Document G6bL9jv5y5BDz4qyr7y0R2Ezv

Br. J. Cancer (1974) 29, 252 their : fcransf THE EFFECTS OF THE INHALATION OF ASBESTOS IN RATS for thi Home J. C. WAGNER, G. BERRY, J. W. SKIDMORE and V. TIMBRELL were histok From the, Medical Research Council's Pneumoconiosis Unit, Uandough Hospital, Penarth, Glamorgan of dus As able si Received 5 September 1973. Accepted 29 November 1973 of dus assess, Summary.--Two experiments in which SPF Wistar rats were exposed by inhalation value to dust clond-Tot-the tl MJC st^dOTd~referenctrsgmples for"periods of between one the r. was u which conBAUfid..~tQlDrogresa-flfter..remoYal from.,exposure but only a litGe fibrosis al., 1 was observed in control rats. Lung tumours, ranging in seyefity^ffdm adendmata ' Af td~St}uam<ms'eafelnematar were-produeed by aiT samples but in the controls there threes werermiyafewadegomatg-afirdwgeoftjirmgfeseriqijstumour?. -Ofthe20tumours suppli whiciThietaSfa9lzedrI6_bccurfed"'after exposure to one or other of the 2 chrysotile ehami samples. In addition, a..total of 11 mesothellomata occurred, 4 of which were with separi crocidolite and 4 with Canadian chrysotile. Two of thelnesotKettOTHata occurred prieta Wtthroaiy-enedayVexposttfeTo asbestos-There was a positive association between water asoestosis and- ' killint it diei Wagner, Berry and TMbrell (1973) The asbestos samples used consisted of full n< reported the results of experiments in which rats were inoculated intrapieurally with samples of asbestos. In the dis cussion it was mentioned that two experi ments in which rats had been exposed to dust clouds of the UICC reference samples had been carried out, and the results of these are now presented. Some preliminary results of one of these experi the 5 UICC standard reference samples (Timbrel!, Gilson and Webster, 1968) which were prepared following recommendations of the UICC. These samples are of 3 amphibole types--amosite, anthophyllite and crocido lite--and 2 ehrysotiles--a Canadian and a Rhodesian sample. The rats were exposed in I >4 m3 inhala tion chambers (Timbrell el al,, 1970) which contained 8 cages, each of which could hold H micro, killinj anaes the t.' The 1. in for right exam dilate ments were reported by Wagner (1972). 6 rats, or, for a short period with young were In. both experiments the rats were rats, there was room for a seventh. Five exam exposed to similar dust concentrations chambers were used, one for each type of had 1 I and the dose varied by exposing rats for various lengths of time. The main aim was to establish the relationship between the development of malignant tumours asbestos. The chambers were constructed so that the rats could be tended without the chambers being opened. The dust clouds were generated using a specially devised dispenser (Timbrell, Hyefct trimn remai the c F. lungs in the lungs and the dose and type of and Skidmore, 1968). The clouds were form) asbestos dust but, additionally, the generated for 7 hours a day and 5 days per sagitl amount of fibrosis was assessed. week. The respirable dust concentrations whoh were measured using size selective gravi lobes metric dust samples (Cassella Type 114A). other MATERIALS AND METHODS The collected samples were evaluated at the end of each daily session. In order to other exam Caesarean derived rats of the Wistar achieve the required dosage, calculated as Ii strain were used which had been bred at the product of concentration and time, haen: the Unit from SPE stocks given to us by variations occurring in the concentrations the Imperial Chemical Industries, Pharma were corrected on the following days. At were lagen ceutical Division at Alderiey Edge, Cheshire the end of exposure the rats were left in the cases in 1964 and 1968. chambers for a few days, to allow time for A by inhalation ` between one ed asbestosis little fibrosis m adenomata ontrols there le 20 tumours e 2 chrysotile tch were with lata occurred ttion between d consisted of rence samples r, U)68) which imendations of of 3 amphibole and croddonadian and a inhalawhich ternnTe<o7uuhld hole, d with young eventh. Five each type of e constructed d without the rated using a imbreli. Kyett clouds were nd 5 days per concentrations lective srravi- Type fl4A). evaluated at In order to calculated as n and time, oncentratior..ing days. At ere left in the dlow time for EFFECTS OF INHALATION OF ASBESTOS IN BATS 253 their fur to become clear of asbestos before transferring them to a clean environment for the remainder of their lives. At transfer some rats, generally *2 or 3 of each sex. were killed and the lungs removed for histological examination and determination of dust content. As-each type of asbestos has a consider able silicate content (about 50%) the amount of dust in the lungs was determined by first assessing its silica content and referring this value to that of rhe respirable dust to which the rat had been exposed. This method was used in previous experiments (Morris et al,, 1967). After exposure the rats were caged in threes or fours isolated in a special unit supplied with filtered air. The inhalation chambers were also in this unit but in a separate room. They were fed on a pro prietary brand of autoclaved cubes, and water ad libitum. Except for the scheduled killings, each rat was allowed to live until it died or appeared to be ilistressed, and a full necropsy examination was carried out. Histological preparation, staining and microscopic methods.