Diet, Tobacco Use, and Fatal Prostate Cancer - Cancer Research

Diet, Tobacco Use, and Fatal Prostate Cancer - Cancer Research

(CANCER RESEARCH 50. 6836-6840, November I, I990| Diet, Tobacco Use, and Fatal Prostate Cancer: Results from the Lutheran Brotherhood Cohort Study An...

1MB Sizes 0 Downloads 17 Views

Recommend Documents

Increased Risk of Prostate Cancer and Benign - Cancer Research
Oct 15, 2000 - Department of Urology [T. H., Z. L., T. S., R. S., N. T., H. T., N. S., S. S., K. S., T. Kat.], Akita Uni

Pim kinase inhibitors sensitize prostate cancer - Cancer Research
Andrew S. Kraft, Department of Medicine, Medical University of South Carolina, 86. Jonathan Lucas Street, Charleston, SC

Prostate Cancer 2017: The Coming Sea-Change of Prostate Cancer
Aug 17, 2017 - Prostate Cancer 2017: The Coming Sea-Change of Prostate Cancer Care | Simms/Mann-UCLA Center for Integrat

AccelerAting Discovery - Prostate Cancer Foundation
Lawrence J. and Joyce Stupski. Tarnopol Family Foundation .... Laurence and Karen Mandelbaum. Bernard Marcus ... Henry a

Cancer Registration - Cancer Research UK
Achievements made possible by cancer registration information: • research showing that there are at least. 10 differen

Thank you - Prostate Cancer UK
The Milly Apthorp Charitable Trust. - The Open Agency. - The Rangers Charity Foundation. - The Simon Gibson Charitable T

banana bread - Prostate Cancer UK
Angela Hartnett MBE. BANANA BREAD. Makes one loaf (12 slices). Ingredients. • 100g Sultanas. • 75ml rum. • 4 Small

international cancer research conference - Windsor Cancer Research
Nov 22, 2014 - Justin Peterson*, Adam Farag, Trevor Szekeres, Eli Gibson, Aaron Ward, Joseph Chin, Stephen. Pautler, Gle

Aspirin and Colorectal Cancer - Clinical Cancer Research
Mar 1, 2014 - Aarnio M, Sankila R, Pukkala E, Salovaara R, Aaltonen LA, de la. Chapelle .... Stark LA, Reid K, Sansom OJ

Unproven Therapies in Prostate Cancer
Theranos was performing tests on patients without having published peer-reviewed research – a cardinal science – and

(CANCER RESEARCH 50. 6836-6840, November I, I990|

Diet, Tobacco Use, and Fatal Prostate Cancer: Results from the Lutheran Brotherhood Cohort Study Ann W. Hsing,1 Joseph K. McLaughlin, Leonard M. Schuman, Erik Bjelke, Gloria Gridley, Sholom Wacholder, Harvey T. Co Chien, and William J. Blot National Cancer Institute, Betttesda, Maryland 20X92 ¡A.W. H., J. K. M., G. G., S. W., W. J. B.J; Division of Epidemiology, School of Public Health, University of Minnesota, Minneapolis, Minnesota [L. M. SJ; Center for Epidemiologie Research, I 'niversity of Herren, Norway //;'. H./; and H estât.Inc., Rockville, Maryland 20850 ¡H.T. C. C.]

ABSTRACT

before the diagnosis of cancer, thus avoiding recall bias, a potential problem in case-control studies of cancer.

A cohort of 17,633 »hite males age 35 and older responded to a mailed epidemiológica! questionnaire in 1966 and was followed until 1986 to determine the risk of cancer associated with diet, tobacco use, and other factors. During the 20-year follow-up, 149 fatal prostate cancer cases were identified. Relative risks for prostate cancer were significantly elevated among cigarette smokers (relative risk, 1.8; 95% confidence interval, 1.1 -2.9) and users of smokeless tobacco (relative risk, 2.1; 95% confidence interval, 1.1-4.1). No significant associations were found with frequency of consumption of meats, dairy products, fruits, or vegetables, there were no overall significant associations between consumption of vitamin A from animal sources (retinol) and provitamin A from plant sources (carotene) and risk, but positive trends were seen for ages under 75, while inverse associations were found at older ages. Beverage con sumption, including drinking coffee and alcohol, was unrelated to risk. Marital status, education, rural/urban status, and farming residence were also unrelated to the risk of fatal prostate cancer. The findings add to limited evidence that tobacco may be a risk factor for prostate cancer, but fail to provide clues to dietary or other risk factors.

