Health Hazard Assessment of Nonasbestos Fibers
             Vanessa T. Vu, Ph.D.

   Health and Environmental Review Division
          Office of Toxic Substances
     U.S. Environmental Protection Agency
               December  30,  1988
                  Final  Draft

-------
                        TABLE OF  CONTENTS
                                                              PAGE
      ACKNOWLEDGMENTS	vii
      REVIEWERS	viii
      EXECUTIVE SUMMARY	1

I.     Introduction	26

II.   Man-Made Mineral Fibers	27

      1.0.   Fibrous Glass	28
            1.1.  Fiber Deposition, Clearance and Retention.... 29
            1.2.  Effects on Experimental Animals.....	34
                  1.2.1.  Oncogenicity	35
                  1.2.2.  Fibrogenicity	51
            1,3.  In Vitro Studies.	57
                  1.3.1.  Genotoxicity	57
                  1.3.2.  Cytotoxicity	63
            1.4.  Assessment of Health Effects	68
            1.5.  Recommendations.	 .78

      2.0.   Mineral Wool	79
            2.1.  Fiber Deposition, Clearance and Retention.... 79
            2.2.  Effects on Experimental Animals	82
                  2.2.1.  Oncogenicity	82
                  2.2.2.  Fibrogenicity.	86
            2.3.  In Vitro Studies	87
                  2.3.1.  Genotoxicity	87
                  2.3.2.  Cytotoxicity	87
            2.4.  Assessment of Health Effects	89
            2.5.  Recommendations	94

-------
                               iii
                   TABLE OF CONTENTS (continued)
                                                               PAGE
      3.0.   Ceramic Fibers	94
            3.1.   Fiber Deposition,  Clearance and Retention.... 95
            3.2.   Effects on Experimental Animals	97
                  3.2.1.   Oncogen icity	98
                  3.2.2.   Fibrogen icity	104
            3.3.   In Vitro Studies	107
                  3.3.1.   Genot ox icity	107
                  3.3.2.   Cytotoxicity	107
            3.4.   Assessment of Health Effects	109
            3.5.   Recommendations	Ill

III.   Naturally Occurring Fibers	Ill

      1.0.   Erionite	Ill
            1.1.   Fiber Deposition,  Clearance and Retention...112
            1.2.   Effects on Experimental Animals	113
                  1.2.1.   Oncogen icity	113
                  1.2.2.   Fibrogen icity	118
            1.3.   In Vitro Studies	119
                  1.3.1.   Genotox icity	119
                  1.3.2.   Cytotoxicity	122
            1.4.   Assessment of Health Effects	123
            1.5.   Recommendations	126

      2.0.   Wollastonite..	126
            2.1.   Fiber Deposition,  Clearance and Retention...127
            2.2.   Effects on Experimental Animals	127
                  2.2.1.   Oncogenicity	127
                  2.2.2.   Fibrogenicity	129

-------
                                IV
                   TABLE OF CONTENTS (continued)
                                                              PAGE
            2.3.   In Vitro Studies	129
                  2.3.1.   Genotoxicity	129
                  2.3.2.   Cytotoxicity	129
            2.4.   Assessment of Health Effects	133
            2.5.   Recommendations	134

      3.0.   Attapulgite	135
            3.1.   Fiber Deposition,  Translocation
                    and Clearance	135
            3.2.   Effects on Experimental Animals	136
                  3.2.1.   Oncogenicity	137
                  3.2.2.   Fibrogenicity	141
            3.3.   In Vitro Studies	143
                  3.3.1.   Genotoxicity	143
                  3.3.2.   Cytotoxicity	144
            3.4.   Assessment of Health Effects	147
            3.5.   Recommendations	152

IV.   Synthetic Fibers	152

      1.0.   Aramid Fibers	152
            1.1.   Fiber Deposition and Clearance 	153
            1.2.   Effects on Experimental Animals	154
                  1.2.1.   Oncogenicity	154
                  1.2.2.   Fibrogenicity	156
            1.3.   In Vitro Studies	160
                  3.3.1.   Genotoxicity	160
                  3.3.2.   Cytotoxicity	160
            1.4.   Assessment of Health Effects	161
            1.5.   Recommendations	164

-------
                                V
                   TABLE OF CONTENTS  (continued)
                                                              PAGE
      2.0.   Carbon Fibers	164
            2.1.   Fiber Deposition, Clearance and Retention...165
            2.2.   Effects on Experimental Animals	166
                  2.2.1.  Oncogenicity	166
                  2.2.2.  Fibrogenicity	170
            2.3.   In Vitro Studies	174
                  2.3.1.  Genotoxicity	174
                  2.3.2.  Cytotoxicity	175
            2.4.   Assessment of Health Effects	177
            2.5.   Recommendations	180

      3.0.   Polyolefin. Fibers	180
            3.1.   Fiber Deposition, Clearance and Retention...181
            3.2.   Effects in Experimental Animals	181
                  3.2.1.  Oncogenicity	181
                  3.2.2.  Fibrogenicity	183
            3.3.   In Vitro Studies	185
                  3.3.1.  Genotoxicity	185
                  3.3.2.  Cytotoxicity	185
            3.4.   Assessment of Health Effects	186
            3.5.   Recommendations	188

V.    Mechanisms of Fiber-Induced Diseases:  Relationship
      Between Fiber Properties and Pathogenicity	188

VI.   References	197

-------
                                VI
                    TABLE  OF  CONTENTS  (continued)
                                                              PAGE
VII.  Appendix	214
Table
Table
Table
Table
Table
Table
Table
Table
Table

1.
2.
3.
4.
5.
6.
7.
8.
9.

Summary
Summary
Summary
Summary
Summary
Summary
Summary
Summary
Summary

of
of
of
of
of
of
of
of
of

animal
animal
animal
animal
animal
animal
animal
animal
animal

studies
stud ies
studies
stud ies
stud ies
stud ies
studies
studies
studies

on
on
on
on
on
on
on
on
on

Fibrous Glass. .

Ceramic Fibers.



Aramid Fibers..
Carbon Fibers..
Polyolef in

..215
230
..234
240
243
244
..246
..248

. ,251

-------
                               Vll
                          ACKNOWLEDGMENTS









     Dr. Kerry Dearfield, (U.S. Environmental Protection Agency)  is



gratefully acknowledged for the review and evaluation of



genotoxicity data on fibers.



     The author wishes to thank Mary Argus, Ph.D./ Charles Auer,



Diane Beal/ Ph.D., Karl Baetcke, Ph.D., Kerry Dearfield, Ph.D.,



Ernest Falke, Ph.D., Penelope Fenner-Crisp, Ph.D., Stephanie R.



Irene, Ph.D., Elizabeth Margoshes, Bruce Means, Karen Milne, Steven



Shapiro and Bruce Sidwell (U.S. Environmental Protection Agency)  for



their critical review of the document.

-------
                               Vlll
                             REVIEWERS
     The early draft of the document has been peer reviewed for
scientific and technical merit by the following scientists and their
valuable comments are greatly appreciated.

        Dr. David L. Bayliss, Office of Health and
        Environmental Assessment, U.S. Environmental
        Protection Agency/ Washington, DC,  U.S.A.

        Dr. Steven Bayard, Office of Health and
        Environmental Assessment, U.S. Environmental
        Protection Agency, Washington, DC,  U.S.A.

        Dr. David L. Coffin, Health Effects Research
        Laboratory, U.S. Environmental Protection Agency,
        Research Triangle Park, NC, U.S.A.

        Dr. J.M.G. Davis, Pathology Branch, Institute of
        Occupational Medicine, Edingburg, Scotland, United
        Kingdom

        Dr. John M. Dement, National Institute of
        Environmental Health Sciences, Research Triangle
        Park, NC, U.S.A.

        Ms. Caroline S. Freeman, Office of Standards Review,
        Occupational Safety and Health Administration,
        Washington, DC, U.S.A.

        Dr. David Groth, National Institute for Occupational
        Safety and Health, Cincinnati, OH, U.S.A.

        Dr. Ulrich F. Gruber, University of Basle, Basle,
        Switzerland

        Dr. Peter F. Infante, Office of Standards Review,
        Occupational Safety and Health Administration,
        Washington, DC, U.S.A.

        Dr. Jon L. Konzen, Owens Corning Fiberglass
        Corporation, Toledo, OH, U.S.A. v

        Dr. Aparna Koppikar, Office of Health and
        Environmental Assessment, U.S. Environmental
        Protection Agency, Washington, DC, U.S.A.

        Dr. Dennis J. Kotchmar, Environmental Criteria and
        Assessment Office, U.S. Environmental Protection
        Agency, Research Triangle Park, NC, U.S.A.

-------
                        IX
            REVIEWERS  (continued)

Dr. Arthur Langer, Mount Sinai School of Medicine,
New York, NY, U.S.A.

Dr. William Pepelko, Office of Health and
Environmental Assessment, U.S. Environmental
Protection Agency, Washington, DC, U.S.A.

Dr. J.C. Wagner, Llandough Hospital, Pernath,
Glamorgan, United Kingdom

-------
EXECUTIVE SUMMARY



     The inhalation of asbestos fibers including amosite,



chrysotile, and crocidolite has been associated with the



development of lung cancer, mesothelioma, pulmonary fibrosis and



other nonmalignant pleural diseases in humans.  Because the



pathogenic effects of asbestos are attributed/ in general, to its



fibrous nature, human health concern extends to the use of other



fibrous substances.  This document assesses the health effects of



nine non-asbestos fibers and attempts to determine the nature and



magnitude of the health hazard as compared to asbestos.  The



fibers chosen for investigation were selected for one or more of



the following reasons:  a) they are commercially important; b)



they are considered potential asbestos substitutes; c) they



represent fiber types with broadly different physical and



chemical characteristics; and, d) health data are available on



them.  The fibers evaluated in this report include fibrous glass,



mineral wool, ceramic fibers, erionite, wollastonite,



attapulgite, aramid fibers, carbon fibers, and polyolefin fibers.



     Available data suggest some similarities in the health



effects of asbestos and some nonasbestos fibers but the degree of



the health effects may differ substantially among fiber types.



The differences in the biological activity may be associated with



the specific characteristics of each fiber type including fiber



morphology, size distribution, chemical constitution, surface



properties and durability.



     A basic property which allows a fiber's potential toxicity to



be expressed is its respirability, i.e., its ability to penetrate



into the smaller conducting airways of the tracheobronchilar tree

-------
and the alveolar region of the lung.  It is clear that fiber



diameter is the most important factor in determinig the



respirability of the fiber.  Fiber length and morphology also



affect the respirability of the fiber but to a lesser extent.



Also, it would appear that the fiber needs to be retained and



persist in the tissue in order to cause toxicity.  Fiber length



is an important determinant of fiber retention, with shorter



fibers being cleared more readily.  Fiber retention is also



determined by the biological solubility of fibers which is



directly related to their chemical composition and physical



characteristics.



     To date, the exact role of various fiber properties in



relation to biological activity and pathogenicity is not clearly



understood.  It is clear, however, that different types of fibers



with similar size properties (e.g. erionite and asbestos) could



have very different biological activity, although there is



increasing evidence suggesting that for a given fiber type,  fiber



size is an important factor, i.e., the thinner and longer the



fiber, the more hazardous it is.  Additional research is



necessary to examine further the importance of fiber properties



in mediating the induction of disease and investigate the



mechanisms by which fibrous materials cause disease.



     It is difficult to definitively assess the relative



biological activity and pathogenicity of nonasbestos fibers  in



comparison to asbestos because of limited data bases.  Major



limitations include a lack of comparable dose-response data  as



well as information available on the characteristics of the



tested fibers, particularly, fiber morphology and size

-------
distribution and the number of fibers in each size category.



However, on the basis of available information, it may be



concluded at this time that with the possible exception of



erionite, the other eight fibers reviewed in this report do not



pose a health hazard of similar magnitude as asbestos.



Additional studies are needed to conclusively determine the



health effects of each fiber type.  Erionite, which may be more



hazardous than asbestos, is not a major concern because of its



limited production and use.



     A summary of the hazard assessment of the oncogenic and



fibrogenic effects of these fibers, and the testing



recommendation(s) to fill data gaps for each fiber is presented



in the following sections.  The assessment of the potential for



carcinogenicity of fibers in humans is based on the current U.S.



EPA classification system for categorizing the overall weight-of-



evidence as determined from human, animal and other supporTing



data (USEPA, 1986).



     Inhalation is the major route of exposure to fibers and



exposure via this route of administration has been shown to cause



cancer in humans in the case with asbestos.  Hence, it would seem



most relevant to use the inhalation route for the animal testing



of fiber carcinogenicity.  Positive results from inhalation



studies in animals would be interpreted to have significant



implications for potential hazard in the human since asbestos has



also been found to induce tumors in animals following inhalation



exposure.  The major pathogenic effects associated with the



inhalation of asbestos in humans including lung fibrosis, lung



cancer and mesothelioma have been replicated in rodents exposed

-------
to asbestos by inhalation.  There are, however, shortcomings of



inhalation studies.  One reason is that fiber deposition and



retention in rodents are considerably different from those in



humans.  Rodents, being obligatory nose breathers, have a greater



filtering capacity than humans resulting in a lower alveolar



deposition of fibers in rodents.  As a result, inhalation tests



in rodents may underestimate the hazard potential of fibers to



humans unless it is clear that the number of fibers reaching the



target tissues are comparable to the positive control.



     Experimental procedures other than inhalation exposure



testing have been developed which attempt to accommodate for



these species differences and to achieve comparable target organ



doses.  However, they do have their disadvantages.  In these



studies, the test fiber is artificially introduced in large



"bolus" dose(s) directly into the target tissue such as the



mesothelium as in the cases with intraperitoneal and intrapleural



administrations, or near major targets including the lung and



pleural mesothelium in the case of intratracheal instillation.



Caution must be excercised in extrapolating the findings from



parental administration studies in animals to humans, since the



results from such studies may not be predictive of inhalational



hazard.  Injection studies bypass the normal physiological



deposition and clearance mechanisms and lead to non-random



accumulations of test substances at the site of deposition.



Thus, respirability characteristics, which are routinely taken



into account in an inhalation study, are not operative following



injection.  Nevertheless, injection studies are of value by



providing useful information regarding the intrinsic biological

-------
activity of the test fiber under conditions where the material is



in direct contact with the cells at risk.








1.   Fibrous Glass
     There is no evidence in available epidemiologic studies that



peritoneal or pleural mesotheliomas are associated with



occupational exposure to man-made mineral fibers (fibrous glass



and mineral wool).  With regard to the respiratory cancer risk,



there was no excess of such cancers among continuous glass



filament workers in either the U.S. or Europe.  For glass wool



production workers, there was no significant increase in



mortality from respiratory cancer (or lung cancer) compared with



regional rates in either the U.S. or European cohorts/ though



there were statistically significant small increases compared



with national rates in the U.S. study.  In both investigations,



mortality from respiratory cancer increased nonsignificantly with



time from first exposure.  However, it was not related to the



duration of employment or cumulative fiber exposure in the U.S.



study.  Also, in the European study, it was not related to the



duration of exposure or to different technological phases



reflecting differences in the intensity and quality of



exposure.  A lack of dose-related trends might be due in part to



the very low exposure experienced by the cohorts.



     Among glass wool workers in the U.S. cohort who were ever



exposed to small diameter fibers (<3.0 urn)/ there was a



nonsignificant excess of respiratory cancer mortality which



increased nonsignificantly with time since first exposure,



compared to those who had never been exposed.  A third study

-------
reported a statistically significant increase of lung cancer



mortality among Canadian glass wool workers, but this was not



related to the time since first exposure nor to the duration of



exposure.



     On the basis of available information, the evidence for



carcinogenicity of small diameter glass fibers, glass wool and



glass filament from studies in humans is considered inadequate.



Still, the epidemiologic findings seem to suggest that workers



engaged in the manufacture of glass wool and small diameter



fibers might be at increased risk of developing respiratory



cancer; additional studies are necessary to clarify the health



effects of fibrous glass in humans.



     A number of long-term inhalation studies have not provided



evidence of lung tumor or mesothelioma in several animal species



exposed to glasswool (typically 3-10 urn in diameter), fine fiber-



glass (1-3 urn in diameter) or to very fine fibrous glass (also



known as glass microfibers; <1 urn in diameter).  Shortcomings of



these investigations include the use of small numbers of animals,



relatively short fibers, low numerical concentrations of fibers,



limited study duration, and/or inadequate positive control.  In



contrast to the inhalation studies, many animal studies involving



the intrapleural injection/implantation, or intraperitoneal



injection of fine glass fibers or glass microfibers consistently



demonstrate that these fibers are capable of producing



mesothelioma in rats, hamsters, and mice when they are introduced



directly into the body cavity.  Glass wool has also been shown to



produce low incidences of pleural tumors in a few intrapleural



implantation/injection studies in rats.  In addition, an

-------
increased incidence of both lung tumors and pleural mesothelioma



has been reported in one study following intratracheal



administration of glass microfibers to hamsters; this indicates



that under certain conditions, glass fibers can pass through the



lung and incite reactions in the pleura.  In another intratra-



cheal instillation study, glass microfibers also caused lung



tumors in the rat.  However, several other intratracheal



instillation studies in hamsters and rats have not reported tumor



formation with glass wool, fine glass fibers or glass



microfibers.



     In the absence of positive findings from available



inhalation studies, the evidence for human carcinogenicity of



very fine and fine fibrous glass and glass wool from animal



studies is considered limited because only non-physiological



routes of administration are associated with carcinogenic



findings.  However, the repeated observation of tumors following



these administrations do indicate the biological activity of the



test fibers when deposited in high enough quantity at or near the



target tissue.  The animal data are supported by positive



findings from a few genotoxicity studies which  indicate that fine



fiberglass and glass microfibers cause similar weak genotoxic



effects (clastogenicity and cell transformation) generally seen



with asbestos.  Thus, considering all available data (human,



animal and supporting evidence), the Office of Toxic Substances



(OTS) of the U.S. Environmental Protection Agency  (USEPA)



proposes to classify fine and very fine fibrous glass and glass



wool in-Category C, i.e., possible human carcinogen, mainly based



on inadequate evidence of carcinogenicity in humans and

-------
                                 8






limited evidence in animals.  Others might interpret the same



information as supporting a B2 (probable human carcinogen)



designation/ mainly because they think the animal injection



studies should be afforded more weight.  Irrespective of these



differences in classification, all would agree that the existing



evidence supporting a human carcinogen hazard for fibrous glass



is much less convincing than for asbestos.



     As for the continuous glass filament (nominal diameters of



6-15 urn), there is inadequate evidence of carcinogenicity in lab-



oratory animals.  The results of a few available intrapleural



implantation studies showed that large diameter glass fibers did



not induce mesothelioma in rats.  Glass filament is therefore not



classifiable as to human carcinogenicity on the basis of



inadequate evidence of carcinogenicity in humans and animals



(Category D).



     There does not appear to be any convincing evidence for



increased risks of non-malignant respiratory disease (NMRD) among



fibrous glass workers.  In the European study/ there was no



excess mortality from NMRD in the continuous glass filament or



glass wool cohort, nor was there a trend with time since first



exposure or duration of employment.  Similarly/ in the U.S.



study/ no significant excess of NMRD mortality was found among



glass filament workers compared with either local or national



rates.  For U.S. glass wool workers, there was no significant



increase in NMRD mortality based on local rates though there was



a statistically significant excess compared with national



rates.  Further analyses of NMRD mortality showed no apparent



dose-related trends.  Among glass wool workers ever exposed to

-------
small diameter fibers/ no excess of NMRD mortality was observed



but there was a nonsignificant increase with time since first



exposure.  The results of a respiratory morbidity study showed



some evidence of radiographic opacities in the lung of a limited



number of fibrous glass workers but there was no evidence of



pulmonary fibrosis, no increase in respiratory symptoms and no



impaired lung function.



     Long-term inhalation studies have not provided definitive



evidence for the development of lung fibrosis in laboratory



animals exposed to fine glass fibers or glass wool.  However, the



positive findings from several injection studies in animals and



in vitro cytotoxicity studies indicate that fine and very fine



fiberglass may be fibrogenic.



     Overall, it may be concluded that a possible health hazard



exists from inhalation exposure to fine and very fine fibrous



glass, i.e., fibers with diameters consistently below 3



microns.  A low health concern is also raised for exposure to



glass wool which does contain some respirable fine fibers.  As



for continous glass filaments which are generally nonrespirable,



they would appear to pose little or no hazard to exposed



humans.  On the basis of available experimental data, it is



concluded that fibrous glass appears to be less pathogenic than



asbestos.  Although the fibrogenicity and oncogenicity of fine



fibrous  glass and glass wool have been extensively investigated,



none of  the available inhalation studies are considered



adequate.  Furthermore, since considerable data gaps still exist,



particularly a lack of comparative dose-response effects with



 pbestos, additional  inhalation/injection studies would be

-------
                                 10






useful.   It  also  appears  necessary  to  conduct additional



epidemiological studies to  conclusively determine the health



hazard potential  of  fibrous glass in humans.








2.   Mineral Wool
      Small  excesses of mortality due  to  respiratory cancer have



been  observed  among rock wool/slag wool  workers  in the U.S. and



in  Europe.   In the U.S. study/  the excess of  respiratory cancer



mortality was  statistically  significant  when  compared to both



local and national rates.  There was  no  clear trend with time



since first exposure  and there  was no relationship with duration



of  exposure, cumulative fiber exposure or average intensity of



exposure.   The results of  a  nested case-control  study using cases



from  the U.S.  cohort  showed  a weak but positive  trend between



mineral wool exposure and  respiratory cancer  when confounding by



cigarette smoking was considered.



      In the European  study,  the lung  cancer excess found among



rock  wool/slag wool workers  was not statistically significant



compared with  either  local or national mortality rates.  There



was also a  statistically nonsignificant  increased mortality with



time  since  first exposure  but there was  no relationship between



lung  cancer mortality and  duration of exposure.  The highest and



statistically  significant  lung  cancer rates were found among



workers after  more than  20 years  first exposed in the early



technological  phases, during which fiber airborne levels were



presumably  higher than  in  later production phases.  The presence



of  workplace contaminants  such  as bitumen, pitch or asbestos



•could not explain the observed  lung cancer excess.

-------
                                11

     Overall/ the available epidemiological findings suggest that
mineral wool workers are at increased risk of respiratory
cancer.  The evidence for mineral wool as an etiological agent
includes the consistent elevated risk observed in several rock
wool/slag wool facilities, and the higher cancer risks found.
among workers who had twenty or more years elapse since first
exposure.  The evidence not supporting an etiological
relationship is the lack of a consistent dose-response trend.
This might be due in part to the low levels of fiber exposure and
the potential exposure misclassification.  On the basis of
available information/ the evidence for carcinogenicity of
mineral wool from epidemiological studies is considered limited.
     The results of three limited long-term studies showed that
mineral wool did not produce tumors in rats or hamsters when
administered by inhalation.  However/ mineral wool has been shown
in a few studies to induce varying tumor yields in rats via
either the intrapleural (pleural mesothelioma) or intraperitoneal
route  (peritoneal mesothelioma) of exposure.  Overall, the
experimental evidence for the carcinogenic potential of mineral
wool is considered to be limited.  Thus/ OTS is proposing to
classify mineral wool as a probable human carcinogen  (Category
Bl) on the basis of limited evidence of carcinogenicity from
epidemiological studies and limited evidence from animal
studies.  There is no genotoxicity information available on
mineral wool.
     There is inadequate epidemiological evidence for an
association  between the development of non-malignant  respiratory
diseases  (NMRD) and exposure to mineral wool.  No increased

-------
                                12






mortality from NMRD was found for the European rock wool/slag



wool workers.  In the U.S. study, a statistically nonsignificant



excess of NMRD mortality was observed among mineral wool workers



based on local or national rates.  However/ there was no



relationship with time since first exposure, duration of



exposure, average intensity of exposure, or estimated cumulative



level of exposure.  Furthermore, the results of a respiratory



morbidity study in the U.S. showed no evidence for impaired lung



functions or radiographic lung abnormalities associated with



mineral wool exposure.



     There is little experimental evidence for the fibrogenicity



of mineral wool.  Mineral wool was not found to cause lung



fibrosis in three long-term inhalation studies but focal fibrosis



was reported in a very limited inhalation study involving only



two rats.  The results of two in vitro studies showed that



mineral wool was cytotoxic in cells in culture.  In view of these



findings, concerns for possible development of pulmonary fibrosis



associated with mineral wool exposure cannot be entirely ruled



out at this time.



     Based on the limited data base in animals, mineral wool



appears to be less biologically active and less pathogenic than



asbestos fibers.  It is concluded at this time that mineral wool



fibers may present a health hazard to exposed humans but not to



the same magnitude as asbestos.  Since the pathogenic effects of



mineral wool have not been adequately characterized, additional



epidemiological studies and animal testing are needed.

-------
                                13






3.   Ceramic Fibers



     There are no studies available on the potential health



effects from exposure to ceramic fibers in humans.  The



pathogenicity of ceramic fibers in laboratory animals appears to



vary considerably for different fiber types which may be a



function of variation in fiber size distribution.



     An increased incidence of lung tumors have been observed



after chronic inhalation exposure to ceramic aluminum silicate



glass in one study using rats.  Another inhalation study produced



no tumors in rats, but one mesothelioma in a hamster.  An



intratracheal instillation study conducted by the same laboratory



showed no tumor induction with refractory aluminum silicate



fibers.  However, these fibers have been shown in several long-



term studies to cause mesothelioma in rats and hamsters by



intrapleural or intraperitoneal injection.  Based on the



sufficient evidence of carcinogenicity in animals in multiple



experiments with different routes of administration, but in the



absence of human data, OTS proposes to classify ceramic aluminum



silicate fiber as a probable human carcinogen (Category B2).



     The experimental evidence of fibrogenicity of ceramic



aluminum silicate fibers is limited.  The positive results of a



chronic inhalation study suggest that long-term inhalation of



ceramic aluminum silicate glass may produce mild  interstitial



lung fibrosis in humans.  This finding is further supported by



positive findings from an  in vitro cytotoxicity  study of ceramic



aluminum silicate glass.

-------
                                14






     In view of available findings and since ceramic aluminum



silicate fibers are respirable and durable, it may be concluded



that this ceramic fiber type may present a health hazard to the



exposed humans.  Because of the variable results from available



in vivo and in vitro studies on ceramic aluminum silicate, its



relative pathogenicity in comparison to asbestos cannot yet be



made at this time.  In order to further evaluate the health



effects of ceramic aluminum silicate fibers, it is recommended



that epidemiological studies of exposed workers be initiated.  No



additional animal tests are recommended at this time since a



large-scale animal study by various routes of exposure is



currently being conducted at a private laboratory.



     Available animal studies have not provided evidence of the



carcinogenicity and fibrogenicity for refractory alumina oxide



and zirconia oxide fibers.  It has been shown in several studies



that these fibers did not produce tumors nor fibrosis in rats via



chronic inhalation exposure or by intracavitary injection.  The



lack of experimental pathogenic effects of these fibers may be



attributable to the test fibers being largely nonrespirable.



Similarly, the cytotoxicity of these fibers in rat peritoneal



macrophages is low.  These refractory fibers are therefore not



classifiable as to human carcinogenicity (Category D) on the



basis of inadequate evidence of carcinogenicity in animal studies



and in the absence of human data.  Based on available findings,



it would appear that refractory alumina and zirconia fibrous



products containing mostly nonrespirable fibers would not pose



significant health hazard in exposed humans.

-------
                                15






4.   Erionite



     Available epidemiological data show that populations from



South Central Turkey have an excessive incidence of malignant



pleural mesothelioma and nonmalignant pleural diseases (chronic



pleurisy fibrosis, pleural thickening and pleural plaques).  The



etiology of these diseases is uncertain but there is limited



evidence to indicate that erionite fibers may be the major



etiological factor.  All of the experimental studies conducted to



date have confirmed that erionite from Turkey and deposits in the



U.S. causes a significant increase in malignant meseothelioma in



animals by several routes of exposure including inhalation.



Animal data are also supported by findings that erionite is



genotoxic, in causing DNA damage and repair and inducing cell



transformation in culture.  Thus, OTS is proposing to classify



erionite as a probable human carcinogen (Category Bl) on the



basis of limited evidence of carcinogenicity from studies in



humans and sufficient evidence of carcinogenicity from animal



studies.



     There is no information available on the ability of erionite



to  induce fibrotic diseases in animals by inhalation.  However,



erionite has been shown to cause fibrogenic effects  in animals by



the injection method.  Furthermore, available in vitro studies



demonstrate that erionite is hemolytic and highly cytotoxic.



Thus, it is concluded that erionite is potentially fibrogenic in



view of the limited evidence from epidemiological studies and



limited evidence from experimental studies.



     Overall, there is sufficient evidence to conclude that



erionite potentially poses a significant health hazard to exposed

-------
                                16






humans.  However, if practical, additional epidemiological



studies should be conducted to further evaluate the association



between erionite environmental exposure and development of



malignant and nonmalignant respiratory diseases.  Since the



toxicological profile of erionite has been adequately



characterized in animals, no further testing is recommended.



Based on the available experimental data, erionite appears to be



at least as hazardous as asbestos.








5.   Wollastonite



     None of the available epidemiological studies were designed



to assess the risk of lung cancer or mesothelioma associated with



wollastonite exposure.  One case of mesothelioma has been



reported in a worker who had been exposed to wollastonite, but no



cause and effect relationship can be drawn based on a single case



report.  Preliminary information on an inhalation oncogenicity



study of wollastonite in rats indicates the lack of a tumorigenic



response.  The results of an intrapleural implantation study



showed that wollastonite was weakly tumorigenic in rats; whereas,



in another long-term study in rats, wollastonite caused no tumors



when injected into the peritoneal cavity.  Thus, based on limited



evidence of carcinogenicity in animals and inadequate human data,



OTS is proposing to classify wollastonite as a possible human



carcinogen (Category C).  No other supporting evidence (e.g.,



genotoxicity data) of oncogenicity is available.



     Available data are  inadequate to evaluate the fibrogenic



potential of wollastonite.  A preliminary report of an NTP



bioassay indicates no evidence of pulmonary fibrosis in rats

-------
                                17






following chronic inhalation of wollastonite but data are not yet



available for a full evaluation.  Available epidemiological



studies indicate a possible association between wollastonite



exposure and some nonmalignant diseases such as impaired



ventilatory capacity, mild fibrosis of the lung, pleural



thickening and chronic bronchitis.  However, because of a number



of limitations, they do not provide convincing evidence of a



causal relationship of nonmalignant respiratory diseases and



wollastonite exposure.  Nevertheless, these epidemiological



findings do raise a health concern, particularly in view of



positive results from in vitro cytotoxicity assays which are



thought to be indicative of fibrogenic activity.



     Overall, there is some evidence supporting a concern for a



possible health hazard from exposure to wollastonite.  However,



it would appear that wollastonite is probably less hazardous than



asbestos since available experimental data indicate that



wollastonite is much less biologically active than asbestos.  In



order to fully assess the health effects of wollastonite, it is



necessary to seek additional epidemiological studies and to fully



evaluate the results of an inhalation bioassay recently completed



by the National Toxicology Program.








6.   Attapulgite



     There is inadequate evidence of carcinogenicity of short-



fibered attapulgite from available studies in humans.  The



results of a single small cohort study in the U.S. showed an



excess of lung cancer among some groups of attapulgite workers.



However, due to several limitations, this study did not provide

-------
                                18






convincing evidence of a fiber etiology.  Several experimental



studies showed that short attapulgite fibers (<2 yum) in



commercial use from the U.S., France, and Spain did not produce



mesothelioma in rats by the intrapleural or intraperitoneal



route.  In addition, short attapulgite fibers from Spain did not



induce tumors in rats following prolonged inhalation exposure.



There is also no evidence of carcinogenicity in mice following



life-time feeding with short-fibered attapulgite.  These data are



supported by negative findings from a single genotoxicity study



on short attapulgite fibers.  Short-fibered attapulgite is,



therefore, not classifiable as to human carcinogenicity (Category



D) on the basis of inadequate evidence of carcinogenicity from



epidemiological and animal data.



     In contrast, attapulgite samples from other geographical



locations (e.g. Spain, U.K.) which contain considerable numbers



of long fibers (>5 urn) have been shown to be tumorigenic in rats,



causing the induction of lung tumors and mesotheliomas by inhala-



tion, as well as pleural mesothelioma following  intrapleural



injection and abdominal tumors via the intraperitoneal route.



Therefore, based on sufficient evidence of carcinogenicity in



animals but in the absence of human data, OTS is proposing to



classify long-fibered attapulgite as a probable human carcinogen



(Category B2).



     Available data have not provided evidence of fibrogenic



effects for short-fibered attapulgite.  The results of three



studies in humans provide inadequate evidence of the development



of nonmalignant respiratory diseases associated  with exposure to



short-fibered American attapulgite.  The results of a long-term

-------
                                19






animal study showed that short attapulgite fibers from Spain did



not induce lung fibrosis in rats via inhalation.  Morever, none



of the available injection studies with short-fibered attapulgite



from various geographical locations have reported any fibrotic



lesions in treated rats.  However/ positive findings of several



in vitro cytotoxicity studies suggest a possible fibrogenic



concern for short-fibered attapulgite.  In contrast, based on the



positive results of a chronic inhalation study in rats with long-



fibered attapulgite, it is concluded that attapulgite samples



containing long fibers (>5 urn long) may induce lung fibrosis in



humans.



     In view of available findings, it would appear that the



toxicological properties of attapulgite may depend on fiber



length.  Overall, there is insufficient evidence to support a



health concern for short-fibered attapulgite in commercial use in



the U.S.  However, because these fibers are highly respirable,



and appear to be biologically active in in vitro, adverse health



effects remain a possibility.  On the other hand, there is a



reasonable basis to support a health concern for long-fibered



attapulgite.  Available animal data are not sufficient to allow a



definitive assessment on the relative pathogenicity of long-



fibered attapulgite compared to asbestos.  However, since these



fibers are not widely available for commercial use, they are not '



expected to pose significant health risks to humans.  In order to



fully assess the health effects of short-fibered American



attapulgite, it is necessary to obtain additional epidemiological



data and to conduct long-term inhalation studies in animals.

-------
                                20






7.   Aramid Fibers



     There is no information available on the health effects of



para-aramid fibers in humans.  In the female rat, long-term



inhalation of ultrafine para-aramid (Kevlar®) fibrils caused a



dose-related production of lung tumors.  Although there are no



oncogenicity data on ultrafine para-aramid in animals via the



intracavitary route, weak tumorigenic responses were observed in



rats in two intraperitoneal injection studies with Kevlar® fiber



and pulp containing a considerable number of fine fibrils.  Thus,



based on the sufficient evidence of the carcinogenicity in



animals but in the absence of human data, OTS is proposing to



classify ultrafine para-aramid as a probable human carcinogen



(Category B2).  There is no genotoxicity information available on



ultrafine para-aramid.



     Data from the same chronic inhalation study also indicate



that ultrafine para-aramid (Kevlar®) is weakly fibrogenic in



rats.  The positive findings from an in vitro cytotoxicity study



on short, thin Kevlar® fibers further support the concern for the



fibrogenic potential of ultrafine para-aramid.



     In view of these findings, it may be concluded that



ultrafine para-aramid is potentially pathogenic.  This fibrous



material, however, does not pose a health risk to humans because



it is not available in commerce.  Available  data, however, are



not sufficient to provide definitive assessment  on the



comparative pathogenicity of ultrafine para-aramid to asbestos.



     The positive results of two intraperitoneal injection



studies in rats indicate that para-aramid pulps  or fibers may



have a low carcinogenic and  fibrogenic potential.  Thus, based on

-------
                                21






the limited evidence of carcinogenicity in animals and in the



absence of human data, OTS proposes to classify commercial grade



para-aramid as a possible human carcinogen (Category C).  Because



of the generally nonrespirable characteristic of commercial grade



para-aramid fiber and pulp, it would appear that the hazard



potential of para-aramid is probably much lower than that of



asbestos.  However, it should be pointed out that since small



numbers of para-aramid fibrils can result from peeling off the



para-aramid fiber matrix and may become airborne, a possible



health hazard may exist for exposure to para-aramid, particularly



to the pulp form.  In order to further assess the potential



health effects of para-aramid, additional animal testing is



recommended.



     There are insufficient data to assess the health effects of



Nomex® aramid fibers.  Nomex® is not classifiable as to human



carcinogenicity because of lack of data in humans and animals



(Category D).  Based on the fact that no effects were observed in



a single long-term intratracheal instillation study in the rat,



and that Nomex® is nonrespirable, it would appear that Nomex®



poses no significant health hazard to humans.  Because of a low



health concern, no additional animal testing is recommended for



Nomex®.








8.   Carbon Fibers
     There is no information available on the potential



development of respiratory neoplasms in humans from exposure to



carbon fibers.  Furthermore, no data are available on the



oncogenicity of carbon fibers in animals by inhalation.  However,

-------
                                22






carbon fibers were not found to induce tumors in rats following



intratracheal instillation, intraperitoneal injection, or



intramuscular implantation.  The only studies that reported



positive results were those from a subcutaneous study in which an



increased production of local sarcomas was found in rats, and



from a dermal bioassay demonstrating that benzene extracts of



pitch-based carbon fibers were weakly oncogenic in mice.



However, because there was no information available on the



characteristics of the test materials, particularly particle size



and morphology, the significance of these findings is question-



able and the overall experimental evidence of carcinogenicity is



considered to be inadequate.  Carbon fibers are, therefore, not



classifiable as to human carcinogenicity (Category D) on the



basis of inadequate evidence from animal studies and in the



absence of human data.  The oncogenic potential of carbon fibers,



however, is supported by available genotoxicity data which



indicate that benzene extracts of pitch-based carbon fibers are



clastogenic and induce DNA damage and repair.  On the other hand,



the evidence of clastogenicity of benzene extracts of



polyacrylonitrile  (PAN)-based carbon fibers is only suggestive.



     There is inadequate evidence of fibrogenicity for carbon



fibers.  A small cross-sectional study conducted to date showed



no evidence of pathological effects in the lungs of workers in a



PAN-based carbon fiber production plant.  With regard to



experimental studies, there is no information available on the



long-term inhalation toxicity of carbon fibers in animals.  With



the exception of one study which reported in an abstract that



polyacrylonitrile  (PAN)-based carbon fibers induced lung fibrosis

-------
                                23






in rats via intratracheal instillation, several other animal



studies showed that carbon fibers did not induce fibrosis in



laboratory animals following subchronic inhalation exposure,



intratracheal instillation, or intraperitoneal injection.  Most



of these studies, however, are of little value for the evaluation



of the fibrogenic potential of carbon fibers because of limited



scope, lack of particle size and morphology data of the test



materials, and/or lack of details available on study design and



findings.  Furthermore, both negative and positive findings have



been reported regarding the in vitro cytotoxicity of carbon



fibers.



     Although currently available data are insufficient to



evaluate the potential health effects of carbon fibers, the data



taken together suggest that carbon fibers do not appear to



present a serious health hazard.  Nevertheless, the marginally



positive tumorigenic effects in a dermal study and the positive



clastogenic effects in genotoxicity tests induced by pitch-based



carbon fibers, suggest that a weak oncogenic potential for



certain types of carbon fibers may exist.  Because carbon fibers



are much less respirable and less biologically active than



asbestos, it would appear that they pose a lower degree of health



hazard compared to asbestos.  In order to further assess the



health hazard of carbon fibers, it is necessary to seek results



of an inhalation study now conducted at a private laboratory.



Since the endpoint of this study is fibrosis,  it is further



recommended that a chronic animal study capable of detecting



oncogenic effects be conducted if carbon fibers of respirable



size enter the marketplace.

-------
                                24






9.   Polyolefin Fibers



     There are no available epidemiological studies which examine



the potential oncogenic effect of polyolefin fibers.  Further-



more, there are no data available on the oncogenicity of poly-



olefin fibers in animals by inhalation.  The results of an intra-



tracheal insufflation study showed that both polyethylene and



polypropylene fibers did not induce tumors in rats.  However, the



lack of information on the characteristics of the fibers, the



dosages, and the specific methods of administration precludes any



definitive assessment of the oncogenicity of these fibers under



the conditions of the study.  In a long-term intraperitoneal



injection study in rats, polypropylene fibers were found to be



weakly oncogenic.  These results were only preliminary and a full



evaluation cannot be made at this time.  Therefore, polyolefin



fibers are not classifiable as to human carcinogenicity  (Category



D) on the basis of inadequate evidence of carcinogenicity in



animals and no human data.



     No epidemiological studies have been conducted to determine



the nonmalignant respiratory effects in humans from exposure to



polyolefin fibers.  There is no information available on the



long-term inhalation toxicity of polyolefin fibers  in animals.



Available animal injection studies have provided inconclusive



results.  Polyethylene and polypropylene fibers did not  induce



fibrosis in rats in a long-term intratracheal insufflation study



and in a short-term intraperitoneal injection study.  These



results are supported by the finding from a single  in vitro study



that polyethylene and polypropylene dusts exhibited very low



cytotoxicity.  However, the lack of information on  the

-------
                                25





characteristics of the test materials makes it difficult to draw



any definitive conclusion on the fibrogenic potential of this



fiber category.  On the other hand, preliminary results of a



long-term intraperitoneal injection study in rats with thin, long



polypropylene fibers showed a strong degree of adhesions of the



abdominal organs.  However, in the absence of histological data,



a full evaluation of this study cannot be made at this time.



Overall, available data are inadequate to determine conclusively



whether polyolefin fibers are fibrogenic.  However, they seem to



suggest a low fibrogenic potential for polyolefin microfibers.



     In summary, available studies do not provide adequate data



for a definitive assessment of potential health effects in humans



exposed to polyolefin fibers by inhalation.  However, the



inhalation of polyolefin fibers or pulp may pose little or no



health hazard because they are generally not respirable and would



not be expected to produce lung diseases even if the material has



some intrinsic activity.  On the other hand, a possible health



hazard potential may exist for polyolefin microfibers since they



may be respirable.  Additional animal testing is therefore



recommended for polyolefin microfibers.  Because of a low concern



for the potential health effects of polyolefin fibers and pulps,
                                               ••f


further animal testing is not recommended at the present time.

-------
                                26





I.   Introduction



     Human exposure to airborne asbestos fibers including



amosite, chrysotile/ and crocidolite has been associated with the



development of malignant (e.g. lung cancer/ mesothelioma) and



nonmalignant (e.g. interstitial pulmonary fibrosis, also known as



asbestosis) diseases.  These diseases have also been induced



experimentally in laboratory animals exposed to asbestos.  As a



result/ concern has risen with the increasing development and use



of other respirable fibrous substances.  Nonasbestos fibers have



come under considerable investigation primarily because they



possess some asbestos-like characteristics (e.g. fiberlike



morphology, dimensional range, durability) suspected to be



important factors in the initiating of diseases.  The objective



of this report is to assess the human health effects associated



with exposure to nonasbestos fibers and to evaluate the



hypothesis that nonasbestos fibers may induce asbestos-like



diseases.



     The fibers under review comprise three categories:  man-made



mineral fibers (fibrous glass, mineral wool, ceramic fibers),



naturally occurring fibers (erionite, attapulgite, wollastonite),



and synthetic fibers (aramid fibers, carbon fibers, polyolefin



fibers).  These fibers were selected because of one or more of



the following reasons:  1) they are commercially important;



2) they are considered potential asbestos substitutes; 3) they



represent fiber types with broadly different physical and



chemical characteristics; and 4) some health data are available



on them.

-------
                                27






     This document reviews available data on pulmonary



deposition/ clearance and retention, in vivo toxicity, and in



vitro biological activity of each of the nine fibrous materials.



It also assesses the human health effects, primarily the



potential development of malignant and nonmalignant respiratory



diseases associated with inhalation exposure to each fiber, based



on the combined available epidemiological and experimental



evidence.  Finally, it determines the adequacy of data for each



of these fibers and makes testing recommendations to fill data



gaps.  A detailed review of the key epidemiological studies on



the health effects posed by most of these fibers is presented in



a separate document by Battelle (1988).  Summaries and



conclusions regarding human data have been derived from this



report and are used in the overall hazard assessment of each



fiber.  The assessment of the carcinogenicity of fibers in humans



is based on the U.S. EPA classification system for categorizing



overall weight-of-evidence for carcinogenicity from human,



animal, and supporting data (USEPA, 1986).



     The last section of the document discusses overall findings



about the whole fiber category and briefly evaluates the role of



physicochemical properties of fibers in relation to biological



activity and pathogenicity.








II.  Man-Made Mineral Fibers (MMMF)
     MMMF comprise three groups:  fibrous glass, mineral wool,



and ceramic fibers.  MMMF have glassy structures rather than



crystalline.  Their length and diameter distribution differ

-------
                                28






considerably and are dependent on the method of production and



the chemical composition.  In general, commercially produced MMMF



are much coarser than asbestos fibers, although specialized



samples have been produced with dimensions similar to those of



asbestos.  MMMF are usually coated with binding materials to



produce fabricated shapes and forms.  MMMF are monofilamentous,



and thus do not split longitudinally into thinner fibrils, but



may break transversely into shorter segments (NRC, 1984).  Based



on available data, the health effects of MMMF appear to vary



substantially.








II.1.  Fibrous Glass
     Fibrous glass is made by forcing molten glass through an



orifice, followed by air, steam, or flame attenuation.  There are



three major classes of fibrous glass:  wool, textile, and



special-purpose fibers.  Glass wool fibers comprise approximately



90 percent of the total fibrous glass production and their major



use is in thermal and acoustical insulation.  They are typically



3-10 urn diameter but may range from 1-25 um diameter, and



therefore, may generate respirable airborne fibers.  Textile



fibers or continuous glass filament which account for 5-10



percent of the total fibrous glass are used in the manufacture of



textile products and as reinforcements in plastics, rubber and



paper.  Textile fibers are, in general, nonrespirable because



they have fairly large diameters with nominal diameters ranging



from 6-15 um.  Special-purpose fibers with small diameters,



representing less than 1 percent of fibrous glass production, are

-------
                                29






manufactured for certain highly specialized uses in thermal



insulation in aerospace vehicles and filter materials.  This



group includes fine fibers which have nominal diameters of 1-3 urn



and very fine glass fibers (or microfibers) with diameters less



than 1 jam.  These fibers are highly respirable (NRC, 1984).








II.1.1.  Fiber Deposition, Clearance and Retention



     Available information regarding the inhalation, deposition,



and clearance of glass fibers is fairly limited.  The results of



available studies suggest that fiber dimension is the most



important factor in the deposition and elimination of glass



fibers.  Coarse glass fibers thicker than 1.5 jum are likely to be



deposited mainly in the upper respiratory tract (nasopharyngeal



and tracheobronchial regions) and would have little chance for



alveolar deposition.  Further, longer fibers (>10 jum) are  less



able to penetrate the alveolar region of the lung.  Like other



fibrous particles, glass fibers are probably eliminated rapidly



from the upper airway via mucociliary clearance whereas fibers



deposited in the alveolar space appear to be cleared more  slowly,



primarily by phagocytosis and to a lesser extent via



translocation and possibly by dissolution.  Short fibers  «5 pm)



are believed to be removed mainly by macrophage uptake whereas



longer fibers may be cleared at a slower rate by dissolution.  In



general, short fibers are cleared more rapidly than longer



fibers, suggesting that fiber per fiber, short fibers are  less



likely to pose a toxicological concern.

-------
                                30






     The regional deposition of inhaled glass fibers has been



studied by Morgan et al. (1980) and Morgan and Holmes (1984a).



In these studies, rats were exposed for several hours by



inhalation (nose-only) to glass fibers of different diameters



(1.5 urn or 3 pm) and lengths (5, 10/ 30 or 60 urn).  The results



of these studies showed that for fibers with 1.5 urn diameter and



longer than 10 urn, fiber deposition in the lower respiratory



tract and alveolar region was low and decreased with increasing



fiber length.  Moreover, alveolar deposition of thicker fibers  (3



urn) was about one third of that of fibers of 1.5 urn diameter of



similar lengths.  These results, together with the previously



reported data on other asbestiform mineral fibers (Morgan, 1979),



indicated that alveolar deposition of fibers in the rat was



optimal with an aerodynamics diameter of 2 urn, which is



equivalent to an actual fiber diameter of approximately 0.5 urn.