--For the scheduled killings, animals were killed by chloroform anaesthesia and following exsanguination the thorax opened and the lungs removed. The left lung was air inflated and suspended in formalin. Representative portions of the right lung were taken for electron microscopic examination and the remainder of the lung dilated with neutral buffered formalin. Slices were taken from both lungs for histological examination. After the histological sections had been cut. the embedded tissue and the trimmings were dewaxed and added to the remainder of the lungs which were used for the chemical estimates. For other animals, at post mortem the lungs were dilated with neutral buffered formalin and after fixation were sliced sagittally, routine sections being taken of the whole left lung and the upper and lower lobes of the right lung. In addition, any other suspicious lesions from the lungs or other organs were taken for histological examination. In all cases, sections were stained with haematoxylin and eosin and the lung sections were stained for elastin, reticulin and col lagen. Special stains were used in some cases as an aid to diagnosis of the tumours. Assessment of the severity of asbestosis.-- Sections of both lungs were examined without knowledge of the duration or type of asbestos exposure and with the animals in random order. The sections were observed on a viewing screen of a Projectina Microscope 4013 BK. using a <7 objective. At this magnification a large proportion of the lung could be assessed in a single field and. as it was not possible to observe asbestos fibres at this' magnification, the results were not biased by knowing the type of asbestos to which the animal had been exposed. Experiments In both experiments and for all doses there were groups which were exposed to all 5 reference samples; there were also control groups which were not exposed. Rats were allocated to treatments at random. At the start of the exposure the majority of the rats were between 5 and 7 weeks old. a few being slightly older or younger, and there were approximately equal numbers of males and females. Experiment 1.--There were 2 time inter vals of exposure: in the first, rats were exposed for 3 months starting in May 1967 and in the second, groups of rats were exposed for one day only in August 1967.' In addition to the killings at the end of exposure, in the 3-month group there were also intermediate sacrifices at 5, 8 and 10 weeks. Experiment 2.--There were 3 time inter vals of exposure--6 months, 12 months and 24 months.. The experiment started in January 1969 and after 6 months half of the rats were removed from the cabinets. They were replaced and a year later these replacement animals were in turn removed. They were replaced by animals to be used for special,electron microscopy examinations which will not be reported in this paper. The remaining animals were removed in January 1971 after 2 years' exposure. In the 6-month groups, in addition to the killings at the end of exposure, rats were also killed after 2 years, i.e. 18 months after removal from exposure. The numbers of rats are given in Table I. In the 24-month groups, overcrowding amongst the males tended to occur due to increase in size after about a year, and some were removed prematurely after 134 months. For analysis these rats have been m xz .wi tflW AM Elr > Length nl` exposure l bay :! months ti months 12 months '24 months Airostt> 49 02 24 25 21 Authophyllito 49 .>2 24 2S 19 * 'mculnlftt* 4!l .">2 Ir'T 2t> 20 i 'hrvsofile 49 r*2 24 23 24 i 'hrysoule (Rhodesian 1 Control 49 48 52 >S 25 1 27 ) W 20 J Length of exposure l day :t months rf months 12 months 24 months Table II.---Mean Respirable Dust Concentration (mgjm3) and ('nnmlative Dose (mgjm3 hours) Amosito .. (4-1 12-4 11-2 10-8 (0-6 99 .i)00 ssso 17000 .`>3500 Anthophvilito 12-8 13-5 109 11-4 10-6 90 .>240 8540 17100 33700 Crocidolite 12-5 (2< 9 10 - 7 10-6 10'3 ss 0030 8430 17000 33200 Ohrysotile (Canadian) 9-7 12-1 LU-2 10-7 10-1 88 4931) 8240 1710O 33200 Chrysorile (Rhodesian 1 14-7 12-3 to-7 10-9 (0-1 103 0070 8390 17100 33000 included in the 12-month groups and hence the number of rats in these groups is slightly higher, and in the 24-month, groups slightly lower, than planned. '' [n Table II the mean respirable dust concentrations and the cumulative doses, the products of concentration and time, are shown. The one-day exposure was 7 hours for all dusts. For the other 4 time intervals of exposure, there were slight variations in the number of hours required to achieve approximately equal doses but the mean times were 402, 788, 1574 and 3237 hours respectively. The mean concentrations were usually higher in Experiment 1 than Experi ment 2. and the 3-month group had an average dose of 80% of that of the 6-month group. Reasonable equality of dose between the dusts was achieved for all the lengths of exposure, except for the one-day which was too short to allow any adjustments. Interpretation of histological findings Classification of asbestosis.--The lesions seen in the lungs of rats exposed to all types of asbestos were similar to those described in guinea-pigs (Wagner, 1963. 1965). There were 2 main differences; firstly, asbestos bodies were never seen in the lung tissue of the rat although they are frequently seen in the pleural granulomata which follow the intra-pleurai inoculation of amphibole fibres; secondly, there was a far greater production of granular pneumocytes (type El) alveolar epithelial cells in the rat. The lesions consist initially of a deposi tion of asbestos fibres, alveolar macrophages and cell debris in the alveoli arising directly from the respiratory bronchioles. These deposits become organized firstly by being enmeshed in a thin reticulin network which coarsens with time and becomes replaced by collagen fibres. The alveolar epithelium reacts with replacement of the type I cells, and becomes completely lined by granular pneumocytes. In some of the alveoli the epithelium is shed into the lamina, in others there is a walling of the alveoli by these cells. This usually occurs at the bifurcations where groups of alveoli are closed off from the lumina of the respiratory bronchioles, giving the so-called pseudo-acinar ap pearance. In the guinea-pigs these small cystic spaces contained asbestos fibre and degenerating macrophages, but in the rats numerous