INTRODUCTION Prostate cancer is a major form of cancer among men, accounting for 20% of cancer incidence and 11% of cancer deaths in the United States (1). Worldwide, the highest rates of prostate cancer occur among United States blacks and the lowest among Asians (2). Despite its high incidence and mor tality, very little is known about the etiology of prostate cancer. Previous epidemiológica! studies have suggested a number of risk factors, including socioeconomic and marital status, farm ing residence, alcoholic beverages, exposure to cadmium, ve nereal disease, number of sexual partners, and androgen levels (3-5). A few investigations have linked cigarette smoking (68) to an increased risk of prostate cancer, but this tumor is generally not considered to be tobacco related (9, 10). The role of diet in the etiology of prostate cancer has been investigated in a number of studies with equivocal results. Increased intake of fat and animal products has been associated with increased risk (11-14), although not always consistently (15-17). Some investigators have found increased dietary intake of vitamin A to enhance prostate cancer risk, particularly at older ages (17-20), while others have reported that high serum levels of retinol (21) and consumption of /i-carotene-containing foods lower the risk (22, 23). In this report, we present results from a cohort study of 17,633 men to determine associations of diet, tobacco use, beverage consumption, and demographic characteristics with the risk of fatal prostate cancer. Such a study design permits the measurement of diet and other factors Received 3/12/90; accepted 7/18/90. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 II.S.C. Section 1734 solely to indicate this fact. 1To whom requests for reprints should be addressed, at Epidemiology and Biostatistics Program. Division of Cancer Etiology. National Cancer Institute. Executive Plaza North. Room 415. Bethesda. MD 20892.

MATERIALS AND METHODS The Lutheran Brotherhood Cohort. In 1966. a cohort (n = 26,030) of white male policy holders of the Lutheran Brotherhood Insurance Society who were at least 35 years of age was selected for a mortality follow-up study (median age at entry into the cohort, 51). In September 1966, a questionnaire was mailed to members of the cohort with questions on demographic background, frequency of use of dietary items, and tobacco habits. A response rate of 68.5% was achieved after three mailings. A comparison of respondents and nonrespondents, using original insurance records, found little difference in age, urban/ rural residence, or policy status (24). Since the LBS2 cohort was a self-selected group, having purchased life insurance, several characteristics of the cohort members were dif ferent from those of the United States population. The cohort was primarily from the upper midwest, particularly Minnesota, and north eastern areas of the United States (45 and 39%, respectively), with 30% of Norwegian and other Scandinavian heritage. In addition, compared to United States white males in the mid-1960s, the LBS cohort was more rural (37 versus 29%). and more were farmers (38 versus 4%) (25). Mortality. The LBS cohort was followed for mortality from 1966 to 1986. for a total of 286,731 person-years. Death certificates, which were received semiannually from the Lutheran Brotherhood Insurance Society, were coded for underlying cause of death, all other contributory causes of death, and all other significant conditions by the nosologist of the Minnesota State Department of Health. During the 20-year follow-up, 4513 deaths, including 1033 cancer deaths, were identified. The two major causes of death were heart disease and cancer. Among the cancer deaths, the distribution of anatomic sites was similar to that among United States males, with the most common being malignancies of the lung (n = 203), prostate (n = 149). and large intestine (n = 120). The overall mortality experience of this cohort was similar to that of United States white males, although the number of prostate cancer deaths was slightly lower than expected (standard mortality ratio. 0.92) (26). During the follow-up, 4027 subjects (23%) were lost to follow-up due to lapsed policies or discontinuation of policies after premium maturity. A comparison of cancer mortality at 11.5 years of follow-up showed no significant differences between original respondents and nonrespondents and those whose policies lapsed (24). Tobacco and Alcohol. Information on tobacco use. including the use of cigarettes, pipes, cigars, and smokeless tobacco (snuff or tobacco chewing), was elicited in the 1966 questionnaire. Since no subsequent information on tobacco use was obtained, smoking exposure in the analysis pertained to the smoking experience of cohort members in 1966. Thus, current smokers were defined as those smoking in 1966. Information on the use of beer and whiskey was also asked in the 1966 questionnaire: persons who used either beer or liquor (whiskey, gin, cognac, etc.) at least 6 times a year were defined as users. Dietary Data. The respondents were asked about the frequency of their current (in 1966) monthly intake of foods. Thirty-five individual food items, including vegetables (potatoes, cabbage, rutabagas, carrots. : The abbreviations used are: LBS, Lutheran Brotherhood Insurance Society; RR. relative risk; CI. confidence interval.

6836

Downloaded from cancerres.aacrjournals.org on December 28, 2017. © 1990 American Association for Cancer Research.

