Environmental Carcinogenesis and Cancers* - Cancer Research

Environmental Carcinogenesis and Cancers*

W. C. HuEPER

(National Cancer Institute, Bethesda, Md.)

I. T H E N A T U R E AND SCOPE OF E N V I R O N - M E N T A L CANCER HAZARDS

Almost ~00 years have passed since the first environmental carcinogen, coal soot, was recognized as the cause of cancer of the human skin. The environmental carcinogenic spectrum com- posed of accepted, suspected, and potentially carcinogenic agents present in the natural and artificial human environment has grown considerably during the intervening years. Despite the impres- sive increase therefrom, resulting in the number, diversity, and spread of environmental cancer hazards among the general and especially the in- dustrially employed and urban populations and despite the ready availability of an abundant amount of reliable facts and observations on the causation of human cancers by environmental agents, governmental, private, and professional parties directly concerned with the protection and maintenance of human health have displayed until recently an astounding indifference and aloofness toward this important aspect of the general cancer problem. A partial awakening from this general state of lethargy has, however, occurred during the last decade because of the impact produced by the strong reaction of the general public to widely dis- seminated information relative to the increasing and probably dangerous contamination of the human environment and of many products of daily consumption with carcinogenic agents.

This rising concern with mainly industry-related environmental cancer hazard was aroused by three recent developments:

1. The growing contamination of air, water, and soil as well as of foodstuffs with radioactive mat ter from activities in the nuclear energy field.

~. The increasing pollution of the inhaled air with various chemical carcinogens contained in in-

* Based upon a paper presented at a Symposium on the Current Status of Cancer Research at a meeting of the Ameri- can Association for Cancer Research in Atlantic City on April 8, 1961.

Received for publication March 17, 1961.

dustrial effluents and in automobile exhaust as well as associated with cigarette smoking.

3. The demonstration of a considerable and rising number of carcinogenic chemicals used as in- tentional and nonintentional food additives and produced in the food because of certain processing procedures as well as employed as ingredients, or present as impurities in drugs, cosmetics, economic poisons, and other consumer goods.

I t is significant that, in response to these developments, federal, state and municipal legislatures and public health authorities have enacted laws or issued regulations during the last decade providing increased protection of the general public against cancer hazards resulting from an environmental dissemination of carcinogens in foods, drugs, and cosmetics and in effluents of industrial plants and automobiles. For the first time during the modern industrial era, protective legal measures which, up to that time, had been applied to occupational cancer hazards only were extended to those of general environmental nature. Formal recognition was given thereby to the long neglected fact that many of the known environmental carcinogens initially encountered in certain occupational activities are subsequently being introduced into the general human environment as pollutants of the air, water, and soil and as constituents and impurities of many consumer goods, and are creating through this mechanism a serious public health problem.

While mankind has had contact, during the greater part of its existence on earth, with a number of natural as well as man-made environmental carcinogens such as sunlight, ionizing radiations, arsenicals, infections with Schistosoma hematobium, soot, and possibly tobacco products, the human race has had a chance through the thousands of years of exposure to develop against these carcinogens some defense mechanisms in the form of detoxication and chemoimmunity reactions and other protective biologic responses, such as increased pigmentation. Mankind, on the other hand, has not had any opportunities for develop-

84~

Research. on April 11, 2019. © 1961 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

HuEeER~Environmental Carcinogenesis 843

ing any reactions of biologic adaptation to the environmental carcinogens of the modern industrial era, because of their introduction during a period of only a few decades. The protection of modern mankind against the growing number of these man-made chemical and physical carcinogens must rely, therefore, entirely on methods and measures artificially devised.

The extent to which the contamination of the human environment with new artificial, man-made physical and chemical carcinogens has progressed during the past century is strikingly illustrated by the long list of known, suspected, and potential human carcinogens, the various routes of contact with them, the numerous opportunities of exposure to them, the different sites of cancers produced by them, and the progressive march of occupational cancers through the individual countries subsequent to the development of an industrial economy (Table 1) (4, 17, ~ , 56, 68-70, 7~, 73). The presently known spectrum of recognized, suspected, and potential human carcinogens exhibits a remarkable degree of diversity and complexity of its component members. I t is comprised of several parasitic and viral organisms, of a growing number of carbon and silicon macromolecular polymerized chemicals, of various specific organic aromatic and aliphatic chemicals, of several metals and minerals, and of various nonionizing and ionizing radiations which exert an actinochemical action on the chemical constituents of tissues being exposed to these radiations.

I t has often been argued in the past that, because of the multitude and diversity of carcinogenic agents, cancers do not represent an anatomic reaction product of tissues to a specific carcinogenic agent which retains in its biologic respon- siveness and properties a direct relation to the specific causal agent. The prevailing concept con- siders cancers as biologic manifestations which develop in response to carcinogenic stimuli and which assume, once produced, complete biologic independence from the causal agent, thereby be- coming disease entities per se. Almost the entire modern diagnostic and chemotherapeutic research in cancer control is based on this concept. This approach is rather surprising and unique in the annals of medicine and contrasts sharply with the efforts made in the past for controlling infectious diseases. These efforts were directed toward influ- encing the causative agents and their toxic products by biologic and chemical means. I t has not occurred to investigators engaged in such tasks to a t tempt the development of curative agents against these diseases by first obtaining detailed biochemical information on their anatomic reae-

tion products, such as tubercles and gummas, and then to use such information for devising chemical means against some of the biologic mechanisms or some special constituents found in such circum- scribed anatomic reaction products, or for developing diagnostic tests for infectious granulomas, ir- respective of their specific etiology.

Because of the past failure to obtain, by the application of this biologic concept of cancer, any reliable specific diagnostic test for cancer and any effective chemotherapeutic agent against cancerous tissue, one is left wondering whether it is not appropriate and timely to put to work in this research the many facts and observations on environmental carcinogenesis and cancers, and to re- turn to those scientific principles which brought success to the control of infectious diseases. The advisability to explore more fully the potentialities

TABLE 1

TIMETABLE OF THE DISCOVERY OF AROMATIC AMINE CANCERS OF TIIE BLADDER IN DIFFERENT

COUNTRIES

Country Discoverer Year

Germany Switzerland Great Britain Russia Austria United States I ta ly .Japan France

Rehn Schedler ROSS Rosenbaum & Gottl ieb Schtiller Ferguson el al. di Maio Nagayo & Kinosita Billiard-Duchesne

1895 1905 1918 1926 1932 1934 1936 1940 1946

of the etiology-specific concept of cancer for its diagnosis, prophylaxis, and treatment is suggested by the many similarities existing between the spectra of environmental carcinogens and of pathogenic microorganisms.

Both major systems of disease-producing agents include a wide range of different component members varying markedly in their degree of potency or virulence among each other--i.e., from producing a 100 per cent at tack rate in the population to the risk of affecting only a fraction of it or the ex- ceptional individual. There exist, moreover, definite and epidemiologically important differences in their pathogenicity or their carcinogenicity, respectively, for different species, strains, races, and individuals of the same race or strain.

This observation on the significance of the host reaction for the occurrence of a carcinogenic response has led to the contention that the host relationship might be more important in this respect than the carcinogen proper and that, therefore, a control of cancer hazards might be achieved by



844 Cancer Research Vol. ~1, A u g u s t 1961

modifying in the right direction the constitution- ally conditioned reaction of the host organism to the carcinogen. I t may be pointed out in this connection that despite distinct congenitally or acquired constitutional differences among equally exposed individuals, the realization of a cancerous reaction or of a tissue response to an infectious agent, like the tubercle bacillus, is fundamentally determined by the action of the pathogen. With- out the primary action of the pathogen the host reaction is immaterial. Past experiences with con- stitutionally conditioned or genetic diseases have, moreover, shown that their control is most difficult. There is also no sound scientific reason in support of the recently revived, speculative allegation that, in contrast to pathogenic microorganisms, environmental chemical carcinogens are not primary carcinogens but act in the process of carcinogenesis merely as agents triggering the specific action of genes or even viruses. The entire symptomatic pattern of manifestations associated with environmental and experimental chemical and physical carcinogenesis militates against the validity of such a concept.

Additional similarities between environmental carcinogens and pathogenic microorganisms are related to the fact that both are capable under certain conditions of eliciting chemo-specific allergic reactions, indicating thereby symptomatical- ly the occurrence of an interaction of the two types of pathogens with proteins of the host organism. The degree of response, as well as the site of reaction in the host for both pathogens, depends to some extent on the particular route of contact, on the intensity and duration of exposure, and on the modifying influences which other exogenous factors, such as diet or ehemotoxie manifestations, might exert on the realization and type of tissue changes resulting. Finally, it may be noted tha t man is in contact throughout his entire lifetime not only with a variety of microorganisms (some with known pathogenicity to man) without developing from all exposures and in all instances an infectious disease, but also to diverse chemical and physical carcinogens acting on him from various sources and at different times and failing to elicit during his lifetime, in most instances, a cancerous response of some tissue. I t seems that the total effective carcinogenic burden of an individual thus is, in part, determined by the various factors mentioned; to another part it seems to depend upon the degree of synergistic, antagonistic, or indiffer- ent interaction of the various carcinogens and upon their interplay with noneareinogenie endoge- non s and exogenous factors acting in the host organism. Similar interrelationships are operating in

determining the activity of pathogenic microorganisms, such as, for instance, tubercle bacilli, in the host organism.