granular pneumocytes were also present. The initial lesions were confined to occasional discrete individual respiratory bronchioles scattered throughout the lung substance. After further exposure, more and more respiratory bronchioles become involved and all the respiratory bronchioles arising from terminal bronchioles become thickened as the ribrous tissue network extends into the wall of the respiratory bronchiole, and this interstitial reaction spreads down into the peripheral elements of the primary unit, involving the alveolar ducts, atria and finally the air sacs and alveoli. With progression, the individual lesions tend to coalesce, leading to the development of a EFFECTS OF INHALATION' OF ASBESTOS IN BATS a deposirophages i directly These by being >rk which placed by pitheiium >o L cells, granular veoli the i~ j^ahers 'irUwWon.j off iron' uchioles, aar ap'e tall ibre and the rats ere also confined -piratory 'he lung core and involved arising sickened ods into <>le. and wn into primary atria - With tend to nfc of a Wro __Slight asbeatosis (Grade 4). Higher power showing numerous retractile crucidotite fibres m the-alvooli.of a respiratory bronchiole. Asbestos bodies are rarely seen m the lungs of rats. Illumination reduced to illustrate the fibres. H. & E. aOO. wr -30 - J. C. WAGYER, a. BERRY, J. W. SKIDMORE AYD V. TIMBRELL Fin. 3.--Moderate asbestoais (Grade 6). A lower power projection illustrating that the lesion although still mainly involving the respiratory bronchiole is now diffuse. Aggregations of type II pneumocytes are seen in the lumina as well as investing the walla. H. & JS. < 80. EFFECTS OF INHALATION OF ASBESTOS fX RATS diffuse interstitial fibrosis with gradual in crease in the density of the fibrous tissue, ultimately resulting in the replacement of most of the lung parenchyma hv a dense collagen network surrounding distorted air spaces, many of which contain large clumps of granular- pneumocytes which, in some areas, have lysed leaving foci of lipo-alveoiar proteinosis. The assessment of these lesions was based on 4 grades of fibrosis--minimal, slight, moderate and severe. Typical ex amples of slight, moderate and severe asbestosis are shown in Fig. 1-4. In addi tion, when carrying out the assessment it was found convenient to introduce the intermediate categories minimal/slight, slight/ moderate and moderate/severe. The 7 re sulting categories were scored 2-8 and the normal lung, in which there was no sign of asbestosis, was scored 1. and the assessments were averaged for each group of rats. The repeatability of this form of assessment was tested on 154 sections, reassessed in a different random order and 90% were assigned to within one category of the first reading. Classification of tumours.--The tumours found in the lungs of rats after exposure to the various types of asbestos dusts were peripheral adenomata, widespread adeno matosis. adenocarcinomata and squamous carcinomata. The rarity of pulmonary tu mours in rats has been stressed in the reviews by Kuschner and Laskin (1970) and Shabad and Pylev (1970). Further, these authors have described the development and morphological features of the tumours that we are reporting. All these tumours were peripheral and appeared to arise from the region of the respiratory bronchioles in which the asbestos fibre had accumulated. The origin of the adenomata appeared to be from accumulation of type II epithelial cells that proliferated in the alveoli of the respiratory bronchioles (Fig. 5. 6). In many of the exposed animals these tumours were multiple, and in a number of animals, particularly those with the more severe grades of asbestosis, there seemed to be adenomata arising from numerous adjoining respiratory bronchioles, giving an impression of contiguous adenomata invading large areas of the lung. Dr Harold Stewart (personal communication) suggested that this type of_lesion should be referred to as adenomatosis. In contrast to this, a few control animals were seen to have solitary adenomata which were small in size. The adenocarcinomata were of the same type and origin as the " alveolar adenocarcino mata " described by Shabad and Pylev (1970); many of these tumours were papillary adenoma-carcinomata. The squamous car cinomata appeared to originate from foci of squamous metaplasia occurring in the asbestotic lesions in the respiratory bron chioles (Fig. 7, 8). As far as can be ascer tained. all these tumours were peripheral in origin and not bronchial papillomata. The classification, of these tumours was discussed in some detail in Session Vt at the Gatlinburg Conference on the Morphology of Experimental Respiratory Carcinogenesis in 1970; and the chapter by M. F. Stanton (1974) in the IARC Monograph on Pathology of Tumours in Laboratory Animals contains detailed descriptions of the tumours that we have illustrated. The rats used in this experiment are from a caesarean derived, barrier maintained colony and fortunately they have been kept free of rat bronchitis: therefore the squamous metaplasia was not associated with bronchi ectasis. Mefcastases in the thoracic cavity to the chest wall, diaphragm, pericardium or the tracheo-bfonchial lymph glands were seen in 14 animals; $he majority had lesions invading 3 of the sites and in only 2 animals were metastases observed in the tracheo bronchial glands. In one animal secondary deposits were seen in sections from a kidney. Eight adenocarcinomata and 6 squamous tumours had metastasized. RESULTS All except 3-of the rats in the groups with exposure of 12 months or less survived for the whole of their planned exposure. In the 24-month group there was appreciable mortality before the end of exposure and only 53% survived for the full period. Out of 1013 rats it was impossible to obtain adequate histological material in only 8 because of cannibalism. Dust retention The mean weights of asbestos dust in the lungs of animals killed at the r.x 0.