Abstract
Toxicologic pathologists need to understand the comparative oncology of lung tumors because lung cancer is a common and serious cancer in the human population. Lung cancer in humans is known to be caused by cigarette smoke and a number of other carcinogens in the environment. Animal studies are needed to elucidate possible interactions with other potential carcinogens in environmental or occupational settings. In addition, knowledge of dose-response relationships and potential synergistic effects are needed to minimize harmful effects. Understanding the pathogenesis of common lung tumors will also aid in the prevention, diagnosis and treatment of the disease. Toxicologic pathologists need to remember several important points about lung tumors. The lung cancer response varies among species. Important factors in this variation are the nature of the administered carcinogen, the tissue dose of the carcinogen, the mode of exposure, the sensitivity of the test animal species and the similarity to the human response. Studies of molecular changes are important new tools to understanding lung carcinogenesis. For example, the molecular changes in lung tumors of mice and humans have a number of similarities that may be important in evaluating the significance of compound-induced lung tumors in mice.
Keywords
Introduction
Cancer of the lung is an important disease because of its high incidence in the human population and its poor response to treatment. The lung cancer incidence in women in the United States is increasing, even among non-smokers. In spite of earlier diagnosis and new treatment regimes, response to treatment remains poor. Thus lung cancer remains a high mortality disease. Lung cancer is also associated with several occupations and with environmental exposures.
Studies in animals are important to identify causative agents, effective concentrations of carcinogens, potential synergistic effects and to determine the pathogenesis that aids in the development of new diagnosis and treatment procedures.
A number of factors affect the lung tumor response of laboratory animals. The key points are the species and strain of the animal, the type of carcinogen, the dose of carcinogen to the lung, the sensitivity of the test animal species and the similarity to the human response. Studies of molecular changes are important new tools to understanding lung carcinogenesis.
Species Affects Lung Tumor Response
A number of factors affect the lung tumor response of laboratory animals. A major factor is the species and strain of animal. Table 1 shows the spontaneous incidence of lung tumors in the strains of rats and mice used in the NTP bioassay program and in Syrian hamsters (Hahn, 1995, 1999). The observations times are for 22–24 months. Lifetime observations result in higher lung tumor incidence as the incidence increases fairly rapidly at the end of the life span. Syrian hamsters have a much lower incidence of spontaneous lung tumors. They are also relatively resistant to lung cancer from inhaled, but not instilled carcinogens. This curious feature has resulted in Syrian hamsters being used infrequently in lung carcinogenesis studies.
Another feature shown in the table is the difference between the spontaneous incidence of males and females. Males have a higher incidence of spontaneous lung cancer in most strains of rodents, as in most species. It is well recognized that strains of mice can vary widely in their incidence of various tumors and this holds true for lung tumors. Table 2 shows the wide range of lung tumor incidence in 3 popular mouse strains. The incidence for the C57 and B6 mice are for 2 years of age. The strain A mouse has a high incidence of lung tumors, frequently over 50% by 6 months of age. The strain A mouse is not used for cancer bioassays because of this high incidence, but is valuable for screening and mechanistic studies (Witschi, 2005).
Tissue Dose Varies Among Species
Tissue dose to the lung varies among species based on deposition of the material in the lung and clearance of material from the lung (either by physical means or via metabolism). The deposition is affected by the method of administration and the physical nature of the material (Hahn, 1999). Inhalation is the most natural route for most lung carcinogens and results in the most uniform distribution in the lung. Intratracheal instillation does administer the material to the tissues of interest but is much more likely to cause foci of high concentration due to nonuniform distribution. Such high depositions may overload metabolism mechanisms or cause local toxic reactions that are unrelated to natural exposures.
The physical nature of the administered compound has important effects on the amount and location of material deposited in the respiratory tract (Snipes, 1989). Larger-sized particulates (>5 micron aerodynamic diameter) are deposited primarily in the upper respiratory tract whereas fine particles are deposited in the small airways and alveoli. For gases and vapors solubility is the key factor in determining the site of deposition. More soluble compounds are rapidly absorbed in the upper respiratory tract, while less soluble compounds give a more uniform distribution.