From these considerations it may, therefore, appear wise to pattern the future at tack on environmental cancer hazards on the principles and methods of investigation and the concepts of biologic mechanisms employed in past studies of the causation and the control of infectious diseases.

II. F U N D A M E N T A L ASPECTS OF ENVI- R O N M E N T A L C A R C I N O G E N E S I S

I t is one of the fortuitous and fortunate by- products of the recent emergence and growing ap- preciation of environmental cancer hazards as major public health problems that many of their scientifically and practically important but often ill defined and badly understood facets werc placed into sharper focus. The immediate need for considerably increased and comprehensive investigations on many fundamental and applied aspects of the cnvironmental cancer problem became rather painfully apparent during the recent hearings of several Congressional Committees (11-16) which led to the inclusion of the Delaney cancer clausc into thc laws enactcd in relation to health hazards from food additives and food and cos- tactic colors. The various points of dispute which arosc at thcse occasions illuminated thc large gaps in factual knowledge on environmental cancers in man and on carcinogenesis by environmental agcnts in experimental animals and demonstratcd the appreciable uncertainty which cxists regarding somc of the most important scientific concepts on these subjects.

DEFINITION OF A CARCI:NOGEN

During these discussions, as at other occasions, difficulties were encountered in arriving at a generally acceptable definition of a carcinogen. Al- though a definition of this basic aspect of environmental carcinogenesis which fulfills all scientific re- quirements does not exist at the present time, it is nevertheless possible to devise a definition which is satisfactory for most scientific conditions and which is practical and applicable for medicolegal purposes.

Carcinogens may be defined as chemical and physical agents which are capable, under proper conditions of exposure, of producing in animals, including man, cancers which would not occur without the intervention of thesc agents. Carcino- gens thus do not merely produce a significant increase in cancer incidence when administered at any dosage level, by any route of administration, and to any spccies or strain, but elicit cancers lo-



HuEeErt--Environmental Carcinogenesis 845

cated ordinarily at sites associated with induced carcinogenesis. This rather inclusive definition of a carcinogen which is applicable also to so-called conditional (Steiner) (75), (Nau) (57), functional (Hueper) (33, 38), hormone-dependent (Morris) (53), and indirect carcinogens avoids the difficulties connected with the use of definitions containing more or less arbitrary restrictions.

I t is obvious that several physiologically active hormones and nutrients such as estrogens, iron, cobalt, and, possibly, selenium must be considered according to this definition as carcinogens. Their carcinogenic properties may become manifest if these substances are produced in or introduced in- to the organism in abnormal amounts or under abnormal circumstances, such as those prevailing at times under occupational, medicinal, or dietary exposures, or created under experimental conditions. Since the physiologic nature of these chemicals has been advanced as an argument against their role as carcinogens and has been used for sup- porting the demand that they be exempted from any legal restrictions as food additives and con- taminants, it may be pointed out that chemicals of animate and inanimate origin occurring in and produced by nature are in no way superior to or different from those made by man in carcinogenic respects (ultraviolet and ionizing radiations, arsenicals, chromium, nickel, asbestos, ergot alkaloid, senecio alkaloids, chili, crude petroleum, viruses).

The claim has recently been advanced by several investigators (Eckardt [19], Bonser 1) that sarcomas produced in rats at the site of the subcutaneous introduction of test materials of any kind represent by themselves not adequate proof for the carcinogenicity of the substances eliciting such cancers. Such allegations have become, during recent years, of more than purely scientific importance, because they have been employed to dis- credit the carcinogenic significance of several chemicals (dyes, plastics, paraffin, etc.) which elicited cancers in rats and which are present in consumer goods. The legal objections against their use as food additives were thereby removed in some instances. The Food and Drug Administra- tion, U.S. Department of Health, Education and Welfare, for instance, adopted this concept some time ago in its interpretation of the legal significance of the sarcomas produced in the subcutaneous tissue of rats given injections repeatedly of various food dyes (light green, SF, brilliant blue, fast green) and with carboxymethylcellulose, and has for this reason permitted the continued incor-

1 Personal communication, Dr. G. Bonser, University of Leeds, Leeds, Eng.

poration of these carcinogenic chemicals into human foodstuffs.

The same considerations apparently have recently been applied by the Food and Drug Adminis- tration and by the Council on Drugs (9) of the American Medical Association in justifying the continued parenteral medicinal use of an iron- dextran complex (Imferon), although several British investigators, including Haddow (~6), had shown that repeated intramuscular injection of large doses of this preparation into rats and mice produced sarcomas at the site of deposition, and despite the fact that Imferon had been withdrawn from the British market. In fact, the initial but later revoked withdrawal of Imferon from the American market was not based on an application of the Delaney cancer clause to the Food Additive Amendment. Only time will tell whether or not an arbitrary, unwise, and perhaps expedient interpretation of experimental evidence and the refusal to learn from previous experiences may not result in the repetition of a rather disastrous situation similar to that which developed during the past decade as a sequela of the ill considered use of thorium dioxide in past decades.

I t is noteworthy that, according to observations made by investigators of the Food and Drug Ad- ministration (Nelson et al.) (58), which for many years had employed this screening technic, only some, but not all, of the various subcutaneously administered food dyes elicited sarcomas at the site of deposition. Evidence obtained on several thousand rats which received different chemicals by the subcutaneous, intramuscular, intrapleural, intraperitoneal, and intravenous routes have clearly demonstrated, moreover, that the mere presence of a mechanically irritating material in the connective tissue of rats does no t provide a carcinogenic stimulus unless the chemical introduced possesses carcinogenic properties (Hueper and Payne [40]; Hueper [3~, 35, 86]). I t has been found, furthermore, that many of the chemicals showing carcinogenic properties when given to rats by the subcutaneous route also produce sarcomas in mice under identical experimental conditions (Payne [63, 64]; Hueper and Payne [39]). In fact, a sarcomatogenic action of plastics implanted into the subcutaneous tissue is not restricted to rats and mice but has also been noted in hamsters.

I t can scarcely be maintained that the objections raised against the carcinogenic significance of tissue responses produced by chemicals in the connective tissue of rats should be limited to connective tissue located in the subcutaneous area. They cannot plausibly be applied to other mesen-



846 Cancer Research Vol. s A u g u s t 1961

chymatous or epithelial tissues of the rat, such as epidermis, muscle tissue, mucosal epithelium of the bladder and liver, for which, in part, similar differences in species-specific susceptibility to carcinogens have been noted. Despite the need for additional information on these aspects of environmental and experimental carcinogenesis, it would be most unwise to deny that the formation of subcutaneous sarcomas in rats following a subcutaneous deposition or repeated introduction of some test material is not an indication of the carcinogenic property of the chemical tested (Hueper and Payne [39, 40]; Hueper [3~, 35, 36]).

The adoption of such a thesis as a matter of scientific policy and experimental practice would indeed lead to an abandonment of the subcutaneous route of administration of test materials in a l l species. Since such a procedure has practically been suggested in a recent treatise (Publ. 749) (e3) on technics suitable for the experimental evaluation of carcinogenic hazards from the use of food additives prepared by the Food Protection Committee, it is appropriate to emphasize that such advice is scientifically unsound and practically operating in favor of the inclusion of carcinogens into the general food supply as advocated by some representatives of the food processing industries as well as by some of their scientific sympathizers and con- sultants.

Closely related to the controversy of carcinogenic pi'operties, in general, of environmental chemicals is the dispute whether or not the demonstration of carcinogenic responses in animals or man from chemicals applied to the skin or injected into the subcutaneous tissue represents a valid or probable indication that the test material may exert, a similar effect upon the other tissues of contact in the host organism, especially when it is either inhaled, such as cigarette smoke, or ingested, such as food additives. In fact, the demand has been made to distinguish between universal carcinogens, such as ionizing radiations and possibly ~-acetylaminofluorene, capable of producing cancers in all types of tissues upon proper exposure, and carcinogens whose specific action is limited to certain tissues and organs and whose local action depends in part on the route of introduction of the carcinogen, in part upon special tissue affinities. The majority of the known carcinogenic chemicals, therefore, would evidently belong to the lat- ter class, although most of them have not been adequately tested in this respect. The designation of a chemical as a carcinogen, according to this concept, would have to be specified and comple- mented by the additional information on the particular species, strain, tissue, and route of intro-

duction in which and by which the carcinogenic property of the test material became manifest.

Because of the highly defective information available on the target organ of the carcinogenic action of chemicals tested in animals by various routes, the adoption of such a classification of carcinogens and, particularly, its practical application for the determination of potential cancer hazards in man is unwarranted at this time. Under the prevailing conditions decisions based on the above principles would reflect mainly lack of pertinent factual information and would not be derived from concrete positive or negative evidence obtained from properly designed experiments.

