Bull. Org. mond. Sante' 1962, 26, 649-659Bull. Wid Hlth Org.

The Role of Experimental Research in the Studyof the Prevention of Malignant Tumours

L. M. SHABAD1

The author discusses the role of experimental oncological research in the prevention ofmalignant neoplasms, with special reference to the conclusions drawn from such researchin the USSR. He points out that experimental research can contribute to cancer preventionin two ways: (a) by supplying information on the occurrence of carcinogenic substances inthe human environment-in the atmosphere, in industry and infoodstuffs-and thus providinga rational basis for the introduction of measures to prevent cancer from arising; and (b) bythrowing light on the series of tissue changes that may precede the development of themalignant tumour and hence making it possible, through the timely treatment and cure ofknown precancerous conditions, to prevent cancer from developing.

In order to prevent cancer it is essential to knowits causes. There is no doubt that experimentalresearch plays an outstanding role in the study of theetiology and pathogenesis of tumours, and experi-mental oncology may therefore serve the cause ofcancer prophylaxis. However, it is not enoughmerely to assert the fact. It seems to us that it hasnow become both possible and necessary to outlinecertain definite methods of preventing malignanttumours and that in this respect experimental on-cology has an absolutely determinate and importanttask to perform. In this paper, an attempt will bemade to demonstrate this on the basis of someexamples taken mainly from the work of the authorand his colleagues.

In our opinion, there are two lines of approach tothe prevention of malignant neoplasms:

(1) preventing cancer from arising (" hygienic "prophylaxis);

(2) preventing cancer from developing (" clinical "prophylaxis).

"CLINICAL" PROPHYLAXIS

To begin with the second of these lines of app-roach, clinical prophylaxis, it was noted long ago,at least in regard to cancer of the skin and lower lipand some other forms, that cancer is not of suddenoccurrence but is merely the last link in a long chain

1 Member of the USSR Academy of Medical Sciences;Head, Department for the Study of Carcinogens, Institute ofExperimental and Clinical Oncology, USSR Academy ofMedical Sciences, Moscow.

of changes developing over a comparatively lengthyperiod. On the basis of these observed facts, the ideaof pretumoral changes and of precancer took rootin clinical medicine.

There is still much that is obscure in the conceptof precancer. At the same time, much has becomeclear through experimental cancer research, whichhas made it possible to study, step by step, systema-tically and on the basis of extensive material, thecourse of development of malignant tumours indifferent organs in various animals, and to giveconcrete meaning to the term " precancer ".

Material has been accumulated over many yearsin our laboratory on the experimental study of pre-cancer in various animals. The development ofcancer of the skin, lungs, stomach, pancreas, liver,kidneys, thyroid gland, mammary glands, uterus,ovaries, testes, prostate and seminal vesicles, etc.has been investigated from this point of view andchanges have also been described for which, in ouropinion, " presarcoma " would be a suitable name.Firm experimental establishment of the fact that

precancer possesses certain regular features is anencouragement to diligent search for precancerousconditions in human pathology and to their thoroughstudy. Such instances are meant as the relationshipof adenocarcinoma of the stomach or rectum toadenomatous polyps, or of mammary-gland cancerto proliferative fibroadenomatosis or fibrocysticdisease of the breast with proliferation of the epithe-lium, or ofcancer ofthe uterine cervix to non-healingpseudoerosions (changes that resemble erosion only

1128 -649

L. M. SHABAD

macroscopically-e.g., papillary and follicular ero-sion) of the cervix accompanied by epithelial prolife-ration, or of cancer of the larynx to pachydermalaryngis, etc. However, the changes which regularlyprecede a number of types of human tumour are notyet known or have been little studied. The greatestdifficulty lies in settling the problem of whether amalignant tumour occurs only against a backgroundof particular changes or whether it develops fromthose changes. Only in the latter case would it becorrect to say that the changes were significant asprecancer in the strict sense of the word.

In a number of instances, it has been possible onthe basis of our own material to compare the resultsof experimental research with observations of pre-cancer in man. This applies to mammary-glandcancer, cancer of the prostate, cancer of the urinarybladder and a number of precancerous changes inother organs and tissues. As a result, it has nowbecome possible to draw a number of conclusions.Each cancer has its own precancer. In other

words, not a single neoplasm whose beginning hasbeen experimentally studied on a sufficiently widescale has begun suddenly; every cancer has beenpreceded by definite morphological changes de-veloping over a comparatively long period.

Cancer often develops from several centres and,accordingly, precancerous changes are usuallymultiple. As a rule, there are more precancerousfoci than there are foci of the resultant cancer. Themulticentric nature of tumour occurrence can easilybe established by experimental study of tumours inany site and of any kind, whether they are tumoursof the skin or subcutaneous cellular tissue, the lungsor the mammary glands, the liver or the ovaries, thestomach or the uterus, etc. This same multicentricityis encountered in man. The best illustration of thisfact is the picture of precancerous changes in themammary gland.The morphological picture of precancerous

changes in various organs differs. Sometimes in oneand the same organ different forms of precancerouschange can be found. Pretumoral changes in varioustissues differ from each other as much as, or rathereven more than, the tumours that develop fromthem-they constitute different nosological entities.At the same time, they reflect a single pathologicalprocess-the process of cancer development-andtherefore some general features can be detected inpretumoral changes in the most diverse organs. Thisis particularly clear if certain stages in the develop-ment of cancer are considered.

