[CANCER RESEARCH35,3374-3383,November19751

Summary

Current awareness of the importance of environmentalfactors such as diet in the etiology of human cancer hasstimulated renewed interest in animal models for studyingeffects ofdiet on tumorigenesis. Diet can influence cancer inanimals by affecting the initiation or subsequent preneoplastic stage of tumorigenesis, but it has less effect on tumorgrowth. Caloric restriction has a general inhibitory influence on tumorigenesis. Dietary fat, on the other hand, tendsto promote tumorigenesis, but only certain types of tumors,such as mammary tumors, are affected. Both caloricrestriction and dietary fat appear to act primarily during thepreneoplastic stage, and their effects on hormone-dependenttumors may be mediated through changes in the hormonalenvironment. Variations in other dietary factors, such asprotein, vitamins, or minerals, above the levels required fornormal maintenance seem to have little influence on thegenesis or growth of tumors.

Introduction

The role of diet in tumorigenesis was a popular field ofinvestigation during the 1940's and early 1950's. It iscurrently enjoying a revival of interest because of increasingawareness of the role of environmental influences on cancermortality in human populations (33, 43) and the possibilitythat diet may be an important factor in this regard (15. 25,55, 106). In discussing the experimental evidence obtainedfrom work with animals, consideration will therefore begiven to its possible relevance to carcinogenesisjn humans.

Most of the experimental work on diet in relation tohormone-dependent cancers has been concerned with mammary tumors in mice or rats. Some investigators have usedspontaneous tumors (2, 80, 8 1), and others have usedtumors induced by various carcinogenic agents (27, 30, 38).The effects ofdifferent dietary components, such as protein,fat, vitamins, and minerals, have been investigated, as wellas the overall caloric intake; and a number of reviews of thiswork have been published (12, 15, 17, 74, 86, 87, 92, 94, 99).

In discussing nutritional influences, one must distinguishclearly between the genesis and the growth of tumors (87).Although there is good evidence that diet can affecttumorigenesis, it has much less influence on the growth ofestablished tumors. In general, overnutrition favors andundernutrition inhibits tumor growth (87), but underfeedingor caloric restriction are not practical methods of controlling the growth of tumors.

The process of tumorigenesis itself can be divided into atleast 2 stages. This concept was derived originally fromwork with skin tumors (3, 35), but it probably applies toother tumors, such as mammary tumors, as well (6, 17, 22,70). In the 1st or initiating stage, normal cells are convertedinto potentially neoplastic cells by interaction with acarcinogenic agent. This is followed by a developmental orpreneoplastic stage during which these cells undergo furtherchanges leading to tumor formation. This period mayactually comprise more than I stage in the development oftumor cells (6).

There are various ways in which diet may affect theinitiating stage of tumorigenesis. Some foods contain smallamounts of substances such as polycyclic hydrocarbons,aflatoxin, azoxyglucosides, and nitrosamines, which arepotentially capable of causing cancer (33, 41, 59, 68, 104).Diet can also influence the metabolism ofcarcinogens in thebody. In some cases this may lead to more rapid inactivation of carcinogens, and in other cases it may promote theirformation from precursor molecules. Such effects might bemediated by the influence of dietary components on inducible enzymes ( 19, 96) or by altering the intestinal microflora(25, 44, 45).

After cells have been exposed to a carcinogenic stimulus,it usually takes some time before tumors begin to develop.

During this preneoplastic stage the process seems to berelatively susceptible to external influences, and it is thenthat diet appears to exert the greatest effects on tumorigenesis. The following discussion will therefore be concernedprimarily with this aspect ofthe role ofdiet in tumorigenesisand will refer mainly to mammary tumors, since they havebeen investigated much more thoroughly than any otherhormone-dependent tumors in experimental animals.

1 Presented at the Conference on Nutrition in the Causation of Cancer,

May 19 to 22, 1975, Key Biscayne, Fla. Original research reported in thispaper was supported by Grant MT-l561 from the Medical ResearchCouncil of Canada.

2 Medical Research Associate of the Medical Research Council of

Canada.

Caloric Intake

Caloric restriction inhibits formation of spontaneousmammary tumors in mice (Table 1). Similar results have

CANCER RESEARCH VOL. 353374

Experimental Evidence of Dietary Factors and Hormone-dependent1

K•nn•thK. Carroll2Department of Biochemistry, University of Western Ontario, London, Ontario, Canada N6A 5C1

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

DegreeofKind

ofrestriction5restriction

(as% ofcontrol diet)Duration

ofstudy(mos.)Strain

ofmouseTumor

incidence(%)Ref.ControlRestrictedU5018DBA20/50

(40)1/50(2)Tannenbaum(79)C+F7516C3H30/44(67)0/51(0)Visscheretal.(95)C+F5022C3H32/32(100)3/19(16)Whiteezal.(l02)C6723DBA27/50(54)0/50(0)Tannenbaum(80)C7515C3H17/30(58)0/30 (0)Tannenbaum(85)

Diet and Hormone-dependent Cancers

been obtained with induced mammary tumors (27, 62) andspontaneous pituitary adenomas (69) in rats. A variety ofother mouse tumors, including hepatoma, primary lungadenoma, leukemia, skin tumors, and sarcoma are alsoinhibited by reducing caloric intake. It therefore appears tobe a general phenomenon, although some tumors seem to beaffected more than others by caloric restriction (86, 87, 92,94).

In many of these experiments, food intake was restrictedto between one-half and two-thirds of normal, but theeffects on tumorigenesis do not seem to be due to inanition,since the restricted animals were generally healthy andactive and had a longer life-span than unrestricted animals.Both underfeeding, in which the intake of all nutrients isreduced, and caloric restriction per se, in which only thecarbohydrate or carbohydrate and fat intake are reduced,have been shown to inhibit tumorigenesis (Table 1). Thegreatest inhibition seems to result when the caloric intake isreduced to less than 80% of the normal intake (84, 86).Restriction to 50% of normal intake is about the maximumthat can be tolerated for long-term experiments.

The experimental evidence suggests that the effects ofcaloric restriction are exerted not on the initiation of tumorsbut at some later stage of their development. Thus, Tannenbaum (80) found that spontaneous mammary carcinoma inmice were significantly inhibited by instituting caloricrestriction at varying times before the tumors began toappear, when the mice were 2, 5, or 9 months old.According to the results of Tannenbaum (82), experimentswith skin tumors induced by benzpyrene also showed thatrestriction during the period after treatment with carcinogenhad the greatest inhibitory effect, although this conclusionwas questioned by Clayson (17). Intermittent restriction,consisting of twice-weekly fasts of 24-hr duration, withfeeding ad libitum between fasts, had no inhibitory effect onformation of spontaneous mammary carcinoma in mice(89).