There exist only some fragmentary data concerning such relationships which are useful in these respects.

Observations in man and animals indicate that agents causing cancer of the skin, such as arsenicals, mustard gas, and radioactive chemicals, may when inhaled produce cancer of the lung. Arseni- cals and radioactive chemicals and possibly also mineral oils, when ingested, may elicit cancers of the digestive tract, such as the liver and the intestine or of other nondigestive organs, such as the bones, respectively. Whereas benzidine, 4-aminodiphenyl, and several of its derivatives which are bladder carcinogens to man and dogs, when subcutaneously injected into rats, have given rise to the development of carcinomas of the intestine, evidence is lacking whether or not a similar effect can be produced in rats or in other species when such chemicals are fed, although it is suspected that benzidine may elicit, under occupational conditions of exposure in man, not only bladder carcinomas but also carcinomas of other organs, such as the intestine and lung. Since, under such circumstances, benzidine penetrates the skin, is inhaled, and is ingested, it is not possible to state whether such multifocal effects are attributable to the introduction of benzidine through one or several of those routes.

General experimental experience indicates that the respiratory and digestive systems of rodents are apparently less susceptible to the action of various human and experimental carcinogens than those of man. Because of the distinct limitations to be placed upon the value of results obtained by the introduction of test materials through these routes in experimental animals, it appears to be most unwise, if not dangerous, to attach any decisive importance as to their significance to human cancer hazards on a negative outcome of tests in which the alimentary and respiratory routes were employed. At the present state of our knowledge on these phases of experimental carcinogenesis the



HuEPER--Environmental Carcinogenesis 847

apparent inability of a chemical given by mouth to produce cancers in experimental animals does not justify the conclusion tha t such a chemical does not exert such an effect in man and thus is not a dietary carcinogen.

As long as these and other methodological un- certainties concerning the identification of carcinogens prevail, it is advisable, as a matter of wise precaution, to exclude as far as possible all environmental carcinogens from contact with man so as to avoid any subsequent epidemic-like occurrence of cancers attributable to an indiscriminate decision made 10-80 years ago.

CRITERIA OF CANCER

I t is doubtless correct that the majority of cancers occurring spontaneously or induced in man and animals fulfill the various criteria of cancer developed on an empirical basis and generally ac' cepted as characteristics of malignant neoplasms. I t has been known, on the other hand, that excep- tions to these rules are not infrequent. The histologic differential diagnosis between a squamous- cell carcinoma of the skin and a keratoacanthoma, both recently observed in workers having contact with hydrogenated coal oils (Sexton) (74), is often difficult or impossible. This distinction can be made at times only on the basis of the subsequent benign or malignant course which a particular tumor assumes. Clinical experiences in coal tar workers and aromatic amine workers also have shown that polypous growths of the skin and of the bladder mucosa, respectively, presenting the gross appearance of cancerous reactions, may regress spontaneously (false positives) (Hueper [~]; Gold- blatt [~5]). Ulcerative lesions in the skin of mon- keys produced by the prolonged application of coal tar and exhibiting, upon histologie examina- tion of biopsy material, the characteristics of epi- dermoid cancers likewise may ultimately heal, despite the continued application of the carcinogenic material. These observations evidently indicate that the histologie demonstration of malignancy is not always proof of the biologic malignancy of a neoplastic lesion.

Another striking and rather frequent exception from the criteria characterizing cancers is repre- sented by basal-cell carcinomas of the skin, which may develop following an occupational or environmental exposure to sunlight, coal tar, and arsenicals. Basal-cell cancers of the skin do not produce metastases and thereby lack one of the most important properties of malignant growths. I t may be pointed out in this connection that an appreciable number of environmentally important carcinogens (asphalt, coal tar, uranium, nickel,

chromates, shale oil, hydrogenated coal oils, coffee soot, 8,4-benzpyrene), when subcutaneously or intramuscularly introduced in rats and mice, will produce cancers only at the site of administration but often do not form metastases, although they show extensive infiltrative and destructive growth and often a high degree of anaplasia (Hueper and co-workers [3~, 35, 36, 39, 40, 65]).

These observations have distinct scientific and medicolegal importance since Brunner (7) recently demanded that the designation of "sarcoma" should be withheld from the malignant mesenchym- atous neoplasms which develop around subcutaneous implants of various plastics because of their low tendency to form metastases. Since the majority of cancers of the bladder produced in man and dogs by contact with carcinogenic aromatic amines do not form metastases in regional lymph nodes or remote organs, the absence or low tendency of such manifestations of discontinued cancerous growth cannot be considered as a scientifically sound reason for assigning to such neoplasms a special position and for placing them among the noncancerous growths. The carcinogenic nature of the chemicals causing these cancers would be denied by such a decision.

During recent years repeated attempts have been made to relate specific histologic types of cancers to a specific etiologic agent (Hueper) (34). I t was, for instance, contended that all or most squamous-cell carcinomas of the lung were due to exposures to cigarette smoke and that adenocarci-

n o m a s were of a different etiology. I t may be pointed out here tha t abundant observations with various specific occupational respiratory carcinogens attest to the fact that tile same agent may elicit different histologie types of lung cancer. Experiences gained with various environmental carcinogens tested on animals have also demonstrated the absence of correlations between these two aspects. Recent reports based on a detailed, if not minute, histologie and histoehemieal analysis of hmg cancers observed in this country, such as Los Angeles, and abroad, such as Japan and I ta ly (Kreyberg) (46), have, moreover, shown that the increase in lung cancers affected in these areas mainly the adenocarcinomatous variety.

Because of these irreconcilable discrepancies it may, therefore, be wise to investigate the possibility tha t variations in the general and local histologic type of cancers may depend upon the speed with which a cancer develops from the tissue of origin and with which it grows later in various parts. I t seems to be reasonable to assume that a cancer which develops rapidly from a tissue exposed to a carcinogen and not, or little, damaged



848 Cancer Research Vol. ~1, Augus t 1961

by this agent would duplicate the original tissue from which it is derived--i.e., for instance, glandu- lar tissue in the bronchial mucosa. If, on the other hand, a cancer originates from a tissue which has undergone for some time chronic, reparatory, and metaplastic alterations, it is likely that the cancer would reflect in its structure the metaplastic tissue from which it is derived. Such conditions may prevail whenever a bronchiogenic carcinoma develops from metaplastic areas of the bronchial mucosa affected by chronic inflammatory processes of chemical, bacterial, or viral nature. Similar correlations apparently prevail in determining the predomi- nant histologic type of cancers of the bladder attributable to exposures to aromatic amines and to Schistosoma hematobium infections. The aromatic amino cancers originating from a bladder mucosa which usually does not display any chronic inflammatory reactions are, as a rule, of the transitional- cell type. Cancers of the bladder appearing in in(ti- viduals suffering from chronic Schistosomiasis of this viscus associated with severe inflammatory and leukoplakic reactions, on the other hand, are predominantly of the squamous-cell type.

The histologic type of a particular cancer, therefore, seems to be the combined result of the relative potency of the carcinogenic action and the reactive status of the host organism.

CAUSATIVE ~[ECHANISMS The various theories developed through the

years concerning the causative mechanisms by which chemicals of various types interact with constituents of the cell in the canccrization process have been, in part, of a generalized and, by necessity, rather diffuse type, such as the chronic irri- tation theory, the somatic mutation theory, the allergic theory, the Warburg's theory of anoxic fermentation, the hypotheses concerning the role of free radicals, peroxides, and sulfhydryl groups, and the postulate of the activation of abnormal carcinogenic genes or specific viruses by carcinogenic chemicals. Other theories are of more restricted nature applying to some specific type of carcinogenic chemical, such as the polycyclic aromatic compounds--i.e., the Pullman electronic theory; and some structural characteristics of the members of these chemical groups, as well as of some carcinogenic aromatic amines--i.e., the formation of ortho-hydroxymetabolites. Whereas some of the general type theories have also been used for explaining the action of carcinogenic metals, such as the sulfhydryl theory and the hy- pothesis on the formation of intracellular protein complexes, only Warburg's theory of cellular anoxia and the theory on the formation of protein

complexes have so far been offered in an a t tempt to account for the development of polymer cancers (Hueper) (37), if one disregards the recently advanced speculative assumption that some mys- terious and nonspecific forces emanating from the surfaces of implanted plastic films are responsible for the neoplastic phenomena observed (Noth- durft [59]; Oppenheimer et al. [6~]).

Apart from the purely scientific benefits ensuing from a better understanding of the causative mechanism or mechanisms of carcinogenic agents, the availability of such information would be of distinct advantage for distinguishing between carcinogenic and noncarcinogenic environmental chemicals and woul(t facilitate the development of more efficient, more reliable, and faster screening procedures. At the present time none of the various theories is of any real value for this purpose.