-WAGNER. G. BERRY. J. W. SKIDMORE AND V. TIMBRELL Fxo. 6.--Electron-micrograph of an adenoma showing typo II pneumoovtes. EFFECTS of IXHALATTOX <*F ASBESTOS IX RATS Table III.--Dust Retained in Lungs {mg} Length of Mean dose Chrysotile Chrysotile * exposure mg/m3 hours Amosite Anthophyllite Crocidolite (Canadian) (Rhodesian) 5 weeks 1 880 1-0 1-3 l-l 0-1 0-1 I S weeks 2430 0-9 1-8 1-8 0-1 0-3 ! 10 weeks 3290 2-0 2'8 2-1 0-5 0-5 j 3 mouths 5050 3-7 3-5 3-0 0-8 0-7 ! 6 mouths 8470 4-7 4-4 4-5 0-4 u-4 ! 12 months 17100 3-3 9-6 -3 0-8 1-4 24 months 33400 18-8 13-8 14-9 0-3 0-8 6 months -- 1-3 2-8 1-2 0-0 0-1 (after 18 months non exposure) Length of exposure l day 3 months ' 8 months 12 months 24 months Table IV;>--Mean Survival* after First Exposure {days) Amosite 804 771 763 692 SO? Anthophyllite 306 823 886 759 77S Crocidolite 795 317 788 776 738 Chrysotile (Canadian) 783 790 869 778 585 Chrysotile (Rhodesian) Control 753 803 857 793 786 ) 826 } 754 758 I i * Adjusted to be independent of sacrifices. X I scheduled times are given in Table III. More dust was usually found in males than in females and on average the female lungs contained only 70% as much dust as the male lungs. The values in Table III are the averages of the male and female the day the rats were first exposed are given in Table IV. The survival times have been estimated so as to be inde pendent of the sacrifices. The short survival of the 24-month group exposed to Canadian chrysotile was largely due 1 means. For the 3 amphiboles there was to 8 rats dying before Day 400 (in the a similar pattern, with an almost pro other four 24-month groups only 2 rats portional increase of lung dust with died before Day 400). These early deaths r\i dose. The 2 chrysotiles were similar to were not due to exposure, since 5 died one another but much less dust was due to an infection in one cage and 2 found than with the amphiboles; also the were killed in a fight. Discounting these s of chrvsotile figures did not show the same 8 deaths, the mean survival was 698 lif clear increase with doBe. The main days, which was still the lowest of the If features are summarized in Fig. 9. The 24-month groups. When the mean was \ dust in the lungs of the animals which taken over all the lengths of exposure, had 6 months5 exposure had been partially eliminated 18 months after removal from exposure. The proportions eliminated were 74% for amosite, 73% for crocidolite but only 41% for anthophyllite. How ever, the lower- -elimination of antho phyllite was not significantly different from the amosite and crocidolite figures. the Canadian chrysotile groups showed least survival, but only a month less than the control groups. The amosite and anthophyllite groups had mean sur vivals only a few days less than the controls, while the crocidolite and Rho desian chrysotile groups had longer sur vivals of 2 and 3 weeks respectively. Hence, there is very little indication that the exposure had any effect on the A t'c in 3 T ei o s< Survival overall survival of the animals. This is in marked contrast to our intrapleural The mean lengths of survival from inoculation experiments in which injection j i rrl Weight of dut in lungs (mg) 15 -1 EFFECTS OF INHALATION' <)F ASBESTOS IN' RATS if) I ,\>4 wed are d times ie inde1 short exposed ely due (in the mea and 2 ig these as 698 of the an was cposure, showed ith iess umosite an. surian the d Rholer suretively. lication on the i. This ipleurai ljection 10 *4 Ampbiboles After removal from exposure 0J Chrysotiles '0 S 12 --------------------1--1----------r 24 Time (months) JL 10000 20000 Cumulative dose (mg/ m3 hours) Fro. 9.--Mean weight of dost in lungs of rata in relation to 'dose and time. of asbestos reduces the expectation of Life by several months (Wagner el al., 1973)." Asbestosis The amount of asbestosis was assessed for ail the rats killed at scheduled times in Experiment 2 and after 8 weeks' and 3 months' exposure in Experiment l. There were 5 or 6 rats per treatment for each exposure, except that there were only 3 after 8 weeks. Overall the 2 sexes had similar amounts of asbestosis and they have, therefore, been combined to give the mean asbestosis scores in Table V, which are summarized in Fig. 10. Except for some inconsistency between 20 ......... : ' " the 3- and 6-month means, there was an increase of asbestosis with exposure for ail the dusts. Also, following 6 months' exposure, there was progression during the following 18 months without exposure for all the asbestos types, but these rats did not fare as badly as those which continued exposure. There were signi ficant differences between the asbestos types (P < 0-01): amosite invariably gave the least asbestosis throughout; anfchophyllite and Canadian chrysotile showed most asbestosis after 6 months' or longer exposure; crocidolite and Rhodesian chry sotile were intermediate. The mean asbestosis scores of the rats which were allowed to live out their lives are given in Table VI. Those rats 2(i2 J. C. WAUXER, G. KERRY. J. \V. KKID MORE AND V. TIMBRELL ;\sbe*to*l grade Moderate -i Slight -I Minimal H Nil J r o Controls 12 24 Time (month*) 1 1 1 '' i 10000 20000 30000 Cumulative dose (mg/nr* hours) Fto. 10.--Asbestosis in sacrificed rats in relation to dose and time. Length of exposure 8 weeks 3 months 6 months 12 months 18 months 24 months 8 months (after 18 months non-exposure) Table V.--Mean Asbeetosi# Scores* of Sacrificed Rata Amosite Anthophyllite Crocidolite Chrysotile Chrysotile (Canadian) (Rhodesian) 2-0 2-0 2-5 2-7 2-2 3-2 4-0 5-2 -- -- 4-3 6-2 3-2 5-0 2-0 2-0 2-7 2-8 2-7 30 2-8 3-0 2-8 4-3 4-3 4-3 -- -- --. 4-8 6-0 5-8 3-7 3-5 3-7 * l:nil. 2: minimal. 