Available data also demonstrated that in general, increasing



fiber length decreases the proportion of inhaled fibers deposited



in the alveolar region  (Harris and Timbrell, 1977; Harris and



Fraser, 1976).



     Immediately following deposition, there is a rapid decline



in the lung content of glass fibers.  Griffis et al. (1981)



reported that 41-48 percent of lung burden of glass fibers in



rats was cleared between daily exposures.  The initial decline



presumably represents early clearance from the upper respiratory



airways, with a half time of less than one day.  Fibers deposited



in the upper airways are cleared by mucociliary activity which



transports the fibers toward the oralpharynx.  Fibers are then

-------
                                31






swallowed/ passed into the gastrointestinal tract and excreted



into the feces.  It has been shown that dogs excreted



approximately 77 percent of the initial total burden of glass



fibers within 4 days after inhalation exposure (Griffis et al.,



1983).  Similarily, in the rat, more than 95 percent of the total



burden of glass fibers was associated with the gastrointestinal



tract following a 2-hour exposure (nose-only), which was all



excreted in the feces two days later (Morgan et al., 1980).



     The elimination of fibers from the alveolar region is much



slower than those in the upper airways via mucociliary



clearance.  The half time alveolar clearance of "TEL" glass



fibers in the rat was reported to be approximately 44 days



(Friedberg and Ullmer, 1984).  Short fibers appeared to be



cleared more efficiently than longer fibers.  Morgan et al.



(1982) showed that in the rat, more than 80 percent of glass



fibers less than 5 urn in length were cleared by one year



following intratracheal instillation whereas no significant



clearance of fibers greater than 10 urn length could be detected



over the same period.  Bellmann et al. (1986) also found that



short glass fibers (_<5 urn) cleared faster than longer fibers  (>5



urn) from the rat lung following intratracheal dosing.  This



study, however, showed that long glass fibers do clear from the



lung while long crocidolite asbestos fibers  (>5 urn) apparently do



not clear from the rat lung over one year.  On the other hand,



long chrysotile asbestos fibers appear to split into fibrils  as



reflected by the observed increase in the number of fibers over a



6-month period.

-------
                                32






     Fibers are cleared from the alveolar region by a variety of



mechanisms.  The major pathway involves the removal of fibers by



macrophage uptake.  It is believed that fiber-laden macrophages



(dust cells) move to the terminal bronchioles and are transported



by the mucociliary system to the upper respiratory tract.  These



dust cells could then be swallowed.  It would appear that the



difference in the lung clearance between short and long fibers



could be due to the fact that short fibers of less than 5 urn are



efficiently removed by phagocytosis whereas the macrophage-



mediated clearance is ineffective for fibers longer than 10 pmf



due to the inability of macrophages to completely engulf the



longer fibers (Bernstein et al., 1980; 1984; Morgan et al., 1982;



Morgan and Holmes, 1984a).



     The second pathway of fiber clearance from the alveoli



involves the lymphatic system.  Fibrous particles in the alveolar



space are removed, either by macrophages or by themselves via an



unknown mechanism, to the lymph nodes.  The fate of the fibers  in



the lymph nodes is not known although they may escape the lymph



nodes and enter the lymphatic and blood circulation, and may



migrate to other tissues.  There are few data available regarding



the translocation of glass fibers.  Glass fibers were found in



the tracheobronchial and mediastinal lymph nodes of animals at



different time periods after exposure to the mineral dusts by



inhalation or intratracheal instillation (Lee et al., 1981;



Bernstein et al., 1980, 1984; Wright and Kuschner, 1977).



Furthermore, it appears that short fibers are more readily



transported to the lymph nodes than longer fibers.  In the study

-------
                                33






by Morgan et al. (1982), measurements of the fiber content of the



hilar lymph nodes of rats killed after one year following



intratracheal instillation showed that approximately 4 percent of



5 urn glass fibers had been transferred from the lung to the lymph



nodes.  Smaller proportions of the 10 urn and 30 urn fibers had



been transported and no 60 urn fibers were detected.  With regard



to the translocation of fibers to other organ tissues/ only



minimal amounts of glass fibers were found in the liver, spleen,



and blood of animals exposed to the fibrous dust by inhalation



(Lee et al., 1979, Griffis et al., 1983).  Glass fibers were also



detected in the spleen of rats after 2 years following



intratracheal instillation.  Further, Monchaux et al. (1982)



reported recovery of fibers from all organs (blood, liver,



kidney, brain) at 90 days after intrapleural injection of glass



microfibers.  However, increased pressure caused by this method



of administration may have been partly responsible for these



results.



     It has been suggested that fibrous particles may also be



cleared by dissolution.  For glass fibers, the suggested evidence



comes from morphological observations showing limited breakage



and etching of the fibers retained over a long period following



dosing, and chemical analysis of the recoverd fibers showing some



changes of elemental composition (Johnson et al., 1984a; Le



Bouffant et al., 1984; Spurny et al., 1983).  These processes



would result in shorter, thinner fragments which then could be



cleared more efficiently by phagocytosis.  The solubility of



glass fibers in lung tissues appears to be dependent on fiber

-------
                                34






size.  In studies with rats, longer glass fibers dissolved more



rapidly than shorter ones (Morgan et al., 1982; Morgan and Holmes



1984a; Bernstein et al., 1980, 1984).  It has been suggested that



the dependency of dissolution on fiber length may be due to



differences in the intracellular and extracellular pH.  The



shorter fibers within macrophages are exposed to a lower pH of



7.17, while those outside are exposed to a higher extracellular



pH of 7.4 (Morgan and Holmes, 1984a).



     The solubility of glass fibers in lung tissues and in



physiological fluids has been shown to be greater than that of



amphibole fibers but may be similar or less then chrysotile



(Forster, 1984; Spurny, 1983a; Spurny et al., 1983).  The results



of other in vitro studies also indicate that glass fibers have



marked solubility rates in physiological fluids (Griffis et al,



1981; Leineweber, 1984; Klingholz and Steinkopf, 1984).  Glass



fibers of fine diameters degraded more rapidly than coarser ones



(Spurny et al., 1983; Forster, 1984).  Futhermore, the



dissolution of long glass fibers (50 jum) in saline was much



faster than that of short fibers (5 urn).  These results indicate



that the in vitro dissolution rate of glass fibers is



proportional to the surface area of the fibers (Leineweber,



1984).








II.1.2  Effects on Experimental Animals



     Fibrous glass has been extensively tested in laboratory



animals for the ability to induce lung tumor, mesothelioma, and



fibrosis.  Information on the design and results of available

-------
                                35





animal studies on glass fibers is summarized in Table 1



(pages 215-229).  Such investigations have been conducted by



several routes of exposure including inhalation, intratracheal



instillation, intrapleural injection/implantation, and



intraperitoneal injection.  Animal exposure by inhalation



represents the most relevant method for the assessment of risks



to man.  However, because of the technical difficulties and high



costs, fewer long-term inhalation studies have been conducted in



comparison with studies using the injection or implantation



method which are more sensitive and generally more



reproducible.  Injection studies are of value in screening the



test fiber for carcinogenicity and providing useful information



regarding the intrinsic biological activity and carcinogenicity



of the test fiber.








II.1.2.1  Oncogenicity



     None of the available long-term studies have provided



evidence of pulmonary or mesothelial carcinogenicity in animals



exposed to fine glass fibers,  glass microfibers.  or larger



diameter glass fibers (e.g., glass wool) by inhalation.  In



contrast, many studies involving intrapleural or intraperitoneal



administration of these fibers to animals have resulted in



increases in mesothelioma of the pleura or peritoneum,



respectively.  In addition, two of several intratracheal instil-



lation studies on glass microfibers also reported tumor induction



with both lung tumors and mesothelioma in hamsters and lung



tumors alone in the rat.  By using the intrapleural implantation

-------
                                36






method, Stanton and coworkers demonstrated that glass fibers less



than 0.25 urn diameter and greater than 8 urn length have



carcinogenic potential equal to that of asbestos fibers of



similar size distribution.  Other investigators also found that



long, thin glass fibers are highly carcinogenic by the injection



routes of exposure but are less effective than asbestos.








II.1.2.1.1  Inhalation Studies



     The earliest studies with fibrous glass were those by



Schepers and coworkers (Schepers and Delahunt, 1955; Schepers,



1955; Schepers, 1959a; Schepers, 1959b; Schepers, 1961) which



were summarized in a final report in 1976 (Schepers, 1976).  In



one series of experiments, guinea pigs and rats were exposed to



fairly large diameter glass wool fibers (average diameter close


                                                       •5

to 5 urn) at an average mass concentration of 0.145 mg/m  for



44 months, and at 0.03 mg/m  for 28 months, respectively.  In



another series of studies, guinea pigs, rabbits, rats, and



monkeys were exposed to dust from two types of glass fiber


                                      3            3
reinforced plastics at either 3.8 mg/m  or 4.6 mg/m  for various



time periods ranging from 8-24 months.  No pulmonary tumors were



reported in any exposed group.  These studies, however, were



inadequate to determine whether the fibrous products tested were



carcinogenic in animals by inhalation due to 1) extremely low



levels of fiber exposure, particularly for glass wool;



2) insufficient information on fiber size distribution  in the



dust cloud; and 3) poor survival of treated and unexposed



animals.

-------
                                37






     Gross et al. (Gross et al., 1970; Gross, 1976) reported



studies in which rats and hamsters were exposed for 2 years to a



very high concentration of uncoated glass fibers (135 mg/irr),



glass fibers coated with a phenol-formaldehyde resin (106 mg/nr),



or glass fibers coated with a starch binder (113 mg/m^).  All



three types of glass fibers in the dust cloud had an average



diameter of 0.5 jam and an average length of about 10 jum.  None of



the rats or hamsters exposed to any of the fiberglass products



developed lung or pleural tumors.  However/ it is not clear



whether there was a sufficient number of animals at risk from



late developing tumors due to a small number of animals and



apparent poor survival of exposed animals.  The survival pattern



of unexposed control animals was not available for comparison.



     Morrison et al. (1981) reported that 5 of 12 male A-strain



mice developed bronchogenic or septal cell tumors 90 days after



exposure to "crushed" glass insulation (80 percent were 6-11 jjm



long and 2-5 urn diameter) mixed in bedding material every 3 days



for 30 days.  However, the results of this study were



inconclusive because of 1) insufficient information on the actual



airborne glass fiber concentration; 2) lack of control animals



caged in normal bedding; and 3) short exposure period and small



number of exposed animals.



     In 1979 and subsequently in 1981, Lee et al. reported



studies in which rats and guinea pigs were exposed to glass fiber



aerosol at an extremely high dust mass of 400 mg/m  for



90 days.  The airborne dust particles had an average diameter of



1.2 urn and most particles were less than 2 urn long; thus, the

-------
                                38






dust particles were predominantly nonfibrous.  After 18 and 24



months post exposure, 2 of 19 rats and 2 of 8 guinea pigs



developed bronchial alveolar adenoma while none of 13 rats and 6



guinea pigs exposed to clean air as controls had pulmonary



tumors.  Since these findings were based on a small number of



animals, meaningful conclusions cannot be drawn from this



study.  Other limitations such as short exposure period further



limit the conclusion that can be made about this study.



     Goldstein et al. (1983, 1984) studied the effects of



inhalation exposure of very fine fibrous glass in male baboons



and compared them with the effects produced by crocidolite



asbestos.  Animals were exposed to a blend of Johns-Manville code



102 and code 104 glass microfibers (median diameter of airborne



fiber of 0.6 urn) at a mass concentration of 7.54 mg/m  (1,122



fibers/mL) for 35 months or UICC crodolite asbestos (median



diameter of airborne fiber of 0.38 urn) at a mass concentration of



15.8 mg/m3 (1,128 fibers/mL) for 40 months.  A total of 10



animals were used; the numbers in glass fibers exposed or



positive control group were not specified.  No neoplasms occurred



with either of the dusts at 6-7 months following the end of



exposure.  Since the exposure and observation periods were short



in relation to the lifespan of baboons, this study  is considered



inadequate for the evaluation of the oncogenic potential of



fibrous glass.  It should also be noted that neoplasms have been



rare even in previous studies of asbestos-exposed monkeys and



baboons.

-------
                                39






     Two studies by Wagner et al. (1984) and McConnell et al.



(1984) were undertaken as a joint effort to compare the



carcinogenic effects of glass microfibers with chrysotile



asbestos/ and to assess the comparability of results of similar



inhalation studies at two different locations.  In the study by



Wagner et al. (1984), specific pathogen-free (SPF) male and



female Fischer 344 (F344) rats (28 of each sex per group) were



exposed for 3 or 12 months to coated or uncoated glass wool,



glass microfibers, or UICC Canadian chrysotile asbestos.



Chrysotile and glass microfibers were highly respirable with



airborne fiber diameters ranging from 0.03 urn to 2 urn  (mean



diameter of 0.3 urn) while glass wool had larger airborne fiber



diameters ranging between 0.3 urn and 3 urn (mean diameter of



0.8 urn).  The respirable dust mass was 10 mg/m  in all cases.



The concentrations of airborne respirable fibers (diameter <3 urn,



length >5 urn) were 240 fibers/mL for uncoated glass wool,



323 fibers/mL for coated glass wool, 1436 fibers/mL for glass



microfibers, and 3822 fibers/mL for chrysotile.  Pulmonary



response was assessed in rats sacrificed at 3, 12, and 24 months,



and in animals that were allowed to live out their natural



lifespan.  One case of lung adenocarcinoma was found in animals



exposed to glass wool with resin (1/48) and glass microfibers



(1/48) while exposure to glass wool without resin resulted in one



case of benign lung adenoma (1/48).  In contrast, a total of 12



lung tumors  (11 adenocarcinoma, 1 adenoma) were produced in  the



chrysotile group.  All of the neoplasms were reported  to occur

-------
                                40






within 500-1,000 days after the start of exposure.  Unexposed



control animals developed no tumors.



     Comparable results were obtained in the study by McConnell



et al. (1984).  In this study, male and female SPF F344 rats were



exposed to the same glass microfibers or chrysotile asbestos



preparation as used by Wagner et al. (1984), targeted at a



respirable dust mass concentration of 10 mg/m  for 1 year.



However, the actual cumulative dose of glass microfibers was



approximately one half of that in the study by Wagner et al.



(1984).  Increased incidences of lung neoplasms were observed in



11 of 56 animals exposed to chrysotile but no tumors were found



in the glass microfiber group (0/55).  Two of 53 unexposed



animals had lung adenocarcinoma.  Most of the tumors were found



after 24 months.



     Analysis of the findings from these two inhalation studies



showed that there was no statistically significant difference in



tumor incidence between the unexposed controls and rats exposed



to glass microfibers (Rossiter, 1982).  In the study by Wagner et



al. (1984), there was also no significant difference in tumor



incidence between animals exposed to coated or uncoated glass



wool fibers and the negative controls.  However, both studies are



limited with regard to study design including the use of a



relatively small number of animals and short duration of



exposure.  Despite these limitations, these two studies



demonstrated that while glass microfibers and glass wool fibers



were not carcinogenic in rats under the conditions tested,



similar mass concentration of chrysotile asbestos produced a

-------
                                41



significant increase in the incidence of benign and malignant


pulmonary neoplasms in the rat.


     Smith et al. (1984, 1986) also found that glass microfibers


and large diameter glass fibers caused neither lung tumor nor


mesothelioma when inhaled by rats and hamsters.  As a part of a


comprehensive study, groups of male Syrian hamsters and female


Osborne-Mendel rats (50-70 animals/group) were exposed "nose-only"


for 24 months to one of the following dusts:   (1) highly


respirable glass microfibers (fiber product with mean diameter


0.45 urn) at a mean mass concentration of 3.0 +_ 0.6 mg/m


(approximately 3,000 fibers/mL) or 0.3_+_0.1 mg/m3 (300


fibers/mL); (2) fibrous glass  "blowing wool" (fiber product with


3.1 urn mean diameter) targeted at 10 mg/m  (100 fibers/mL);


(3) flame attenuated fibrous glass (fiber product with 5.4 urn mean


diameter) at either 12 mg/m3 (100 fibers/mL) or 1.32 mg/m3 (10


fibers/mL); (4) fibrous glass  insulation building (fiber product


with 6.1 urn mean diameter) at  9.0 mg/m   (25 fibers/mL).  Positive


control animals were exposed to UICC crocidolite asbestos at a

                            o
mass concentration of 7 mg/m   (3,000 fibers/mL).  One negative


control group was exposed to clean air (sham controls).  Following


the exposure period, test and  sham control animals were maintained


for their natural lifespans.   Another negative untreated control


group remained in cages throughout their lives.


     No primary lung tumors were  found in rats or hamsters exposed


to any of the fibrous glass dusts.  On the other hand, one


mesothelioma and two cases of  bronchoalveolar  tumors were detected


in 57 asbestos-exposed rats.   None of the hamsters exposed to

-------
                                42






crocidolite asbestos developed lung tumors or mesothelioma.



However, bronchoalveolar metaplasia/ possibly a preneoplastic



event in the development of epithelial tumors/ was significantly



elevated in hamsters exposed to crocidolite asbestos.  With the



exception of the occurrence of a bronchoalveolar tumor in a sham



control hamster/ none of the other sham control or unexposed



control animals developed lung tumors.  Thus, under the conditions



of these lifetime studies there was no evidence of carcinogencity



in rats or hamsters exposed to glass microfibers or large diameter



fiberglass.  The lack of significant tumorigenic response by



crocidolite asbestos observed in this study might well be due to



the use of a short-fibered material (approximately 95-97 percent



were less than 5 um long).



     The long-term effects of inhalation of glass microfibers and



glass wool were also studied in rats by Le Bouffant et al.



(1984).  Groups of 48 Wistar IOPS AF/Han rats (24 animals of each



sex) were exposed to French commercial resin-free glass wool



(Saint-Gobain)/ American produced glass microfibers  (Johns-



Manville code 100) or Canadian chrysotile asbestos at a respirable



dust mass of approximately 5 mg/m  for 12 or 24 months.  Because



of differences in fiber size distribution and proportion of non-



fibrous material present in the aerosols, the numerical concentra-



tions of respirable fibers greater than 5 um length as determined



by optical microscopy were varied, ranging from 48 fibers/mL for



glass wool (68 percent <1 um diameter), 332 fibers/mL for glass



microfibers (51 percent with diameters from 0.2-0.5 um), to 5,901

-------
                                43






fibers/mL for chrysotile (fiber diameter distribution not



specified).  The animals were sacrificed at 12, 24, and 28 months.



     No pulmonary tumors were found in animals exposed to glass



microfibers (0/48).  A single lung tumor was observed at 24 months



with the glass wool group (1/45).  In contrast, nine cases of lung



tumors were detected in the positive control group exposed to



chrysotile (9/47).  Negative control animals (unexposed) had no



pulmonary tumors.  Although there were no significant increases in



lung tumors in rats exposed to either glass microfibers or glass



wool fibers, this study is considered limited based on small



numbers of animals and a relatively low level of fiber exposure.



This study, however, demonstrated that under similar experimental



conditions and mass concentrations, chrysotile asbestos was more



potent in inducing lung tumors in rats than glass fibers.



     Mitchell et al. (1986) also reported no evidence of pulmonary



or mesothelial neoplasms in rats and monkeys following chronic



inhalation of fibrous glass of varying geometry and mass



concentrations.  In this study, groups of F344 rats (50 animals of



each sex per group) and male cynomolgus monkeys (15 per group)



were exposed to  (1) large diameter and long glass fibers with



binder (4-6 urn in diameter and >20 urn long) at approximately 15



mg/m  (Group I); small diameter and long glass fibers with binder



(0.5-3.5 urn in diameter and >10 jam long) at 15 mg/m3 (Group II);



(3) small diameter and long uncoated glass fibers  (<3.5 urn in



diameter and >10 urn long) at 5 mg/m3 (Group III); and (4) small



diameter and short uncoated glass fibers «3.5 urn and <10 urn long)



at 5 mg/m3 (Group IV).  Control animals (Group V) were exposed  to

-------
                                44






filtered air.  The rats were exposed for a total of 86 weeks while



the monkeys were dusted for only the 72 weeks.  The animals were



sacrificed following the termination of exposure.



     Neither pulmonary tumors nor mesothelioma were detected in



any treated monkey or rat groups.  However, short treatment and



study duration may have excluded observation of late developing



tumorigenic effects/ particularly in the monkeys.  Furthermore/



there was a low survival among treated and control rats.



Approximately 37 percent of rats (187 of 500 animals) died



spontaneously or were killed in a moribund condition before the



termination of study.  Many of the spontaneous early deaths were



due to mononuclear cell leukemia (MCL).  It was reported that



there was a statistically increased incidence of MCL in each glass



fiber exposed rat group.  However, the investigators performed



statistical analyses on the combined incidence in both males and



females rather than analyzing the incidence data for the male and



female populations separately.  Reanalysis of data using Fisher



exact test showed that for the male population only Group 3



(p = 0.024) and Group 4 (p = 0.002) displayed a significant



increase in MCL.  In the females/ the incidence was significant



only in Group I (p = 0.047).  The biological significance of this



finding remains uncertain since spontaneous increase in MCL is



commonly seen in aged F344 rats.








II.1.2.1.2  Intrapleural Implantation/Injection Studies



     Stanton and his colleagues reported a series of experiments



(Stanton and Wrench/ 1972; Stanton et al./ 1977 and 1981) in which

-------
                                45






they tested the ability of fibrous glass and other mineral fibers



(including asbestos) of diverse dimensional distributions, to



induce malignant neoplasms in female Osborne-Mendel rats by



intrapleural implantation of the mineral dusts.  Pledgets of



coarse fibrous glass were coated with 40 mg of the test fibers



suspended in gelatin and the pledgets were placed over the



visceral pleura of the rats after open thoracotomy.  Animals were



observed for two years at which time survived animals were



killed.  The greatest increase in pleural sarcomas was observed



for fibers with diameters less than 0.25 jam and lengths greater



than 8 urn, although relatively high tumor yields were also



produced with fibers having diameters up to 1.5 urn with lengths



greater than 4 urn.  These studies demonstrated that glass fiber



with dimensional distribution similar to that of asbestos was



equally carcinogenic as asbestos by intrapleural implantation.



     Similar findings were obtained in the study by Smith et al.



(1980) in which the tumorigenic effects of six fiberglass samples



were tested in hamsters by intrapleural injection of a single dose



of 25 mg of the test fiber.  Intrathoracic tumors occurred in 9 of



60 animals which received fibers with a mean diameter of 0.1 ium



and 82 percent longer than 20 urn.  Fibers with a mean diameter of



0.33 urn and 46 percent longer than 20 urn induced tumors in 2/60.



Fibers with a mean diameter of 1.23 um and 34 percent longer than



20 um also induced tumors in 2/60.  No tumors were found in groups



treated with the other three preparations containing fibers of



similar diameter range but shorter lengths with only 0-2 percent



longer than 10 um.  These results suggest that carcinogenicity is

-------
                                46






associated with length and diameter of fibers; the thinner and



longer the fiber, the more tumorigenic it is.



     Wagner et al. (1973, 1976, and 1984) also tested the



carcinogenicity of fiberglass of various types and size



distributions by intrapleural injection in rats and confirmed that



thin fiberglass was carcinogenic.  Glass microfibers (Johns-



Manville code 100), when injected as a single dose of 20 mg into



the pleura of rats produced a significant increase in pleural



tumors.  In the 1976 study, 4 of 32 Wistar rats (p = 0.01)



developed pleural mesothelioma while none of 32 control animals



had tumors.  Similarly, in the 1984 study, 4 of 48 Sprague-Dawley



rats treated with glass microfibers developed mesothelioma.  In



contrast, coarse glass fibers (Johns-Manville code 110) produced



no tumors in rats (Wagner et al., 1973, 1976) and only one case of



mesothelioma was found among 48 rats injected with glass wool



(Wagner et al., 1984).  When comparing these results with those



obtained with various types of asbestos in earlier experiments



reported in Wagner et al. (1973) using identical intrapleural



injection technique, finer glass fibers were considerably less



carcinogenic than some of the asbestos samples, while coarse glass



fibers were not tumorgenic.



     Monchaux and coworkers (Lafuma et al., 1980; Monchaux et al.,



1981) also reported induction of pleural tumors in rats following



intrapleural injection of 20 mg of fine glass fibers (JM 104).



The mean length of the test glass fibers was 5.8 urn and mean



diameter of 0.229 urn.  Pleural mesotheliomas were observed in 6 of



44  (14 percent) animals.  Higher tumor incidences were produced by

-------
                                47






UICC crocidolite (54 percent) and chrysotile (45 percent) asbestos



while control animals receiving saline alone had no tumors.  These



findings were consistent with results of other intrapleural



studies which showed that at a similar mass dose, crocidolite and



chrysotile asbestos were more potent in inducing mesothelioma in



rats than fine fibrous glass.








II.1.2.1.3  Intraperitoneal Injection Studies



     In a series of studies, Pott and coworkers investigated the



ability of fibrous glass to induce abdominal tumors in the rat by



the intraperitoneal (i.p.) route of exposure.  In the first series



of experiments as reported in 1972, 1974 and 1976 (Pott and



Friedrichs, 1972; Pott et al. 1974, 1976), a dose-related tumor



induction (2.5 - 57.5 percent) was produced in female Wistar rats



(40 per group) following intraperitoneal injection of a single



dose of 2 or 10 mg, or 4 doses of 25 mg of fibrous glass (S + S



106; mean diameter 0.5 urn; 72 percent <5 um long).  Positive



control animals receiving UICC chrysotile asbestos also developed



tumors in the peritoneal cavity in a dose-related manner (15-67



percent).  Histologically, nearly all the tumors from fibrous



glass or chrysotile treated animals were sarcomatous



mesothelioma.  In both treated groups, the latency period  for



tumor development was inversely related to the dose of fibers



injected.  No tumors were observed in negative control animals



receiving saline.



     Similar results were obtained with uncoated glass fibers of



type MN 104  (50 percent <0.2 um in diameter; 50 percent  <11 um

-------
                                48






long).  This study showed a dose-related increase in mainly



peritoneal mesothelioma in Wistar rats (80 animals/group)



following intraperitoneal injection of the test fiber at a single



dose of 2 or 10 mg, or 2 doses of 25 mg.  Glass fibers of type MN



112 (50 percent <1 urn in diameter; 50 percent <28 urn long)



produced a tumor incidence of 27.5 percent following an i.p. dose



of 20 mg (Pott et al. 1976).



     In a subsequent study, Pott et al. (1980) reported that



intraperitoneal injection of 10 mg of glass microfibers (JM 104)



to rats of 4 different strains resulted in different tumor rates



ranging from 51 to 79.6 percent.  The rat strains used in this



study included Wistar (Ivanovas)/ SIV (Ivanovas), Sprague-Dawley



(Hagemann)/ and Wistar (Hagemann).  No other details of the



experiments were available.



     These results were confirmed in a later study by Pott et al.



(1984) which showed a production of high incidences (40-70



percent) of abdominal tumors, primarily sarcoma or mesothelioma,



in Wistar or Sprague-Dawley rats following intraperitoneal dosing



with 2.5 or 10 mg of long glass microfibers (JM 104).  Shorter



glass microfibers  (JM 100) induced lower incidences of tumors (2-



10 percent).



     Comparable findings were reported by other investigators



using similar injection techniques.  Davis (1976) injected into



the peritoneal cavity of Balb/C mice and rats  (strain unspecified)



very fine glass fibers with an average diameter of 0.05 urn as a



single dose of 10 and 25 mg, respectively.  Three of 25 mice and 3



of 18 rats developed peritoneal tumors.  It was reported  that

-------
                                49






these tumors appeared identical to those produced in the



peritoneal cavities of rats and mice by injection of crocidolite



asbestos, as reported in earlier studies (Davis, 1974).



     Recently, Smith et al. (1986) reported a 32 percent incidence



of abdominal mesothelioma (8/25) in female Osborne-Mendel rats



following an intraperitoneal injection of 25 mg of 0.45 urn mean



diameter fiber.  UICC crocidolite asbestos produced tumors in 80



percent of the animals while no tumors were observed in saline



controls or untreated animals.








II.1.2.1.4.  Intratracheal Instillation Studies
     Variable results on the carcinogenicity of fibrous glass via



the intratracheal route have been reported.  Tumor induction by



fibrous glass was reported in one study by Mohr et al. (1984).  In



this study, groups of 136 male Syrian golden hamsters received



eight weekly intratracheal instillations of 1 mg of the dusts.



Thin fibrous glass (JM 104) of two different size lengths (mean



diameter of 0.3 jum, and mean length of either 7 urn or 4.2 urn) and



UICC crocidolite asbestos, were tested.  Neoplasms, including lung



carcinomas (4 percent), mesothelioma (27 percent) and thoracic



sarcoma (4 percent), were found in hamsters treated with glass



fiber samples at comparable rates.  Interestingly, the incidence



of mesothelioma in asbestos-treated animals was considerably lower



than that of fiberglass.  Control animals treated with titanium



dioxide nonfibrous dusts developed no mesothelioma or lung



carcinomas.

-------
                                50






     A subsequent intratracheal instillation study by the same



group of investigators (Pott et al., 1987a) reported a low



incidence of lung tumors in the rat treated with glass micro-



fibers.  Female Wistar rats were administered 20 weekly doses of



0.05 mg of JM 104/Tempstran 475 glass fibers (50% <3.2 urn long;



50% <0.18 urn in diameter).   Five cases of lung tumors (1 adenoma,



2 adenocarcinomas,  2 squamous cell carcinomas) were found among 34



treated animals.  In rats treated similarly with crocidolite



asbestos (50% <2.1  urn long; 50% <0.20 urn in diameter), there were



11 cases of lung tumors out of 35 animals examined.



     In contrast, several other intratracheal studies with fine



glass fibers have not produced positive results.  Gross et al.



(1976) found no tumors in rats or hamsters injected



intratracheally with multiple doses of uncoated glass fibers,



glass fibers coated with resin, or starch binder.  Glass fibers



tested in this study had an average diameter of 0.5 urn and



average length of 10 pm.  Wright and Kuschner (1976 and 1977)



also reported no tumor induction in guinea pigs injected with



12.5 mg of either thin, long glass fibers (90 percent >10 pm in



length) or 25 mg of shorter fibers (90 percent <10 urn) of similar



diameter (mean diameter <1 urn).



     Recently, Feron et al. (1985) reported no mesothelioma or



other tumors of the respiratory tract in Syrian golden hamsters



treated with JM 104 glass microfibers (31 percent <0.25 pm in



diameter; 89 percent <12 urn long) via intratracheal instillation



(1 mg every 2 weeks for 52 weeks).  Smith et al. (1986) also



found no tumors in female Osborne-Mendel rats following

-------
                                51






intratracheal instillation of 0.45 urn mean diameter glass fibers



(1 mg weekly for 5 weeks).








II.1.2.2  Fibrogenicity



     Fine fibrous glass and glass wool have been shown in several



animal studies to produce minimal interstitial dust cell reaction



without fibrosis following chronic inhalation.  The pulmonary



responses generally consist of macrophage infiltration with



alveolar dust cell collections, alveolar proteinosis, and



granuloma formation.  In one study/ fine fiberglass was reported



to produce focal fibrosis in baboons.  However, the small number



of animals and the lack of unexposed animal control group limit



the conclusions which can be made from this study.



     In contrast, extensive pulmonary fibrosis has been induced



in animals by intratracheal instillation and intrapleural



injection of fine fibrous glass.  Furthermore, marked peritoneal



fibrosis has been produced via injection of fine glass fibers



into the abdominal cavity of animals.  The results of these



injection studies showed that long, thin glass fibers are more



fibrogenic than short, thin glass fibers, while thick glass



fibers are apparently relatively inert, producing no significant



pulmonary response.  Pulmonary pathology induced by glass fibers



by these routes of exposure, including inhalation, are much less



severe than that produced by asbestos fibers in concurrent



experiments.



     Since the experimental details of most available studies are



already presented in the discussion of oncogenicity and are

-------
                                52

summarized in Table 1 (pages 215-229) only relevant information
and test results on the fibrogenic effects are discussed in the
following sections.


II. 1.2. 2.1   Inhalation Studies
     Gross et al.  (1976) found no development of pulmonary
fibrosis in rats and hamsters exposed to very high dose levels
(100 mg/m3) of uncoated or coated glass fibers (mean length of 10
urn; mean diameter of 0.5 urn).  The survival rate of treated
animals/ however, was poor.  Schepers et al. (1976) also reported
that glass wool and glass fiber reinforced plastics did not
induce fibrosis in rats/ guinea pigs, rabbits/ and monkeys/
following a two-year inhalation exposure to various
concentrations of the dusts  (0.03-4.6 mg/m  ) .  This study also
had a high incidence of mortality.
     In 1979 and subsequently in 1981 / Lee et al. reported that
the major pathological lesion found  in rats, hamsters, and guinea
pigs which were exposed for  90 days  to a very high glass fiber
dust cloud (400 mg/m3) with  a full lifespan follow up, was
alveolar proteinosis.  Very  slight alveolar interstitial fibrosis
occurred in a few old animals.  It should be noted that these
experiments used fibers of small aspect ratios (3:1) with only 7
percent of the fibers considered fibrous in shape.  Furthermore,
the exposure period was relatively short.
                                            WAGrfif, v-~ .•  -
                                               (202)200-3;},,"

-------
                                53



     Morissett et al. (1979) reported that a group of 20 male


albino mice which were exposed to respirable glass fiber «3 urn


diameter and <10 urn length) at 1,070 fibers/mL for six weeks did


not develop pulmonary fibrosis.  This study is considered limited


because of the short duration of exposure.


     In 1980, Johnson and Wagner reported that electron


microscopic examination of lung tissues of rats (two SPF Fischer


rats/group) exposed to 10 mg/m  of glass microfibers, resin


coated glass wool or uncoated glass wool for 50 weeks revealed


focal fibrosis.  However, pulmonary fibrosis was not found in the


more extensive investigation by Wagner et al. (1984).  Fischer


344 rats developed minimal interstitial cellular reaction (grade


3.4) following a one-year exposure to glass microfibers (10

    o
mg/mj) and a one-year follow up.  Animals exposed to glass wool


either with or without resin at similar mass concentration (10


mg/m3) had no significant pulmonary responses (grade 2.6 and 2.4,


respectively).  On the other hand, animals exposed to UICC


Canadian chrysotile showed evidence of early interstitial


fibrosis (grade 4.1).


     McConnell et al. (1984) also found no evidence of pulmonary


fibrosis in Fischer 344 rats following a one-year exposure to


very fine JM 100 glass fibers, obtained from the same source as


described in Wagner et al.  (1984).  Animals which were exposed to


UICC crocidolite developed mild pulmonary fibrosis.  It should be


pointed out the accumulative exposure of glass fibers in this


study was only half of that achieved in Wagner's study.

-------
                                54






     Le Bouffant et al. (1984) found only minimal pulmonary



response in rats after one year exposure to fine fibrous glass or



glass wool at 5 mg/m .  They were limited to alveolar and



interstitial macrophage reactions, with mild septal fibrosis in



the case of glass microfibers.  In the case of chrysotile fibers/



only hyperplastic changes of the alveolar lining were observed;



pulmonary fibrosis was not detected.



     Smith et al. (1986) exposed groups of rats and hamsters to



fine fibrous glass (0.45 um mean diameter; 300 or 3,000



fibers/mL) and coarse fibrous glass (>3 um diameter; 25-150



fibers/mL) for 2 years.  Pulmonary fibrosis was not found in any



of the treated animals (50-60 animals/group) with fibrous



glass.  Many of the rats and hamsters exposed to UICC crocidolite



asbestos, however, developed prominant pulmonary fibrosis.



     Baboons were reported to develop focal peribronchiolar



fibrosis following exposure to respirable glass microfiber dust



clouds for 35 months at 1,122 fibers/ml (7.54 mg/m ).  The test



fibers had a mean diameter of 0.5 um and median length of 6 um.



Pulmonary lesions induced by glass microfibers were morphologi-



cally similar to those produced by crocidolite asbestos; however,



the incidence and severity in unexposed control animals were not



reported (Goldstein et al., 1983).



     In the study by Mitchell et al. (1986), cynomolgus monkeys



and Fischer 344 rats were exposed via inhalation to dust clouds



of fibrous glass of varying geometry and concentrations (5 or




15 mg/m3) for 13 months and 21 months, respectively.  There was



no evidence of lung fibrosis in either species.  Both species had

-------
                                55
pulmonary macrophage aggregates and granulomas containing fibrous



glass.  The rats had grossly visible pleural plaques which were



not seen in the monkeys.  No positive controls were included in



this study.







II.1.2.2.2  Intratracheal Instillation Studies
     Kuschner and Wright (1977) reported that long/ thin glass



fibers (92 percent >10 urn in length; <1 urn in diameter) produced



a marked fibrotic reaction in the guinea pigs following



intratracheal injection of a single dose of 12 mg.  The



instillation of short, thin glass fibers (93 percent <10 yum) at



similar doses produced only macrophage aggregation in the



alveoli.



     In the study by Pickrell et al. (1983), groups of 20 male



Syrian hamsters were intratracheally instilled with one of two



uncoated glass microfibers (2 and 7 mg; 0.1-0.2 urn diameter) or



one of three commercial glass fiber samples (17-21 mg; 2.3-4.1 um



diameter), or UICC crocidolite (6 mg; 0.24 jam diameter).  The



thinner glass microfiber (0.1 um) caused deaths from pulmonary



edema shortly after instillation.  Increased collagen deposition



and mild pulmonary fibrosis were observed in animals treated with



the thicker glass microfiber (0.2 um) and one type.of commercial



glass fiber (2.3 m) at 11 months after instillation.  However,



the microfibers produced a greater effect than the commercial



type, while crocidolite asbestos induced the greatest response.



No pulmonary responses were observed in animals treated with the

-------
                                56
other two types of commercial glass fibers which had larger



diameter (3-4 urn).



     Marked lung fibrosis was also found in 27 percent of female



Osborne-Mendel rats (7/22) treated intratracheally with 2 mg of



0.45 urn mean diameter glass fibers once a week for five weeks



compared to saline control animals.  However, the incidence of



pulmonary fibrosis in positive control rats instilled with



crocidolite asbestos was much higher and the lesions were more



severe (Smith et al./ 1986).








II.1.2.2.3.  Intrapleural Injection Studies



     The relationship between fiber dimension and fibrogenicity



was also demonstrated by Davis (1976).  Groups of 25 Balb/c mice



received 10 mg of one of 4 samples of glass fibers of varying



lengths and diameter, by the intrapleural route.  Short fiber



samples «20 pm) of both large (3.5 urn) and small diameter (0.05



urn) produced only small discrete granulomas with minimal



fibrosis.  Long fiber samples (>100 urn) produced massive



fibrosis, which was comparable to that induced by asbestos.







II.1.2.2.4  Intraperitoneal Injection Studies



     Pott et al.  (1974) reported that glass fibers (average



diameter of 0.5 urn; 72 percent less than 5 urn in length), when



injected into the abdominal cavity of rats as a single dose of



50 mg, produced marked peritoneal fibrosis.  The effect was less



extensive at lower doses (2 and 10 mg).  Similarly, Smith et al.



(1986) found extensive peritoneal fibrosis in female Osborne-

-------
                                57
Mendel rats injected intraperitoneally with 25 mg of 0.45 jjm mean



diameter glass fibers.
II.1.3.  In Vitro studies
II.1.3.1  Genotoxicity



     Several studies on glass fibers have been performed, ranging



from bacterial mutation tests to transformation studies in



mammalian cells.  Most studies dealt with Code 100 and Code 110



fiberglasses of various lengths.  These two glass fibers differ



in their diameters with mean diameters of roughly 0.12 urn and



1.9 jum, respectively.  Glass fibers do not appear to induce gene



mutations in bacterial cells, although this evidence is very



limited.  The two major effects that consistently appear with



Code 100 fiberglass are aberrations and transformation in



cultured cells.  The cytogenetic effects seen by Code 100



fiberglass appear to be less effective than chrysotile asbestos,



although sometimes comparable to crocidolite asbestos.  Other



cytogenetic effects, such as induction of sister chromatid



exchanges (SCE) and micronuclei, were not seen; again however,



only a very limited number of studies were available.  Code 110



fiberglass does not appear to have effects comparable to Code 100



fiberglass.








II.1.3.1.1.  Mutational Effects
     Glass fibers and several asbestos samples were tested in two



bacterial mutation tests (Chamberlain and Tarmy, 1977).  The two



glass fibers examined were Code 100 (mean length of 2.7 urn, mean

-------
                                58
diameter of 0.12 urn) and Code 110 (mean length of 26 |im, mean



diameter of l.S^m) fiberglasses.  The asbestos samples included



UICC Canadian chrysotile/ UICC crocidolite as well as a "cleaned



crocidolite" (Magnetite), UICC amosite, UICC anthophyllite and



SFA chrysotile.  All samples were found negative in the



Salmonella/mammalian activation test in strains TA1535 and TA1538



with and without activation.  The asbestos samples were tested at



0.1 and 1.0 mg/plate, but the glass fiber concentrations were not



specifically stated.  Similar negative results were found in the



_E_. coli mutation test with strains B/r, WP2, WP2 uvrA and WP2



uvrA pol A.  All samples were tested up to 1,000 ug/plate in the



_E_. coli test, except for Code 110 fiberglass (100 jug/plate top



dose).  The authors suggest that the negative results may be due



to a lack of phagocytosis of fibers by bacteria (bacteria



apparently do not phagocytize).  Also, in general, bacteria



appear resistant to the cytotoxic effects of fibers whereas



mammalian cells are sensitive.  Despite the problems of not



specifically stating the concentrations for glass fibers and the



lack of data using other strains (e.g. TA98, TAlOO) in the



Salmonella assay, it appears that bacterial systems may not be



appropriate to assay the potential mutagenicity of fibers.








II.1.3.1.2.  Chromosomal Effects
     Glass fibers were examined  in several  studies  for



chromosomal effects and were compared  to  the  effect induced  by



chrysotile and/or crocidolite asbestos.   The  first  study  to



examine the potential effects of glass fiber  and  glass  powder in

-------
                               59
Chinese hamster ovary (CHO) cells was reported by Sincock and



Seabright (1975).  They found that chrysotile and crocidolite at



0.01 mg/mL both induced polyploid cells, cells with fragments/



and other chromosomal changes (such as breaks and double minutes)



as well as an increase in the percentage of abnormal cells.



Glass fiber and glass powder at 0.01 mg/mL appeared to cause no



effect different from controls.  However/ an effect by



specifically sized glass fibers cannot be ruled out by this



study, as the dimensions of the glass fibers were not specified.



     In subsequent studies, Sincock (1977) and Sincock et al.



(1982) examined the potential chromosomal effects of glass fibers



in CHO cells in more detail.  Asbestos samples/ including



chrysotile/ crocidolite/ amosite, and anthophyllite were also



examined.  Chrysotile and crocidolite up to 0.1 mg/mL for 48 or



72-hour cell exposure consistently induced high levels of



chromosome damage including increases in breaks/ dicentrics,



inversions/ rings/ percent abnormal cells/ and polyploidy.  Code



100  (up to 0.1 mg/mL; lengths 2.7-26 urn) induced a significant



increase in the same parameters, but at levels usually less than



that for chrysotile and crocidolite.  It produced polyploidy at



levels similar to amosite and anthophyllite.  Code 110 (lengths



2.7-26 urn), glass powder (coarse borosilicate) and glass of 2 urn



diameters but of specific lengths (<10 urn, 25 urn, 50 urn and



100  urn) all had no effect.



     Sincock et al. (1982) also examined the potential effect of



the  asbestos and glass fiber samples  (described above) on



cultured human cells.  Five different fibroblast cell strains

-------
                                60
(not exceeding passage 15) and two different lymphoblastoid cell



lines were used.  No increase in chromosomal damage was noted for



any sample assayed.  The authors also searched for fiber-induced



micronuclei in the lymphoblastoid lines, but noted no increase



over controls.  It was shown in this report that cultured CHO



cells sustained damage to fibers, but the cultured human cells



were not overtly damaged.  There is no apparent explanation,



although the authors suggest that the difference may be due to



differences in excision repair, the nature of the transformed



lines, or the species of origin.  The phenomenon! seems similar



for the asbestos and glass fibers tested by these authors.



     Code 110 fiberglass and its respirable fraction were tested



in Chinese hamster lung (V79) cells (Brown et al., 1979b).  The



fibers were milled to lengths <200 jam.  The respirable fraction



was obtained from collecting respirable dust and was designated



110R.  Code 110 had 9 x 109 fibers/g and 110R had 25.2 x 109



fibers/g of which a 20 ug/mL concentration for both was tested in



the V79 cells.  The 110R sample presumably had similar diameters



as 110, but had shorter lengths.  In comparison to a crocidolite



positive control, 110 fibers had no observed chromosomal



effect.  110R however, induced increased fragments, breaks, and



percent of cells with abnormal spreads over the negative



control.  The 110 sample was weakly cytotoxic to V79 cells, but



the 110R sample induced noticeable cytotoxicity.



     Oshimura et al. (1984) examined the cytogenetic effects of



several asbestos and fiberglass samples on tertiary cultures of



Syrian hamster embryos.  Chrysotile induced time- and

-------
                               61
concentration-dependent increases in frequencies of cells with


numerical changes (aneuploidy and tetraploidy), chromosomal


aberrations (breaks/ fragmentation, exchanges, dicentrics),


binuclei, and micronuclei after 24- and 48-hour treatments

                          *\
(concentrations at 2 jjg/um1').  Code 100 fiberglass and


crocidolite had similar effects (48 hour exposure; 2 ug/^m^), but


both of these were less effective in inducing cytogenetic changes


than chrysotile.  Code 110 fiberglass and alpha-quartz (non-


fibrous mineral dust) were without significant effects.  Milling


of code 100 samples reduces the length (not given) as well as the


cytogenetic effect.  This suggests that the appropriate length,


not the chemical composition, of Code 100 fiberglass is


responsible for the cytogenetic effects seen in this study.


     Glass fiber and asbestos samples were tested for their


ability to induce sister chromatid exchanges (SCE) in CHO and


cultured human cells (Casey, 1983).  Chrysotile, crocidolite,


Code 100, and Code 110 fiberglasses were added to cell cultures


under two regimes.  Concentrations of 0.001, 0.01 and 0.05 mg/mL


were added to cells one hour after seeding in procedure one, and


cells in suspension were exposed to 0.01 mg/mL before seeding in


procedure two.  By either treatment, no fiber tested induced SCE


over the control frequency in CHO cells, human lymphoblastoid


cells, or primary human fibroblasts (8-10 passages).  No cell


cycle effect was seen in the fibroblasts under either procedure


and a mitotic delay was seen in CHO cells only with procedure two


(except for Code 110).  It appears that while glass fibers may

-------
                               62
induce aberrational and ploidy type changes in cells, glass


fibers do not induce SCE, at least based on this study.





II.1.3.1.3.  Transformation Effects


     Two studies report results from transformation studies using


asbestos and glass fibers.  In the earlier report, Sincock (1977)


described the effect of chrysotile, crocidolite, and coarse


fiberglass (Code 110) on murine 3T3 cells.  The two asbestos


fibers (0.01 mg/mL) induced foci indicative of transformation


after only 7 days of exposure.  Coarse fiberglass had no apparent


effect.