I t is also probable that an adequate knowledge of the causative mechanism of environmental car- cinogcns would help in the development of etiology-specific diagnostic procedures and of rational and effective prophylatic and therapeutic measures. Since some of these carcinogens are recognized human allergens, such as the various metals, and others arc known to form complexes with cellular proteins (various azodyes, quinones formed from carcinogenic aromatic amines, carcinogenic polycyclic aromatic hydrocarbons) (Hueper [~9, 30]; Mayer [48]; Miller and Miller [49]; IIeidel- berger et al. [~7]; and others) and thereby may become antigens, a comprehensive experimental ex- ploration of a chemo-specific allergenic mechanism of cnviromnental carcinogenesis seems to be worth while. Experimentation with noncarcino- genie chemicals structurally similar to carcinogenic chemicals for testing the usefulness of structural blockade as a prophylactic measure and for determining the importance of specific spatial con- figuration and qualitative and quantitative structural factors for the interaction of exogenous and endogenous agents in the cancerization process offer an additional promising approach for advanc- ing the control of environmental cancer hazards and cancers.

Impor tant facets in these efforts, which depend in part upon a more exact and detailed information on the causative mechanism of carcinogens, are associated with the selection of tile proper animal species suitable for the screening of chemicals for carcinogenic properties and with the extrapolation of observations made in animals to man in both qualitative and quantitative respects.

The decisive significance which specific mctabo- lites, formed by some species but not by others, possess in determining the occurrence as well as



HuEPER--Environmental Carcinogenesis 849

the site of a carcinogenic response has been strikingly illustrated by the role which ortho-hydroxy- derivatives of certain carcinogenic aromatic amines, such as fl-naphthylamine, 4-aminodiphenyl, and benzidine, play in the production of bladder cancers in man and dogs, and in the absence of such neoplastic reactions in mice and rats in which these metabolites are essentially absent. The development of cancers at other sites in these species, when given the aromatic amines mentioned, on the other hand, indicates that these species apparently produce metabolites which have

alterations in carcinogenic properties and relative potency of several chemicals which result from a shift in the position and the relative length of their sidechains (polycyclic aromatic hydrocarbons: benzpyrene, dibenzanthracene) and occur subsequent to the introduction of fluorine into the ring (aromatic amino compounds) (Miller and Miller) (49).

The increased reliance of the modern economy on products of the chemical industry and the growing awareness of the carcinogenic nature of some of its products make an early acquisition of

T A B L E

TARGET ORGANS OF CANCERS PRODUCED BY AROMATIC AND AZO COMPOUNDS IN DIFFERENT SPECIES

CI~EMICAL

Benzidine 3-Naphthylamine -Aminodiphenyl

9-Dimethylamino- azobenzene

LAminofluorene -Aeetylaminofluorene

V-Hydroxy-~', 3-azotoluene

~,S'-Dimethyl- azobenzene

IN. N-Diacetyl-o- aminoazotoluene

-Amino-2',8-dimethyl- azobenzene

t~cetyl-o-aminoazotoluene

~-Toluidine Chrysoidin ?onceau R Dil Orange XX Drange SS Ponceau 3R

MAN DoG RAT

Bladder Liver Intest. Bladder

X?

Liver

MOUSE

Intest. Bladder Liver Infest. Bladder Liver Intest.

x x

x

x

x

X X

X X X

an affinity to and exert a carcinogenic action on other organs such as the liver and the intestine. Because of these shifts in target organ (Table 2) and in carcinogenic response to various aromatic amines by different species, experimental results in animals would become more meaningful, as far as man is concerned, if the fundamental biologic principles underlying these fluctuations were known and could be systematized, permitting thereby a more valid and reliable extrapolation of the results of screening procedures to man than is possible and justifiable at present.

Additional advantages would accrue from such knowledge, not only to the study of the environmental cancers but also to the future development of noncarcinogenic economically important chemicals by industry, if the reason were known for the

an adequate knowledge of the basic factors underlying, modifying, and controlling chemical carcinogenesis an urgent necessity. The availability of sufficient factual data which may serve as guides for avoiding the development of carcinogens in the production of new chemical compounds would protect the chemical and related industries against engaging in research which might prove to be needlessly costly, when it inadvertently leads to the development of carcinogenic compounds.

POTENCY OF CARCINOGENS AND " S A F E " D O S E

The emphasis which has recently been placed upon the determination of the absolute and com- parative potency of carcinogens in experimental animals, for the purpose of making educated



850 Cancer Research Vol. ~1, A u g u s t 1961

guesses at their so-called "safe dose" to man, has resulted in uncovering the many methodological difficulties existing in this respect and in demonstrat ing the distinct lack of fundamental facts for arriving at a "safe dose" which is safe for the great major i ty of members of the human populations at risk--i .e. , mainly the general public and the general consumer of industrial goods.

Abundant experimental experience has clearly shown tha t the absolute tumor yield, as well as the length of the latent period for carcinogenesis by a part icular carcinogen, may vary in wide limits not only among different species but even among different strains of the same species and among the two sexes of the same strain, al though identical conditions of exposure are being used. Almost 100 per cent of dogs, for instance, developed bladder cancer within ~ years when given optimal doses of ~-naphthylamine by month, whereas only about 10 per cent of mice fed this carcinogen showed, within this period, cancers of the liver (Bonser et al. [5]; Hueper et al. [41]). Another carcinogenic

aromatic amine, namely, !~-acetylaminofluorene, administered orally to rats, elicited cancers in various organs (liver, intestine, ear duct, lung, etc.) in most of the t reated rats within a year (Bielschowsky et al. [3]; Morris et al. [54]), whereas dogs given this chemical by the same route re- quired for the development of cancers of the bladder and of the liver a minimum exposure period of over 5 years (Morris and Eyestone) (55).

Recent experiments with a number of hexava- lent chromium salts have shown tha t the physical properties, i.e., the degree of water solubility of the individual compounds, control both the tumor yield and the length of latent period (Table 8). The relative biologic availabili ty of carcinogens when in contact with the host organism, therefore, influences the potency of a carcinogen by controlling either the relative amounts of carcinogen released within s tandard periods from depots or by determining the length of time it s tays in contact with the tissues at any one site because of its excretion or removal rate.

TABLE 3

CANCEROUS RESPONSES AT THE SITES OF IMPLANTATION OF SEVERAL C]JROMIUM COMPOUNDS IN THE THIGH MUSCLE AND PLEURAL CAVITY OF RATS

ROUTE C(}MI]OUND 85 ~ATS

CUMULATIVE TOTAL OF CANCERS/NUMBERS OF S~s RATIO AT MONTHS AFTER START

0-3 4-6 7-9 10-12 1 8 - 1 5 1 6 - 1 8 19-~1

Chromic chromate I .M. 11/~0- 16/16 I.P1. 0/34 8,/~3 15/15

Calcimn chromate I.M. 0,/32 5/~25 8 /~ I.P1. 0/8~ 1/80 16/10 ~0/0

Sintered calcium I.M. 0/84 5/;~7 7/0-5 8/~q* chromate I.P1. 0/33 13/16 17/9

Strontium I.M. 1/38 3/~9 9/~0 chromate I.P1. 0/~8 ~ / ~ 6 7 / ~ 0 13/9

Barium chromate I.M. 0/34 0/38 0 / 3 1 0/30 I.P1. 0/'81 O/30

Lead chromate I.M. 1/33 1/~8 I.P1. 0/84 0/3~

Sodium dichromate I.M. 0/38 0/81 0/~5 1.el. 0/~6 0/~4 0 / ~ 1 0/~0

Chromite roast I.M. 0/84 0/3~ 0/3~ residue I.P1. 0/32 1/~8

Zinc yellow I.M. 1/34 9/~6 11/~ I.P1. 0/33 1/32 7/t3 17/11

Chromium acetate I.M. 0/34 0/38 0/30 I.el. 0/34 0/83 0 / 3 1 0/34

Sheep fat I.M. 0/3~ 0/80 I.PI. 0/34 0/8~ 0/30

LATENT PERIOD

i r162 !r

8/~0

i

9/18 , 17/6

13/11 14/7

0 / ' ~ 7 1/~0

1/~ 0/+30

0/~0 o/11

0/~7 3/r

13/16 ~1/6

0/~0 0/18

0/~3 o /~

Per cent 22-24 ] 25-~7 yield I

[

~8/5 ~3/5 ~5/~

9/13 9/9 9/7

9/11 17/s

15/4 16/4

o/17 1/14

1/18 1/~o

o/16

0/15 3/18

18/12

1/17 0/16

lO/4 17/1

15/o 16/3

0/14 1/8

1/12 s/15

1/lO 5/15

16/9 ~ /3

1/17 1/15

0/lO 0/11

1~/0 17/0

17/o

o/7 1/5

1/8 5/lO

o/6 O/5

! ~ 6 / 0

9/0

o/o 1/0

1/~ 5/0

0/0 0/0

I

o/15 0/18

I 7O i 75

25 i 60

85 50

[ 45 5O

J 0 5

0 0

3 15

45 65

Mean Minima

14 14

18 16

19 i 19 13 i 1~2

6 5 10 6

1.6 16 19



HvEPER--Environmental Carcinogenesis 851

From the results of recent experiments undertaken for comparing the relative carcinogenic ef- fectiveness on mice of a potent carcinogen, 3,4- benzpyrene, when standardized amounts were administered subcutaneously in a single injection with that obtained when the same total amount of carcinogen was given by repeated monthly injections, it appeared that the effects produced de- pended not only on the amount administered but also on whether it was given in a single dose or in multiple doses. The carcinogenic response under these conditions increased when the total dose was administered in increments (Chart 1).