Several stages can be distinguished in the develop-ment of cancer.The first of them is a characteristic diffuse, uneven

and irregular hyperplasia which differs from inflam-matory or regenerative hyperplasia and is function-ally and morphologically distorted. This can be seenfrom the lack of correspondence between the mor-phological manifestations and the functional condi-tion. A clear example is the diffuse hyperplasia of themammary-gland epithelium in male rats, or invirgin female rats without lactation, induced by theadministration of large quantities of oestrogens.This type of epithelial change can be regarded as apathological neoplastic variant of the process ofwhich the physiological prototype is the hyperplasiaof the mammary-gland epithelium during lactation.The second stage in the development of malignant

tumours, in our opinion, is that of focal and nodularproliferations, differing from the surrounding tissueonly in that their general structure and componentcells are to some degree atypical. As our researchhas shown, such proliferations may possess theirown stroma.The third stage in the development of cancer is

that of the so-called "benign" tumours-papillomas,adenomas, etc. Experimental research has de-monstrated with particular clarity that the benig-nancy of such tumours is only relative 'and that it isvery often possible that they will develop into cancer.At the same time, it should not be forgotten thatthere is a possibility that a malignant tumour willarise from a nodular precancerous proliferationdirect, that is, without passing through the thirdstage of development of the process.The fourth-and final-stage in the process is

cancer itself-the malignant tumour.The stages in the development of cancer men-

tioned above can be discovered, as a rule, in all formsof induced cancer subjected to sufficient study. How-ever, it is not always possible to establish exactly theboundaries between the stages since they merge intoone another, being parts of a continuous process.Numerous experiments carried out by Mona-

styrskaya (1952), Prokofyeva (1952) and othercolleagues of ours have shown that inflammationdoes not promote the development of cancer and isnot an obligatory stage in the phenomena leadingto it; in other words, it is not of precanceroussignificance. Carcinogenic hydrocarbons inhibit in-flammation, distort its character and destroy theinterrelationships between the epithelium and theconnective tissue (Vasilyev, 1959).

650

ROLE OF EXPERIMENTAL RESEARCH IN THE PREVENTION OF MALIGNANT TUMOURS

Cancer and inflammation are different patholo-gical processes and this in itself is enough to indicatethat inflammation is not significant as a precanceroussign. This does not mean, of course, that a com-bination of the two processes cannot occur in anumber of cases. Recently, in our laboratory, cancerof the lungs was induced in rats by intratrachealadministration of large doses of 9,10-dimethyl-1,2-benzanthracene (Pylev, 1961). In this instance,the precancerous changes developed in some casesfrom bronchiectases with chronic inflammation,while in other cases they had no connexion withinflammatory foci. Thus, although cancer may some-times occur against a background of inflammationor its after-effects, this does not mean that inflamma-tion plays a role as a precursor of cancer.

In assessing the role of inflammation in cancer itis useless to speak in general terms. It is essential todetermine the exact significance that may be attachedto a particular form of inflammation at particularstages in the induction, development and progressionof the tumour. At the same time, account must betaken of differences in the form and site of differentneoplasms. The multiplicity of neoplasms pre-supposes a multiplicity of possible interrelationshipsbetween cancer and inflammation.

Thus, for example, while inflammation is not anessential link in the chain of phenomena leading tocancer, its presence at the period when proliferationof the cancer cells has already begun may play animportant role. In some cases, perifocal inflamma-tion may further the growth of the tumour and theformation of its stroma, giving tumour-growthmechanical and nutritional support, as was shown,for example, in the experiments of Vasilyev (1958).In other cases, inflammation may further the break-down and destruction of at least part of a tumour,as is often seen in clinical practice when tumourtissue suppurates and undergoes liquefactive de-generation.From the morphological point of view the name

" precancer " could be given to microscopic multi-centric and often multiple foci of non-inflammatory,atypical proliferation of immature epithelium (orfibroblastic or osteoid tissue in the case of pre-sarcoma) with a tendency to infiltrative growth butwithout, as yet, destruction of the tissue. The name" precancer " in the narrow sense of the word shouldbe given to these focal proliferations and so-called"benign " tumours, which particularly often anddirectly turn into cancer. The diffuse, uneven,irregular hyperplasia of tissue which precedes focal

proliferations but is not precancer in the strict senseof the word may, nevertheless, be considered as apretumoral change.

Thus, in our interpretation, the concept of pre-cancer is made narrower and more concrete. Inour opinion it is better to be reproached for con-fusing the notion of precancer with the beginning ofcancer than to allow the concept to be loosely usedfor a mass of non-specific changes, so-called " back-ground processes ", various precursive diseases, etc.In our view non-specific changes of this type,preceding or accompanying cancer, should not beconsidered as precancer, a term that should be re-served for defining the beginning of the onset ofcancer.The various pretumoral changes, like benign and

malignant tumours themselves of course, may pro-gress at different rates. Thus skin papillomas in-duced in mice by painting the skin with carcinogenichydrocarbons usually grow considerably moreslowly if the carcinogen is given in a small dose or ata low strength. The reverse is true when considerabledoses of strong carcinogens are applied. Thus it ispossible, to a certain extent, to regulate the rate ofdevelopment of precancerous changes experimen-tally.