4: slight, 6: moderate, 8: severe. Control 1-3 1-3 1-2 -- 1-2 1-8 -- scheduled for 24 months have been divided into those that died before completion of exposure and those that survived for a period of non-exposure. The amount of asbestosis found in the rats exposed for one day was no more than that found in control rats. Comparing Tables V and VI for the rats which completed their exposure, progression had occurred be tween the end of exposure and death with all dusts, the single exception being the 3 months' exposure of Rhodesian chrvso tile. The rats which died before com pleting 24 months' exposure had more asbestosis than those sacrificed after 24 months' exposure for amosite, anthophyllite and Rhodesian chrvsotile. This was not the case for crocidolite and Canadian chrysotile, for which those rats that died during exposure had shorter mean survivals than for the other dusts. Meaned over all dusts, those rats that EFFECTS OF INHALATION OF ASBESTOS IN RATS Table VL--Mean Asbestosii Scores* of Survivors (Mean Survival in Months) Length of exposure I day :> months (1 months 12 months Cp to 24 munths 24 months Amosite 1-3 (20) 2-9 (25) 3 :s (24) 4-8 (23) 8-0 (23) 8-3 (28) Anthophyllito 1-3 (20) 3-2 (27) 4-2 (20) 6-0 (25) 8-4 (22) Crocidolite 1-2 (2b) 3-1 (27) 3-2 (24) 5-8 (26) 4-2 (14) 7-0 (28) 6-6 (29) Chrysotile (Canadian) L2 (25) 3-3 (20) 3-7 (20) 5-1 (25) 5-1 (16) Chrysotile (Rhodesian) 1-4 (23) 2*8 (28) 4-2 (23) ! (27) 8-1 (22) -- 6-8 (28) * I; nit. 2; minimal. 4: .slight, 6: moderate, 8: severe. 2rt3 <.lied during exposure had slightly more asbestosis than would be expected from the sacrifice rats, consistent with the more severely affected animals having the shorter survivals. However, the effect was very slight and. as observed earlier, the exposure did not affect survival to any extent. In Table VI there is again less asbestosis for amosite than the other dusts although the difference is not as large as in Table V. For rats which completed their exposure, the difference between amosite and the other 4 asbestos types had a mean of 0-7 for rats sacrificed and 0*5 for survivors. The results in Table VI do not support the findings in Table V that anthophyllite and Canadian chrysotile produce more asbestosis than crocidolite and Rhodesian chrysotile, and we conclude, therefore, that there were no important differences in the amount of asbestosis produced by these 4 samples. Tumours of the lung Lung tumours were observed in 247 of the rats exposed to asbestos. The total numbers of each kind for each dust are shown in Table VTI, where for those rats with more than one tumour of the lung, classification is by the more severe condition. No tumours of the lung were observed within 300 days of the start of exposure and therefore only rats which survived this initial period are considered to have been at risk. Apart from the scheduled killings, only 13 rats died within the first 300 days. There were 7 20 control rats out of 84 survivors in Experi ment I with adenomata, but in Experi ment 2 there were no lung tumours out of 42 control rats. There were slightly more male than female rats with tumours --128 compared with 119--but the oniv 2 tumour types for which there was any major difference between the sexes were adenocarcinoma and squamous carcin oma. Out of 50 adenocarcinomafca, 35 occurred in males whereas 30 of the 40 squamous carcinomata were in females. Metastases occurred in 20 rats. 10 of each sex. There were also 11 mesotheliomata (Table VII), 7 in males. Two of the mesotheliomata occurred with only one day's exposure, 1 with 3 months', none with 0 nlpnths', 6 with 12 months' and 2 with 24 months'. The meso thelioma which occurred with 3 months' exposure to crocidolite was a peritoneal tumour; the others were all of pleural origin. The distribution of the lung tumours with time after first exposure are shown in Fig. 11 for all dusts and all lengths of exposure except one day. In the 5 groups exposed to asbestos for one day there were 14 adenomata and, compared with the 4 in the corresponding controls, there was clearly no evidence that these adenomata were a consequence of expo sure to asbestos. There were 5 more serious tumours; 2 of these were meso theliomata, one with amosite after 715 days and the other with crocidolite after 551 days. There were also 3 adenocarcinomata, one with crocidolite after 2r)4 J. C. WAGNER. G. BERRY, d. W. SKIDMORE AND V. TIMBRELL Table YIL--Number of Animals icilk Lung Tumours or Mesotheliomata Exposure AmosUc. 1 dav 3 months 6 months 13 months 24 months Total N\ *r rats at ri.-jk* No. with Inn# tumour Type or' luntt tumour AdenoAdenoma Adenomatosis CHrcinomaf No. with .Squamous ineso- carcinomaf thelioma 45 3 3 rt 0 37 18 10 1 7 i 3 0 0 i 25 10 5 4 i 21 L3 3 1 3 146 38 19 8 0 01 00 00 00 60 6l .4 nthophyllite 1 dav 3 months 6 months 44 2 37 6 18 ... 6 12 months 28 20 24 months Total 18 145 16 50 6 3 9 > 22 (1 0 l 6 5 12 0 0 i 4 ID 3 8 (1) 0 0 1 1 6 8 0 0 0 i i 2 Crocidolitt I day 3 months 6 months 12 months 24 months Total Chrysolite (Canadian) 43 6 36 t4 18 4 28 18 18 13 141 . 55 44 5 10 2 5 4 26 0 .1 > 4 5 13 1 1 0 3 2 111 7(1) 0 1 (1) 0 6 2 (1) 9(2) l 1 0 o 0 4 l day 3 months 0 months 12 months 24 months Total 42 34 17 23 21 137 1 18 5 11 10 45 0 15 2 I 2 20 0 l (1) 0 0 03 20 00 10 3 6(1) I (1) 3 3 1 U) 4(2) 1 8 11 (3) 6(3) 4 Chrysotile (Rhodesian) 1 dav 45 5 4 0 1 00 3 months 6 months. 12 months 38 19 27 16 8 19 11 o 2 2 3 4 3 U) 3 (1) 7(2) 0 0 6(4) 0 0 0 24 months 17 11 o 1 5 (2) 5 0 ' Total 144 59 19 10 19 (6) 11 (4) 0 Control 1 day 44 4 4 0 0 3 months 40 3 3 0 0 6-24 months 42 0 0 0 0 Total 126 7 7 0 0 00 00 00 00 * Rata whioh survived at least 300 dayB after start of exposure, f Numbers in brackets are those with metastases. S07 days, another .with Rhodesian chrysotile after 719 days and one which metastasized with Canadian chrysotile after 838 days. In interpreting Table VII and Fig. 11, it has to be borne in mind that there was a greater tendency for rats to develop adenomata in Experiment l than in Experiment 2, as shown in the controls. Therefore, the higher proportion of ani mals with adenomata after 3 months' exposure than after 6 months' exposure is probably an artefact. We do not know why adenomata occurred in the < ho-ltoma. L / i) . 