     Hesterberg and Barrett (1984) examined the effect of many


more fiber samples on tertiary cultures of Syrian hamster


embryos.  Chrysotile and crocidolite produced linear increases in

                                                     o
transformation frequency  (concentrations 0.25-2 ug/cm  ) with


chrysotile more potent than and twice as cytotoxic as


crocidolite.  Extracted chrysotile actually induced a  3-fold


higher frequency than unextracted samples, indicating that


possible contaminating organics may not have a role in fiber-

                                                     ^
induced transformation.  Code 100 fiberglass (2 ug/cm  ; 9.5-16 pm


length) was as active as chrysotile in transforming cells.


Milled Code 100 (0.95-1.7 urn length) and milled chrysotile both


exhibited greatly reduced transformation and cytotoxic


activities.  The reduction is presumably due to the reduction in


fiber length.  Code 110 fiberglass was less toxic than Code 100


and was 20-fold less potent than Code 100 for transformation.


Two nonfibrous mineral dusts  (alpha-quartz and Min-U-Sil) also

-------
                                63
induced concentration dependent increases in transformation, but



at much higher concentrations (10-75 ^ug/cm ) than chrysotile and



Code 100 fiberglass.  They were also much less toxic than



chrysotile.  The authors suggest that the slopes of the response



curves indicate a one-hit mechanism for transformation that would



suggest a direct effect by the fibers.
II.1.3.2.  Cytotoxicity



     Fibrous glass was tested in several in vitro studies to



determine its cytotoxic potential in various cell culture



systems.  Glass microfibers were found highly cytotoxic to lung



and peritoneal macrophages, P388D1 macrophage-like cells,



phagocytic ascites tumor cells, Chinese hamster lung fibroblast



V79-4 cells, rabbit lung fibroblasts, type II human alevolar A549



cells, and human bronchial epithelial cells.  The cytotoxic



effect of thin glass fibers appears to approach that of asbestos



fibers.  On the other hand, coarse fiberglass (code 110) had



little or no cytotoxic effects although its respirable fraction



was found to have some cytotoxic activities in one study.  These



studies also indicated that long, thin glass fibers are more



cytotoxic than short, thin fibers.  This may be due to the fact



that long fibers are incompletely phagocytized and, as a result,



may damage the cell membrane and may cause subsequent release of



enzymes, followed by cell death.  The ability of fibrous glass to



cause cellular membrane damage, as measured by hemolysis of red



blood cells, has been reported, but a varying degree of hemolytic



activity was found.

-------
                                64
II.1.3.2.1.  Erythrocytes



     Available data on the hemolytic activity of glass fibers are



limited.  Jaurand and Bignon (1979) reported that glass fibers



had a poor hemolytic effect, compared to UICC chrysotile, when



incubated with human red blood cells.  However, the hemolytic



activity of glass fibers was similar to that of UICC crocidolite.



In contrast, Ottolenghi et al. (1983) showed that Pyrex glass



fibers (dimensions not specified) at 100 ug/mL did not cause



hemolysis in chicken erythrocytes.  Amosite also had no effect,



while chrysotile asbestos induced high hemolytic effect.  In an



abstract, Nadeau et al. (1983) reported that glass microfibers



(diameter of 0.2 urn; length of 221 urn) induced marked hemolytic



activity in rat erythrocytes.  No further details were available



for a conclusive evaluation.








II.1.3.2.2.  Phagocytic cells



     Tilkes and Beck (1983a) examined the cytotoxic effects of



glass microfibers of different size distributions in guinea pig



and rat lung macrophages.  The release of lactate dehydrogenase



(to demonstrate plasma membrane permeability) and beta-



glucuronidase (to indicate lysosomal permeability) were measured



as indicators of cytotoxicity.  It was shown that macrophage



toxicity (100 ug/mL) was length dependent; the highest toxicity



was seen with fibers longer than 5 urn.  For fibers of similar



physical size dimension, the cytotoxic effects of amosite and



crocidolite asbestos and glass fibers were equivalent when doses



were gravimetrically equivalent.  Glass microfibers (JM 100) also

-------
                                65
caused a significant depression in phagocytic activity of



macrophages.  Phagocytosis was assayed quantitatively by



determining the-amount of luminescence produced after the



addition of serum-opsonized zymosan A particles.



     These results confirmed the earlier findings in the study by



Beck et al. (1972) showing that fine fiberglass (0.25-1.0 urn



diameter; 1-20 urn length) induced an increase in cell membrane



permeability of guinea pig alveolar and peritoneal macrophages/



as measured by increased release of lactate dehydrogenase and



lactate levels.



     Pickrell et al. (1983) tested the in vitro cytotoxicity of



two uncoated glass microfiber insulation materials (0.1-0.2 urn



diameter), three types of fibrous glass-containing household



insulation (2-4 urn diameter) and crocidolite asbestos (0.25 urn



diameter) in pulmonary alveolar macrophages isolated from Beagle



dogs.  It was reported that the most cytotoxic of the fibers



tested was crocidolite asbestos.  Household insulations were not



cytotoxic at the highest concentration tested.  Both types of



glass microfibers had cytotoxicities intermediate between



household insulation and crocidolite asbestos.



     In an abstract/ Nadeau et al. (1983) also reported that long



glass microfibers (0.2 um x 221 turn) were highly cytotoxic to rat



pulmonary alveolar macrophages.  However, the experimental



details were not provided in this report.



     Brown et al. (1979a) studied the cytotoxicity of JM code 100



glass microfibers (nominal diameter of 0.05-0.09 um) and JM code



110 glass fibers  (nominal diameter of 1.5-2.49 um) in mouse

-------
                                66
peritoneal macrophages.  Code 100 microfibers were more cytotoxic



than code 110 fibers as reflected by 2-3-fold differences in the



release of lactate dehydrogenase and beta-glucuronidase levels at



160 ug/mL.  Similar findings were obtained by Davies (1980) who



demonstrated that fine glass fibers (JM 100) at a concentration of



160 ug/mL were cytotoxic toward mouse peritoneal macrophages.



Coarse glass fibers code JM 110 had no cytotoxic effects but the



respirable fraction of JM 110 glass fibers had some cytotoxic



activities.



     Tilkes and Beck (1980, 1983b) investigated the cytotoxicity



of fibrous glass in phagocytic ascites tumor cells derived from



Wistar rats/ as measured by the release of lactic dehydrogenase



and the inhibition of cell proliferation as determined by cell



count, DNA, RNA, and protein synthesis.  It was found that for



long glass fibers (>20 urn), the thinner the fiber, the greater the



toxicity  in this cell culture system.  In addition, a glass fiber



fraction with comparable geometry to a UICC chrysotile asbestos



fraction exhibited the same high cytotoxicity.



     In the study by Lipkin (1980), borosilicate glass fiber



(dimension unspecified) which was obtained from the same lots used



by Stanton et al. (1977) was found highly toxic to P388D1



macrophage-like cells at a concentration of 100 ug/mL.  The



cytotoxicity of glass fibers in this cell culture was well



correlated with potency in pleural sarcoma induction reported by



Stanton et al.  (1977) in intrapleural  implantation studies.

-------
                                67
II.1.3.2.3.  Nonphagocytic cells



     Brown et al. (1979b) studied the effect of glass wool on



Chinese hamster lung fibroblasts (V79-4 cells) and human alveolar



type II (A549) cells.  Respirable fractions of coated or uncoated



glass wool produced a dose-dependent inhibition of cell growth



(10-50 ug/mL) of V79-4 cells.  Uncoated glass wool also produced a



significant increase in the number of giant cells when added to



the A549 cell cultures at 200 ug/mL.



     Chamberlain et al. (1980) reported that code 100 glass



microfibers (dimensions and concentrations not provided) reduced



the colony forming ability of V79-4 cells and induced giant cells



in A549 cell cultures.  In contrast, code 110 coarse glass fibers



had no effect.  The actual data were not provided in this study to



fully evaluate the findings.



     In a study by Richards and Jacoby (1976), glass fibers



(dimensions unspecified) caused a slight cytotoxicity to rabbit



lung fibroblasts when added to cell cultures at 50 jug/mL.



Fiberglass also induced morphological changes and alterations in



reticulin deposition in the fibroblast cultures.  In contrast,



UICC chrysotile asbestos was highly cytotoxic to fibroblasts and



caused more extensive morphological changes in these cultures at a



similar mass concentration.



     Glass fibers (1-100 ug/mL) were also found to induce a dose-



dependent inhibition of clonal growth rate of human bronchial



epithelial cells (Haugen et al., 1982).  In this cell culture



system, UICC chrysotile was more cytotoxic than glass fibers by



more than 100-fold.

-------
                                68
II.1.4.  Assessment of Health Effects
     Existing studies have provided no clear evidence of a
carcinogenic or fibrogenic hazard in humans.  However, available
animal studies show that fine fibrous glass is carcinogenic and
fibrogenic by the injection route of exposure.  Thus, there remains
a concern for possible health hazards from inhalation exposure to
fine fibrous glass, i.e., fibers with diameters less than 3 urn.  A
low health concern is also raised for exposure to glass wool, which
does contain some respirable fine fibers.  As for textile fibers
(continuous glass filaments) which are generally nonrespirable,
they would appear to pose little hazard to exposed humans.  On the
basis of available animal data, it is concluded that all fiberglass
categories appears to be less pathogenic than asbestos.

II.1.4.1  Oncogenicity
     Available health and toxicological information seems to
indicate that the oncogenicity of fibrous glass varies for the
three major categories i.e., fine fibrous glass, glass wool, and
continuous glass filament.  The variable oncogenic potential for
these classes of fibrous glass appear to be related to their
different fiber size distributions.
     By using the U.S. EPA weight-of-evidence criteria for car-
cinogenicity (USEPA, 1986), fine fibrous glass and glass wool may
be categorized as possible human carcinogens  (Category C) on the
basis of inadequate evidence of carcinogenicity in humans and
limited evidence in animal studies.  On the other hand, continuous
glass filament is not classifiable as to human carcinogenicity
(Category D) due to inadequate evidence of carcinogenicity from
epidemiological  and  animal  data.

-------
                                69
     Available data from recent cohort studies suggest that



workers engaged in the manufacture of glass wool and small



diameter fibers might be at increased risk of developing



respiratory cancer.  Small excesses of respiratory cancer death



have been observed among workers exposed to glass wool and small



diameter glass fibers but no excess of respiratory cancer has



been found among glass filament workers.  A dose-related trend



has not been found although it should be noted that exposure to



fibrous glass has been extremely low.  The causal relationship



between fibrous glass exposure and the development of respiratory



cancer is therefore not considered credible at this time.  There



is also inadequate evidence of an increased mortality from



mesothelioma in available MMMF cohorts.  On the basis of



available information, the weight of evidence of carcinogenicity



of fibrous glass, i.e., glass wool, continuous glass filament,



and small diameter fibers (fine fiberglass), from studies in



humans is considered inadequate.  Since the results of relevant



epidemiological studies on fibrous glass have been reviewed in



details and assessed in a report by Battelle (1988), only a brief



description of the study design and findings are presented here.



     Enterline et al. conducted a large cohort study on  fibrous



glass workers from 11 plants in the U.S..  These workers had at



least one year exposure between 1945 and 1963.  For those working



in facilities where small diameter fibers were prevalent, the



criterion was greater than six months of exposure. The cohort's



mortality experience was traced through 1977 in the early study



(Enterline et al, 1983) and in the subsequent studies mortality

-------
                                70
was followed to 1982 (Enterline et al., 1986; 1987).  The average



level of exposure to glass fibers (<3 urn in diameter) for all



fiberglass plants was 0.039 fibers/mL.



     In this study/ a slight excess of mortality from respiratory



cancer was observed among glass wool workers which was



nonsignificant based on local rates but was statistically



significant compared to national rates.  Mortality from



respiratory cancer increased nonsignificantly with time from



exposure but was not related to duration of exposure/ cumulative



exposure, or average intensity of exposure.  In the glass



filament subcohort, there was no excess of respiratory cancer and



no upward trend with time since first exposure, duration of



exposure, or average intensity of exposure.  Among workers in 4



fiberglass plants ever exposed to small diameter glass fibers (<3



urn), there was a nonsignificant excess of respiratory cancer



mortality which increased nonsignificantly with more than 30



years since onset of exposure.  However, the small number of



deaths limits any definitive conclusion regarding the



relationship between fine fiberglass exposure and respiratory



cancer.  The results of a nested-case control study using



respiratory cancer cases among fibrous glass (type unspecified)



workers showed a statistically significant association between



respiratory cancer and smoking but not between respiratory cancer



and cumulative fiber dose (Enterline et al., 1986; 1987).



     Similar results were obtained in the European study.



Simonato et al. (1985; 1986a; 1986b) also performed a historic



cohort investigation of glass wool workers from five plants and

-------
                                71
continuous glass filament workers from two facilities in Europe.



This study cohort consisted of men and women employed with at



least one year of employment from 1933-46.  Mortality was



followed to 1982 and risks were also examined for early, middle



and late production phases.  In the glass wool cohort/ there was



no overall excess of lung cancer deaths by using the local



mortality rates but there was a small nonsignificant excess when



compared to national rates.  Mortality from lung cancer increased



nonsignificantly with time since first exposure but was not



related to duration of exposure or to different technological



phases, reflecting differences in the intensity and quality of



exposure.  Among glass filament workers there was no excess of



lung cancer and no upward trend with time since first exposure or



duration of exposure.



     A third study by Shannon et al. (1986) reported a



significantly elevated risk for lung cancer in a small Canadian



glass wool cohort.  However, analyses of lung cancer deaths by



duration of employment and time since first exposure indicated no



consistent dose-related trends.



     There were no excessive mesothelioma deaths reported in the



two large cohort studies on MMMF workers exposed to fiberglass and



mineral wool.  Simonato et al. (1985) observed one death due to



mesothelioma in the European study of 24,000 workers.  Enterline



et al. (1986) reported two mesothelioma deaths in a cohort of



16,000 workers followed for 36 years.  However, an investigation



by Engholm et al. (1986) reported an excess number of mesothelioma



in the Swedish construction industry.  The study population

-------
                               72
consisted of 135,000 male workers exposed to MMMF (no distinction



between exposure to fibrous glass and mineral wool).   There was a



significantly increased mortality from pleural- mesothelioma in the



Swedish cohort.   However/ possible confounding by asbestos



exposure and several limitations of the study (e.g.  exposure



defined by job category and no monitoring data to define



categories) limit the conclusions that can be made about this



finding.



     Experimentally, there is insufficient evidence for the



carcinogenesis of fibrous glass in animals by inhalation.  Fine



fibrous glass (including glass microfibers) and glass wool have



been tested in several long-term inhalation studies, in several



animal species including the rat (Wagner et al., 1984; McConnell



et al., 1984; Le Bouffant et al., 1984; Smith et al., 1986),



hamster (Smith et al., 1986), monkey (Mitchell et al., 1986) and



baboon (Goldstein et al., 1983).  There was no statistically



significant increase in the incidence of lung tumors or pleural



mesothelioma in any of these studies; only a few tumors of the



respiratory tract occurred in some experiments in rats (Wagner et



al., 1984; Le Bouffant et al., 1984).  Although none of the



available inhalation bioassays is considered adequately studied,



collectively they do demonstrate that at equal mass concentrations



and similar experimental conditions, chrysotile asbestos generally



induced significant increases in lung tumors while fine fibrous



glass and glass wool did not cause significant tumorigenic



responses in laboratory animals following chronic inhalation



exposure.

-------
                               73
     However, data from studies in which glass fibers were



administered by nonphysiological routes indicate that a carcino-



genic hazard potential does exist for glass wool and fine fibrous



glass, in particular, for glass microfibers which contain a



considerable number of long, thin fibers.  Glass wool, fine



fibrous glass and glass microfibers were not found to cause tumors



in a number of intratracheal instillation studies in rats (Smith



et al. 1986; Gross et al., 1976), hamsters (Feron et al., 1985;



Gross et al., 1976) and guinea pigs (Wright and Kuschner, 1976,



1977).  However, in one study, lung tumors and pleural mesotheli-



omas were observed in hamsters by intratracheal instillation of



glass fibers with a median diameter of 0.3 urn in hamsters (Mohr et



al., 1984).  In another study by the same laboratory, lung tumors



were also induced in rats instilled intratracheally with glass



microfibers  (Pott et. al., 1987a).  In studies where various



samples of glass microfibers «1 urn diameter) were tested by



intrapleural implantation (Stanton et al., 1977, 1981) or



injection (Smith et al., 1980; Wagner et al., 1976, 1984; Monchaux



et al., 1981) variable incidences of pleural tumors were induced



in rats.  Furthermore, peritoneal mesotheliomas or sarcomas were



found in the abdominal cavity in rats following intraperitoneal



injection of glass microfibers  (Davis, 1976; Pott et al., 1974,



1976, 1980,  1984).  By the intrepleural route, glass wool also



caused low incidences of mesothelioma in a few studies (Stanton et



al., 1981, 1977; Wagner et al., 1984) while other studies have



produced no mesothelioma  (Wagner et al., 1973; 1976).  Stanton and



coworkers also demonstrated that glass fibers less than 0.25 urn

-------
                                74
diameter and greater than 8 urn length have carcinogenic potential



equal to that of asbestos fibers.  Similarly, other investigators



found that long/ thin gl-ass fibers are highly carcinogenic by the



injection routes of exposure but are generally less effective than



asbestos at equal mass doses.



     The relevance of the injection method with regard to human



exposure is considered questionable considering that it bypasses



normal physiological deposition and clearance mechanisms in the



respiratory tract.  Positive results from studies using



intrapleural or intraperitoneal injection/implantation method in



the absence of positive findings from inhalation experiments do



not indicate that these fibers will produce tumors in man upon



inhalation.  However, positive results from such injection studies



as found in the case of fine glass fibers and glass wool indicate



that they have the potential to induce tumors when introduced to



the target tissues in sufficient quantity.  Furthermore, the fact



that in two studies involving intratracheal instillation of small



doses of glass microfibers  (to mimic the inhalation exposure



condition) resulted in the  induction of tumors distal to the



administration site (lung tumors and mesothelioma) indicate that



fine fiberglass can reach the critical target tissues (lung and



pleural mesothelium) if a sufficient amount of fibers can



penetrate the upper respiratory airways.  Whether or not these



materials when inhaled will  indeed reach the target tissues in



sufficient quantity to cause tumors depends on the respirability



characteristics of the fibers, which are not operative in the

-------
                                75
injection study.  Thus, in the absence of positive findings from



available inhalation studies, the weight-of-evidence for the



carcinogenicity -of fine fibrous glass and glass wool in animal



studies is considered limited.



     In vivo animal data are supported by positive findings from a



few genotoxicity studies showing that fine glass fibers appear to



have similar genotoxic effects (clastogenicity and transformation)



as asbestos (Sincock et al., 1982; Oshimura et al., 1984;



Hesterberg and Barrett, 1984).



     There are no studies available that examined the



carcinogenicity of glass filaments in animals via inhalation.



Moreover, large diameter glass fibers did not produce mesothelioma



in rats via the intrapleural route (Stanton and Wrench, 1972;



Stanton et al., 1977; 1981).  Thus, the weight-of-evidence of



carcinogenicity for glass filament in animal studies is considered



inadequate.







II.1.4.2.  Fibrogenicity



     There does not appear to be any convincing evidence for



increased risks of non-malignant respiratory disease (NMRD) from



exopsure to fibrous glass.  There is also no definitive evidence



for the development of lung fibrosis in animals inhalation.



However, the positive findings from several injection studies in



animals and in vitro cytotoxicity studies indicate that fine



fibrous glass may be fibrogenic.



     Available epidemiological studies have provided weak or no



evidence of excess mortality  from NMRD in fibrous glass workers.

-------
                               76
In the large cohort study in the U.S. (Enterline et al., 1986;



1987; Enterline/ 1987)/ there was no significant excess mortality



from NMRD among glass wool workers compared with local rates,



although there was a statistically excess based on national



rates.  However, there was no relationship with time from first



exposure, or duration of exposure.  Among glass wool workers "ever



exposed" to small diameter fibers, there was no excess of NMRD



mortality but there was a slight nonsignificant increase with time



since first exposure.  In the continuous glass filament cohort, no



excess NMRD mortality was observed based on either local or



national rates.



     Other investigators have not observed an association of NMRD



and fibrous glass exposure.  In the large European study (Simonato



et al., 1985; 1986a; 1986b), there was no excess mortality from



NMRD in the glass wool or continuous filament cohort, nor was



there a relationship with time from first exposure or duration of



exposure.  Shannon et al. (1986) also did not find an excess risk



of NMRD in their study of Canadian glass wool workers.  A deficit



in risk for NMRD was reported by Engholm et al. (1986)  in their



study of Swedish workers exposed to MMMF (fibrous glass and



mineral wool).



     The results of a respiratory morbidity study (Weil et al.,



1983) showed some evidence of radiographic opacities in the  lung



of a limited number of fibrous glass workers.  However, this study



showed no evidence of pulmonary fibrosis, no increase in



respiratory symptoms, and no impaired lung function.

-------
                                77
     Overall/ there is inadequate experimental evidence of



fibrogenicity for fine fiberglass and glass wool via inhalation



exposure.  It has been shown in several studies that chronic



inhalation exposure to fine glass fibers or glass wool produced



only minimal interstitial dust cell reaction without fibrosis in



rats (Wagner et al., 1984; Smith et al., 1986; McConnell et al.,



1984; Mitchell et al., 1986; Le Bouffant et al., 1984), hamsters



(Smith et al. 1986) and monkeys (Mitchell et al., 1986).  One



study reported the development of focal fibrosis in baboons



exposed to fine glass fibers (Goldstein et al., 1983).  However,



the small number of animals and the lack of unexposed control



animals limit the conclusions that can be made from this study.



     In contrast, more extensive pulmonary fibrosis was induced in



animals by intratracheal instillation (Wright and Kuscher, 1977),



Pickrell et al., 1983; Smith et al., 1986) or intrapleural



injection (Davis, 1976).  Furthermore, intraperitoneal injection



of fine glass fibers generally resulted in marked peritoneal



fibrosis (Pott et al., 1974; Smith et al., 1986).  The results of



these injection studies also indicate that long, thin glass fibers



are more fibrogenic than short, thin fibers, while thick glass



fibers (>3 urn diameter) are relatively inert.  In general, glass



fibers produced less severe and less progressive pulmonary lesions



than those induced by asbestos via either inhalation or injection



at equal mass concentrations or doses.



     The in vivo findings are further supported by results from



several in vitro studies showing that fibrous glass is cytotoxic



to various cell types in culture.  The cytotoxicity of fibrous

-------
                                78
glass was found to be a function of fiber dimension, with longer



(>10 urn), thinner (<1 urn) fibers being most cytotoxic/ whereas



coarse fibrous glass with fairly large diameter fibers (>3 urn)



were less cytotoxic (Tilkes and Beck, 1980, 1983a, 1983b; Brown et



al., 1979a; Davies, 1980; Pickrell et al., 1983).








II.1.5.  Recommendations
     Although glass wool and fine fibrous glass have been tested



extensively, none of available inhalation studies are considered



to be adequate.  In addition, there are no data available



regarding the comparative dose-response effects with asbestos.



Thus, it would be useful to conduct additional long-term animal



inhalation or injection studies on fibrous glass.  Further



epidemiological studies are also necessary to clarify the



pathogenicity of fibrous glass in humans.

-------
                                79
II.2.  Mineral Wool



     There are two major types of mineral wool, i.e., rock wool



and slag wool.  Rock wool is made by melting natural igneous rocks



and then drawing, blowing, or centrifuging the melts into



fibers.  Slag wool is produced by a similar process from blast



furnace slag.  Mineral wools are primarily used for thermal



insulation but are also used for sound dampening and reinforcement



of other materials.  Slag and rock wools have nominal diameters



ranging from 6-9 um, but also contain a relatively high proportion



of respirable fibers (diameter _<3 um).  Thus, mineral wools are



likely to generate respirable airborne fibers during production



and processing (ICF, 1986).








II.2.1.  Fiber Deposition, Clearance and Retention



     There is limited information on the pulmonary deposition,



clearance, and retention of mineral wool fibers.  However, the



results of available animal studies suggest that mineral wool



deposition and clearance are dependent on both fiber length and



diameter.  The lung clearance of mineral wools appears to be rapid



soon after inhalation, presumably via the mucocillary system and



by phagocytosis.  Translocation of inhaled mineral wool fibers to



regional lymph nodes and abdominal organs appears to be limited.



During later periods, mineral wools are eliminated slowly,



presumably by dissolution.



     Hammad  (1984) studied the pulmonary deposition and clearance



of mineral wool  in the rat.  A group of 49 male rats were exposed



"nose only"  to aerosol of mineral wool  (fiber  type not specified)

-------
                                80
at 300 fibers/mL for 6 consecutive days.  Count median length and



median diameter of airborne fibers were 13 urn and 1.2 urn,



respectively.  Rats in groups of seven were sacrificed at 5, 30/



90, 180, and 270 days after the last day of exposure.  The



pulmonary deposition of fibers, as approximated by fiber retention



after 5 days of clearance, appeared to be dependent on both fiber



length and diameter.  Fibers of diameters less than 1.3 urn and



shorter than 50 urn were found much more frequently in the lung



than thicker and longer fibers.  The pulmonary clearance of



mineral wools was multiphasic.  During the 5-30 and 30-90 day



periods, mineral wool had a half-life of 38 days, whereas the



half-lives increased to 68 days and 175 days for the 90-180 and



180-270 day periods, respectively.  Approximately 3-7 percent of



mineral wool fibers deposited in the lung were retained after 270



days.  Most of the fibers which were retained in the lung had



fairly large diameters (1.0-1.3 urn) and were relatively short



«5 jam).



     Fiber retention has also been demonstrated in the rat lung



following long term inhalation exposure to mineral wools (Johnson



et al., 1984a; Le Bouffant et al., 1984).  In the lung, rock wool



fibers were found predominantly in the alveolar or interstitial



macrophages (Johnson et al., 1984a).  Fiber translocation from the



lung to the tracheobronchial glands and diaphragm has been



observed for rock wool as evidenced by the presence of small



amounts of fibers in these organs  (Le Bouffant et al., 1984).



Migration of slag wool fibers outside the lung to the abdominal



organs including the spleen, liver, and diaphragm have also been

-------
                               81
reported after instillation of fibers into the trachea of rats,



hamsters, and rabbits (Spurny et al., 1984).



     Mineral wool fibers appear to dissolve in the rat lung after



a long period of a time (1-2 years) following inhalation exposure



(Hammad et al., 1985; Johnson et al., 1984a; Wagner et al., 1984),



or via inoculation into the trachea (Morgan and Holmes, 1984b;



Spurny et al., 1984).  Apparent fiber dissolution in abdominal



tissues has also been observed after injection of mineral wool



fibers into the peritoneum of various rodent species (Pott et al.,



1984; Spurny et al., 1984).



     Slag and rock wools appear to have different solubility



properties.  Morgan and Holmes (1984b) showed that rock wool



appeared to dissolve relatively slowly in the lung and that



dissolution apparently occurred more rapidly at the end of fibers



than in the middle, where the diameter was essentially



unchanged.  Similar findings have been obtained by other



investigators reporting that etching of fibers in the lung was



minimal for rock wool (Johnson et al., 1984a; Wagner et al,



1984).  On the other hand, slag wool fibers in the lung (Spurny et



al., 1984) and in abdominal tissues (Pott et al., 1984; Spurny et



al., 1984) showed a considerable degree of corrosion.  Analyses of



the retained slag wool fibers by various spectroscopic methods



revealed a partial to complete loss of alkali metals and alkaline



earth ions (Spurny et al., 1984).



     Results from in vitro experiments which studied the



dissolution of mineral wools in simulated physiological fluid, are



in agreement with the in vivo observations on fiber durability.

-------
                               82
Slag wool fibers have been shown to undergo rapid and extensive



leaching in physiological fluid, while the leaching of rock wool



fibers was much less extensive (Forster, 1984; Klingholz and



Steinkopf, 1984).  Leineweber (1984) also demonstrated that in



physiological saline, mineral wool fibers dissolved at a slow



rate, but it is not clear whether the tested fibers in this study



were rock wool or slag wool.








II.2.2.  Effects on Experimental Animals



     A number of studies have been conducted to evaluate the



tumorigenic and fibrogenic effects of mineral wools in laboratory



animals.  The experimental protocols and findings of available



studies are summarized in Table 2 (pages 230-233).



II.2.2.1.  Oncogenicity



     The results of available long-term studies have not provided



evidence of pulmonary or mesothelial carcinogenicity in rats or



hamsters exposed to mineral wool fibers by inhalation (Wagner et



al., 1984; Le Bouffant et al., 1984; Smith et al., 1986).  On the



other hand, malignant mesothelioma of the pleura or peritoneum



have been produced in rats following either intrapleural (Wagner



et al., 1984) or intraperitoneal (Pott et al., 1984, 1987b)



injection of various types of mineral wool at varying yields.



However, significantly more neoplasms were found in animals



exposed to asbestos fibers by either inhalation or injection route



of exposure at equal mass concentrations or doses.

-------
                                83
II.2.2.1.1.  Inhalation studies



     In the study by Wagner et al. (1984), a group of 56 SPF



Fischer 344 male and female rats were exposed to a dust cloud of



rock wool (58 percent <1 urn diameter; 64 percent >10 um long) at a



mass concentration of 10 mg/m  (equivalent to approximately 227



fibers/mL) for 12 months.  Of the 48 exposed animals which were



allowed to live out their full lifespans, there were two cases of



lung tumors (one benign adenoma and one adenoma with some



malignant features).  Unexposed control animals had no tumors.



Taking into account the known occurrence of spontaneous species-



specific lung neoplasms in F344 rats, there were no significant



differences in tumor incidence between unexposed and exposed



rats.  In contrast, UICC chrysotile asbestos produced 12 cases of



lung neoplasms (11 adenocarcinomas, 1 adenoma with some malignant



features).



     In a limited inhalation study by Le Bouffant et al. (1984), a



group of male and female Wistar IOPS rats were exposed to rock



wool at concentration of 5 mg/m  of dust for 24 months.  The rock



wool fibers had fairly large diameters (only 22.7 percent <1



um).  Because the rock wool fiber dust also contained a large



proportion of nonfibrous particles and fragments that did not



conform with the definition of a fiber, the numerical rock wool



fiber concentration was very low (11 fibers/mL) in comparison with



that of Canadian chrysotile (9,978 fibers/mL) used as control.  No



tumors were found in the exposed males (0/24) or female (0/23)



rats.  On the other hand, nine cases of lung tumors were observed

-------
                                84
in the chrysotile exposed animals at 24 months.  Unexposed control



animals had no lung tumors.



     In a study reported by Smith et al. (1986) , -female Osborne-



Mendel rats and male Syrian hamsters were exposed "nose-only" to a



mineral wool dust cloud at a mass concentration of 12 mg/m  (200



fibers/mL) for 24 months.  The tested mineral wool fibers had



fairly large diameters with a median value of 2.7 urn.  No primary



lung tumors were found in the exposed hamsters (0/69) and rats



(0/55).  Tumors were not observed in any of the untreated control



rats (0/125) or hamsters (0/112) nor in rats exposed to clean air



(0/59), and only one bronchoalveolar tumor (1/58) was found in the



control hamsters group exposed to air.  No significant production



of neoplasms were found in positive control hamsters (0/58) or



rats (3/57; 1 mesothelioma, 2 bronchoalveolar tumors) exposed to



UICC crocidolite asbestos.  It was believed that the lack of a



significant tumorigenic response by crocidolite asbestos observed



in this study could have been due to the use of relatively short-



fibered material (95 percent <5  im long).
II. 2. 2. 1.2.  Intrapleural Injection Studies



     Wagner et al. (1984) injected 20 mg of either rock wool  (with



or without resin) or slag wool  (with or without resin) into the



pleural cavity of groups of 48  SPF Fischer 344 rats.



Approximately 70-80 percent of  rock wool and slag wool, either



with or without resin, were less than 5 urn long and less than



1 urn diameter.  Three cases of  mesotheliomas were produced in the



animal group treated with rock  wool with resin, and two cases by

-------
                                85
rock wool without resin.  The slag wool produced no mesotheliomas



while UICC chrysotile asbestos induced mesothelioma in six



animals.  It should be pointed out th-at the injected rock wool and



slag wool dusts contained more nonfibrous particles (i.e., aspect



ratio less than 3) than fibrous particles.








II.2.2.1.2.  Intraperitoneal Injection Studies



     Pott et al. (1984) reported very low tumor yields in female



Sprague-Dawley and Wistar rats (40-60 animals per group) following



intraperitoneal injection of mineral wools.  A single dose of 5 mg



of slag wool containing a high proportion of very thin fibers (90



percent <0.28 urn diameter) and long fibers (9 percent >10 urn



length) produced only a tumor rate (sarcoma or mesothelioma) of 5



percent, which was not statistically different from zero.  Rock



wool of fairly large diameter fibers (50 percent <1.9 urn diameter)



produced a 16 percent incidence of sarcoma/mesothelioma of the



peritoneum of rats treated with three doses of 25 mg of dust,



while no neoplasms were observed in rats after 15 months following



a single dose of 10 mg of thin rock fibers (50 percent <0.64 urn in



diameter).  Basalt wool (50 percent <0.52 urn diameter, 50 percent



<5.8 urn long) also did not produce neoplasms in female Wistar rats



after 15 months following a single i.p. dose of 5 mg.



     Recently, Pott et al. (1987b) reported that relatively thick



basalt wool produced a high incidence of abdominal tumors in rats



following repeated intraperitoneal injections.  In this study,



female Wistar rats received five weekly injections of 15 mg of



basalt wool (50 percent <1.8 urn diameter; 50 percent <20 yam

-------
                                86
long) suspended in saline.  The animals were kept for their entire


lifespan.  Peritoneal mesothelioma/sarcoma were found in 30 of 53


treated animals.  High tumor yields were also obtained with


UlCC/Canadian chrysotile asbestos at considerably lower doses,


11/36 at 0.05 mg, 21/34 at 0.25 rag, and 30/36 at 1.0 ing).  The


tumor incidence in negative saline controls was 1/102.





II.2.2.2.  Fibrogenicity


     The experimental data on the fibrogenic effect of mineral


wools are limited.  The majority of studies showed a lack of


fibrogenic response following chronic inhalation exposure.


However, the results of a limited inhalation study suggest that a


low fibrogenic potential may exist for mineral wools.


     In a limited study by Johnson and Wagner (1980), groups of


two SPF Fischer 344 rats were exposed to dust clouds of either


respirable rock wool (length >5 urn) or UICC chrysotile B asbestos,

          •j
at 10 mg/m  for 50 weeks.  The animals were sacrificed 4 months


following the inhalation period.  Rock wool and chrysotile


asbestos produced focal fibrosis.  However, the effect was more


marked following inhalation of chrysotile than after exposure to


rock wool.  Two unexposed rats had normal lungs.


     On the other hand, lung fibrosis was not observed in the more


extensive study subsequently reported by Wagner et al. (1984).


Rats which were exposed to rock wool dust clouds (10 mg/m3) for 12


months developed interstitial cellular reactions to the dusts


without fibrosis (grade 3) at 12 and 24 months upon sacrifice or


following spontaneous death.  In contrast, animals which were

-------
                                87
exposed to UICC chrysotile asbestos showed evidence of early


interstitial fibrosis (grade 4) similar to those seen in human


asbestosis.  U-nexposed control animals had normal lungs (grade 1).


     A lack of fibrogenic effects was also reported in the chronic


inhalation study by Smith et al. (1986).  Hamsters and rats did


not develop lung fibrosis following a 2-year "nose-only" exposure


to mineral wool fibers (12 mg/m ).  Similarly/ there was no


evidence of lung damage in rats exposed to Saint-Gobain rock wool

                       q
for 24 months at 5 mg/mj as reported by Le Bouffant et al. (1984).




II.2.3.  In Vitro Studies


II.2.3.1.  Genotoxicity


     There is no information available on the genotoxicity of


mineral wools.  Thus, the genotoxic potential of mineral wools


cannot be assessed at the present time.




II.2.3.2.  Cytotoxicity


     The results from two in vitro studies indicate that rock wool


and slag wool are cytotoxic to phagocytic and nonphagocytic


cells.  In general, mineral wools appear to be less cytotoxic than


crocidolite asbestos.


     Davies (1980) studied the cytotoxic effect of rock wool, slag


wool, and UICC crocidolite asbestos on mouse peritoneal


macrophages.  The macrophages were exposed to 160 ug/mL of the


respirable fraction of the test fibers  (fiber dimension


unspecified).  Cytotoxicity was measured by determining the


release of lysozomal enzyme beta-glucuronidase (BGD) and

-------
                               88
cytoplasraic enzyme lactic dehydrogenase (LDH).  Both slag wool and



rock wool caused a significant release of BCD (2.5- and 1.8-fold,



respectively) when added to the cell culture.  Rock wool also



induced a 1.7-fold increase in the release of LDH, while UICC



crocidolite asbestos produced a 3.2- and 2.9-fold increase in LDH



and BCD levels, respectively.  It was also reported that removal



of binder material from rock and slag wools had no effect on their



activity; however, no data were presented to support this



conclusion.



     Brown et al. (1979b) studied the effects of respirable



fractions of rock and slag wools (size distribution unspecified)



on Chinese hamster (V79-4) lung cells and human alveolar



epithelial type II lung tumor cells (A549).  Both slag and rock



wools, with or without resin (10-50 ug/mL) caused a dose-dependent



cytotoxic response toward V79-4 cells, but the uncoated samples



were slightly more cytotoxic than the equivalent resin coated



samples.  Uncoated slag and rock wools, when added to the A549



cell cultures at 200 ug/mL, produced a significant increase in the



formation of giant cells, i.e., clusters of 200 cells or more.



The effect of UICC crocidolite asbestos on A549 cells, however,



was much greater than that of mineral wools.



     Nadeau et al. (1983) reported in an abstract that mineral



wools were more or less unreactive toward rat pulmonary alveolar



macrophages and rat erythrocytes.  However, no other information



was provided to evaluate these findings.

-------
                                89
II.2.4.  Assessment of Health Effects



     Data from available epidemiological and experimental studies



indicate that mineral wools are potentially carcinogenic and



possibly fibrogenic.  Thus, there is a reasonable basis to



conclude that mineral wool fibers may present a health hazard to



exposed humans.  However, based on available experimental data,



mineral wools appear to be much less biologically active than



crocidolite asbestos in a few cytotoxicity studies and less



carcinogenic than chrysotile asbestos fibers in a few limited



inhalation studies, and therefore would pose a health hazard of



less magnitude than that of asbestos.








II.2.4.1.  Oncogenicity



     By using the weight-of-evidence criteria for carcinogenicity,



mineral wool may be classified as a probable human carcinogen



(Category Bl) on the basis of limited evidence of carcinogenicity



from epidemiological studies and limited evidence from animal



studies.



     The earlier epidemiological evidence relating mineral wool



exposure as reviewed by the National Research Council (NRC, 1984)



suggests an association with respiratory cancer.  More recent data



from two large cohort mortality studies (Enterline et al., 1986;



1987; Simonato et al., 1985; 1986a; 1986b) now indicate that



mineral wool workers are at increased risk of developing



respiratory cancer.  Overall, there are no large excesses of



deaths from respiratory cancer in any of available studies.



Evidence supporting an etiological relationship between respira-

-------
                                90
tory cancer and mineral wool exposure includes the consistently



elevated respiratory cancer risks seen in several mineral wool



plants from different countries, and the higher risks in workers



who have had 20 or more years elapse since first exposure.  In



addition/ in a nested-case control study in which confounding by



smoking was controlled/ there was a weak but significant trend



observed between mineral wool exposure and respiratory cancer.



However/ consistent dose-response relationships have not been



observed among available studies.  It should be pointed out that



the low levels of exposure in nearly all plants studied and the



potential exposure misclassification could have contributed to the



apparent lack of dose-response relationships. On the basis of



available findings, the weight-of-evidence for a causal



association between exposure to mineral wool and occurrence of



respiratory cancer is considered limited (Battelle/ 1988).



     The results of key epidemiological studies on mineral wools



have been reviewed in detail in a report by Battelle (1988).



Briefly/ the studies by Enterline et al. examined deaths due to



respiratory cancer (malignant neoplasms of the bronchus/ trachea



and lung) among male mineral workers from 6 plants in the United



States.  These workers were employed for at least 1 year from



1941-63.  In the early study (Enterline et al., 1983), the cohort



was followed from 1941-1977.  An update of the study included a 5



additional years of follow up from 1946 through 1982 (Enterline et



al./ 1986; 1987).  The mean fiber exposure level in the mineral



wool plants was 0.3 fibers/mL.  It was found that respiratory



cancer death was significantly elevated using both national and

-------
                                91
local mortality rates.  Analyses of data by duration of exposue,



cumulative exposure and average intensity of exposure showed no



dose-related trend.  However, a significant excess was observed



for 20 or more years from first exposure.  The investigators also



conducted a small case-referent study which controlled for



cigarette smoking.  It was found that there was a statistically



significant relationship between fiber exposure and respiratory



cancer for mineral wools even after smoking was considered



(Enterline et al., 1987).



     Simonato et al.  (1985; 1986a; 1986b) conducted an historical



cohort study on mineral wool workers from seven rock/slag wool



facilities in 4 European countries.  The study was followed from



the time that production started (1937-1943) through 1982.  The



study cohort consisted of men and women with at least one year



employment.  There was a nonsignificant elevated increased risk of



lung cancer mortality rates among rock/slag wool workers compared



to national and regional rates.  There was no relationship between



mortality from lung cancer and duration of exposure.  However/



significant excess of lung cancer death was found after more than



20 years follow up among workers first exposed during the early



technological phase,  i.e., before the introduction of oil binders



and presumably dustier conditions existed.  Exposure such as



smoking or other occupational substances are considered unlikely



to provide a sufficient explanation for the excess of lung cancer



risk.  However, possible effects of such exposure either alone or



in combination with fiber exposure cannot be excluded on the basis



of available information.

-------
                               92
     The experimental evidence for the oncogenicity of mineral



wool is considered to be limited.   In two long-term inhalation



studies in which rats were exposed to rock wool,  no statistically



significant increase in lung tumor incidence was  observed (Wagner



et al., 1984; Le Bouffant et al.f  1984).   In a third study in



which rats and hamsters were exposed chronically  to mineral wool



(fiber type not specified) containing fairly large diameter



fibers, no lung tumors were found  (Smith et al.,  1986).  Although



available studies have not provided evidence of oncogenicity via



inhalation exposure, it should be  noted that due  to a number of



experimental limitations, none of  the studies on  mineral wool are



considered adequately studied.  However,  rock wool was shown to



produce a low incidence of pleural mesothelioma in rats via



intrapleural injection (Wagner et  al., 1984) and  both rock wool



and basalt wool induced abdominal  tumors following intraperitoneal



injection (Pott et al., 1984, 19875).  Slag wool, on the other



hand, produced no pleural tumors in rats by intrapleural



inoculation (Wagner et al., 1984)  and only a statistically



nonsignificant increase in peritoneal tumors via intraperitoneal



injection (Pott et al., 1984).  In contrast, at equal mass



concentrations or doses chrysotile asbestos induced high



incidences of lung tumors in rats via inhalation (Le Bouffant et



al., 1984; Wagner et al., 1984) and pleural mesothelioma by



intrapleural injection (Wagner et al., 1984).



     There were no reports available that examined the



genotoxicity of mineral wools.

-------
                               93
II.2.4.2.  Fibrogenicity



     The epidemiological evidence of an association between



increased risks of nonmalignant respiratory diseases and mineral



wool exposure is considered inadequate.  However,  since there is



limited evidence from experimental studies indicating that mineral



wool may have some cytotoxic and fibrogenic effects, there remains



a concern for possible development of pulmonary fibrosis



associated with mineral wool exposure.



     No increased mortality from nonmalignant respiratory diseases



was found for the European rock wool workers (Simonato et al.,



1985, 1986a, 1986b).  However, in the U.S. study (Enterline et



al., 1986; 1987; Enterline, 1987), a statistically nonsignificant



excess of deaths from nonmalignant respiratory diseases was



observed among mineral wool workers based on local or national



rates, but the data did not establish a relationship with interval



from onset of employment nor was there a dose-related trend using



exposure indices including duration of employment, cumulative



level of exposure, and average intensity of exposure.



Furthermore, the results of a morbidity study (Weill et al., 1983)



showed no evidence for impaired lung functions or radiographic



lung abnormalities associated with exposure to mineral wools.



     The experimental evidence for the fibrogenic potential  is



limited.  Three inhalation studies showed that mineral wools did



not produce pulmonary fibrosis in rats or hamsters following



chronic inhalation exposure (Wagner et al., 1984; Le Bouffant et



al., 1984; Smith et al., 1986), whereas chrysotile asbestos  used



as the positive control induced extensive lung fibrosis  in the

-------
                               94
exposed rats (Wagner et al., 1984).   However,  the results of a



limited inhalation study involving two rats which showed a



production of focal fibrosis following chronic exposure to rock



wool, provided suggestive evidence of a fibrogenic potential



(Johnson and Wagner, 1980).  The fibrogenic concern is further



supported by the findings that respirable fractions of mineral



wools were cytotoxic to cells in culture (Brown et al., 1979b;



Davies, 1980).  However, mineral wools were less cytotoxic than



crocidolite asbestos.








II.2.5.  Recommendations
     Since the effects of mineral wools have not been adequately



tested in animals, additional long-term inhalation studies on



mineral wools are recommended.  Additional epidemiological studies



are also needed in order to fully assess the health hazard



potential of mineral wools in humans.







II.3.  Ceramic Fibers



     Ceramic aluminum silicate glasses are vitreous substances,



made by melting kaolin clay or a mixture of alumina and silica,



and then blowing the melt to form the fibers.  Alumina and



zirconia fibers are monocrystalline substances, composed mainly of



aluminum oxide and zirconium oxide, respectively.  Ceramic fibers



have high temperature resistance, and for that reason are often



referred to as refractory ceramic fibers.  Ceramic products are



primarily used in industrial settings as high temperature



insulation.  The desired range of fiber diameters for industrial

-------
                               95
applications is 2-3.5 urn/ but the diameters can range from less



than 1 urn to 12 jum.  Thus, ceramic fibers are likely to generate



respirable airborne fibers (NRC, 1984).








II.3.1.  Fiber Deposition/ Clearance, and Retention



     Very limited information is available on the lung deposition,



clearance and retention of ceramic fibers.  The results of



available studies indicate that only a small fraction of inhaled



ceramic aluminosilicate fibers was deposited in the rat lung.



Fibers deposited in the lung were predominantly short «10 jum) and



thin (<1 jum).  Ceramic fibers were found to be cleared slowly from



the rat lung.  Like other fibrous materials, ceramic fibers are



presumably cleared by macrophage uptake, transport to the



lymphatic system and possibly by slow dissolution and



fragmentation.



     The pulmonary deposition of ceramic fibers has been studied



in the rat.  Rowani and Hammad  (1984) exposed 19 albino male rats



(nose-only) to a ceramic aluminium silicate dust cloud with an



average concentration of 709 fibers/mL for 5 consecutive days.



The airborne fibers had a median length of 3.7 pn and a median



diameter of 0.53 pm.  The pulmonary deposition of fibers was



measured 5 days after the last day of exposure; this period was



allowed for clearance of approximately 95 percent of fibers



deposited in the ciliated airways.  Fiber deposition of the entire



lung was 6.7 percent.  Fibers that were deposited in all



lobes were predominantly thin (<1 urn) and short  (<10 urn).  Large



diameter and long fibers which  constituted only a small fraction

-------
                               96
of total lung deposition, were found to be higher in the lobes



with relatively short path length from the trachea to the terminal



bronchioles (left lung and right apical lobe).  However, the



overall fiber size distributions in the various lobes were not



significantly different.