Precancerous changes may not only be halted intheir development, as has just been mentioned, butmay disappear altogether. The possibility of thedisappearance of manifest precancerous changes caneasily be demonstrated by careful observation ofskin papillomas induced in mice by means of carci-nogens. It is a matter of general knowledge that thenumber of papillomas occurring at the site ofpainting is considerably greater than the number ofcancerous tumours that develop there. This state ofaffairs is due to the disappearance of individualpapillomas as well as to their confluence.Thus diffuse hyperplastic changes and even pre-

cancerous focal proliferations may disappear al-together. Pretumoral and precancerous changes maynot develop into cancer. This indicates the con-ditional nature of the very concept of precancer.The possibility that what is known as a " sponta-neous cure " of precancer may occur has veryimportant implications for its successful rationaltreatment.At the same time, it is quite clear that the prompt

cure of precancer represents a very considerablestep towards cancer prophylaxis. The idea of theprocess of tumour development as a single patholo-gical process, passing through a number of stages

651

L. M. SHABAD

and phases, makes it possible, by eliminating onestage or phase, to interrupt the whole process.Thus we see that experimental cancer research,

by giving a practical foundation to the precancerconcept, has paved the way to the prevention of thedevelopment of cancer from precancerous and pre-tumoral changes. The concept of precancer, whicharose in clinical practice and has now been checkedexperimentally, returns to clinical practice enrichedby many facts. At the present time it is far fromhaving merely a theoretical interest; it is of greatpractical importance in both the individual and themass prophylaxis of cancer.

In ordinary clinical practice a number of suchpathological phenomena as papillomas and ad-enomas of various kinds, various focal proliferationsof the epithelium, hyperkeratoses, etc. are nowregarded not merely as " benign " tumours, but alsoas a possible stage on the road to cancer, and areconsequently subjected to radical treatment. Thus,individual clinical cancer prevention is beingpractised every day.Mass clinical prophylaxis against the development

of tumours has become possible as a result of theprophylactic medical examination of wide sectionsof the population, and particularly of special groupsexposed to particular carcinogenic factors. Thusclinical cancer prophylaxis comes into contact with" hygienic " prophylaxis, the problems of which willnow be discussed.

" HYGIENIC t PROPHYLAXIS

Possible methods of preventing the occurrence ofany disease are evolved as a result of a thoroughstudy of its etiology. The more that is known ofthe causes of a particular disease, the greater thepossibility of eliminating it. At the present time theetiology and pathogenesis of tumours have not beenfully elucidated, but the facts accumulated never-theless make it possible to indicate ways of pre-venting the occurrence of tumours. These ways arebased on the concept that blastomogenous agentsexist which may act upon man, a concept that aroseas a result of observations of what are known asoccupational tumours.

In the workers in some occupations particularforms of tumour are encountered more frequentlythan in the ordinary population of their district.Thus, as long ago as 1775 the well-known Englishdoctor, Percival Pott, described cancer of the skinin chimney-sweeps, localized in most cases in theregion of the scrotum, the penis and the thighs,

that is, in places subjected to the greatest contamina-tion with the products of the distillation of coal.It will be remembered that at that time boys climbedup inside the chimneys in order to clean them.

" Consumption " of miners in the Schneeberg inSaxony and at Jachimov in Bohemia had also beenknown for a very long time. It has proved to bean occupational cancer of the lungs caused byradioactive substances.At the end of the last century observations were

published on the very high frequency of bladdercancer in workers in aniline-dye factories in Switzer-land and Germany. Later, cases of this type ofbladder cancer were described in all countries of theworld where the industry existed, and it was estab-lished that its occurrence was connected with cer-tain chemical substances-f,-naphthylamine, ben-zidine, dianisidine, etc.At the beginning of the twentieth century, some

years after X-rays and other forms of radiation hadbegun to be extensively used in medicine andindustry, reports appeared of cases of occupationalcancer occurring amongst radiologists as the resultof excessive irradiation without proper protection.Malignant tumours of the jaws, observed in theUnited States of America among women paintingluminous watch dials, were caused by the action ofradioactive salts.

In the twenties in Great Britain, occupationalcancer of the skin was reported as occurring amongmule-spinners, who leaned with part of the bodyagainst machine parts abundantly lubricated withshale oils.At the present time a considerable number of

forms of occupational tumours have been described.The examples quoted are enough to show that thereare forms of tumour in man whose causes areknown. The elimination of those causes or even areduction in the amount of the harmful agent willhelp to prevent the tumours. It should be em-phasized that all known observations of occupationalcancer and all the vast amount of experimentalmaterial gained from the study of carcinogens showsthat the growth of tumours depends to a largeextent on the dose of the blastomogenous agent.The smaller the dose, the smaller the number ofpeople or animals in whom cancer develops and thelater its onset. Therefore, not only the completeelimination but even some reduction in the quantityof blastomogenous substances to which man isexposed is a way of preventing tumours. Nowadayschimney-sweeps' cancer, radiologists' cancer and

652

ROLE OF EXPERIMENTAL RESEARCH IN THE PREVENTION OF MALIGNANT TUMOURS

some other forms of occupational cancer havepractically disappeared as a result of changes inworking conditions, the introduction of new in-dustrial techniques, protective measures and meas-ures of personal hygiene, which have helped toeliminate the effect on the organism of blastomogen-ous agents. These facts show that cancer prophylaxisis in principle a possibility.

The'study of occupational tumours naturally ledto extensive experimental research on the carcino-genic effect of various chemical substances that weresuspected of causing various types of occupationalcancer. A great deal of modem experimentaloncology has developed from such research and verymany facts have been discovered that can be used forcancer prophylaxis.

Thus, during the period when cancer was beinginduced experimentally by painting the skin ofanimals with coal-tar, a large number of differenttars, pitches, bitumens and mineral oils obtainedfrom various sources and produced by variousmethods were investigated. It was discovered that aparticular product may differ in its carcinogeniceffect. For example, when various Soviet shale tarswere investigated by my colleagues and myself(Larionov, Soboleva & Shabad, 1934), it was foundthat some of them induced many more tumoursthan others. The same applies to some mineral oilsand to the " coolants " used in the engineeringindustry (Prokofyeva, 1938; Verkhovskaya, 1941).The work of Bogovski (1960) showed that differentshale products from Estonia had differing degrees ofcarcinogenic effect. On the basis of these data it ispossible in a number of cases to substitute a slightlycarcinogenic product for a highly carcinogenic one.