DIET, TOBACCO USE, AND FATAL PROSTATE CANCER

Table 1 Relative risks of fatal prostate cancer by quartile levels of selected food groups in the Lutheran Brotherhood cohort, 1966-1986

cauliflower, corn, green salads, tomatoes, peas, and beans), vegetable and fruit soups, fruits (oranges, apples, grapes, bananas, canned fruit), fruit juices, fish, meat, dairy products (milk and ice cream), cereals, bread, and alcohol were included in the questionnaire. These 35 food items were combined into 9 food groups. Information on portion size (sex specific) and nutrient values was derived from the Second National Health and Nutritional Examination Survey and from data of the United States Department of Agriculture, respectively (27-28). For example, the carotene value for a specific food item for each subject was derived by multiplying the reported consumption frequency by the average portion size and the carotene content of that food item. The total carotene index for a specific individual was derived by summing the values from all carotene-containing foods. One hundred eighty-five individuals were excluded because they had more than 10 unknown responses to the food item questions. The excluded individuals were older, more likely to reside in rural areas, and less educated as compared to the 17,633 subjects who remained in the final analysis. For the included subjects. 71% had no missing data on any food items; of the remaining 29% (87% of whom had fewer than 5 items missing), we imputed intake for missing food items, using the median values of the remaining subjects, stratified by urban/rural, education, and age categories. The imputation is unlikely to greatly affect the nutrient analysis, since, for example, it contributed to only about 1% of the vitamin A index. Consumption of food groups and nutrients was divided into quartiles based on the intake of all 17.633 subjects. Statistical Analysis. A Poisson regression program for modeling hazard functions with grouped data was used to calculate age-adjusted relative risks (29-30). Since the number of prostate cancer deaths in this cohort is small relative to the size of the cohort, the prostate cancer deaths occurring during the 20-year follow-up may be regarded as statistically independent Poisson variables. As with other grouped survival analyses, the major assumption for this Poisson regression method is that the hazard function is constant in each time interval. Five-year age intervals (attained age) were used for the grouping of the data; RRs for prostate cancer were calculated for each age stratum and summarized over all strata for the selected variables. Persons-years for study subjects were accumulated up to death, loss to follow-up, or the end of the follow-up in 1986.

RESULTS Among the 149 cohort members who died of prostate cancer, the median age at death was 73, compared to 70 for all deaths in the cohort. There were no significant associations for prostate cancer risk with marital status, farming residence, or education. Table 1 shows the risk of fatal prostate cancer associated with intake of 9 food groups. No significant associations were found with consumption of meat, poultry, fish, eggs, dairy products, vegetables, cruciferous vegetables, fruits, or breads. In addition, no individual food item was found to increase or decrease risk significantly. No significant trends in risk were found in association with total vitamin A, retinol, or ß-caroteneintake. The smokingadjusted RRs for the 4 quartiles were 1.0, 1.2, 0.8, and 1.1 for total vitamin A; 1.0, 0.8, 0.9, and 1.2 for retinol; and 1.0, 1.2, 1.3, and 0.9 for /3-carotene; however, because increased intake of vitamin A and/or carotenes has been reported to enhance the risk among older men (age at diagnosis >70), intake of total vitamin A, retinol, and /i-carotene was analyzed for the two age groups separately (age at death <75 or >75) (Table 2). For age <75, increased consumption of total vitamin A en hanced the risk of prostate cancer: smoking-adjusted RRs for the four quartiles were 1.0, 2.3, 1.7, and 2.8, respectively. However, for older ages, increased consumption of total vitamin A was associated with a decreased risk of prostate cancer; RRs

Food groups of (times/mo)"'*Meat<1718-2627-39>39Poultrv<0.50.5-1.51.6-4.0>4.0Fish<0.80.8-1.71.8-4.0>4 CI0.6-1.40.7-1.60.5-1.30.8-2.60.6-1 deaths5035372714655020315139284140452346473323483045263

vegetables<1.21.2-2.22.3-4.5>4.5Fruits<29.329.3-46.546.6-67.0>67.0Breads<8888-138.8138

°Consumption frequency per month. * Food items included in each food group are as follows: meat: beef, bacon, fresh pork, and smoked ham: poultry: chicken; fish: fresh fish, salted fish, and fish balls; dairy group: milk and ice cream: vegetables: potatoes, green salad, tomatoes, carrots, peas. corn, beans, vegetable soup, cabbage, cauliflower, pea soup, and rutabaga: cruciferous vegetables: cabbage, cauliflower, and rutabaga: fruits: fruit juice, canned fruit, apples, bananas, oranges, grapes, and fruit soup; bread group: bread, packaged cereal, cooked cereal, pancakes, and flat bread. c Adjusted for age and tobacco use.

were 1.0, 0.7, 0.5, and 0.4, respectively. There were similar risk patterns by age for retinol and /3-carotene. Similar age effects were found when 70 (age at death) was used as the cutoff. All forms of tobacco use were associated with an increased risk of prostate cancer (Table 3). Risks were significantly ele vated among persons who ever used any form of tobacco (RR = 1.8, 95% CI, 1.1-2.9), both among cigarette smokers and users of smokeless tobacco. Risks according to number of cigarettes smoked and regularity of smokeless tobacco use are presented in Table 4. There was no clear dose response for amount of cigarettes smoked, and there was little difference after adjustment for the use of smokeless tobacco. Among current smokers, however, the risk was elevated among those who inhaled compared to those who did not (RR = 2.0; 95% CI, 0.7-5.8). Among persons who had ever used smokeless tobacco, the risk was highest among persons who dipped snuff or chewed tobacco regularly (RR = 2.4; 95% CI, 1.3-4.9), as compared to those who had never used any tobacco. A review of contributory and other causes of death on the

6837

Downloaded from cancerres.aacrjournals.org on December 28, 2017. © 1990 American Association for Cancer Research.

