Tobacco Documents Online

J Experimental Tumor Research Vol. 1 Editor: F. Id O M B U R G E R, Cambridge, Mase. BASEL (Switzerland) S. KARGER NEW YORK Separatum Vol. 1 (1960) PPrinted ia 3witzerl.ad HeerMrder F. asd Tregler A.: Modifying faotors in vreiaogenesis. . Progr. eV. Tumor Ret,, voL z, pp. St x-3sY (Karger, Basel J.New York ig6o). . Bio-Research Institute, Cambridge, Mass. Modifying Factors in Carcinogenesis* F. HOMBURGER and A. TREGIER CONTENTS Sex Modifying Carcinogenesis ., . . . . . . . . . . . ., . . . . . . 318 Immunologic Factors Modifying Carcinogenesis . . . . . . . . . . . . . . 321' Carcinogens Inhibiting Carcinogen Induced Tumors . . . . . . . . . . . 323 Discussion and Conclusions _ . . . . . . . . . . . . . . . . . . . . . . 326 References . . . . . . . . . . . . . . ., . . . . . . . . . . . . . . . 327 In recent years interest in the study of chemical carcinogens has increased and investigators in this field are devoting consider- able time, thought and experimentation to methodologic problems not hitherto thoroughly explored. Many discussion meetings such as the Gordon Conference and two sessions sponsored by the Tobac- co Industry Research Committee have been held and their deliber- ations, while unpublished, have had impact on the work of those in this field. An issue of the British Medical Bulletin (MAY, 1958) has been largely devoted to this problem, and a recent Ciba Foundation Symposium from which we shall repeatedly quote below, has re- vealed the thinking of a group of workers from eleven countries on the subject' of Carcinogenesis: Mechanisms of Action, WOLSTEN- HOLME AND O'CONNOR (1959) ~ The present state of knowledge-or ignorance-in this field has been succinctly summarized by I. HIEGER (',Ioc. cit. 1959) as follows: "While we have at present a number of theories of carcinogenesis, * These studies supportedi in part, by grants from the Tobacco Industry Research Committee, New York, N. Y., and by a Public Health Service Research Grant (C4338) from the Nationali Cancer Institute, U. S. Public Health Service.

312 H o m b u r g c r, T r e g i e r, Modifying Factors in Carcinogenesis there is as yet' no theory of carcinogenesis; that' these hypotheses have scarcely proved more than restatements of the facts of experi- ment or observation;, and that the very tentative nature of these ideas is a: measure of the difficulties of our formidable and wonderful problem. HEIDELBERGER (1oc. cit. 1959)~ aptly observed that "the mechanism of carcinogenesis is a mirror into which we look and see ourselves," implying that the immunologist sees an immunologic theory, the biochemist a biochemical theory, and so forth. All theories, however, no matter how esoteric or how pertinent, must be based upon or find their confirmation in biologic experi- ment. Inherent in the biologic experiment,, (whether it'~ be conducted in humans, such as studies on occupational carcinogenesis, or in other mammals, lower animals or plants, or in tissue culture), there are a number of intrinsic variables which are difficult to control, the very presence of which renders complex the interpretation of data obtained. MUHLBOCK (toc. cit. 1959)', has stated this for hormonal car- cinogenesis as follows: "In studying the carcinogenic effect of hor- mones it must be kept in mind, however, that an organ~ exposed to hormonal influence does not function isolated in the body; but' is invariably interrelated with other organs and tissues which! may exert an important influence on this function." HACKMANN (1OC. cit'. 1959), speaking in broader terms, notes that "we cannot overlook the important fact that the development of malignant tumors is not only governed by the degeneration of the cells but is also influenced by other factors,, the total evaluation of which is a considerable prob- lem. Thus, it is difficult to determine the role played in the devel- opment of malignant tumors by the chemical substance used or by other fact'ors."' The decision of whether a substance is carcinogenic or not car- cinogenic and if so, to what degree, is customarily based on the evaluation of end results; that is to say that one determines how many tumors appear in a given set of experimental conditions, how rapidly and where they develop, and how they may alter the sur- vival of the test animal. All of these criteria merely represent the end result of the interaction of many variables which singly, or in combination, may alter the final effect of a given substance under a given set of rigidly controlled conditions. One widely accepted view of carcinogenesis holds that funda- mentally, carcinogenesis is a two-step process consisting in : 1) "initi- C N N ~ CD N C~