Differences in the carcinogenicity of coal-tar andother tars, and the possibility of substituting a lessdangerous product for a dangerous one, are ofgreat topical interest in many fields-for example,in evaluating various building materials from ahygienic point of view, in building roads, etc. In theUSSR regulations are in force under which thebitumens used for road surfacing in inhabitedplaces must contain a considerably smaller quantityof blastomogenous hydrocarbons than coal-tars(Malyugina, 1955).At the present time the question is being raised of

"decarcinogenizing" industrially important tars-that is, eliminating their carcinogenic properties.It can be considered as proved that this process istheoretically possible (Danetskaya, 1952; Bogovski,1960).

Since the discovery of chemically pure carcinogenicsubstances, a number of facts that can be used forcancer prophylaxis have come to light as a result ofexperimental study. Several research schemes of thiskind have been carried out in our laboratory. ThusKleinenberg (1939) showed that while 3,4,8,9-dibenz-pyrene is a strong carcinogen, its quinone, which isthe basis for one of the dyes used in the textileindustry, does not induce tumours. Recently,Pliss (1959), by systematic research on a number ofbenzidine derivatives, has demonstrated the highlycarcinogenic properties of 3,3'-dichlorobenzidine.Oral administration or subcutaneous injection of thissubstance in rats and mice caused various malignanttumours, often mutltiple, in about 80% of the ratsand 40% of the mice. both at the site of the hypo-dermic injection (sarcoma) and at some distance fromit (cancerous tumours of the mammary glands, thesebaceous g!ands and the liver, papillomas of thebladder, etc.). Other work by Pliss (1958) hasshown the comparatively slight, but undoubted,carcinogenic activity of dicyclohexylamine and itsnitrite.

It is impossible to dwell at greater length here onindividual pieces of experimental research that haveprovided facts which can be used for tumour pro-phylaxis. One general proposition, however, may bestated.

In view of the fact that the chemical industry iscontinually producing a large number of completelynew substances with which people will sometimescome to be in close and lengthy contact, both atwork and in the home, and that ever-increasingquantities of new chemicals are being used in agri-culture, the food industry, cosmetics production,etc., the question arises of whether the new substancesmay not include products with blastomogenousactivity. Early testing of such new substances inanimal experiments is an important task. It shouldbe remembered that the latent period of developmentof a tumour, according to our data, is equal toabout one-fifth of the longest life-span, thus con-stituting 15-18 years in man but only about 6 or at themost 12 months for the laboratory mouse or rat,which has a life-span of not more than 21/2 years.Thus it is possible to detect blastomogenous pro-perties in a particular substance experimentallylong before it manifests its effect on people.So far it is mainly questions connected with

harmful occupational factors that have been dealtwith. This is natural because, historically, observa-tion of occupational tumours led to experimental

653

L. M. SHABAD

study of carcinogens. Systematic research in animalexperiments on the possible carcinogenic effect of anumber of new products of the chemical industry isundoubtedly of assistance in evolving measures toprevent occupational cancer. However, if somewhatmore profound consideration is given to the possibleresults of such work, it will be realized that theyrange far beyond the mere control of occupationalhazards. Harmful blastomogenous factors maydevelop from occupational factors into generalenvironmental factors, affecting large numbers ofthe people and not merely a restricted group ofworkers in the branch of industry concerned. Thismay happen in the case of many products of theaniline-dye industry and many other dyes. Forexample, at one time a colouring agent known as" Buttergelb " or " butter yellow " was, in a numberof countries, added to foodstuffs, particularly butter,margarine, flour and macaroni; this agent was none

other than dimethylaminoazobenzene, a carcinogenicsubstance that induces liver tumours in rats and mice,as was established by Kinosita as long ago as 1937.Among the numerous dyes and all the various

preservatives used in the food industry today theremay still be substances with possible blastomogenousactivity. It is no accident that, as mentioned later, a

number of special international conferences havebeen devoted to this problem.Another example of the transformation of an

occupational factor into a general environmentalfactor is the classic case of chimney-sweeps' cancer.

There is no doubt that this form of occupationalcancer was caused by the products of fuel com-

bustion: tars, soot, and smoke. Now smoke fromhouse and factory chimneys still enters the atmos-phere. Consequently, it naturally springs to mind thatthe atmosphere may be polluted with carcinogens,particularly those carcinogenic hydrocarbons con-

tained in coal-tar and similar products. The wonderis not that such pollution does actually occur, butthat it has only recently been proved to occur bymeans of the modern, precise methods of spectralfluorescence analysis.About ten years ago, the author and a group of

co-workers began a systematic study of atmosphericpollution with the powerful carcinogen 3,4-benz-pyrene, and established that it was present inthe air of a number of Soviet cities (Gurinov, Zore,Ilyina & Shabad, 1953). Later, many facts were

accumulated in extensive research by a number ofSoviet workers concerning the quantitative distribu-tion of 3,4-benzpyrene in various cities and various

districts of the same city, and the relationship be-tween this pollution and many sources of contamina-tion of the human environment with 3,4-benzpyrene(Shabad & Dikun, 1959).