0 4 0 0 0 0 <J 0 nan m mtrois. of aninonths' cpoaure lo not in the 3 MONTHS AMOSITE EFFECTS OF INHALATION OF ASBESTOS IN RATS 6 MONTHS 12 MONTHS 26.> 500 750 1000 ANTHOPHYUITE 500 750 CROOOOtlTE WOO P 500 750 1000 Jd A 500 750 WOO SCO 750 WOO 560 "" 750 tooo 500 /50 WOO a 56? ^^o^iobo 500 -750 WOO CHRYSOTIIE Canadian a...q r~x 500 "3rWOO WOO 500 750 500 750 WOO CHRYSOTILE Rhodian Mi SOO 750 WOO WOO WOO - ji*. 500 *750 500 500 ! Squamous carcinoma 9 Adonoeordnoma 0 Adenoma/ adenomatosis Other causes Fio. II.--Distribution of survival times in days after first exposure. WOO controls in one experiment and not in the other but as the finding, is significant (P = 0-06) in its own right and is sup ported by the results from the exposed animals, it is unlikely to be due to chance. There was a higher incidence of tumours with 12 months' exposure than with 6 months' but little difference between the 12 and 24 months' exposure. Half of the 8 mesotheliomata in Experiment 2 occurred with Canadian chrysotile, so that in total crocidolite and Canadian chrysotile produced 4 mesothefiomata each. Of the 20 tumours which metastasized, 16 were after expo sure to a chrysotile (10 with the Rhodesian sample and 6 with the Canadian). Three others were with crocidolite and one with anthophyllite. Two of the meso theliomata in the 12-mouth groups occur red within 400 days after first exposure, one with crocidolite after 399 days and one with Canadian chrysotile after 355 days (the only rat which failed to survive for its scheduled 12 months' exposure). wpi^NMir'1 266 J.-G. WAGNER, G. BERRY, J. W. SKIDMORE AND V. TIMBRELL Asbestosis and lung tumours An analysis was carried out to deter mine whether there was any relationship between the grade of asbestosis and the presence of lung tumours. Since the asbestosis grade depends on survival, it was necessary to standardize to a constant survival' time. This was achieved by calculating the regression coefficients of asbestosis grade on survival time for rats without lung tumours, exposed for 3 months or more and with survival of at least 400 days after first exposure. Differences in these coefficients between the 5 types of asbestos and. the 4 lengths of exposure were not significant and the pooled coefficient of 0*00304 0*00065 grade units per day was used. The asbestosis grade of each rat was then adjusted using this slope to an arbitrary survival. The adjusted mean asbestosis grades were then calculated for each of the 20 groups, for those with and without lung tumours. In 15 of these groups the mean asbestosis grade was higher in the animals with lung tumours; in the other 5 groups the opposite occurred but only slightly so in 4 cases. The accuracy of an estimate of the difference in asbestosis between those with and those without a tumour is dependent on the number of anjmals in each category which varies between groups, and weighting each group to- take account of this gave a mean difference of 0*71 -- 0*13. Hence overall the animals with lung tumours had significantly (i5 < 0*001) more asbestosis than those without. Differences between the dusts were not significant but the wide range in means--anthophyllite 0*25, Canadian chrysotile 0*42, crocidolite 0*66, amosite 0*85 and Rhodesian chrysotile 1*22--shows that there are insufficient data to reach" any firm conclusions on this question. The groups exposed for only one day provide supporting evidence of a relation ship between lung tumours and .sbestosis. There was very little asbestosis in these groups (Table VI) and restricting atten tion to animals which survived for at least 600 days, the mean survival times of those with and without lung tumours were very similar. There were 17 lung tumours in 201 rats: only 6 of these occurred in 157 rats without asbestosis (3*8%) while 11 occurred in the 44 rats with minimal or slight asbestosis (25%), a highly significant difference (P<0*001). Tumours at sites other than lung A total of 412 tumours, other than lung tumours or mesothelioma of the pleura or peritoneum, were observed. The majority of these were adenomata of the breast or pituitary adenomata, which were common post-mortem find ings. Both of these adenomata occurred 4 times as frequently in females as in males. The numbers of these tumours, other benign tumours and malignant tumours for each type of asbestos are shown in Table VIII. For none of the Table VIII.--Number of Tumours at Sites Other than the Lung Amosite Anthophyllite Crocidolite Chrysotile (Canadian) Chrysotile (Rhodesian) Control Benign tumoura ,----------------- 1------------------ 1 Malig- Pituit- mint Breast ary Other tumours 22 40 4 8 16 38 4 15 20 31 4 10 23 26 7 7 19 37 9 3 20 34 2 13 tumour types was the difference between the control and the asbestos treated significant. In Table IX more detail is given of the sites of the tumours with all types of asbestos combined. The largest differences between the treated and con trol rats were for tumours of the ovary, 10 in treated and none in controls, and tumours of male genito-urinary organs, 11 in treated and none in controls. However, neither difference was significant. A few rats had multiple malignant tumours: 2 of the rats with mesothelioma of the pleura also had a lung carcinoma and one rat with a squamous carcinoma EFFECTS OF INHALATION OF ASBESTOS IN BATS i'O 7 Table IX. --dites of Tumours Other than Lung Asbestos treated Control Site/Tumour typo Digestive organa and. peritoneum Bone and akin Breast Ovary Other female genito-urmary organa Wale genito-urmary organa Intracranial Thvmoma Lymphoma, leukaemia Others Benign. Malignant 43 34 100 0 :i 1 s 33 173 1 l --s 53 thyroid mediastinum adrenal (4) salivary gland thyroid Benign. I ft {) o /. 