DIET, TOBACCO USE, AND FATAL PROSTATE CANCER

Table 2 Relative risks" of fatal prostate cancer, by age and quartile levels of nutrient indices in the Lutheran Brotherhood Cohort, 1966-1986

Table 4 Relative risks of fatal prostate cancer associated with level of tobacco use in the Lutheran Brotherhood Cohort, 1966-1986

vita

Quartiles*Age<75*(n Af1.02.3(1.1-4.9)1.7(0.8-3.8)2.8(1.4-5.8)P" min 78)'1 = (low/'234Age

useCigarettes'Never Tobacco

of per deaths"19915212113194213524Cohort son-years58,888190,64087,21529.40436,58815,73 CI1.1-2.91.1-3.30.8-3.30.8-3

tobaccoEver used any used (0.5-2.3)2.0(1.0-3.8)1.7(0.9-3.3)P (0.6-2.5).6(0.8-3.1).9(1.0-3.7)0cigarettes''Occasional ex-smoker1-19 and cigarettes/day20-29 0.051.00.7(0.4-1.3)0.5(0.3-1.0)0.4 < 0.05.0.1 < cigarettes/day30 i1.00.6(0.3-1.2)0.4 < 0.05 71)'1 3 75 (n = cigarettes/daySmokeless + tobacco'Never (low)234Total (0.6-2.1).1 tobaccoEver used any tobacco*Ex-usersOccasionalRegularNo. used smokeless (0.2-0.8)0.9(0.5-1.7)(0.6-1.9)).2 (0.2-0.9)P (P (0.1-0.6)°<0.01 < 0.01Retino!1.01.1 > 0.05 i/3-carotene.0.2 " Adjusted for age and tobacco smoking. °Missing data not included. * Cutoff points for quartile were 95,289, 128,007, and 167,479 IU/month for * Adjusted for age. total vitamin A; 9,585, 13.672, and 19,084 Retinol Equivalent/month for total ' Thirty-four subjects who used pipes, cigars, and smokeless tobacco only were retinol; and 11,517. 19,100, and 30,165 ¿ig/monthfor d-carotene. ' Represents vitamin A from both plant and animal sources. not included. d Including some cigarette smokers who used other forms of tobacco. d Age at death for prostate cancer cases. ' Snuff and chewing tobacco. 78 subjects who used only other forms of tobacco ' Number of prostate cancer deaths. ^Reference category. were not included. ^Adjusted for cigarette smoking. * Number of prostate cancer deaths by quartile, among those age <75 were 10, * Including some smokeless tobacco users who used cigarettes. 23, 17, and 28 for total vitamin A; 14, 15, 27, and 22 for retinol; and 13, 16, 22, and 27 for rf-carotene; among those age ^15 were 29, 18, 13, and 11 for total vitamin A; 27, 17, 8, and 19 for retinol; and 20, 22. 25, and 4 for /I-carotene. current users of beer were 1.7 (95% CI, 1.0-2.9) and 1.2 (95% Number of study subjects among those age <75 were 3,252, 3,389, 3,384, and 3,389 for total vitamin A; 3,240, 3,328. 3,397, and 3,449 for retinol; and 3,424. CI, 0.8-1.7), respectively, while the risk for ex- and current 3,381, 3,316, and 3,293 for fi-carotene; among those age >75 were 1,156, 1,019, users of liquor were 0.7 (95% CI, 0.3-1.5) and 1.0 (95% CI, 1,025, and 1,019 for total vitamin A; 1.168, 1,081, 1,011, and 959 for retinol; 0.7-1.4), respectively. and 3,294, 3,369, 3,405, and 3,346 for (l-carotene. * Linear trend test.