Details of the study of atmospheric pollution withcarcinogens need not be given here; the question isnow attracting the attention of many researchworkers and has obviously already become arecognized feature of general and municipal hygieneand experimental oncology. Only one aspect of thematter needs emphasis here-the possibility ofcarrying out successful campaigns against suchpollution and of preventing the discharge of 3,4-benzpyrene into the air or considerably reducing thequantity so discharged. This is a matter of taking anumber of technological and sanitary-engineeringmeasures to improve the conditions of fuel combus-tion and to trap the fumes discharged. Rationaltown-planning and the large-scale use of districtheating and electrification are also of great import-ance. A special investigation carried out by Dikun& Nikberg (1958) in the neighbourhood of a coke-and-pitch factory of an obsolete type in the city ofMakeyevka showed that, after the design of the cokeovens had been improved, there was a sharp reduc-tion in atmospheric pollution with 3,4-benzpyrene.Comparative studies of snow samples in two neigh-bouring cities of Siberia-Irkutsk and Angarsk-showed that in the new, properly planned town ofAngarsk, where the residential quarters are separatedfrom the factory districts by a green belt and areserved by a district-heating plant, only very smallquantities of 3,4-benzpyrene were found, whereas inthe old city of Irkutsk the air contained quantitiesof the pollutant equally as large as those present inother big industrial cities (Shabad & Dikun, 1959).The study of atmospheric pollution with 3,4-

benzpyrene is of great interest in connexion with theproblem of lung cancer. It is now generally recog-nized that the frequency of lung cancer has greatlyincreased. Many workers (including ourselves)consider that one of the reasons for this increase inthe incidence of lung cancer is the presence of carci-nogens in the air inhaled by human beings. It iswell known that city dwellers contract lung cancermore often than the rural population, and that it isencountered more often in large cities than in smallones. It is characteristic that contamination of theair with 3,4-benzpyrene, as our data in particularhave shown (Shabad & Dikun, 1959), is considerablymore marked in large cities than in smaller cities andsmall towns, and is practically non-existent away

654

ROLE OF EXPERIMENTAL RESEARCH IN THE PREVENTION OF MALIGNANT TUMOURS

from towns and large industrial areas. Thus,campaigning for cleansing the air of carcinogens isone method of approach to the prophylaxis of lungcancer.

Carcinogenic hydrocarbons may enter the humanlungs not only through atmospheric pollution butalso through the smoking of tobacco. The idea thatthere is a possible connexion between lung cancerand smoking is based on clinical statistics andphysico-chemical research.The data obtained from clinical statistics, mainly

in the United States and the United Kingdom, leadto the conclusion that cancer of the lungs is foundconsiderably more often among cigarette smokersthan among non-smokers.

Spectral fluorescence analysis techniques have beenused by a number of authors to detect in tobaccosmoke that same well-known carcinogen-3,4-benz-pyrene. Thus, according to the data of Dikun &Chushkin (1959), which fully correspond to thefindings of research workers in other countries,1.6 y of 3,4-benzpyrene is found in the smoke from100 cigarettes without filter tips and 1.1 y in thesmoke from 100 Russian papirosy (cigarettes withlong cardboard holders). Thus, a carcinogen is con-tained in the products of smoking, although in smallquantities. This is also confirmed by the results ofexperiments on animals. According to Gritsyute &Mironova (1960), painting the skin of mice withtobacco tar led to the development of tumours onlyin rare instances (in 1.5-2%. of cases only).

Experimental research on the part played bysmoking in the causation of lung cancer should becontinued along various lines: for example, a searchshould be made for harmful substances other than3,4-benzpyrene in the products of smoking; and amore thorough study should be made of the relation-ship between the appearance of carcinogens intobacco smoke and the combustion conditions of thetobacco and the nature of the smoking. However,the facts already accumulated provide a basis for anumber of methods of prophylaxis against thiscarcinogenic factor: on the one hand, obviously, areduction in smoking or the giving-up of smokingaltogether; on the other hand, improvements in theproduction of smokers' requisites, such as thedesign of special filters to trap the carcinogens, specialtreatment of the cigarette paper, etc.

It is of great importance for solving the problemof lung cancer prophylaxis to devise an experimentalmodel as close as possible to bronchogenic cancer ofthe human lung. Without dwelling here in detail on

this problem, on which a number of researches havebeen carried out in our laboratory at various times,it should be recalled that many research workersover a period of some forty years have tried to induceexperimental lung cancer by introducing carcinogensinto lung tissue, but without success. Apparently,the most important factor in the induction of lungcancer is the deposition of the carcinogen over asufficiently long period. Very recently, in ourlaboratory (Pylev, 1961) success was achieved inregularly inducing lung cancer in 30% of rats bymeans of intrabronchial administration, by intuba-tion, of 6-10 mg of 9,10-dimethyl-1 ,2-benzanthracene,incorporated in Indian ink to improve its adsorption.Above, a great deal has been said about the con-

tent of carcinogens, particularly 3,4-benzpyrene, invarious types of smoke. It was natural to consideralso the possibility of carcinogens being present insmoke-cured foodstuffs, which are sometimes sub-jected to lengthy smoke treatment. The componentsof the smoke may not only affect the surface of thefoodstuff but also penetrate into the inside.

In 1954, the Czech research worker gula and hiscolleagues detected 3,4-benzpyrene in smoked meat(Dobes, Hopp & gula, 1954). Gorelova & Dikun(1958) carried out systematic quantitative determina-tions of 3,4-benzpyrene in several forms of smokedfish, and in smoked sausages. Spectral fluorescenceanalysis showed that 1 kg of fish smoked by theordinary hot process may contain between 1.7 yand 7.5 y of 3,4-benzpyrene. At the same time, itwas found that as much as 38% of the total amountof 3,4-benzpyrene detected in smoked fish maypenetrate inside the fish when it is smoked whole,and as much as 87% when it is smoked in fillets.