     In another study, Hammad (1984) studied the clearance of



ceramic fibers from the rat lung.  In this study, a group of 49



male rats were exposed by "nose-only" to ceramic fiber (aluminum



silicate) dust clouds with a mean concentration of 300 fibers/mL



for five consecutive days.  The mean diameter and length of



airborne fibers were 0.7 and 9.0 urn, respectively.  Rats in groups



of seven were sacrificed 5, 30, 90, 180, and 270 days after the



last day of exposure.  The clearance of ceramic fibers from lung



tissues was determined by the percentage of initial fiber



retention at day 5 after exposure.  The results of this study



indicate that ceramic fibers are cleared slowly from the rat



lung.  The fiber clearance during the 5-30 and 30-90 day periods



had the same half-lives of 85 days.  The fiber clearance for the



90-180 and 180-270 day periods was much slower, with half-lives of



157 and 196 days, respectively.  After 270 days following



exposure, about 25 percent of the ceramic fibers were still



present in the lung tissue.



     There is very limited information on the clearance mechanisms



of ceramic fibers.  In the rat, alumina fibers were detected  in



the alveolar macrophages and in the mediastinal lymph nodes



following inhalation exposure.  These results suggest that  these



fibers may be transported from the  lung via macrophages  into  the

-------
                                97
lymphatic system (Pigott et al., 1981).  Furthermore, large



number of fibers were found in the sternal and mesenteric lymph



nodes following injection of alumina or zirconia fibers into the



abdominal cavity of rats.  These findings also indicate that



these fibers can migrate from the peritoneum into the lymph nodes



and can be subsequently removed by the lymphatic system (Pigott



and Ishmael, 1981;  Styles and Wilson, 1976).



     Ceramic fibers may also be cleared by dissolution and



fragmentation.  Hammad et al. (1985) reported that considerable



alterations in the elemental composition were found in ceramic



fibers (aluminum silicate) recovered from the rat lung after a



long period of 270 days following short-term exposure to the



fibers via inhalation.  Furthermore, Pigott et al. (1981)



observed that alumina fibers retained in the rat lung showed a



high degree of fragmentation.  These in vivo observations are



consistent with results obtained from an in vitro study which



showed considerable dissolution of ceramic fibers in



physiological saline solution.  However, ceramic fibers are



relatively more resistant to solvent attack in comparison with



other man-made mineral fibers including fibrous glass and mineral



wools (Leineweber,  1984).








II.3.2.  Effects on Experimental Animals



     The potential oncogenic and fibrogenic effects of ceramic



fibers have been evaluated in several animal studies via



inhalation and injection routes of exposure.  The experimental

-------
                                98
design and results of available studies are summarized in Table 3



(pages 234-239).








II.3.2.1.  Oncogenicity



     The oncogenicity of ceramic fibers in laboratory animals



appears to vary considerably for different fiber types.   In one



inhalation study (Davis et al., 1984), an increased incidence of



lung tumors was observed in rats after chronic exposure to



ceramic aluminum silicate glass.  In another inhalation study, no



lung tumors were produced in rats, but a single case of malignant



mesothelioma was found in hamsters (Smith et al., 1986).  Alumi-



num silicate fibers also produced mesothelioma in rats and ham-



sters by intrapleural or intraperitoneal injection (Davis et al.,



1984; Smith et al., 1986; Wagner et al., 1973; Pott et al.,



1987).  In contrast, refractory alumina oxide and zirconia oxide



fibers did not induce tumors in rats by either inhalation or



intrapleural implantation (Pigott et al., 1981; Stanton et al.,



1981).








II.3.2.1.1.  Inhalation studies



     Davis et al.  (1984) studied the effects of long-term inhala-



tion exposure to ceramic aluminum silicate glass in rats.  In



this study, a group of 48 SPF Wistar rats of the AF/HAN strain



were exposed to the test dust cloud at a mass concentration of



8.4 mg/m3 for 12 months.  Approximately 95 fibers/mL were longer



than 5 urn and less than 3 urn in diameter.  Animals were sacri-



ficed at 12, 18, and 36 months.  The survival of treated and

-------
                                99
control groups was similar.  Eight exposed animals developed



pulmonary neoplasms (1 adenoma, 3 carcinomas, 4 malignant histi-



ocytomas).  It should be noted that the pattern of tumor develop-



ment appeared to be different than that for asbestos exposure



since malignant histiocytomas have not generally been associated



with asbestos exposure.  One case of peritoneal mesothelioma was



also observed in the exposed group.  No pulmonary tumors of any



types were found in the 40 unexposed control animals.



     Smith et al. (1986) also studied the long-term health



effects of refractory ceramic fibers in rats and hamsters.  Male



Syrian hamsters and female Osborne-Mendel rats were exposed



"nose-only" to dust cloud of refractory ceramic fibers at a mass



concentration of 12 mg/m3  (equivalent to 200 fibers/mL) for 24



months.  Approximately 83 percent of the exposure aerosol was



greater than 10 ^um long and 86 percent less than 2 urn in diameter



(69 percent >10 pm long and <2 urn in diameter).  The ceramic



fiber type was not specified in the report but it was presumed to



be aluminum silicate fibers (Kaolin) similar to the type that was



tested by Davis et al. (1984).



     In contrast to the test results of the study by Davis et al.



(1984), no pulmonary neoplasms were observed in the exposed rats



(0/55) in this study.  None of the exposed hamsters developed



primary lung tumors (0/70) but malignant mesothelioma was found



in one hamster (1/70).  This finding was not statistically



significant although the possibility that the tumor was



associated with ceramic fiber exposure could not be  ruled out,



since positive tumor responses have been observed in rats and

-------
                               100
hamsters by the injection routes of exposure.  With the exception



of one bronchoalveolar tumor in a sham control hamster (1/58),



none of the other sham controls or unexposed control animals



developed pulmonary or pleural tumors.  Thus, under the experi-



mental conditions of this study, there was no evidence of carci-



nogenicity in rats exposed to refractory ceramic fibers and there



was only suggestive evidence of carcinogenicity in exposed ham-



sters.  It should be noted that in this study UICC crocidolite



asbestos only produced a low tumor incidence in rats (3/57; 1



mesothelioma, 2 bronchoalveolar tumors) and no tumor response in



hamsters.  It was suggested that the lack of significant tumori-



genic effects of crocidolite asbestos observed in this study



could be due to the use of relatively short fibers (95-97 percent



<5 jjm long).



     Pigott et al. (1981) investigated the effects of chronic



inhalation exposure to refractory alumina fibers in rats and



reported that they were not carcinogenic.  Groups of 50 albino



Wistar derived rats (25 of each sex) were exposed by inhalation



to dust cloud of refractory alumina fibers (Saffil fibers),



either as manufactured or in a thermally aged form at a mean



respirable dust concentration of 2.18 and 2.45 mg/m , respec-



tively, for 86 weeks, after which the animals were maintained to



85 percent mortality.  No pulmonary tumors were reported in



either animal groups exposed to manufactured Saffil fibers  (0/42)



or aged Saffil fibers (0/38).  It should be pointed out  that a



major limitation of this study was the low levels of respirable



dust clouds.  Furthermore, both types of Saffil fibers had  fairly

-------
                               101
large diameters with median value of around 3 urn.  The tumor



incidence found in positive control animals exposed to UICC



chrysotile A asbestos at 4.54 mg/m3 was comparable with those



reported in other studies.  Pulmonary neoplasms were found in 9



of 38 animals (5 adenomas/ 3 squamous-cell carcinomas,



1 adenocarcinoma).








II.3.2.1.2.  Intrapleural Injection/Implantation Studies



     Wagner et al. (1973) found that the carcinogenic potency of



ceramic aluminum silicate fibers, when administered via the



intrapleural route, was considerably less than that of SFA chry-



sotile asbestos.  Groups of 36 SPF Wistar rats (24 males and 12



females) were administered a single dose of 20 mg of ceramic or



chrysotile fibers via intrapleural inoculation.  The diameters of



the tested ceramic fibers were between 0.5 and 1.0 um (lengths



unspecified).  Pleural mesotheliomas were observed in 3 of 31



rats treated with ceramic fibers.  The survival time to the first



mesothelioma was 743 days.  In contrast, significantly higher



incidences of mesothelioma were detected in rats treated with SFA



chrysotile samples (21/32 with one sample and 23/36 in a second



sample) as early as 325 days after injection.  A concurrent



vehicle control was not used.



     Stanton et al. (1981) tested two ceramic glasses (presumably



alumina and zirconia ceramic fibers) for carcinogenicity after



intrapleural implantation (40 mg) into rats  (50 female Osborne-



Mendel rats/group).  These were large diameter fibers, with



microcrystalline aluminum oxide content >80 percent  (glass 21),

-------
                               102
and with microcrystalline zirconia oxide content greater than 90



percent (glass 22).  Both ceramic fiber types were considered



noncarcinogenic; 2/47 rats receiving glass 21 and 1/45 treated



with glass 22 developed pleural neoplasms, compared to 3/491



untreated controls and 17/615 negative controls receiving



noncarcinogenic pleural implants.  As concluded by the



investigators, the carcinogenicity of fibrous dust appeared to be



related to fiber size; the lack of tumorigenic responses seen



with the two ceramic glasses samples were probably due to the



fact that they were composed of large diameter fibers.








II.3.2.1.3.  Intraperitoneal Injection Studies



     Davis et al.  (1984) found that ceramic aluminum silicate



glass was carcinogenic in rats by the intraperitoneal route.  In



this study, a group of 32 SPF Wistar rats of the AF/HAN strain



(sex unspecified)  received a single intraperitoneal injection of



25 mg of ceramic fibers.  The injected fibers which were col-



lected from the inhalation exposure system were predominantly



short and thin (90 percent <3 urn long and <0.3 jurn in diameter).



Three animals developed peritoneal tumors (1 mesothelioma,  2



fibrosarcomas) approximately 850 days after injection.  Negative



control animals were not included in this study.



     Smith et al.  (1986) also reported that refractory ceramic



fibers were carcinogenic in male Syrian hamsters and  female



Osborne-Mendel rats when injected as a single dose of 25 mg into



the abdominal cavity of the animals.  Fibers used for injections



were collected airborne materials from the inhalation exposure

-------
                               103
chambers with mean diameter of 1.8 urn.   The incidences of peri-



toneal mesbthelioma in hamsters injected with refractory ceramic



fibers were 13 percent (2/15) in one group and 24 percent (5/21)



in a second group.  The incidence of abdominal tumors in treated



rats was 83 percent (19/23).  On the other hand, 40 percent



(8/25) of the hamsters and 80 percent (20/25) of the rats



injected with crocidolite asbestos had abdominal mesotheliomas at



their deaths.  Negative saline controls and unmanipulated control



animals had no tumors.



     A recent study by Pott et al. (1987b) also showed that



ceramic aluminum silicate fibers are highly carcinogenic in rats



via the intraperitoneal route.  Two ceramic fibrous samples were



tested in this study.   Suspensions of either Ceramic "Fiberfrax"



dust (50 percent <8.3  jam in length; 50 percent <0.91 urn in



diameter) or ceramic "MAN"  (50 percent <6.9 urn in length; 50



percent <1.1 urn in diameter) were injected into the abdominal



cavity of female Wistar rats at 5 weekly doses of 9 mg  (a total



dose of 45 mg) and 15  mg, (a total dose of 75 mg), respectively.



High incidences of abdominal tumors were observed in both animal



groups treated with ceramic Fiberfrax (33/47) and ceramic MAN



(11/54).  Under similar experimental conditions, UICC chrysotile



asbestos induced comparable incidences of peritoneal tumors in a



dose-related manner but at  considerably lower doses,  (11/36,



21/34, 30/36 at 6.05,  0.25, 1.0 mg, respectively).  The saline



control group had one tumor  (1/102).

-------
                               104
II.3.2.1.4.  Intratracheal Instillation Studies



     Smith et al. (1986) also tested the carcinogenicity of



refractory ceramic fibers in male Syrian hamsters and female



Osborne-Mendel rats via intratracheal instillation of 2 mg of the



test fiber (mean diameter of 1.8 jum) once a week for 5 weeks (a



total dose of 10 mg).  All animals were maintained for the dura-



tion of their lives.  Six of the 22 rats had bronchoalveolar



metaplasia but none of the rats (0/22) nor hamsters (0/25)



instilled intratracheally with refractory ceramic fibers developed



primary tumors.  In contrast, 74 percent (20/27) of the hamsters



and 8 percent (2/25) of the rats instilled intratracheally with



crocidolite asbestos developed bronchoalveolar tumors.








II.3.2.2.  Fibrogenicity



     The experimental data on the fibrogenicity of ceramic



aluminum silicate glass are limited but suggestive of a



fibrogenic potential.  This ceramic fiber type was shown in one



limited study to cause mild interstitial pulmonary fibrosis in



rats by inhalation exposure.  In contrast/ available evidence



indicates that large diameter alumina and zirconia fibers do not



appear to cause fibrosis in rats via  inhalation or by injection.








II.3.2.2.1.  Inhalation Studies



     In an inhalation study by Davis  et al. (1984), as discussed



under "Oncogenicity", ceramic aluminum silicate glass was found



to induce low levels of interstitial  pulmonary  fibrosis  in  rats.



Large areas of alveolar proteinosis and small amounts of  inter-

-------
                               105
stitial fibrosis were detected in the lungs of rats killed at the



end of the 12 months exposure period and 6 months later (at 18



months).  The fibrotic lesions were~more severe and extensive



(0.2 - 14.5 percent of total lung area) in older rats killed at



32 months, while only two of the control animals had very small



areas of interstitial fibrosis (<0.01 percent of total lung



tissue area).



     Smith et al. (1986)/ however, did not observe lung fibrosis



induction in rats or hamsters chronically inhaling (nose-only)



refractory ceramic fibers for 24 months.  In contrast, the number



of hamsters and rats in the crocidolite asbestos exposure group



with fibrous pulmonary lesions was statistically higher than



either that of chamber or unmanipulated control group.



     As discussed under "Oncogenicity", Pigott et al.  (1981)



reported that inhalation of Saffil or  "aged" Saffil fibers



(refractory alumina fibers) did not produce pulmonary  fibrosis in



the rat.  Pulmonary reactions to both  forms of alumina fibers



were confined to minimal alveolar epithelialization.  On the



other hand, moderate fibrosis was produced in most rats exposed



to UICC chrysotile asbestos.  These results are not surprising



considering that the tested ceramic fibers were relatively thick



and contained low respirable fractions.  Furthermore,  the



exposure concentrations of alumina fibers were low and almost



twofold less than that of chrysotile.

-------
                               106
II.3.2.2.2.  Intraperitoneal Injection Studies



     In a study by Pigott and Ishmael (1981), groups of 24 SPF



Wistar rats (12 of each-sex) were each injected intraperitoneally



with 20 mg of refractory alumina fibers, either of Saffil type A



(15.5 x 2.75 jjm) or type B (17 x 3.7 yum), or UICC chrysotile



asbestos (7.7 x 1.1 urn).   Rats which were treated with either



alumina fibers showed mild chronic inflammatory reactions and



deposition of small amounts of collagen in the abdominal tissues



after 3, 6, and 12 months following treatment.  There was no



evidence of progressive peritoneal fibrosis in these animals.  In



contrast, marked fibrosis was observed in rats given chrysotile



at 6 and 12 months, whereas animals which were injected with



vehicle control (saline)  showed normal findings.



     Styles and Wilson (1976) tested the fibrogenic potential of



Saffil alumina and Saffil zirconia fibers (median diameter of 3.6



and 2.5 urn, respectively) and found that under the experimental



conditions, these fibers were not fibrogenic in the rat.  Groups



of 40 Wistar rats (20 of each sex) were each injected intraperi-



toneally with a single 20 mg dose of either test materials or



UICC chrysotile asbestos in saline (mean diameter of 1.1 urn).



Control animals received saline only.  All animals were  killed 6



months after injection.  Considerable number of white nodules



containing fibroblasts and mononuclear cells and small amounts of



collagen were detected in the abdominal cavity of 36 rats which



were dosed with either Saffil zirconia or Saffil alumina



fibers.  However, there was slightly more collagen  in animals



treated with alumina fibers than  in those dosed with zirconia

-------
                               107
fibers.  In contrast/ animals which were treated with chrysotile



showed evidence of marked peritoneal fibrosis.







II.3.2.2.3.  Intrathoracic Injection Studies



     In a short-term study by Davis et al. (1970), guinea pigs



(number, sex and strain unspecified) were each administered by



intrathoracic injection with 25 mg of ceramic aluminum silicate



glass (mean diameter of 2 urn; 50 percent <75 jam in length).



After 6 weeks following treatment there was considerable



formation of ferrunginous bodies and large granulomas consisting



of macrophages, giant cells and fibroblasts,  in treated



animals.  These pulmonary reactions appeared  similar to those



induced by asbestos reported in other previous studies.



Induction of lung fibrosis by ceramic fibers  was not observed in



this study.  This finding is certainly not unexpected since the



observation period of this study was quite short.  In general,



fibrotic lesions are produced by most mineral fibers including



asbestos only after a minimum of 6 months following treatment.







II.3.3.  In Vitro Studies
II.3.3.1.  Genotoxicity



     There is no information available with regard to the



genotoxicity of ceramic fibers.







II.3.3.2.  Cytotoxicity



     Available in vitro data indicate that fibrous ceramic



aluminum silicate is not cytotoxic to macrophage-like cells

-------
                               108
 (P388D1)  and  lung  fibroblasts  (V79/4 cells) but is active toward



 the  human alveolar tumor cell  line  (A549).  The results of a



^single  study  showed that alumina and zirconia ceramic  fibers are



 slightly  toxic  to  macrophages.



     A  respirable  sample of ceramic aluminum silicate  fibers



 collected from  the inhalation  chambers by an elutriation process



 was  not found to affect the viability of P388D1 cells  following a



 48 h treatment  at  50 mg/mL of  the dust (Gormley et al., 1985).



 In contrast,  almost all asbestos samples tested showed a wide



 range of  cytotoxicity  in this  assay.



     In a subsequent study (Brown et al., 1986),  the same group



 of investigators tested the same ceramic fiber sample  in two



 nonphagocytic assays.  In the  assay using Chinese hamster lung



 fibroblast  (V79/4) cell line,  respirable fibrous  ceramic aluminum



 silicate  was  found to  be inactive;  the ED^Q, the  concentration of



 dust causing  a  50  percent reduction in the  cloning efficiency of



 V79/4 cells was greater than 100 mg dust/mL.  On  the other hand,



 the  same  fiber  sample  displayed  some activity  in  the assay using



 human alveolar  type II lung tumor cell line, as measured by the



 induction of  growth and giant  cell  formation at 25 and 50 mg



 dust/mL.  Ceramic  aluminum silicate fibers  ranked in the mid



 range in  this assay; it was slightly more active  than  some of  the



 amphibole asbestos tested but  much  less  active than several



 chrysotile  samples.



     Styles and Wilson (1976)  reported that Saffil alumina and



 Saffil  zirconia induced low cytotoxicity in rat peritoneal



 macrophages whereas UICC chrysotile asbestos was  highly

-------
                               109
cytotoxic.  It should be noted that both types of ceramic fibers



contained predominantly large fibers (2-6 urn in diameter) while



UICC chrysotile asbestos contained thin fibers (median diameter



of 1.1 urn).  The experimental details were not provided in this



report.








II.3.4.  Assessment of Health Effects



     Available experimental studies indicate that ceramic



aluminum silicate glass is carcinogenic and weakly fibrogenic in



animals whereas large diameter refractory aluminum oxide and



zirconia oxide do not appear to be tumorigenic nor fibrogenic.



Therefore, it is concluded that ceramic aluminum silicate fibers



may present a health hazard to exposed humans.  Because of the



variable results seen in available animal studies/ a comparison



of the carcinogenicty between ceramic aluminum silicate and



asbestos cannot yet be determined.  Based on available



information/ it would appear that refactory zirconia oxide and



alumina oxide fibers would not pose significant health hazard in



humans.  The discrepant results concerning the pathogenicity of



various types of ceramic fibers may be a function of variation  in



fiber size distribution.



     There are no epidemiological studies available on the health



effects from exposure to ceramic  fibers.  Available experimental



studies indicate that ceramic aluminum silicate glass is



carcinogenic in laboratory animals.  The results of a study by



Davis et al. (1984) showed that chronic  inhalation exposure to



ceramic aluminum silicate glass produced an increased incidence

-------
                               110
of lung tumors in rats.  In addition, injection of these fibers



into the pleural cavity or abdomen of rats or hamsters have



resulted in the production of mesothelioma of the pleura or



peritoneum, respectively (Wagner et al., 1973; Davis et al.f



1984; Smith et al., 1986; Pott et al., 1987b).  Thus, based on



sufficient evidence of carcinogenicity of ceramic aluminum



silicate glass in multiple experiments with different routes of



administration, but in the absence of human data, this ceramic



fiber type may be categorized as a probable human carcinogen



(Category B2).



     Experimental evidence of fibrogenicity of ceramic aluminum



silicate fibers is limited.  However, the results of one chronic



inhalation study by Davis et al. (1984) which showed low levels



of lung fibrosis in the exposed rats suggest a low fibrogenic



potential for ceramic aluminum silicate fibers.  This finding is



supported by in vitro test data showing that ceramic aluminum



silicate is active toward the human cell line A549 (Brown et al.,



1986).



     With regard to other ceramic fiber types, refractory alumina



oxide and zirconia fibers did not cause tumor or fibrosis in rats



via inhalation exposure  (Pigott et al., 1981) or intracavitary



injection (Stanton et al., 1981; Pigott and Ishamael, 1981;



Styles and Wilson, 1976).  Similarly, the cytotoxicity of these



fibers is low (Styles and Wilson, 1976).  On the basis of



available information, refractory alumina oxide and  refractory,



zirconia fibers are not classifiable as to human carcinogenicity



due to inadequate evidence of carcinogenicity  in animals and no

-------
                               Ill
human data (Category D).  A lack of pathogenic effects of these



fibers may be because test fibers were largely nonrespirable.







II.3.5.  Recommendations



     It is recommended that epidemiological studies of exposed



workers be initiated.  No additional animal tests are recommended



at this time.  A large-scale animal study sponsored by industry



is being conducted at a private laboratory.  The study is



designed to subject rats and hamsters to common types of ceramic



fibers for long periods by inhalation and injection methods.







Ill.   Naturally-Occurring Mineral Fibers



III.l.  Erionite



     Erionite is a naturally-occurring mineral of the fibrous



zeolite class.  Erionite and other natural zeolites are crystal-



line minerals that contain alkaline metal and alkaline earth



elements in a hydrated aluminium silicate structure.  In the



United States, fibrous erionite is found in several well-defined



deposits in Arizopa, Nevada, Oregon, and Utah, where  it occurs as



thin, pure beds with sedimentary tuft sequences or as outcrops in



the desert valleys of the intermountain region (Rom et al.,



1983).  Fibrous erionite has also been identified in  the volcanic



tuft located in the central region of Turkey  (Spurny, 1983b).



     Erionite can occur as either single needles or in



clusters.  Erionite fibers are, on the average, shorter than



asbestos fibers with a maximum length of approximately 50 urn.



Fiber diameter generally ranges from 0.25 urn-to 1.5 )im (Wright et

-------
                               112
al., 1983) although fibers with diameters of 0.01 to 5 urn have



been reported (Suzuki, 1982).   Airborne erionite fibers are



generally respirable.



     Erionite is rarely rained  and has very limited commercial



uses, mainly as a molecular sieve in ion-exchange processes.



Erionite was once, but is no longer, used as a catalyst in



petroleum cracking.  Instead,  synthetic nonfibrous zeolites are



used extensively for these types of applications (ICF, 1986).







III.1.1.  Fiber Deposition, Clearance and Retention



     Little is known about the lung deposition of erionite



fibers.  A number of epidemiological studies have reported the



detection of erionite  in the pleural and parenchymal tissues in



some Turkish villagers exposed to low ambient erionite



concentrations (Baris  et al.,  1978; Rohl et al., 1982; Sebastien



et al., 1981).  Most of the erionite fibers found in the lung



tissues were uncoated, with a mean diameter of 0.3 urn and maximum



length of approximately 9 urn.   Some ferrunginous bodies



containing erionite fibers (zeolite bodies) which were similar to



typical asbestos bodies were also detected  (Sebastien et al.,



1981).  Experimentally, it was reported in an abstract that



erionite fibers were well distributed in the rat lung soon after



intratracheal instillation, followed by the development of



macrophage and giant cell granulomatous reactions, and the



production of ferrunginous bodies after 1 week  (Moatamed et  al.,



1981).  These findings taken together indicate that airborne



erionite fibers can penetrate the lung and pleura and appear to

-------
                               113
elicit early lung tissue responses similar to those induced by



asbestos.  There is no information available on the clearance of



erionite.







III.1.2.  Effects on Experimental Animals



     Erionite has been evaluated in several studies for potential



carcinogenic and fibrogenic effects in animals by several routes



of exposure.  Table 4 (pages 240-242)  summarizes the experimental



protocols and results of available studies on erionite.







III.1.2.1.  Oncogenicity



     Erionite from different geographical sources has been shown



to be extremely carcinogenic in rats by inhalation/ and in both



rats and mice following injection.  Erionite appears to be more



potent in inducing mesothelioma than either crocidolite or



chrysotile asbestos.







III.1.2.1.1.  Inhalation Studies



     Wagner et al. (1985) tested samples of natural erionite and



synthetic nonfibrous erionite for carcinogenicity  in Fischer 344



rats via inhalation.  Groups of 20 male and 20 female rats were



exposed to mean respirable dust concentrations of  10 mg/m  for



7 hours/day/ 5 days/week for 12 months.  The tested dusts



included Oregon erionite (86 percent <0.4 urn in diameter;



92 percent <10 um long) at 354 fibers/mL, synthetic nonfibrous



Zeolite (>0.5 jjm) of similar chemical composition  to erionite at



1040 particles/mL, and UICC crocidolite asbestos (95 percent <0.4

-------
                               114
urn in diameter; 86 percent <10 urn long) at 1,630 fibers/mL.

Twelve rats in each group were sacrificed at 12 months to study

dust accumulation.  The remaining animals were observed until

death.  An extremely high incidence of pleural mesothelioma was

induced in rats (27/28) exposed to Oregon erionite.  The average

induction time was 580 days.  One mesothelioma (1/28) and one

adenocarcinoma (1/28) occurred in the rats exposed to the

synthetic nonfibrous zeolite.  It is interesting to note that no

mesotheliomas were produced in any of the positive control rats

exposed to crocidolite, and only one squamous carcinoma of the

lung was observed (1/28).  Unexposed control rats had no tumors

(0/28).

     Johnson et al. (1984b) examined the histopathology and

ultrastructure of seven pleural tumors that had been induced in

rats by inhalation to Oregon erionite.  The tumors were

epithelial, fibrosarcomatous, or mixed epithelial/sarcomatous in

form.  However, the majority of tumors were of mixed features

with either the fibrous or epithelial component being more

prominent.  In general, erionite-induced mesothelioma appeared to

be morphologically similar to human mesothelioma and to

experimentally induced mesothelioma in rats by inoculation of

asbestos into the pleural or peritoneal cavity.



III.1.2.1.2.  Intrapleural Injection Studies

     Three intrapleural studies also document  the oncogenicity of
                  /
erionite in rats.  In the study by Wagner et al.  (1985), groups

of 40 Fischer 344 rats (20 of each sex) were each  inoculated

-------
                               115
intrapleurally with 20 mg of Oregon erionite (75 percent



<6 jLim long; 92 percent <0.2 /jm in diameter), Turkish (Karain)



rock fiber (91 percent <0.2 ^im in diameter; 86 percent



<6 jum long), synthetic nonfibrous zeolite or chrysotile



asbestos.  The test dusts were suspended in saline.  A negative



control group received only saline.  All of the rats treated with



Oregon erionite developed mesothelioma (40/40 rats).  Karain rock



fiber induced 38 mesotheliomas (38/40) while only two



mesothelioma (2/40) occurred with the nonfibrous zeolite, which



had similar chemical composition as erionite.  Chrysotile



asbestos produced a total of 19 mesotheliomas and one



mesothelioma was found in the negative control group.  The mean



survival times in the Oregon erionite group (390 days) and Karain



rock fiber (435 days) were considerably shorter than that in the



chrysotile group (678 days).



     Maltoni et al. (1982a) also reported  induction of pleural



mesothelioma in rats by erionite by intrapleural injection.



Groups of 40 Sprague-Dawley rats (20 of each sex) received a



25 mg dose of either erionite or crocidolite asbestos in water.



The dimensions of tested fibers were not specified.  Among the 40



rats treated with erionite, 10 animals died within  53 weeks, 9



with pleural mesotheliomas.  No pleural tumors were found in the



crocidolite group after 53 weeks.  None of  the vehicle controls



developed tumors.  Followup data on the study have  not been



published.



     Preliminary results from intrapleural  studies  by Palekar  and



Coffin (1986) showed that erionite induced  pleural  mesothelioma

-------
                               116
in a dose-related manner in Fischer 344 rats (25 animals/dose



level).  A dose response (0.5-32 mg)  was obtained for two samples



of erionite tested, erionite I (mean length of 2.2 urn; mean width



of 0.25 urn) and erionite II (mean length of 1.4 urn; mean diameter



of 0.17 urn).  The tumor response of both erionite samples was



much greater than that of chrysotile and crocidolite when the



data were expressed as the mass or the number of fibers.








III.1.2.1.3.  Intraperitoneal Injection Studies



     The evidence of the carcinogenicity of erionite following



intraperitoneal injection is demonstrated by three studies in



mice.  Suzuki (1982) studied the carcinogenicity of erionite



after a single intraperitoneal injection in male Swiss albino



mice.  In the first experiment, a group of 12 mice were each



treated with 10 mg of erionite (20 percent <1 urn and 95 percent



<8 urn in length; 19 percent <0.1 urn and 94 percent <1 urn in



diameter).  Seven mice served as untreated controls.  Six treated



mice were sacrificed at 2-3 months after injection to determine



early pathological lesions.  One treated mouse died from



intestinal obstruction due to severe peritoneal fibrosis.  Among



the remaining five animals which died between 8 and 15 months



after treatment, two had malignant peritoneal tumors.  No tumors



were found in untreated control animals.  In a second experiment,



groups of five mice received either 10 or 30 mg of erionite.  A



positive control group that also consisted of 5 animals were



treated with 10 mg of chrysotile asbestos.  A negative control



group of 6 animals remained untreated.  Four of five mice  in  the

-------
                               117
low dose erionite group and 2/5 in the chrysotile group developed



malignant peritoneal tumors.  All mice in the high dose erionite



group died with intestinal obstructions due to adhesion of tfhe



intestine.  Untreated control mice had no tumors.



     Suzuki and Kohyama (1984) subsequently reported a high



incidence of peritoneal tumors, mainly malignant mesotheliomas,



in male Balb/c mice treated with a single intraperitoneal dose of



erionite.  Two samples of erionite were tested.  A group of 50



mice were administered 10 mg of erionite I (90 percent <8 jjm and



6 percent >9.5 jjm in length; 85 percent <1 pm and 8.7 percent



>1.4 um in diameter) while groups of 20, 50 and 75 mice received



a single dose of 0.5, 2, or 10 mg of erionite II (95 percent



<8 um and 4 percent >9.5 um in length; 82 percent <0.5 um and 100



percent <1 um in diameter), respectively.  In the animal group



treated with erionite I, 21 of 42 (50 percent) had malignant



peritoneal tumors.  Of the three groups treated with erionite II,



6 of 18 (33 percent), 24 of 44 (55 percent) and



3 of 8 (37.5 percent) had malignant tumors, respectively.



Animals treated with 2 mg of chrysotile had no tumors  (0/22)



while 6 of 32 (18 percent) animals treated with 20 mg of



chrysotile developed malignant peritoneal tumors.  The erionite-



induced mesotheliomas were similar to those induced by chrysotile



asbestos in gross appearance and histology.  Saline controls and



untreated controls had no tumors (0/118 and 0/37, respectively).



     Ozesmi et al. (1985) also tested dust from the village of



Karain (Turkey) containing both fibrous and nonfibrous erionite



for induction of tumors in Swiss albino mice using the

-------
                               118
intraperitoneal route.  Groups of Swiss albino mice (37-98/group)



received a 5, 10, 15, 20, 30 or 40 mg dose of Karain dust in



saline and were followed until death (up to 32 Inonths).



Mesothelioma developed in 41/321 of the dosed mice, malignant



lymphomas in 31 and both lymphomas and mesothelioma in 11



animals, within 9 to 32 months after injection of dust.  The



incidence of tumors did not appear to be dose-related.  Three



mesotheliomas and one lymphoma occurred in 55 saline controls.








III.1.2.2.  Fibrogenicity



     There is no information available on the ability of erionite



to induce fibrotic disease in animals by inhalation.  However,



erionite has been shown to cause fibrogenic effects in animals by



the injection method.








III.1.2.2.1.  Intrapleural Injection Studies



     Erionite has been reported to produce a fibrogenic reaction



when administered by intrapleural injection.  In the study by



Maltoni et al. (1982a), Sprague-Dawley rats were injected



intrapleurally with 25 mg of erionite (dimension unspecified).



Among the 40 treated animals, 10 died within 53 weeks, 9 with



pleural mesotheliomas.  Upon gross examination, the visceral,



parietal, and diaphragmatic pleura appeared thickened  and



whitish.  In addition, several hard, whitish or yellowish nodules



from 2-10 mm in diameter were found scattered at different sites



of the serosal surfaces.  Deposits of erionite surrounded by



granulomatous reaction were seen within the neoplastic tissue.

-------
                               119
III.1.2.2.2.  Intraperitoneal Injection Studies



     Erionite has been shown to possess fibrogenic properties



similar to asbestos following intraperitoneal injection.  Severe



fibrotic lesions were observed in the peritoneum of Swiss albino



mice treated with single intraperitoneal injections of either 10



or 30 mg of erionite (Suzuki, 1982).  Similar findings were



observed in a followup study by Suzuki and Kohyama (1984), which



reported that two different samples of erionite produced marked



peritoneal fibrosis in mice by intraperitoneal injection, with a



severity similar to that produced by chrysotile asbestos.  The



experimental details of these two studies are presented in the



oncogenicity section.








III.1.3.  In Vitro Studies
III.1.3.1.  Genotoxicity



     Available data indicate that erionite is genotoxic.  The



major genotoxic effects seen with erionite include DNA damage and



repair, induction of cell transformation, clastogenicity and



aneuploidy.  Like asbestos fibers, erionite does not appear to



cause detectable gene mutations.



     Fibrous erionite (Oregon origin) was tested in CSHIOT1/^ cells



for transformation and unscheduled DNA synthesis (UDS) and in a



human lung cell line (A549) for UDS  (Poole et al., 1983b).  The



count median length was 1.7 ^am and diameter was 0.2 urn.  Erionite



induced the appearance of transformed type III foci at



concentrations >10 ug/mL (up to 30 ug/mL).  These same



investigators report that while erionite is not more cytotoxic

-------
                               120
than asbestos, in another article they demonstrate that



crocidolite and amosite asbestos do not transform 101^ cells



(Poole et al., 1983a).



     Erionite was examined by two methods for UDS, the



autoradiographic method with 10T^ cells and the liquid



scintillation method with lOT^/2and A549 cells (Poole et al.,



1983b).  With the autoradiographic method, erionite induced a



significant increase in UDS over controls at 100, 150 and



200 ug/mL.  However, at higher concentrations (250 and 500



pg/mL), no increases were seen, suggesting a cytopathic effect



not measured in this study as no apparent cytotoxicity was



noted.  UDS was induced at all concentrations (50, 100, and 200



jug/mL) in both cell types with the scintillation method.  The



authors favored this method for use with fibers for two



reasons:  1) fibers can obscure the nucleus in the



autoradiographic method and therefore not allow a truly random



selection of nuclei for counting; and 2) the scintillation method



allows a larger number of cells to be assayed.  This may reduce



variability seen in autoradiographic UDS as cells do not receive



a homogenous exposure to similar fiber lengths and diameters.



Overall, erionite produced a significant level of UDS  in two



different cultured cell types indicating induced DNA damage and



repair.



     Palekar et al. (1987) reported that erionite fibers were at



least as effective as, if not more than, asbestos in producing



aneuploidy in exposed V79-4 cells.  A comparable or lesser



clastogenic effect than asbestos was also noted.  A significant

-------
                               121
reduction in diploid cells and a parallel increase over controls



in aneuploid and polyploid cells were observed in cultures



treated with erionite at all concentration levels ranging from



10-100 ug/mL, whereas in UICC crocidolite and UICC chrysotile-



treated cultures, significant increases in aneuploidy were



observed at all exposure levels except the low concentration,  10



ug/mL.  When the effects were compared on the basis of number of



fibers per dose, fewer erionite fibers than those of crocidolite



and chrysotile were required to produce similar aneuploidy.



Erionite treatment at 100 ug/mL also produced chromatid



aberrations but the clastogenic effect of erionite was comparable



to that of crocidolite, but weaker than that of chrysotile



asbestos.



     Further evidence that erionite fibers are clastogens and are



capable of altering the ploidy of cultured cells is provided by



Kelsey et al. (1986).  These investigators studied the



cytogenetic effects of Oregon erionite and crocidolite asbestos



in Chinese hamster ovary (CHO) fibroblasts.  Treatment with



erionite at concentrations of 5-50 ug/mL induced a slight but



significant elevation in sister chromatid exchanges (SCE) in



cultures of synchronous CHO cells, while crocidolite at the same



concentrations failed to significantly increase the frequency of



SCE.  However, both fiber types induced a low level of



chromosomal aberrations in CHO cells.  In addition, an increase



in the relative percent of tetraploid CHO cells after treatment



with either erionite or crocidolite was observed.

-------
                               122
     Kelsey et al. (1986) also tested the mutagenicity of



erionite and crocidolite fibers in a human lymphoblastoid cell



line (TK6) at either the HGPRT (hypoxanthine guanine



phosphoribosyl transferase) or thymidine kinase loci.  Both



erionite and crocidolite were negative in these assays.







III.1.3.2.  Cytotoxicity



     Experimental evidence has demonstrated that erionite is



hemolytic and cytotoxic to various cell types.  In an abstract,



Nadeau et al. (1983) reported that erionite caused hemolysis to



rat erythrocytes.  The hemolytic activity of erionite was lower



than that of chrysotile but higher than that of amphibole



asbestos (crocidolite/ amosite, anthophyllite).  It was also



reported that erionite was cytotoxic to rat pulmonary alveolar



macrophages in a dose-related manner.  However, the experimental



details including fiber dimensions and dose levels were not



reported.



     Palekar et al. (1985) studied the cytotoxicity of erionite



and asbestos fibers in Chinese hamster ovary  (CHO) and Chinese



hamster lung V79-4 cell cultures at concentrations of 10-100



jug/mL during a 6-day exposure.  In CHO cells, erionite and



chrysotile asbestos were cytotoxic at 20 ug/mL and crocidolite



asbestos was cytotoxic at 40 ug/mL.  Similarly, V79 cytotoxicity



was induced by erionite and chrysotile at 40 jLig/mL while



crocidolite was cytotoxic at 100 ^ag/mL.



     Subsequent findings by Palekar and Coffin (1987) confirmed



their previous results that erionite was cytotoxic to V79

-------
                               123
cells.  Erionite samples containing long erionite fibers (median



length of 1.6 ^im) exhibited similar cytotoxicity to UICC



chrysotile asbestos by weight while shorter erionite fibers



(median length of 0.99 urn) were less cytotoxic than chrysotile.



However, both erionite samples were more cytotoxic to V79-4 cells



than chrysotile when the cytotoxicity was expressed as a function



of number of fibers.  Both erionite samples were more cytotoxic



to V79-4 cells than UICC crocidolite asbestos by either methods



of data analysis.








III.1.4.  Assessment of Health Effects
     Erionite has been shown to be a potent carcinogen in animals



and is potentially fibrogenic.  Thus, there is sufficient



evidence to conclude that erionite potentially poses a



significant health hazard to the exposed humans.  Based on



experimental data, erionite appears to be at least as hazardous



as asbestos.







III.1.4.1.  Oncogenicity



     Erionite may be categorized as a probable human carcinogen



(Category Bl) based on limited evidence of carcinogenicity from



studies in humans and sufficient evidence from experimental



studies.



     Available epidemiological studies have shown that



populations from several locations in South Central Turkey have a



large excess incidence of malignant mesothelioma  (Baris et al.,



1978, 1981; Artvinli and Baris, 1979, 1982; Rohl  et al., 1982).

-------
                               124
There is limited evidence to suggest that erionite may be the



major etiological factor.  This is based on the findings that



erionite was the major fibrous material present in the



surroundings and in the air of these affected areas,  and in the



pleural and parenchymal tissues of individuals with pleural



disease(s).  However, since asbestos and other zeolite fibers



were also found in environmental and tissues samples taken in  one



of the affected villages (Rohl et al., 1982; Baris et al., 1978;



Boman et al., 1982), it is possible that asbestos and other



fibrous agents could also be involved in the etiology of this



malignant mesothelioma (Battelle, 1988).



     Experimental studies have confirmed that erionite from



Turkey and the United States is extremely carcinogenic in animals



by several routes of exposure.  In an inhalation study, a 96



percent incidence of malignant mesothelioma of the pleura was



produced in rats following 1-year exposure to Oregon erionite



(Wagner et al., 1985).  Intrapleural inoculation of Oregon or



Karain erionite also produced very high incidences of pleural



mesothelioma (53-100 percent) in rats (Maltoni et al., 1982a;



Wagner et al., 1985; Palekar and Coffin, 1986).  In these



studies, erionite caused more mesothelioma than either



crocidolite or chrysotile asbestos by inhalation or intrapleural



inoculation.  Furthermore, the latency periods for mesothelioma



induced by erionite were much shorter than those induced  by



asbestos.  In mice, intraperitoneal  injection of erionite



resulted in the production of malignant mesotheliomas of  the



peritoneum at high yields, comparable to those induced by

-------
                               125
asbestos (Suzuki, 1982;  Suzuki and Kohyama,  1984;  Ozesmi  et  al.,



1985).



     Positive findings from a few genotoxicity studies further



support a carcinogenic concern.  Erionite has been shown  to  cause



DNA damage and repair (Poole et al., 1983b), cytogenetic  changes



including aneuploidy, chromosomal aberrations, sister chromatid



exchanges (Kelsey et al., 1986; Palekar et al., 1987) and



morphologic transformation of cells in culture (Poole et  al.,



1983b).








III.1.4.2.  Fibrogenicity



     In view of limited evidence from epidemiological studies and



limited evidence from experimental studies,  erionite is



considered to be potentially fibrogenic.



     Epidemiological evidence collected over the past several



years from a limited geographical area in South Central Turkey



where erionite was present indicated significant incidences of



nonmalignant pleural diseases as well as malignant pleural



mesothelioma.  These nonmalignant chest diseases included



calcified plaques, chronic pleural fibrosis, and pleural



thickening (Baris et al., 1978, 1981; Artvinli and Baris, 1979,



1982).  The evidence for erionite as the major etiological factor



is considered to be limited because possible exposure to asbestos



and other fibrous material cannot be excluded  (Battelle, 1988).



     No information is available on the ability of erionite to



induce fibrotic diseases in animals by inhalation.  However,



erionite has been shown to cause fibrogenic effects  in animals by

-------
                               126
injection.  Whitish or yellowish nodules and plaques associated



with the visceral, parietal and diaphragmatic pleura,  and pleural



thickening were observed in rats given an intrapleural dose of



erionite (Maltoni et al., 1982a).  In mice, intraperitoneal


injection of erionite produced severe peritoneal fibrosis which



was intimately associated with the observed peritoneal tumors


(Suzuki, 1982; Suzuki and Kohyama, 1984).  The in vivo results



are further supported by positive findings from in vitro studies



showing that erionite is hemolytic and highly cytotoxic (Nadeau



et al., 1983; Palekar et al., 1985; Palekar and Coffin, 1987).






III.1.5.  Recommendations
     Since erionite has been adequately tested in animals, no
   v,


further animal testings are thought necessary.  However,



additional epidemiological studies should be conducted, if



practical, to further evaluate the association between erionite



environmental exposure and development of malignant and



nonmalignant respiratory diseases.






III.2.  Wollastonite
     Wollastonite is an acicular or fibrous natural monocalcium


silicate mineral.  The largest deposits of Wollastonite are


located in the United States, Mexico and Finland.  Wollastonite


is widely used in ceramics, and as a substitute  for asbestos in


insulation, wallboard, and brake linings.  Wollastonite exhibits


similar heat resistance properties as asbestos but has lower


tensile strength.  Wollastonite fibers range from 1 to 10 urn in

-------
                               127
diameter with an average diameter of 3.5 urn.   The median fiber



size of airborne wollastonite is 0.22 urn in diameter and 2.5  urn



in length.  Approximately" 92-97 percent of total airborne fibers



during mining and milling operations are considered respirable



(ICF, 1986).








III.2.1.  Fiber Deposition, Clearance and Retention



     No information is available on the deposition, clearance,



and retention of wollastonite.








III.2.2.  Effects on Experimental Animals



     Very few studies have been conducted to examine the



oncogenic and fibrogenic effects of wollastonite in animals.



Table 5 (page 243) summarizes the experimental protocols and test



findings of these studies.








III.2.2.1.  Oncogenicity



     There is no information available regarding the oncogenicity



of wollastonite in animals via inhalation exposure.  However,



wollastonite has been shown to be weakly tumorigenic by intra-



pleural implantation and nontumorigenic by the intraperitoneal



route in rats.








III.2.2.1.1.  Inhalation Studies



     Recently, a chronic inhalation study in male  Fischer 344



rats was conducted by the National Toxicology Program  (NTP)  to



test for the oncogenicity of wollastonite.  The experiment has

-------
                               128
been completed but full results are not yet available.  However,



the authors reported that inhalation exposure to wollastonite



produced no adverse effects on the animal survival (Adkins and



McConnell, 1985) and no tumorigenic response (McConnell, 1988).








III.2.2.1.2.  Intrapleural Implantation Studies



     Stanton et al. (1981) showed that wollastonite was weakly



carcinogenic in rats following intrapleural implantation with a



40 mg dose of particles.  Four samples of wollastonite from a



Canadian mine were tested.  These fibers were relatively large



and only one of these samples was completely fibrous.  The



incidences of pleural sarcoma observed at 2 years following



treatment were:  grade 1, 5/20; grade 2, 2/25; grade 3, 3/21;



grade 4, 0/24.  The tumor incidence in groups receiving grades 1



and 3 was statistically significantly higher (p <0.05, Fisher



exact test) than that of control animals implanted with



noncarcinogenic materials (17/615).








III.2.2.1.3.  Intraperitoneal Injection Studies



     Pott et al. (1987b) recently reported that wollastonite was



not tumorigenic in rats following intraperitoneal injection.  In



this study, female Wistar rats were injected intraperitoneally



with 5 weekly 20 mg doses of wollastonite in saline.  The test



dust was obtained from India, with 50 percent of the fibers



having diameters less than 1.1 pm and lengths less than 5.2 ^am.



The animals were observed for full life span.  No peritoneal



tumors (0/54) were found.

-------
                               129
III.2.2.2.  Fibrogenicity



     There are no data available for evaluating the fibrogenic



potential of wollastonite in laboratory animals.  As mentioned



above, a 2-year inhalation study of wollastonite has been



completed to study the development of chronic respiratory disease



in rats (Adkins and McConnell/ 1985).  According to a preliminary



report, no evidence of pulmonary fibrosis was found in this study



(McConnell, 1988) but data are not yet available for a full



evaluation of the study.  In the long-term intraperitoneal



injection study by Pott et al. (1987b), wollastonite from India



was found to cause a low degree of adhesions of abdominal organs



as observed macroscopically in rats.  However,  it was not clear



whether there were any developments of fibrosis in treated



rats.  Histological data are not yet available  for a complete



evaluation of this preliminary finding.








III.2.3.  In Vitro Studies



III.2.3.1.  Genotoxicity



     There are no genotoxicity data available on wollastonite.