Most of the experimental work has been concerned witheffects of undernutrition, but Waxler et a!. (98) used micemade obese by treatment with gold thioglucose and foundthat the mice developed spontaneous mammary carcinomasat a faster rate than did controls. However, ifthe obese micewere restricted in diet so that they weighed only as much asthe controls, they developed fewer tumors than did thecontrols (97). In these experiments, there seemed to be a

correlation between body weight and tumorigenesis, andsome supporting evidence for this has been reported byTannenbaum and Silverstone (92). On the other hand, theseworkers found no correlation between body weight andincidence or time of appearance of spontaneous mammarycarcinomas in 250 mice fed ad libitum on adequate stockdiet (92). The maximum weights attained by the miceranged from 22 to 42 g. In the experiments of Moore andTittle (62), rats in which mammary tumor formation wasinhibited by caloric restriction or forced exercise weresmaller and had less body fat than controls.

The diversity of tumors affected by caloric restrictionsuggests some common mechanism of action, such asinhibition of tumorigenesis due simply to lack of availableenergy or metabolites required for the process. There is stillthe possibility, however, that different mechanisms areinvolved for different types of tumors. Mice and rats onrestricted rations show changes in ovaries, uteri, andmammary gland analogous to those seen in hypophysectomized animals; and inhibition of spontaneous mammarytumors might be a result of decreased estrogen productionand/or diminished responsiveness of mammary tissue toestrogen (49). Alternatively, Boutwell et a!. (7) suggestedthat adrenal hyperfunction in restricted animals might beresponsible for the inhibitory effect on tumorigenesis.Bullough (10) observed that mitotic activity is inhibited bycaloric restriction, and he concluded that cell division islimited by the availability of metabolites to provide energyfor mitosis. The developmental stage of carcinogenesiscould thus be influenced by an inhibitory effect on mitoticactivity, including that of latent cancer cells.

The inhibitory effect of caloric restriction on inducedtumors can be largely overcome by increasing the dose ofcarcinogen to the point at which nearly all animals in bothcontrol and restricted groups develop tumors (94). Theexperiments of White and White (100) and of King et a!.(52)providepossibleexamplesof this overridingeffectof astrong carcinogenic stimulus on mammary tumorigenesis.

Dietary Fat

Dietary fat intake has also been shown to influence tumorincidence, but in this case only certain types of tumors seemto be affected (15, 87, 92, 94). Numerous studies have shown

Table I

Inhibitory effect of caloric restriction on incidence of spontaneous mammary tumors in mice@z

a Adapted from Tannenbaum and Silverstone (94).a @,underfeeding; C + F, carbohydrate and fat restricted; C, carbohydrate restricted.

NOVEMBER 1975 3375

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Typeof fatLevel

in diet(% by wt)Carcinogenic agentDuration

of study(mos.)StrainTumor

incidence(%)Ref.Low

fat'HighfatMiceKremit512None18DBA16/50(32)32/50(64)Tannenbaum(8l)Kremax516None21C3H40/54(74)48/54(89)Silverstoneand

Tannenbaum(75)RatsCrisco546Stilbestrol13A

x C line99359/12

(75)I 1/12 (92)Dunninge:a/.(27)Lard16-302-Acetylaminofluorene7AES2/31

(6)55/71 (78)Engel and Copeland(30)Olive

oil20None10Sprague-Dawley3/25 (12)5/13 (39)Benson et al.(2)Cornoil20DM BA4Sprague-Dawley15/21 (7 1)21 /22 (96)Gammal et al. (38)

Level of dietary fatTumor

incidence(%)BeforeDMBAAfterDMBA20

0.5200.520

200.50.529/30

(97)28/30(93)22/30 (73)22/30 (73)

K. K. Carroll

Table 2

Enhancement of mammary tumor incidence in mice and rats by dietary fat

a The low-fat diets contained from 0.5 to 3% fat except in the experiments of Dunning et a!., where the level was 6.5%.b Partially hydrogenated cottonseed or cottonseed-soybean oil.

that high-fat diets increase the incidence of skin tumors andmammary tumors in mice and rats. Some of the resultsobtained with mammary tumors are shown in Table 2.There are also isolated reports that dietary fat enhancestheincidence of other tumors such as spontaneous hepatomas(76) and intracranial tumors (78). On the other hand,various other tumors, particularly sarcoma induced byinjection of carcinogenic hydrocarbons, appear to be unaffected by dietary fat ( I5).

In the experiments with skin tumors and mammarytumors, a number of different fats and oils were used;Tannenbaum (87) concluded that the enhancement of tumoryields was related to amount rather than type of dietary fat.Our experiments with mammary tumors induced in rats byDMBA3 showed that unsaturated fats generally gave highertumor yields than did saturated fats (14), but this was duemainly to an increase in number of tumors per tumor-bearing rat (Chart I).

High-fat diets tend to be high in calories as well, butpaired feeding experiments indicated that dietary fat has aneffect that is independent of caloric intake (85). Lavik andBaumann (54), working with skin tumors in mice, obtainedevidence that enhancement of tumorigenesis was related tothe triglyceride component of dietary fat, rather than thenonsaponifiable fraction. This does not seem to have beeninvestigated with mammary tumors, perhaps because of theeffort and expense involved in carrying out such experiments.

Dietary fat, like caloric intake, appears to affect primarily the developmental stage of tumorigenesis ( I 5). Ourstudies with mammary tumors indicated that dietary fat wasmore effective when fed after administration of the carcinogen (Table 3), and similar results were obtained in earlierexperiments with skin tumors (83). Dietary fat may thus beconsidered to exert a promoting effect on these 2 types oftumors ( 15). It is difficult to rule out completely the

PERCENTOF 4515WiTH TUMORS

TOTAL. NO OF TUMORSPER 30 RATS

COTTONSEED OIL ___________

OLIVE OIL Ill

CORN Oft. 110

SOYBEAN OIL 103

LARD 97

BUTTER 88

TALLOW 72

COCONUT OIL 7.

LOR FAT CONTROL 75

Relation of time offeeding dietary fat to incidence of tumors induced inrats by DMBA'

a Data from Carroll and Khor (13).

3 The abbreviation used is: DM BA, 7,l2-dimethylbenz(a)anthracene.

3376 CANCER RESEARCH VOL. 35

P40.OF TUMORSPER TUMOR-@EMP1GRAT

SUNFLORER @EDOIL 30

0 50 00 0 2 4 6

Chart I . Effect of different dietary fats on incidence of mammarytumors induced in female Sprague-Dawley rats by a single 5-mg dose ofDMBA p.o.Thefatswerefedas20%by weightof a semisyntheticdiet.The low-fat control diet contained 0.5% corn oil. Data are taken fromCarroll and Khor (14).

Table3

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Diet and Hormone-dependent Cancers

differential between animals on low-fat and high-fat diets.Treatment with the antiestrogen drug, U 1l,lOOA, alsoreduced the tumor incidence, but it failed to abolish thedifferential due to different levels of dietary fat. This latterresult tends to argue against the suggestion of Hill et a!. (46)that dietary fat affects breast cancer by altering estrogenlevels as a result of changes in gut bacteria.