Translocation (physical movement of materials out of the lung) is an important factor in determining the dose to the lung. More insoluble materials can be carried out of the lung by the mucociliary escalator to the digestive tract. More soluble materials may enter the blood stream and be carried to all parts of the body where they may be excreted (i.e., by the kidney or liver) or retained for long periods (e.g., the bone).
Metabolism of compounds is another important factor in determining local dose to the lung (Bond, 1993). The respiratory tract is entirely lined by epithelium that has metabolic capabilities. The nasal epithelium metabolizes inhaled trichloroethylene to the proximate carcinogen for cancers induced in the nasal cavity of mice. The Clara cells in the small airways are responsible for metabolizing inhaled naphthalene to the proximate carcinogen for lung cancers in mice. Neither one of these chemicals induce nasal or lung cancers in rats because rats do not metabolize the compounds to proximate carcinogens in those organs.
Human Pulmonary Carcinogens That Cause Lung Tumors in Rats and Mice
Over 30 carcinogens are designated by the IARC as causing lung cancer in humans. (Hahn, 1998; Gold et al., 2001) An important consideration is how well studies in laboratory rodents correlate with the findings in humans. The designation as a human carcinogen is multifaceted but is based primarily on epidemiologic studies in humans. Bioassay studies in animals are used as indicators of potentially carcinogenic compounds in humans.
Of the 30 known human lung carcinogens, 15 have been tested by 2-year bioassay in rats and/or mice (Table 3). Metals and minerals are asbestos, and compounds of arsenic, beryllium, cadmium, chromium and nickel. Organic compounds are bis (chloromethyl) ether, chloromethyl ether, sulfur mustard and vinyl chloride. Organic hydrocarbons are mixtures: coal tars, coal tar pitches and soot (diesel). Radioactive materials are particles of alpha-emitting compounds and radon gas.
Of the 15 human lung carcinogens tested in rats, 11 caused significant increases in lung tumors. In contrast, only 5 of 15 caused significant increases of lung tumors in mice. None of the metal or mineral compounds gave a positive result in mice. Rats appear to correlate better with the response of humans to lung carcinogens than do mice based primarily on the lack of response of mice to metal or mineral compounds.
Morphologic Comparisons of Human and Rodent Lung Tumors
There are important differences in the morphologic features of rodent lung tumors compared with those of humans (Hahn, 1996, 1998; Nikitin et al., 2004; Table 4). Hyperplasia and adenomas of the alveolar epithelium are very common in mice and much less frequent in rats and humans. The true incidence in humans is clouded by the difference in sampling at autopsy compared with rodents. Lung adenomas and even some focal hyperplalsias in mice may be palpated or seen grossly. The larger lung of humans makes palpation or visualization of small proliferative lesions difficult. In humans alveolar hyperplasia and adenoma are often found as incidental, multifocal lesions in patients who die from primary lung cancer. In lung cancer patients who smoked, anywhere from 10% of those with nonadenocarcinomas to 40% of those with adenocarcinomas have alveolar hyperplasia and adenomas. In contrast, 3% of lung cancer patients who never smoked had such lesions (Wistuba, 2006). Squamous differentiation is common in human lung tumors, especially tumors of the bronchi associated with cigarette smoking.
Squamous differentiation is rare in rats and mice and is generally related to exposure to high concentrations of specific carcinogens, such as inhaled irritant particles in rats or methyl cholanthrene instilled in the trachea of mice. Neuroendocrine differentiation, seen as small cell carcinomas in humans, is not seen in rats or mice. This is an important difference since small cell carcinomas account for a significant fraction (~20%) of the lung cancers in humans. Salivary gland differentiation is infrequently seen in human lung tumors, but is not reported in rats or mice. Combined phenotypes, such as adenosquamous carcinomas, are not uncommon in humans but are seen in rats but not in mice. Metastasis of lung tumors is rare in rats and mice and very rare outside of the thoracic cavity. Human lung cancers are frequently aggressive and tend to metastasize widely throughout the body.