Of distinct significance in regard to the deter- ruination of carcinogenic potencies was also the observation of a plateau effect seen in mice given 3,4-benzpyrene by single injection after an optimal quantitative exposure had been reached. In contrast, there was a continued rise in the cancer yield with a progressive increase in the total dose of carcinogen injected when it was given in frac- tionated doses (Payne and Hueper [65]). Since a low-level, recurring exposure to a carcinogen, according to these findings, is more hazardous than a single exposure to the same total amount, and

because man sustains, as a rule, the first type of exposure to environmental carcinogens, any potency rates obtained in experimental animals by a single administration of a carcinogenic test material would probably give results with little or no significance to man.

The possible importance of a third factor which may determine the outcome of potency studies of carcinogens was suggested by the results of experiments on the character and type of metabolites appearing in the urine of rats following an oral administration of different amounts of a certified food dye, Citrus Red No. 2 (Conway and Lethco [8]). I t was found that the number of demonstrable urinary metabolites increased with an increase of the dose of the dye given and that there occurred a shift in the quantitative relations between the chemically different metabolites under these experimental conditions. In view of the fact that the chemical character and the relative quantity of individual metabolites of urinary carcinogens seems to influence the carcinogenic response produced in the bladder by some of the known carcinogenic amines, it is probable that purely quantitative factors may decisively influence not only the outcome of experimental potency studies but also the degree of cancer hazard from such chemicals in exposed human populations. These observations, moreover, illuminate the frequently raised problem concerning the existence of marked dif- .fere~:ces in the susceptibility to environmental

carcinogenic agents by different members of the general population, since it consists not only of so- called normal individuals but also of a considerable proportion of metabolically abnormal ones who may possess a reactivity to carcinogens different from that of the normal group.

The importance of the reactive status of the host organism in determining the potency of environmental carcinogens and, consequently, any calculation which one might dare to make in regard to a "safe dose" has recently received added attention by the demonstration of a greatly increased susceptibility of newborn mice to carcinogenic polycvclic aromatic hydrocarbons (Pietra et al. [66, 67]). Since previous experiences with a trans-

I I I I I I 9 0

E- z 80

~...~ Administered in 7 0 . - , " " 12 m o n t h l y d o s e s -

~ 6 0 f - / k - / m /

50 -- // /

/ O3 4 0 - / g I

I A d m i n i s t e r e d in I.-- 20 / f s i n g l e d o s e _ o _

/ I0

-~5: o / I I I , I I. 0 0.I 0 .2 0,3 0.4 0-5

T O T A L D O S E OF" B E N Z P Y R E N E ,

0,6 0.7

M I L L I G R A M S

CHART 1.--Production of tumors in mice with varying amounts of benzo[a]pyrene in single and repeated monthly doses. Per cent of mice with tumors and total dose.

placental transfer of urethan from pregnant mice to their offspring resulting in an early and accentuated appearance of puhnonary tumors in the young has a similar connotation, experimental potency determinations with prospective application to man must take these observations into proper consideration. In fact, infants and children, if these findings in animals should apply also to man, would be particularly vulnerable to prolonged exposures to rather small and perhaps minute doses of environmental carcinogens present in foodstuffs, including the maternal milk, in the air, and in drinking water. Because of their prospective long lifespan, infants also face the additional liability of a long latent period, usually necessary for weak carcinogenic exposures to result in manifest tumor formation. There is an additional health hazard from such intrauterine exposures of the



85~ Cancer Research Vol. ~1, August 1961

fetus to carcinogenic agents, such as x-rays, radioactive substances, trypan blue, and others, because they may result in the development of congenital malformations.

These observations and considerations of a methodologic nature, as well as the absence of any dose-response curves for human carcinogens in experimental animals, do not justify the claim that any valid conclusions can be drawn from dose- responses obtained with a few experimental carcinogens in experimental animals on the reactivity of environmental carcinogens of all kinds and of all types of contact in members of the human population. The control of environmental cancer hazards could easily be made by the adoption of a "safe dose" of carcinogens into a lottery of human lives.

Added to the methodological objections to this conccpt must be the fact that the exposure conditions as to source, variety, and total dose of environmental carcinogens are, in most instances, not adequately controllcd or even sufficiently control- lable for keeping the carcinogenic burden within set limits as far as the individual is concerned. The "safe dose," if adopted, therefore, would become because of these circumstances a most unrealistic standard without any practical significance.

Finally, it must bc emphasized that thcre does not exist any correlation between the carcinogenicity of a chemical and its toxic properties. The degree and type of the toxic reactions of a chemical therefore do not provide any qualitative or quantitative index of its prospective carcinogenic quali- ties and its relative potency in man or animals. This fact deserves special mention, because phar- macologists not sufficiently familiar with the specific aspects of chemical carcinogenesis have contended that bioassays of chemicals for carcinogenic reactions arc equivalents or modifications of chronic toxicity tests.

III . EPIDEMIOLOGIC ASPECTS In the past the initial recognition of environ-

mental and especially occupational cancers and carcinogens by cpidemiologic observations was subsequently confirmed by appropriate experi- mcntal observations. During the last few years this sequence of events has been reversed in coimection with the demonstration of carcinogenic hazards from cutaneous contact with hydrogenated coal oils produced by the Bcrgius process (Huepcr [3s Sexton [74]) and with respiratory contact to mustard gas (Heston [s Case and Lea [7]; Bcebe [s Yamada et al. [76]). The experimentally induced occurrence of cancers in animals by the use of a chemical with economic importance should thus

furnish an adequate justification for conducting comprehensive epidemiologic investigations on population groups exposed to this agent for explor- ing the existence of a similar environmental cancer hazard to man.

A similar connotation has the observation of any epidemic-like appearance of cancers in animals, especially if they are food animals and if they become exposed to environmental factors, particularly dietary constituents, which are identical with or similar to those introduced into the human food supply. While the occasional appearance of cancers in domesticated and wild animals, including fish exposed to human carcinogens, such as radioactive matter, petroleum oils, and arsenicals, had previously been reported, the scope of such environmental pollutions to animals and man was demonstrated by the recent observation of an epidemic of primary liver cancer among rainbow trout kept in hatcheries throughout the eountry (Hueper and Payne; 2 Rueker, Yasutake and Wolf [71]). I t was found that such cancers occurred in animals as young as 6-7 months old and affeeted about 80-90 per cent of brood trout kept for over 3 years. The evidence indicated that the carcinogenic agent active in the production of these liver cancers was evidently highly potent and apparently acted upon the very young animal. A prelimi- nary assessment of the circumstances under which this epidemie has made its appearance suggests that a dietary factor (arsenical amebieides, sul- fonamides, oxidized fatty acids, antioxidants, vita- minie and proteinie imbalanees, etc.) is to be in- criminated, which is eontained in or associated with the processed fish feed introduced into the industry some years ago.

This tentative explanation is made rather probable because of the fact that various chemicals introduced into animal and human foodstuffs and dietary deficiencies have been shown to elicit primary cancer in experimental animals (chlorinated hydrocarbon pesticides, such as DDT and Ara- mite; fungicides, such as thiourea and thiourea derivatives and thioacetamide; the sweetening agent, phenetyl urea [l)ulcin]; the flavoring agent, safrol; the food contaminant, selenium; the pesti- cide, arsenic; several food dyes, oil orange E; oil yellow HA, butter yellow; and dietary methionine and protein deficiencies). Since some of the various dietary ingredients mentioned are not only present in many of the foodstuffs used for human consumption but are also stored in the hmnan body and thus capable of exerting a prolonged and pro- gressively accentuated action, the occurrence of a

Unpublished observations, ])rs. W. C. Hueper & W. W. Payne, National Cancer Institute, Bethesda, Md.



H u E P E R - - E n v i r o n m e n t a l Carcinogenesis 85S

catastrophic cancer epidemic among rainbow trout should provide a most forceful stimulus for organ- izing comprehensive epidemiologic studies on the incidence and site distribution of cancers in human populations especially strongly exposed, for some reason, to the various dietary carcinogens to be suspect. This occurrence also furnished additional justification for an extension of the rather limited efforts of screening food additives and contami- nants as well as processed foodstuffs for carcinogenic properties, so as to avoid setting the stage for a future repetition of a similar cancer epidemic with some man-made carcinogenic environmental contaminant. Since Japanese investigators recently demonstrated the carcinogenic action of the antibiotic actinomycin which produced sarcomas at the site of subcutaneous injection in mice (Ka- wamata et al. [44, 45]) as well as of penicillium islandicum which contaminated rice and elicited carcinomas of the liver when fed to rats (Miyake et al. [51]), antibiotics used in the processing of foodstuffs or formed in them by fungal contami- nants should be included in experimental and epidemiologic investigations of this type.