Dietary Protein

Tannenbaum and Silverstone (88) found that varying thecasein content of semipurified rations from 9 to 45% had noparticulareffect on the formation ofspontaneous mammarycarcinoma in mice. Similar results were obtained with skintumors and sarcoma (87). On the other hand, White andAndervont (103) found that severe restriction to 4% caseinprevented formation of spontaneous mammary tumors inC3H mice. Feeding a diet containing 18% gliadin, which islow in lysine, reduced the tumor yield to 25% of normal(101). These effects are similar to the reductions in tumoryield that can be obtained by caloric restriction to givecomparable reductions in body weight. However, Tannenbaum and Silverstone (93) were able to demonstrate aneffect of dietary protein independent of body weight byregulating food intake to equalize body weights in mice onlow- and high-protein diets. Ross et a!. (69) reported thatthe incidence of pituitary adenomas was directly related tothe protein content of the diet when rats were fed restrictedbut equal amounts of the different rations.

Although restriction of dietary protein inhibits tumorigenesis, it is still possible to obtain a relatively high tumoryield under such circumstances. White and White (100)obtained a tumor incidence of 45% by s.c. implantation ofpellets containing diethylstilbestrol in mice on 4% caseindiet.

Dunning et a!. (28) investigated the effect of dietarytryptophan on the occurrence of diethylstilbestrol-inducedmammary cancer in rats. They used semisynthetic diets inwhich the protein requirement was supplied by either caseinor a tryptophan-free casein hydrolysate. Addition of I .4%tryptophan to the latter diet gave a higher tumor yield andaddition of 4.3% tryptophan a lower tumor yield than didthe control casein diet. An inhibitory effect on body growthmay explain the latter result.

Increasing the level of casein in the diet reduced theincidence of mammary carcinoma induced in rats by2-acetylaminofluorene, provided dietary intake was restricted to the amount eaten when the diet contained lessprotein (32). In these experiments the carcinogen was addedto the diet, and the increased food intake in unrestrictedanimals on high-protein diet therefore resulted in exposureto larger amounts of carcinogen, which apparently maskedthe inhibitory effect of casein. Gilbert et a!. (40) observedthat Wistar-type rats fed a diet containing 77% casein andless than 4% carbohydrate had a significantly lower mcidence of spontaneous tumors than did rats on 3 other dietscontaining from 12 to 15% protein. The rats on high-proteindiet gained less weight than the other 3 groups, and thereduced tumor yield may have been partly due to caloric

possibility that dietary fat exerts some effect on thecarcinogenic agent (39), but it seems more probable that thefat acts by providing a more favorable environment fordevelopment of latent tumor cells (15).

With hormone-dependent tumors, this leads one to thinkabout possible effects on the hormonal environment. Mammary tumors and the normal developrrent ofthe mammarygland are both influenced strongly by hormones, particularly estrogens and prolactn (23, 60. 61, 77). Dunning et a!.(27) observed a stimulatory effect of dietary fat on mammary glands of rats implanted with pellets containing stilbestrol. Rats on high-fat diet showed histological evidenceof more hypertrophy and secretory activity than did rats ondiets containing lower levels of fat. However, a stimulatedgland may actually be more refractory to mammary tumorinduction, since tumorigenesis can be inhibited by treatmentwith hormones (48) or by pregnancy ensuing soon aftertreatment with a carcinogenic hydrocarbon (22). On theother hand, pregnancy occurring at a later stage causes amarked stimulation of tumor growth (21). Age is also afactor. Female rats 50 to 60 days old are more prone todevelop tumors after treatment with a carcinogenic hydrocarbon than are either younger or older rats (20, 47).Recently, Nagasawa and Yanai (65) reported a correlationbetween mammary cell division, as measured by incorporation of [3Hjthymidine into the DNA of mammary gland,and age-related variation in sensitivity to tumor development.

Preliminary studies in our laboratory to investigatepossible effects of dietary fat on distribution and metabolism of injected [3Hjestradiol showed a tendency towardhigher tissue levels in rats on low-fat diet. This somewhatunexpected result could be interpreted to mean that a largerproportion of the tissue receptors are already occupied byendogenous hormone in rats on high-fat diet ( I5). Competition for tissue receptors between noncarcinogenic (estriol)and carcinogenic estrogens (estrone, estradiol) may play arole in breast cancer. Cole and MacMahon ( I8) proposedthat low ratios of estriol to estrone and estradiol in thedecade or so after puberty are associated with high risk ofbreast cancer, and studies of urine and estrogen profiles inAsian and North American women have provided evidencein support of this concept (57, 58).

Prolactin has also been implicated in mammary cancer(36, 66). Boyns et a!. (8) observed a positive correlationbetween the basal blood level of prolactin in 3 strains of ratsand their susceptibility to development of mammary tumorsafter treatment with a carcinogenic hydrocarbon. Pituitaryprolactin levels were found to be elevated in a small series offemale dogs bearing spontaneous mammary tumors (71),and there is some evidence that serum prolactin levels areelevated in women with breast cancer (4, 64); but Boyns eta!. found no such difference (9). High levels of prolactinmay inhibit induction of mammary tumors in the rat (37).

Recent studies of Chan and Cohen (16) suggest that theeffects of dietary fat on mammary cancer incidence may bemediated through prolactin. Treatment with the antiprolactin drug, CB-154, reduced the incidence of mammarytumors induced in rats by DMBA and abolished the

NOVEMBER 1975 3377

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

K. K. Carroll

restriction. However, a group of rats on restricted amountsof a ration containing 15% protein gained less weight butdeveloped more tumors than did those on the high-proteindiet. Tumors observed in this study included pheochromocytoma, fibroadenoma of the breast, and pituitaryadenoma.

Shay el al. (73) reported that rats on a semipurified dietcontaining casein at levels of 27 or 64% developed moremammary tumors than rats on stock diet, following treatment with 3-methylcholanthrene. When ovalbumin (64%)was used as source of protein, the rats did not grow well andthe tumor yield was lower than in rats on stock diet. A dietcontaining 64%lactalbumin had little or no effect on tumoryield.

In our experiments with mammary cancer induced in ratswith DMBA, it was found that fewer tumors developed inanimals on commerical feed compared to others fed semisynthetic diet containing approximately the same level ofdietary fat (Chart 2). The semisynthetic diet containedanimal protein (casein) as opposed to the commercial feed,in which much of the protein is derived from plant sources.Since mortality from breast cancer in humans shows astrong positive correlation with animal protein intake aswell as with dietary fat (15), preliminary experiments werecarried out to compare the effects on tumor yield of animaland plant protein fed in a semisynthetic diet. Rats given adiet in which the protein was supplied by soy protein isolate(Promine-R) grew more slowly than rats on a comparablediet with casein as source of protein, but they developed atleast as many tumors (Table 4). These experiments therefore did not provide any indication that the lower incidenceof tumors in rats on commercial feed is related to theprotein component of the diet.

