Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
7 vitamins, Minerals, and Nonnutritive Inhibitors of Carc~nogenesis Many microconstituents of the diet have been found to have an impact on carcinogenesis. Only three groups of these substances are discussed in this chapter: vitamins A, C, and E and the carotenes; selenium and selected trace elements; and some nonnutritive compounds (i.e., indoles, phenols, aromatic isothiocyanates, methylated flavones, protease inhibitors, and plant sterols) that inhibit carcinogenesis in experimental systems. The ubiquity of these compounds is such that it is impossible for humans not to consume some of them. Evidence relat- ing these and other microconstituents to carcinogenesis was discussed in Chapters 9, 10, and 15 of the committee's first report (National Research Council, 1982~. Data relating other microconstituents (e.g., the B vitamins) to carcinogenesis are extremely limited, as explained in the first report. A striking feature of many of the compounds in the three groups mentioned above is their capacity to prevent or retard the occurrence of neoplasia. However, there is a paucity of data concerning the conditions under which inhibition occurs, the mechanisms of inhibition, and the precise impact of these microconstituents on humans. Better data on these compounds will assist in the formulation of more defini- tive conclusions and recommendations to reduce the risk of cancer. VITAMINS, CAROTENES, AND RETINOIDS . . . The dosages of vitamins used in experimental work and in human studies may range from levels recommended for optimal nutrition to megadoses. If research on vitamins and carcinogenesis suggests that large doses of specific vitamins or their analogs are needed to achieve effects, it should be noted that such doses are pharmacological in nature and may have deleterious consequences. Vitamin A, Carotenes, and Retinoids As explained In the first report, several epidemiological investi- gations indicate that there is an inverse relationship between esti- mated "vitamin A" intake and the occurrence of a variety of cancers. With few exceptions, the estimates of vitamin A intake in such studies were based on the frequency of ingestion of certain food groups, especially green and yellow vegetables that contain carotene (a pro- vitamin that is enzymatically converted to vitamin A in vivo) and a
Vitamins, Minerals, and Nonnutritive Inhibitors of Carcinogenesis 37 few foods, such as whole milk and liver, containing preformed retinal (vitamin A). Thus, these studies do not clearly distinguish between the effect of vitamin A Jeer _ and that of carotene. Experimental studies of the effects of vitamin A on carcinogenesis were also reviewed in the committee's first report (National Research Council, 1982, Chapter 9~. Animal experiments indicate that increased intake of this vitamin has a protective effect against the induction of cancer by chemical carcinogens in most but not in all instances. Only a few experiments have been published on the capacity of carotenes to inhibit chemically induced carcinogenesis. In contrast, their inhibi- tory effect on ultraviolet light-induced neoplasia of the skin has been well documented. There is a need to gain a detailed understanding of the effects of carotenes on carcinogenesis and the conditions under which vitamin A alters responses to neoplastic agents. In other animal experiments, certain nonnutritive compounds (i.e., indoles, phenols, aromatic isothiocyanates, methylated flavones, and plant sterols) present in fruits and vegetables were found to inhibit carcinogenesis. The specific agents responsible for the lower cancer incidence observed in populations that frequently consumed fruits and carotene-containing vegetables have not been identified (National Research Council, 1982, Chapter 9~. Although it is entirely appro- priate to propose interim dietary guidelines on the basis of current knowledge, identification of the responsible compounds is extremely important to assist in designing better epidemiological studies and in planning more precisely focused intervention studies concerning the consumption of green and yellow vegetables and certain fruits and the occurrence of cancer. Early studies indicating that vitamin A inhibits the occurrence of neoplasia in animals but that it can be toxic in high doses led to the synthesis of analogs of vitamin A--the retinoids. Certain retinoids are less toxic than vitamin A and can be targeted to specific organs where they exhibit inhibitory effects on carcinogenesis. Because of these properties, retinoids are an especially useful group of compounds for further investigation. Vitamin C (Ascorbic Acid) . Epidemiological studies have suggested that frequent consumption of foods containing vitamin C may be associated with a lower risk of cancer in humans, especially in the esophagus and stomach. However, there have been no systematic studies of populations consuming pharma- cological doses of vitamin C. Vitamin C has also been studied under a variety of experimental conditions for its effects on carcinogenesis. It has been well established that this vitamin can inhibit the forma- tion of nitroso carcinogens from precursor substances. Investigations