H o m b ux g e r, T r e g i e r, Modifying Factors iniCarcinogenesis 313 ation", and 2) "promotion" BERENBLUM (lbc. cit. 1959). This ig- nores the third phase through which every tumor must pass before it becomes visible to microscopic examination or grossly palpable, namely that of "progression";* for only those malignant changes which have progressed sufficiently to be detected are taken into ac- count in determining the carcinogenic potency of any substance. It is not always possible to ascertain which of the three phases of tumor formation may be affected by any given variable which alters the end result of a carcinogen~study. This is especially difficult in differentiating factors altering the cells' sensitivity to initiation from those affecting promotion, but it would be technically feasiblee to distinguish these factors from those which are determining the rate of progression. Thus it is clear that if a given variable alters only the incidence of tumors and not their growth rate,,it is morelikely to act by chang- ing induction and promotion, whereas: if only the rate of tumor growth is affecteds the mechanism of such a change should be sought in factors affecting progression. This distinction, however, is not absolute since a variable sup- pressing the progression of some of the tumors that have been in- duced and promoted but failed to reach palpable dimensions could' simulate a lowered incidence of malignancy. It is, therefore, suggested that studies on~ carcinogenicity should! include careful microscopic examination of the areas exposed to test substances during early phases of such experiments. It has been found! in the writer's studies of the carcinogenic ef. fects of single subcutaneous doses of 3, 4, 9, 10-dibenzpyrene that morphologic criteria of malignancy become microscopically detect- able (Fig. 1) as early as 5 weeks after injection of the carcinogen, whereas palpable tumors appear rarely before 9 weeks after in- jection. What keeps in check the morphologically malignant cells for weeks before they proliferate is not clear and is one of the un- known variables likely to affect progression, rather than initiation or promotion. Unless microscopic studies are done and rule out the presence of * The term "progression" is here used in its medicaLsense meaning, "going, for- ward, advancing", and' is here considered synonymous with growth, proliferation4 extension in space, and not in the sense in which Fovnos (1954)I employed it as "indicating development of a tumor by way of permanent, irreversible qualitative change in~one or more of its characters."

314 H o mb u r g e r, T r e g i e r, Modifying Factors inCarcinogenesis a Fig, 1. (a) Earliest detectable change in the histogenesis of a fibrosarcoma. Note greatt increase in nuclear size and vesicular appearance of transforming cell (A) in compari: son with normal fibroblast (B)i C57Br/bd mouse 5 weeks after subcutaneous injection of 3, 4', 9, 101 dibenzpyrene. (b) Fully developed fibrosarcoma in a C57 Br/cd' mouse 12 weeks after subcutaneous injection of 3, 4, 9, 10 dibenzpyrene for comparison with Figure laI

I+P o m b u r g e r, T r e g i e r„ Modifying Factors in Carcinogenesis 315 morphologically malignant cells during the so-called "latent" phase after induction of carcinogenesis, the mechanism of so-called "pro- moting" factors cannot be elucidated and it is always possible that they act either by impairing some host defense which holds in check an already malignant cell or by actually intensifying, whatever chemical process it may be that transforms a normal cell into a malignant one. In other words there cam be acceleration of pro- gression by removing obstacles to malignant growth or by intensi- fying malignancy. What are these biologic factors that modify thusly the phenom- enology of carcinogen-induced tumors a It is possible for any of thee several factors which have been shown to alter the outcome of ex- periments in carcinogenesis to act in this fashion during the phase of progression, and it! is difficult to demonstrate the exact mode of action of such factors. Those of greatest int'erest'~ to us have been sex, immunologic factors and the effect of the carcinogen itse f The original' studies on which are based most considerations in this essay were carried out using the relatively new carcinogen 3, 4, 9, 101dibenzpyrene first described in 1957 by LACASSACNE, ZAD- JELA, Buu-Hoi' and CiIALVET. These authors injected 0:6 mg of 3, 4, 9, 10-dibenzpyrene in peanut oil, three times at monthly intervals into the subcutaneous tissue of eleven mice of Strain XVII and! ob- tained sarcomas appearing in some animals as early as on the 42 nd day of the experiment, and leading to t!he death of all animals within 135 days.. WARAVDEKAR and RANADINE (1958) confirmed these findings in eight male and eight female F2 hybrids of the XVII and C57BI strains, which were given single subcutaneous injections of 2 mg of 3, 4, 9, 10-dibenzpyrene in propylene glycol'. Tumors appeared im all animals 2-3 months after treatment, with an average time of latency of 74 days. Morphologically, these tumors were fibrosarcomas. Their transplantability was not ascertained, but they had all characteris- tics of cytologic malignancy and invasiveness. A first supply of 3, 4, 9, 10-dfbenzpyrene was graciously given us by Dr. LACASS.acNE. Later experiments were carried out using& material prepared for us by Arthur D. Little, Inc., Cambridge, as sample number C-61947. The infrared spectrum of this material was superimposable on that! of the previous French sample. Most of our studies dealt with single subcutaneous injections of 3, 4, 9~ 1!0-dibenzpyrene in peanut oil or tricaprylate and to date, Ms j~ . -~P:4

