5
TOXICOLOGICAL STUDIES
IN support of its selection of an occupational exposure standard of 2 fibers/cm3 for manufactured vitreous fibers (MVF), the Navy reviewed much of the available toxicological literature published before 1997. It presents this information as taken from in vitro studies, epidemiological studies (discussed in the next chapter), and animal toxicity studies. The animal toxicity studies are grouped by fiber type and route of administration. The published literature cited by the Navy are current through 1997. However, some relevant toxicological material published since the Navy's 1997 report might inform the Navy's selection of an occupational exposure limit.
The Navy correctly notes that inhalation studies have yielded the most relevant data, as they were conducted using the route of administration that most closely mimics expected human exposures. It also acknowledges the controversy with regard to some aspects of animal toxicity testing of MVF, including the validity of intrapleural and intraperitoneal administration. The subcommittee agrees with the Navy that the route of administration is one of the most controversial aspects of toxicity studies of MVF. Although the Navy does mention some of the controversy and limitations of the toxicity studies it reviewed, it does not elaborate on the limitations of the noninhalation studies. In general, an assessment of the toxic effects of inhaled fibers requires consideration of both the animal model and the fibers' characteristics, including its dimensions, durability, biopersistence, and surface characteristics.
Inhalation, intratracheal instillation, and intracavitary injection studies in animals have been used for estimating the biopersistence and hence potential toxicity and carcinogenicity of inhaled MVF in humans. Each
kind of study has advantages and limitations, as discussed in McClellan et al. (1992) and McConnell (1995) and briefly presented below.
INHALATION STUDIES
Experimental data are essential for providing basic information on the physiological and pathophysiological pulmonary responses to inhaled particles. Because various rodent species respond differently to selected inhaled materials, it is essential to consider numerous factors—such as anatomy and deposition patterns, physiology and macrophage clearance efficiency, biochemistry and inflammation and fibrogenic potential—when extrapolating the results of animal inhalation studies to humans. Therefore, knowledge of morphological and functional pulmonary characteristics is essential for full understanding of structure-function relationships among species but it is also necessary if one is to develop accurate risk estimates with regard to the toxicity of inhaled particles in exposed humans.
Several rodent species are commonly used in particle and fiber inhalation-toxicity studies designed to simulate human exposures and to evaluate lung responses to inhaled dusts. But experimental animals and humans differ with respect to lung anatomy and physiology and these differences influence particle deposition and corresponding lung-clearance responses. For example, humans have relatively symmetrical dichotomous airway branching that favors concentrated deposition on branch points, or bifurcations; rodents have highly asymmetric, monopodal branching that theoretically should reduce the tendency for concentrated deposition. Distal airways are fundamentally different between humans and rodents: humans have several generations of nonrespiratory bronchioles and three generations of respiratory bronchioles and alveolar ducts; guinea pigs and hamsters have poorly develop respiratory bronchioles, and mice and rats generally lack them. Humans and rodents have different pleural tissue anatomy. And rodents are obligate nasal breathers, whereas humans can favor oral breathing while speaking or during strenuous activity, thus permitting enhanced particle penetration to the lungs.
Several studies have used rats and hamsters as the primary species for assessing the chronic effects of inhaled fibers (Mast et al. 1994; Mast et al. 1995a; McConnell et al. 1999). Some have demonstrated clear
interspecies differences in lung-tumor and mesothelioma responses to inhaled synthetic fibers. Rats appear to be more likely to develop lung tumors after exposure to refractory ceramic fibers (RCF) than hamsters, which have greater sensitivity for developing mesotheliomas (Mast et al. 1994; McConnell et al. 1994). Hamsters appear to be resistant to the development of lung tumors after chronic exposure but appear to be extremely sensitive to mesothelioma induction after exposure to selected fiber types. Because few chronic fiber inhalation studies of appropriate reference materials have been conducted in hamsters, it is difficult to determine whether the hamster is a relevant model for humans. Similarly, interpretations of lung-tumor response in chronically exposed mice are difficult because of the high incidence of spontaneous lung tumors.