38 DIET, NUTRITION, AND CANCER: DIRECTIONS FOR RESEARCH of its effects on already-for~ed carcinogens have yielded less impres- sive results, with the possible exception of chemically induced neo- plasia of the large bowel. It should be pointed out, however, that almost all studies of the inhibitory properties of vitamin C have been conducted in species that synthesize their own vitamin C, unlike humans, who require a dietary source of this vitamin. There is some indication that vitamin C has mutagenic activity. However, ~n-vitro studies have generally indicated that vitamin C can prevent or, under some circum- stances, retard the manifestations of malignancy in cultured cells. Vi tamin E There are no epidemiological data concerning vitamin E and the risk of cancer in human populations. Such data may prove difficult to obtain because of the widespread occurrence of the vitamin in foods and the lack of a clear-cut deficiency syndrome in humans. For many years, efforts have been made to inhibit neoplasia by administering large amounts of vitamin E to laboratory animals, but the results of such studies have been inconclusive. However, because some of these report s indicate that vitamin E may inhibit carcinogenesi s and the mounting evidence that antioxidants may play a role in inhibiting neoplasia (Wattenberg, 1981), further investigations of the effects of vitamin E a re warranted . SELENIUM AND OTHER TRACE MINERALS Although humans require very low levels of dietary trace minerals, these micronutrients are as essential to good health as protein and energy sources. Several diseases of previously unknown etiology have been identified as trace mineral deficiencies. The prevention of goiter in the United States by supplementation of food supplies with the missing micronutrient iodine ranks among the most successful public health measures undertaken. Because minerals function in very funda- rnental biochemical processe s, including immune reactions, detoxif i- cation, and free-radical trapping, it is reasonable to postulate that they may influence mechanisms that affect the development of cancer. On the other hand, although there is an adequate understanding of the mechanisms and sites of the physiological action of the trace elements iron, iodine, zinc, copper, chromium, and selenium, there is no precise knowledge of the mechanisms that might be involved in their effects on carcinogenesis. It is also unknown whether mechanisms for the action of pharmacological doses of trace minerals are the same as or different from those of phy Biological levels that are needed f or normal nutri- tion. Furthermore, although the requirements for trace elements that are accepted as being essential for humans are well defined when protection against deficiency is used as a criterion, there is great controversy about whether higher intakes provide additional health benef it s or pose the risk of adverse effect s. The safety of trace
Vitamins, Minerals, arid, Nonnutritive Inhibitors of Carcinogenesis 39 element intake s substantially exceeding the Recommended Dietary Allow- ances (National Research Council, 1980a) has not been established. There i s very little reliable information on the trace element composition of foods, except for their iron content. But even the reported and widely used data on this element have been recognized as erroneous and are being revised by the U.S. Department of Agriculture. Data on other important trace element s, such as selenium, are sparse . Moreover, their reliability i s questionable because most have not been validated by the use of standard reference materials. In addition, since most of the analytical studies were not designed to detect regional differences, their results cannot be regarded as a reliable basis for epidemiological correlations. Finally, the rapid sequence in which trace element s with essential nutritional functions have been discovered during recent decades suggests that our pre sent knowl- edge of essential trace minerals may still not be complete (National Re search Council, 1982, Chapter 10) . The bioevailability of different chemical forms of trace elements can vary widely. For