DISCUSSION

Table 3 Relative risks of fatal prostate cancer associated with tobacco use in the Lutheran Brotherhood Cohort, 1966-1986 use"Never Tobacco

In contrast to some studies that have linked per of deaths*1911692210358416Cohort son-years58,888217,30013,67648,8234,02593,5447,6132,72922.896RR'1.01.81.62.04.51.72.91.41.695% CI1.1-2.90.7-3.51.1-3.72.1-9.71.0-2.91.3-6.50.5-4.10.8-3.1 animal products with increased prostate cancer

tobaccoUsed used any tobacco''Pipes any form of onlyCigarettes and/or cigars onlySmokeless onlyCigarettes tobacco' pipesand/or and cigarsCigarettes smokelesstobaccoSmokeless and andpipes tobacco cigarsCigarettes and/or smokelesstobacco and and/or and pipes cigarsNo. " All categories refer to ever users. '' Missing data not included. ' Adjusted for age. 'included 12 subjects who used some form of tobacco, but for whom no detailed information was available for further classification. ' Snuff and chewing tobacco.

death certificates revealed 58 subjects for whom prostate cancer was not the underlying case of death. The 58 cases were much older than the 149 fatal cases (median age at death 81 versus 73). In this group, a positive association with smokeless tobacco use was also found. Fourteen of the cases were regular users of smokeless tobacco (RR = 2.3; 95% CI, 1.0-5.2), including 8 who dipped snuff or chewed tobacco exclusively (RR = 2.5; 95% CI, 1.0-6.5). The risk associated with cigarette smoking in this group, however, was not significantly elevated (RR = 1.1; 95% CI, 0.5-2.3). Risk of prostate cancer was also assessed for consumption of coffee and alcoholic beverages. For persons who drank 3-4 and 5 or more cups of coffee/day, the RRs were 0.8 (95% CI, 0.61.2) and 1.0 (95% CI, 0.6-1.6), respectively, as compared to those who drank less than 3 cups of coffee/day. Increased consumption of beer or hard liquor was weakly associated with an increased risk. Smoking-adjusted risks for ex-users and

dietary fat and risk, we found no association with consumption of eggs, milk, pork, or beef. In addition, we found no positive effect upon prostate cancer risk with the intake of green vegetables. There also was no overall association with retinol or /3-carotene intake. The ob served higher risks among those under 75 years and lower risks among those 75 years and older for consumption of total vitamin A, retinol, and 0-carotene stand in contrast to studies that reported increased risk with vitamin A intake in older men (17-20). The etiological implication of our findings on vitamin A or those of previous researchers are not clear at this time. The lack of an association between dietary factors and the risk of prostate cancer may be due partially to the limited number and nature of the dietary items in the 1966 foodfrequency questionnaire. The foods in the self-administered instrument in this study were adapted from dietary question naires of earlier case-control studies in Minnesota and Norway, which were specifically designed to elicit sufficient information to discriminate between persons with light and heavy consump tion of these food items (31). However, because certain food items that are major contributors of vitamin A (both retinol and carotene) or dietary fat in the American diet (such as liver, cheese, butter, processed meat, broccoli, spinach, and canta loupe) were not included in the questionnaire, we were unable to capture all sources for each nutrient. Hence there may be random misclassification of dietary intakes, which may tend to dampen the relative risk estimates but should not account for the opposite trends associated with vitamin A in men under versus over age 75. In recent reviews of tobacco smoking and cancer (9, 10), prostate cancer is considered not to be a tobacco-related cancer. Most epidemiological studies, including 8 prospective studies (9) and 7 case-control studies (32-38), have reported no in creased risk of prostate cancer among smokers. The lack of an

6838

Downloaded from cancerres.aacrjournals.org on December 28, 2017. © 1990 American Association for Cancer Research.

DIET.TOBACCO USE,AND FATAL PROSTATE CANCER

association with cigarette smoking in these case-control studies could be due in part to the high prevalence of cigarette smoking in control groups because most used hospital patients as the comparison group and a moderate association with cigarette smoking may be masked by the prevalence of smoking in hospital controls. It is noteworthy that when neighborhood controls were used for comparison, a positive association with cigarette smoking was observed (7, 8). Cohort studies, however, do not have this potential methodological limitation, and all but the follow-up study of United States veterans (6) showed no excess of prostate cancer mortality among tobacco users. In the LBS cohort, risk was increased nearly 80% among smokers. The increase may have been more easily detected in this cohort, since the prevalence of current cigarette smoking (32%) was low compared to that of United States middle-aged white males in 1965 (44%) (39). We, however, found no evidence of doseresponse trends, and it is possible that the base-line reference group of the nonusers of tobacco differed in other ways that relate to prostate cancer risk. If real, the association of cigarette smoking with prostate cancer could have a hormonal basis, since cigarette smoking is associated with an antiestrogenic effect (40, 41), and male smokers are reported to have elevated levels of circulating androsterone and testosterone (42, 43), which may increase the risk of prostate cancer. Further research is needed to clarify whether cigarette smoking is related to the risk of prostate cancer. The prevalence of smokeless tobacco use in the LBS cohort was 17%, which is similar to the national prevalence (20%) reported in the 1960s (44). Although it is known that the use of smokeless tobacco is related to oral cancer, its relationship with other cancers is less clear (44). Various parts of the body may be exposed to components of smokeless tobacco, such as nitrosamines, via the blood stream. The amounts of certain tobacco-specific nitrosamines are greater in snuff than in ciga rettes (45), and some /V-nitroso compounds [/V-nitrosobis(2hydroxypropyl)amine and A'-nitrosobis(2-oxypropyl)amine] can