An additional and probably the most promising field for greatly intensified epidemiologic efforts is presented by the occupational cancer hazards. In comparison to the epidemiologic evidence available on the occurrence and incidence of occupational cancers American epidemiologic-statistical data are highly defective, if not pitiful, and, therefore, if taken at their face value, misleading and not adequately reflecting the actual situation. This deplorable situation has recently been complicated by the publication of two occupational cancer studies in which objectionable methods were employed. In industry-inspired epidemiologic studies on the liability of railroad employees and of workers in asbestos operations (Kaplan [48]; Braun and Truan [6]) epidemiologic conclusions have been drawn, therefore, which do not conform with the results of other properly conducted investigations and which contain the incorrect allegation that an increased liability to cancer of the lung of the worker groups analyzed could not be demonstrated (Mancuso; 30'Donnell [60]; Doll [18]). This lack of an adequate number of trustworthy and competent epidemiologic studies on occupational cancers in the United States is also responsible for the fact that the eminent carcinogenic role of benzidine in the causation of cancers of the bladder,

3 Personal communication on asbestosis cancer of the lung presented before tim Meeting of the American Academy of Preventive Medicine, March 1961, Dr. Thomas F. Mancuso, Chief, Industrial Hygiene, Ohio Dept. Health, Columbus, Ohio.

which is generally recognized throughout the world, has been challenged by some American investigators (Eckardt [~0]) who followed in this respect the lead given some years ago by Gehr- mann (~4).

Through such practices the false impression is supported that occupational cancers are uncom- mon (Eckardt [~1]; Oettel [61]) and that, for this reason, this particular but highly important kind of environmental cancer is only of limited importance as a public health problem. The fact is that according to world-wide experience the study of occupational cancer hazards deserves greatly increased attention not only because they elicit important and often fatal industrial diseases, but also because the special circumstances related to exposures to carcinogenic agents present for occupational population groups greatly facilitate the demonstration of the action and sometimes of the character of an industrial carcinogen. Such a carcinogen, through its subsequent introduction into the general human environment, might also represent an environmental carcinogen which then has usually a rather diffuse distribution and possibly widely scattered carcinogenic effects not readily shown by available epidemiologic methods. Epi- demiologic studies on occupational cancer hazards, therefore, may provide the key to the recognition of environmental cancer hazards.

Retrospective epidemiologic studies are indicated for the multitude of occupational population groups whose members have been exposed for many years to the various known carcinogens. Such studies are especially indicated whenever the available evidence for a particular carcinogenic hazard has remained totally or mainly restricted to non-American sources, such as cancers in hema- tite miners and foundry workers, in nickel smelter workers, in employees of asbestos mines, mills and textile plants, in miners of arsenic-containing ores, arsenic smelter workers, and insecticide makers and users, in uranium ore miners and millers, chemical and pharmaceutical producers and users of various aromatic amino- and nitro- compounds and their users in industry (particularly rubber plants and processing and preparation of foodstuffs), makers of synthetic estrogens and the producers and users of estrogenized animal feeds, employees of oil refineries and carbon black and carbon electrode plants, and workers on gas retorts in gas plants and coke ovens, as well as the numerous industrial users of coal tar and pitch.

Prospective epidemiologic investigations are needed for the even more numerous occupational populations engaged in the manufacture and use of the many potential human carcinogens for



854 C a n c e r R e s e a r c h Vol. ~1, August 1961

which, at present, any type of human evidence is lacking. Retrospective and prospective epidemiologic studies can most profitably be pursued at present on worker groups of great stability (long- term employment) who are employed in factories located in rather small communities and represent- ing their main local industrial establishment.

Such surveys on human populations undertaken for the discovery and assessment of occupational cancer hazards should be supplemented by long- term surveys on wild and domesticated animals living on the ground and in bodies of water in the vicinity of industrial operations with carcinogenic hazards, since industrial effluents released into the air and into bodies of water near such plants may contain carcinogens which may elicit carcinogenic responses among such animals. In fact, observations of cancers among these animals may serve as early warning signs of the existence of an industry-related environmental cancer hazard to man breathing the air, drinking the water, and eating the plants and fruits grown in the soil polluted with carcinogenic chemical dirt from nearby industrial plants, and means of motorized transpor- tation. The reported more frequent occurrence of lung cancer in dogs living in urban areas in the U.S.S.R. than in those of rural regions, if confirmed, furnishes a good illustration of such correlations, especially since they would point to the importance of carcinogenic factors in urban air pollutants as significant agents in the causation of this particular type of cancer (Leake [47]; Hueper [31]; Hueper et al.). 4

In the etiology-specific interpretation of such statistical data, when applied to cancer of the lung in man, it appears wise to consider the proved polyetiology of many cancers, including lung cancers, as well as the probability that several car- einogenie and co-carcinogenic factors might com- bine in elieiting a cancerous response. The results of recent epidemiologie and pathologic observations have revived the former interest in the relation of tuberculosis and cancer of the lung. In view of the fact that the oral administration of the sedative, urethan (Jaff6 [4~]) and of the tuberculostatie isonicotinic acid hydrazid (Mori et al. [5s resulted in an excessive development of pulmonary adenomas in mice, it may be indicated to investigate with statistical methods the possible role which the various types of medications administered for different reasons to persons with tuberculosis of the

4 W. C Hueper, P. Kotin, E. C. Tabor, W. W. Payne, H. Falk, and E. Sawicki. Carcinogenic Bioassays on Air Pol- lutants. Presented at the 58d Annual Meeting of the Air Pol- lution Control Association, May, 1960. Arch. Path. (in press).

lung may have played in the development of cancers in tuberculous lung (Bal6 [1]).

In the interpretation of the statistical data obtained from epidemiologic studies of occupational groups it is unrealistic to adopt the concept that " I t is scientifically unsound to attribute a given type of cancer to an occupation unless it can be established with statistical significance that the incidence of this type of cancer among employees is higher than in the general population, or in an employee group comparable in all other characteristics except the occupation" (Eckardt [19]). Such decisions do not depend entirely on the outcome of statistical investigations but require also the application of sound medical judgment concerning the symptomatic and pathologic manifestations noted in members of worker groups suspect of being subjected to an occupational cancer hazard. Individuals exposed to benzol and ionizing radiations, for instance, exhibit a hematic syndrome reflecting strikingly the ambivalent aplastic and hyperplastic effects, including leukemia, produced by these agents upon the blood-forming tissues. Since the entire range of such reactions can be elicited with these agents in experimental animals there can exist no reasonable scientific doubt concerning the causal relation between contact with benzol and ionizing radiations and the subsequent development of leukemia, despite the fact that the presently available epidemiologic data may not be entirely sufficient to support such a conclusion. The objections raised lately by several commercial- ly interested parties to further consider arsenicals as human carcinogens because of the absence of adequate epidemiologic and experimental evidence in support of such a concept, held for many years by numerous competent and experienced derma- tologists and industrial physicians, also totally dis- regard the characteristic symptomatic pattern of chronic arsenic poisoning presented by the type, sequence, and location of cutaneous changes pre- ceding the development of cancers of the skin, lung, and liver often seen in individuals who were exposed to arsenicals for occupational, medicinal, and dietary reasons.

A serious shortcoming of the presently available epidemiologic-statistical methods is related to the fact that the information obtained by such procedures usually relies on the demonstration of ab- normalities in the quantitative pattern of the cancer panorama found in the population group ex- amined, i.e., as a rule an incidence rate of a certain type of cancer which is excessive in relation to that found in a standard or control population. How- ever, conclusions drawn from such comparisons are subject to the vagaries of the so-called "stand-



HvEPEa--Environmenlal Carcinogenesis 855

a rd" cancer rates, since these va ry wi th in ra ther wide l imits for different popula t ion groups according to region and occupation. If one should give full credence to the claims of the advocates of the cigaret te theory of lung cancers, they va ry for these cancers also according to the prevalence and in tens i ty of the smoking habi ts observed by the members of the par t icu la r popula t ion group analyzed which, in turn , are alleged to de termine thei r p r edominan t histologic variety. " S t a n d a r d " cancer rates thus are ra ther a rb i t ra ry values, which, when ind iscr imina te ly and improper ly applied, m igh t obscure the demons t ra t ion of an occupa- t ional cancer hazard.

An example of such an obli terated correlation is being offered by the gradual d isappearance of the formerly marked discrepancy in lung cancer incidence rates between Engl ish coke oven and gashouse workers exposed to an inhala t ion of coal tar fumes and the Engl ish popula t ion in general, beconfing lately exposed to an increasing degree to carcinogenic const i tuents present in indust r ia l air pol lu tants and in automobi le exhaust (Hueper [31]; Doll [18]). Similar objections app ly to the va l id i ty of da ta on cancer incidencc in so-called normal control groups, which can also be selccted in such a way tha t they m a y or m a y not differ fun- damen ta l l y f rom the test group as far as cancer ra les are concerned, a l though they are appa ren t ly chosen by the general ly accepted rules. The decision concerning the absence or presence of an occupat ional cancer haza rd among members of a test populat ion, therefore, should not ent i re ly depend on the demons t ra t ion of an excessive cancer incidence rate, because this rate, for the reasons given, m a y be within the l imi t of " s t a n d a r d " values, despite the existence of a dis t inct occupat ional cancer hazard.