316 H o m b u r g e r„ T r e g i' e r, Modifying Factors in Carcinogenesi's over 10,000 tumors have been produced in mice. Those of these tumors that'~ have been examined histologically were found to be fibrosarcomas and leiomyosarcornas. One rhabdo-myosarcoma was also found. In an additional series of intradermal injections, two epidermal carcinomas were produced (Fig. 2).* In contrast to the great susceptibility of subcutaneous tissue and epidermis to 3, 4, 9, 10-dibenzpyrene carcinogenicity, the endo- metrial lining appears to be resistant. So far, none of 30 animals receiving 500 gamma of the carcinogen in intrauterine injections have developed any tumors, although the chemical' has remained in utero for up to 32 weeks. * The cooperation of AcxES B. RussFtELD;, M. D., Ph. D.,, iin the stud{y of the histopathology of these tumors is gratefulNy acknowledged. Fig. 2. Epidermoid carcinoma developed 15 weeks after intradermall injection of 3, 4, 9, 10-dibenzpyrene inia C57 Br mouse. O w ~ ~ ~ W CJ Q

Csis , of these ,nd to be :oma was ions, two )us tissue the endo- ) animals injections nained in study of the H o m b u r g e r, T r e g i e r, Modifying Factors in Carcinogenesis 317 The minimal effective subcutaneous dose was determined in male C57Br/cd mice obtained from the Jackson Memorial Lab- oratory, Bar Harbor, Maine, and the results of this experiment are shown in Figure 3 and Table I. It appears from this experiment (limited to groups of 12 animals each), that no tumors are produced at dosages below 1 gamma and that there is a dose-effect relation- ship up to 50 gamma beyond which dosage all animals develop tumors within 18 weeks. This last point has been re-emphasized in a large scale experiment involving 8,850 males of the C57B1/6 strain wherein nearly 100 °;; of the animals developed tumors within 22 weeks following a single injection of 500 gamma of 3, 4, 9, 10-dibenzpyrene in peanut oil. There have been no regressions, and these tumors pro- gress rapidly in size, killing the host within 4 weeks from the time when the tumor becomes first palpable. No metastases were observ- ed and all tumors tested proved to be transplantable within the strain of origin. Injections into the axilla produced tumors slightly more rapidly than injections into the groin. This difference may be due to the anatomical characteristics of these areas which permit perhaps more ready palpation of tumors in the axilla. Since tumors tend to invade and adhere to the chest wall in this site, whereas they remain more discreet in the groin, all later injections were made in the inguinal region. 100'r 90 80, m70- 0 E 60- ~ ~ 50 n ~ 40: n a 30- 20- I1°o 0 ° 1.0Y ~ 10~ ~ x) x 00 2 4 6 8 10 12 14 16 18 20 22 24 2654 56 Time after injection (weeks) Fig. 3. Curves showing the incidence of subcutaneous sarcomas arising from single m of 3, 4. 9, subcutaneous injections of 3, 4, 9, 10-dibenzpyrene in C57BrJcd mice and the time (in weeks) at which these neoplasms occur (see also Table I).

318 H o m b u r g,e r, T r e g i e r, ModiB~ing Factors in Carcinogenesis Table I Single Subcutaneous Dose, 3S 4,~ 9. ~ o- Dtbenzpyrene. . in Gammas No. A'nimals~~ at Start I No. Weeks Elapsed'~when fo %~, Animals ~ Had Tumors~ No. Weeks Elapsed ~when~ All Animals . Had Tumors~. I No: Atumals ~ Dead of Tumors~ No. Animal9~ Dead withoat Tumors No. Animals~~ Ahve withour~. Tumors~~ at 66 weeks ~ 600 11 7 17 11 - - 100 11 5-6 14, 11 - - 50 12 12-13 17 12 - - 25 12 13-14 - Ill 1 - 12.5 11 18-19 - 7 4 - 6.25 12' 66' - 6 11 5 2.0 12' - - 4 7 1 1.0 11 - - 1 4: 6 0.5 12 - - - 12 - 0.1 12 - - - 12 - 0:05 12 - - - 8 4' 0:01 10 - - - 4 6 Tumor Formation Following,Singie Subcutaneous Ihjfction of 3, 4, 9, 10-dibenz- pyrene (see also fig. 3)', The vehicles, peanut oil,, tricaprylate, or tricaprylate with: had'' no significant effect on tumor development and, therefore, peanut oil of the same batch has been used throughout. From the beginning it was evident that there are strain and species differences in susceptibility. A rhesus monkey (female, one year old) received 3.5 mg of 3, 4, 9, I9-dibenzpyrene subcutaneously on May 15, 1958, and has so far shown no tumor. Sarcomas were induced in rats and hamsters with 1.2 mg given subcutaneously, but some a'nimals in these two species failed to develop tumors. Among mice, the Swiss strain seemed most resistant; the C57Br/cd strain most susceptible. Sex Modify~ing Carcinogenesis The effecls of sex on the outcome of carcinogen experiments' has long been overlooked, probably because relatively high doses of car- cinogenic chemicals were used, resulting in a maximal tumor in- cidence in males and females. Studying the subcutaneous carcino- O ~ bJ .~ c0 W N gene> injec no si; and I betw, n}a1e, 2 me effect ever, tumo is the femal co-ca (1937 ed th malig ceptil' pyren out ir esis a ferenc mentr mal t vestig the C that r maless oping S can in but ai C posses induc+ 1 benzp C57'Bn days . metlic cell su~