example, some selenium compounds have no bio- logical activity, wherea s others have pronounced physiological effects. The bioavailability of dietary iron compounds can differ by at least a factor of 10 (Bowering _ al., 1976~. For most micronu- trients, the scientific basis for these differences has not been deter- mined and the degree of bioavailability has not been quantif led. Food processing is known to have a strong influence on the concentration and bioavailability of most trace elements, but this influence has not been adequately quantif fed . Hundreds of nutrient-nutrient interactions have been described qualitatively, and many more may yet be discovered. These interactions are strong determinants of bioavailability and are involved in mecha- nisms that lead to the early signs of chronic toxicity. For example, large amounts of vitamin C greatly increase the bioavailability of iron compound s, while simultaneously reducing selenite to the biologically unavailable elemental selenium (Monsen et al., 1978; Newberry and Christian, 1965~. High but not excessive supplements of certain trace elements, although not necessarily toxic by themselves, can interfere with the metabolism of other elements to create secondary deficiencies Levander and Cheng, 1980) . Such ef f ect s have been clearly demon- strated for only one trace element, zinc, but they can be expected for others whenever high intakes of trace element s are maintained over prolonged period s. Selenium Selenium has two known biochemical modes of action: as a constit- uent of glutathione peroxide se, it prevents free-radical damage to
40 DIET, NUTRITION, AND CANCER: DIRECTIONS FOR RESEARCH cell constituents, and it acts as a potent antagonist of chronic and acute toxicity resulting from exposure to cadmium and mercury. Although both these modes of action may be related to reduction in the risk of cancer, it is not known whether these are the mechanisms by which selen- ium affects carcinogenesis, nor is it known at which stage of carcino- genesis selenium might be effective. There is only limited knowledge about the effect of different doses of selenium on the reduction of cancer incidence in animals. Much of the information on the element's inhibitory effects has been obtained with doses that are close to toxic levels. One study suggests that selenium-deficient diets increase the risk of cancer, compared to diets containing nutritionally adequate but not excessive levels (Ip and Sinha, 1981~. Because of the relatively narrow range between toxic and optimum levels of selenium (National Research Council, 1980a), it is essential to define a range of selenium intake that does not result in chronic toxicity but is effective in reducing cancer risk. Experiments to establish nutrient requirements have demonstrated that the amount of selenium required for optimum nutrition is dependent on dietary levels of fat, vitamin E, and certain heavy metals. There is also evidence that at least the interaction between selenium and fat is an important determinant of the element's effect on cancer. However, these inter- actions have not been quantified (Levander and Cheng, 1980~. The results of a few epide~iological studies suggest a correlation between exposure to high levels of selenium and a reduction in the risk of certain cancers (National Research Council, 1982, Chapter 10~. But these data are not conclusive, partly because the data base on the selenium content of foods is poor. Because selenium concentrations in food are dependent on the geochemical environment, dietary selenium levels vary widely from one region to another and cannot be calculated accurately on the basis of universal food composition data. Other reservations apply to blood concentrations of selenium as indicators of nutritional status, since there are no standard reference materials to safeguard analytical accuracy and comparability of results obtained by different laboratories. Populations in certain geographical areas of the world have substantially different levels of selenium intake, al- though other nutrient intakes are virtually identical. Epidemiological and/or intervention studies in such areas appear to be more promising than those in populations either with adequate selenium intakes or with relatively small differences in intake. The metabolisms of selenium depends on the chemical fore of the element. These metabolic differences may affect the impact of selenium on carcinogenesis. Very little is known about the long-ter~ conse- quences of elevated intakes of different forms of selenium, and there are no adequate means for diagnosing subclinical pathogenesis resulting from the accumulation of selenium in tissues.