Effects of Other Dietary Components

Some studies have been carried out to test effects of

0 0 20 30 40 50 60 70 80 90 00 110 20DAYSAFTERDMBAADMINISTRATION

Chart 2. Cumulative number of palpable mammary tumors induced ingroups of 30 female Sprague-Dawley rats by a single 5-mg dose of DMBA

p.o. given at 50 days of age. Both groups were on commercial feed until Iweek after the DMBA was given, at which time I group was transferred tothe semisynthetic diet containing 5% corn oil. The composition of this dietwas described previously (14).

different levels of vitamins and minerals in the diet ontumorigenesis in animals (87, 94). Varying the level of Bvitamins from amounts just adequate for growth to from 3to 9 times these amounts had little effect on incidence ofspontaneous carcinoma in mice (90). A lower incidence ofspontaneous carcinoma was reported by Morris (63) in micedeficient in riboflavin, but body weight and food consumption were also reduced and the effect may have been due toinanition.

Varying the level of a standard salt mix from 2 to 8% in asemisynthetic diet had no apparent effect on incidence ofspontaneous mammary tumors in mice (91). A specificinstance of tumorigenesis being influenced by deficiency ofan inorganic dietary constituent is provided, however, by thereport of Axelrad and Leblond ( I ). They observed thatprolonged iodine deficiency leads to production of thyroidtumors in the rat. There is also some evidence that iodinedeficiency may increase incidence of thyroid cancer inhumans (87, 94).

Rele@'anceof Animal Experiments to Human Cancer

Within the past 10 years, a number of authors have calledattention to the positive correlation between dietary fatintake and mortality from certain types of cancer indifferent countries of the world (12, 15, 24—26,46, 55,56, 105—107).The correlation is strongest with breast cancer, intestinal cancer, and cancer of the prostate, but it isnot seen with other types such as stomach cancer (15).

Correlations of this sort must be regarded with cautionbecause of the difficulties of collecting reliable epidemiological data and because associations between variables may becoincidental. However, when a correlation such as thatbetween fat intake and breast cancer mortality is supportedby experimental evidence from animal models, it seemsworthy of more serious consideration. Dietary fat has alsobeen reported recently to enhance the yield of induced colontumors in rats (67). It is thus of interest to consider thepossible relevance of results obtained with animal models tothe role of diet in human cancer.

In the animal model used for our experiments, theincidence of mammary tumors induced in rats by a single5-mg dose of DMBA was increased from about 70 to 95%.by raising the level of dietary fat from 0.5 to 20%by weight(14). Attempts to increase this differential in tumor mcidence by lowering the dose of carcinogen were unsuccessful(13). These results appear to be representative of thoseobtained with other animal models (Table 2), although agreater differential has been reported in some studies. Onthe other hand, the age-adjusted mortality from breastcancer in humans shows a 5- to 10-fold difference betweencountries with a per capita fat intake of 50 g/day or less andcountries with an intake of 140 to 150 g/day (Chart 3).

As pointed out earlier, our experiments suggest thatunsaturated fats have a greater effect on tumor yield than dosaturated fats (see Chart I ). This may not be true for otheranimal models, but systematic studies have not been carriedout. The epidemiological data for humans indicate that

00

90@ SEMISYNTHET1C DIET — 5% CORN OIL

.—. cOMMERCIAL FEED

80

70

60

50

40

30

20

10

3378 CANCER RESEARCH VOL. 35

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

AgewhenDMBAgivenDietaryBody

wtTumoryieldAt

DMBA(days)proteinInitialtreatmentFinalPalpableTotal―50Casein

Soyprotein―49 49157 142246 24412 15201960Casein

Soyprotein47 48189 165246 22620 27243475Casein

Soyprotein47 47207 190244 24414 1816 23

I

0

Diet and Hormone-depzndent Cancers

Table4Yields of mammary tumors induced by DMBA in rats on semisynthetic diets containing casein or

soy protein isolate

Groups of 10 female Sprague-Dawley rats were placed on the diets at weaning. The diets contained 5%corn oil and had the same basal composition as in previous experiments (14) except that a different vitaminmix (42) and salt mix (5) were used.

a Includes tumors found when the rats were autopsied 4 months after treatment with DMBA.

b Soy protein isolate (Promine-R) kindly supplied by Dr. E. W. Meyer of Central Soya, Chicago, Ill.

.u KGANLOAS •w@aZEA&NIC

SWITZERLKNOIREL@NO?• @s

•LLuSTRAUAS SWEDEN

&JSTRAS •SGERMAN@5ITALY NORWSY S FRKNCE

SCZECII

SPORT(@GA1

FEMALE25 -

20

‘5

S NETHERLANDS

CAN*.D@ •DENMARICSNEWZEALAND

SWITZERLANE@@ IRELANDBELGIUM SSUS A

SSWEDENSAUSTRALIA•@J@GERMANY

IA •NORWAYSITALY •FRANCE

S CZEO4OSLOVAXIASFINLAND

FEMALE

.•PORTUG*L HUNGARY

HONGKONG •POLARD8ULGML@S

j@MANLA

@,4 •PUERTORICOPHILIP@INES8 COLUM8IA

•@XICO@ TAIWAN

IAJLANO@S•EL,SALVAD0@@ I

20 40 60 80 100 20 40 60ANIMAL FAT INTAKE (g/day)

FEMALE •NETHERLANDSa 25@ B

§ .NEW ZEALAND CANADABELGIUM•@ . RELAND

@@208 AUSTRALIA

—. •SWEDEN

LU NORWAY• AUSTRIA••GERMANY@ •FRANCE.ITALY@ IS

I FINLAND••HUNGARY@ . . PORTUGAL

LU

@ 10 POLANG•HONGKONGSLUI-.@ GREECE

SSPAIN(I)

•PUERTORICO YUGOSLA/IA

4 5 PHILIPP1NES• 5 MEXICOLU AN•

TH4ILANOS@ SEL SALVADOR@ I0 ID 20 30 40 50 60 70

VEGETAL FAT INTAKE (g/day)

Chart 4. Correlation between per capita consumption of (A ) animal fatand (B) vegetal fat, and age-adjusted mortality from breast cancer indifferent countries. Data obtained from the same sources (34. 72) as thosein Chart 3.

cancer, prostatic cancer, and intestinal cancer in differentcountries can be explained on the basis of differences in fatintake alone. It is therefore of interest to consider possibleeffects of other dietary constituents on carcinogenesis.

Analysis ofdata from FAO Food Balance Sheets for 132different countries showed that total fat intake was strongly

I0 • HO5GKOr@G•p@Ar@C,-@tLE@

@@ZL8LA@S @NiAc@@NA@

•PUERTORICO5 . •COWM8IA

•PHIUPPINL5@ CO@ •TAJw@:

.csvL@'@ TP@LAt@

0@@@ - -@@@ -@@@ -- I -@@

0 20 40 60 80 00 20 40 ‘60 80TOTAL DIEThRY FLIT NT@KE (g /doy)

Chart 3. Correlation between per capita consumption of dietary fat(34) and age-adjusted mortality from breast cancer in different countries(72). The values for dietary fat are averages for 1964 to 1966 and those forcancer mortality are for 1964 to 1965, except in a few cases where datawereavailableonly for 1960to 1961or 1962to 1963.

animal fat intake does not correlate as well as total dietaryfat with breast cancer mortality, and intake of vegetal fatshows little, if any, correlation (Chart 4). Vegetal fat may beexpected to be, if anything, more unsaturated than animalfat, although some plant fats such as coconut oil arerelatively saturated and others may be partially hydrogenated for edible use. This classification, therefore,does not provide a reliable guide to possible differencesbetween effects of saturated and unsaturated fats on humanbreast cancer. However, death rates from breast cancer,prostatic cancer, and intestinal cancer are much higher inAmericans than in Japanese, although both have similardietary intakes of unsaturated fat in the form of linoleicacid. Analysis of adipose tissue also showed that Japanesehave a higher proportion of linoleate than do Americans(50).

