Dietary $-Carotene and Cancer of the Prostate: A Case...

ICANCER RESEARCH 48, 1331-1336, March 1, 1988)

whether only fat, or protein, or both factors, either independently or in interaction, contribute to the risk for prostate cancer,though fat appears to be more important than protein. And itcould not be ascertained whether vitamin A and vitamin C arerelated to prostate cancer risk in humans (7). Among theepidemiological data available today, the effect of diets rich invitamin A on prostate cancer risk is in controversy. In a notablylow incidence area, Japan, green/yellow vegetables were negatively associated with the risk (8, 9). In a high incidence area,the United States, however, diets characterized by high intakeofvitamin A enhanced the risk (1012).

To explore this conflicting dietary issue in the etiology ofprostate cancer, we here present a dietary part of results ohtamed from a Japanese-Dutch case-control study. To ourknowledge, this is the first study in Japan in which dietarynutrients were investigated in relation to prostate cancer risk.

MATERIALS AND METhODS

We conductedthis study from January1981 to December 1984 intwo University hospitals and 11 affiliated hospitals in Kyoto and sixnearby cities. Study subjectswere cases and controls defined as follows.

Cases were 5079-year-oldpatients newly diagnosed as having cliiiical prostate cancer (TNM 1974: T1T4)ordiffuselatent prostate cancer(T0, pT2-pT4). Clinical prostate cancer was histologically confirmed bybiopsy. Diffuse latent cancer was pathologically diagnosed, when theprostate gland, taken bysimple prostatectomy or transurethral resectionunderthe diagnosisof BPH,3showedcancerousregions in more than5% of the examined areas (13). When cases had other malignancies,liver disease, or hormonal disorders, they were excluded. Controls aretwo series: BPH patients and general hospital patients. We denotehereafter the former as BPH controls and the latter as hospital controls.BPH controls are 5079-year-oldpatients whose operative prostaticspecimen showed BPH only and who were not affected by othermalignancies, liver disease, and hormonal disorders. Hospital controlsare 50-79-year-old patients, confirmed as having neither prostate cancer nor BPH by digital examination.Those with other malignancies,liver disease, hormonal disorders, other prostatic disease, or lowerurinary tract symptoms were excluded.

Studysubjectswereconstructedas one hundredtripletsof 1:1:1andmatched to each other on hospital admittance, age (3years) and dateofadmission (3months). Routine demographic, epidemiological, anddietary information was obtained during admission by direct interviewofthe study subject himself, using a standardized questionnaire, by oneurologist (K. 0.) and two well-trained nutritionists. Information onsexuallife was,however,collectedbya self-administeredquestionnaire,and blood samples were drawnto determinethe serum levels of sexhormones and vitamins.

Dietary information, the only part that is reported in this paper, wasobtained by a quantitative food frequency technique for estimatingusual diet. The technique is similar to that used by Hankin et a!. (14),though modified and made applicable to Japanese diets. At the beginningof the dietaryinterview,a nutritionistaskeda subjectto recallallpossible patterns of breakfast, lunch, and dinner taken usually 5 yearsbefore the interview. He was then asked about the weeklyfrequency of

3 The abbreviations used are: BPH, benign prostatic hyperplasis, RR, relative

risk; Cl, confidence interval.

ABSTRACT

One hundred patients with prostate cancer and two different controlseries 1100 benign prostatic hyperplasis (BPH) patients and 100 generalbospital patientsj were matched to each other upon hospital admittance,age (3years) and date of admission (3months), and directly interviewed during admission from 1981 to 1984 in Kyoto, Japan. Majordietary findings derived from a quantitative food frequency technique forestimating usual diet are as follows.(a) The smaller the dietary Intakeof fl-caroteneand vitaminA as well, the higherthe risk, with a highlysignificant linear trend. From the fl-carotene analyses, the relative risk(95% confidence Interval) for the lowest Intake quartile relative to thehighest was 2.10 (O.98-4.47)for the uncorrectedIntake, 2.35(1.085.12)for the Intake per kg, and 2.94 (134-6.44) for the intake per kcal in thecomparison with BPH patients 2.88 (131-632), 2.56 (1.14-5.76), and3.50 (1324.06), respectIvely, in the comparison with hospital controls.The correspondingrelativerisk obtainedfrom the vitaminA analyseswas 2.82 (130-6.14), 2.64 (1.245.60),and 3.29 (1.477.35)Indue orderIn the comparison with BPH patients 2.69 (1.22-5.94), 4.78 (1.98-11.52), and 330 (132-8.06) In the comparison with hospital controls.(b) fl-CaroteneaswellasvitaminA containedingreen/yellowvegetablesweresignificantlyprotective,and those Inseaweedsand kelp suggestivelyprotective.But those in fruits appeared to enhance the risk. (c) The riskreductionby dietary fl-caroteneand vitaminA wassignificantin the oldermen (7079years), but not in the younger men (50-69 years). (d) Totalenergy Intake and the dietary Intake of fat, protein, carbohydrate, water,fiber, ash, such vitamins as retinal, B3, B@,C, and niacin, and suchminerals as calcium, potassium, sodium, phosphorus, and iron were notlinked with prostate cancer risk. (e) A protective effect of dietary ficarotene and vitamin A against prostate cancer could be related to thelowoverall fat intake In Japan.