Attrition is a major problem in most prospective studies. During the 20-year period, 4027 (23%) members of the cohort were lost to follow-up, due to lapsed policies or discontinuation of policies after premium maturity. We examined dietary in take, tobacco use, alcohol, and coffee consumption for these subjects lost to follow-up at 20 years relative to those remaining in the cohort and found no significant differences. In addition, after 11.5 years of follow-up, a special investigation of cancer mortality among lapsed and discontinued subjects found no difference from that of the remainder of the cohort (24). Thus, selection bias introduced by attrition in the study is likely to be minimal. The LBS cohort is a self-selected group, with more farmers, rural residents, and individuals of Scandinavian descent than in the United States as a whole (25), perhaps limiting the generalizability of the results. In addition, because mortality is used in the study as the end point, it is probable that subjects who had a diagnosis of advanced prostate cancer are more likely to die of prostate cancer and be included in the study; thus the results may not generalize to the less invasive and aggressive types of prostate cancer. In summary, our findings suggest that the use of cigarettes and smokeless tobacco is significantly associated with an in creased risk of fatal prostate cancer, providing renewed incen tive for evaluation of tobacco as a cause of this cancer. No significant relationships were found for dietary factors evalu ated in the study, but the possible role of vitamin A and ßcarotene deserves further attention.

REFERENCES

induce papillomas and squamous cell carcinomas of the prostate in rats (46, 47), while adenocarcinoma can be induced in both intact and castrated rats when testosterone is given along with Ar-nitrosobis(2-oxopropyl)amine (48). Our results suggest that the risk of prostate cancer associated with smokeless tobacco may be greater than that for cigarette smoking. However, we were unable to evaluate the intensity and duration of smokeless tobacco use because such information was not collected. Fur thermore, the effect of tobacco chewing could not be separated from that of using snuff in this study, since the questionnaire inquired about their combined use. Consistent with other stud ies, the use of smokeless tobacco in the LBS cohort is more prevalent among farmers and persons who reside in rural areas. Although studies (37, 49) have reported an association of farming residence with prostate cancer, we found no increased risk among farmers or rural residents. Although the study is prospective in design, information on dietary habits and tobacco use was obtained only once, in 1966. Misclassification of exposures thus could occur with those subjects who changed their tobacco or dietary habits in the follow-up period. For example, men who smoked in 1966 would still be classified as smokers, when as many as 40% of them may have quit smoking in the ensuing 20 years as they aged (39). Similarly, dietary patterns could have changed with time, and misclassification of dietary exposure is thus likely. How ever, such misclassification is likely to be nondifferential and would lead to an underestimation of the true effect.

Silvcrbcrg, E.. Boring, C. C., and Squires, T. S. Cancer statistics, 1990. CAA Cancer J. Clin., 40: 9-26, 1990. Waterhouse. J., Muir, C.. Corron. P., and Dowell, J. Cancer Incidence in Five Continents. Vol. 3. IARC Scientific Publication No. 15. Lyon. France: International Agency for Research on Cancer. 1976. Mandel. J. S.. and Schuman. L. M. Epidemiology of cancer of the prostate. In: A. M. Lilienfeld (ed.). Reviews in Cancer Epidemiology, pp. 1-83. Amsterdam: Elsevier/North-Holland. 1980. Flanders. W. D. Review: prostate cancer epidemiology. Prostate, 5:621-629, 1984. Owen. W. L. Cancer of the prostate: a literature review. J. Chronic Dis.. 29: 89-114. 1976. Rogot. E., and Murray, J. L. Smoking and causes of death among U.S. veterans: 16 years of observation. Public Health Rep., 95: 213-222, 1980. Honda. G. P., Bernstein, L., Ross, R. K., Greenland. S., Gerlans, V., and Henderson, B. E. Vasectomy, cigarette smoking and age at first sexual intercourse as risk factors for prostate cancer in middle-aged men. Br. J. Cancer, 57: 326-331. 1988. Schuman. L. M.. Mandel, J., Blackard. C.. Bauer, H., Scarlett, J., and McHugh. R. Epidemiologie study of prostate cancer: preliminary report. Cancer Treat. Rep.. 61: 181-186. 1977. International Agency for Research on Cancer. Tobacco Smoking. IARC Monogr. Eval. Carcinogen. Risk Chem. Hum., 38: 199-298, 1986. United States Department of Health and Human Services. The Health Consequences of Smoking. Cancer. A Report of the Surgeon General. United States Department of Health and Public Health Services. Office on Smoking and Health. Washington. DC: Department of Health and Human Services. Publication No. 82-50179. 1982. Kolonel, L. N., Yoshizawa, C. N., and Hankin, J. Diet and prostate cancer: a case-control study in Hawaii. Am. J. Epidemici., 127: 999-1212. 1988. Kaul, L.. Heshmat. M.. Kovi, J.. Jackson, M. A., Jackson, A. G., Jones, G. W., Enterline. J. P.. Worrell, R. G., and Perry, S. L. The role of diet in prostate cancer. Nutr. Cancer. 9: 123-128. 1987. Ross. R. K.. Shimizu. H.. Paganini-Hill, A.. Honda. G.. and Henderson. B. E. Case-control studies of prostate cancer in blacks and whiles in southern California. J. Nati. Cancer Inst., 78: 869-874, 1987. Snowdon, D. A., Phillips, R. L.. and Choi, W. Diet, obesity, and risk of fatal prostate cancer. Am. J. Epidemiol.. 120: 244-250. 1984. Severson. R. K., Nomura. A. M. Y., Grove, J. S., and Stemmermann, G. N. A prospective study of demographics, diet, and prostate cancer among men of Japanese ancestry in Hawaii. Cancer Res.. 49: 1857-1860. 1989. Phillips. G., and Snowdon, D. Association of meat and coffee use with cancers of the large bowel, breast, and prostate among Seventh-Day Adventists: preliminary results. Cancer Res.. 43: 2403s-2408s, 1983.