H orn b u r g e r, T r e g i' e r,, Modifying Factors in Carcinogenesis 319 genesis with methylcholanthrene, BOYLAND and WARREN (1937),, injecting 1 mg of inethylcholanthrene subcutaneously in lard, found no significant sex differences. In the experiments of WARAVDEKAR and' RANADINE with~ 3, 4, 9; 10-dibenzpyrene, there was no difference between the tumor incidence and other characteristics seen in males and those in females. In the light of our data, their dose of 2 mg, per mouse appears an overwhelming one. When minimal effective doses were employed by LEtTER and SHEAR (1943) how- ever, significant differences became apparent' with males having, a tumor incid'ence 44% greater than that observe& in females. Whatt is the mechanism of this difference ? It would be more logical if females had~ greater susceptibility to induction of tumors since the co-carcinogenic effects of estrogens appear established. GILMOUR ( 1937). However, it should be noted that GILMOUR already conclud- ed that follicular sex hormone does not accelerate the growth of malignant cells,, even of the mamma, but that it' increased the sus- ceptibility of mouse skin to the carcinogenic effect of 3, 4-benz- pyrene, hastening the appearance of the tumor. WOLF (1952) points out in this connection that, "The processes involved in carcinogen- esis~ are probably much more complex than a mere direct inter- ference by the hormone... as might be expected in su& funda- mental processes as growth or abnormal growth, the whole abnor- mal balance of the organism comes into play." In the present in- vestigation, the incidence of induced tumors in males and females in the C57Br/cd strain was studied and it was observed consistently that at any given time, tumors in females were smaller than in males, and that, with small doses, the percentage of tumors devel- oping in females was significantly less than in males. Since this difference persisted~ in castrated males and females, it can~ indeed not be explained on the basis of hormone effects alone, but an explanation must be sought elsewhere. Consequently, it was decided to determine whether females possessed defense mechanisms against the growth (progression) of induced tumors when these are implanted subcutaneously. Tumors induced in C57Br/cd'males by means of 3, 4, 9, 10-di- benzpyrene were therefore implanted subcutaneously into female C57Br/cd' mice and male C57Br/cd mice, allowed to grow for 14 days and then dissected out and weighed. The transplantation method was that described by SNELL (1953) employing a 6 ~~ single cell suspension in Ringer's solutionprepared by means ofa cytosieve.

320 H o m~b u r g e r, T r e i g e r, Modifying Factors in Carcinogenesis The results are shown in Table II. While the statistical signi- ficance of the differences is borderline in.the individual groups, the pooled data are strongly significant and show that there is a more favorable environment for tumor growth in the male than in the female. Table II Age ofMice. ( Tumor Weight in Tumor IX'tight in I Signifi¢ance of DilBerence (Months). to9Gdes(Gm.). tuFemales(Gm.). I P I t 5-6 1.6 1.0 0.025-0.05 1L8 3-4 2.7 1.6 0.025-0.05 1 L86 2-3 1.65 1.14 0.025-0.05 1 L88 ' Tumor weights in males and females 14 days after cytosieve transplant of 3, 4, 9,, 10' dibenzpyrene-induced tumor Fig. 4. Showing the electrophoretic patterns seen im mice bearing tumors, in control' mice, and'in mice bearing implants of embryonic tissue. The histologic appearance of cross-sections of these mice, taken at the levels of the implants, are also shown. 0 ~ N ~ ~ ~ 1a