INTRODUCTION

Descriptive epidemiology of prostate cancer indicates a wideinternational and ethnic variation in incidence and mortality(1, 2), its changing rate as humans move from low risk to highrisk areas (3), and its varying rate over years within a population(3). These descriptive features cannot be explained solely bygenetic susceptibility supported by familial aggregation (4), butcan be explained by postulating an etiological role for nongenetic factors (5).

Nongenetic factors of possible importance in the etiology ofprostate cancer currently include diet, endocrine function, someaspects of sexual behavior, and occupational exposure to cadmium (6). According to Bosland (7), who evaluated the strengthand consistency of the epidemiological data for each food itemor nutrient in relation to prostate cancer risk, it was not clear

Received5/1/87; revised 11/1 1/87; accepted12/3/87.The costsof publicationof this article weredefrayedin part by the payment

of pagecharges.Thisarticlemustthereforebe herebymarkedadvertisementinaccordancewith 18U.S.C.Section1734solelyto indicatethis fact.

I Supported by a Grant-in-Aid (60010013) for Cancer Research from the

Ministry of Education, Scienceand Culture, Japan, and grants from the ToyotaFoundation; the Dutch Cancer Foundation, the Netherlands, and the Internstional Agency for Researchon Cancer, Lyon, France. Presentedat the FifthSymposium on Epidemiology and Cancer Registries in the Pacific Basin, Novemher 1621,1986, Kauai, HI.

2Towhomrequestsforreprintsshouldbeaddressed.1331

Dietary $-Carotene and Cancer of the Prostate: A Case-Control Study

in Kyoto, Japan'

Yoshiyuki Ohno,2 Osamu Yoshida, Kenji Oishi, Kenichiro Okada, Hirohiko Yamabe, and Fritz H. Schroeder

TheDepartmentofPublicHealth,NagoyaCityUniversityMedicalSchool,Kawazumi,Mizuho.cho,Mizuho-lcu,Nagoya,467,JapanfY.0.);theDepartmentofUrology,Kyoto University Faculty ofMedicine, Kawahara-cho, Shogoin, Sakyo-ku, Kyoto, 606, Japan (0. Y., K. 0., K. 0.1; the Laboratory ofAnatomic Pathology, KyotoUniversity Hospital, Kawahara-cho, Shogoin, Sakyo-ku, Kyoto, 606, Japan (H. Y.J;and the Department of Urology, Erasmus University, Ronerdam, the Netherlands(F. H. S.]

on April 30, 2018. 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Table 2 Meanvaluesofdaily mtakeofdietarynutrientsBPHHospitalCasescontrolscontrolsNutrient

(unit)(n = 100)(n = 98)(n =99)Energy(kcal)205119812059Protein(g)706870Fat(g)464744Carbohydrate(g)291280289Water(g@l324@12051274Fiber(g)3.83.63.6Ash(g)16.416.215.5Retinol(pg)228310257/1-Carotene

(pg)1509C18891950VitaminA(mU)'160920961950VitaminB(mg)0.880.860.82VitaminB2(mg)1.101.091.10Vitamin

C(mg)1039692Niacin(mg)13.412.613.0Calcium(mg)445418420Sodium(mg)355636043325Potassium(mg)238522802323Phosphorus

(mg)922880900Iron(mg)9.69.59.3

Table 3 Relative risksand 95% confidenceintervalsforlowerdaily intakeofdietaryfatControls@Cases(BPH/hospi

tel)CrudeRR5Age-adjustedRRCNutrient

(g)