6839

Downloaded from cancerres.aacrjournals.org on December 28, 2017. © 1990 American Association for Cancer Research.

DIET.TOBACCO USE.AND FATAL PROSTATE CANCER 17. Graham, S.. Haughey, B., Marshall. J., Priore, R., Byers, T., Rzepka, T.. Mettlin, C., and Pontes. J. E. Diet in the epidemiology of carcinoma of the prostate gland. J. Nati. Cancer Inst.. 70: 687-692. 1983. 18. Kolonel. L. N., Hankin, J. H., and Voshizawa, C. N. Vitamin A and prostate cancer in elderly men: enhancement of risk. Cancer Res.. 47: 2982-2985, 1987. 19. Heshmat, M. V., Kaul, L., Kovi, J., Jackson. M. A.. Jackson. A. F.. Jones, G. W.. Edson, M., Enterline, J. P.. Worrell, R. G., and Perry, S. L. Nutrition and prostate cancer: a case-control study. Prostate, 67: 7-17. 1985. 20. Kolonel, L. N., Nomura, A. M. Y.. Hinds. M. W., Hirohata, T., and Hankin, J. H. Role of diet in cancer incidence in Hawaii. Cancer Res. (Suppl.), 43: 2397s-2402s, 1983. 21. Hayes, R., Bogdanovicz. J., Schroeder, F.. Bruijn. A.. Raatgever, J. W., Van Der Mass, P. J., Oishi, K., and Yoshida. O. Serum retinol and prostate cancer. Cancer (Phila.), 62: 2021-2026, 1988. 22. Hirayama. T. Epidemiology of prostate cancer with special reference to the role of diet. J. Nati. Cancer Inst., 53: 149-155, 1979. 23. Ohno, Y., Yoshida, O.. Oishi, K., Okada. K., Yamabe, H.. and Schroeder. F. H. Dietary /i-carotene and cancer of the prostate: a case-control study in Kyoto, Japan. Cancer Res., 48: 1331-1336. 1988. 24. Snowden. D. A. Alcohol use and mortality from cancer and heart disease among members of the Lutheran Brotherhood cohort. Ph.D. Dissertation, University of Minnesota, 1981. 25. United States Bureau of the Census: Census of the Population: 1970. Vol. 1, Characteristics of the Population. Part 1. United States Summary, Section 2, Table 224. Washington, DC: U.S. Government Printing Office, 1973. 26. Horm. J. W., Asiré, A. J., Young, J. L., Jr., and Pollack. E. S. (eds.). SEER Program: Cancer Incidence and Mortality in the United States. 1973-1981. N1H Publication No. 85-1837. Bethesda. MD: NIH. 1984. 27. National Center for Health Statistics. Plan and operation of the second National Health and Nutrition Examination Survey, 1976-1980. Vital and Health Statistics. Programs and Collection Procedures. Series 1, No. 15. Hyattsville, MD: United States Department of Health and Human Services, 1981. 28. United States Department of Agriculture. Composition of Foods: Raw, Processed. Prepared. Agriculture Handbook Nos. 8-1 to 8-10. Washington. DC: United States Government Printing Office. 1976-1983. 29. Preston, D. L., Kopecky. K. J.. and Kato, H. Analysis of mortality and disease incidence among atomic bomb survivors. In: W. J. Blot. Y. Hirayama, and D. G. Hoel (eds.). Statistical Methods in Cancer Epidemiology. Hiro shima. Japan: Radiation Effects Research Foundation. 1985. 30. Breslow, N. E., and Day, N. E. Statistical Methods in Cancer Research. Vol. 2. pp. 120-176. Lyon, France: IARC Scientific Publication No. 82, 1987. 31. Bjelke, E. Epidemiologie studies of cancer of the stomach, colon, and rectum with special emphasis on the role of diet. Vols. I-1V. Dissertation. University of Minnesota. Ann Arbor, MI: University Microfilms, 1973. 32. Ross. R. K., Paganini-Hill, A., and Henderson. B. E. The etiology of prostate