III.2.3.2.  Cytotoxicity



     Several in vitro studies have been conducted to compare the



biological effects of wollastonite with asbestos with regard to



the hemolytic activity to erythrocytes and cytotoxicity to



macrophages.  A number of investigators have shown that wollasto-



nite was weakly to moderately hemolytic to human and rat



erythrocytes while others have reported that wollastonite was

-------
                               130
more or less unreactive to rat erythrocytes.  Similarly,



wollastonite was found to induce varying degrees of cytotoxicity



in rat or rabbit alveolar macrophages, ranging from noncytotoxic



to moderately cytotoxic.  However, wollastonite was far less



hemolytic and cytotoxic than asbestos.  The conflicting in vitro



results with wollastonite appeared to be related to different



experimental conditions and the nature of the materials tested,



especially particle morphology and size distribution.








III.2.3.2.1.  Erythrocytes



     Skaug and Gylseth (1983) tested the hemolytic activity of



two samples of naturally-occurring wollastonite and three samples



of synthetic nonfibrous calcium silicate in human red blood



cells.  Both wollastonite samples {one fibrous and the other



mostly nonfibrous) were found to be weakly hemolytic while



synthetic compounds were far more reactive.  The particle size



distribution of the tested fibers and dusts were not specified.



     Hefner and Gehring (1975) studied the hemolytic activity of



two wollastonite samples using rat erythrocytes.  One sample was



fibrous with a mean fiber length of 200 urn and the other sample



was nonfibrous with a mean particle size of 4 pm.  Both wollasto-



nite samples were hemolytic but the large nonrespirable 200 pm



wollastonite (fiber diameter not specified) had a much slower



rate of hemolysis.



     Similar results were obtained in a study by Potts et al.



(1978) who studied the ability of two wollastonite samples  to



induce hemolysis in rat red blood cells.  Large particle

-------
                               131
wollastonite (200 pm) induced weak hemolytic activity whereas



small particle wollastonite (6.73 jam) was moderately hemolytic.



In contrast, chrysotile asbestos was found to be strongly



hemolytic in this in vitro system.



     In an abstract, Vallyathan et al. (1984) also reported that



wollastonite is moderately hemolytic.  Wollastonite fibers tested



had lengths less than 10 jam.  The hemolytic effects of chrysotile



«21 jjm), amosite (41 jam), and crocidolite «10 jum) were more



pronounced than that of wollastonite.  The source of red blood



cells used in this assay was not specified.



     In constrast, Nadeau et al.  (1983) reported in an abstract



that wollastonite was more or less unreactive in hemolytic assays



using rat erythrocytes, whereas crocidolite asbestos was highly



hemolytic followed by anthophyllite, amosite, and crocidolite.



No other experimental details were provided in this report for



evaluation.








III.2.3.2.2.  Macrophages



     Pailes et al. (1984) found that wollastonite had no



cytotoxic activity in rabbit alveolar macrophages whereas



chrysotile was strongly cytotoxic.  Exposure of alveolar



macrophages to wollastonite (fiber size distribution not



specified) as much as 250 ug/mL did not induce lysosomal enzyme



release (beta-glucuronidase, beta-galactosidase, acid



phosphatase, and N[-acetylglucosaminidase)  or alter membrane



integrity as measured by trypan blue exclusion and the  release  of



the cytosolic enzyme, lactate dehydrogenase  (LDH).  On  the other

-------
                               132
hand, treatment of alveolar macrophages with as little as 25



|jg/mL of chrysotile asbestos caused the release of lysosomal



enzymes and decreased membrane integrity.



     Similarly, Nadeau et al. (1983) reported in an abstract that



wollastonite (fiber size distribution unspecified) was unreactive



to rat pulmonary alveolar macrophages whereas dose-response



relationships for cytotoxicity were observed with all asbestos



samples.  Cytotoxicity was evaluated by the release of



cytoplasmic LDH and lysosomal alpha-galactosidase enzymes.



     Warheit et al. (1984) also found that exposure of rat



pulmonary macrophages to wollastonite did not affect cell



viability or morphology but wollastonite exposure did result in a



diminished phagocytic capacity of the cells.  On the other hand,



crocidolite asbestos affected both macrophage morphology and



phagocytic activity without affecting macrophage viability.



Wollastonite fibers were large and long compared to crocidolite



fibers which were short and thin.



     Vallyathan et al. (1984) reported in an abstract that the



cytotoxicity of wollastonite appeared to be dependent on fiber



length.  Wollastonite containing fibers less than 10 urn long was



moderately cytotoxic in macrophage enzyme release studies  (LDH,



beta-glucuronidase, beta-H-acetylglucosaminidase) whereas  shorter



fibers  (<5 urn) were only mildly cytotoxic.  The  cytotoxic  effects



of long-fibered chrysotile  (<21 pm), crocodolite  (<10 /am)  and



amosite «41 ^jm) in alveolar macrophages were more pronounced



than those of both wollastonite samples.

-------
                               133
III.2.4.  Assessment of Health Effects
     Overall, there is some evidence to support a possible health
hazard from exposure to wollastonite.  However, results from
available experimental studies indicate that wollastonite is much
less biologically active than asbestos, suggesting that
wollastonite may pose a lesser health hazard than asbestos.


III.2.4.1.  Oncogenicity
     Wollastonite may be classified as a possible human
carcinogen (Category C) on the basis of limited experimental
evidence of carcinogenicity and inadequate human data.
     None of available epidemiological studies were designed to
assess the risk of lung cancer or mesothelioma associated with
wollastonite exposure.  One case of retroperitoneal malignant
mesothelioma has been reported in Finland in one worker who had
been exposed to wollastonite for twenty years  (Huusk.onnen et al.,
1983).  However/ no cause and effect relationship can be drawn
based on a single case report.  Preliminary information on an
inhalation study of wallastonite in rats indicates the lack of
tumorigenic response (McConnell, 1988).  However, wollastonite
has been shown in one study to produce weak tumorigenic response
in rats via intrapleural implantation  (Stanton et al., 1981) but
does not induce tumors in rats via the intraperitoneal route
(Pott et al., 1987b).


III.2.4.2.  Fibrogenicity
     Available data are inadequate to assess the fibrogenic
potential of wollastonite.  There were no reports available that

-------
                               134
examined the fibrogenicity of wollastonite in animals.   Limited



epidemiological studies conducted to date on quarry workers in



the U.S. (Shasby et al., 1977, 1979; Hanke et al., 1984) and



Finland (Huuskonen et al. 1983, 1984) indicate a possible



association between wollastonite exposure and some nonmalignant



diseases such as impaired ventilatory capacity, mild fibrosis of



the lung, pleural thickening, and chronic bronchitis.  However,



these studies do not provide conclusive evidence of nonmalignant



respiratory disease following exposure since the sample size was



small and exposure was relatively short (Battelle, 1988).



     Nevertheless, available epidemiological findings do raise a



concern for potential fibrogenicity of wollastonite, particularly



in light of positive results from in vitro cytotoxicity assays,



which are thought to be indicative of fibrogenic activity.



Wollastonite has been shown to induce varying degrees of



hemolytic and cytotoxic activity although it is far less active



than asbestos (Skaug and Gylseth, 1983; Hefner and Gehring, 1975;



Potts et al., 1978; Vallyathan et al., 1984).  These in vitro



findings suggest that wollastonite may be considerably  less



fibrogenic than asbestos.







III.2.5.  Recommendations
     In order to fully assess the health effect of wollastonite,



additional epidemiological studies are needed.  In addition, the



results of the NTP inhalation bioassay should be evaluated.

-------
                               135
III.3.  Attapulgite



     Attapulgite is a naturally-occurring sorptive and gelling



clay made up of fibrous aluminum and magnesium silicate.



Although attapulgite is mined commercially in several countries,



the United States is a leading producer of attapulgite, the



majority of which is mined in the areas of Quincy, FL and



Attapulgus, GA.  Attapulgite is used in a wide variety of



applications as an absorbent and thickening agent, and to a



lesser extent as a substitute for asbestos in friction products



and other materials (NRC, 1984).



     Attapulgite morphology can vary greatly depending on where



the material is mined.  The attapulgite in commercial use in the



United States consists of short fibers (0.1-2.5 jam) with mean



diameter of 0.07 jum (0.02 - 0.1 pm) (Zumwalde, 1977).  French



attapulgite fibers are also short «1.2 jum) (Bignon et al., 1980)



while attapulgite samples from Spain can either be long or short



(Wagner, et al., 1987).  All attapulgite fiber types are of



respirable size.








III.3.1.  Fiber Deposition, Translocation, and Clearance



     There is very little information available on the



deposition, translocation, and clearance of attapulgite.  The



results of two limited case studies suggest that  attapulgite



fibers are capable of penetrating into the alveolar spaces



following inhalation, and that after ingestion, attapulgite can



be transported to the kidney and excreted  in the  urine.  There is



also some experimental evidence suggesting that short  attapulgite

-------
                               136
fibers are readily cleared from the lung while longer attapulgite



fibers appear to be retained longer in the lung.



     Bignon et al. (1980) reported the findings of two case



studies.  In the first case, a high concentration of attapulgite



fibers (42,000 fibers/mL) were found in lung washing fluid



recovered by bronchoalveolar lavage from a 41-year-old patient



with lung fibrosis, who had been exposed to attapulgite for 3



years during mining and processing of attapulgite.  Mean length



and diameter of the attapulgite fibers in lung washing fluids



were 1.5 urn and 0.11 urn, respectively.  In a second case, a very



high concentration of attapulgite was found in the urine of a 60-



year-old woman treated orally for 6 months with an attapulgite-



containing drug at a fairly large dosage (6-9 g/day).



     Wagner et al. (1987) examined the recovered dusts from the



lungs of rats following 12 months exposure to two attapulgite



dusts at 10 mg/m .  Examination of both macerated lung tissue and



ashed lung sections from animals exposed to short-fibered



attapulgite (all fibers <2 pm long) from Lebrija  (Spain) showed a



complete absence of short fibers.  On the other hand,  in animals



exposed to long-fibered attapulgite (palygorskite from Leicester,



U.K.), fibers up to 25 jam in length and with diameters less than



0.2 jum were found in the macerated lung and ashed sections.








III.3.2.  Effects on Experimental Animals



     There is considerable  information available  on  the  effects



of attapulgite in laboratory animals.  The experimental  protocols



and results of available studies are  summarized  in Table  6  (pages

-------
                               137
244-245).  Attapulgite fibers from various geographical locations



appear to have different pathogenic potential which may be



related to differences in fiber size distribution.








III.3.2.1.  Oncogenicity



     A number of studies have been conducted to evaluate the



oncogenic potential of attapulgite by different routes of



administration.  In a long-term inhalation study, attapulgite



from Spain (Lebrija) which consisted of short fibers (all <2 ^im)



did not induce tumors in rats.  Several injection studies also



showed that short attapulgite fibers (<2 pm) from the United



States (U.S.), Spain (Lebrija), and France did not cause



mesothelioma in rats by the intrapleural or intraperitoneal



routes of administration.  Tumors were also not observed in mice



following lifetime feeding with short-fibered attapulgite.  In



contrast, inhalation exposure to attapulgite (also known as



palygorskite) from the United Kingdom (Leicester) which contained



a considerable number of long-fibers (>6 urn) resulted in low



incidences of lung tumors and mesothelioma in rats.  In addition,



long-fibered attapulgite samples from the U.K. (Leicester), Spain



(Torrejon) and an unknown source produced high incidences of



mesothelioma in rats via intrapleural or intraperitoneal



injection.








III.3.2.1.1  Inhalation Studies



     In a recent report, Wagner et al. (1987) provided  the



completed findings of a series of  inhalation experiments on

-------
                               138
attapulgite and asbestos.  Two samples of attapulgite were
tested/ Lebrija attapulgite from Southern Spain, and palygorskite
(synonymous to attapulgite) from the U.K. (Leicester).  All
fibers in the Lebrija attapulgite sample were less than 2 yum in
length whereas the palygorskite sample from Leicester also
consisted of long, thin fibers (18 percent >6 urn in length and
                                           •_•»  ^
<2.0 urn in diameter).  In this study, groups of 40 SPF Fischer
rats (20 of each sex) were exposed to a dust cloud of either
attapulgite sample or UICC crocidolite asbestos at 10 mg/rrr for 6
hours a day, five days a week for up to 12 months.  Four animals
(2 of each sex) were sacrificed after 3, 6 and 12 months of
exposure to assess the severity of pulmonary fibrosis.  The
remaining 28 animals were allowed to live out their normal life
span.
     No significant tumor response was found in the animal group
exposed to short-fibered attapulgite from Lebrija.  Among the 40
exposed animals, there was only one peritoneal mesothelioma and 3
bronchoalveolar hyperplasia (BAH).  It should be noted that BAH
is considered to be a reaction to an irritant and not a tumor.
In contrast, there was some evidence of carcinogenicity for the
long-fiber palygorskite sample from Leicester in rats.  Three
mesotheliomas (two pleural and one peritoneal), one malignant
alveolar tumor  (MAT, accepted as an early carcinoma), two benign
alveolar tumors (BAT) and 8 BAH  (one BAH with MAT) were found
among the 40 animals exposed to  the palygorskite sample.  A
comparison of the carcinogenic potency between  this  attapulgite
sample and asbestos could not be made based on  the results of

-------
                               139
this study alone because of the lack of significant tumorigenic



response in the positive control group exposed to UICC



crocidolite asbestos (1/40 adenocarcinoma/ 2/40 BAH, 1/40 BAH



with adenocarcinoma).  No tumors were found in the unexposed



controls (0/40) and the negative controls exposed to nonfibrous



kaolin dust (0/40).








III.3.2.1.2.  Intrapleural Injection Studies



     Stanton et al. (1981) tested two samples of American



attapulgite (Attapulgus, GA) for carcinogenicity in female



Osborne-Mendel rats using an intrapleural implantation method.



Both samples were composed of small-diameter short fibers.  No



excess of tumors was found in animals treated with 40 mg of



either sample (2/29 for both samples) compared to untreated



animals (3/491) and negative controls treated with noncarcino-



genic implants (17/615).



     Renier et al. (1987) also found no tumorigenic response with



short, thin attapulgite from France in an oncogenic intrapleural



bioassay.  In this study, 20 mg of attapulgite fibers suspended



in saline were injected into the pleural cavity of Sprague-Dawley



rats  (sex and number of animals not specified).  The test fibers



had a mean diameter of 0.06 urn and a mean length of 0.77 jum.



After 24 months of treatment, no pleural tumors were found.



Positive control animals treated with UICC chrysotile or Canadian



chrysotile asbestos had a 19 percent and 48 percent mesothelioma



incidence, respectively.  Vehicle control animals injected with



saline had no tumors.

-------
                               140
     Wagner et al. (1982, 1987) conducted a series of injection



studies with three samples of attapulgite and showed that two



attapulgite samples which consisted of variable proportions of



long fibers were highly carcinogenic by the intrapleural route



whereas no excess tumors were produced by short attapulgite



fibers.  Forty SPF Fischer 344 rats, 20 of each sex, were



inoculated with a single injection of one of the following dusts



suspended in saline (dose unspecified) including Lebrija (Spain)



attapulgite (<2 jam long), Torrejon (Spain) attapulgite (0.54



percent _>L 6 jum long and <0.5 jam in diameter), palygorskite (18



percent _>. 6 jam in length and <0.2 jam in diameter) from Leicester



(U.K.), UICC crocidolite asbestos and chrysotile B asbestos.



High incidences of pleural mesothelioma were found among animals



treated with the palygorskite sample (30/32) and Torrejon



attapulgite (14/40).  These long-fibered attapulgite samples



appeared to have comparable carcinogenic potency as UICC



crocidolite and chrysotile B asbestos which induced 34/40 and



19/40 cases of mesothelioma, respectively.  In the group treated



with short-fibered attapulgite from Lebrija there were two



mesotheliomas (2/40), one pleural and one peritoneal.  The saline



control^group had one pleural mesothelioma  (1/40).








III.3.2.1.3.  Intraperitoneal Injection Studies



     Pott et al.  (1974) found that attapulgite from an unknown



source which contained a large proportion of long fibers (30



percent >5 um long) was tumorigenic when injected



intraperitoneally in rats.  In this study, a group of 40 Wistar

-------
                               141
rats received three doses of 25 mg of attapulgite dusts at weekly



intervals.  A tumor incidence of 65 percent (peritoneal



mesothelioma) was found for the attapulgite group, and the first



tumor appeared at day 275.  In animals treated with chrysotile



asbestos/ 30-67 percent had mesotheliomas.  No peritoneal tumors



were reported for saline control animals.



     More recently Pott et al. (1985) reported that short, thin



attapulgite fibers from France, Spain, and the United States



induced no excess tumors in rats after intraperitoneal



injection.  However, no experimental details and results were



available for evaluation.








III.3.2.1.4.  Oral Studies
     Brune and Deutsch-Wenzel (1983) reported that attapulgite



was not tumorigenic in mice following lifespan feeding.  In this



study, groups of 60 male and 60 female NMRI mice were fed for 25



months with 1 percent or 3 percent of attapulgite (mean length



<1 pm) admixed in a pelleted diet.  Untreated animals served as



controls.  The animals were sacrificed at the end of the



treatment period.  Mortality rates were not influenced by the



treatment of attapulgite.  No toxic effects nor increase of



tumors in any organs were observed.








III.3.2.2.  Fibrogenicity



     The results of a recent inhalation study indicates that



long-fibered palygorskite (attapulgite) is fibrogenic in rats



following long-term exposure.  Short-fibered attapulgite, on the

-------
                               142
other hand/ appears to be nonfibrogenic under the same



experimental conditions.



     In the study by Wagner et al. (1987), 40 SPF Fischer 344



rats (20 of each sex) were exposed to either palygorskite fibrous



dust from Leicester (U.K.) which contained a considerable



proportion of long, thin fibers (18 percent _>6 urn in length and



<0.2 urn in diameter) or short attapulgite fibers (all <2 urn long)



from Lebrija (Spain) at 10 mg/m^ for 6 hours daily, five days a



week for up to 12 months.  A group of 40 positive control animals



were exposed to UICC crocidolite asbestos at similar experimental



conditions.  Four animals (two of each sex) were sacrificed after



3, 6, 12 and 24 months to assess the severity of pulmonary



fibrosis.  The remaining animals were allowed to live out their



normal life span and the oncogenic response was then evaluated



upon sacrifice.



     The mean fibrosis gradings evaluated from four animals



killed at 3, 6 and 12 months after exposure to the long-fibered



palygorskite were 3.0, 3.1, 4.0, respectively.  Since a grading



of 4.0 represents first signs of fibrosis, these results indicate



that this palygorskite sample should be considered to be



potentially fibrogenic in humans.  It should be noted that a



conclusive evaluation of the extent and progression of pulmonary



fibrosis induced by this attapulgite sample in comparison to



asbestos could not yet be made because (1) the gradings of



pulmonary responses were not done for the palygorskite sample at



24 months; and (2) the low fibrogenic response in crocidolite



asbestos exposed animals.  In the majority of inhalation studies

-------
                               143
with asbestos, extensive lung fibrosis with gradings greater than



4.0 is generally seen in exposed animals.  In this study,



however, the mean fibrosis gradings at 3, 6, 12 and 24 months for



crocidolite exposed animals were only 4.1, 3.3, 3.1 and 3.8,



respectively.



     In contrast, the pulmonary response to short-fibered



attapulgite from Spain was confined to the presence of dust-laden



macrophages (grade 2) and early interstitial reaction (grade



3).  The mean gradings of tissue response in the lungs of rats



exposed to Lebrija attapulgite at 3, 6, 12, and 24 months were



reported to be 3.1, 2.6, 3.2, and 3.2 respectively.  Unexposed



animals had normal lungs throughout the study period (gradings of



1.25-1.75).








III.3.3.  In Vitro Studies



III.3.3.1.  Genotoxicity



     Little information is available for the genotoxicity of



attapulgite.  In one study, short attapulgite fibers were not



found to induce unscheduled DNA synthesis (UDS) in primary



cultures of rat hepatocytes.



     Denizeau et al. (1985) tested the ability of short, thin



attapulgite to induce DNA damage in primary rat hepatocyte



cultures by measuring its capacity to induce UDS.  Ninety-six



percent of the attapulgite fiber had a diameter between 0.01 and



0.1 pm with an average length of 0.8 urn.  The liquid



scintillation UDS method was used.  Attapulgite produced no UDS



effect over controls at 10 ug/mL.  The fiber did not induce

-------
                               144
cytotoxicity, as assessed by the release of lactate



dehydrogenase.








III.3.3.2.  Cytotoxicity



     In general/ attapulgite from various sources was hemolytic



to red blood cells and cytotoxic to macrophages; these in vitro



effects were somewhat comparable to those induced by asbestos.



Long-fibered attapulgite was found to induce cytotoxicity in



nonmacrophage cells, whereas short attapulgite fibers were



relatively inert.  However, short attapulgite fibers were found



to cause a nonsignificant increase in squamous metaplasia of



hamster tracheal organ cultures.








III.3.3.2.1.  Erythrocytes



     A number of studies have shown that attapulgite was



hemolytic to red blood cells from humans and animals.  Jaurand



and Bignon (1979) reported that palygorskite (attapulgite) was



highly hemolytic.  However, the information on the source and



fiber size distribution of the attapulgite tested was not



provided.  Bignon et al. (1980) subsequently reported that long-



fibered Spanish attapulgite (unspecified fiber length



distribution) was more hemolytic to human red blood cells than



UICC chrysotile asbestos.  Three drugs containing short French



attapulgite  (mean length of 0.8 pn) were weakly hemolytic to



human red blood cells.  Other investigators found that



attapulgite was as hemolytic as chrysotile asbestos to rat

-------
                               145
erythrocytes (Nadeau et al., 1983) and sheep erythrocytes (Harvey



et al.,  1984).








III.3.3.2.2.  Phagocytic cells



     In an abstract, Nadeau et al. (1983) reported that



attapulgite induced cytotoxicity in a dose-related manner to rat



pulmonary alveolar macrophages similar to that induced by



asbestos fibers.  Cytotoxicity was evaluated by the release of



cytoplasmic enzyme lactate dehydrogenase (LDH) and lysosomal



enzyme alpha-galactosidase.  Fiber dimensions and dose levels



were not reported.



     Similar findings were reported by Bignon et al. (1980) who



studied the cytotoxicity of two drugs containing short



attapulgite fibers (mean length of 0.6 urn) in rabbit alveolar



macrophages.  Both drugs produced a 39-45 percent of LDH and 24-



31 percent of beta-galactosidase release at 300 or 200 ug/mL.



     Chamberlain et al. (1982) found that both short-fibered



attapulgite and long-fibered attapulgite were cytotoxic to rat



peritoneal macrophages.  Treatment of cell culture with short-



fibered attapulgite at 150 ug/mL caused a 58 percent of LDH



release/ while after treatment with long-fibered attapulgite



there was a 29 percent of LDH release.  Specific fiber dimensions



were not provided in this study.



     By using P388D1 macrophage-like cells/ Harvey et al.  (1984)



found that attapulgite treatment at a relatively high



concentration (1 mg/mL) for 4 hours caused considerable



cytoxicity.  However, attapulgite was slightly less cytotoxic

-------
                               146
than UICC chrysotile A or Canadian chrysotile asbestos but was



markedly more cytotoxic than crocidolite asbestos.  This study



did not provide the fiber size distribution of the attapulgite



sample.



     In contrast, Lipkin (1985) found no evidence of cytotoxicity



of short-fibered French or American attapulgite in P388D1



macrophage-like cells.  The maximum fiber length of both



attapulgite samples was 1.2 and 1.6 urn, respectively.  UICC



amosite asbestos showed a dose-dependent cytotoxic effect on the



macrophage system while both attapulgite samples had no effect at



10, 50 or 100 ug/mL, as measured by reduction in cell number over



a 72-hour period.








III.3.3.2.3.  Nonphagocytic cells



     Chamberlain et al. (1982) reported that short fibered



attapulgite at concentrations greater than 100 pg/mL induced no



effect on colony formation of Chinese hamsters of V79-4 cells nor



on the ability of human type II alveolar tumor (A549) cells to



form giant cells, i.e., colonies containing more than 200



cells.  Long-fibered attapulgite, on the other hand, reduced



cloning efficiency of V79-4 cells by 50 percent at 52 ug/mL.



UICC crocidolite asbestos was more potent than either attapulgite



samples in inducing cytotoxic effects in both cell types.



     Using 1-407 human embryo cells, Reiss et al.  (1980) found



that short-fibered attapulgite from the United States



(Attapulgus/ GA) caused only minimal inhibition of colony

-------
                               147
formation.  At equal doses, amosite asbestos was considerably



more cytotoxic than attapulgite.








III.3.3.2.3.4.  Tracheal Organ Cultures



     Woodworth et al. (1983) examined the ability of short



attapulgite fibers to induce metaplastic changes in trachea



mucosa of the Syrian hamster.  Ninty four percent of attapulgite



fibers were shorter than 1 pm long.  The investigators reported



that attapulgite treated explants underwent proliferative and



metaplastic alterations.  However, metaplastic changes were not



statistically significant at 1, 4 or 16 mg/mL of attapulgite.  In



contrast, both long and short fibers of chrysotile asbestos



induced a significant increase in metaplasia at low concen-



trations (1.0, 4.0 mg/mL).








III.3.4.  Assessment of Health Effects
     The toxicological properties of attapulgite may depend on



fiber length.  There is inadequate evidence of carcinogenicity



and fibrogenicity of short attapulgite fibers «2 urn long) in



humans and animals.  Based on an apparent lack of significant



effects from long-term animal studies via inhalation and by the



intrapleural or intraperitoneal route, it would appear that



short-fibered attapulgite from commercial deposits in the United



States is less hazardous than asbestos.  In contrast, there is



sufficient experimental evidence of carcinogenicity and



fibrogenicity for attapulgite samples containing long fibers (>5



urn in length).  Available data bases, however, are not sufficient

-------
                               148
to provide a definitive assessment with regard to the comparative



pathogenicity between long-fibered attapulgite and asbestos.



Fortunately, one of -the samples (palygorskite from the U.K.)  is



of no commercial interest and the other sample is a Spanish



product (Torrejon attapulgite) and is being used in the



preparation of drilling mud in the exploration of oil deposits in



the North Sea and Persian Gulf.








III.3.4.1.  Oncogenicity



     With regard to human carcinogenicity, short-fibered



attapulgite «2 pm in length) from the United States, France, and



Spain, is not classifiable (Category D) because of inadequate



evidence from epidemiological studies and insufficient evidence



from animal studies.  On the other hand, attapulgite containing



long fibers (>5 urn in length) from Spain, and the U.K. may be



categorized as a probable human carcinogen (Category B2) based on



sufficient evidence of carcinogenicity in animal studies in the



absence of human data.



     The results of a single available cohort study (Waxweiler et



al.f 1985) provide inadequate evidence of human carcinogenicity



of short-fibered attapulgite.  This study examined the mortality



trends among workers at one Georgia attapulgite operation.  Lung



cancer mortality in the total cohort was slightly elevated but a



statistically significant excess of mortality due to lung cancer



was observed among white workers while a deficit of risk was



found in the nonwhite subgroup.  The excess risk among white



employees could have been related to exposure to attapulgite

-------
                               149
because there was an increased risk among employees with presumed



highest exposure levels, and an increased risk among those with



longer duration of employment and time since first exposure.



However, the relatively small size of the cohort and several



other limitations such as the inability to confirm the



completeness of the cohort, limit the conclusions that can be



made about this study (Battelle, 1988).



     Experimental studies indicate that short-fibered attapulgite



fibers (<2 jam long) did not induce tumors in rats following



chronic inhalation (Wagner et al., 1987) or by the intrapleural



route (Stanton et al., 1981; Renier et al., 1987; Wagner et al.,



1987), or intraperitoneal route (Pott et al., 1985).  Short-



fibered attapulgite also did not induce tumors in mice following



lifetime feeding (Brune and Deutsch-Wenzel, 1983).



     The negative results from a genotoxicity study provide



supporting evidence for a lack of tumorigenic effect of short-



fibered attapulgite.  These fibers did not induce DNA damage in



primary rat hepatocytes as reflected by a lack of an induction of



unscheduled DNA synthesis (Denizeau et al., 1985).



     In contrast, materials containing long attapulgite (>5 jum in



length) have tested positive in long-term animal studies via



various routes of exposure.  Palygorskite from the U.K.



(Leicester) was shown to induce low incidences of lung tumors and



mesothelioma in rats following chronic inhalation and very high



incidences of pleural mesothelioma by intrapleural injection



(Wagner et al., 1987).  Other attapulgite samples containing long



fibers such as the one from a small deposit in Torrejon, Spain,

-------
                               150
and an unknown source were also found to induce pleural



mesothelioma by intrapleural injection (Wagner et al., 1987), and



abdominal tumors by intraperitoneal injection (Pott et al.,



1974), respectively.  In all of these studies, the tumorigenic



responses of long-fibered attapulgite were comparable to those



induced by chrysotile and crocidolite asbestos.








III.3.4.2.  Fibrogenicity



     In view of inadequate evidence of fibrogenicity in humans



and laboratory animals, there is an insufficient basis to support



a health hazard concern for potential fibrogenic effects of



short-fibered attapulgite.  However, positive findings of several



in vitro cytotoxicity studies on these fibers suggest a



possibility of a fibrogenic hazard.  As for the long-fibered



attapulgite, there is sufficient evidence to conclude that



prolonged exposure to the dust may cause the development of lung



fibrosis in humans.



     The results of three available studies reporting the effects



of attapulgite exposure provide inadequate evidence of fibro-



genicity of short-fibered attapulgite in humans (Battelle,



1988).  One case of lung fibrosis in a worker who had been



exposed to attapulgite for two years was reported, indicating a



possible link between occupational exposure and fibrosis (Sors et



al., 1979).  However, the results of a morbidity study showed no



consistent relationship between attapulgite exposure and



respiratory symptoms (Gamble et al., 1985).  Furthermore,



Waxweiler et al. (1985) reported a deficit mortality risk for

-------
                               151
nonmalignant respiratory diseases among attapulgite workers in a



Georgia plant.



     Experimentally, there are no studies available that examined



the ability of short-fibered attapulgite in commercial use in the



United States in inducing lung fibrosis in animals via



inhalation.  However, the results of a long-term inhalation study



showed that attapulgite from Lebrija (Spain) which consisted of



only short fibers did not induce fibrosis in rats (Wagner et al.



1987).  Furthermore, none of the available injection studies with



short-fibered attapulgite have reported the production of



fibrotic lesions in rats via the intrapleural route (Stanton et



al., 1981; Renier et al., 1987; Wagner et al., 1987) or



intraperitoneal route (Pott et al., 1985).  These results taken



together suggest that short-fibered attapulgite is not likely to



induce severe fibrogenic effects in humans.  On the other hand,



the fact that these fibers are hemolytic (Bignon et al., 1980;



Jaurand and Bignon, 1979; Harvey et al., 1984; Nadeau et al.,



1983) and cytotoxic to macrophages (Chamberlain et al., 1982;



Bignon et al., 1980; Nadeau et al., 1983) suggests that a



fibrogenic potential may exist.



     With regard to long-fibered attapulgite, the results from a



recent inhalation study (Wagner, et al., 1987) showed that the



palygorskite from the U.K. (Leicester) that contained a



considerable number of long particles caused lung fibrosis in



rats following 12 months of exposure, comparable to that induced



by crocidolite asbestos.  These findings indicate that long-



fibered attapulgite is potentially fibrogenic.

-------
                               152
III.3.5.  Recommendations



     A long-term inhalation study in animals should be conducted


to fully assess the chronic toxicity and oncogenic effects of



attapulgite from a United States commercial deposit.  Additional


epidemiological studies should also be performed to further


determine the health effects of attapulgite in humans.






IV.   Synthetic Fibers


IV.1.  Aramid Fibers


     These synthetic fibers are formed from aromatic



polyamides.  Aramid fibers are characterized by high tensile


strength, and chemical and flame resistance.  They are used as


replacements for asbestos in a number of applications such as



insulation, flame barriers, thermal protective clothing and


friction products.



     There are two major types of aramid fibers produced in the


United States, Kevlar® and Nomex®.  Para-aramid Kevlar® is
                  4

produced as continuous filament yarn, staple fiber (38-100 mm),


short fiber (6-12 mm) or pulp (2-4 mm), with a nominal diameter


of 12 urn.  Thus, Kevlar® fibers generally tend to fall outside of


the respirable range.  However, Kevlar® pulp, which is frequently


used to replace asbestos, has fine curled or tangled fibrils «1


urn in diameter) attached to the surface of the core fiber, and


these fibrils may break off and potentially become airborne upon


abrasion during the manufacturing processes.  Nomex® is


manufactured as continuous filament, staple fiber and short


fiber, with a nominal diameter of 12 urn.  However, unlike

-------
                               153
Kevlar®, Noraex® does not have the tendency to form fine fibrils



and therefore is not respirable (ICF, 1986).








IV.1.1.  Fiber Deposition and Clearance



     Available information on the pulmonary deposition and



clearance of aramid fibers is very limited.  Results of



intratracheal instillation and inhalation studies indicate that



the deposition of aramid fibers is dependent on fiber dimension



and is dose-related.  Short aramid fibers are mostly phagocytized



by alveolar macrophages which are cleared from the lung via



transport to the lymph nodes.



     Reinhart (1980) reported that following intratracheal



instillation (25 mg of Kevlar® polymer dust), large para-aramid



fibers  (100-150 urn in diameter) remained in the terminal



bronchioles whereas small particulates (approximately 5 urn)



penetrated deep in the aleveolar region of the rat lung.



Following inhalation exposure to 0.1-18 mg/m3 of ultrafine



Kevlar® fibrils for 2 weeks, fiber deposition and macrophage



response in the rat lung were found to be dose-related (Lee et



al., 1983).  At high concentrations, fiber dust and fiber-laden



macrophages (dust cells) accumulated in the respiratory



bronchioles and alveolar region immediately following exposure.



Some dust cells containing short fibers «2 urn long) were found



in the peribronchial lymphoid tissue or tracheobronchial lymph



nodes by 6 months post exposure.  These results suggest that



short Kevlar® fibers are cleared via phagocytosis followed by



transport to the lymphatic system.

-------
                               154
IV.1.2.  Effects on Experimental Animals



     The pathogenic potential of aramid fibers has been



investigated in a single long-term inhalation study an-d a number



of short-term inhalation studies as well as injection studies.



The experimental protocols and results of available studies are



summarized in Table 7 (pages 246-247).








IV.1.2.1.  Oncogenicity



     It has been shown that chronic inhalation to a dust cloud of



ultrafine para-aramid (Kevlar®) fibrils resulted in increased



lung tumor formation in rats.  In addition, intraperitoneal



injection of para-aramid (Kevlar®) pulp or fibers caused low



incidences of peritoneal tumors in rats.








IV.1.2.1.1.  Inhalation Studies
     The results of a long-term inhalation study which



investigated the oncogenic potential of ultrafine Kevlar® fibrils



in rats are summarized in an unpublished report by Reinhardt



(1986).  In this study, 5 groups of male and female Sprague-



Dawley rats (100 of each sex per group) were exposed to dust



clouds containing ultrafine Kevlar® fibrils (90 percent <1.5 urn



in width; more than 75 percent less than 20 pm long) at targeted



concentrations of 2.5, 25, 100, or 400 fibrils/mL.  The means of



weekly Kevlar® fibril counts during the exposure period were 2.4,



25.4, 112.2 or 435.5 fibrils/mL, respectively (approximately



equivalent to 0.08, 0.32, 0.63 or 2.23 mg/m3, respectively).



Rats were exposed for two years (6 hours/day, 5 days/week) with

-------
                               155
the exception of males and females in the highest dose group



which were exposed for only one year and then maintained for one



year without exposure.  This was due to high mortality and



apparent lung toxicity found following one year of exposure at



the highest dose.  By the end of first year of exposure, 34 males



and 15 females in the high dose group were found dead or were



sacrificed in extremis.



     Lung tumors identified as cystic keratinizing squamous cell



carcinomas were observed in rats exposed to the two highest dose



levels; in the group exposed to 400 fibrils/mL, 1 of 36 males



(3%) and 6/56 females  (11%) developed lung tumors.  At 100



fibrils/mL, none of the exposed males had tumors  (0/68) but lung



tumors were found in 4 of 69 females (6%).  Squamous cell



metaplasias were also observed in 6 females exposed to 100



fibrils/mL.  Control animals had no lung tumors.








IV.1.2.1.2.  Intraperitoneal Injection Studies



     Two intraperitoneal injection studies have also been



conducted to determine the oncogenic potential of Kevlar® in



laboratory animals.  Pott et al. (1987b) reported a low tumor



yield with Kevlar® in rats.  Female Wistar rats (8 weeks of age)



were administered 4 weekly intraperitoneal doses of 5 mg of



Kevlar® (50% <3.9 urn and 90% <11 jum long; 50% <0.47^um and 90%



<0.75 urn in diameter) suspended in saline.  Surviving animals



were sacrificed at 130 weeks after first treatment.  Three of 53



animals were found to have peritoneal tumors.  The tumor



incidence in the vehicle controls was 1/102.

-------
                               156
     The results of another intraperitoneal study conducted by



Davis (1987) also showed a low tumor yield in rats with Kevlar®



pulp.  Three groups of male A-F/Han strain rats (3 months of age)



received a single intraperitoneal injection of a preparation of



disaggregated filaments of Kevlar® pulp suspended in phosphate



buffered saline at either 25, 2.5, or 0.25 mg of the test



material.  By mass the bulk of the injected material consisted of



aggregates of large fibers.  However, a small proportion of the



injected sample was composed of free fibrils which were within



the respirable range (96 percent <1 pm and 56 percent <0.25 /am in



diameter).  A group of untreated animals were maintained as



controls.  There were no significant differences in the survival



between the treated and untreated groups.  More than 50 percent



of animals survived for more than 800 days and the oldest



survivors exceeding the age of three years.  Two animals in the



high dose group consisting of 32 animals which received 25 mg of



Kevlar® developed peritoneal mesothelioma.  Both mesotheliomas



were typical of those induced by asbestos.  No peritoneal



mesotheliomas were found in the two  low dose treated group (0/32



in the 2.5 mg and 0/48 in the 0.25 mg dose groups) and the



untreated controls (0/48).








IV.1.2.2.  Fibrogenicity



     Results of two inhalation studies indicate that ultrafine



para-aramid (Kevlar®) fibrils induced low fibrogenic activity in



rats.  Animal studies with commercial grade para-aramid (Kevlar®)



by inhalation and intratracheal instillation and Nomex® by the

-------
                               157
intratracheal route did not produce lung fibrosis in rats.



However, injection of para-aramid (Kevlar®) pulp into the



peritoneal cavity of rats produced strong tissue reactions and a



minimal degree of fibrosis.








IV.1.2.2.1.  Inhalation Studies



     Lee et al. (1983) studied the pulmonary response of male



Crl:CD rats exposed by inhalation to ultrafine Kevlar® fibrils at



O.I/ 0.5, 3.0 or 18 mg/m  and commercial Kevlar® fiber at 18



mg/m  for 2 weeks.  Fiber length distribution at different



exposure concentrations of ultrafine Kevlar fibers showed 60-70



percent of fibers were 10-30 jam in length and less than 1 urn in



diameter.  Only 13 percent of airborne commercial Kevlar® fibers



were respirable.  Five rats from each group were sacrificed at



the end of 2 weeks of exposure.  Subsequently, 5 rats from each



group at 0.1, 0.5, 3.0 mg/m  exposed to ultrafine Kevlar®, and 5



rats exposed to commercial Kevlar® at 18 mg/m  were killed at 2



weeks, 3 months and 6 months post exposure.  Five rats at 18



mg/m  exposure were sacrificed at 4 and 14 days, 3 and 6 months



after exposure.  Control animals exposed to air alone were



sacrificed at the same intervals as the exposed rats.



     The pulmonary response in rats exposed to commercial Kevlar®



at 18 mg/m3 and ultrafine Kevlar® at 0.1, 0.5, and 3.0 mg/m3



essentially satisfied biological criteria for nuisance dusts,



i.e., they did not produce significant collagen formation or



permanent alteration of basic lung structure, and tissue



reactions were reversible.  However, ultrafine Kevlar® fibrils at

-------
                               158
concentrations of 18 mg/m  produced minimal collagen fiber



formation in the alveolar duct region where dust particles



accumulated.  It should be noted that this study was an



assessment of the effects of short-term exposure and the results



alone cannot be extrapolated to assess long-term hazard.



     Reinhardt (1986) also reported a low level of lung fibrosis



in rats chronically exposed to ultrafine Kevlar® at 2.5, 25 or



100 fibers/mL for 104 weeks or 400 fibers/mL for 52 weeks.  Lung



lesions observed in male and female rats included alveolar type



II cell hyperplasia, bronchoalveolar hyperplasia, collagen fiber



granulomas, cholesterol-containing granulomas, and formation of



ciliated columnar cells in the alveolar ducts (alveolar



bronchiolarization).  A dose-response trend was evident with



fewer, less severe lesions occurring in groups exposed to 25



fibers/mL and the highest incidences and most severe lesions



present in rats exposed to 400 fibers/mL.  No pathological events



were evident in rats exposed to 2.5 fibers/mL.  Experimental



details of the study are provided in Section IV.1.2.1.1.








IV.1.2.2.2.  Intratracheal Instillation Study



     Reinhardt (1980) reported that intratracheal instillation of



Kevlar® polymer dust produced only a non-specific dust cell



reaction in rats.  Rats (strain, sex, and total number of animals



unspecified) were treated intratracheally with 25 mg of the



polymer dust which contained a low proportion of respirable



fibrous particles (<1.5 ^im in diameter and between 5-60 urn in



length) and a large proportion of larger nonrespirable particles

-------
                               159
ranging up to 150 urn in diameter.  Rats were maintained without
further treatment and were sacrificed at 2, 1 and 49 days and 3,
6, 12 and 21 months after treatment.  A group of control rats
received saline only.  No differences in mortality rates,
clinical observations and gross autopsy results were observed
throughout the study.  Initially there was a nonspecific
inflammatory response in the rat lung which then subsided within
one week.  In later sacrifices, however, foreign body granulomas
containing dust particles were found along with a negligible
amount of collagen.  All tissue responses to dust particles
decreased with increasing recovery periods.
     Similarly, Reinhardt (1980) reported that intratracheal
instillation of fibrous dust of Nomex® did not show any
progressive pulmonary fibrosis in rats.  Rats (strain, sex, and
total number not specified) were instilled intratracheally with
2.5 mg of Nomex® suspended in physiological saline.  The test
material contained circular, oblong or rod-shaped particles
varying in size from 2-100 urn in length and 2-30 urn in
diameter.  Groups of rats (number unspecified) were sacrificed at
2 and 7 days, 3 and 6 months, and 1 and 2 years following
exposure.  Initial transitory acute inflammation followed by
foreign body granuloma formation was produced by larger
nonrespirable particles (3-100 jam).  Respirable particles
«10 urn) produced only negligible dust cell reaction similar to
that seen with nuisance particulates.  After 2 years post-
exposure, lungs appeared normal.  It should be pointed out that
the actual data of both intratracheal instillation studies were
not provided in the report.

-------
                               160
IV.1.2.2.3.  Intraperitoneal Injection Study



     In a long-term intraperitoneal injection study by Davis



(1987) / Kevlar® pulp was found to cause a low level of peritoneal



fibrosis in rats.  Histological examination of peritoneal tissues



from rats injected with 25 mg disaggregated Kevlar® pulp taken at



varying time periods between one week and 9 months after



injection showed the formation of cellular granulomas.  These



granulomas consisted of macrophages and fibroblasts and foreign



giant cells.  There was also a small amount of fibrosis with



deposition of reticulin and collagen fibers.








IV.1.3.  In Vitro Studies
IV.1.3.1.  Genotoxicity



     There is no information available on the genotoxicity of



aramid fibers.








IV.1.3.2.  Cytotoxicity



     The results of an in vitro study by Dunnigan et al. (1984)



indicate that short, thin aramid fibers are at least as cytotoxic



as chrysotile asbestos to rat pulmonary alveolar macrophages.



Short, thin aramid fibers were extracted from commercial grade



Kevlar®.  Ninty percent of the counted fibers were less than 5 jam



long and smaller than 0.25 pm in diameter.  Average fiber length



and diameter were 2.72 pi and 0.138 jLtm, respectively.  The test



fibers were added to freshly harvested pulmonary alveolar



macrophages or cultured macrophages obtained from adult male



Long-Evans black hooded rats, at concentrations of 0, 25, 50,

-------
                               161
100, or 200 ug/mL.  After an 18-hour incubation period,



cytotoxicity was assessed by measuring lactate dehydrogenase



(LDH) and beta-galactosidase enzyme released into the incubation



medium.  Releases of LDH (10-55%) and beta-galactosidase (0-48%)



from pulmonary alveolar macrophages were essentially identical



when either aramid or chrysotile fibers were incubated with



freshly harvested cells.  With cultured macrophages, the



cytotoxic response was even higher with aramid fibers than



chrysotile.








IV.1.4  Assessment of Health Effects
     There is sufficient experimental evidence to conclude that



ultrafine para-aramid is potentially carcinogenic and



fibrogenic.  Due to limited comparative data bases, it is not



possible at this time to definitively assess the pathogenicity of



ultrafine para-aramid relative to that of asbestos.  This



material, however, does not pose a health risk to humans because



it is not available in commerce.  For the commercial grades of



para-aramid fiber and pulp to which humans are exposed, there is



limited experimental evidence suggesting that they may have a



low oncogenic and fibrogenic potential.  Thus, a possible health



hazard exists from exposure to commercial grade para-aramid



particularly the pulp form which may generate respirable airborne



fine fibrils upon abraison.



     Nomex® aramid which contains mainly nonrespirable fibers do



not appear to pose a significant health hazard to humans.

-------
                               162
IV.1.4.1.  Oncoqenicity



     Ultrafine para-aramid may be classified as a probable human



carcinogen (category B2) on the basis of sufficient evidence of



para-aramid carcinogenicity in animal studies and in the absence



of human data.



     There is no information available on the oncogenicity of



para-aramid fibers in humans.  The results of an inhalation study



indicate that ultrafine para-aramid (Kevlar®) is tumorigenic in



rats via inhalation (Reinhardt, 1986).  It was shown that chronic



inhalation exposure to ultrafine Kevlar® resulted in the



development of malignant lung tumors in female rats in a dose-



related manner (100 and 400 fibers/mL).  No injection data are



available on ultrafine Kevlar® to further assess its relative



carcinogenic potency in comparison with asbestos.  However/ the



positive findings from the inhalation bioassay are further



supported by the weak tumorigenic responses observed in rats



treated intraperitoneally with commercial grade para-aramid



(Kevlar®) fibers (Pott et al., 1987b) and pulp (Davis, 1987)



which contained small numbers of thin fibrils.



     In the absence of epidemiological data and based on the



limited evidence of carcinogenicity in animals, commercial grade



para-aramid may be classified as a possible human carcinogen



(Category C).  The limited evidence of carcinogenicity in



laboratory animals is provided by the positive results of two



intraperitoneal injection studies which showed that Kevlar® fiber



and pulp induced low incidences of peritoneal tumors in rats



(Pott et al., 1987b; Davis, 1987).  The authors attributed the

-------
                               163
low tumorigenic responses of aramid fibers and pulps to technical



difficulty in administering the test fibers, since they tend to



aggregrate and form large clumps.



     No experimental studies were available that evaluated the



oncogenic potential of Nomex® aramid fibers.  Thus/ Nomex® is not



classifiable as to human carcinogenicity (category D) due to



inadequate data in humans and animals.