Nevertheless, mammalian inhalation tests have some obvious advantages over other tests. The route of exposure is similar to that in humans, and the exposure to fibrous materials is directed to the intact pulmonary system, including all natural defense mechanisms. In rats, the incidences of fibrosis, lung cancer, and mesothelioma after exposure to asbestos are comparable with those in humans (Warheit and Hartsky 1994).
Disadvantages of animal inhalation studies include species differences in respiratory anatomy and function noted above, and species-specific pathological responses in control and treated animals (especially, in the latter, to exposures that result in overloading of the animals ' capacity to clear deposited particles and fibers). Animal inhalation studies for fiber toxicity screening tend to be time-consuming, are expensive, and cannot necessarily elucidate the details of cellular and molecular events.
Despite the limitations, a panel of the World Health Organization (WHO) has concluded that inhalation studies constitute the best available laboratory model for assessing the human health risks posed by exposures to fibers (McClellan et al. 1992; WHO 1992).
Subchronic and chronic inhalation tests are typically used to study health effects and dose-response relationships. Recently, short-term inhalation studies (about 1-day to 2-weeks) with extended followup have been used to study biopersistence, cellular reactions, proliferative reactions, and repair and clearance mechanisms. For studying biopersistence of MVF in this fashion, methods for digesting the lung must be validated. Some techniques for validating the methods, such as low-temperature ashing and digestion with strong acids or bases, have limitations. For instance, low-temperature ashing can make the fibers brittle or artifi
cially break them, and digestion with strong acids or bases can destroy retained fibers or alter their composition. Therefore, alternative digestion techniques that preserve each fiber type need to be developed.
That physical characteristics of the fibers, such as fiber dimensions, play an important role in the pathogenesis of fiber-associated lung disease was demonstrated clearly by Davis et al. (1986), who compared the effects of short and long amosite asbestos fibers at equivalent mass concentrations. Rats were exposed for 1 year by inhalation to aerosols of specially prepared short amosite asbestos fibers (shorter than 5 µm) or long amosite asbestos fibers (longer than 20 µm); the two preparations were derived from the same source and at equivalent gravimetric concentrations. As a result, rats were exposed to greater numbers of short fibers than long fibers. After exposure, no significant histopathological effects were observed in the lungs of rats exposed to the short fibers, but one-third of the rats exposed to the long fibers developed lung tumors. Nearly all the rats exposed to the long fibers also developed diffuse pulmonary fibrosis.
Inhalation toxicity studies in rodents must be extrapolated to humans cautiously. Rats or other rodent species generally are experimentally exposed to high concentrations of preparations of long fibers by enriching the aerosol with the fibers. But, such exposures might not adequately simulate occupational or environmental exposures to lower fiber concentrations or to mixtures of fibers of varied lengths; rather, they are designed to represent a potential worst-case scenario.
INTRATRACHEAL INSTILLATION
Studies that use intratracheal instillation as a route of rodent exposure to fibers are generally regarded as easier and less expensive than inhalation studies. Bolus administration often leads to uneven distribution of fiber-shaped particles throughout the lung and localized overloading (ECETOC 1996). Nevertheless, these types of studies might have value for the initial screening of fibrous compounds. A European Commission (EC) directive for classification and labeling of synthetic mineral fibers (Commission Directive 97/96/EC of December 5, 1997) allows for the use of either the short-term inhalation biopersistence assay or the intratracheal-instillation biopersistence assay in exonerating fibers from classification as a carcinogen. The protocols for performing those tests
have been defined by the European Chemical Bureau (EC 1999). The biopersistence protocols are accepted by the EC for interim use and are being validated in a multicenter ring test. The subcommittee believes that instillation tests are useful for ranking the biopersistence of MVF fibers, but their validation will require data on more fiber types than are presently available. Validation should include the deposition of instilled fibrous material into the alveolar regions of the lung, and correlation of biopersistence of the instilled material (as defined by the investigators) with the development of pathological pulmonary effects.
In spite of the limited data available from these studies, intratracheal instillation of materials remains a popular alternative to inhalation exposure for several practical reasons: small quantities of the test compound can be used, thus reducing waste and increasing safety when hazardous materials are being tested; the technique is inexpensive because it does not require expensive exposure chambers and elaborate vapor or aerosol generation apparatus; complex technical support is not necessary for producing and monitoring vapor or aerosol exposures; and high concentrations of particles or fibers can be administered to the respiratory tract at numerous doses with precise control and measurement.