34 ft -- 1 suprarenal Malignant 23 i 0 4 0 ft t> 0 of the lung had a mesothelioma tunicaMlfrom the worn hammer of the mill used vaginalis. Mesotheliomata of the vagin-7to produce the respirable fibre could alis were seen in this and one other rat: have been a factor in the causation of there is no evidence to suggest an associa the tumours. However, our results now tion with exposure to asbestos. Also, show that there is no need to invoke in some cases there were secondaries, for such a hypothesis to explain the high example, a synovioma had spread to the rate of lung tumours. lung. In all such cases tumours have The amount.ofLchrysotile j-etainedjn been classified by the primary site and the lungs did not show any clear increase in the few cases of multiple malignant with dose in rats exposed" fof~Ionger ' tumours there was no difficulty in recog than 3 months,! Iff two earlier expert-' - nizing the distinct types, i.e. one was meats" '(Wagner and Skidmore, 1965: not a secondary of the other. Morris et al., 1967) a higher airborne dust concentration was used to give a cumulative doep in 6 weeks similar to DISCUSSION that given in the present experiment Our finding that the asbestosis pro over 3 months. The weight of asbestos duced by exposure progressed after cessa found in the lungs of rats exposed to tion of exposure is in agreement with amphibole was 3 times greater than in human experience but contrasts with those exposed to chrysotile. In the the early inhalation experiments reported present experiments the ratio was 6 to 1 by Vorwald, Durkan and Pratt (1951) after 3 months, but increased with in which progression did not occur. continuing exposure as the weight of Wagner (1963) reported more asbestosis amphibole in"'the lungs continued to with amosite than with chrysotile in increase, but the amount of chrysotile guinea-pigs, rats and monkeys but our did not. The previous experiments had experiments show that of the TJICC shown that the rate of elimination of standard reference samples amosite is the dust from the lungs was much greater least fibrogenic in rats. for chrysotile than for the amphiboies. Gross et al. (1967) found lung cancers The present results may be explained on in 25 of 72 rats which survived 16 months' this basis, the weight of chrysotile having exposure to chrysotile dost at a mean reached equilibrium level, i.e. the rate of concentration of 86 mg/m3 for 30 hours elimination equalling the rate of retention. a week. They considered that contami There are a number of features of the nation of the asbestos by trace metals results presented above which we found 268 J. c. WAGNER. G. BERRY, S. W. SKIDMORE AND V. TIMBRELL surprising. First, in Experiment I two of sites other than the lung is equivocal. mesotheliomata occurred with the one Although an association with gastro day exposure compared with only one intestinal tumours has been found epi- with the 3 months' exposure, which had a demiologicallv, it is not yet regarded as dosage more than 50 times greater. dearly established (Selikoff, Hammond If the incidence of mesotheliomata was and Churg, 1972; Xewhouse, 1974), and proportional to dose, as is indicated for is of lower magnitude than the excess the inoculation experiments (Wagner et lung cancer risk. Our experiments pro al.. 1073), then the probability of such vide no support for such an association. an extreme result occurring by chance The experimental work of Graham and would be about 2 in a 1000. Graham (1-967) suggested that intra- Secondly, there was no evidence of peritoneai injection of tremolite asbestos either teas carcinogenicity or less asbesfosis .. could produce ovarian tumours, but the in" tKe groups exposed to chrysotile than follow-up of women asbestos workers those~exposed to the'"amphiboles/ `even, reported by Newhouse et al. (1972) pro though the"amdunts of dust in the- lungs duced no definite conclusions on this were so different. In particular, the question because of the rarity of the TJICC "Canadian chrysotile produced as tumour. Our experiments do give some many mesothehonaata aS the UICC ui'PCf- support to an association between asbestos cbtjtttfer The 2 UICU samples of chrysotile exposure and ovarian tumours as well produced 12 of the 14 tumours with as tumours of~~bhe male genito-urinary metastases. However, much less dust 'system. .Although neither was significant^ was retained in the lungs of rats exposed this could be because of the relatively to chrysotile than amphiboles (Fig. 10). small size of the control group. To Moreover, after intrapleural inoculation overcome this, we have included the the risk of a mesothelioma occurring control rats from some of our other with UICC crocidolite is 3 times the risk experiments, thus increasing the non- with chrysotile (Wagner et al., 1973). treated group to 403 rats of the same Therefore, allowing for the greater reten strain. This larger group contained 2 tion of crocidolite after inhalation, we malignant tumours of the ovary and 5 might have expected the risk with eroci- tumours of the genito-urinary tract in doUte to have been of the order of 20 males. In over 700 rats exposed to times that of chrysotile. asbestos there were 10 ovarian tumours, .Two of the mesotheliomata occurred 7 of which were malignant, and 11 within 400 days of the start of exposure. tumours of the genito-urinary tract in This may be compared with our injection males. Hence, based on the larger set experiments in which only 20 out of 803 of controls, the association between asbes occurred within 400 days (Wagner and tos exposure and ovarian tumours is Berry, 1969; Wagner et al., 1973). Also, weak and non-significant, whereas there is the earliest mesothelioma occurred after no support for an association with tumours 355 days and we observed only 3 within of the male genito-urinary system. this period in our injection experiments. The UICC chrysotile samples are finer The positive association between