IV.1.4.2.  Fibrogenicity



     There were no epidemiological studies available that



examined the potential fibrogenicity of aramid fibers.  Results



of two inhalation studies showed that ultrafine para-aramid



(Kevlar®) is weakly fibrogenic in rats.  Minimal pulmonary



fibrosis was induced in rats by 6 months following a 2-week



inhalation exposure to high concentrations (18 mg/m3) of



ultrafine Kevlar® fibrils (Lee et al., 1983).  Furthermore, dose-



related pathological lung effects including alveolar type II



hyperplasia, alveolar broncholarization, and collagenized



fibrosis were also observed in rats following long-term



inhalation exposure to ultrafine Kevlar® at 25, 100 and 400



fibrils/mL (equivalent to approximately 0.3, 0.6 and 2.23 mg/m3,



respectively (Reinhardt, 1986).  The concern for the fibrogenic



potential of ultrafine Kevlar® is further supported by findings



of an in vitro study demonstrating that short, thin aramid fibers



extracted from commercial grade Kevlar® are as cytotoxic as



chrysotile asbestos to rat pulmonary macrophages (Dunnigan et



al., 1984).

-------
                               164
     Lung fibrosis was not found for commercial Kevlar® in a



short-term inhalation study in rats (Lee et al., 1983).  However,



a low fibrogenic effect has been demonstrated for commercial



Kevlar® pulp via injection.  Davis (1987) showed that inoculation



of Kevlar® pulp into the abdominal cavity of rats resulted in a



low level of peritoneal fibrosis.  On the other hand, no fibrogenic



effects were observed in rats instilled intratracheally with large



nonrespirable Kevlar® or Nomex® aramid fibers (Reinhardt, 1980).








IV.1.5.  Recommendations
     In order to fully assess the potential health effects of



para-aramid fibers or pulps, additional animal testings by



inhalation or injection appear necessary.  Because of a low



health concern, no further testing is recommended for Nomex®.








IV.2.  Carbon Fibers
     Carbon fibers are synthetic fibers which are characterized



by light weight, high tensile strength, flexibility, good



electrical conductivity, thermal resistance, and chemical



inertness (except to oxidation).  Carbon fibers are mainly used



as reinforcing materials in structural composites.  They are



currently not used as asbestos substitutes but may replace



asbestos in thermal and electrical insulation, textiles, and



friction products.



     Carbon fibers (92 percent carbon by weight) are made by the



carbonization (i.e., pyrolysis) of precursor polyacrylonitrile



(PAN), rayon, or pitch fibers, with PAN-based carbon fibers being

-------
                               165
the most common.  Carbon fibers are manufactured as continuous or



chopped fibers.  The nominal diameter of carbon fibers range from



5-8 urn, which fall outside the resprrable range.  However, less



than 25 percent of these fibers have diameters less than 3 pm and



shorter than 80 jam, which are considered respirable.



Furthermore, upon mechanical or thermal stress, carbon fibers may



split longitudinally to finer respirable fibers (ICF, 1986).








IV.2.1.  Fiber Deposition, Clearance and Retention



     There is very limited information on the deposition and



clearance of carbon fibers.  Results of available inhalation



studies in guinea pigs indicate that inhaled carbon fibers are



capable of penetrating the alveoli.  In the lung, nonfibrous



carbon particles are phagocytized by alveolar macrophages while



uncoated carbon fibers longer than 5 jjm are found in the



extracellular matrix.  Carbon fibers appear to be cleared from



the lung slowly as evidenced by the detection of uncoated carbon



fibers and dust-laden macrophages in the lung even after 6 months



to 2 years following exposure.



     Holt and Horne (1978) exposed guinea pigs to high



concentrations of a respirable dust cloud of carbon fibers for



7-24 hours.  Most of the respirable dust (99 percent) was



nonfibrous (370 particles/mL) and the airborne concentration of



respirable fibers (1-2.5 urn in diameter and lengths up to 15 urn)



was very low (2.9 fibers/mL).  Examination of lung tissues



revealed the presence of carbon particles to be intracellular in



the cytoplasm of macrophages.  The few carbon fibers found in the

-------
                               166
lung that were longer than 5 urn were still extracellular after 27


weeks post-exposure.  These fibers were uncoated.


     In a subsequent experiment, Holt (1982) exposed guinea pigs


to dust of carbon fibers for 100 hours.  The carbon dust was


reported to be submicron in size and mainly nonfibrous.  However,


dust concentration and particle dimensions were not given.  It


was reported that phagocytosis of the dust particles commenced


immediately one day after exposure but proceeded slowly, with the


number of dust-laden macrophages continuing to increase up to 400


days post exposure.  Macrophages containing dust began to decline


after that but were still evident even after 2 years following


exposure.




IV.2.2.  Effects on Experimental Animals


     A number of studies have been conducted to evaluate the


oncogenic and fibrogenic potential of carbon fibers in laboratory


animals via various routes of exposure.  Table 8 (pages 248-250)


summarizes the experimental protocols and findings of relevant


studies on carbon fibers.
       s~



IV.2.2.1.  Oncogenicity


     No information is available on the oncogenicity of carbon


fibers in animals via inhalation.  Studies of carbon fibers in


rats by intratracheal instillation, intraperitoneal injection,


and intramuscular implantation have reported no tumorigenic


response.  Petroleum pitch-based continuous fibers were reported


to be weakly oncogenic in mice by the dermal route, and

-------
                               167
subcutaneous implantation of carbon fibers were reported to



produce local sarcomas in rats.  However, the results of these



studies are questionable in view of inadequate reporting of the



test results and/or the nature of the materials tested.








IV.2.2.1.1.  Intratracheal Instillation Studies



     A small intratracheal instillation study was conducted at



the U.S. Air Force Aerospace Materials Research Laboratory by C.



Olson.  Carbon fibers were reduced to respirable size  (20 percent



<1 pm in diameter/ with varying lengths) by partial oxidation at



a high temperature and then injected intratracheally into the



lungs of male Fischer rats.  The rats were maintained over a



2-year period.  As reported by Parnell (1987), no lung tumors



were found at 200 days, 1 year or 2 years after treatment.  This



was a preliminary report and details of the study design (e.g.,



dosage, number of animals) and results were not given for full



evaluation.








IV.2.2.1.2.  Intraperitoneal Injection Studies



     Parnell (1987) also reported that no tumor response was



observed in male Fischer rats treated intraperitoneally with



respirable carbon fibers.  Mesotheliomas were not found in any of



the treated rats at 200 days or 2 years post-treatment.  No other



experimental details were provided.

-------
                               168
IV.2.2.1.3.  Intramuscular Studies



     Tayton et al. (1982) investigated the carcinogenic potential



of intramuscular implantation of carbon fiber in strand and



powdered forms in rats and found no signs of any malignant



changes.  In one experiment, a 1.5 urn length carbon fiber



Graffil®, was inserted intramuscularly into the left gluteal



muscle of 50 rats (unspecified sex and strain).  A group of 50



control animals received an implant of black braided silk suture



material.  In a second experiment, groups of 10 rats had either a



5 jjm length of carbon fiber (Graffil®) or the black silk (for



control) tied around the periosteum of the left femur.  In a



third study, a group of 50 rats each received intragluteal



injection of a suspension of powdered carbon fibers (unspecified



particle size).  All surviving animals were sacrificed at 18



months and morphological and histological examinations of the



implants were performed.  In all cases, minimal tissue reactions



were observed and there was no evidence of malignant changes.  It



should be noted that the route of administration used in this



study may not be relevant in providing information with regard to



the potential of carbon fibers in inducing lung toxicity and



carcinogenicity via inhalation.








IV.2.2.1.4.  Subcutaneous Implantation Studies



     In 1982, Maltoni et al. (1982b) reported initiation of



testing of carbon fibers for oncogenicity in rats by subcutaneous



implantation.  Male and female Sprague-Dawley rats (40 per sex)



were each subjected to subcutaneous implantation of a 2 cm

-------
                               169
diameter disc containing 25 rag of carbon fibers (fiber dimension



not specified).  Animals were to be kept under observation until



spontaneous death and a complete necropsy and histopathologic



evaluation of the tissues were to be performed.  Recently,



Maltoni et al. (1987) reported in an abstract that the



preliminary results of this study showed an induction of local



sarcomas in carbon fiber treated rats, but no other details



regarding the nature of the test material and test results were



available for a full evaluation of the findings.








IV.2.2.1.5.  Dermal Studies
     DePass (1982) evaluated the potential of carbon fibers in



inducing cancer of the skin in mice.  Four types of carbon fibers



were tested:  (1)  continuous filament (CF) pitch-based, (2)



pitch-based carbon fiber mat (MAT), (3)  polyacrylonitrile



continuous fibers (PAN-based), and (4) oxidized PAN-based (PAN-



oxidized) fibers.  No tumorigenic response appeared to be



elicited by PAN-based, MAT, or PAN-oxidized fibers.  The CF



pitched-based, however, were judged to produce a weak tumorigenic



response.



     Groups of 40 male C3H/HeJ mice each received a 25 uL



application of the test material suspended in benzene (10% w/v),



to the clipped skin of the back three times weekly until death.



Each of the four fiber types was ground by mortar and pestle



(particle size not specified).  A group of negative controls



received benzene only while positive control animals were treated



with 0.1 percent methylcholanthrene in acetone.  No skin tumors

-------
                               170
were found in the groups treated with PAN-based fibers or vehicle



control (benzene).  In the CF pitch-based group, one papilloma



(1/40) and one squamous cell carcinoma (1/40) of the skin at the



application sites were found.  Because of extremely low



historical control incidence (0/285 in C3H/HeJ mice used as



benzene controls at that laboratory), CF pitch-based fibers were



considered to be marginally oncogenic under the conditions of the



study.  A very low incidence of various types of skin tumors were



found in rats treated with MAT or PAN-oxidized fibers.  However,



MAT and PAN-oxidized fibers were considered to have questionable



oncogenic potential because the tumors observed were distal to



the application site.








IV.2.2.2.  Fibrogenicity



     A number of studies have been conducted to determine the



fibrogenic potential of carbon fibers by various routes of



exposure.  With the exception of a report that carbon fiber is



fibrogenic in rats via intratracheal instillation, other studies



have not produced positive results.  It should be noted, however,



that available studies are of little value in evaluating the



fibrogenicity of carbon fibers because of their limited scope or



experimental design, and/or lack of information on the test



material or test results.







IV.2.2.2.1.  Inhalation Studies
     Two inhalation studies were conducted in an attempt to



evaluate the effects of short-term exposure to chopped PAN-

-------
                               171
oxidized carbon fibers in guinea pigs.  In the first study,  Holt



and Home (1978) exposed 13 specific pathogen-free guinea pigs to



carbon dusts for up to 104 hours.  No pathological effects were



found in the lung of exposed guinea pigs examined at 1-144 days



post-exposure.   It should be pointed out that 99 percent of the



respirable dust generated was nonfibrous (370 particles/mL)  and



the levels of respirable carbon fibers were extremely low (2.9



fibers/mL).   These fibers had diameters of 1.0-2.5 urn and lengths



up to 15 urn.



     In a subsequent study, Holt (1982) also reported no evidence



of pathological changes in the lungs of guinea pigs exposed to



submicron carbon dusts.  Specific pathogen-free guinea pigs (2-9



per group) were exposed to carbon dust for 7-12 hours or 100



hours and single animals were killed at intervals after one to



720 days.  The dust was reported to be submicron in size and



mainly nonfibrous.  Dimensional characterization of the dust was



not provided, nor were dust concentrations reported.  Exposed



guinea pigs showed no lung fibrosis nor other pathology.



However, it should be pointed out that these two studies only



demonstrated that short-term inhalation of mostly nonfibrous



respirable fragments of carbon fibers caused no adverse effects



to guinea pigs.



     Recently, Owen et al. (1986) conducted a subchronic



inhalation toxicity study on PAN-based carbon fibers and reported



no systemic toxicity nor progressive pulmonary dysfunction in the



exposed rats.  Four groups of 10 male Sprague-Dawley rats were



exposed to carbon fibers for 6 hours daily, five days a week for

-------
                               172
4, 8, 12, or 16 weeks and were sacrificed at the end of the
exposure.  A fifth group consisting of 20 animals were exposed
for 16 weeks and were kept for 32 weeks post-exposure.  The mean
atmospheric concentration of carbon fibers was 20 mg/m , with a
range of 16-23 mg/m .  Carbon fibers had a mean diameter of 7 urn
and lengths ranging from 20-60 urn.  A similar number of control
rats were exposed to air only and were sacrificed at similar
schedules.  One death occurred during the sixth week of exposure
but was not considered related to treatment.  Pulmonary function
tests conducted prior to animal sacrifice did not show any
significant or consistent changes in airway resistance.
Histologic examination revealed no inflammatory or fibrogenic
reaction in the lungs of exposed rats.  A lack of an effect was
not unexpected considering that the dust cloud contained mostly
large nonrespirable fibers.

IV.2.2.2.2.  Intratracheal Instillation
     Troitskaya et al. (1984) reported findings on comparative
fibrogenicity of carbon fibers and asbestos.  Rats each received
a single intratracheal administration of either chrysotile
asbestos or one of two preparations of polyacrylonitrile-
reinforced carbon fibers.  The animals were examined 1-9 months
after the administration of the dust.  It was reported that
chrysotile asbestos was several-fold more fibrogenic than either
of the carbon fiber samples.  No further details were provided.
     In a study by Swensson (1979), as reported by Gross and
Braun (1984), a mixture of carbon fibers (size distribution not

-------
                               173
specified)  and unspecified plastic was injected intratracheally



into Sprague-Dawley rats (dose and number of animals not



specified).   The animals were mainta-ined for eight months.  Aside



from an acute foreign body reaction during the first month after



instillation/ there was no indication of obstructive lung disease



at 8 months  as judged by analysis of collagen content in the



lungs.   No other details were given for conclusive evaluation.



     Parnell (1987) recently reported that there was no evidence



of any adverse effects in male Fischer 344 rats following



intratracheal instillation of respirable carbon fibers.  No



treatment-related degenerative lesions of the lungs were observed



in this long-term study.  This was only a preliminary report and



full results were not available for evaluation.








IV.2.2.2.3.   Intraperitoneal Injection Studies



     Styles  and Wilson (1973) reported that carbon fiber was not



fibrogenic in rats following intraperitoneal injection.  A group



of 6 male and 6 female SPF albino Wistar rats (200-250g) were



injected intraperitoneally with carbon fibers at a dose of 50



mg/kg (10-15 mg per animal).  Particle size of the test material



ranged from 0.2-15 urn in diameter.  No pathological lesions were



found at 1 and 3 months after treatment.  On the other hand, rats



injected with chrysotile asbestos developed diffuse fibrosis of



the peritoneum after 1-3 months post-treatment.  It should be



noted that a major limitation of this study was the short



observation  period and the use of a small number of animals.  It



was also not clear as to whether the test carbon dust was in




fibrous or nonfibrous form.

-------
                               174
     Parnell (1987) also reported negative findings with two



samples of respirable carbon fibers following injection into the



peritoneal cavity of male Fischer 344 rats.  No treatment-related



degenerative lesions were observed in either treated animal group



at 200 days or 2 years after treatment.  This was only a



preliminary report and full results were not available for



evaluation.








IV.2.2.2.4.  Other Studies
     Neugebauer et al. (1981) conducted a series of experiments



to determine the reaction of tissues to carbon fibers.  In this



study/ 50 mg of carbon fiber reinforced carbon (CFRC) fragments



(diameter of 7 um and lengths between 20-100 urn) were injected



into the femoral medullary canal of 16 rabbits of the CHBB:CH



strain.  Tissue reactions were evaluated at 2 and 12 weeks post-



treatment.  A small amount of fibrosis and foreign body giant



cell reactions were found in the medullary cavity.  It was also



reported that previous experiments in rats involving intravenous,



intraperitoneal or intra-articular injections of carbon fiber



particles (1-8 um) also showed no evidence of tissue reaction.



No additional details were available.








IV.2.3.  In Vitro Studies
IV.2.3.1.  Genotoxicity



     Genotoxicity tests have been conducted on two types of



carbon fibers.  The carbon fibers are the acetone reconstituted



benzene extracts of pitch-based carbon fibers and poly-

-------
                               175
acrylonitrile (PAN)-based carbon fibers.  Available data indicate



that neither type of carbon fiber appears to cause gene



mutations. - However, pitch-based carbon fibers appear to be



clastogenic while a clastogenic mechanism cannot be entirely



ruled out for PAN-based carbon fibers.



     Both carbon fibers were negative in the Salmonella/mammalian



activation assay and the Chinese hamster ovary/hypoxanthine-



guanine-phosphoribosyltransferase (CHO/HPRT) mutation assay



(Litton Bionetics, 1980; Union Carbide, 1983a, 1983b, 1983c,



1983d).  Pitch-based carbon fibers induced significant



concentration dependent increases of sister chromatid exchanges



(SCE) in CHO cells and unscheduled DNA synthesis (UDS) in primary



rat hepatocytes (Union Carbide, 1983c).  In the SCE assay,



chromosomal aberrations were also noted for this carbon fiber.



On the other hand, PAN-based carbon fiber did not induce



significant increases in UDS in primary rat hepatocytes and the



frequency of SCE in CHO cells.  However, several types of



chromosomal aberrations were observed in the SCE assay (Union



Carbide, 1983d).  These results are consistent with other fiber



studies where mineral fibers do not appear to cause gene



mutations, but are clastogenic.








IV.2.3.2.  Cytotoxicity



     Both positive and negative results were obtained from in



vitro cytotoxicity studies on carbon fibers.  Carbon fibers were



reported in one study to be non-hemolytic to rabbit erythrocytes



but highly cytotoxic to rabbit alveolar macrophages.  However, in

-------
                               176
another study, carbon fibers were found to cause no cytotoxic



effects to either rat alveolar or peritoneal macrophages.  Carbon



fibers also did not affect rabbit lung fibroblast cultures.  It



would appear that the discrepancy in the observed in vitro



biological activity of carbon fibers might be related to



differences in fiber type and/or size distribution of the test



materials.







IV.2.3.2.1.  Erythrocytes



     Richards and Hunt (1983) reported that carbon fibers had



little or no hemolytic activity in rabbit erythrocytes.  The test



fibers were obtained by grinding carbon fiber cloth.  Ninety



percent of the fibers were less than 10 urn in length.  Hemolysis



was observed only at a relatively high dose compared to



chrysotile asbestos.  No other details were available.







IV.2.3.2.2.  Phagocytic Cells



     Richards and Hunt (1983) also reported that carbon fibers



(90 percent <10 um long) were highly cytotoxic to rabbit alveolar



macrophages following one hour of incubation.  Experimental



details were not provided for full evaluation.



     On the other hand, Styles and Wilson (1973) found that



carbon dusts (0.2-15 jam in diameter) were not cytotoxic to rat



peritoneal macrophages or alveolar macrophages.  The test dusts



were incubated with cells for 2 hours and cell viability was



assessed at 0, 1 and 2 hours after addition of dust.  Less than



2-5 percent of cells were killed following phagocytosis of carbon

-------
                               177
dust.  In contrast, chrysotile asbestos induced a high degree of
cytotoxicity.

IV.2.3.2.3.  Fibroblasts
     Richards and Hunt (1983) tested the effect of ground carbon
fiber cloth (90 percent <10 pm long) on rabbit lung fibroblast in
culture.  The amount of DNA and hydroxyproline levels in the
culture were measured after 24 days of exposure.  Treatment of
fibroblast cultures with carbon fibers affected neither parameter.

IV.2.4.  Assessment of Health Effects
     Currently available data provide inadequate evidence of
carcinogenicity and fibrogenicity for carbon fibers.  However/
based on suggestive evidence from a dermal study and positive
clastogenic effects in genotoxicity tests with pitch-based carbon
fibers, a weak oncogenic potential for certain types of respirable
carbon fibers may exist.  Overall, carbon fibers appear to pose a
lower degree of health hazard compared to asbestos because they
are less respirable, and less biologically active than asbestos as
demonstrated in the few available comparative studies.

IV.2.4.1.  Oncogenicity
     Carbon fiber is not classifiable as a human carcinogen
(Category D) based on inadequate evidence of carcinogenicity from
animal studies and in the absence of human data.
     No information is available on the potential development of
respiratory malignant diseases in humans from exposure to carbon
fibers.  Moreover, there are no animal data on the oncogenic

-------
                               178
potential of carbon fibers via inhalation.  However, negative



results have been reported in rats via intratracheal instillation



(Parnell, 1987), intraperitoneal injection (Parnell, 1987), and



intramuscular implantation (Tayton et al., 1982).  The studies



that reported positive results were those of a subcutaneous study



with carbon fibers in which an increased production of local



sarcomas were found in rats (Maltoni et al., 1987) and a lifetime



skin painting study with pitch-based carbon fibers showing a



nonstatistically significant increase of skin tumors in mice



(DePass, 1982).  The biological significance of these findings



remains uncertain in light of the absence of particle size and



morphology data, the weak tumorigenic response in the dermal



study, the lack of data reported in the subcutaneous injection



study as well as the questionable relevance of its method of



administration to human exposure at the workplace.  On the other



hand, the positive clastogenic effects of benzene-extracts of



pitch-based carbon fibers (Union Carbide, 1983c) tend to support



an oncogenicity concern for this carbon fiber type.  Additional



data are needed to conclusively evaluate the oncogenicity of



carbon fibers.








IV.2.4.2.  Fibrogenicity



     Available data are insufficient to evaluate the fibrogenic



potential of carbon fibers.  There  is one single small cross-



sectional study which showed no evidence of pathological effects



in the lungs of workers in a PAN-based carbon fiber production



plant, based on respiratory symptoms, spirometric and chest

-------
                               179
radiographic data (Jones et al., 1982).  It should be noted,



however, that respirable fiber concentrations in this facility



were low and that the duration of exposure to carbon fibe-rs was



relatively short.  Thus, the results of this study do not provide



conclusive evidence of a negative effect (Battelle, 1988).



     With regard to experimental studies, there are no data



available on the long-term effects of inhalation of respirable



carbon fibers in animals.  The results of a subchronic inhalation



toxicity study showed no evidence of lung pathology in rats



exposed to large diameter carbon fibers (Owen et al., 1986).  In



addition, several studies have reported that carbon fibers were



not fibrogenic in rats via intratracheal instillation (Parnell,



1987; Swenson, 1979), intraperitoneal injection (Parnell, 1987;



Styles and Wilson, 1973) or injection into the medullary cavity



of femur bone (Neugebauer et al., 1981).  On the other hand, it



was reported that PAN-based carbon fibers induced lung fibrosis



in rats following intratracheal instillation (Troitskaya et al.,



1984).  Most of these studies, however, are of little value for



the evaluation of the fibrogenic potential of carbon fibers



because of limited scope, lack of particle size and morphology



data of the test materials, and/or no details available on study



design and findings.  Furthermore, both negative and positive



findings have been reported regarding the in vitro cytotoxicity



of carbon fibers  (Styles and Wilson, 1973; Richards and Hunt,



1983).  Thus, available animal and in vitro studies do not



provide conclusive evidence for or against a fibrogenic effect



for carbon fibers.  They do suggest, however, that carbon fibers

-------
                               180
at most have low fibrogenic potential, as supported by results of



a few studies showing that chrysotile asbestos was more



fibrogenic, hemolytic and cytotoxic than c-arbon dust fibers under



the same experimental conditions (Troitskaya et al., 1984; Styles



and Wilson, 1973; Richards and Hunt, 1983).








IV.2.5.  Recommendations



     A chronic inhalation toxicity study was recently conducted



at a private laboratory.  When results become available, they



should be assessed to see if further study is warranted on this



fiber.








IV.3.  Polyolefin Fibers



     Polyolefin fibers are manufactured from long-chain,



synthetic polymers of ethylene, propylene or other olefin



units.  Approximately 95 percent of polyolefin fibers are made



from polypropylene, while most of the rest is from polyethylene.



Polyolefins are manufactured as monofilament yarn (greater than



153 urn in diameter), multifilament yarn (5-20 urn in diameter),



tape and fibrillated film yarn (continuous sheet), spun-bonded



fabric, staple fiber (chopped multifilament), synthetic pulp



(5-40 urn in diameter, 2.5-3 mm long), and microfiber (1-5 urn in



diameter).  Applications of polyolefin fibers and pulp as



substitutes for asbestos include roof sealant, asphalt solvent,



caulks, joint cement, adhesive, textile compounds, filter,



flooring felts, and roofing felts.  With the possible exception



of polyolefin microfibers, the likelihood that polyolefin fibers

-------
                               181
and pulp generate airborne respirable fibers appears small since



they generally fall outside the respirable range.  Furthermore/



it is unlikely that polyolefin fibers and pulp would split



longitudinally to produce finer respirable fibers (ICF, 1986).








IV.3.1.  Fiber Deposition, Clearance and Retention



     There is no information available on the lung deposition,



clearance and retention of polyolefin fibers.








IV.3.2.  Effects on Experimental Animals



     Very few studies have been conducted to determine the



oncogenic and fibrogenic potential of polyolefin fibers in



animals.  Table 9 (page 251) summarizes the experimental



protocols and results of available animal studies on polyolefin



fibers.








IV.3.2.1.  Oncogenicity



     There is no information available on the oncogenic potential



of polyolefin fibers in animals via inhalation.  Preliminary



results of an intraperitoneal injection study showed that



polypropylene fibers induced a low incidence of peritoneal tumors



in rats.  In a limited intratracheal insufflation study, both



polyethylene and polypropylene fibers were not tumorigenic in



rats.  Other forms of polyethylene and polypropylene including



disc,  film, rod fragment and powder produced local sarcomas in



mice and rats following subcutaneous or intraperitoneal



implantation (as reported in IARC, 1979).

-------
                               182
IV.3.2.1.1.  Intraperitoneal Injection Studies



     Pott et al. (1987b) recently reported preliminary findings



of a study of polypropylene fibers in rats via intraperitoneal



injection.  Female Wistar rats received 5 weekly injections of 10



mg of polypropylene fibers in saline.  Ninety percent of the test



fibers were less than 2.1 urn in diameter and less than 23 urn



long.  The animals were observed for full lifespan.  Peritoneal



tumors were found in 2/53 treated animals compared to 1/102



negative controls (saline).  In contrast, animals treated with



chrysotile asbestos showed a dose-dependent tumorigenic response



at extremely low doses.  The tumor incidences (identified as



mesothelioma or sarcoma) in the chrysotile group were 11/36,



21/34, and 30/36 at a single dose of 0.05, 0.25, and 1.0 mg,



respectively.








IV.3.2.1.2.  Intratracheal Insufflation Studies



     MB Research Laboratories (1980) conducted a long-term study



of the effects of intratracheal insufflation of polyolefin fibers



in rats.  Groups of 40 male Long-Evans rats were administered a



single dose of ozonized polyethylene SHFF, ozonized polypropylene



SHFF, or HHF polypropylene.  Control animals received vehicle



only (Tween 60).  All surviving animals were sacrificed at 21



months following administration of the test material.  A number



of deaths were observed in both control and treated groups.  The



cause of early deaths was attributed to dosing technique and that



of later deaths to infectious diseases.  Histologic examinations



showed the development of lung granulomas in all treatment

-------
                               183
groups.  No lung tumors were found in any test groups.  It should



be pointed out that in the absence of available information on



the characteristics of the test materials, specific dosages and



methods of administration, this study is of little value for the



evaluation of the oncogenic potential of polyolefin fibers.








IV.3.2.2.  Fibrogenicity



     There is no information available on the fibrogenic effects



of polyolefin fibers in animals via inhalation.  In a 3-month



study, peritoneal fibrosis was not observed in rats following a



single intraperitoneal dosing of polyethylene or polypropylene



dusts.  In a limited intratracheal study, lung fibrosis was not



produced in rats treated with either polyethlyene or poly-



propylene fibers.  However, preliminary results of a lifespan



study  in rats reported a strong degree of adhesions of the



abdominal organs following intraperitoneal injection of



polypropylene fibers.








IV.3.2.2.1.  Intraperitoneal Injection Studies



     In the study by Pott et al. (1987b) that was described in



the oncogenicity section, a strong degree of adhesions of



abdominal organs was observed macroscopically in rats treated



with 5 weekly doses of 10 mg of polypropylene fibers  (90 percent



<2.1 um in diameter; 90 percent <23 um long).  It was not clear



whether there were any developments of fibrosis in treated



rats.  Histological data are not yet available for a full



evaluation of this preliminary finding.

-------
                               184
     Styles and Wilson (1973) reported that polyethylene and



polypropylene dusts were not fibrogenic in rats via the intraper-



itoneal route of exposure.  In this study, groups of 6 male and 6



female albino Wistar rats (200-250g) were intraperitoneally



administered a single dose of either polyethylene (3-75 urn in



diameter) or polypropylene (4-50 jum in diameter) dusts at 50



mg/kg (10-15 mg/animal).  No fibrosis were observed at 1 or 3



months after treatment in either treated group.  On the other



hand, chrysotile asbestos at lower doses produced characteristics



of fibrotic nodules after 1 month and diffuse fibrosis by 3



months.  It is difficult to evaluate these findings in view of



the small numbers of animals used in this study and short



duration of the study.  Moreover, it was not clear whether the



test dusts were fibrous or nonfibrous.








IV.3.2.2.2.  Intratracheal Insufflation Studies



     In the long-term study by MB Research Laboratories (1980)



that was described in the oncogenicity section, lung fibrosis was



not found in rats at 21 months following intratracheal insuffla-



tion of ozonized polyethylene SHFF, ozonized polypropylene SHFF,



or HHF polypropylene.  No information with regard to the charac-



teristics of the test materials, the particle size distribution



and administered doses was provided. These findings are therefore



inadequate for definitive assessment.

-------
                               185
IV.3.3.  In Vitro Studies



IV.3.3.1.  genotoxicity



     Polyethylene extracts were tested in the Salmonella/mamma-



liam activation assay in strains TA98, TA100 and TA1537 (Fevolden



and Holler, 1978).  This report is an abstract that provides no



details, therefore it is not completely adequate for assessment.








IV.3.3.2.  Cytotoxicity



     Results of a single in vitro study showed that polypropylene



and polyethylene dusts were significantly less cytotoxic to



alveolar or peritoneal macrophages than was chrysotile



asbestos.  However, the authors did not specify whether the



materials tested were fibrous or nonfibrous.



     In the study by Styles and Wilson (1973), peritoneal and



pulmonary macrophages (10  cells/mL) obtained from male and



female Wistar rats were treated with either polyethylene or



polypropylene dust at 500 ug/mL.  The particle size of poly-



ethylene and polypropylene dusts ranged between 3-75 urn and 4-50



urn in diameter, respectively.  Negative control cultures were



untreated while positive controls were treated similarly with



chrysotile asbestos.  Results of cell culture experiments indi-



cated that cells treated with asbestos had the highest mortality



(10-60 percent) while those tested with either polyethylene or



polypropylene dust had the lowest mortality  (less than 2 percent



to 5 percent), as measured by the ratio of percentage of living



to dead cells after 1 and 2 hours post-incubation.

-------
                               186
IV.3.4.  Assessment of Health Effects



     Available data are inadequate for a conclusive assessment of



potential carcinogenic and fibrogenic effects of pblyolefin



fibers, but they do suggest that polyolefin microfibers may have



low fibrogenic potentials.  Because polyolefin fibers or pulp are



generally not respirable, inhalation of these fibrous materials



would pose little or no health hazard to humans.  On the other



hand/ a health hazard potential for polyolefin microfibers may



exist since these may be respirable.








IV.3.4.1.  Oncogenicity



     Polyolefin fibers are not classifiable as to human carcino-



genicity (Category D) on the basis of inadequate evidence of



carcinogenicity in animal studies and no human data.



     There are no available epidemiological or animal inhalation



studies that examine the oncogenic potential of polyolefin



fibers.  The results of a limited intratracheal insufflation and



an apparently well-conducted intraperitoneal injection



oncogenicity study have not provided conclusive evidence of



carcinogenicity for polyolefin fibers in animals.   In the long-



term intratracheal insufflation study, both polyethylene and



polypropylene fibers did not induce tumor in rats  (MB Research



Laboratories, 1980).  However, the lack of information on the



nature, size distribution and dosage of the test materials



precludes any definitive assessment of the oncogenicity of these



fibers under the conditions of the study.  In the  long-term



intraperitoneal injection study in rats, a low tumor incidence

-------
                               187
was obtained with polypropylene microfibers (Pott et al.,



1987b).  These results, however, were only preliminary and a full



evaluation cannot yet be made.



     In contrast, the results of subcutaneous or intraperitoneal



implantation studies showed that polyethylene/polypropylene disc,



film, rod or fragments cause local sarcomas in mice and rats



(IARC, 1979).  However, because the test materials were not in



fibrous form, these findings are not considered relevant to the



assessment of the oncogenicity of fibrous polyolefin per se.








IV.3.4.2.  Fibrogenicity



     There is no information available on the fibrogenic effects



of polyolefin fibers in humans and animals via inhalation



exposure.  Available animal studies by the injection/insufflation



method and a single in vitro cytotoxicity study provide



inconclusive data, and thus definitive assessment of the



fibrogenic potential of polyolefin fibers cannot be made,



although these studies seem to suggest a lower fibrogenic



potential than that of asbestos.



     In a long-term intratracheal study with polyethylene and



polypropylene fibers, lung fibrosis was not observed in rats



following 21 months (MB Research Laboratories, 1980).  In a



short-term study, peritoneal fibrosis was not found in rats



treated with either polyethylene or polypropylene dusts, whereas



under similar experimental conditions but at lower doses,



chrysotile asbestos induced a low level of fibrosis after 3



months (Styles and Wilson, 1973).  These in vitro results are

-------
                               188
supported by the finding of a single in vitro study that both

polypropylene and polypropylene dusts had significantly lower

cytotoxicity in rat alveolar/peritone.al macrophages than did

chrysotile asbestos (Styles and Wilson, 1973).  However, the lack

of information on the characteristics of the tested fibers makes

it difficult to draw any definitive conclusions for this fiber

category.  On the other hand, preliminary results of an

apparently well-conducted long-term intraperitoneal injection

study in rats with long, thin polypropylene fibers showed a

strong degree of adhesions of the abdominal organs.  However,

without histological data, it is not yet known whether or not

fibrosis was also induced by polypropylene fibers  (Pott et al.,

1987b).



IV.3.5.  Recommendations
     A chronic inhalation study is recommended to further

evaluate the oncogenic and fibrogenic potential of polyolefin

microfibers.  Because of a low health hazard associated with

inhalation exposure to polyolefin fibers and pulps, additional

animal tests do not appear necessary at this time.



V.  Mechanisms of Fiber-Induced Diseases:  Relationships between
    Fiber Properties and Pathogenicity

     Epidemiological and experimental evidence accumulated thus

far suggests that inhalation of fibrous dust other than asbestos

might also be associated with malignant and nonmalignant

pulmonary diseases in humans.  However, the carcinogenicity and

fibrogenicity of nonasbestos fibers appear to be variable.  While

-------
                               189
erionite seems to be at least as potent as asbestos, if not more



so, fibrous glass and mineral wools are probably less hazardous



than asbestos.  On the o_ther hand, available experimental data



suggest that ceramic aluminum silicate glass, ultrafine aramid



(Kevlar®) fibrils, and long-fibered attapulgite are potentially



pathogenic, but conclusive assessment of the comparative



oncogenic potential of these fibers and that of asbestos cannot



yet be determined.  As for wollastonite, short-fibered



attapulgite, carbon fibers, and polyolefin fibers, these fibers



appear to exhibit considerably lower pathogenic potential than



asbestos.  It should be stressed, however, that these assessments



are by no means definitive because of incomplete data bases.



     Although it seems that certain asbestiform fibers can cause



asbestos-related diseases, there is some evidence suggesting that



the pattern of diseases may vary with different fiber types.  For



example, available epidemiological evidence suggests that



erionite exposure appears to be associated mainly with malignant



mesothelioma of the pleura and peritoneum.   This is consistent



with animal evidence showing that inhalation exposure to erionite



produces very high rates of mesothelioma in animals.  The



development of lung cancer, however, has not been demonstrated



experimentally nor has it been conclusively established in human



studies.  On the other hand, for man-made mineral fibers, there



is some epidemiological evidence for a possible association of



lung cancer and occupational exposure to fibrous glass or mineral



wool, but a risk of mesothelioma is not apparent.

-------
                               190
     An explanation of these differences may lie in the different



intrinsic fiber properties which may control biological



activity.- It should be pointed out, however, that mechanisms by



which mineral fibers, including asbestos, produce pathogenicity



are not understood.  Furthermore, little is known of the



physicochemical properties that determine pathogenicity.



However, it is still important to briefly discuss the



relationships of fiber characteristics with possible mechanisms



of fiber-induced diseases so that speculation about their



importance can be focused in terms of research needs, and



qualitative ranking of the hazard of fibers.



     Clearly, fiber dimension is an important determinant for the



development of any type of fiber-induced disease because it



governs the entry and bioavailability of a fiber at target



tissues.  Fiber diameter is the most important factor in



determining the respirability of the fiber.  The thinner the



fiber, the more respirable it is and the more easily it can



penetrate into the lung.  Fiber length and shape also affect the



respirability and pulmonary deposition of the fiber but to a



lesser degree.  However, fiber length is more important in terms



of fiber retention.  Short fibers «5 jum) are readily cleared by



macrophage uptake while long fibers (>20 pm) which are  not



efficiently removed by phagocytosis may be retained long enough



to cause diseases.



     Fiber retention is probably also determined by the



biological solubility of the fiber.  In the  lung, asbestos and



nonasbestos fibers may undergo physicochemical alterations to

-------
                               191
varying degrees, which could result in fragmentation and



dissolution.  Thus, it would appear that fibers which have low



solubility, i.e., more durable fibers, are potentially more



hazardous because of their long retention at target tissues.   It



has therefore been suggested that ceramic fibers are of



considerable concern because they are relatively durable.  On the



other hand, the lower hazard potential of mineral wool and



fibrous glass might be due to their high solubility.  It should



be noted that the solubility of a fiber is probably largely



determined by fiber chemical characteristics.



     With respect to the role of fiber properties in mediating



biological effects and the development of diseases, fiber size



also appears important for the pathogenesis of malignant



mesothelioma.  In a series of studies by Stanton and co-workers



(Stanton et al., 1977, 1981) involving intrapleural implantation



of various fibrous dusts of diversified chemical, crystallo-



graphic and morphological structures, a correlation was



demonstrated between tumor incidence and the number of fibers



present with lengths greater than 8 ^im and diameters less than



0.25 jjm.  These studies provided the foundation for the  "long,



thin" hypothesis (also known as the Stanton's hypothesis) for the



pathogenesis of mesothelioma.  This hypothesis is supported by



most of the experiments conducted to date by other  investigators,



showing that regardless of fiber characteristics, longer, thinner



fibers are more carcinogenic than short, thick fibers by a



variety of  intracavitary injection methods.  Furthermore, for a



given fiber type, samples containing more long, thin fibers are

-------
                               192
considerably more carcinogenic than those with mostly shorter



thin fibers.



     Bertrand and Pezerat (1980) statistically reanalyzed the



data obtained from Stanton's studies and showed that



carcinogenesis is a result of a continuous function of the aspect



ratio (ratio of fiber length to fiber diameter), and the effects



of fiber length and diameter cannot be separated.  Most of the



fibrous dusts which have been reported to be carcinogenic have



high aspect ratios, and therefore, tend to support this



hypothesis.  However, this hypothesis does not explain the fact



that short, thin attapulgite fibers from various sources with



high aspect ratios are not carcinogenic in animals by either the



intrapleural or intraperitoneal injection method.  The only



attapulgite samples which have been shown to be carcinogenic are



those containing considerable amounts of long, thin fibers.  This



finding appears to argue against the Betrand and Pezerat



hypothesis but does support the role of fiber length in fiber



carcinogenesis.



     The fiber size hypothesis, however, cannot explain the



differential carcinogenic responses observed for various fiber



types with similar fiber size distributions.  A most notable



example is that of erionite, which has comparable fiber size



distribution as that of asbestos, yet is more potent in inducing



mesothelioma in animals fiber per fiber than asbestos by either



inhalation or injection methods.  Thus, it would appear that



other fiber properties such as chemical constitution and/or



surface properties are also important in the development of

-------
                               193
mesothelioma.  However, the fact that synthetic nonfibrous



erionite, which has identical chemical composition as naturally-



occurring fibrous erionite, is not carcinogenic argues against



the direct role of chemical constitution but rather supports the



importance of fiber morphology in fiber carcinogenesis.  It



should also be noted that other nonfibrous particles do not



generally cause mesothelioma.



     Mesothelioma, which is not known to be associated with



cigarette smoking, could be mediated by different mechanism(s)



than those by lung cancer.  It has been postulated that mineral



fibers may behave as complete carcinogens in mesothelial cells



and fibroblasts, the progenitors of mesotheliomas and pleural



sarcomas (Mossman et al., 1983).  This hypothesis is supported by



the observation that erionite, which is the most potent



mesothelioma-inducing  fibrous agent, is highly genotoxic while



asbestos and other fibers such as glass fibers are only weakly



genotoxic.  Based on the limited data base, the genotoxicity of



fibers also appear to  be influenced by fiber dimension.  Thin



fibers (e.g., fibrous  glass) generally show some degree of



clastogenicity and cell transformation, whereas coarse fibers



have little or no activity.  Fiber length also appears to affect



not only the ability of fibers to be phagocytized but also the



ability of intracellular fibers to induce cytogenetic damage and



cell transformation.



     There is also evidence to indicate that fiber size appears



to be important in the induction of lung cancer.  Davis et al.



(1986, 1987) recently  demonstrated that long, thin amosite and

-------
                               194
chrysotile asbestos fibers are more potent than short/ thin


fibers in inducing lung tumors in rats via inhalation.  It would


appear that the different tumorigenic responses between short


versus long fibers could be explained by their differences in the


lung residence time and biological activity.  Long fibers of a


number of mineral fibers have been shown to be retained longer in


the lung than short fibers and, moreover, long fibers are


generally more cytotoxic and genotoxic than short fibers.


     Emphasis has also been placed on the importance of surface


properties of asbestos regarding the cocarcinogenic or


promotional ability of asbestos in the development of lung


cancer.  In humans, a potentiating increase in lung cancer risk


associated with asbestos exposure among cigarette smokers has


been well documented.  A synergistic effect has also been
                 \

demonstrated experimentally in animals exposed by intratracheal


instillation to a combination of asbestos and chemical


carcinogens such as polycyclic hydrocarbons (PAHs) found in


cigarette smoke.  It has been hypothesized that asbestos fibers


might serve as a physical carrier of chemical carcinogens,


providing a means for cellular transport and uptake (Mossman and


Craighead, 1979).  It would appear that while the surface area


would influence the quantity of chemical carcinogen adsorbing to


the fiber, the parameters which actually determine the adsorption


of chemical carcinogens onto the fiber, would most likely be the


specific surface chemical characteristics of the fiber.


     It should be noted that a synergistic effect in the


induction of lung cancer has not yet been observed between

-------
                               195
cigarette smoke and exposure to other asbestiforra fibers (e.g.,



fibrous glass, mineral wool).  Furthermore, unlike asbestos,



erionite does not increase the genotoxicity-of chemical



carcinogens (e.g., benzo[a]pyrene),  and extraction of erionite



with organic solvent to remove potential organic contaminants



does not reduce its in vitro genotoxicity (Brown et al., 1987).



These findings suggest that asbestos fibers might have different



adsorptive properties that are not shared by other fibers, which



enable them to act as promotional agents in the development of



lung cancer.



     Fiber length also appears to be an important determinant in



the development of lung fibrosis.  Available experimental studies



have shown that long, thin asbestos fibers are more fibrogenic



than short, thin fibers in animals via inhalation or



intratracheal instillation.  It has been postulated that the



sequence of cellular events leading to fibrosis probably involves



first the interaction between the fiber and macrophage followed



by a macrophage-fibroblast direct interaction and/or via effects



on an intermediary cell type (NRC, 1984).  If the theory is



correct, i.e., at least with regard to the initial step, then the



observation that long fibers are generally more cytotoxic to



macrophages than are short fibers does indeed support the



importance of fiber length in the development of fibrosis.



However, the fact that long, thin fibers of various types, such



as those of asbestos and fibrous glass, do not necessarily have



comparable fibrogenic activity, and that respirable nonfibrous



particles (e.g., silica) are also fibrogenic indicate that dust

-------
                               196
particle characteristics other than morphology and size may also



play an important role in the induction of fibrosis.



     In conclusion, it is now recognized that the inhalation of



durable fibers of certain diameter and length size range may be



associated with the development of malignant and nonmalignant



lung diseases.  However/ the pathogenic response may vary



depending on the nature of the fibrous dustf including chemical



constitution, solubility, surface charge, surface area, fiber



size and morphology.  Because these properties are probably



interrelated, elucidation of the pathogenicity and mechanisms of



fiber-induced diseases has been difficult.  To increase our



understanding of the health hazards of fibrous dusts, it is



therefore necessary to study the common physical and chemical



properties of these fibers in relation to their biological



activity.  This could lead to modifications of the



physicochemical properties of fibers in order to minimize the



production of adverse effects without affecting the desired



properties necessary for industrial and commercial usage.  Until



the gaps in knowledge of this subject are filled, the fiber size



model remains useful for further experimental studies, as well as



environmental and epidemiological studies.  Additional studies



should also be conducted to elucidate the mechanisms for fiber-



induced cytotoxicity, genotoxicity, and pathogenicity.  Further



studies investigating the relationship between fibrogenicity and



carcinogenicity are also needed.

-------
                               197
VI.  References
Adkins B, McConnell EE.  1985.  Inhalation carcinogenesis study
of wollastonite in,rats.  Northrup Services, Inc. and
NIEHS/National Toxicology Program, Research Triangle Park, North
Carolina.

Artvinli M, Baris YI.  1979.  Malignant mesotheliomas in a small
village in the Anatolian region of Turkey:  An epidemiologic
study.  J Natl Cancer Inst 63:17-22.

Artvinli M, Baris YI.  1982.  Environmental fiber-induced pleuro-
pulmonary diseases in an Anatolian village:  An epidemiologic
study.  Arch Environ Health 37:177-181.

Baris YI, Sahin AA, Ozesmi M, et al.  1978.  An outbreak of
pleural mesothelioma and chronic fibrosing pleurisy in the
village of Karain/Urgup in Anatolia.  Thorax 33:181-191.

Baris YI, Saracci R, Simonato L, Skidmore JW, Artvinli M.
1981.  Malignant mesothelioma and radiological chest
abnormalities in two villages in Central Turkey.  Lancet
1:984-987.

Battelle.  1988.  Review of recent epidemiological investigations
on populations exposed to selected non-asbestos fibers.  Final
report.  Washington, DC:  Prepared for USEPA by Battelle,
Columbus Division under contract No. 68-02-42-46.

Bellmann B, Konig H, Muhle H, Pott, F.  1986.  Chemical
durability of asbestos and of man-made mineral fibres in vivo.
J. Aerosol Sci. 17:341-345.

Bernstein DM, Drew RT, Kuschner M.  1980.  Experimental
approaches for exposure to sized glass fibers.  Env Health
Perspect 34:47-57.

Bernstein DM, Drew RT, Schidlosky G, Kuschner M.  1984.  Patho-
genicity of MMMF and the contrasts with natural fibers.  In:
Biological effects of man-made mineral fibers.  Proceedings of a
WHO/IARC Conference, Volume 2, Copenhagen, Denmark, April 20-22,
1982, Denmark, WHO Regional Office for Europe; pp. 169-195.

Bertrand R, Pezerat H.  1980.  Fibrous glass:  Carcinogenicity
and dimensional characteristics.  In:   Biological effects of
mineral fibers, Wagner JC, Davis W, eds.  IARC Sci Publication No
30; pp. 901-911.

Bignon J, Sebastien P, Gaudichet A, Jaurand MC.  1980.
Biological effects of attapulgite.  In:  Biological effects of
mineral fibers.  Wagner JC, Davis W, eds.  IARC Scientific
Publication 30. pp. 163-181.

-------
                               198
Boman G, Schubert V, Svane B, et al.  1982.  Malignant
mesothelioma in Turkish immigrants residing in Sweden.  Scand J
Work Environ Health 8:108-112.

Brown-GM, Cowie H, Davis JMG, Donaldson K.  1986.   In Vitro
assays for detecting carcinogenic mineral fibers:   a comparison
of two assays and the role of fiber size.  Carcinogesis J7_:1971-
1974.