Perhaps it is too much to expect quantitative as well asqualitative agreement between the results of animal experimentation and epidemiological data from human population studies. However, it appears somewhat doubtful thatthe 5- to 10-fold difference in death rates from breast

3379NOVEMBER 1975

25

@2O

LU

@ IS

@ toI..‘I)

a<5LUCD4

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

TotalfatAnimal fatAnimal proteinTotal caloriesTotal proteinVegetal fatCarbohydrateAnimal

fat+0.951(p <0.01)Animal

protein+0.931(p < 0.01)+0.931 (p <0.01)Total

calories+0.878(p < 0.01)+0.841 (p < 0.01)+0.827 (p <0.01)Total

protein+0.801(p < 0.01 )+0.777 (p < 0.01)+0.784 (p < 0.01)+0.867 (p <0.01)Vegetal

fat+0.402(p < 0.01)—0.099 (NS)b+0.224 (@,< 0.05)+0.335(p < 0.01)+0.286(p <0.01)Carbohydrate

Vegetal protein—0.078(NS)

—0.268(p < 0.01)—0.070

(NS)—0.300

(p < 0.01)—0.075

(NS)—0.360

(p < 0.01)+0.071

(NS)+0.018(NS)+0.182

(p < 0.05)+0.264

(p <0.01)—0.044

(NS)+0.069(NS)+0.413 (p < 0.01)

K. K. Carroll

Table 5

Correlation coefficients between dietary variables―

a Based on Food Balance Sheets for 132 countries (34).b NS, not significant.

and cancer of the colon (5 1). Natural diets fed to animalsnormally give a much higher residue than do purified diets,and tumor yields tend to be lower on the former type of diet(3 1, 94). Our own experience is in agreement with this(Chart 2).

In human diets, variations in one component are almostalways associated with variations in others, and this makesit difficult to assess the effect of any one dietary component.For example, low-residue diets frequently contain higherlevels of refined carbohydrate, and dietary fat is correlatedwith animal protein intake. This problem also extends toanimal studies. Thus, the fat content of the diet cannot beincreased without decreasing either the carbohydrate orprotein, and it is often difficult to be certain which change isresponsible for observed effects.

Environmental factors are considered to have a muchgreater influence on cancer incidence than do genetic traits(43), and diet may be one of the more important environmental variables. The large geographic differences in humancancer mortality give a measure of the potential for alteringcancer incidence if these environmental factors can beidentified and their mechanism of action determined. Because of the magnitude of the cancer problem in terms ofhuman suffering and economic loss (33), it seems verydesirable to explore such leads as rapidly as possible. Thiscan be done by collecting more and better information onrelationships between environmental variables and cancerincidence or mortality in human populations, by continuingto study effects of diet and other environmental variables oncancer incidence in animal models, and by comparingresults obtained in these models with the epidemiologicaldata on humans.

References

I . Axelrad, A. A., and Leblond, C. P. Induction of Thyroid Tumors inRats by a Low Iodine Diet. Cancer, 8: 339-367, 1955.

2. Benson, J., Lev, M., and Grand, C. G. Enhancement of MammaryFibroadenomasin the FemaleRat by a High Fat Diet. CancerRes.,

correlated with animal fat, animal protein, and total caloricintake and, to a lesser extent, with total protein and vegetalfat (Table 5). Since mortality from breast cancer shows astrong positive correlation with total fat intake, it is to beexpected that it will also show positive correlations withother dietary variables such as animal protein and totalcalories, and this is in fact observed (15).

In the animal experimentsdiscussedearlier, restriction ofcaloric intake had an inhibiting effect on tumorigenesis, butrather severe restriction was required in most instances toproduce a marked reduction in tumor yield, and the effectdid not appear to be specific for any particular type oftumor. It seems reasonable to expect that caloric restrictionmay also have an inhibitory effect on human carcinogenesisin countries where large segments of the population areundernourished, but the data from animal experimentswould lead one to expect a relatively unselective effect withregard to different cancer sites. Although caloric intake maybe a factor in human carcinogenesis, it does not appear tooffer a practical approach to the problem.

A positive correlation between animal protein intake andbreast cancer mortality has been noted by Drasar and Irving(26) as well as by ourselves (15), but the results of animalexperiments indicate that the protein content of the diet canbe varied over a wide range without much effect on tumoryield. Preliminary studies also failed to show differences intumor yield as a result of feeding different types of dietaryprotein (Table 4).

There is little evidencethat vitamins or minerals play asignificant role in tumorigenesis, and effects ofother normalingredients of the diet have not been investigated experimentally to any extent. There has been much speculationrecently about the possible beneficial effects ofdietary fiber(1 1, 29, 53). High-residue diets accelerate passage ofmaterial through the digestive tract and may reduce tumorincidence in humans. Drasar and Irving (26) were unable tofind any correlation between dietary fiber and cancers of thebreast and colon, but a more detailed subsequent analysisshowed a small negative correlation between cereal fiber

3380 CANCER RESEARCH VOL. 35

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Diet and Hormone-dependent Cancers

Dietary Fat and Carbohydrate on Diethylstilbestrol-induced Mammary Cancer in Rats. Cancer Res., 9: 354-361, 1949.

28. Dunning, W. F., Curtis, M. R., and Maun, M. E. The Effect ofAddedDietaryTryptophaneontheOccurrenceof Diethylstilbestrolinduced Mammary Cancer in Rats. Cancer Res., 10: 319- 323, 1950.

29. Eastwood, M. A. Dietary Fibre in Human Nutrition. J. Sci. FoodAgric.,25: 1523-1527,1974.

30. Engel, R. W., and Copeland, D. H. Influence of Diet on the RelativeIncidence of Eye, Mammary, Ear-Duct, and Liver Tumors in RatsFed 2-Acetylaminofluorene. Cancer Res., I I: 180 - I83, 195I.

31. Engel, R. W., and Copeland, D. H. Protective Action of Stock Dietsagainst the Cancer-Inducing Action of 2-Acetylaminofluorene inRats. Cancer Res., 12: 211-215, 1952.

32. Engel, R. W., and Copeland, D. H. The Influence of Dietary CaseinLevel on Tumor Induction by 2-Acetylaminofluorene. Cancer Res.,/2: 905-908, 1952.