as- than the chrysotile which has been used bestosis and lung-tumours which we have in industry in the past. However, there established in the animals is in agreement is a trend for industry to use finer chry with epidemiological findings (e.g. Minister sotile (Wright, 1969) and so the experi of Labour and National -Service, 1949; mental results may be more relevant to Knox et al., 1968; Elmes and Simpson, the eurrent situation than to the past. 1971). We are investigating the effects of inhala The failure to establish any association tion of chrysotile in more detail in an between asbestos exposure and tumours experiment involving UICC Canadian ! t i } t! St jf'-ocal. o- >iircieci as tmmond. T-i), and 0 excess uts proociation. iatn and fc intra asbestos but- the workers '72} proon this of the ve some asbestos as well -urinary .tiificant, elatively up. To ded the ir other he nonhe same w osed to umours, and 11 tract in rgcr set n asbes' IOUT3 is < there is tumours I. are finer en used t, there e-r ehry experiwant tc fie past, f inhalaii in an "anadian EFFECTS OF INHALATION OF ASBESTOS IN RATS 2S9 ehrvsotile, a grade 7 sample from a Canadian mine and the superfine sample which proved the most carcinogenic of the materials which we inoculated intrapieuraliy (Wagner et al,, 1973). The experiments we report have given results which in several respects cor respond to those found in man. Thus, this experimental method is established as a valid tool for the investigation of the biological effects of asbestos. We are grateful to all our colleagues who over a number of years were res ponsible for the daily attention necessary in carrying out the experiments. We would also like to acknowledge our thanks to Dr Harold Stewart of the National Cancer Institute who advised us on the classification of tumours. We are also grateful to our former colleague, Dr A. Walter, who had classified the non-lung tumours occurring in some of our earlier experiments which we mentioned for comparison. REFERENCES Elmes. P. C. & Simpson, M. J. C. (1971) Insulation Workers in Belfast. 3. Mortality 1940--96. Br. ind. Med., 28, 228. Graham. J. <& Graham, R. (1967) Ovarian Cancer and. Asbestos. Envir. Ret., t, 115. Gross, P., ds Trevillb, R. T. P., Tosher, E. B., Kaschak, M. & Babyak, M. A. (1967) Experi mental Aabestosis. The Development of Lung Cancer in Rats with. Pulmonary Deposits of Chrysotile Asbestos Dust. Archs envir, Slth, 15. 343. Knox, J. F., Holmes, S., Doll, R. & Hill. I. D. (1968) Mortality from Lung Cancer and Other Causes Among Workers in an Asbestos Textile Factory. Br. J. ind. Med., 25. 293. Kuschnbr, M. & Larkin, S. (1970) Pulmonary Epithelial Tumors and Tumor-like Proliferation in the Rat. In Morphology f Experimental Respiratory Carcinogenesis. Proo. Conf. Gatliaburg 13-16 May 1970. EcL P. Nettesheim, M. G. Hanna, Jr. and J. W. Deatherage, Jr. TT.S. Atomic Energy Commission, Symposium Series, No. 21. p. 203. Minister or Labour and National Service (1949) Annual Report of the Chief Inspector of Factoriesfor the Year 1947 (Cmd. 7621). London: HMSO. Morris. T. G-, Roberts, W. H., Sjxverton, R. E., Skidmore, J. W. Wagner, J. C. 4 Cook, G. W. <19671 Comparison of Dust Retention in Specific Pathogen Free and Standard Rata~ In Inhaled Particles and Vapours II. Ed. C. N. Davies. Oxford: Pergamon. p. 205. Xewhouse, M. L. (1974) Cancer Among Workers in the Asbestos Textile Industry. In Biological Effects of Asbestos. Lyon. 2-5 October 1972. Ed. P. Bogovski, J. C, Gilson, V. Timbrell and J. O. Wanner, (IARC Scientific Publications, No. 8). In print. Newhocse, M. L.. Berry, G-. Wagner. J. C. 4 Ti."rok, M. E. (1972) A Study of the Mortality of Female Asbestos Workers. Rr. J. ind. Med., 29. 134. Selikoff. I. J., Hammond, E. C. 4 Chcro, J. (1972) Carcinogenicity of Amosita Asbestos. Archs envir. HUh, 25, 183. Shtabad, L. M. & Pylev, L. N. (1970) Morphological Lesions in Rat Lungs Induced by Polycyclic Hydrocarbons. In Morphology of Experimental Respiratory Carcinogenesis, Proc. Conf.. Gatlinbitrg, 13-16 May, 1970. Ed. P. Nettesheim, M. G. Hanna, Jr. and J. W. Deatherage Jr. U.S. Atomic Energy Commission. .Symposium Series No. 21. p. 227. Stanton, M. F. (1974) In Pathology of Tumours in Laboratory Animals. Vol l. Tumours of the Rat. Part 2. I.A.R.C. Scientific Publications. No. 6. Lyon: International Agency for Research Cancer. In preparation. Timbrell, V., Gilson, J. C. 4 Webster, I. (1968) UICC Standard Reference Samples of Asbestos. tnt. J. Cancer, 3, 406. Timbrell, V\, Hyett, A. W. & Skidmore, J. W. (1968) A Simple Dispenser for Generating Dust Clouds from Standard Reference Samples of Asbestos. Ann. occup. Hyg., II, 273. Timbrell, V., Skidmore, J. W., Hyett, A. W. & Waoner, J. C, (1970) Exposure Chambers for Inhalation, Experiments with Standard Reference Samples of Asbestos of the Inter national Union Against Cancer (UICC). Aerosol Sci., I, 215. Vorwald, A. J., Durham, T. M. 4. Pratt, P. C. (1951) Experimental Studies of Aabestosis. A.M.A. Archs ind. Hyg., 3. 1. Wagner, J. C. (1963) Aabestosis in Experimental Animals. Br. J. ind. Med., 20. 1. Wagner, J. C. (1965) The Sequelae of Exposure to Asbestos Dust. Ann. N.Y. Acad. Sci., 132, 691. Waoner, J. C. (1972) The Significance of Asbestos in Tissue. In Recent Results in Cancer Research, Vol. 39, Current Problems in the Epidemiology of Cancer and Lymphomas. Ed. E. Grundmann. and H. Tulinius. New York: Springer-Verlag. p. 37. Waoner, J. C. & Berry, G. (1989) Mesotheliomas in Rats Following Inoculation with Asbestos. Br. J. Cancer, 23, 567. Waoner, J. C., Berry, G. & Timbrell, V. (1973) Mesotheliomas in Rats Following Inoculation with Asbestos and Other Materials. Br. J. Cancer, 28, 173. Waoner. J. C. & Skidmore. J. W. (1985) Asbestos Dust Deposition and Retention in Rats. Ann. N.Y. Acad. Sci., 132, 77. Wright, G. W. (1989) Asbestos and Health in 1969. Am. Rev. resp. Die., 100, 467. HMEWIWJllltlllilll.MII.MH