Table 4 Relativerisksand 95% confidenceintervalsofdaily intakeofdietary @-caro:enebyquartileCases

vs.BPH controls Casesvs.hospitalcontrolsa-Carotene

Uncorrected Per kg Per kcal Uncorrected Per kgPerkcal

FoodsCases

vs.BPH controlsCases vs.hospitalcontrolsUncorrectedPer

kgPer kcalUncorrectedPer kgPerkcalEggs,

Milks

White vegetables0.97(0.551.71)

1.10(0.631.94)0.99

(0.571.75)1.07

(0.611.88)0.93

(0.531.64)1.26

(0.722.20)0.80

(0.451.40)0.67

(0.381.17)0.85

(0.481.49)0.71

(0.411.25)0.80

(0.461.40)0.73

(0.421.28)Green/yellowvegetables

Fruits2.65(1.484.75)

1.26(0.722.20)2.26

(1.264.04)1.07

(0.611.88)2.97

(1.645.39)1.37

(0.782.39)2.20

(1.233.92)0.50

(0.280.89)2.24

(1.264.00)0.57

(0.321.01)2.41

(1.354.32)0.52

(0.300.92)Seaweeds,kelp

Others1.84(1.053.23)

1.20(0.682.12)1.46

(0.832.57)1.20

(0.692.10)1.50

(0.852.63)1.20

(0.692.11)1.19

(0.682.09)1.14

(0.652.01)1.07

(0.611.88)0.95

(0.541.66)1.04

(0.601.83)1.07

(0.611.88)

DIETARY si-CAROTENE AND PROSTATE CANCER

Table6 Age-adjustedrelativerisksand 95% CI for lowerdaily intakeoff3-carotenecontainedin specifiedfoodsSeeTable 3,footnote c, and Table 5, legend.

this type of prostate cancer is clinicopathologically equal to oreven more malignant than clinical prostate cancer (13). Nevertheless, we reanalyzed the data after exclusion of these ninepatients and the corresponding controls. The results obtainedwere almost exactly like those presented here. The findings inour study, therefore, could be regarded as those for clinicalprostate cancer, and might indicate the dietary effects probablyspecific to the late stage promotion or progression of prostatecancer. Cases in our study were relatively small in number. Thisis simply due to the fact that prostate cancer occurs in a verylow incidence in Japan: approximately one-twentieth of theincidence in the United States (1). Controls in our study weretwo series. The rationale for the use of both BPH controls andhospital controls is as follows. In general, population controlsrandomly selected would be more preferred, since they arepresumed to represent the nondiseased population. For a casecontrol study of prostate cancer in particular, however, thismay not be true, because population controls, if sampled frommen aged 70 years or older, may include those with BPH inabout 70%, and those with BPH may have latent prostate cancerin about 20% (19), and because prostate cancer patients coincidentally suffer from BPH in about 70% (20). Because of thisbiological background we decided not to use population controls with the diseased prostate, but to use the two series ofcontrols strictly defined.

Our dietary survey method was a quantitative food frequencytechnique for estimating usual diet. It was, however, technicallydifferent from a food frequency method usually used by otherinvestigators (1012),because we paid attention not to someselected food items, but to all food items in all meals usuallytaken during a week. This approach which is assisted by mealphotographs is adopted, because it is more likely to disclose avariety of dietary patterns among Japanese, and food itemsactually consumed could be detected with much more reality ascompared to assessing some selected food items. It was presumed by our dietary method that we could usefully distinguishcases and controls, at least, in very broad categories along acontinuum of amount of ingestion, e.g., high or low, or fromhigh to moderate to low to very low.

Our mean values a day for total energy, protein, fat, carbohydrate, vitamin A, vitamin B1, vitamin B2, and vitamin C were19812059kcal, 6870g, 4447g, 280291g, 16092096IU,0.820.86 mg, 1.091.10mg, and 92103mg, respectively(Table 2). The corresponding figures from National NutritionSurvey were 2134 kcal, 81 g, 59.5 g, 300 g, 2226 IU, 1.26 mg,1.29 mg, and 133 mg in the Kansai area (Kyoto, Osaka, andKobe) in 1983 (21). Although in National Nutrition Survey thenutrients were derived from the quantities of all food itemsactually consumed in 3 days, the mean value is surprisingly

close to ours in total energy and carbohydrate, and differencesin the mean value for the other nutrients are in a magnitude ofapproximately 20%. The smaller mean values in our studymight possibly result from the differences in the dietary surveymethod and in the age distribution of the study subjects; National Nutrition Survey included those aged less than 50 years.As a consequence of this comparison, we could have greaterconfidence in the usefulness of our dietary survey method.

