/Qmfidd UCAREF00024362