33. Epstein, S. S. Environmental Determinants of Human Cancer.Cancer Res.,34:2425-2435,1974.

34. Food Balance Sheets, 1964-66 Average. Rome: Food and AgricultureOrganization of the United Nations, 1971.

35. Friedewald, W. F., and Rous, P. The Initiating and PromotingElements in Tumor Production. An Analysis of the Effects of Tar,Benzpyrene, and Methylcholanthrene on Rabbit Skin. J. Exptl.Med.,80:101-126,1944.

36. Furth, J. Prolactin and Carcinogenesis, in. A. R. Boyns and K.Griffiths (eds.), Fourth Tenovus Workshop, Prolactin and Carcinogenesis, pp. 137-142, Cardiff, Wales: Alpha Omega Alpha, 1972.

37. Gala, R. R., and Loginsky, S. J. Correlation between SerumProlactin Levels and Incidence of Mammary Tumors Induced by7,12-Dimethylbenz(a)anthracene in the Rat. J. NatI. Cancer Inst.,51:593 597,1973.

38. Gammal, E. B., Carroll, K. K., and Plunkett, E. R. Effects of DietaryFat on Mammary Carcinogenesis by 7,12-Dimethylbenz(a)anthracene in Rats. Cancer Res., 27: 1737- 1742, 1967.

39. Gammal, E. B., Carroll, K. K., and Plunkett, E. R. Effects of DietaryFat on the Uptake and Clearance of 7,12-Dimethylbenz(a)anthracene by Rat Mammary Tissue. Cancer Res., 28: 384 385, 1968.

40. Gilbert, C., Gillman, J., Loustalot, P., and Lutz, W. The ModifyingInfluence of Diet and the Physical Environment on SpontaneousTumour Frequency in Rats. Brit. J. Cancer, 12: 565-593, 1958.

41. Grasso,P. The Rangeof CarcinogenicSubstancesin Human Foodsand the Problems of Testing for Them. Proc. Roy. Soc. Med. 66:26—27,1973.

42. Greenfield, H., Briggs, G. M., Watson, R. H. J., and Yudkin, J. AnImproved Diet for Carbohydrate Preference Studies with Rats: SomeCriticisms of Experimental Diets. Proc. Nutr. Soc., 28: 43A-44A,1969.

43. Higginson,J. PresentTrends in Cancer Epidemiology.Can. CancerConf., 8:40-75, 1969.

44. Hill, M. J. Bacteria and the Etiology ofColonic Cancer. Cancer, 34:815-818,1974.

45. Hill, M. J., and Drasar, B. S. Bacteria and the Etiology of Cancer ofthe Large Intestine. In: A. Balows, R. M. DeHaan, V. R. Dowell, Jr.,and L. B. Guze (eds.), Anaerobic Bacteria: Role in Disease, pp.I 19-133. Springfield, Ill.: Charles C Thomas, Publisher, 1974.

46. Hill, M. J., Goddard, P., and Williams, R. E. 0. Gut Bacteria andAetiology of Cancer of the Breast. Lancet, 2: 472-473, 1971.

47. Huggins, C., Grand, L. C., and Brillantes, F. P. Mammary CancerInduced by a Single Feeding of Polynuclear Hydrocarbons, and ItsSuppression. Nature, 189: 204-207, 1961.

48. Huggins,C., Moon, R. C., and Morii, S. Extinction of ExperimentalMammary Cancer. I. Estradiol-17$ and Progesterone.Proc. NatI.Aced.Sci.U.S.,48:379-386,1962.

49. Huseby, R. A., Ball, Z. B., and Visscher, M. B. Further Observationson the Influence of Simple Caloric Restriction on Mammary Cancer

16: 135-137, 1956.

3. Berenblum, I., and Shubik, P. A New, Quantitative, Approach to theStudy ofthe StagesofChemical Carcinogenesisin the Mouse'sSkin.Brit. J. Cancer, I: 383-391, 1947.

4. Berle, P., and Voigt, K. D. Evidence of Plasma Prolactin Levels inPatients with Breast Cancer. Am. J. Obstet. Gynecol., 114.1101-1102,1972.

5. Bernhart, F. W., and Tomarelli, R. M. A Salt Mixture Supplying theNational Research Council Estimates of the Mineral Requirementsof the Rat. J. Nutr., 89: 495-500, 1966.

6. Boutwell, R. K. Some Biological AspectsofSkin Carcinogenesis. In:F. Homburger(ed), Progr.Exptl. Tumor Res.,4: 207-250,1964.

7. Boutwell, R. K., Brush, M. K., and Rusch, H. P. Some PhysiologicalEffects Associated with Chronic Caloric Restriction. Am. J. Physiol.,154:517-524,1948.

8. Boyns, A. R., Buchan, R., Cole, E. N., Forrest, A. P. M., andGriffiths, K. Basal Prolactin Blood Levels in Three Strains of Ratwith Differing Incidence of 7,12-Dimethylbenz(a)anthracene Induced Mammary Tumours. EuropeanJ. Cancer,9: 169-Ill, 1973.

9. Boyns, A. R., Cole, E. N., Grifliths, K., Roberts, M. M., Buchan, R.,Wilson, R. G., and Forrest, A. P. M. Plasma Prolactin in BreastCancer. European J. Cancer, 9: 99-102, 1973.

10. Bullough, W. S. Mitotic Activity and Carcinogenesis. Brit. J. Cancer,4:329-336,1950.

I I . Burkitt, D. P., Walker, A. R. P., and Painter, N. S. Dietary Fiberand Disease. J. Am. Med. Assoc., 229: 1068-1074, 1974.

12. Carroll, K. K., Gammal, E. B., and Plunkett, E. R. Dietary Fat andMammary Cancer.Can. Med. Assoc.J., 98: 590-594, 1968.

13. Carroll, K. K., and Khor, H. T. Effects of Dietary Fat and DoseLevel of 7,12-Dimethylbenz(a)anthraceneon Mammary TumorIncidence in Rats. Cancer Res., 30. 2260-2264, 1970.

14. Carroll, K. K., and Khor, H. 1. Effects of Level and Type of DietaryFat on Incidence of Mammary Tumors Induced in FemaleSprague-Dawley Rats by 7,12-Dimethylbenz(a)anthracene. Lipids, 6:415—420,1971.

15. Carroll, K. K., and Khor, I-I. T. Dietary Fat in Relation toTumorigenesis. Progr. Biochem. Pharmacol., 10: 308-353, 1975.

16. Chan, P., and Cohen, L. A. Effect of Dietary Fat, Antiestrogen, andAntiprolactin on the Development of Mammary Tumors in Rats. J.Nail.Cancer Inst.,52:25-30,1974.

17. Clayson, D. B. Chemical Carcinogenesis. London: J & A Churchill.1962.

18. Cole, P., and MacMahon, B. Oestrogen Fractions during EarlyReproductiveLife in the Aetiology of BreastCancer.Lancet, 1.604—606, 1969.

19.Conney,A. H. Editorial, CarcinogenMetabolismand HumanCancer. New Engl. J. Med., 289: 971-973, 1973.

20. Dao, T. L. Mammary Cancer Induction by 7,12-Dimethylbenz(a)anthracene: Relation to Age. Science, /65: 810-81 I, 1969.