The cases ingested more water than BPH controls. Perhapsmost likely is the possibility that BPH patients ingested lesswater in order to avoid such uncomfortable symptoms as hesitancy and straining to initiate micturation, norturia, and feelingof urinary retention. The mean water intake was not differentbetween cases and hospital controls, although 29% of hospitalcontrols were the patients with renal calculi. We found nodifferences in the daily mean intake ofprotein and carbohydrateas well as fat when cases were compared to either BPH orhospital controls. The RR analysis by dichotomization with orwithout age-adjustment and the quartile analysis as well failedto find any significant associations with these macronutrients.One may, however, argue that the failure to find a difference infat, in particular, might reflect unusually high fat intakes by thetwo series of controls relative to the general population. Thisis apparently least likely as suggested in the earlier discussion

on our dietary survey method. Socioeconomic status is in general associated with diets, and might have produced the ohserved results for fat, vitamin A, and fl-carotene. Such anargument is also not likely, because several indirect indicators

of socioeconomic status such as educational level, occupation,marital status, and religion were not significantly differentamong the three study subjects.

The association between vitamin A/@3-caroteneand prostatecancer risk is conflicting. Graham et aL (10) reported a directassociation of an index of vitamin A (without distinction between the preformed and precursor forms) intake with prostatecancer risk: the larger the monthly intake, the higher the risk,in men aged 70 or older, but not in men aged 69 or younger.In the older men, the RR increased from the lowest to thehighest quartile of monthly vitamin A intake as 1.00, 1.59,2.74, and 1.97 with a linear trend of significance. Their recentinvestigation (22), however, failed to show a positive doseresponse effect for dietary vitamin A in prostate cancer, whencontrolled for age, alcohol consumption, and smoking exposureby multiple logistic regression analysis. Kolonel and his associates (1 1) also reported an increase in prostate cancer risk withincreased ingestion of dietary vitamin A (retinol and carotenein foods plus retinol supplements) in the older men (70 yearsold or more), but not in the younger men. The RR derived frommultiple logistic regression with adjustment for age, ethnicity,

1334



DIETARY 13-CAROTENE AND PROSTATE CANCER

marital status, and total fat intake was 1.8 (1 .13.0)for thehighest quartile of weekly intake relative to the lowest in theolder men, but 0.7 (0.41.2)in the younger men. Their quartileanalysis for carotene intake, however, did not show any significant effect on prostate cancer risk. Heshmat and his colleagues(1 2) also reported an increased prostate cancer risk for theblacks with increased daily intake of vitamin A (without distinction between the preformed and precursor forms). Thisassociation was significant for the age period 3049years andonly suggestive for the age period 50 years and older. Thus, forthe U. S. white and black men, an increased ingestion of vitaminA enhanced the risk of prostate cancer. In contrast, we found asignificantly reduced risk of prostate cancer with increasedintake of dietary vitamin A (carotene precursors plus retinol)from all food sources as well as from green/yellow vegetablesin the older men (7079years), but not in the younger men(5069 years). In Japan, a reduction of prostate cancer risk was

linked with the consumption of large amounts of green/yellowvegetables (8, 9). Vitamin A-rich food (carrots) was also suggested to reduce the risk in the United States (23, 24). Of moreinterest in our study is that dietary a-carotene from all foodsources as well as from green/yellow vegetables reduced theprostate cancer risk in the older men, although Kolonel et a!.(1 1) and Shekelle et a!. (25) showed no evidence of an effect ofdietary carotene on prostate cancer risk. A protective effect ofdietary vitamin A observed in our analysis could probably beconcluded as specific to @3-carotene,since the separate analysesfor dietary retinol yielded no evidence of a risk reduction. Wehave greater confidence in these results for vitamin A and @-carotene, because an overall consistency was notable across thetwo different control series as well as the three different indicesof intake (uncorrected, per kg, and per kcal).

What are the possible explanations for the different riskmodification by vitamin A between Japan and the UnitedStates?