In various varieties of sausage, Gorelova & Dikundetected between 1.9 y and 10.5 y of 3,4-benzpyreneper kg; 65% of the amount of substance foundpenetrated inside the sausage.The above data show that carcinogens are actually

present in smoked foodstuffs. What is their patho-genic importance? It must be emphasized that thequantities of 3,4-benzpyrene found are extremelysmall, but cannot be neglected, especially in caseswhen, for some reason or other, smoked productsmake up a particularly large proportion of the diet.This proposition can be illustrated in the followingway.

Voitelovich, Dikun, Dymarski & Shabad (1957)studied the frequency of malignant tumours in theTukum District of the Latvian SSR. The data forthe fishing population in two places on the coast were

655

L. M. SHABAD

compared with data for the population in six agri-cultural areas well inland, but in the same district.It was found that the mean annual morbidity frommalignant tumours, in terms of cases per 100 000inhabitants, was 318 in the places on the coast,whereas in the inland areas it was only 149. Thisdifference is due to the considerably higher frequencyof malignant tumours of the digestive tract among thepopulation in the coastal places; such tumours arethree or four times as common among the fishingpopulation as among the agricultural population ofthe inland areas. It is interesting to note that thefrequency of mammary-gland tumours and tumoursof the uterus and ovaries is identical in the twopopulation groups.Another example which suggests that extensive

consumption of smoked products possibly plays apart in the causation of cancer, and particularlycancer of the digestive organs, in man is the studyby Kaufman, Mironova & Shabad (1959) of thecauses of death among workers in sausage and fish-curing factories, who in the nature of things eatlarge quantitites of smoked food. It was found thatthese workers do indeed contract cancer with parti-cular frequency, especially cancer of the gastro-intestinal tract.

Thus, the observed facts suggest that smokedfoods, if consumed in large quantities, may be ofimportance as a carcinogenic agent for man. Thismakes it necessary to look for ways and means ofreducing the quantity of such known carcinogens as3,4-benzpyrene in food products, or of completelyeliminating them.

Present food-curing techniques are based on veryancient principles-the same principles, in fact,that were applied by primitive man when he hungpieces of meat over the smoke of his camp fire.Naturally it is essential now to devise new curingtechniques, based on the latest discoveries ofphysics and chemistry. In doing so, it must beremembered that it is essential to prevent carcino-genic admixtures in foodstuffs. It is difficult toimagine any way of reducing the quantity of 3,4-benzpyrene or, even more, any way of ensuring itscomplete absence, in smoke, which consists of a verycomplex mixture of substances, difficult to control.For that reason, our attention has been concentratedrecently on methods of curing in which either smokeis not used at all or the period of action of the smokeis reduced to a minimum. What is aimed at is, so tospeak, " smokeless " curing or curing with littlesmoke. There is a possibility of doing this by treat-

ing foodstuffs with curing fluids by the so-called" wet-curing " method.

Recently, Gorelova, Dikun & Lapshin (1959)reported that fish treated with curing fluids containsfar less 3,4-benzpyrene than fish smoke-cured in theusual way. The use of new methods of curing, andparticularly of curing fluids, which can be analysed toensure that they contain only a very small quantityof 3,4-benzpyrene (or none at all), is one of the waysof preventing exposure of the human organism tocarcinogens.

Further improvements in curing methods requirestill more experimental work. The effect of variouscuring fluids or other means of preparing smokedproducts can be investigated in animal experiments.There is no doubt that in this case also the results ofexperimental research will promote tumour pro-phylaxis among human beings.As stated above, the problem of preventing human

exposure to carcinogens is in the forefront ofthe prob-lems of occupational, municipal and food hygiene.Special attention is being paid to the possibilityof the presence of carcinogenic additives in food-stuffs and to measures directed towards eliminatingsuch additives.The interests of food hygiene in regard to blasto-

mogenic substances are by no means only concernedwith the problem of food curing. At the presenttime, the food industry uses large quantities ofcolouring agents, preservatives, flavourings and othersubstances needed in the production of certainfoodstuffs. Naturally, any colouring agents that mayarouse even a mere suspicion of having blastomo-genic properties should be withdrawn from the foodindustry. The colouring agent " butter yellow "(dimethylaminoazobenzene), which, as mentionedabove, has been shown experimentally to induceliver tumours, is now prohibited in all countries (ithas never been used in the food industry of theUSSR). Systematic efforts are now being made tofind substitutes even for those food colours which,although experimentally they have not inducedtumours, are closely similar to blastomogenicsubstances in their chemical structure. Thus,measures have been taken to find substitutes fornaphthol yellow, various forms of Sudan, etc.

Special interest is being aroused by the problem ofinsecticides, herbicides, substances to inhibit germi-nation in stored vegetables, and many other chemi-cals that may enter human food together with agri-cultural products. Without going into detail on thissubject, it may be mentioned that many compounds

656

ROLE OF EXPERIMENTAL RESEARCH IN THE PREVENTION OF MALIGNANT TUMOURS

allied to urethane, which is known to induce certaintumours under experimental conditions, have beensuggested for preventing germination in storedvegetables. Systematic preliminary experimentalresearch on a number of substances to prevent ger-

mination will make it possible to reject dangeroussubstances and substitute harmless ones for them(Shabad & Naumova, 1956).The problem of the possible presence of carcino-

genic additives in food has attracted the attention ofa number of international organizations, particularlyFAO and WHO. The Joint FAO/WHO ExpertCommittee on Food Additives (1961) adopted a

number of general basic principles, including thefollowing:

". . . Food additives should be used for the benefit ofthe consumer; they should not be permitted for any

deceptive or misleading purpose.