21. Dao, T. L. Studies on Mechanism ofCarcinogenesis in the MammaryGland. Progr. Exptl. Tumor Res., ii: 235-261, 1969.

22. Dao, T. L. Inhibition of Tumor Induction in Chemica Carcinogene515 in the Mammary Gland. Progr. Exptl. Tumor Res., 14: 59—88,

1971.23. Dao, T. L., and Sinha, D. Oestrogen and Prolactin in Mammary

Carcinogenesis: In Vivo and In Vitro Studies. In: A. R. Boyns and K.Griffiths (eds.), Fourth Tenovus Workshop, Prolactin and Carcinogenesis, pp. 189-194. Cardiff, Wales: Alpha Omega Alpha, 1972.

24. De Waard, F. The Epidemiology of Breast Cancer: Review andProspects.Intern. J. Cancer,4: 577-586, 1969.

25. Drasar, B. S., and Hill, M. J. Intestinal Bacteria and Cancer. Am. J.Clin. Nutr., 25: 1399-1404, 1972.

26. Drasar, B. S., and Irving, D. Environmental Factors and Cancer ofthe Colon and Breast. Brit. J. Cancer, 27: 167- 172, 1973.

27. Dunning, W. F., Curtis, M. R., and Maun, M. E. The Effect of

NOVEMBER 1975 3381

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

K. K. Carroll

Incidence and Related Phenomena in C3H Mice. Cancer Res., 5:40-46, 1945.

50. Insull, W., Jr., Land, P. D., Hsi, B. P., and Yoshimura, S. Studies ofArteriosclerosis in Japanese and American Men. I. Comparison ofFatty Acid Composition of Adipose Tissue. J. Clin. Invest., 48:1313-1327,1969.

SI. Irving, D., and Drasar, B. S. Fibre and Cancerofthe Colon. Brit. J.Cancer,28:462-463,1973.

52. King, J. T., Casas, C. B., and Visscher, M. B. The Influence ofEstrogen on Cancer Incidence and Adrenal Changes in Ovariectomized Mice on Calorie Restriction. Cancer Res., 9: 436—437,1949.

53. Kritchevsky, D., Tepper, S. A., and Story, J. A. Nonnutritive Fiberand Lipid Metabolism. J. Food Sci., 40: 8-11, 1975.

54. Lavik, P. S., and Baumann, C. A. Dietary Fat and TumorFormation. Cancer Res., 1: 181-187, 1941.

55. Lea, A. J. Dietary Factors Associated with Death-Rates from CertainNeoplasms in Man. Lancet, 2: 332-333, 1966.

56. Lea, A. J. Neoplasms and Environmental Factors. Ann. Roy. CoIl.Surg. EngI., 41: 432-438, 1967.

57. MacMahon, B., Cole, P., Brown, J. B., Aoki, K., Lin, T. M.,Morgan, R. W., and Woo, N-C. Oestrogen Profiles of Asian andNorth American Women. Lancet, 2: 900-902, 1971.

58. MacMahon, B., Cole, P., Brown, J. B., Aoki, K., Lin, T. M.,Morgan, R. W., and Woo, N-C. Urine Oestrogen Profiles of Asianand North American Women. Intern. J. Cancer, 14: 161- 167, 1974.

59. Masuda, Y., and Kuratsune, M. Polycyclic Aromatic Hydrocarbonsin Smoked Fish “Katsuobushi.―Gann, 62: 27-30, 1971.

60. Meites, J. Relation of Prolactin and Estrogen to Mammary Tumorigenesis in the Rat. J. NatI. Cancer Inst., 48: 1217-1224, 1972.

61. Meites, J. Relation of Prolactin to Mammary TumourigenesisandGrowth in Rats. In: A. R. Boyns and K. Griffiths (eds.), FourthTenovus Workshop, Prolactin and Carcinogenesis, pp. 54-63. Cardiff, Wales: Alpha Omega Alpha, 1972.

62. Moore, C., and Tittle, P. W. Muscle Activity, Body Fat, and InducedRat Mammary Tumors. Surgery, 73: 329-332, 1973.

63. Morris, H. P. Effects on the Genesis and Growth of TumorsAssociated with Vitamin Intake. Ann. N. Y. Acad. Sci., 49: 119- 140,1947.

64. Murray, R. M. L., Mozaffarian, G., and Pearson, 0. H. ProlactinLevels with i-Dopa Treatment in Metastatic Breast Carcinoma. In:A. R. Boyns and K. Grifliths (eds.), Fourth Tenovus Workshop,Prolactin and Carcinogenesis, pp. 158-161. Cardiff, Wales: AlphaOmega Alpha, 1972.

65. Nagasawa, H., and Yanai, R. Frequency of Mammary Cell Divisionin Relation to Age: Its Significance in the Induction of Mammary

Tumors by Carcinogen in Rats. J. NatI. Cancer Inst., 52: 609@6I0,1974.

66. Pearson, 0. H., Murray, R., Mozaffarian, G., and Pensky, J.Prolactin and Experimental Breast Cancer. in: A. R. Boyns and K.Griffiths (eds.), Fourth Tenovus Workshop, Prolactin and Carcinogenesis, pp. I 54 I 57. Cardiff, Wales: Alpha Omega Alpha, 1972.

67. Reddy, B. S., Weisburger, J. H., and Wynder, E. R. Effects ofDietary Fat Level and Dimethylhydrazine on Fecal Acid and NeutralSterol Excretion and Colon Carcingenesis in Rats. J. Natl. CancerInst.,52:507-5I1, 1974.

68. Roe, F. J. C. Carcinogens and Food: A Broad Assessment. Proc.Roy. Soc. Med., 66: 23-26, 1973.

69. Ross, M. H., Bras, G., and Ragbeer, M. S. Influence of Protein andCaloric Intake upon Spontaneous Tumor Incidence of the AnteriorPituitary Gland ofthe Rat. J. Nutr., 1@. 177-189, 1970.

70. Salaman, M. H., and Roe, F. J. C. Cocarcinogenesis. Brit. Med.Bull., 20: 139-144, 1964.

71. Saluja, P. G., Hamilton, J. M., Gronow, M., and Misdorp, W.Pituitary Prolactin Levels in Canine Mammary Cancer. European J.

Cancer, 10: 63-66, 1974.

72. Segi, M., Kurihara, M., and Matsuyama, T. Cancer Mortality forSelected Sites in 24 Countries. No. 5 (1964-1965). Sendai, Japan:Department of Public Health, Tohoku University School of Medicine, 1969.

73. Shay, H., Gruenstein, M., and Shimkin, M. B. Effect of Casein,Lactabumin. and Ovalbumin on 3-Methylcholanthrene-InducedMammary Carcinoma in Rats. J. Natl. Cancer Inst., 33: 243-253,1964.

74. Shils, M. E Nutrition and Neoplasia. in: R. S. Goodhart and M. E.Shils (eds.), Modern Nutrition in Health and Disease. Dietotherapy,Ed. 5, pp. 981-996. Philadelphia: Lea & Febiger, 1973.