One may suspect the difference, if any, in genetic susceptibility to vitamin A. This may, however, not be a likely explanation,since an increased risk with increased ingestion of vitamin Awas observed in Caucasians (10) and Negroids (12) as well.Methodological differences in study design such as inclusionand exclusion criteria for cases and controls, and dietary surveymethod might in part be related. As discussed earlier, however,our methodological limitations, particularly of dietary surveymethod, are believed not to be so serious as to produce aspurious finding as well as to reverse a direction of risk modification.

In Japan, the daily amounts of per capita consumption ofprotein, carbohydrate, fat, and vitamin A were 79.6 g, 329 g,55.2 g, and 2060 IU, respectively, in 1976 (26). In the UnitedStates, the corresponding figures were 101 g, 376 g, 157 g, and2430 @ig(equivalent to 8100 IU) (27). The United States consumed more fat by 2.8 times and more vitamin A by 3.9 timesthan Japan. In 1984, daily per capita amount of fat in the foodsupply was 79.9 g in Japan (28) and 166 g in the United States(29): about a two-fold difference in fat supply. The daily meanintake of dietary vitamin A was 16092096IU in our studysubjects (Table 2), whereas, in 82 Pittsburgh men (30), that was6944 IU by the 24hrecall, 5 137 IU by the 3-day food record,and 7876 IU by the food frequency questionnaire. Regardlessof dietary survey methods, Pittsburgh men ingested more vitamm A by approximately 3.5 times than our study subjects. Inthe light of these comparisons, it is presumed that such a hugedifference in fat consumption might be a most plausible explanation for the different risk modification between the two

countries, because the ingestion of large amounts of fat issuggested to alter the hormonal milieu to increase prostatecancer risk (12) and perhaps even to overwhelm a possibleprotective effect of vitamin A in the United States. Or alternatively, vitamin A might protect against prostate cancer in moderate amounts, but could enhance the risk in large amounts.The men in the United States appear to receive an overwhelming dose of vitamin A because of greater bioavailability withhigh fat ingestion, since vitamin A is a fat-soluble vitamin,whereas this might not be the case in Japan. In such an instance,the Japanese could reveal a decrease in risk. These are ourspeculations, but it may possibly be supported by quite a few

studies using experimental animals and human data as well thatindicate the toxic effects and even tumor-promotion effects ofvitamin A given in large doses (7, 31, 32, 33).

f3-Carotene in fruits was positively associated with prostatecancer risk when cases were compared to hospital controls.This was observed in the older age group, which showed apositive association with vitamin A in the U. S. studies (10,I 1). This particular finding might possibly imply that all $-carotene is not alike, and/or that another component of fruitsthat is highly correlated with $-carotene actually increase prostate cancer risk, since fl-carotene is contained far less in fruitsthan in green/yellow vegetables. @3-Carotenein a unique Japanese food (seaweeds and kelp) was suggested to reduce the riskin the older men, when the uncorrected intake was comparedbetween cases and BPH controls. This finding, though probablyascribable to chance, might indicate the usefulness ofa separateexamination for nutrient components by food item, becausefood sources of vitamin A and/or @9-carotenemay probably bedifferent between Japan and the United States. In general,however, analysis of dietary nutrients rather than food itemshas an advantage of protecting against the possible effects ofother nutrients contained in the same food item (12).

The mechanism of risk modification by dietary vitamin A isnot well understood particularly for prostate cancer. Grahamet a!. (10, 34) suggested a different risk modification according

to the cell type of tumors. Heshmat et a!. (12) etiologicallyrelated an increased risk to a disturbance in zinc-retinol bindingprotein-vitamin A axis. Our study might possibly suggest anunfavorable effect of huge fat ingestion on a protective effect ofdietary vitamin A against prostate cancer.

Finally, the prostate cancer risk was inversely associated withdietary fl-carotene and vitamin A as well, but not associatedwith dietary fat, protein and retinol in our study. This wasrevealed in Japan, a country where fat has been increasinglyconsumed in the past two decades (21), but is still ingested onlyat approximately one-third of the quantities consumed in theUnited States. Our study seems to be the first epidemiologicalstudy that demonstrated a reduced risk of prostate cancer withincreased intake of dietary $-carotene and vitamin A as well.More data are undoubtedly required which should focus on therole of these dietary nutrients particularly in the light of different dietary levels offat and other nutrients. This must be studiedin depth utilizing all possible methods, in vitro, in vivo, andepidemiological.