Cigarette Smoke-Induced Lung Tumors—Mouse
Recent studies have shown that lifetime exposure of mice to cigarette smoke results in a significant increase in lung tumors (Hutt et al., 2005). Strain B6C3F1 female mice were exposed, whole body, to mainstream cigarette smoke at 250 mg total particulate matter/m3, 6 hours a day, 5 days a week for 30 months. The survival of the exposed mice was significantly greater than the sham exposed control mice, likely the result of a reduction in body weight. The lung tumor types were papillary adenoma, bronchiolar papilloma and adenocarcinoma (papillary, solid, and mixed patterns; Figure 1). Focal alveolar hyperplasia, a proliferative lesion that typically precedes adenomas was fairly common, occurring in about 18% of the smoked mice.
Bronchiolar epithelial hyperplasia, however, was widespread in nearly all of the exposed mice. Papillomas of the bronchioles were tumors not found in the controls and are rarely reported in mice. The tumors were likely related to the deposition of cigarette smoke in the small bronchioles. Lung tumors were commonly multiple, and occasionally invaded the pulmonary vessel walls, but distant metastasis occurred in <2% of the malignant tumors. Intrapulmonary metastasis occurred in 25–30% of those with adenocarcinomas. The incidence of proliferative lesions, benign and malignant lung tumors was significantly increased in the exposed mice (Table 5). In the exposed mice (n = 330), the incidence of benign tumors was 30.9% and of malignant tumors was 20% (number with tumor/number of mice in group). The incidence of benign and malignant lung tumors was 48% (some mice had both types of tumors). In the control mice (n = 326) the incidence of benign lung tumors was 6.7% and of malignant tumors 2.8%. The incidence for both tumor types was 9.5%.
Cigarette Smoke-Induced Lung Tumors—Rat
Recent studies have shown that lifetime exposure of rats to cigarette smoke results in a significant increase in lung tumors (Mauderly et al., 2004). Strain F344 male and female rats were exposed, whole body, to mainstream cigarette smoke at 250 mg total particulate matter/m3, 6 hours a day, 5 days a week for 30 months. The survival of the high-dose-exposed rats was not significantly different from the control rats. The lung tumor types were bronchioloalveolar adenoma and bronchioloalveolar adenocarcinoma (Figure 2).
These cancer phenotypes are similar to those seen in mice exposed to cigarette smoke. No squamous cancers were noted. Focal alveolar epithelial hyperplasia was common and appeared to be a precursor of the tumors. The lung tumors occasionally invaded the pulmonary vessel walls, but none metastasized outside the lung. The incidence of proliferative lesions, benign and malignant lung tumors was significantly increased in the exposed female rats, but not the male rats (Table 6). In the exposed females (n = 81), the incidence of benign tumors was 8.6% and of malignant tumors was 4.9% for a total of 13.5% for all lung neoplasms. No lung neoplasms were found in the control rats (n = 119).
Molecular Changes in Lung Cancer
Study of the molecular changes in Iung cancer is an expanding area of tumor biology that profoundly affects tumor diagnosis and treatment. The molecular changes in lung cancers depend on the inducing carcinogen, the tumor type and the species of animal (Hahn, 1998).
Dependence on Carcinogen
The frequency of molecular changes in rat lung tumors is dependant on the nature of the inducing carcinogen (Table 7). The variation in frequency of K-ras point mutations, a common mutation in human and rodent lung tumors, ranges from 0 to 100% in rats. For example, a potent pulmonary carcinogen in rats, inhaled tetranitromethane, results in K-ras mutations in 100% of the induced lung tumors. In contrast, the injection of NNK, a carcinogenic nitrosamine common in cigarette smoke, results in lung tumors with no K-ras mutations. This wide variation illustrates that there is not a single, unique molecular pathway to lung cancer. Interestingly, the location of K-ras mutations also seems to depend upon the carcinogen; i.e., codon 12, 13, or 61.