"... Food materials and all chemicals and processes

at present in use for the preparation and distribution offood should be subjected to adequate examination toensure minimum risk in use, if this has not already beendone, or unless previous knowledge indicates that this isunnecessary.

.... All chemicals and processes proposed in the futureshould be subjected to adequate examination to minimizerisk before being accepted for use.

". . The presence of carcinogenic substances in foodmight be a significant factor in the occurrence of what isconsidered to be spontaneous cancer in man and animals.

"... Since dose-response relationships have beendemonstrated in the case of carcinogenic agents, thereduction of carcinogenic substances in food to the lowestpracticable level may be one of the effective measures

towards cancer prevention.'.... Many factors may influence dose-response in

carcinogenesis. Their complexities are such that it isagreed that no assuredly safe level for carcinogens inhuman food can be determined from experimentalfindings at the present time."

These quotations show the very great importanceof the campaign against carcinogenic food additives.It is clear that experimental research may play a

very significant role in furthering the cause of cancer

prophylaxis. In the planning of such research, thepurpose of the experiments and, more important,their range, need to be clearly defined. For example,in order to evaluate the carcinogenic effect of a par-

ticular food additive, it is not enough to set up

experiments based on administering the substancesubcutaneously to animals or painting it on theirskins; it is essential to introduce the additive intotheir digestive tract. The fact that sarcomas are

obtained at the site of subcutaneous injection doesnot mean that the same product will have a carcino-genic effect if given by mouth. To illustrate thisstatement, it is enough to refer to the well-knownresults of experiments involving the subcutaneousinsertion in rats of polymer plates. The fact thatsarcomas developed around the plates would neverlead anybody to suspect plastic tableware of havinga carcinogenic effect.A very important but also very difficult problem,

particularly in relation to carcinogenic food addi-tives, is that of the possibility of setting hygienicstandards for carcinogens. This problem has beendiscussed on many occasions at a number of nationaland international conferences, and the general con-clusion is that it is impossible to lay down maximumpermissible doses of carcinogens. This is not onlybecause it is extremely difficult or even impossibleto make quantitative determinations of manycarcinogens. The important point is the fact thatthe information at present available on the mecha-nism of action of carcinogens suggests that it isimpossible to determine what would be permissibledoses. For instance, it has been proved that carci-nogenic aminoazo compounds and hydrocarbonscombine with proteins. Factors which must not beforgotten are the accumulation of carcinogens in theorganism and the possibility of co-carcinogenesis,that is, the potentiation of the effect of carcinogensby non-specific factors.The impossibility of establishing maximum per-

missible doses for carcinogens in no way means thatit is impossible to devise prophylactic measures.Cancer prophylaxis is served by reducing to thegreatest possible extent the doses of carcinogens towhich man may be exposed. This has been soundlydemonstrated in many thousands of animal experi-ments, and in the case of human beings, in regard tooccupational neoplasms and tumours among smokers.The lower the dose of the carcinogen, the smaller thenumber of people who contract the disease and thelater its onset. It should be borne in mind that ifonly very small doses of carcinogens are ingested,they may induce cancer so late that it will practicallyfall outside the normal human life-span.

It has naturally not been possible in this review tooutline all the aspects of experimental oncology thatmay be put to use for cancer prophylaxis. Thus, nomention has been made of radiation-inducedtumours: attention has been concentrated onchemical carcinogens alone. However, the materialquoted seems to us sufficient to throw light on the

657

658 L. M. SHABAD

principles of the problem. Experimental researchdoes, indeed, contribute to the prevention of tu-mours, since it discovers carcinogenic agents, and,once a substance is known to be a carcinogen,various measures can be taken against it. For ex-ample, it may be eliminated altogether from thehuman environment by the prohibition of its usein industry or in the home; its effect on human beingsmay be minimized by such means as sealing offproduction processes and thus reducing human

exposure to contact with it; it may be renderedharmless (" decarcinogenized"); it may be replacedby some less harmful or entirely harmless product;the dose or length of time to which people areexposed to it may be reduced. All such prophylacticmeasures should be based on the correspondingexperimental findings. It is in this way that experi-mental oncology, which grew up as a result of thepractical observations of occupational tumours inman, is now repaying its debt to practical medicine.

RtSUMt

D'apres les travaux de l'auteur et de ses collaborateurs,la prophylaxie des tumeurs, consideree du point de vuedu pathologiste et de l'experimentateur, revet deuxaspects: d'une part l'hygiene anticancereuse, destinee 'aprevenir l'apparition du cancer; d'autre part, la prophy-laxie clinique, dont le but est d'enrayer le developpementdu cancer.La prophylaxie clinique consiste a deceler et ai traiter

a temps les manifestations precancereuses, car tout cancera son # precancer #. Les alterations precancereuses desdivers organes presentent un tableau morphologiquedifferent, mais comportent des traits communs. On peutdistinguer quatre phases principales dans le develop-pement des tumeurs malignes: 1. hyperplasie diffuse,irreguliere; 2. proliferations nodulaires; 3. tumeursrelativement benignes; 4. tumeurs malignes. 11 importede distinguer les processus precancereux des alterationsinflammatoires et r6g6n6ratrices. L'inflammation n'est pasnecessairement l'un des maillons qui conduisent au cancer.