ACKNOWLEDGMENTS

We thank M. Kato and J. Macno at Kyoto University, for dietaryinterview, and M. Arafuka at Nagoya City University, for technicalassistance of data analysis.

1335



grammableCalculator, pp. 510.Washington, DC: U. S. Government Printing Office, 1979.

18. Mantel, N. Chi-square tests with one degree of freedom: extensions ofMantel-Haenszel procedure.J. Am. Stat. Assoc., 58: 690-700, 1963.

19. Yamabe, H., ten Kate, F. J. W., GaIIee, M. P., Schroeder, F. H., Oishi, K.,Okada, K., and Yoshida, 0. StageA prostatic cancer a comparative studyin Japan and the Netherlands. World J. Urol., 4: 136-140, 1986.

20. Wendel,E. F., Bannen,G. E., Putong,P. B., and Grayhack,J. T. The effectoforchiectomy and estrogenson benign prostatic hyperplasia.J. Urol., IO&116119,1972.

21. Ministry of Health and Welfare, Japan, Health Service Bureau, HealthPromotion and Nutrition Division. National Nutrition at Present,ResultsofNational Nutrition Survey in 1983.Tokyo: Diich-Shupan, Inc., 1985.

22. Middleton, B., Byers, T., Marshall, J., and Graham, S. Dietary vitamin Aand canceramultisitecase-controlstudy. Nutr. Cancer,8: 107-116, 1986.

23. Schuman, L M., Mandel, J. S., Radke, A, Seal, U., and Halberg, F. Someselectedfeaturesof the epidemiology of prostatic cancer MinneapolisSt.Paul,Minnesotacase.controlstudy,19761979.In. K. Magnus(ed.),Trendsin Cancer Incidence: causesand practical implications, pp. 345-354. Washington,DC:Hemisphere,1982.

24. Ross,R. K., Paganini-Hill,A., andHenderson,B. E. The etiologyof prostatecancer what doesthe epidemiology suggest?Prostate, 4: 333-344, 1983.

25. Shekelle, R. B., Lepper, M., Liu, S., MaIiZa, C., Raynor, W. J., Jr., Rossof,A. H., Paul,0., Shryock,A. M., andStamler,J. DietaryvitaminA andriskofcancer in the westernelectric study. Lancet, 2: 11851190,1981.

26. Ministry of Health and Welfare,Japan,Health ServiceBureau,HealthPromotion and Nutrition Division. National Nutrition at Present.ResultsofNational Nutrition Survey in 1976.Tokyo: Diich-Shupan, Inc., 1978.

27. Page, L, and Friend, B. The changing United States diet. Bioscience, 28:192197,1978.

28. National Life Center (ed.). Life Statistics 85,pp. 211. Tokyo: Ministry ofFinance Printing Off@ice,1985.

29. Raper, N. R., and Marston, R. M. Levelsand sourcesoffat in the U. S. foodsupply. In: C. Ip., D. F. Birt, A. E. Rogers, and C. Mettlin (eds.), DietaryFat andCancer,pp. 127-152. New York: Alan R. Lisa,Inc., 1986.

30. Russell-Briefel,R., Caggiula,A. W., andKuller, L H. A comparisonof threedietary methods for estimating vitamin A intake. Am. J. Epidomiol., 122:628636,1985.

31. Schroder, E. W., and Black, P. H. Retinoids tumor preventers or tumorenhancer?J. NatI.CancerInst.,65:671674,1980.

32. Yenna, A. K., Conrad, E. A., and Boutwell, R. K. Differential effects ofretunoicacidand7,8-benzoflavoneontheinductionof mouseskintumorsbythecompletecarcinogenesisprocessand by the initiation-promotion regimen.Cancer Res., 42: 35193525,1982.

33. Ohshima, M., Ward, J. M., and Wenk, M. L Preventive and enhancingeffectsofretunoids on the developmentofnaturally occurring tumors of skin,prostate gland, and endocrine pancreas in aged male ACI/segHapBR rats. J.NatI. Cancer Inst., 74: 517524,1985.

34. Graham, S. Epidemiology of retinoids and cancer. J. Natl. Cancer Inst., 73:14231428,1984.

DIETARY a-CAROTENE AND PROSTATE CANCER

1336

REFERENCES

I . Waterhouse,J., Muir, C., Shanmugaratunam,K., and Powell, J. (eds.),CancerIncidencein Five Continents,Vol. 4, pp. 746747.Lyon: International Agency for Researchon Cancer, 1982.