Characteristic mutational spectra in K-ras point mutations have been found in lung tumors induced in mice by tetranitromethane, butadiene, and urethane. These characteristic mutations are important as they may reveal mechanism underlying tumor initiation by these chemicals. In addition, characteristic mutational spectra may serve as signatures of exposure or of dose.
Dependence on Lung Tumor Type
The lung tumor morphology is correlated with the molecular changes. Figure 3 illustrates the development of two different tumor types with different molecular changes of plutonium-induced lung tumors in rats. K-ras point mutations are the early molecular changes in the proliferative alveolar epithelium. If the proliferative lesion progress into squamous cell carcinoma, p53 is dysfunctional. However, if adenocarcinomas develop, p53 dysfunctions are not found (Hahn, 1998).
Dependence of Species
The molecular changes in lung cancer among the species are frequently not the same, but recent studies have shown that some of the major pathways to lung cancer may be similar in mice and humans. For example, with cigarette smoke-induced lung cancer both mice and humans have several similar molecular changes in cancers originating in the alveolar region of the lung (Figure 4). In mice, the K-ras point mutations of codon 12 occur early and are present in the focal hyperplastic alveolar lesions. Hypermethylation of the death associated protein (DAP) kinase and retinoic acid receptor (RAR) β gene promoters are present in the adenocarcinomas that subsequently develop (Hutt et al., 2005). The K-ras mutations are not different in frequency compared to the lung tumors in control mice. However, the malignant tumors from smoke-exposed mice have a greater mutation spectrum at codon 12. In addition, the frequency of DAP-kinase and RAR- β hypermethylation is greater in the smoke exposed mice.
A similar pattern of gene mutation and epigenetic silencing is seen in the adenocarcinomas associated with cigarette smoking in humans. In smokers, the K-ras mutation pathway is favored in lung adenocarcinomas. However, in adenocarcinomas related to other carcinogens, mutations in the epidermal growth factor pathway occur. The 2 pathways seem mutually exclusive (Wistuba and Gazdar, 2006). Mutational inactivation of the p53 tumor suppressor gene is a frequent additional event in lung cancer of humans. The p53 mutations are present in about half of all non-small-cell lung cancers but are infrequent in the earlier lesions (Westra, 2000).
Summary
Toxicologic pathologists need to remember that lung cancer response to administered carcinogens varies among species, strains within a species and gender. Important factors in this variation are the nature of the administered carcinogen, the tissue dose of the carcinogen and the sensitivity of the test animal species. Another important consideration is the similarity to the human response and animals to lung carcinogens. Studies of molecular changes are important new tools to understanding lung carcinogenesis.
Tissue dose to the lung varies among the species based on deposition of the material in the lung and clearance of material from the lung either by physical means or metabolism. Deposition is affected by the method of administration and the physical nature of the material. The physical nature of the administered compound determines the amount and location of material deposited in the respiratory tract. Translocation (physical movement of materials out of the lung) is also important in determining the dose to the lung. Metabolism of compounds is another important factor in determining local dose to the lung.
Of the 15 human lung carcinogens tested in rats, 11 caused significant increases lung tumors. In contrast, only 5 of 15 caused significant increases of lung tumors in mice. Rats appear to correlate better with the response of humans to lung carcinogens than do mice based primarily on the lack of response of mice to metal or mineral compounds.
There are important differences in the morphologic features of rodent lung tumors compared with those of humans. Recent studies have shown that life-time exposure of female mice or rats to cigarette smoke results in a significant increase in lung tumors. These studies have been used to show that some of the major molecular pathways to lung cancer may be similar in mice and humans. For example, with cigarette smoke-induced lung cancer both mice and humans have several similar molecular changes in the cancer originating in the alveolar region of the lung.