cancer: what does the epidemiology suggest? Prostate, 4: 333-344. 1983. 33. Mishina. T., Watanabe, H., Araki, H., and Nakao, M. Epidemiologie study of prostatic cancer by matched-pair analysis. Prostate, 6: 423-436. 1985. 34. Wynder, E. L.. Mabuchi, K.. and Whitmore, W. F. Epidemiology of cancer of the prostate. Cancer (Phila.). 28: 344-360. 1971. 35. Niijima, T., and Koiso, K. Incidence of prostate cancer in Japan and Asia. Scand. J. Urol. Nephrol. (Suppl.). 55: 7-21, 1980. 36. Newell, G. R., Fueger, J. J., Spitz, M. R., and Babaian, R. J. A case-control study of prostate cancer. Am. J. Epidemiol.. 130: 395-398. 1989. 37. Checkoway, H., Diferdinando, G., Hulka, B. S., and Mickey, D. D. Medical, life-style and occupational risk factors for prostate cancer. Prostate, 10: 7988, 1987. 38. Williams. R. R.. and Horm, J. W. Association of cancer sites with tobacco and alcohol consumption and socioeconomic status of patients: interview study from the Third National Cancer Survey. J. Nati. Cancer Inst.. 58: 525547. 1977. 39. United States Department of Health and Human Services. Reducing the Health Consequences of Smoking. 25 Years of progress. A Report of the Surgeon General. Washington. DC: Department of Health and Human Services (Center for Disease Control) 89-8411, 1989. 40. Michnovicz, J. J.. Hershcopf. R. J.. Naganuma. H.. Bradlow. H. L.. and Fishman. J. Increased 2-hydroxylation of estradiol as a possible mechanism for the anti-estrogenic effect of cigarette smoking. N. Engl. J. Med., 315: 1305-1309, 1986. 41. MacMahon, B.. Trichopoulos. D.. Cole, P., and Brown, J. Cigarette smoking and urinary estrogens. N. Engl. J. Med., 307: 1062-1065, 1982. 42. Dai, W. S.. Gutai, J. P., Kuller, L. H., and Cauley, J. A. Cigarette smoking and serum sex hormones in men. Am. J. Epidemiol., 128: 796-805, 1988. 43. Shaarawy. M.. and Mahmoud. K. Z. Endocrine profile and semen character istics in male smokers. Fértil.Steril., 38: 255-257. 1982. 44. United States Department of Health and Human Services. The health con sequences of using smokeless tobacco: a report of the advisory committee to the surgeon general. NIH Publication No. 86-3874, 1986. 45. International Agency for Research on Cancer. Tobacco habits other than smoking. IARC Monogr. Eval. Carcinogen. Risk Chem. Hum.. 37: 62-116, 1985. 46. Pour. P. A new prostatic cancer mode: systemic induction of prostatic cancer in rats by a nitrosamine. Cancer Lett., 13: 303-308, 1981. 47. Pour, P. Prostatic cancer induced in MRC rates by /V-nitrosobis (2-oxopropyl)amine and iV-nitrosobis (2-hydroxpropyl) amines. Carcinogenesis (Lond.). 4: 49-55. 1983. 48. Pour, P.. and Stepan, K. Induction of prostatic carcinoma and lower urinary tract neoplasms by combination treatment of intact and castrated rats with testosterone propionate and A'-nitrosobis(2-oxypropyl)amine. Cancer Res., 47:5699-5706. 1987. 49. Blair. A.. Malker, H.. Cantor. K. P., Burmeister, L., and Wiblund. K. Cancer among farmers: a review. Scand. J. Work Environ. Health, //.- 397-407. 1985.

6840

Downloaded from cancerres.aacrjournals.org on December 28, 2017. © 1990 American Association for Cancer Research.

Diet, Tobacco Use, and Fatal Prostate Cancer: Results from the Lutheran Brotherhood Cohort Study Ann W. Hsing, Joseph K. McLaughlin, Leonard M. Schuman, et al. Cancer Res 1990;50:6836-6840.

Updated version

E-mail alerts Reprints and Subscriptions Permissions

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/50/21/6836

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected] To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/50/21/6836. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on December 28, 2017. © 1990 American Association for Cancer Research.