Les mesures d'hygiene anticancereuse tentent de pre-venir l'apparition du cancer en empechant les agentscancerogenes, en particulier ceux du milieu ambiant,d'agir sur l'organisme humain. L'id6e de l'action cance-rogene de telles substances se fonde sur de nombreusesobservations de cancers professionnels. I1 suffit en effetde supprimer les agents en cause, ou de reduire leuraction, pour voir diminuer la frequence de ces cancersprofessionnels. C'est ce qui permet de parler d'unehygiene anticancereuse.Le nombre croissant de nouveaux produits chimiques

avec lesquels l'homme entre en contact a l'heure actuelle,

impose l'6tude de 1'eventuel pouvoir cancerogene de cessubstances, sur l'animal d'experience. Les additifs alimen-taires sont l'une des premieres classes de produits a sur-veiller. L'auteur insiste notamment sur la presence debenzo-3,4 pyrene dans les poissons et les saucissesfumes. On a constate que le cancer de l'appareil digestifest plus fr6quent chez les personnes qui consomment desaliments fum6s en grande quantit6 (p6cheurs, ouvriersdes entreprises de fumage), que dans la population envi-ronnante. On met actuellement au point des techniquesde fumage qui permettent de r6duire les hydrocarburescancerogenes dans les aliments fumes.Le pouvoir cancerogene des polluants de l'atmosphere

est l'objet de vastes recherches. L'auteur et ses collabo-rateurs ont cherch6 l'origine de plusieurs de ces substan-ces; ils ont procede a une etude comparative de la situa-tion d'un grand nombre de villes et ont demontre que,dans certains cas, un am6nagement plus rationnel desvilles et la combustion plus complete des combustiblespermettraient de diminuer la quantite des polluantscancerogenes. On a egalement etudie les substances can-cerogenes que l'on risque d'absorber en fumant du tabac,et les mesures prophylactiques qu'il y aurait lieu deprendre. On est parvenu a provoquer des tumeurs expe-rimentales chez le rat, par injection d'hydrocarburescancerogenes par voie intratracheale, ce qui pourra servirde base a la recherche de moyens preventifs applicables'a 'homme.

L'auteur se d6clare convaincu que la prophylaxie ducancer est possible, et que les recherches et experiencesentreprises lui donnent une base scientifique.

REFERENCES

Bogovski, P. A. (1960) [Occupational skin tumoursinduced by products of the processing of mineral fuels],Moscow, Medgz

Danetskaya, 0. L. (1952) Gig. i Sanit., No. 10, p. 26Dikun, P. P. & Chushkin, S. G. (1959) Vop. Onkol., 5,No. 7, p. 34

Dikun, P. P. & Nikberg, I. 1. (1958) Vop. Onkol., 4,No. 6, p. 669

Dobeg, M., Hopp, K. & gula, J. (1954) esl. Onkol., 1,No. 3-4, p. 254

Gorelova, N. D. & Dikun, P. P. (1958) Vop. OnkoL, 4,No. 4, pp. 398, 405

ROLE OF EXPERIMENTAL RESEARCH IN THE PREVENTION OF MALIGNANT TUMOURS 659

Gorelova, N. D., Dikun, P. P. & Lapshin, I. I. (1959)Vop. Onkol., 5, No. 9, p. 341

Gritsyute, L. A. & Mironova, A. I. (1960) Vop. Onkol.,6, No. 8, p. 25

Gurinov, B. P., Zore, V. A., Ilyina, A. A. & Shabad, L. M.(1953) Gig. i Sanit., No. 2, p. 10

Joint FAO/WHO Expert Committee on Food Additives(1961) Wld Hlth Org. techn. Rep. Ser., 220, 4

Kaufman, B. D., Mironova, A. I. & Shabad, L. M. (1959)Vop. Onkol., 5, No. 9, p. 314

Kinosita, R. (1937) Trans. Soc. path. jap., 27, 665Kleinenberg, G. E. (1939) Arh. biol. Nauk, 56, No. 3,

pp. 39, 48Larionov, L. F., Soboleva, N. G. & Shabad, L. M. (1934)

Vestn. Rentgenol. Radiol., 13, No. 1-2, p. 131Malyugina, L. L. (1955) Vop. Onkol., 1, No. 3, p. 105Monastyrskaya, B. I. (1952) [Some new experimental dataon the prophylactic role of chronic inflammation in thegenesis of skin cancer induced by carcinogens]. In:[Symposium on Problems of Oncology], vol. 4, p. 136

Pliss, G. B. (1958) Vop. Onkol., 4, No. 6, p. 659Pliss, G. B. (1959) Vop. Onkol., 5, No. 7, p. 101Prokofyeva, 0. G. (1938) Arh. biol. Nauk, 51, No. 3,

p. 120Prokofyeva, 0. G. (1952) [The effect of carcinogens on

inflammatory proliferations of epithelium]. In: [Sym-posium on Problems of Oncology], vol. 5, p. 170

Pylev, L. N. (1961) Bull. exp. Biol. Med., 52, No. 11,p.99

Shabad, L. M. & Dikun, P. P. (1959) [Atmospheric pollu-tion with the carcinogen 3,4-benzpyrene], Moscow,Medgiz

Shabad, L. M. & Naumova, L. P. (1956) Vop. Pitan.,No. 2, p. 27

Vasilyev, Y. M. (1958) Brit. J. Cancer, 12, 524Vasilyev, Y. M. (1959) J. nat. Cancer Inst., 23, 441Verkhovskaya, R. I. (1941) Arh. biol. Nauk., 61, No. 2,

p. 47Voitelovich, E. A., Dikun, P. P., Dymarski, L. Y. &

Shabad, L. M. (1957) Vop. Onkol., 3, No. 3, p. 351