There are also disadvantages to instillation that stem from the differential distribution in the lung of instilled particles compared with inhaled particles. Instilled particles move to the gravity-dependent portions of the lung because the injected material settles, whereas inhaled airborne particles tend to be well distributed throughout the respiratory system, particularly in the small airways. The high local concentration of instillates or their carrier liquids can cause local tissue damage, particularly at high particle or fiber doses. That can lead to local hemorrhage and even death by mechanisms not directly relevant to the study. The acute inflammatory response that develops in response to the high particle burden and liquid suspension of the carrier could actually contribute to the formation of lesions observed in instillation studies. In contrast, the inhalation technique avoids these local and regional overload effects because the lungs of the exposed animals do not receive the full bolus of particles in one dose. Inhalation models best simulate human exposure because only respirable particles reach the lung parenchyma. Instillation techniques, in contrast, can result in the delivery of nonrespirable (large) particles to the alveolar regions, where they normally would not deposit.
Instillation is an acceptable form of dosing in many cases and might
be the only practical mean of dosing but it cannot substitute for a properly performed inhalation study. This type of toxicity study has been used in fiber clearance and biopersistence studies.
INTRACAVITARY INJECTION
Intracavitary tests, such as intraperitoneal and intrapleural fiber-injection studies, are conducted primarily in rats. In many cases, rats are given abdominal or pleural injections of a bolus that contains from 106-109 fibers and are then evaluated at the end of their lifespan or when a tumor is identified. These tests are known to produce a high incidence of mesotheliomas. Intracavitary models have been advocated as relatively inexpensive and highly sensitive tests to predict the carcinogenicity of fibers (Stanton et al. 1981; Pott 1980). However, the route of administration bypasses all natural defenses, and a single dose (or a few repeated doses) early in life might not necessarily produce the physiological responses that would be observed at lower doses and longer exposures. There is considerable concern that intracavitary models can give false-positive results, even for the prediction of mesothelioma risk, and there is no agreement over their predictive value for lung cancer. The subcommittee agrees with a WHO scientific panel's conclusion that the intraperitoneal model should not be used for quantitative risk assessment or for comparing relative hazards posed by different fibers (WHO 1992).
CONCLUSIONS
It appears reasonable to conclude that extrapolations from animal toxicity data to humans for MVF can best be made when experimental animals are exposed to fibers via inhalation. Studies using instilled doses are valuable insofar as they provide a rough estimate of the pulmonary toxicity of materials, but they should not be used for hazard assessments when setting exposure limits.
Intracavitary exposures, via either intraperitoneal or intrapleural injections, can produce a high incidence of mesotheliomas. Such exposures have been advocated as relatively inexpensive and highly sensitive tests to predict the carcinogenicity of inhaled fibers (Pott et al. 1989). However, this route of administration bypasses all natural pulmonary
defenses, and the single dose (or a few repeated doses) is not physiologically based and can create an overload in the peritoneal or pleural cavity. Intracavitary tests can also yield false-positive results, for the assessment of lung cancer and mesothelioma risks. The WHO consultation (WHO 1992) concluded that the intracavitary model should not be used for quantitative risk assessment or for hazard evaluation of fibers.
Of the three types of tests that can be used to screen for fiber toxicity—inhalation, instillation, and intracavitary—one might be more advantageous than another. Intracavitary tests are not recommended because of the numerous deficiencies discussed above. Results of instillation studies are qualitatively similar to those of inhalation studies (Henderson et al. 1995) and are adequate for short-term estimates of toxicity and fiber-clearance studies, but they cannot substitute for inhalation models for setting dose levels. Short-term inhalation testing should be used for estimating toxicity, evaluating mechanisms, and setting doses for subchronic or chronic inhalation studies. With regard to the latter goal, it is likely that the data generated from short-term inhalation tests could be used to set dose levels for 90-day inhalation studies, thus obviating costly 2-week or 28-day dose-setting inhalation studies.