Brown RC, Chamberlain M, Skidmore JW.  1979a.  In vitro effects
of man-made mineral fibres.  Ann Occup Hyg 22:175-179.

Brown RC, Chamberlain M, Davies R, Gaffen J, Skidmore JW.
1979b.  In vitro biological effects of glass fibers.  J Env
Pathol Toxicol 2:1369-1383.

Brown RC/ Davies R, Road AP.  1987.  Modifications to fibrous
Oregon erionite and their effects on in vitro activity.
Presented at the Symposium on Mineral Fibers in the Nonoccupa-
tional Environment, Lyon, France, Sept 8-10, 1987.

Brune H, Deutsch-Wenzel.  1983.  Lifespan feeding study with
attapulgite in mice.  Biomeasure Inc., Hopkintan,  Massachusetts.

Casey G.  1983.  Sister-chromatid exchange and cell kinetics in
CHO-K1 cells, human fibroblasts and lymphoblastoid cells exposed
in vitro to asbestos and glass fibre.  Mutat Res 16:369-377.

Chamberlain M, Tarmy E.  1977.  Asbestos and glass fibres in
bacterial mutation tests.  Mutat Res 43:159-164.

Chamberlain M, Brown RC.  1978.  The cytotoxic effects of
asbestos and other mineral dust in tissue culture cell lines.
British Journal of Experimental Pathology 59:183-189.

Chamberlain M, Brown RC, Griffiths DM.  1980.  The correlation
between the carcinogenic activities in vivo and the cytopathic
effects in vitro of mineral dusts.  Brown RC, et al., eds.  The
in vitro effects of mineral dusts, pp. 345-349.

Chamberlain M, Davies R, Brown RC, Griffiths DM.  1982.  In vitro
tests for the pathogenicity of mineral dusts.  Ann  Occup Hyg
26:583-592.

Davies R.  1980.  The effects of mineral fibers on macrophages.
In:  Biological effects of mineral fibers.  Wagner JC, ed. Volume
If IARC Scientific Publication number 30, pp 419-425.

Davis JMG.  1974.  Histogenesis and fine structure of peritoneal
tumors produced in animals by injections of asbestos.  J. Natl.
Cancer Inst. 52:1823-1837.

Davis JMG.  1976.  Pathological aspects of the injection of glass
fiber into the pleural and peritoneal cavities of rats and mice.

-------
                               199
In:  Occupational exposure to fibrous glass.  Proceedings of a
Symposium, U.S. Department of Health, Education, and Welfare;
NIOSH Publication No. 76-151; pp. 141-149.

Davis JMG.  1987.  Carcinogenicity of Kevlar Aramid Pulp
following intraperitoneal injection into rats.  Institute of
Occupational Medicine, Edinburg, U.K.  Report No. TM/87/12.

Davis JMG, Gross P, DeTreville RTP.  1970.   "Ferruginous bodies"
in guinea pigs.  Arch Pathol 89:364-373.

Davis JMG, Addison J, Bolton RE, Donaldson K, Jones AD,
Wright A.  1984.  The pathogenic effects of fibrous ceramic
aluminium silicate glass administered to rats by inhalation and
peritoneal injection.  In:  Biological effects of man-made
mineral fibers.  Proceedings of a WHO/IARC Conference, Volume 2,
Copenhagen, Denmark, April 20-22, 1982, WHO Regional Office for
Europe; pp. 303-322.

Davis JMG, Addison J, Bolton RE, Donaldson K, Jones AD, Smith T.
1986.  The pathogenicity of long versus short fibre samples of
amosite asbestos administered to rats by inhalation and intra-
peritoneal injection.  Br J Expl Pathol 67:415-430.

Davis JMG, Jones AD, Smith T.  1987.  Comparisons of the path-
ogenicity of long and short fibers of chrysotile asbestos in
rats.  In Press.

Denizeau F, Marion M, Chevalier G, Cote M.  1985.  Genetoxicity
of dimethynitrosamine in the presence of chrysotile asbestos UICC
B and xonotlite.  Carcinogenesis 6:1815-1817.

DePass LR.  1982.  Evaluation of the dermal carcinogenicity of
four carbon fiber materials in male C3H/HCJ mice.  Bushy Run
Research Center, Export, Pennsylvania.

Dunnigan J, Nadeau D, Paradis D.  1984.  Cytotoxic effects of
aramid fibers on rat pulmonary macrophages:  Comparison with
chrysotile asbestos.  Toxicol Lett 20:277-282.

Engholm G, Englund A, Fletcher T, Hallin N.  1986.  Respira-
tory cancer incidence in Swedish construction workers exposed to
man-made mineral fibers.  Presented at International Symposium on
Man-Made Mineral Fibers in the Working Environment, Copenhagen,
Denmark, October, World Health Organization, Regional Office for
Europe; pp. 311-339.

Enterline P.  1987.  Personal Communication to Dr. Fedor Valic,
London, England.

Enterline P, Marsh G, Esmen N.  1983.  Respiratory disease among
workers exposed to man-made mineral fibers.  Am Rev Resp  Dis
128:1-7.

-------
                               200
Enterline PE, Marsh GM, Henderson V.  1986.  Mortality update of
a cohort of U.S.  man-made mineral fiber workers.   Presented at
the International Symposium on Man-Made Mineral Fibers in the
Working Environment, Copenhagen, Denmark, October 29, 1986.

Enterline PE, Marsh GM, Henderson V, Callahan C.   1987.
Mortality update of a cohort of U.S. man-made mineral fiber
workers.  8EHQ-0887-0553, USEPA.

Feron VJ, Scherrenberg PM, Immel HR, Spit BJ.  1985.  Pulmonary
response of hamsters to fibrous glass:  Chronic effects of
repeated intratracheal instillation with or without
benzo[a]pyrene.  Carcinogenesis 6:1495-1499.

Fevolden S, Moller M.  1978.  Lack of mutagenicity of extracts
from polyethylene (PE) in strains of Salmonella typhimurium.
Mutat Res 53:186-187.  (Abstract).

Forster, H.  1984.  The behaviour of mineral fibers in physiolo-
gical solution.  In:  Biological effects of man-made mineral
fibers.  Proceedings of a WHO/IARC Conference, Volume 2, Copen-
hagen, Denmark, April 20-22, 1982, Denmark, WHO Regional Office
for Europe; pp. 27-39.

Friedberg KD, Ullmer S.  1984.  Studies on the elimination of
dust of MMMF from the rat lung.  In:  Biological effects of man-
made mineral fibres.  Proceedings of a WHO/IARC Conference,
Volume 2, Copenhagen, Denmark, April 20-22, 1982, Denmark, WHO
Regional Office for Europe; pp. 18-26.

Gamble J, Sieber W, Wheeler R, Reger, Hall B.  1985.
Environmental-epidemiologic study of U.S. attapulgite workers.
NIOSH Division of Respiratory Disease Studies.  Presented at the
Sixth International Symposium of Inhaled Particles, Cambridge,
England.

Goldstein B, Rendall REG, Webster I.  1983.  A comparison of the
effects of exposure of baboons to crocidolite and fibrous-glass
dusts.  Environ Res 32:344-359.

Goldstein B, Webster I, Rendall REG.  1984.  Changes produced by
the inhalation of glass fibre in non-human primates.  In:
Biological effects of man made mineral fibre.  Proceedings of a
WHO/IARC Conference, Volume 2, Copenhagen, Denmark, April 20-22,
1982, Denmark, WHO Regional Office for Europe; pp. 273-286.

Gormley IP, Brown GM, Cowie H, Wright A, Davis JMG.  1985.  The
effects of fiber length on the in vitro cytotoxicity of asbestos
samples in three different assay systems.  In:  In Vitro Effects
of Mineral Dusts.  Beck EG, Bignon J, eds.  NATO ASI Series, Vol
G3, Springer-Verlag Berlin Heidelberg, pp. 397-404.

-------
                               201
Griffis LC, Henderson TR, Pickrell JA.  1981.  A method for
determining glass in rat lung after exposure aerosol.  Am Ind Hyg
Assoc J 42:566-569.

Griffis LC, Pickrell JA, Carpenter RL, Wolf RK, McAllen
SJ, Yerkes KL.  1983.  Deposition of crocidolite asbestos and
glass raicrofibers inhaled by the beagle dog.  Am Ind Hyg Assoc J
44:216-222.

Gross P.  1976.  The effects of fibrous glass dust on the lung
of animals.  In:  Occupational exposure to fibrous glass.
Proceedings of a Symposium.  Published by the U.S. Dept. of
Health, Education and Welfare.  HEW Publication No.  (NIOSH)
76-151; pp. 166-179.

Gross P, Braun DC.  1984.  Toxic and biomedical effects of
fibers.  Asbestos, talc, inorganic fibers, man-made vitreous
fibers, and organic fibers.  Chap. 4. Inorganic fibers other than
asbestos and man-made vitreous fibers, Noyes Publications, Park
Ridge, NJ; pp. 132-142.

Gross P, Kaschak M, Tolker EB, Fabyak MA, DeTreville RTF.
1970.  The pulmonary reaction to high concentrations of fibrous
glass dust.  Archives of Environmental Health 20:696-704.

Hammad YY.  1984.  Deposition and elimination of MMMF.
In:  Biological effects of man-made mineral fibres.  Pro-
ceedings of a WHO/IARC Conference, Volume 2, Copenhagen,
Denmark, 20-22 April, 1982, Denmark, WHO Regional Office
for Europe; pp. 126-142.

Hammad Y, Simmons W, Abdel-Kader H, Reynolds C, Weill H.  1985.
Effect of chemical composition on pulmonary clearance
of man-made mineral fibers.  Presented at the Sixth International
Symposium of Inhaled Particles, Cambridge, England.

Hanke WE, Sepulveda MJ, Watson A, Jankovic J.  1984.  Respiratory
morbidity in wollastonite workers.  Br J Ind Med 41:474-479.

Harris RL, Eraser DA.  1976.  A model for deposition of fibers in
the human respiratory system.  Amer. Ind. Hyg Ass J. 37:73-89.

Harris RL, Timbrell V.  1977.  The influence of fiber shape and
lung deposition-mathematical estimates.  In:  Walton, Ed. Inhaled
Particles IV, Part 1.  Pergamon Press, Oxford, pp. 75-89.

Harvey G, Page M, Dumas L. 1984.  Binding of environmental
carcinogens to asbestos and mineral fibres.  Br J Med
41:396-400.

Haugen A, Schafer PW, Lechner JF, Stoner GD, Trump BF,
Harris CC.  1982.  Cellular ingestion, toxic effects,
and lesions observed in human bronchial epithelial tissue

-------
                               202
and cells cultured with asbestos and glass fibers.   Int J Cancer
30:265-272.

Hefner RE, Gehring P.  1975.  A comparison of the relative
rates of hemolysis induced by various fibrogenic and non-
fibrogenic particles with washed rat erythrocytes in vitro.   Am
Ind Hyg 1:734-740.

Hesterberg 1, Barrett JC.  1984.  Dependence of asbestos-
and mineral dust-induced transformation of mammalian cells in
culture on fiber dimension.  Cancer Res 44:2170-2180.

Holt PF.  1982.  Submicron carbon dust in the guinea
pig lung.  Environ Res 28:434-442.

Holt PF, Home M.  1978.  Dust from carbon fibre.  Environ Res
17:276-283.

Huuskonnen MS, Tossavainen A, Koskinen H.  1983.  Wollasto-
nite exposure and lung fibrosis.  Environ Res 30:291-304.

Huuskonnen MS, Tossavainen A, Koskinen H, et al.  1984.
Respiratory morbidity of quarry workers exposed to wollasto-
nite.  In:  Occupational Lung Disease.  Bernard J,  Gee L, Morgan
G, Brooks SM, eds. New York, Raven Press.

IARC.  1979.  IARC monographs on the evaluation of carcinogenic
risk of chemicals to man.  Vol 19, Lyon, France.

ICF.  1986.  Durable fiber exposure assessment.  Washington, DC.
Prepared for U.S. EPA Office of Toxic Substances.

Jaurand MC, Bignon J.  1979.  Interaction of fibres with human
and animal cells.  Thorax 341:694.

Johnson NF, Wagner JC.  1980.  A study by electron micro-
scopy of the effects of chrysotile and man-made mineral fibres on
rat lungs.  In:  Biological effects of mineral fibres.  Wagner,
JC, ed.  Volume 1, Lyon, IARC Scientific Publication 30, pp. 293-
303.

Johnson NF, Griffiths DM, Hill RJ.  1984a.  Size distribution
following long-term inhalation on of MMMF.  In:  Biological
effects of man-made mineral fibers. Proceedings of a WHO/IARC
Conference, Volume 2, Copenhagen, Denmark, April 20-22, 1982,
Denmark, WHO Regional Office for Europe; pp. 102-125.

Johnson NF, Edwards RE, Munday DE, Rowe N, Wagner JC.  1984b.
Pluripotential nature of mesotheliomata induced by inhalation of
erionite in rats.  Brit J Exp Path  65:377-388.

Jones HD, Jones TR, Lyle WH.  1982.  Carbon fibers:  Results
of a survey of process workers and their environment in a factory
producing continuous filament.  Ann Occup Hyg 26:861-868.

-------
                               203
Kelsey KT, Yano E, Liber HL, Little JB.  1986.  The in vitro
genetic effects of fibrous and crocidolite asbestos.  Br J Cancer
54:107-114.

Klingholz R, Steinkopf B.  1984.  The reactions of MMMF in a
physiological model fluid and in water.  In:  Biological effects
of man-made mineral fibers.  Proceedings of a WHO/IARC Conference
Volume 2, Copenhagen, Denmark, April 20-22, 1982, Denmark, WHO
Regional Office for Europe;  pp. 60-86.

Lafuma J, Morin M, Poncy JL, et al.  1980.  Mesothelioma induced
by intrapleural injection of different types of fibers in rats;
synergistic effects of other carcinogens.  In:  Biological
effects of mineral fibres.  Wagner JC, ed.  Volume 1, IARC
Scientific Publication 30, pp. 311-320.

Le Bouffant L, Henin JP, Martin JC, Normand C, Tichoux G, Trolard
E.  1984.  Distribution of inhaled MMMF in the rat lung - long
term effects.  In:  Biological effects of man-made mineral
fibers.  Proceedings of a WHO/IARC Conference, Volume 2,
Copenhagen, Denmark, April 20-22, 1982, Denmark, WHO Regional
Office for Europe; pp. 143-169.

Lee KP, Barras CE, Griffith FD, Waritz RS.  1979.  Pulmonary
response to glass fiber by inhalation exposure.  Lab Invest
40:123-133.

Lee KP, Barras CE, Griffith FD, Waritz RS, Lapin CA.  1981.
Comparative pulmonary responses to inhaled inorganic fibers with
asbestos and fiberglass.  Environ Res 24:167-191.

Lee KP, Kelly DP, Kennedy GL, Jr.  1983.  Pulmonary response
to inhaled Kevlar aramid synthetic fibers in rats.  Tox Appl
Pharm 71:242-253.

Leineweber JP.  1984.  Solubility of fiber in vitro and in
vivo.  In:  Biological effects of man-made mineral fibres.  Pro-
ceedings of a WHO/IARC Conference, Volume 2, Copenhagen, Denmark,
April 20-22, 1982, Denmark, WHO Regional Office for Europe; pp.
87-101.

Lipkin LE.  1980.  Cellular effects of asbestos and other fibers:
Correlations with in vivo induction of pleural sarcoma.  Environ
Health Perspect 34:91-102.

Lipkin MD, Lewis E.  1985.  Failure of attapulgite to produce
tumors:  Prediction of this result by in vitro cytotoxicity
test.  In:  In vitro effects of mineral dust.  Beck EG, Bignon J,
eds.  Ecological Sciences, Vol 3, Berlin, Springer Verlag, p.
539.

Litton Bionetics.  1974.  Mutagenic evaluation of compound FDA
71-41.  Calcium silicate.  PB-245 457.

-------
                               204
Maltoni C, Minardi F, Morisi L.  1982a.  Pleural mesotheliomas
in Sprague-Dawley rats by erionite:  First experimental evidence.
Environ Res 29:238-244.

Maltoni C, Minardi F, -Morisi L.  1982b.  The relevance of the
experimental approach in the assessment of the oncogenic risks
from fibrous and non-fibrous particles.  The ongoing project of
the Bologna Institute of Oncology.  Med Lavoro 4:394-407.

Maltoni C, Minardi F.  1987.  Recent results of carcinogen-
icity bioassays of fibers and other particulated materials.
Presented at Symposium on Mineral Fibers in the Nonoccupational
Environment/ Lyon, France, September 8-10, 1987.

MB Research Laboratories, Inc.  1980.  Project Number 78-2806.
Twenty one-month observation after single intratracheal insuff-
lation.

McConnell EE, 1988.  Personal verbal communication from EE
McConnell to J.A. Moore on April 26, 1988.

McConnell EE, Wagner JC, Skidmore JW, Moore JA.  1984.
A comparative study of the fibrogenic and carcinogenic
effects of UICC Canadian chrysotile B asbestos and glass
microfiber (JM 100).  In:  Biological effects of man-made
mineral fibers.  Proceedings of a WHO/IARC Conference,
Volume 2, Copenhagen, Denmark, April 20-22, 1982,
1982, WHO Regional Office for Eurpoe; pp. 243-252.

Mitchell RI, Donofrio DJ, Moorman WJ.  1986.  Chronic inhal-
ation toxicity of fibrous glass in rats and monkeys.  J Amer
College Toxicol 5:545-575.

Moatamed F, Rom W, Casey K, Archer V.  1981.  Early response to
injected fibrous erionite in the mouse peritoneum and rat lung.
Respir Dis 123:26.

Mohr U, Pott F, Vonnahmo FJ.  1984.  Morphological aspects
of mesotheliomas after intratracheal instillation of fibrous
dusts in Syrian golden hamsters.  Exp Path 26:179-183.

Monchaux G, Bignon J, Jaurand MC, et al.  1981.  Mesotheliomas in
rats following inoculation with acid-leached chrysolite asbestos
and other mineral fibres.  Carcinogenesis 2:229-236.

Monchaux G, Bignon J, Hirsch A, Sebastien P.  1982.
Translocation of mineral fibers through the respiratory system
after injection into the pleural cavity of rats.  Ann Occup Hyg
26:309-318.

Morgan A.  1979.  Fiber dimensions:  Their significance in the
deposition and clearance of inhaled fibrous dusts.  In:  Dusts
and disease:  Lehman R, Dement M, eds.  Park Forest South, IL,
Pathotox Publishers, Inc., pp. 87-96.

-------
                               205
Morgan A, Holmes A.  1984a.  The deposition of MMMF in
the respiratory tract of the rat, their subsequent clearance,
solubility in vivo and protein coating.  In:  Biological effects
of man-made mineral fibers.  Proceedings of a WHO/IARC
Conference/ Volume 2, Copenhagen, Denmark, April 20-22, 1982,
Denmark, WHO Regional Office for Europe; pp. 1-17.

Morgan A, Holmes A.  1984b.  Solubility of rockwool fibres in
vivo and the formation of pseudo-asbestos bodies.  Ann Occup  Hyg
28:307-314.

Morgan A, Black A, Evans N, Holmes A, Pitchard JN.  1980.
Deposition of sized glass fibers in the respiratory tract of the
rat.  Ann Occup Hyg 23:353-366.

Morgan A, Holmes A, Davidson W.  1982.  Clearance of sized glass
fibers from the rat lung and solubility in vivo.  Ann Occup
Hyg 25:317-331.

Morriset Y, Pan A, Jegier Z.  1979.  Effect of styrene and fiber-
glass on small airways on mice.  J Tox Env Health 5:943-956.

Morrison DG, Daniel J, Lynd FT, et al.  1981.  Retinyl palmitate
and ascorbic acid inhibit pulmonary neoplasms in mice exposed to
fiberglass dust.  Nutrition and Cancer 3:81-85.

Mossman BT, Craighead JE.  1979.  Use of the hamster tracheal
organ cultures for assessing the cocarcinogenic effects of
inorganic particulates on the respiratory epithelium.  Prog Exptl
Tumor Res 24:37-47.

Mossman B, Light W, Wei E.  1983.  Asbestos:  Mechanisms
of toxicity and carcinogenicity in the respiratory tract.
Ann Rev Pharmacol Toxicol 23:595-615.

Nadeau D, Paradis D, Gaudreau A, Pele JP, Calvert R.  1983.
Biological evaluation of various natural and man-made mineral
fibers:  Cytotoxicity, hemolytic activity and chemilunescence
study.  Environ Health Perspec 51:374  (Abstract).

Neugebauer R, Helbing G, Wolter D, Mohr W, Gistinger, G.  1983.
The body reaction to carbon fibre particles implanted into the
medullary space of rabbits.  Biomaterials 2:182-184.

NRC.  1984.  National Research Council.  Abestiform -
nonoccupational health risks.  Committee on Nonoccupational
Health Risks of Asbestiform fibers. Board of Toxicology and
Environmental Health Hazards, Commission on Life Sciences.
National Academy Press.  Washington D.C.

Olsen J, Jensen 0.  1984.  Cancer Incidence among employees in
one mineral wool production plant in Denmark.  Scand. J. Work
Environ. Health 10:17-24.

-------
                               206
Oshimura M, Hesterberg T, Tsutsui T, Barrett JC.   1984.
Correlation of asbestos-induced cytogenetic effects
with cell transformation of Syrian hamster embryo cells
in culture.  Cancer Res 44:5017-5022.

Ottolenghi AC, Joseph LB, Newman HAI, Stephens RE.  1983.
Interaction of erythrocyte membrane with particulates.  Environ
Health Perspectives 51:523-256.

Owen P, Glaister JR, Ballantyne B, Clary JJ.  1986.  Sub-
chronic Inhalation Toxicology of Carbon fibers. J Occup
Med 28:373-376.

Ozesmi M, Patiroglu TE, Hillerdal G, Ozesmi C.  1985.
Peritoneal  mesothelioma and malignant lymphoma in mice  caused by
fibrous zeolite.  Br J Ind Med 42:746-749.

Pailes WH, Resnick JH, Castranova V.  1984.  Relative effects of
asbestos and wollastonite on alveolar macrophages.  J Toxicol
Environ Health 14:497-510.

Palekar LD, Coffin DL.  1986.  Significance of mineral fibers
induction and V79 cytotoxicity.  Toxicologist 6:6.

Palekar LD, Coffin DL.  1987.  Significance of mineral fibers in
the determination of V79 cytotoxicity.  Environ Res.  In press.

Palekar LD, Brown BG, Coffin DL.  1985.  Correlation between in
vivo tumorigenesis and in vitro cytotoxicity in CHO and  V79 cells
after exposure to mineral fibers.  In:  Short term bioassays in
the analysis of complex environmental mixtures.  Waters  MD,
Sandhu SS, Lewtas J, Claxton L, Strauss G, Nesnow S, eds.  New
York:  Plenum Publishing Corp., pp. 155-169.

Palekar LD, Eyre JF, Most BM, Coffin DL.  1987.  Metaphase and
anaphase analysis of V79 cells exposed to erionite, UICC
chrysotile and UICC crocidolite.  Carcinogenesis 8:555-560.

Parnell M.  1987.  Current application and studies on health
effects of carbon fibers - military point of view.  Presented at
the Toxicology Forum, Aspen, CO., July 13-17, 1987.

Pickrell JA, Hill JO, Carpenter RL, Hahn FF, Rebar AH.
1983.  In vitro and in vivo response after exposure to man-made
mineral and asbestos insulation fibers.  Am Ind Hyg Assoc J
44:557-561.

Pigott GH, Ishmael J.  1981.  An assessment of the fibrogenic
potential of two refractory fibers by intraperitoneal injections
in rats.  Toxicol Lett 8:153-163.

-------
                               207
Pigott GH, Gaskell BA, Ishmael J.   1981.  Effects of long term
inhalation of alumina fibers in rats.   Br J Exp Path 62:323-331.

Poole A, Brown R, Fleming G.  1983a.   Study of the cell
transforming ability of amosite and crocidolite asbestos and the
ability to induce changes in the metabolism and macromolecular
binding of benzo(a)pyrene in C3H10T1/2 cells.   Environ Health
Perspec 51:319-324.

Poole A, Brown R, Turver C, Skidmore J, Griffiths D.  1983b.
In vitro genotoxic activities of fibrous erionite.  Br J Cancer
47:697-705.

Pott F, Friedrichs KH.  1972.  Tumoren der Ratte nach i.p.
injektion faserforminger staube.  Naturwissenschaften 59:318.

Pott F, Huth F and Friedrichs KH.   1974.  Tumorigenic effects of
fibrous dust in experimental animals.   Environ Health Perspec
9:313-315.

Pott F, Huth F, Friedrichs KH.  1976.   Results of animal
carcinogenesis studies after implantation of fibrous glass  and
their implications regarding human exposure.  In:  Occupational
exposure to fibrous glass.  Proceedings of a Symposium.  U.S.
Dept. of Health, Education and Welfare.  NIOSH Publication  No.
76-151.  pp. 183-191.

Pott F, Huth F, Spurny K.  1980.  Tumor induction after
intraperitoneal injection of fibrous dusts.  In:  Biological
effects of mineral fibres.  Wagner JC, ed.  Volume 1, Lyon, IARC
Scientific Publication 30, pp. 337-342.

Pott F, Schlipkoter HW, Ziem U, Spurny K, Huth F.  1984.  New
results from implantation experiments with mineral fibers.   In:
Biological effects of man-made mineral fibers.  Proceedings of a
WHO/IARC Conference, Volume 2, Copenhagen, Denmark, April 20-22,
1982, Copenhagen, WHO Regional Office for Europe; pp. 286-302.

Pott F, Matscheck A, Ziem U, Huhle H,  Huth F.   1985.  Animal
experiments with chemically treated fibers.  Abstract for the
Sixth International Symposium on Inhaled Particles. Cambridge,
England.

Pott F, Ziem U, Reiffer FJ, Huth F, Ernst H, Mohr U.  1987a.
Carcinogenicity studies in fibres, metal compounds and some other
dusts in rats.  Exp. Pathol. 32:129-152.

Pott F, Rolla M, Ziem U, et al.  1987b.  Carcinogenecity studies
on natural and man-made fibers with the intraperitoneal test in
rats.  Presented at the Symposium on Mineral Fibers in the
Nonoccupational Environment, Lyon, France, Sept 8-10, 1987.

-------
                               208
Potts WJ, Lederer TS,  Gehring PJ.  1978.  Hemolysis of washed rat
erythrocytes in vitro  by dust particles.  Am Ind Hyg Assoc J
39:497-502.

Reinhardt CF.  1980.  Toxicology of araraid fibers.  Proceedings
of the National Workshop on Substitutes for Asbestos.  EPA-560
13-80-001, Nov. 1980.   Sponsored by U.S. EPA, CPSC, and IRLG.

Reinhardt CF.  1986.  8EHQ-0986-0550.  Two-year inhalation study
in rats with Kevlar® Fibrils.  USEPA.

Reiss B, Millette JR,  Williams GM.  1980.  The activity of
environmental samples  in a cell culture test for asbestos
toxicity.  Environ Res 22:315.

Renier A, Fleury J, Monchaux G, Nebut M, Bignon J, Jaurand MC.
1987.  In vivo and in  vitro studies of the toxicity of one
attpulgite.  Presented at the Symposium on Mineral Fibers in the
Nonoccupational Environment, Lyon, France, September 8-10, 1987.

Richards RJ, Hunt J.  1983.  Reaction of mineral dusts with
primary lung fibroblast cultures.  Environ Health Perspec
51:61-65.

Richards RJ, Jacoby F.  1976.  Light microscope studies on the
effects of chrysotile  asbestos and fiber glass on the morphology
and reticulin formation of cultured lung fibroblasts.  Env Res
11:112-121.

Rohl A, Langer A, Moncure G, Selikoff I, Fischbein A.  1982.
Endemic pleural disease associated with exposure to mixed fibrous
dust in Turkey.  Science 216:518-520.

Rom WN, Casey KR, Parry WT, Mjaatvedt CH, Moatamed F.  1983.
Health implications of natural fibrous zeolites for the
intermountain west.  Environ Res 30:1-8.

Rossiter CE.  1982.  The MRC and NIEHS Studies:  A further
analysis.  In:  Biological effects of man-made mineral fibers.
Report on a WHO/IARC Meeting, Copenhagen, Denmark, April 20-22,
1982.  EURO Reports and Studies No. 81, pp. 129.

Rowhani F, Hammand YY.  1984.  Lobar deposition of fiber in the
rat.  Am Ind Hyg Assoc J 45:436-439.

Schepers GWH.  1955.  The biological action of glass wool.
A.M.A. Archives of Industrial Health 12:280-287.

Schepers GWH.  1959a.   Pulmonary histologic reactions to inhaled
fibers glass plastic dust.  American Journal of Pathology
35:1169-1189.

-------
                               209
Schepers GWH.  1959b.  The pulmonary reaction to sheet fiber
glass plastic dust.  American Industrial Hygiene Association
Journal 20:73-81.

Schepers GWH.  1961.  The pathogenicity of glass reinforced
plastics.  Archives of Environmental Health 5:278-299.

Schepers GWH.  1976.  The comparative pathogenicity of inhaled
fibrous glass dust.  In:  Occupational exposure to fibrous glass.
Proceedings of a Symposium.  Published by the U.S. Dept.  of
Health, Education and Welfare.  HEW Publication No.
(NIOSH) 76-151; pp. 265-341.

Schepers GWH, Delahunt AB.  1955.  An experimental study of the
effects of glass wool on animal lungs.  AMA Archives of
Industrial Health 12:276-280.

Sebastien P, Gaudichet A, Bignon J, Baris YI.  1981.  Zeolite
bodies in human lungs from Turkey.  Lab Invest 44:420-425.

Shannon HS, Jamieson E, Julian JA, Muir DCF, Walsh C.  1986.
Mortality experience of glass fibre workers — extended follow
up.  Presented at the International Symposium on Man-made Mineral
Fibers in the Working Environment, Cophengagen, Denmark, October
29, 1986.

Shasby DM, Peterson M, Hodous TK, Boehlecke BA.  1977.  Medical
survey of workers at the Interpace Corporation, Willsboro, NY.
Final Report, Nov. 4, 1977.  National Institute for Occupational
Safety and Health, Morgantown, WV.

Shasby DM, Peterson M, Hodous TK, Boehlecke B, Merchant J.
1979.  Respiratory morbidity of workers exposed to wollastonite
through mining and milling.  In:  Dust and disease.  Lemen R,
Dement JM, eds.  Pathotox, Park Forest South, IL.

Simonato L, Fletcher A, Saracci R, Charnay N, Esteve J.  1985.
Historical cohort study on man-made mineral fibre  (MMMF)
production workers in seven European countries.  Final report to
the Joint European Medical Research Board (JEMRB).  Internatl
Agency for Res on Cancer  (IARC), Lyon, France.

Simonato L, Fletcher AC, Cherrie J, et al.  1986a.  The man-made
mineral fiber European historical cohort study — extension of
the follow-up.  Scand J Work Environ Health 12:34-47.

Simonato L, Fletcher AC, Cherrie J, et al.  1986b.  The man-made
mineral fibers (MMMF) European historical cohort study —
extension of the follow-up.  Presented at the International
Symposium of Man-made Mineral Fibers in the Working Environment,
Copenhagen, Denmark, October 29.

-------
                               210
Sincock A.  1977.   Preliminary studies of the in vitro cellular
effects of asbestos and fine glass dusts.  In:  Origins of human
cancer, Book B.  Cold Spring Harbor Laboratory, pp. 941-954.

Sincock A, Seabright M. 1975.  Induction of- chromosome changes in
Chinese hamster cells by exposure to asbestos fibres.  Nature
257:56-58.

Sincock A, Delhanty J, Casey G.  1982.  A comparison of the
cytogenetic response to asbestos and glass fibre in Chinese
hamster and human cell lines.  Demonstration of growth inhibition
of primary human fibroblasts.  Mutat Res 101:257-268.
Skaug V, Gylseth B.  1983.  Hemolytic activity of five different
calcium silicates.  Environ Health Perspect 51:195-203.

Smith WE, Hurbert PD, Sobel HJ.  1980.  Dimension of fibers in
relation to biological activity.  In:  Biological effects of
mineral fibers.  Wagner JC, ed.  Volume 1, Lyon, France, IARC
Scientific Publication 30, pp. 357-360.

Smith DM, Ortiz LW, Archuleta RF.  1984.  Long term exposure of
Syrian hamsters and Osborne-Mendel rats to aerosolized 0.45 urn
mean diameter fibrous glass.  In:  Biological effects of man-made
mineral fibers.  Proceedings of a WHO/IARC Conference, Volume 2,
Copenhagen, Denmark, April 20-22, 1982, Copenhagen,  WHO Regional
Office for Europe; pp. 253-273.

Smith DM, Ortiz LW, Archuleta RF, Johnson NF.  1986.  8EHQ-1186-
0553 Suppl.  Long-term health effects in hamsters and rats
exposed chronically to man-made vitreous fibers.  Office of Toxic
Substances, U.S. Environmental Protection Agency, Washington DC.

Sors H, Gaudichet A, Sebastien P, Bignon J, Even P.  1979.
Lung fibrosis after inhalation of fibrous attapulgite.
Thorax 34:695  (Abstract).

Spurny KR, 1983a.   Measurement and analysis of chemically changed
mineral fibers after experiments in vitro and in vivo.   Env
Health Perspect 51:343-355.

Spurny KR, 1983b.   Natural fibrous zeolites and their
carcinogenicity - a review.  Sci Total Environ 30:147-166.

Spurny KR, Pott F, Stober W, Helmutopiela JS, Weiss G.  1983.
On the chemical changes of asbestos fibers and MMMF  in biological
residence and the environment:  Part 1. Am Ind Hyg Assoc J
44:833-845.

-------
                               211
Spurny KR, Opiela H, Schoerman J, Weiss G.  1984.  Methods and
investigations relative to the stability of mineral fibers in
vitro and vivo.  Staub-Reinhalt. 44:169-180.

Stanton MF, Wrench C.  1972?  Mechanisms of mesothelioma
induction with asbestos and fibrous glass.  J Natl Cancer Inst
48:797-821.

Stanton MF, Layard M, Tegeris A, et al.  1977.  Carcinogenicity
of fibrous glass:  Pleural response in the rat in relation to
fiber dimension.  J Natl Cancer Inst 58:587-603.

Stanton MF, Layard M, Tegeris A, et al.  1981.  Relation of
particle dimension to carcinogencity in amphibole asbestoses and
other fibrous minerals.  J Natl Cancer Inst 67:959-975.

Styles JA, Wilson J.  1973.  Comparison between in vitro toxicity
of polymer and mineral dusts and their fibrogenicity.
Ann Occup Hyg 16:214-250.

Styles JA, Wilson J.  1976.  Comparison between in vitro toxicity
of two novel fibrous mineral dusts and their tissue reactions in
vivo.  Ann Occup Hyg 19:63-68.

Suzuki Y.  1982.  Carcinogenic and fibrogenic effects of
zeolites:  Preliminary observations.  Environ Res 27:433-445.

Suzuki Y, Kohyama N.  1984.  Malignant mesothelioma induced by
asbestos and zeolite in the mouse peritoneal cavity.  Environ
Res 35:277-292.

Swensson A.  1979.  Experimentalla undersoknongar over den
fibrogenetiska effekten av alipdamm fran kolfiber armerad
plast.  National Board of Occupational Safety and Health
(Sweden).  Arbete Och Halsa 6:15 (As reported in Gross and Braun,
1984)

Tayton K, Phillips G, Ralis Z.  1982.  Long-term effects of
carbon fibers on soft tissues.  J Bone Joint Surg 64-8:112-114.

Tilkes F, Beck EG.  1980.  Comparison of length-dependent
cytotoxicity of inhalable asbestos and man made mineral fibres.
In:  Biological effects of mineral fibres.  Wagner JC, ed. Lyon,
France, International Agency for Research on Cancer.  IARC Wagner
JC, Scientific Publication No. 30. pp. 475-483.

Tilkes F, Beck EG.  1983a.  Macrophage functions after exposure
to mineral fibers.  Environ Health Perspec 51:67-72.

-------
                               212
Tilkes F, Beck EG.  1983b.  Influence of well-defined mineral
fibres of proliferating cells.  Environ Health Perspec 51:275-
279.

Troitskaya NA~, Velichkovskii BT, Kogan FM, El'nichnykh LN.
1984.  Comparative fibrogenicity of carbon fibers and asbestos.
Gig Sanit 6:18-20.  (In Russian) (As reported in CAS Online
CA101:145322)

Union Carbide.  1983a.  8EHQ-0483-0423 Supplement.  (1) Mutagenic
potential of benzene extracts from CF and PAN carbon fiber
fragments (memorandum from B. Ballantyne, Corporate Director of
Applied Toxicology, Union Carbide Corporation); (2) pitch-based
carbon fibers (benzene extract) Salmonella/microsome (Ames)
bacterial mutagenicity assay by W.C. Hengler and R.S. Slesinski
(10/21/82);  (3) material safety data sheet- carbon fibers.
Washington, DC:  Office of Toxic Substances, U.S. Environmental
Protection Agency.

Union Carbide.  1983b.  8EHQ-0483-0423 Supplement.  Polyacrylo-
itrile (PAN)- based carbon fibers (benzene extract)
Salmonella/microsome (AMES) bacterial mutagenicity assay by
Hengler WC, Slesinki RS, et al.  (10/21/82)  Washington, DC:
Office of Toxic Substances, U.S. Environmental Protection Agency.

Union Carbide.  1983c.  Union Carbide Corporation.  8EHQ-1285-
0423 FLWP.  Benzene extract of pitch-based carbon fibers in
vitro  mutagenesis studies:  3-test battery.  By Slesinski RS, et
al.  (1/14/83) Washington, DC:  Office of Toxic Substances, U.S.
Environmental Protection Agency.

Union Carbide.  1983d.  Union Carbide Corporation.  8EHQ-0483-
0423 Supp. Benzene extract of polyacrylontrile (PAN)-based carbon
fibers in vitro mutagenesis studies:  3-test battery by Slesinski
RS, et al. (12/22/82)  Washington DC:  Office of Toxic
Substances, U.S. Environmental Protection Agency.

USEPA.  1986.  Guidelines for carcinogenic risk assessment.
Federal Register 51:33992-34003.

Vallyathan V, Robinson V, Reasor M, Stettler L.  1984.
Wollastonite and asbestos cytotoxicity.  Ann Rev Resp Dis
129:A178.

Wagner JC.   1982.  Health hazards of substitutes.  In:
Proceedings of the World Symposium of Asbestos.  May 25-27.
Montreal, Canada, pp. 244-266.

Wagner JC, Berry G, Timbrell V.  1973.  Mesothelioma in rats
after inoculation with asbestos and other minerals.  Br J Cancer
28:173-185.

-------
                               213
Wagner JC, Berry GB, Skidmore JW.  1976.  Studies on the
carcinogenic effects of fiberglass of different diameters
following intrapleural inoculation in experimental animals.
In:  Occupational exposure to fibrous glass.   Proceedings of a
Symposium.  Washington, DC.  US Dept. Health, Education and
Welfare.  NIOSH Publication No. 76-151.  pp.  193-197.

Wagner JC, Berry GB, Hiu RJ, Munday DE, Skidmore JW.  1984.
Animal experiments with MMM(V) fiber-effects  on inhalation and
intrapleural inoculation in rats.  In:  Biological effects of
man-made mineral fibers.  Proceedings of a WHO/IARC Conference,
Volume 2, Copenhagen, Denmark, April 20-22, 1982, Copenhagen,
WHO Regional Office for Europe; pp. 209-233.

Wagner JC, Skidmore JW, Hill RJ, Griffiths DM.  1985:  Erionite
exposure and mesotheliomas in rats.  Br J Cancer.  51:727-730.

Wagner JC, Griffiths DM and Munday DE.  1987.  Experimental
studies with palygorskite dusts.  Br. J. Ind. Med. 44:749-763.

Warheit DB, Hill LH, Brody AR.  1984.   In vitro effects of
crocidolite asbestos and wollastonite on pulmonary macrophages
and serum complement.  Scanning Electron Microscopy 2:919-926.
Waxweiller RJ, Zumwalde RD, Ness GO, Brown DP.  1985.  Mortality
among persons mining and milling attapulgite clay.  Division of
Surveillance, Hazard Evaluations and Field Studies and Division
of Standards Development and Technology Transfer, National
Institute for Occupational Safety and Health.  Presented at
International Symposium on Inhaled Particles, Cambridge, England.

Weill H, Hughes JM, Hammad YY, et al.  1983.   Respiratory health
in workers exposed to man-made vitreous fibers.  Am Rev Resp Dis
28:104-112.

Woodworth GD, Mossman BT, Craighead JE.  1983.  Induction of
squamous metaplasia in organ cultures of hamster trachea by
naturally occurring and synthetic fibers.  Cancer Res 43:4906-
4912.

Wright G, Kuschner M.  1976.  The effects of intratracheal
instillation of glass fiber of varying sizes in guinea pigs.
Proceedings of a symposium, U.S. Department of Health, Education,
and Welfare.  NIOSH Publication No. 76-151; pp. 151-156.

Wright GW, Kuschner M.  1977.  The influence of varying lengths
of glass and asbestos fibers on tissue response in guinea pigs.
Inhaled particles.  Watton WH, ed.  New York:  Pergamon Press,
pp. 455-474.

-------
                               214
Wright WE, Rom WN/ Moatamed F.  1983.  Characterization of
zeolite fiber sizes using scanning electron microscopy.  Arch
Environ Health 38:99-103.

Zumwalde, R.  1977.  Industrial hygiene study of the Englehard
Minerals and Chemicals Corporation, Attapulgus, Georgia.  NIOSH
Report, April 29, 1977.