2. Kurihara, M., Aoki, K., and Tominaga, S. Cancer Mortality Statistics in theWorld, pp. 163. Nagoya The University of Nagoya Press, 1984.

3. Mandel, J. S., and Schuman,L. M. Epidemiologyofcancer of the prostate.In: A. M. Liienfeld (ed.), Reviews in Cancer Epidemiology, pp. 1-83. NewYork, Amsterdam:Elsevier/North-Holland,1980.

4. Woolf, C. M. An investigation of the familial aspectsof carcinoma of theprostate. Cancer (Phila.), 13: 739744,1960.

5. Wynder, E. L, Laakso, K., Sotarauta, M., and Rose, D. P. Metabolicepidemiology ofprostatic cancer. Prostate, 5: 4753,1984.

6. Flanders,W. D. Review:prostatecancerepidemiology.Prostate,5:621629,1984.

7. Sosland, M. C. Diet and cancer of the prostate: epidemiologic and experimental evidence. in: B. S. Reddy and L A. Cohen (eds.), Diet, Nutrition,and Cancer a Critical Evaluation, Volume I: Macronutrients and Cancer,pp. 125149.BocaRaton, florida CRC Press,Inc., 1986.

8. Mishina,T., Watanabe,H., Araki, H., andNakao, M. Epidemiologicalstudyofprostatic cancer by matched-pair analysis. Prostate, 6: 423-436, 1985.

9. Hirayama,T. A largescalecohortstudyon cancerrisksbydietwithspecialreferenceto the risk reducing effects ofgreen-yellow vegetabk consumption.In: Y. Hayashi, M. Nagao, T. Sugimura, et a!. (eds.), Diet, Nutrition andCancer, pp. 4153.Tokyo: Japan Sd. Soc. Press,Utrecht: VNU Sci. PressBY, 1986.

10. Graham, S., Haughey, B., Marshall, J., Priore, R., Byers,T., Rzepka, T.,Mettlin, C., and Pontes,J. E. Diet in the epidemiologyof carcinomaof theprostate gland. J. Nail. Cancer Inst., 70:687-692, 1983.

11. Kolonel,L N., Hinds, M. W., Nomura, A. M. Y., Hankin, J. H., and Lee,J. Relationshipof dietary vitamin A and ascorbicacid intake to the risk forcancers of the lung, bladder, and prostate in Hawaii. Natl. Cancer Inst.Monogr.,69:137142,1985.

12. Heshmat, M. Y., Kaul, L, Kovi, J., Jackson, M. A., Jackson, A. G., Jones,G. W., Edson, M., Enterline, J. P., Worrell, R. G., and Perry, S. L Nutritionand prostate cancer a case.controlstudy. Prostate, 6: 717,1985.

13. Cantrell, B. B., Dc Kierk, D. P., Eggleston,J. C., Boitnott, J. K., and Walsh,P. C. Pathologicalfactorsthat influenceprognosisin stageA prostaticcancerthe influenceofextent versusgrade.J. Urol., 125: 516-520, 1981.

14. Hankin, J. H., Rhoads,G. G., and Glober, G. A dietary method for anepidemiologic study ofgastrointestinal cancer.Am. J. Clin. Nutr., 28: 10551061, 1975.

15. Committee on Resources,Science Bureau, Japan (ed.). Standard Tables ofFoodCompositionin Japan,4thedition,1983.

16. Fleiss, J. L. Statistical Methods for Ratesand Proportions, pp. 2627.NewYork: Wileyand Sons, 1981.

17. Rothman, K. J., and Boice, J. D., Jr. Epidemiologic Analysis with a Pro



1988;48:1331-1336. Cancer Res Yoshiyuki Ohno, Osamu Yoshida, Kenji Oishi, et al. Study in Kyoto, Japan

-Carotene and Cancer of the Prostate: A Case-ControlDietary

Updated version

http://cancerres.aacrjournals.org/content/48/5/1331

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/48/5/1331To request permission to re-use all or part of this article, use this link


http://cancerres.aacrjournals.org/content/48/5/1331http://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/48/5/1331http://cancerres.aacrjournals.org/

Dietary $-Carotene and Cancer of the Prostate: A Case...

Documents

Transcript of Dietary $-Carotene and Cancer of the Prostate: A Case...