75. Silverstone, H., and Tannenbaum, A. The Effect ofthe Proportion ofDietary Fat on the Rate of Formation of Mammary Carcinoma inMice. Cancer Res., 10: 448-453, 1950.

76. Silverstone, H., and Tannenbaum, A. The Influence of Dietary Fatand Riboflavin on the Formation of SpontaneousHepatomasin theMouse. Cancer Res., 11: 200203, 1951.

77. Sinha, D., Cooper, D., and Dao, T. L. The Nature of Estrogen andProlactin Effect on Mammary Tumorigenesis. Cancer Res., 33:411-414,1973.

78. Szepsenwol, J. Intracranial Tumors in Mice ofTwo Different StrainsMaintained on Fat Enriched Diets. European J. Cancer. 7: 529-532.1971.

79. Tannenbaum, A. The Initiation and Growth of Tumors. Introduction.I.Effectsof Underfeeding.Am. J.Cancer,38:335-350,1940.

80. Tannenbaum, A. The Genesis and Growth of Tumors. I I . Effects ofCaloric Restriction per se. Cancer Res., 2: 460-467, 1942.

81. Tannenbaum, A. The Genesis and Growth ofTumors. III. Effects ofa High-Fat Diet. Cancer Res., 2: 468-475, 1942.

82. Tannenbaum, A. The Dependence of the Genesis of Induced SkinTumors on the Caloric Intake during Different Stages of Carcinogenesis. Cancer Res., 4: 673-677, 1944.

83. Tannenbaum, A. The Dependence of the Genesis of Induced SkinTumors on the Fat Content of the Diet during Different Stages ofCarcinogenesis. Cancer Res., 4: 683-687, 1944.

84. Tannenbaum, A. The Dependence of Tumor Formation on theDegree of Caloric Restriction. Cancer Res., 5: 609-615, 1945.

85. Tannenbaum, A. The Dependence of Tumor Formation on the

Composition of the Calorie-Restricted Diet as Well as on the Degreeof Restriction. Cancer Res., 5: 616 625, 1945.

86. Tannenbaum, A. The Role of Nutrition in the Origin and Growth ofTumors. In: F. R. Moulton (ed), Approaches to Tumor Chemotherapy, pp. 96-127. Washington: American Association for the Advancement of Science, 1947.

87. Tannenbaum, A. Nutrition and Cancer. In: F. Homburger (ed), ThePhysiopathology ofCancer, Ed. 2, pp. 517 562. New York: HoeberHarper, 1959.

88. Tannenbaum, A., and Silverstone, H. The Genesis and Growth ofTumors. IV. Effects of Varying the Proportion of Protein (Casein) inthe Diet. Cancer Res., 9. 162-173, 1949.

89. Tannenbaum, A., and Silverstone, H. Failure to Inhibit the Formation of Mammary Carcinoma in Mice by Intermittent Fasting.

Cancer Res., 10: 577-579, 1950.90. Tannenbaum, A., and Silverstone, H. The Genesis and Growth of

Tumors. V. Effects of Varying the Level of B Vitamins in the Diet.Cancer Res., 12: 744-749, 1952.

91. Tannenbaum, A., and Silverstone, H. The Genesis and Growth ofTumors. VI. Effects of Varying the Level of Minerals in the Diet.Cancer Res., 13: 460-463, 1953.

92. Tannenbaum, A., and Silverstone, H. Nutrition in Relation toCancer. Advan. Cancer Res., 1. 451 501, 1953.

93. Tannenbaum, A., and Silverstone, H. Mammary Carcinoma in theMouse. Proc. Am. Assoc. Cancer Res., 1: 56, 1953.

94. Tannenbaum, A., and Silverstone, H. Nutrition and the Genesis ofTumours. In: R. W. Raven (ed), Cancer, Vol. I, pp. 306-334,

3382 CANCER RESEARCH VOL. 35

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Diet and Hormone-dependent Cancers

London: Butterworth & Co., Ltd., 1957.95. Visscher, M. B., Ball, Z. B., Barnes, R. H., and Sivertsen, I. The

Influenceof CaloricRestrictionuponthe Incidenceof SpontaneousMammary Carcinoma in Mice. Surgery, 11: 48-55, 1942.

96. Wattenberg, L. W. Dietary Modification of Intestinal and PulmonaryAryl Hydrocarbon Hydroxylase Activity. Toxicol. AppI. Pharmacol.,23:741-748,1972.

97. WaxIer, S. H. The Effect of Weight Reduction on the Occurrence ofSpontaneousMammary Tumors in Mice. J. NatI. Cancer Inst., /4.1253-1256,1954.

98. WaxIer, S. H., Tabar, P., and Melcher, L. R. ObesityandtheTime ofAppearanceof SpontaneousMammary Carcinoma in C3H Mice.Cancer Res., 13: 276-278, 1953.

99. White, F. R. The Relationship between Underfeeding and TumorFormation, Transplantation, and Growth in Rats and Mice. CancerRes.,2/:28l-290, 1961.

100. White, F. R., and White, J. Effect of Diethylstibestrol on MammaryTumor Formation in Strain C3H Mice Fed a Low Cystine Diet. J.Nail. Cancer Inst., 4: 413-415, 1944.

101. White, F. R., and White, J. Effect of a Low Lysine Diet on

Mammary-Tumor Formation in Strain C3H Mice. J. NatI. CancerInst., 5:41-42, 1944.

102. White, F. R., White, J., Mider, G. B., Kelly, M. G., Heston, W. E.,and David, P. W. Effect ofCaloric Restriction on Mammary-TumorFormation in Strain C3H Mice and on the Responseof Strain DBAto Painting with Methylcholanthrene. J. NatI. Cancer Inst., 5: 43-48,1944.

103. White, J., and Andervont, H. B. Effect of a Diet Relatively Low inCystine on the Production ofSpontaneous Mammary-Gland Tumorsin Strain C3H Female Mice. J. Natl. Cancer Inst., 3: 449-451, 1943.

104. Wogan, (3. N. Naturally Occurring Carcinogens in Foods. In: F.Homburger (ed.), Progr. Exptl. Tumor Res., 1/: 134-162, 1969.

105. Wynder, E. 1., Hyams, L., and Shigematsu, T. Correlations ofInternational Cancer Death Rates. An Epidemiological Exercise.Cancer,20:113-126,1967.

106. Wynder, E. L., and Mabuchi, K. Etiological and Preventive Aspectsof Human Cancer. Prey. Med., 1: 300-334, 1972.

107. Wynder, E. L., and Reddy, B. Guest Editorial. Studies of LargeBowelCancer: Human Leadsto Experimental Application. J. NatI.Cancer Inst., 50: 1099—1106,1973.

NOVEMBER 1975 3383

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1975;35:3374-3383. Cancer Res   Kenneth K. Carroll  Hormone-dependent CancersExperimental Evidence of Dietary Factors and

  Updated version

  http://cancerres.aacrjournals.org/content/35/11_Part_2/3374

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

  .pubs@aacr.orgDepartment 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/35/11_Part_2/3374To request permission to re-use all or part of this article, use this link

on July 18, 2020. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from