VII.  APPENDIX

-------
TABLE I.   SUMMARY  OF ANIMAL  STUDIES ON FIBROUS GLASS
Fiber Type
Glass Moot
Coated Glass
Fibers (calcium
carbonate)
Coated Glass
Fibers (calcium
sulfate)
Glass Fibers
(uncoated)
Coated Glass
Fibers (Phenol
formaldehyde
resin)
Coated Glass
Fibers (starch
binder)
Fiber Dimension
i
Species
20) <2 fat ' Guinea Pigs
diameter; average Rats
diameter near
5 fm; 70K >5 um long
Not specified
Not specified
Average d 1 ameter
0. 5 jjn; average
length 10 urn
(5-20 urn)
As Above
As Above
Monkeys
Guinea Pigs
Rabbits
Rats
Guinea Pigs
Rabbits
Rats
Rats
Hamsters
Rats
Hamsters
Rats
Hamsters
Number
of
Animals
too
50
5
40
18
30
63
18
20
30
30
30
30
30
30
Route Dose/ Frequency/
of Admin 1- Concentra- Treatment
stratlon tlon Duration
Inhalation 0.145 mg/m3 44 mo
0.03 mg/m3 28 mo
Inhalation 4.6 mg/m 8 mo
24 mo
18 mo
18 mo
Inhalation 3.8 mg/m3 15 mo
19 mo
9 mo
Inhalation 135 ng/m3 2 yr
Inhalation 106 mg/m3 2 yr
Inhalation 1 13 mg/m3 2 yr
Duration Results Remarks References
of Study
44 mo Little evidence of dust Poor animal survival; Ion Schepers (1955), I959a.
28 mo reaction In any species; No concentration level 1959b, 1961)
fibres Is; no pulmonary tumors Schepers and Delahunt
(1955); Schepers (1976)
8 mo As above No Information on dust cloud As above-
24 mo
18 mo
18 mo
15 mo As above No Information on dust cloud As above
15 mo
9mo «
Ul
Llfespan No flbrosts or tumors; hlsto- Poor survival Gross et al. (1970)
logical changes limited to Gross (1976)
those seen from Inert dusts
Llfespan As above As above
Llfespan As above As above

-------
                                                             TABLE I.  SUMMARY OF ANIMAL STUDIES ON FIBROUS_GLASS (CONTINUED)
Fiber Type
Fibrous Glass


Fiberglass
Insulation



Fiberglass






Fiber Dimension
801 3



Species
Male Albino
Mice (Charles
River, CDR-t)
Male A-straln
Mice



Male
Charles River
Sprague-Oawley
Rats

Male Albino
Guinea Pigs
Number Route
of of Admin 1-
Anlmals stratlon
20 Inhalation


12 Mixed In
bedding
material
(Inhalation)

46 Inhalation




32 Inhalation

Dose/
Concentra-
tion
1070
f 1 bers/mL

300 mg/
100 mg
bedding


400 mg/m3
(700
fibers/Hi)


400 mg/m3
(700
Frequency/
Treatment Duration Results
Duration of Study
6 veeks 6 weeks No lung damage; no f Ibrosls


30 days 90 days Bronchogenlc tumor (3/12);
septal cell tumor (2/12)



90 days 18-24 mo Bronchoalveolar adenoma (2/19);
control rats had no tumors
(0/13); No flbrosls


90 days 18-24 mo Bronchoalveolar adenoma (2/8);
control animals had no tumor*
Remarks
Short exposure


No unexposed controls; short
exposure; small number of
exposed animals; atypical
method of administration

Short exposure; small number of
animals; mostly nonflbrous
material




References
Mori s jet et al. (1979)


Morrison et al.
(1981)



Lee et al. (1979)
tee et at. (1981)

K>
O*

As above
                                                                           fIbers/mL)
                                                                                                               (0/6); no flbrosls
Glass Fibers
(Johns-ManvlIle
Code 100)
Uncoated Glass
Wool
Coated Glass MooI
Not specified
SPF Fischer Rats  2/group   Inhalation  10 mg/m
SO weeks    16 mo      Focal flbrosls was evident with
                       all the dusts but more marked
                       with chrysotlle; noncoated glass
                       wool more reactive than uncoated
                       glass wool
                                                                                                                                                Smalt number of animals
Johnson and Wagner
(1980)

-------
TABLE I.  SUMMARY OF ANIMAL STUDIES ON FIBRQyUtLASS (CONTINUED)
Fiber Type
Glass Micro-
fibers (A blend
of JM CI02 and
JM C104)





Glass Micro-
fibers (Johns-
Manvllle
Coda 100)

Glass Moot
(uncoated)


Glass Wool
(coated)

Glass Micro-
fibers
(JM Cod* 100)


Fiber Dimension
Median diameter
0.6 urn; iMdlan
length 6. 25 urn






Mean diameter
0.3 fm; 71)
10 pi long
1
0.2-3 urn
diameter; 57 J
>10 um long
Not available
but presumably
similar to that
used In Wagner
et al. (1984)
Number Route
Species of of Admin 1-
Anlmals stratlon
Male Baboons Not Inhalation
(Paplo urslnus) speci-
fied
(10
animals
In total
Including
positive
contro 1 s )
Male and Female 28 of Inhalation
Fischer 344 Rats each sex



Male and Female 28 of Inhalation
Fischer 344 Rats each sex


Male and Female 28 of Inhalation
Fischer 344 Rats each sex

Male and Female Unspe- Inhalation
Fischer 344 Rats clfled



Dose/
Concentra-
tion
7.54 mg/m3
(1122
f Iber/mL)






10 mg/m1
(1436
f Ibers/mL)


10 mg/m3
(240
f Ibers/mL)

10 mg/m3
(323
f Ibers/mL)
10 mg/m3




Frequency/
Treatment Duration Results Remarks
Duration of Study
35 mo 41-42 mo Fibres Is after 18 and 30 mo Short exposure In relation to
exposure; however, lesions llfespan of baboons; no un-
less severe than those In exposed controls
animals exposed to crocl-
dollte; no evidence of
malignancy



12 mo Llfespan No f Ibrosls at 24 mo; No tumors In unexposed controls;
lung adenocarclnoma (1/48) 12 tumors In 48 chrysotlle
exposed animals (It lung
. adenocarclnomas; 1 adenoma)

12 mo Llfespan No f Ibrosls; lung adenocarclnoma
( 1/48)


12 mo Llfespan No f Ibrosls; lung adenoma (1/48)


12 mo Llfespan No f Ibrosls; no neoplasms . Lung tumors In 11 of 56 chryso-
(0/55); lung adenocarclnoma tile exposed rats (7 adenocarcl-
In 2 of 53 negative controls noma; 4 adenoma)


References
Goldstein et al. (1983)
Goldstein et al. (1984)







Wagner et el. (1984)



M
*4
As above



As above


McConnell et al. (1984)





-------
TABLE I.   SUMMARY OF ANIMAL  STUDIES ON  FIBRQUSrfSLASS  (CONTINUED)
Fiber Type
31 as* Micro-
fibers
(JM Cod* 100)


Slass Moot
(uncoated)

3 lass Mlcro-
f 1 bers











Number Route DOS*/ Frequency/
Fiber Dimension Species of of Admlnl- Concentre- Treatment Duration
Animals stratlon tlon Duration of Study
51* with 0.2-O.J Male and Female 24 of Inhalation 5 mg/m3 12 or 24 24 mo
un diameter IOPS AF/Han Rats each sex (332 mo
f Ibers/mL)


68* 
-------
TABLE I.   SUMMARY OF ANIMAL  STUDIES ON FIBRO^B^ASS  (CONTINUED)
Number
rltwr Type Fiber 01 mansion Species of
An Ima Is
3lass Wool 3.1 urn mean Male Syrian 60
diameter Hamsters

Female Osborne- 52
Mendel Rats
;lass Wool 5.4 pm mean Male Syrian 66/69
d 1 ameter Hamsters

Female Osborne- 57/61
Mendel Rats
•lass Moot 6.1 pm mean Male Syrian 99
diameter Hamsters

Female Osborne- 58
Mendel Rats
Route Dose/ Frequency/
of Admlnl- Concentre- Treatment Duration Results Remarks References
strati on tlon Duration of Study
Inhalation 10 mg/m 24 mo Llfespan No f Ibrosls or lung neoplasms Glass wool aerosol dusts had a Smith et al. (1986)
(100 In exposed rats or hamsters large proportion of nonflbroos
flbers/mL) material


Inhalation 12 mg/m 24 mo Llfespan No f Ibrosls or lung neoplasms As above
(100 In exposed rats or hamsters
fibers/mo or
1.2 mg/m3
(10 flbers/mL)
i K>
Inhalation 9 mg/«r 24 mo Llfespan No f Ibrosls or lung neoplasms As above —
(25 In exposed rats or hamsters • "°
flbers/mL)



-------
                                                             TABLE
                                                                       SUMMARY OF ANIMAL STUDIES ON Fl
                                                                                                               SS (CONTINUED)

Fiber Type Flbar Dimension Species

Number
of
Animals
Rout*
of Admini-
stration
Dose/
Concentra-
tion
Frequency/
Treatment
Duration

Duration Results
of Study

Remarks References

Fibrous Glass    4.6 um diameter;
(Red Binder;     >20 um long
Group I)
                                    Male and Female   50/5ex    Inhalation  IS mg/m3
                                    Fischer 344 Rats
                                    Male Cynomolgus    IS
                                    Monkeys
                                                                Inhalation  II mg/«
                                                                                        86 weeks    86 weeks
                                                                                        72 weeks
                                                                                                    72 weeks
                                                                           No fibres Is; no pul nonary
                                                                           tumors or mesothel Ionia In any
                                                                           treated rat or monkey groups;
                                                                           statistical Increase In mono-
                                                                           nuclear cell leukemia In Group I
                                                                           female rats (p-0.047; Fisher
                                                                           exact test). Group III (p-0.024)
                                                                           and Group  IV (p-0.002) male rats.
Early death In 37* rats In        Mitchell et al.  (1986)
treated and control group; short
treatment and study duration
Fibrous Glass    0.5-5.5  um          Male and Female
(Yellow Binder;   diameter; >10 urn    Fischer 344 Rats   50/sex    Inhalation  15 mg/m
(Group II)       long
                                    Male Cynomolgus    15
                                    Monkeys
                                                               Inhalation  IS mg/mj
                                                    86 weeks    86 weeks

                                                    72 weeks    72 weeks
                                                                                                                                                                                  As above
Fibrous Glass    <3.5  um  diameter;   Male and Female   SO/sax   Inhalation  5 mg/m
(Uncoated;
Group III)
                 >IO fm  long
Fischer 344 Rats

Male Cynomolgus
Monkeys
                                                       15
                                                               Inhalation  5 mg/m
                                                                                        86 weeks    86 weeks
                                                                                        72 weeks    72 weeks
                                                                                                                                                                                  As above
Fibrous Glass    <3.S  um diameter;   Male and Female   50/sex   Inhalation  5 mg/m
(Uncoated;
Group IV)
                 <10 um  long
Fischer 344 Rats

Male Cynomolgus   15
Monkeys
                                                                                        86* weeks    86 weeks
                                                               Inhalation  5 mg/m3      72 weeks    72 weeks
                                                                                                                                                                                  As above

-------
                                                             TABLE I.  SUMMARY OF ANIMAL STUDIES ON Fl
                                                                                                               SS (CONTINUED)
Iber Type Fiber Dimension

Ibrous Glass 6 samples of vary-
ing fiber size
distributions


Species

Female pathogen-
free Osborne-
Mendel Rats


Number Route Dose/
of of Admlnl- Concentra-
Anlmals stratlon tlon
Not Intrapleural 40 mg
sped- Implantation
fled


Frequency/
Treatment Duration Results
Duration of Study
Single dose 2 yrs Moderately high Incidence of
mesothelloma In two samples of
fine fiberglass milled to
approach length of asbestos
fibers
Remarks

High Incidence of mesothellom
In amoslte, chrysotlle, and
croc 1 do lite exposed group


References

Stanton and wrench (1972)




 I berg I ass
 Code  110)
30} <2.5 urn
diameter; 60)
>20 pm long
SPF Wlstar Rats   36
(sex unspecified)
Intrapleural   20 mg
Injection
Single dose  Llfespan  No mesothelloma
Mesothelloma In 23/36 rats
treated with SFA chrysotlle
Wagner et a I. (1973)
• lass Fibers
                3.; pm diameter;
                <20 pm long

                3.5 urn diameter;
                >100 pm long

                0.05 pn diameter;
                <20 |M long

                0.05 pm diameter;
                »00 pn long
                    Balb/C Mice       25/group Intrapleural   10 mg
                                               Injection
                                                    Single dose  18 mo
                                                Long fiber samples produced
                                                mass Ive fIbrosIs  while short
                                                flbered samples produced only
                                                discrete granulomas with minimal
                                                fIbrosIs
                                                                                                                                                                  Davis (1976)
 Ibrous Glass    17 samples of       Female Osborne-   30 In    Intrapleural   40 mg
                diverse dimensional  Mendel  Rats      each     Implantation
                distributions                         treated
                                                      group
                                                                        Single dose   2 yrs
                                                                           Highest yield of pleura!  sar-
                                                                           coma with fibers <1.S urn  In
                                                                           diameter and >8 urn In length
                                                                                                                   Stanton et al. (1977)

-------
                                                             TABLE
                                                                       SUMMARY OF ANIMAL STUDIES ON FIB
                                                                                                              |SS (CONTINUED)
Iber
Mass
Code


• lass
Type
Fiber
MO)



'Icrof Iber
Code
100)
Fiber Dimension
17) <1 urn diameter
median dlametef
1.8 urn; median
length 22 urn
99) 20 pi long
Ibrous Glass    Mean diameter       Male Syrian        60
                0.1 urn;  S2t >20 \n  Golden  Hamsters
                long

                Median diameter
                0.09 LMI; 0-2)
                >20 urn long

                Mean diameter 0.33
                urn; 46)  >20 |M long

                Mean diameter 0.41
                )«; 0-2% >20 |im long

                Mean diameter 1.23
                urn; 34)  >20 \m long

                Mean diameter 1.49
                urn; 0-2) >20 Mm long
IntrapI euro I   25 mg
Injections
Single dose   Not       Intrathoraclc tumors  (9/60)
             specified
                                                No tumors
                                                Intrathoraclc tumors  (2/60)
                                                No tumors
                                                Intrathoraclc tumors (2/60)
                                                No tumors
Histology not provided; no
control group reported
                                                                                                                                                                                   Smith at al. (1980)

-------
                                                              TABLE I.   SUMMARY OF ANIMAL  STUDIES ON Fl«"
                                                                                                               •SS (CONTINUED)
Iber Type Fiber Dimension
*lne Glass Mean diameter
"Ibers 0.229 urn; mean
length 5.8 urn
fibrous Glass 22 samples of
diverse dimension-
Species
Male Sprague-
Dawley Rats
Female Osborne-
Mendel Rats
Number Route Dose/
of of Admlnl- Concentra-
Anlmals stratlon tlon
45 Intrapleural 20 mg
Injection
30-50 Intrapleural 40 mg
Implantation
Frequency/
Treatment Duration Results
Duration of Study
Single dose Llfespan 13) Incidence of pleura! tumors
(6/44)
Single dose 2 years Probability of pleura! sarcoma
best correlated with the number

Remarks
No tumors In vehicle controls
(saline); high Incidence of
mesothel loma In chrysotlle (4S)>
and crocldollte (54)) treated
animals


References
Lafuma et al. (1980);
Monchaux et al. (1981)
Stantcm et al. (1981)
                 al  distributions
                                                                                                                of  fibers  with  diameters
                                                                                                                <0.25 urn and length >8 urn;
                                                                                                                relatively high correlations
                                                                                                                also noted for  fibers  with
                                                                                                                diameters  up to 1.5 urn and
                                                                                                                length >4  urn
Slass Micro-
fibers (Johns-
lanvllle
Code 100)
86) <0.6 pm
diameter;
88) <5 urn long
SPF Sprague-
Dawley Rats
                                                       48
                           Intrapleural
                           Injection
              20 mg
Single dose  Llfespan
                                                Pleura!
                                                rats
                                                                                   mesothel loma In 4 of 48
Six cases of pleura I tumors
48 rats treated with UICC
chrysotlle asbestos
In    Wagner et at. (1984)
Slass Moot
(uncoated)
85) 
-------
                                                             TABLE I.  SUMMARY OF ANIMAL STUDIES ON FIBRCj
                                                                                                                  (CONTINUED)
Number Route Dose/ Frequency/
Iber Type Fiber Dimension Species of of Admin 1- Concentre- Treatment Duration Results
Animals stratlon tlon Duration of Study
Ibrous Glass Mean diameter 0.5 Female Wlstar 40 Intraperl- 2 mg Single dose Not
Abdominal tumors (mostly
S + SI06) urn; 72.6) <5 urn Rats toneal specified mesothel loma) 2.5) at 2 mg,
long Injection 10 mg Single dose

4x25 mg Four doses
at weekly
Intervals

10) at 10 mg and 57) at
100 mg; f Ibrosls In
animals receiving 100 mg
of glass fibers; less
severe lesions at 10 mg and
2 mg dose levels
Remarks
Tumor rates of 15-67) In UICC
chysot 1 1 e asbestos treated
group at similar dosing regimen;
latency period Inversely related
to dose; no tumors In saline
treated controls


References
Pott and Frledrlchs
(1972); Pott et al.
(1974) and (1976)





 I berg I ass
:MNI04)
50) <0.2 urn
diameter; 50) 
-------
TABLE I.  SUMMARY OF ANIMAL STUDIES ON Fl
                                                  iSS (CONTINUED)
Fiber Type
Glass
Mlcrof Ibers
Glass Fibers
(JM 104)


Glass
Mlcrof Ibers
(JM 104)
Glass
Mlcrof (bars
(JM 100)
Fibrous Glass
Fiber Dimension Species
Average diameter Balb/c
of 0.05 urn nice
Rats
(unspecified
strain)
Not specified Wlstar Rats
( Ivanovas)
SIV Rats
( Ivanovas)
Sprague-Oawley
Rats (Hagenann)
Ml star Rats
(Hagemann)
501 <0.3 urn Wlstar/Sprague-
dlaneter; Daw lay Rats
90| <13 urn long
501 <0.3 pn Wlstar/Sprague-
dlameter; Daw ley Rats
90S <7 urn long
0.45 urn mean Female Osborne-
dlameter Mendel Rats
Number Route Dose/
of of Admin)- Concentra-
Anlmals strut Ion tlon
25 Intraperl- 25 ng
tonea 1
Injection
18 Intraperl- 10 mg
tonea 1 «
Injection
50 Intraperl- 10 mg
tonea 1
Injection
50
50
50
40-60 Intraperl- 2 or 10 mg
group tonea 1
Injection
40-60 Intraperl- 2, 5, or
group tonea 1 10 mg
Injection
25 Intraperl- 25 mg
tonea 1
Injection
Frequency/
Treatment Duration
Duration of Study
Single dose Not
Specified
Single dose Not
Specified
Single dose Not
specified


Single dose Not
specified
Single dose Not
specified
Single dose Not
specified
Results Remarks References
Peritoneal tumors (3/25) Davis (1976)
Peritoneal tumors (3/28)
Abdominal tumors (25/49; 5t)> No experimental details Pott et al. (1980)
Abdominal tumors (36/50; 721)
to
to
Abdominal tumors (29/49; 59.2)) "
Abdominal tumors
(39/49; 79.6J)
Peritoneal sarcoma and Pott et al. (1984)
mesothelloma ( 40-70* )
Peritoneal sarcoma and As above
mesothelloma (2-IOf)
Abdominal mesothelloma (8/25); Tumor Incidence of 80$ (20/25); Smith et al. (1986)
Abdominal reactive tissue/ In positive controls treated with
fibres Is In 13/17 animals UICC croc 1 do lite asbestos; no
                                                                                   untreated animals

-------
TABLE I.  SIMMARY OF ANIMAL  STUDIES ON  FIG
                                                 6S  (CONTINUED)
Iber Type
"Ibrous Glass
(Uncoated)



~ Ibrous Glass
(Coated with
;esln)



Fibrous Glass
(Coated with
Starch Binder)


Fiber Dimension
Average d 1 ameter
0.5 um; average
length 10 um


Average diameter
0.5 urn; average
length 10 um



Average diameter
0.5 um; average
length 10 |m


Number Route
Species of of Admlnl-
Anlmals stratlon
Rats (strain I5/
unspecified) 30

Hamsters (strain 12
unspecified)
Rats 30


Hamsters 12
12
12
Rats 15


30
Hamsters 12
Intratra-
cheal In-
stillation


Intratra-
cheal In-
stillation



Intratra-
chea 1 1 n-
stl nation


Dosa/ Frequency/
Concentra- Treatment Duration Results
tlon Duration of Study
3 x
3.5 mg or Not Llfespan No tumors In any treated rat
10 x 3.5 mg specified or hamster group

3 x

3 x
or
10 x
1 x
2 x
3 x
3 x


10 x
3 x

3.5 mg

3.5 mg As above As above

3.5 mg
3.5 mg
1.75 mg
3.5 mg
3.5 mg


3.5 mg
3.5 mg
Remarks References
Experimental details not avail- Gross (1976)
able; no positive controls;
small number of animals


As above As above





As above





-------
                                                             TABLE I.  SUMMARY OF ANIMAL STUDIES ON Fl
                                                                                                                S (CONTINUED)
Fiber Type
                 Fiber  Dimension     Species
Number    Route        Dose/      Frequency/
  of     of Admlnl-  Concentra-   Treatment   Duration
Animals  stratlon    tlon         Duration    of Study
                                                                                                                           Results
                                                                                                                                                          Remark*
                                                                                                                                                                                  References
-lass Fibers     Diameter  <1 pm;     Guinea Pigs
(Thin Fibers)     93*  <10 urn  long
                Diameter  <1 urn;     Guinea Pigs
                921  >IO in  long
I: lass Fibers
JVery Thin)
3 lass Fibers
(Thick)
Diameter <0.3 urn
and <5 urn long
Diameter <0.3 urn;
50* >IO urn long
Diameter 2 um;
88) IO um  long
30       Intratra-   2 x 12.? mg  Biweekly    2 years    No fIbrosls nor  tumors
         cheal
         Instillation

30                   3 x 4 mg     Biweekly    2 years    Flbrosls;  no tumors
                                                      30       Intratra-   2 x 12.5 mg  Biweekly    2 years    No f Ibrosls nor tumors
                                                               cheal
                                                               Instillation

                                                      30                   2 x 6 mg     Biweekly    2 years    Flbrosls;  no tumors
                                                      30       Intratra-   2 x 12. $ mg  Biweekly    2 years    No f Ibrosls nor neoplasms
                                                               cheal
                                                               Instillation
                                                      30
                     2 x  12.5 mg   Biweekly     2  years     No  f Ibrosls  nor neoplasms
Long fibers O10 um) of croc I do-  Wright and Kuschner
lite asbestos produced fIbrosls;  (1976, 1977)
short fibers «IO um) did not
produce fIbrosls
                                                                                                                                                                                  As  above
                                                                                                                                                                                  As  above

-------
TABLE I.  SUMMARY OF ANIMAL STUDIES ON Fl
                                                   S  (CONTINUED)
Iber Type Fiber Dimension Species
tacoated Glass O.I um man Male Syrian
"Icroflbers diameter Hamsters
0.25 |M mean
d 1 ameter '
household 2.3 ym mean diameter
nsulat Ion
•lass fibers 3.0 um mean diameter
4.1 pm mean dlamater
3lass SO) <0.3 |m Male Syrian
'Icroflbers diameter; Golden Hamsters
(JM Code 104) 50) <7 um long
SO) <0.3 pm
diameter;
50) <4.2 M«
long
Number Route Dose/
of of Admlnl- Concentra-
Anlmals stratlon tlon
20 Intratra- A total of
cheal In- 7 mg
stl Nation
20 2 mg
20 21 mg
20 18 mg
20 17 mg
136 Intratra- 8 x 1 mg
cheal
Instillation
138 Intratra- 8 x 1 mg
cheal
Instillation
Frequency/
Treatment Duration
Duration of Study
Biweekly It months
Meekly
Not specified
Not specified
Not specified
Weekly 136 weeks
Meekly 136 weeks
Results Remarks References
Mild flbrosls In glass micro- Ptckrell et al. (1983)
fibers treated group at 11
months after Instillation

No flbrosls As above
No flbrosls
No flbrosls
K>
Lung carcinoma (5/136) Tumors In croc 1 do lite asbestos Mohr et al. (1984) £
Pleura! mesothel loma (37/136) positive controls
Thoracic sarcomas (6/136) Lung carcinoma (9/142)
Mesothel loma (8/142)
Thoracic sarcoma (1/142)
Lung carcinoma (6/138) 2 cases of thoracic sarcoma
Mesothel loma (26/138) (2/135) In titanium dioxide
Thoracic sarcomas (6/138) treated group but no
mesothel loma or lung
                                                                                  neoplasms; no saline
                                                                                  control group

-------
TABLE I.  SUMMARY OF ANIMAL  STUDIES ON Fl
                                                   S  (CONTINUED)
Iber Type
• lass
Mcrof Ibers
JM Code 104)
Ibrous Glass
-lass
Mcrof Ibers
JM 104/
Fiber Dimension
31} <0.25 urn
diameter; 58*
<5 urn long
0.45 v» mean
d 1 ameter
50* <3.2 urn
length; 50*
<0. 12 urn
Species
Mala and Female
Syrian Golden
Hamsters
Female Osborne-
Mendal Rats
Female
Wlstar
Rats
Number Route Dose/
of of Admin 1- Concentra-
Anlmals stratlon tlon
35/sex Intratra- 26 x 1 mg
cheat
Instillation
32 Intratra- 5 x 2 mg
cheat
Instillation
34 Intratra- 20 x 0.05 mg
cheat
0.5 mg
Frequency/
Treatment Duration
Duration of Study
Biweekly 85 weeks
for 52
weeks
Meekly Llfespan
Weekly Llfespan
Results
No tumors (0/34 males; 0/30
females)
Slgnf leant f Ibrosls (7/22);
no tumors
Lung tumors In 5/34 animals
(1 adenoma, 2 adenocarclnoma,
2 squamous cell carcinomas)
Remarks
No positive control animals
No lesions In saline controls
and untreated animals
Positive control animals
treated with croc 1 do lite
had 11/35 lung tumor
References
Feroo et al. (1985)
Smith et al. (1986)
Pott et al. (1987a)
  Instilla-
  tion
and 9/142 lung cancer

-------
Table 2.  SLMMARY OF ANIMAL
                                   FON MINERAL WOOL
Fiber
Type
Rock wool






Salnt-Gobaln
rock wool


Fiber
Dimension
58$ IO |im long




22.71 IO um long

Species
Male and
Female
F344 rats




Male and
Female
Wlstar IOPS
rats
Number of Route of Dose/ Frequency/ Duration of
Animals Administration Concentration Treatment Study
Duration
56 Inhalation 10 mg/m3 12 mo Llfespan
(227 flbers/mL;
for d <3 (•»;
and 1 >5 um)



48 Inhalation 5 mg/m3 24 mo 24 mo
(It flbers/mL;
for J^>5 um)

Results
No lung fibres Is;
lung adenoma (2/48)





No pulmonary
changes; no
lung tumors
(0/24 males;
Remarks References
No tumors In Wagner et at. (1984)
unexposed controls;
lung adenocarclnoma
(11/48) and adenoma
(1/48) In chrysotlle
asbestos positive
controls
Very low fiber con- Le Bouffant et al.
centratlon; dust (1984)
cloud contained
mostly nonflbrous
                                                                              particles; nine
                                                                              cases of  lung tumors
                                                                              In Canadian  chryso-
                                                                              tlle asbestos control
                                                                              group (males, 9/24;
                                                                              females.  4/23); no
                                                                              tumors  In unexposed
                                                                              control animal (0/27)

-------
                                  Table 2.   SUMMARY OF ANIMAL STUOI9
                                                                            ERAL WOOL (CONTINUED)
Fiber
Type
Mineral
Wool








Fiber
Dimension
mean diameter
2.7 urn; 75*
>10 urn long







Number of Route of Dose/ Frequency/ Duration of
Species Animals Administration Concentration Treatment Study Results
Duration
Female Osborne- 55 Inhalation 12 mg/m (200 24 mo Llfespan No pulmonary
Mendel rats flbers/mL; flbrosls; No lung
76 flbers/mL tumors (0/55)
for fibers
longer than
10 urn with
diameters
<1 pm)


Remarks References
3/57 UICC crocodo- Smith et al. (1986)
lite positive control
rats developed tumors
(1 mesothel loma.
2 bronchoalveolar
tumors); No tumors
In sham controls
(0/59) and
unexposed rats
(0/125)
Male Syrian
hamsters
                 69
                               InhaI at I on
                                                 as  above
                                                                  24  mo
Llfespan
No lung flbrosls;
no lung tumors
(0/69)
No tumors In UICC
crocodoIIte asbestos
exposed group (0/58)
or In unexposed
controls (0/112);
one bronchoalveolar
tumor (1/58) In
sham control group

-------
Table 2.  SLMMARY OF ANIMAL STUDIt
                                        INERAL WOOL  (CONTINUED)
Fiber
Type
Rock Wool
(with resin)



Rock wool
(without
resin)
Slag Wool
(with resin)

Slag Wool
(without
resin)
Fiber
Dimension
77$ <1 ftm In
diameter; 10%
<5 urn long


82$ <1 urn In
diameter; 69$
<5 pm long
10% 
-------
Table 2.   SIMMARY OF ANIMAL STUD IE
                                         NERAL WOOL  (CONTINUED)
Fiber
Type
Rock Wool


Rock Wool


Slag Wool



Basalt Wool


Basalt Wool









Fiber Species Number of
Dimension Animals
SO) <0.64 urn In Female Sprague- 45
diameter; 50) ' Daw ley rats
<4.l urn long
SO) <1.90 urn Female Sprague- 63
diameter; 50) Daw ley rats
<23 urn long
90) <0.28 urn Female Wlstar 41
diameter; 9) rats
<10 urn long

SO) <0.52 |im Female Wlstar 45
diameter; 50) rats
<58 pm long
50) <1.8 urn Female Wlstar S3
diameter, 50) rats
<20 urn long







Route of Dose/
Administration Concentration
Intraperltoneal 10 mg
Injection

Intraperltoneal 25 mg
1 n ject 1 on

Intraperltoneal 5 mg
Injection


Intraperltoneal 5 mg
Injection

Intraperltoneal 15 mg
Injection








Frequency/ Duration of Results
Treatment Study
Duration
Single dose 15 mo No peritoneal tumors


3 x 25 mg Llfespan Peritoneal sarcoma/
mesothelloma (16))

Single dose Llfespan Peritoneal tumors
(5))


Single dose 15 mo No peritoneal tumors


5 weekly Llfespan Peritoneal tumors
doses In 30/53; negative
sat Ine control had
a tumor Incidence
of 1/102; UICC/
Canadian chrysotlle
produced high
Incidence of tumor
at much lower
doses
Remarks References
Short observation Pott et al. (1984)
period (ongoing
study)
No positive As above
control group*

Small dose; tumor As above
yield not
statistically
significant
Relatively small As above
dose; short
observation period
Relatively large Pott et al. (1987b)
fibers; very
high dose level ;
preliminary
results only






-------
Table 3.  SUMMARY OF ANIMAL
                                       CERAMIC FIBERS
Fiber
Type
Ceramic
Aluminum
Silicate
Gists












Fiber Number of Rout* of Dose/ Frequency/ Duration of
Dimension Species Animals Administration Concentration Treatment Study Results Remarks References
Duration
90% <3 urn long SPF Wlstar rats 48 Inhalation 8.4 mg/m1 12 mo Up to Interstitial f 1 broils Dust cloud contained Davis et al. (1984)
and <0.3 urn (AF/HAN strain; (95 f Ibers/mL; 32 mo occurred to a loiter mostly short, thin
diameter sex unspecified d <3 urn, 1 but not significantly fibers
>5 urn) different degree than
that for chrysotlle
asbestos animals ob-
served In other
studies; lung
tumors In 8 animals
(1 adenoma.
3 carcinomas.
4 mal Ignant
hlstlocytomas); no
tumors In 40
unexposed control
animals

-------
Table 3.  SUMMARY OF ANIMAL STUOIEl
                                           4IC FIBERS (CONTINUED)
Fiber Fiber
Type Dimension
Refractory 83f> 10 Urn long
Ceramic Fiber and 86} <2 urn
diameter






Number of
Species Animals
Female Osborne- ' 55
Mendel rats





Mole Syrian 70
hamsters
Route of Dose/
Administration Concentration
Inhalation ' 12 mg/m3
(200 flbers/mL)





Inhalation 12 mg/m5
(200 ftbers/mL)
Frequency/ Duration of
Treatment Study
Duration
24 mo Llfespan






24 mo Llfespan

Results
No lung flbrosls; no
lung tumors; no tumors
In sham or unexposed
control



No lung flbrosls;
one mesothel loma;
Remarks References
Low tumorlgenlc re- Smith et al. (1986)
ponse In UICC crocldo-
1 Ite asbestos control
rats (3/59;
1 mesothel loma.
2 bronchoalveolar
tumor)
No tumors (0/58)
In positive control
                                                              no primary lung
                                                              tumors;  no tumors
                                                              In unexposed
                                                              control; one case
                                                              of bronchoalveolar
                                                              tumor In sham
                                                              controI
hamsters exposed to
UICC crocldollte
asbestos

-------
Table 3.  SUMMARY OF ANIMAL STUDIES
                                            1C  FIBERS  (CONTINUED)

Fiber
Type
Refractory
Alumina Fibers
(a* manufac-
tured or
"thermally
aged")



Synthetic
Aluminum
Silicate
Fibers
Fiber
Dimension
Median diameter of
3.0 urn; median
length 10.9-62 pm






D 1 ameter between
0.9 and 1.0 |M;
1 ength
unspecified
Number of
Spec 1 es An 1 ma 1 s
Mala and female 25 of each
albino rats of sex/group
the Alder ly
Park (Wlstar
derived strain)




SPF Wlstar 24 males
rats 12 females


Route of Dose/ Frequency/ Duration of
Administration Concentration Treatment Study Results
Duration
Inhalation 2.18-2.45 mg/m 86 weeks >B6 weeks Pulmonary reaction to
both forms of Saff II
was minimal ; no
pulmonary neoplasms





Intrapleural 20 mg Single dose Lifetime Masothel loma In
Inoculation 3/31 rats


Remarks References
Levels of resplrable Plgott et al. (1981)
dust In the atmos-
pheres were low;
positive control
animals exposed to
UICC chrysotlle
asbestos developed
pulmonary neoplasms
(9/38)
Carcinogenic poten- Wagner et al. (1973)
cy of ceramic
fibers were consi-
derably less than
                                                                                    SFA chrysotlle
                                                                                    asbestos

-------
Table 3.   SLMMARY OF ANIMAL STUD IE
                                          1AMIC FIBERS (CONTINUED)
Fiber
Type
Alumina
Oxide Fibers
(glass 21)
Zlrconla oxide
fibers
(glass 22)
Refractory
Alumina
Fiber
(Saffll)





Fiber
Dimension
Not specified


Not specified


Type A:
median diameter
2.75 um;
median length
15.5 um
Type B:
median diameter
3.7 um; median
length 17 um
Species Number of
Animals
Female Osborne- 50
Mendel rats

Female Osborne- 50
Mendel rats

SPF Mlstar 22 males and
rats (Alderley 12 females
Park Strain) per group






Route of Dose/
Administration Concentration
Intrapleural 40 mg
Implantation

Intrapleural 40 mg
Implantation

Intraperltoneal 20 mg
Injection
__ ^






Frequency/ Duration of Results
Treatment Study
Duration
Single dose 2 yr Pleura! neoplasms
In 2/47 rats

Single dose 2 yr Pleura! neoplasm
In 1/45

Single dose up to 12 mo Mild chronic
Inflammatory
response with a
ml Id amount of
cot lagen In the
abdominal tissues



Remarks References
Tumor Incidence Stanton et al.
not statistically (1981)
significant
Tumor Incidence
not statistically
significant
Progressive Plgott and Ishmael
peritoneal fibres Is (1981)
In rats receiving
UICC Rhodes Ian
chrysotlle asbestos





-------
Table 3.  SUMMARY OF ANIMAL STUD I?
t RAM 1C FIBERS (CONTINUED)
Fiber
Type
Saffll
Alumina
Fibers
Satf II
Z 1 rcon 1 a
Fibers
Ceramic
Aluminum
Silicate
Glass
Ceramic
"Flberfrax"
Fiber
Dimension ,
median diameter
3.6 urn; median
length 17 urn
median diameter
2.5 urn; median
1 ength 1 1 urn
90S <3 urn long and
<0.3 urn
diameter
50* <8.3 ym long;
50* <0.9I um
In diameter
Number of
Spec 1 es An 1 ma 1 s
SPF albino 40 (20 of
Wtstar rats each sex)
(Alderley
Park Strain)
SPF albino 40 (20 of
Mlstar rats each sex)
(Alderley
Park Stain)
SPF HI star rats 32
of AF/HAN
stain (sex
unspecified)
Female HI star 47
rats
Route of Dose/ Frequency/ Duration of
Administration Concentration Treatment Study
Duration
Intraperltoneal 20 mg (0.2 mL Single dose 6 mo
Injection of a 10)
suspension of
fibers)
Intraperltoneal 20 ng Single dose 6 mo
Injection
Intraperltoneal 25 mg Single dose Lifetime
Injection
Intraperltoneal 9 mg 5 weekly Llfespan
Injection doses
Results
Nodular deposit of
connective tissue;
no f 1 bros 1 s
Nodular deposit of
connective tissue
conta 1 n 1 ng co 1 1 agen
Peritoneal neoplasms
In 3/32 animals;
first tumor occurred
850 days after
Injection
Abdominal tumors In
33/47 animals; total
number of fibers In-
Remarks References
Harked peritoneal Styles and Ml (son
flbrosls In rats (1976)
treated with UICC
chrysot Me
asbestos
As above
•o
No vehicle control Davis et al. (1984) <"
group
Preliminary results; Pott et al. (1987b>
UICC chrysotlle
Induced dose-related
                                                             Jected (173 x10°) «er
                                                             comparable to that
                                                             of chrysot11• at
                                                             0.25 mg (202 x 106)
                                             Increase  In peritoneal
                                             tumors at much  lower
                                             doses (11/36, 21/34.
                                             30/36 at 0.0$,  0.25;
                                             and  1.0 mg, respec-
                                             tively

-------
                                                                    Table 3.   SIMMARY OF ANIMAL STUD IE
                                                                                                            ERAM 1C FIBERS (CONTINUED)
Flb«r Fiber
Type Dimension
Ceramic 50% < 6.9 pi In
HAN length; 50*

-------
Table 4.   SIMMARY OF  AN I
                                IES ON ERIONITE
Fiber
Typ.
Oregon
Erlonlte



-



Oregon
Erlonlte





•
Kara In
Rock fiber


Erlonlte





Fiber
Dimension
86$ <0.4 urn
diameter;
92$ 
-------
Table 4.  SUMMARY OF ANIMAL STI
                                       ERIONITE (CONTINUED)
Fiber
Type
Erlonlte

















Fiber Number of Route of Dose/ Frequency/ Duration of
Dlrmnslon Species Animals Administration Concentration Treatment Study
Duration
201 <1 urn and Male Swiss 12 Intraperltoneal 10 mg Single dose Six animals
95$ <8 urn albino mice Injection sacrificed at
long; 19$ 2-3 months;
<0.l pm and 95$ remaining 6
<1 pm diameter animals were
maintained
until death
(up to 15
months
Male 5*1 ss 5/group Intraperltoneal 10 mg or 30 mg Single dose Not Specified
albino mice Injection







Results Remarks References
Malignant peritoneal Suzuki (1982)
tumors In 2/5 animals
after 15 months;
flbrotlc lesions
presented at
neoplastlc tissues



Significant flbrosls; Chrysotlle (10 mg)
malignant peritoneal produced
tumors In 4/5 animals peritoneal tumors
receiving 10 mg of In 2/5 animals;
erlonlte; animals no tumors In
receiving 30 mg untreated controls
erlonlte died with
Intestinal
obstruction

-------
Table 4.  SUMMARY OF ANIMAL STI
                                       ERIONITE (CONTINUED)
Fiber
Type
Erlonlte 1
(Colorado)








Erlonlta 1 1
(Nevada)





Karaln
(Turkey)
dust
arlonlte

Fiber
Dimension Species
90% <8 urn and male Balb/C
6} >9.5 urn mice
long; 85| 1.4 urn
diameter





95) <8 urn and Male Balb/C
4J >9. 5 |im mice
length; 82) <0.5
urn and IOOK 
-------
Table 5.   SUMMARY OF AN I MALI
                                 ES ON WOLLASTONITE
Fiber
Type
Wollastonlte



Wo II as ton Ita
(Canada)





Wollastonlte
(India)



Fiber Number of
Dimension Species Animals
Not specified Male Fischer Not
344 rats specified


4 samples Female Osborne- 20-2 5/
consisting of Mendel rats group
most 1 y 1 arge
fibers; only
one sample
was completely
fibrous
10< <2.4 urn and Female Wlstar 94
90} < 13 urn long; rats
\0t <0.62 urn
and 90( <2.3 urn
diameter
Route of Dose/ Frequency/ Duration of
Administration Concentration Treatment Study Results Remarks
Duration
Inhalation to mg/m 12 or 24 Up to 120 Results not yet
mo weeks available; no
adverse effects
on survival
Intrapleural 40 mg Single dose 2 yr Pleura! tumors In
Implantation 0/24, 2/25, 3/21,
5/20 animals




Intraperltoneal 20 mg 5 weekly 130 weeks No peritoneal tumors; Preliminary
Injection doses low degree of findings only; no
adhesion of hlstologlcal data
abdominal organs

References
AdKlns and
McDonnell (198$)


Stanton et al.
(1981)





Pott et al. (1987b)





-------
TABLE 6.  SUMMARY OF AN IN
                                lES ON ATTW>ULGITE
Fiber
Type
Attapulglte
(LebrIJa, Spain)

Attapulglte/
Palygorsklte
(Leicester, U.K.)






Attapulglte
(U.S.)






Attapulglte
(French)


Fiber
Dimension
All fibers
<2 urn
long
18* of fibers
_>6 p« In length
and <0.2 um In
d 1 ameter



I

Two samples
composed
entirely of
short fibers
of smell
d 1 ameters


Mean diameter
0.06 um;
mean length
0.77 pm
Species
SPF Fischer
344 rats

SPF F344
rats







Female
Os borne-
Mendel





Sprague-
Dawley rats
(sex unspe-
cified)
Number Route Frequency/ Duration
of of Dose/ Treatment of
Animals Administration Concentration Duration Study
40 Inhalation 10 mg/m3 12 months Llfespan
(20 of
each sex)
40 Inhalation 10 mg/m3 12 months Llfespan
(20 of
each sex)






30-50 Intrapleural 40 mg Single dose 2 years
group Implantation






Not Intrapleural 20 mg Single dose 2 years
specified Injection


Results
No flbrosls; peritoneal
mesothelloma (1/40)

Flbrosls; mesothelloma
(3/40); malignant alveolar
ttmor (2/40); bronchoal velar
hyperplasla (8/40; 1 BAH
with MAT); UICC cradollte
produced 1/40 adenocarc 1 noma
and 3 BAH (1 BAH with

adenocarc 1 noma )
Pleura! tumor Incidence
2/29 for both samples






No mesothel loma; UICC
and Canadian chrysotlle
asbestos produced I9{
and 48< mesothel loma
Remarks Reference
No significant Wagner et al.
excess of tumors (1987)

Some evldlence of Wagner et al. (1987)
carcinogenic I ty






*>
^
Tumor Incidence Stanton et al.
not statistically (1981)
significant
compared with
controls
treated with
noncarc 1 nogen 1 c
materials (17/61))
Abstract only; Renler et al.
actual data (1987)
not aval (able

                                                       Incidence, respectively

-------
TABLE 6.  SUMMARY OF ANl!
IIES ON ATTAPULGITE  (CONTINUED)
Fiber
Type
Attapulglte
(LebrIJa,
Spain)
Attapulglte
(Torrejon,
Spain)
Attapulglte/
palygorsklte
(Leicester,
U.K.)
Attapulglte
(palygorsklte)
Attapulglte
(France,
Spain,
U.S.)
Attapulglte
Fiber Species
Dimension
Al 1 fibers SPF F344
<2 fM long rats
0.54J of fibers SPF F344
(by mass) j>6 pm rats
In length and
<0.5 urn In
d 1 anwter
18( Of fibers SPF F344
(by mass) >6 urn rats
In length and
<0.2 urn In
d 1 ameter
37.5JC <2um Wlstar
long; 70J <5um Rats
Not specified; Rats
composed of
the short,
thin fibers
mean > NMRI
length <1 pm mice
Number
of
Animals
20 of
each
sex
20 of
each
sex
32 (16 of
each
sex)
40
Not speci-
fied
60 of
each
sex
Route
of Dose/
Administration Concentration
Intrapleural Not specified
Injection
Intrapleural Not specified
Injection
Intrapleural Not specified
Injection
1 ntraper 1 tonea 1 3 x 25 mg
Injection
1 ntraper 1 tonea 1 Not specified
Injection
Feeding If or 3)
Frequency/
Treatment
Duration
Single
dose
Single
dose
Single
dose
Weekly
Single
dose
25 mo
Duration
of Results Remarks Reference
Study
Llfespan Peritoneal mesothel loma Tumor Incidence Wagner et al.
(1/40); Pleural not significant (1987)
mesothel lorn (1/40)
Llfespan Pleural mesothel loma As above
(14/40)
Llfespan Pleural mesothel loma As above
30/32; UICC
Croc 1 do lite produced M
34/40 mesothel loma; »
chrysotlle produced ""
19/40 pleural tumors
Llfespan 65< of animals developed Chrysotlle Pott et al.
peritoneal mesothel lorn; asbestos produced (1974)
first tumor appeared at 30-671 tumor
day 275 Incidence
Not specified No excess tumors observed No experimental Pott et al. (1985)
with 3 types of attapul- details or data re-
glte ported
25 mo No toxlcltles; no Brune and Deutsch-
tumors Wenzel (1983)

-------
Table 7.  SUMMARY OF ANIMAL S
                                       ARAMIO FIBERS
Iber
ype
Utraf Ina
.evlar*








Mtraflne
:evlar«




;evlar*
Ibers

.evlar*
(bars


Flbar
Dlmanslon
90< <1.5 urn
d 1 amater ;
>75J lass
than 20 fim
long





60-70) 
-------
TABLE 7.  SUMMARY OF ANIMAL STUDIES ON ARAM ID FIBERS  (CONTINUED)
Fiber
Type
Kevlar*
pulp







Kevlar*
polymer
dust








Nomex*
aramld


Fiber
Dimension
Mostly aggregates
of large fibers;
a small proportion
composed of f 1 ne
fibrils; (96J of
these fibrils

-------
Table 8.  SUMMARY OF ANIMAL S
                                       CARBON FIBERS
ber
-pe
•.N-based
-rbon
ber

.opped
rbon
Ibers





.opped
-rbon
Ibers


rbon
ISt





-rbon
Ibers


Fiber
Dimension
7 urn
diameter
20-60 urn
long
99$ nonflbrous ,
0.8$ carbon
fibers with
1-2.5 urn
diameter and up
up to 15 urn
long

Not specified




0.2-15 urn
diameter





20$ <1 um and
35-40$ <2 (Mi
d 1 ameter

Species
Male albino
rats of the
CO BR sprague
Daw ley Strain
SPF guinea
pigs






SPF
guinea
pig*


Male and
female
SPF Wlstar
rats



Male Fischer
344 rats


Number Duration/ Duration
of Route of Dose/ Frequency of of
Animals Administration Concentration Treatment Study
10-20 animals Inhalation 20 mg/m3 4,8,12, or Up to 32
per group (16-23 mg/m ) 16 weeks weeks after
16 wk of
exposure
A total of Inhalation 370 nonflbrous 7-104 hr 1-144 days
13 exposed particles/ml after exposure
animals; 2.9 fibers/ml
2 negative
controls



2-9 animals Inhalation Not specified 7-100 hr Up to
per group 2 yr



12 Intraperltoneal 50 mg/kg Single 1 or
Injection (10-15 mg/ dose 3 mo
rat)




Not Intraperltoneal Not 1 single l.p. Up to
reported Injection/ reported dose; 2 Intra- 2 yr
Intratracheal trachea) doses
Instillation
Results
No lung pathology



No pathological
effects observed
In animals at
Interval sacrifice




No evidence of
pathological changes
In the lungs


No evidence
of flbrosls; positive
control animals re-
ceiving chrysotlle
asbestos (2.5 mg)
showed diffuse flbro-
sls by 3 months;
No tumors or toxlclty



Remarks
Large diameter
fibers


Oust cloud was
predominantly
nonflbrous;
short exposure;
small number
of animals


Short exposure;
resplrable fraction
was mainly non-
fibrous
particles
Short observation
period





Prel 1 ml nary oral
report only


References
Owen et al.
( 1986)


Holt and Home
(1978)





to
Holt (1982)




Styles and
Wilson (1973)





Parnell (1987)




-------
                                   TABLE 8.  SUMMARY OF ANIMAL STUDIES
                                                                                 FIBERS  (CONTINUED)
ber Fiber Species
pe Dimension
•lyacryl- Not reported Rats
Itrlle (strain, sex
AN) -based - not specified
rbon fibers




Number
of
Animals
Not
reported






Duration/ Duration
Route of Dose/ Frequency of of
Administration Concentration Treatment Study
Intratracheal Not Single 1-9 mo
Instillation reported dose






Results
Both preparations
of carbon fibers Induced
lung f Ibrosls; the
flbrogenlclty of
chrysotlle asbestos
was several -fold
higher than that of
carbon fibers
Remarks References
No experimental Troltskaya et al.
details were (1984)
provided





mixture
carbon
bers and
•specified
astlc
rbon
bers

-rbon
!ber
lament

Not
reported



Not
reported

1.5 cm long
or 5 cm
long; diameter
not specified
Sprague
Daw 1 ay
rats
(sex unspeci-
fied)
Male & Fe-
male Sprague-
Dawley rats
Wlstar rats
(sex not
specified)
!
Not Intratracheal
reported Injection



4 0/5 ex Subcutaneous
Implantation

10 or 50 rats/ Intramuscular
group Implantation


Not
reported



25 mg
In a
2 cm disc
Not
specified


Not
specified



Single dose


Single dose



rbon
•n  f 1 brous
•wder
                                                                                        Life span
                                                                                        18
particle
sIze not spec I-
fled
Wlstar Rats
                   50 rats
                                    Intramuscular
                                    implantatlon
Not
specified
Single dose     18 months
                                                                                                        No lung fIbrosls; only
                                                                                                        moderate foreign body
                                                                                                        reaction
                                                  Local sarcoma
                                                  (tumor Incidence not
                                                  specified)
                                                                                                        No malignant
                                                                                                        change
                                No malignant
                                change
                                                                             Experimental details
                                                                             and results were not
                                                                             not avallable for
                                                                             evaluation
                                                           Details of findings
                                                           not available
                                                                             Test materials were
                                                                             either nonflbrous or
                                                                             large filament; route
                                                                             of exposure not rele-
                                                                             vant to evaluation of
                                                                             of lung carclnogenlclty
                                                                                    Swensson (1979)
                                                                                    (as reported by
                                                                                    Gross, and Braun,
                                                                                    1984)
                                                                                                                                                                                            Ma I too I et al.
                                                                                                                                                                                            (1982b, 1987)
                                                                                    Tayton et al.
                                                                                    (1982)

-------
TABLE 8.  SLMMARY OF ANIMAL STUOIE
                                              FIBERS (CONTINUED)
Number
'Iber Fiber Species of
'ype Dimension Animals
Jarbon 7 put CHB8:CH B per
:lber r»ln- diameter; rabbits group
orced 20-100 out (sax not
:arbon long specified)


•found Not specified Hale 40 per
:arbon fib- C3H/HeJ group
>ers of 4 types nice

a) continuous
(lament (CF)-
i Itch-based

b) short -fl fa-
ired pltch-
ias«d (MAT)

c) polyacrylc—
iltrlle (PAN)-
lased
d) oxidized
'AN- based
Duration/ Duration
Route of Dose/ Frequency of of Results
Administration Concentration Treatment Study
1 ntramedu 1 1 ary 50 mg Single 2 or Small amount of
Implantation dose 12 weeks H broils around some
carbon fibers



Skin painting 2.5 mg 3 times Llfespan
(25 pL of veekly
10* w/v
f Iber suspen-
sion) Skin tumors at site of
application; squamous
cell carcinoma (1/40);
papllloma (1/40)

Flbrosarcoma (1/40);
hemanglosarcoma (1/40)


No tumors (0/40)

Lelyonyosarcoma (0/40)

Remarks References
Method of exposure Neugebauer
not relevant to the et al. (1981)
determination of the
potential Induction of
lung toxlclty and
flbrosls
In the CF treated group, DePass (1982)
neither tumor Incidence nor
time to onset of tumors was
significant compared to ben-
zene-treated control animals
(0/40) In this study but
were judged to be weakly
oncogen Ic using historical
benzene controls (0/281); "
positive controls receiving 0
O.t( methylenolantnrene
had Increased Incidence
of squamous cell carcinoma
and papllloma (33/38)




-------
TABLE 9.
                    F ANIMAL STUDIES ON POLYOLEFIN FIBERS
Fiber Type
Polypropylene





Polyethylene
dust



Polypropylene
dust
Ozon 1 zed
Polyethylene
SHFF

Ozon 1 zed
Polypropylene
SHFF

HFF
Polypropylene
Fiber Dimension
90S <2.l u«
and 50« 
number of animals
'
M
Ml


Limited study'; No MB Research
data available Laboratories
on .the choree- ' . (1980)
terlsltlc on the \
test materials, >
dose and detalls'on
the method of
administration'' . •



-------