Because of the a m b i v a l e n t effects elicited by m a n y carcinogens, such as ionizing radiat ions, benzol, estrogens, u re thans , and others, i t appears to be impor t an t t ha t considerable epidcmiologic research be conducted on the carcinogenic rela- t ions of various chemotoxic disease mani fes ta t ions character ized by degenerat ive manifes ta t ions , such as aplast ic anemia , agranulocytosis , l iver degenerat ions and necroses, funct ional and ana tomic thyro id hypoplasias , and cutaneous atrophies, which represent ei ther direct chemotoxic reactions or chemoallergic responses.

The rapidly growing industr ial , d ie tary, cos- tactic, and medic ina l use of macromolecular chemicals, such as especial ly plastics and water- soluble carbon and silicon polymers, which have d isp layed carcinogenic propert ies in exper imenta l animals , also seems to jus t i fy the organizat ion of

a registry of pa t ien t s who received paren te ra l im- p lan ts of such plast ic mater ia l s for a subsequent epidemiologic assessment of possible late carcinogenic sequelae a round such implan t s appear ing some 10-30 years af ter such deposi ts were made. As a ma t t e r of wise precaut ion, such studies m a y also be extended to persons wear ing plast ic den- tures, since it is impor t an t to ascer ta in whether or not a prolonged mucosal contact wi th such plast ics m i g h t be related to a delayed fo rmat ion of cancers in the oral cavi ty .

This brief outl ine on the facts, the concepts, and the m a n y unsolved and controversial problems of env i ronmen ta l carcinogenesis and cancers has served its purpose if it has conveyed adequa te ly the message tha t the field of env i ronmen ta l cancer research represents not only one of the most im- po r t an t facets of cancer research but , at the same time, also one of the most reward ing and imme- dia te ly useful ones. Wi th the ava i l ab i l i ty of increased funds and increased interest in these aspects of cancer control especial ly by prophylac t ic and prevent ive measures, it m a y be hoped t ha t the foreseeable fu ture m a y see perhaps a repet i t ion of the successes which this approach brought to the control of communicab le diseases.

REFERENCES 1. BAI/), J. Lungenkarzinom und Lungenadenom, p. 363.

Budapest: Verlag dcr Ungarischcn Akademie dcr Wissen- schaften, 1957.

~. Bv2~mn, G. W. Lung Cancer in World War I Veterans: Pos- sible Relation to Mustard-Gas Injury and 1918 Influenza Epidemic. J. Nat. Canccr Inst., 25:19281-5~, 1960.

3. BIELSCIIOWSKY, F., and IIAI,L, W. It. Carcinogenesis in Paral)iotic Rats. Tumours of Liver and Seminal Vesicle Induced by Acetylfluorene in Normal Males. Brit. J. Can- ccr, 5:106-1~, 1951.

'~. BILLIARD-])UCHESNE, J. L. Les Amino-tumeurs de la vessie. J. d'urol., 53: 4()1-~2~, 1946-47.

5. BONSmL G. M.; CLAYSON, 1). B.; JITLL, J. W.; and PYuA~t, L. N. Carcinogenic Properties of ~-Amino-1- Naphthol Hydrochloridc and Its Parent Aminc--~-Naph- thylamine. Brit. J. Cancer, 6:41~-~4, 195~.

6. BRAUN, ]). C., and TiiUAN, T. D. An Epidemiological Study of Lung Cancer in Asbestos Mincrs. Arch. Indust. ttealth, 17: 634-53, 1958.

7. BRU~'NER, H. Expcrimentelle Ausloesung yon Tumoren durch Implantation yon Polymethyl-methacrylat bei Rat- ten. Arzncimittelforsch. 9:396-99, 1959.

8. CASE, R. A. M., and L~'A, A. J. Mustard Gas Poisoning, Chronic Bronchitis, and Lung Cancer; Investigation into Possibility That Poisoning by Mustard Gas in 1914-18 War Might Bc Factor in Production of Neoplasia. Brit. J. Prcv. & Soc. Med., 9:6~2-7~2, 1955.

9. CONWAY, W. D., and L~.THCO, E. J. Metabolism of Citrus Red No. ~2--an Oil Soluble Food I)yc. Abstracts of Papers, 139th Meeting American Chemical Society, St. Louis, March ~21-30, 1961.

10. Cou~-ClL ON DnUGS. Withdrawal of Iron-Dcxtran Comt)lex (hnferon). J.A.M.A., 175:836, 1961.

11. ])ELANEY CANCER CLAUSE: t[EARING8 BEFORE TIIE IIo(:sn



856 Cancer Research Vol. s August 1961

SELECT COMMITTEE TO INVESTIGATE THE USE OF CHEMI- CALS IN FOOD PRODUCTS. House of Representatives, 82d Congress. Chemicals in Food Products, pp. 582, 878. Washington, D.C.: U.S. Govt. Printing Office, 1951.

12. DELANEY CANCER CLAUSE: HEARINGS BEFORE THE HOUSE SELECT COMMITTEE TO INVESTIGATE THE USE OF CHEMI- CAI.~ IN FOODS AND COSMETICS. House of Representatives, 82d Congress. Chemicals in Foods and Cosmetics, p. 1803. Washington, D.C.: U.S. Govt. Printing Office, 195~.

1~. DELANEY CANCER CLAUSE: HEARINGS BEFORE SUBCOM- MITTEE OF THE COMMITTEE ON INTERSTATE AND FOREIGN COMMERCE. Itouse of Representatives, 85th Congress, ~(i Session. Color Additives, p. 533. Washington, D,C.: U.S. Govt. Printing Office, 1958.

14. DELANEY CANCER CLAUSE: HEARINGS BEFORE COMMITTEE ON INTERSTATE AND FOREIGN COMMERCE. House of Rep- resentatives, 86th Congress, ~d Session. Color Additives, p. 608. Washington, D.C. : U.S. Govt. Printing Office, 1960.

15. DELANEY CANCER CLAUSE: SCIIRI~rENREIHE I)ES BUNDES FUR LEBENSMITTELI~ECnT UND LEBENSMITTELKUNI)E. Heft ~5. Das Lcbcnsmittelgesetz, p. `20. Wiesbaden: B. Behr's Verlag GMBII, 1959.

16. DELANEY CANCEIr CLAUSE: BRITISH MINISTRY OF HEALTh. Carcinogenic Risks in Food Additives and Pesti- cides. Monthly Bull., pp. 108-12, 1960.

17. I)i MAIO, G. Tumore della vcsica da amine, detti da anili- na. Urologia, 3:~39-4`2, 1986.

18. DOLL, R. Etiology of Lung Cancer. Adv. Cancer Research, 3: 1-50, 1955.

19. EC~ARDT, R. E. Industrial Carcinogens, p. 164. New York: Grunc & Stratton, 1959.

`20. - - - - - - . Experimental Carcinogenesis and the Problem of Food Additives. Am. J. Publ. Health, 50:1488-9~, 1960.

~1. - - - - - - . Occupational Cancer: Methods of Epidemiological Study. Arch. Indust. IIcalth, 19:335-~8, 1959.

~ . FERGITSON, R.. S.; GEHHMANN, G. H.; GAY, ]). M.; ANDER- SON, L.; and WAsnmmN, V. Symposium on Aniline Tu- mors of the Bladder. J. Urol., 31: lC21-26, 1934.

03. FOOD PROTECTION COMMITTEE. Problems in the Evalua- tion of Carcinogenic Hazard from Use of Food Additives, Publ. 749, p. 44. Washington, D.C.: National Academy of Sciences, 1959.

~4. GEIIRMANN, G. H. Papilloma and Carcinoma of the Blad- der in Dye Workers. J.A.M.A., 107:1436-39, 1936.

~5. GOLDBLATT, M. W., and GOLDBLATT, J. Industrial Car- cinogenesis and Toxicology, 3 : 185-365. In: E. R. A. MERE- WETHER, Industrial Medic.i,m and Hygiene. London: Butterworth and Co., Ltd., 1956.

`26. HADDOW, A., and HOI~NING, E. S. On the Carcinogenicily of an Iron-Dextran Complex. J. Nat. Cancer Inst., 9.4:109- 47, 1960.

~7. HEIDELBEnGm~, C. Applications of Radioisotopes to Studies of Carcinogenesis and Tumor Metabolism. Adv. Cancer Research, 1:~274-338, 1953.

'28. HESTON, W. E. Carcinogenic Action of Mustards. J. Nat. Cancer Inst., 11 : 415-~3, 1950.

~9. HVEPEIt, W. C. ()cc upational Tumors and Allied Diseases, p. 897. Springfield, I11.: Charles C Thomas, 194- ~

30. - - - - - - . Recent I)evelopmcnts in F, nvironmental Cancer. Arch. Path., 58:~60-99; 474-5~3; 645-8~, 1954.

31. . A Quest into the Environmental Causes of Cancer of the Lung. Public lIealth Monogr. No. 36, Public Heallh Service Publ. No. 45'2, 54. Washington, D.C.: U.S. Govt. Printing Office, 1955.

3~. . Experimental Carcinogenic Studies on Hydroge- nated Coal Oils. I. Bergius Oils. Indust. Mcd. & Surg., 25: 51-55, 1956.

33. - - - - - - . Environmental Factors in the Production of Hu-

man Cancer. In: R. W. RAV~N, Cancer, 1:404-96. Lon- don: Butterworth & Co., Ltd., 1957.

84. . Epidemiologic, Experimental and Histological Studies on Metal Cancers of the Lung. Acta Unio Inter- nat. contra cancrum, 15:424-36, 1959.

35. - - - - . Carcinogenic Studies on Water-soluble and In- soluble Macromolecules. Arch. Path., 67: 589-617, 1959.

36. - - - - . Medicolegal Considerations of Occupational Can- cer; Medicolegal Considerations of Non-Occupational En- vironmental Cancers, p. 558-663. In: Lawyers' Medical Cyclopedia, Vol. 5. Indianapolis: Allen Smith Co., 1960.

37. - - - - - - . Carcinogenic Studies on Water-Soluble Polymers. Path. Microbiol. (Basel), 24:77-107, 1961.

38. - - - - - - . Carcinogens in the Human Environment. Arch. Path., 71:~237-67; 355-80, 1961.

89. IIuEvER, W. C., and PAYNE, W. W. Carcinogenic Studies on Soot of Coffee-Roasting Plants. Arch. Path., 69:716- `27, 1960.

40. - - - - - - . Carcinogenic Studies oil Petroleum Asphalt, Cool- ing Oil, and Coal Tar. Ibid., pp. 37~2-84.

41. HUEPER, W. C.; WILF.Y, F. H.; and WOLFE, H. D. Experi- mental Production of Bladder Tumors in Dogs by Ad- ministration of Beta-Naphthylarnine. J. Indust. Hyg. & Toxicol., 20: 40-84, 1988.

4~. JAFFa, W. G., and ,IAFF~, R. Carcinogenic Action of Ethyl Urethanc on Rats, Including Itistological Findings in Lungs and Livers. Cancer Research, 7: 107-11, 1947.

48. KAPLAN, I. Relationship of Noxious Gases to Carcinoma of the Lung in Railroad Workers. J.A.M.A., 171:~089-43, 1959.

44. KAWAMATA, Z.; NAKABAYANHI, N.; KAWAI, A.; FUJITA, H.; IMANISm, M.; and IKEGAMI, R. Studies on the Carcino- genic Effect of Actinomycin. Bilkcn's J., 2:105-1~, 1959.

45. KAWAMATA, J.; NAKABAYASHI, N.; KAWAI, A.; and I;SHIDA, T. Experimental Production of Sarcoma in Mice with Actinomycin. Med. J. Osaka University, 8:753-6~, 1958.

46. KREYBmCG, L. The Significance of IIistological Typing in the Study of the Epidcmiology of Primary Epithelial Lung Tumours: A Study of 466 Cases. Brit. J. Cancer, 8:199- "208, 1954.

47. LE.(KE, C. D. Lung Cancer in Dogs. J.A.M.A., 173:85--86, 1961.

48. MAYER, R. L. Aromatic Amines and Azo-Dyes in Allergy and Cancer. J. Invest. Dermatol., 10:389-96, 1948.

49. MILLER, E. C.; MILLn~, J. A.; BI~OWN, R. R.; and MAC- DONALD, J. C. On the Protective Action of Certain Poly- cyclic Aromatic Hydrocarbons against Carcinogenesis by Aminoazo-Dyes and 2-Acctylaminofluorene. Cancer Re- search, 18:469-77, 1958.

50. MILLER, J. A., and MILLER, E. C. The Carcinogenic Amino-Azo-I)yes. Adv. Cancer Research, 1:339-96, 1953.

51. MIYAKE, M.; SAITO, M.; ENOMOTO, ~{.; et al. Development of Primary Hepatic Carcinoma in Rats by Long Term Feeding of the Yellow Rice by Pcnicillium Islandicum Sopp. (;,ann, 50:117-18, 1959.

5~. MORI, K. ; YASI:.~-o, A. ; and MATSUMOTO, K. Induction of Puhnonary Tumors in Mice with Isonicotinic Acid Hy- drazid. Gann, 51:83-89, 1960.

58. MORRIS, II. P. The Experimental Development and Mctal)olism of Thyroid Gland Tumors. Adv. Cancer Re- search, 3: 51-115, 1955.

54. MORalS, H. P.; DUNN, T. B.; and DUBNIK, C. S. Two Transplantable l~tt Carcinonms Induced by N-acetyl-~ - aminofluorene. J. Nat. Cancer Inst., 9:~2~5-~~ 1948.

55. MORRIS, II. P., and EYESTONE, W. H. Tumors of Liver and Urinary Bladder of Dog after Ingestion of 2-Acetyl- anfinofluorene. J. Nat. Cancer Inst., 13:1139-65, 1958.



H uEP ER--Env iron men t al Carcinogenesis 857

56. NAoAyo, M., and KINOSITA, R. Some Significant Features of Cancer Incidence in Japan. Yale J. Biol. & Med., 12: 801-8, 1940.

57. NAU, C. A.; NEAL, J. ; and STEM:BRIDGE, V. A. A Study of the Physiologic Effects of Carbon Black. Arch. Environ. Health, 1:512-33, 1960.

58. NELSON, A. A. Carcinogenic Substances in Food. Int. Rec. Med. & G. P. Clin., 169:47-49, 1956.

59. NOTHDURFT, H. Experimentelle Sarkomausloesung durch eingeheilte Fremdkoerper. Strahlentherapie, 100:19~-210, 1956.

60. O'DoNNELL, W. H., and MANN, R. H. Asbestos: An Ex- trinsic Factor in the Pathogenesis of Bronchogenic Carci- noma. Am. J.. Path., 33:610, 1957

61. OETTEL, H. Zur Frage der Verunreinigung yon Lebens- mitteln durch kanzerogene Substanzen. Veraender. d. Nahr. d. Indust. und haushaltsm. Verarb., 5:93-184. Wiss. Veroefftl. Deut. Gcs. f. Ernaehr, 1960.

62. OPPENHEIMER, B. S.; OPPENHEIMER, E. W.; STOUT, A. P.; WILLHITE, M.; arid DANISREFSKY, I. The Latent Period in Carcinogenesis by Plastics in Rats and Its Relation to the Presarcomatous Stage. Cancer, 11:204-18, 1958.

63. PAYNE, W. W. Production of Cancer in Mice and Rats by Chromium Compounds. Arch. Indust. Health, 21: 530-35, 1960.

64. - - - - - - . The Role of Roasted Chromite Ore in the Produc- tion of Cancer. Arch. Environ. Health, 1:20-26, 1960.

65. PAYNE, W. W., and HUEPER, W. C. The Carcinogenic Ef-

fects of Single and Repeated Doses of 8,4-Benzpyrene. Am. Indust. Hyg. Assoc. J., 21:350-55, 1960.

66. PIETRA, G.; RAPPAPORT, H.; and SHUBIK, P. The Effects of Carcinogenic Chemicals in Newborn Mice. Cancer, 14: 308-17, 1961.

67. PIETRA, G.; SPENCER, K.; and SHUBIK, P. Response of Newly Born Mice to a Chemical Carcinogen. Nature, 183:1689, 1959.

68. REHN, L. Blasengeschwuelste bei Fuchsin-Arbeitern. Arch. f. klin. Chir., 50:588-600, 1895.

69. ROSENBAUM, N., and GOTTLIEB, J. Neubildungen der Harnblase bei Textilarbeitern. Gigiona truda, p. 30, 1926.

70. Ross, H. C. Occupational Cancer, J. Cancer Research, 3:321-56, 1918.

71. RUCKER, R. R.; YASUTAKE, W. W.; and WOLF, H. Trout Hepatoma. Progr. Fish Culturist, 23:8-7, 1961.

72. SCHEDLER, R. Zur Kasuistik der Blasentumoren bei Facharbeitern. Inaug. Diss., Basel, 1905.

73. SCHUELLER, H. Blasentumoreu bei Anilinarbcitern. Ztschr. Urol., 26: ~84-86, 193~.

74. SEXTON, R. J., The Hazards to tIealth in the Hydrogena- ation of Coal. Arch. Environ. Health, 1:208-31, 1960.

75. STEINEn, P. Conditional Biological Activity of Carcino- gens in Carbon Black, and Its Elimination. Cancer Re- search, 14:103-10, 1954.

76. YAMADA, A.; HIROSE, F.; NAGAI, M.; and NAKAMURA, T. Five Cases of Cancer of the Larynx Found in Persons Who Suffered from Occupational Mustard Gas Poisoning. Gann, 48: 366-68, 1957.



1961;21:842-857. Cancer Res W. C. Hueper Environmental Carcinogenesis and Cancers

Updated version

http://cancerres.aacrjournals.org/content/21/7/842.citation

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/21/7/842.citationTo request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Environmental Carcinogenesis and Cancers* - Cancer Research