Vitamins, Minerals, arid, Nonnutritive Inhibitors of Carcinogenesis 41 Molybdenum, Zinc, Iron, Iodine, and Arsenic As indicated in the committee's first report, evidence that molyb- denum, zinc, iron, iodine, and arsenic play a role in carcinogenesis is limited and, in some cases, contradictory (National Research Council, 1982, Chapter 10~. Molybdenum is essential for the function of certain enzymes. The next three elements are essential for human nutrition. For example, iron, zinc, and iodine have a profound influence on immune function or on hormonal status, or both, and deficiencies in these ele- ments reported worldwide have presented major health hazards. Arsenic is considered to be essential for growth in animals. Because of their frequent occurrence in the diet of humans, basic research is necessary to formulate definitive conclusions about the role of these trace elements in carcinogenesis. Epidemiological associations of diet with cancer rely on the adequacy with which food intake data are collected and evaluated or on the adequacy with which nutritional status is assessed by direct mea- surements in individuals. The data base for essential trace elements (exclusive of iron) is inadequate and, in part, erroneous. Expanded efforts to analyze trace elements in foods are essential in order to create a reliable and complete data base. Similarly, there are only a few reliable methods for the direct assessment of nutritional status for essential trace elements other than iron and iodine. Any progress in the development of such methods will significantly increase the reliability of epidemiological studies of diet and cancer. NONNUTRITIVE INHIBITORS OF CARCINOGENESIS An increasing number of nonnutritive substances in food have been found to inhibit carcinogenesis in laboratory animals. Included among these substances are phenols, indoles, aromatic isothiocyanates, methylated flavones, protease inhibitors, and plant sterols (National Research Council, 1982, Chapter 15~. The chemical diversity of these inhibitors suggests that other compounds with inhibitory activity are likely to exist in food, but, for the most part, efforts at identifying such substances have been haphazard (Wattenberg, 1983~. It is impor- tant to develop the technology to identify inhibitors of carcinogenesis in food and then to apply it systematically. The occurrence of cancer can be inhibited at three specific stages during the succession of events leading to development of neoplasia. These stages, in sequence, are (1) inhibition of the formation of carcinogens from precursor compounds, (2) prevention of carcinogens from reaching or reacting with critical target sites, (3) inhibition of the postinitiation stages. Some procedures have been developed for identifying inhibitors that are effective at one or more of these stages, but the scope of the procedures is limited. Those that do exist have not been extensively exploited (Wattenberg, 1983~.
42 DIET, NUTRITION, AND CANCER: DIRECTIONS FOR RESEARCH Investigations of the inhibition of carcinogen formation have been limited largely to studies of nitroso compounds in laboratory animals. Extension of such work to investigations in humans, especially in population subgroups that appear to be at high ri sk f ram exposure to nitroso carcinogens, should be explored. Techniques of this nature are discussed in Chapter 5. Food contains many compound s that prevent carcinogens f rom reach- ing or reacting with critical target site s. These so-called "blocking agents" have been identified by virtue of their ability to enhance the activities of enzyme systems that can detoxify chemical carcinogens. One such system i s the glutathione S-transferase enzyme system (Sparnins and Wattenberg, 1981; Sparnins et al., 1982~. Only limited efforts have been made to use this enzyme system to detect blocking agents in foods. Other marker systems exist as well. Research should be conducted to determine which of these systems would be effective in detecting inhibitors. Those systems should then be used for the system- atic identification of inhibitors in food. Current technology makes it feasible to identify dietary compounds that can inhibit neoplasia during the promotion phase. For example, there are three biochemical events that are associated with some facets of tumor promotion in laboratory animals. The induction of ornithine decarboxylase activity occurs in some tumor promotion systems (Boutwell, 1977), and inhibition of the induction of this enzyme's activity could be used as a parameter for identifying putative inhibitors of tumor promotion. The inhibition of free radicals that are formed during tumor promotion could be used as another method for identifying such inhibi- tors (Troll _ al., 1982~. Finally, inhibition of the stages in the arachidonic acid metabolism cascade has also been used to study inhibi- tion of promotion (Verne et al., 1980~. Accordingly, the development of appropriate methodology offers promise that putative dietary inhibi- tors of promotion can be identified. The identification of compounds that can inhibit carcinogenesis after exposure to neoplastic agents would be very important. The pro- totype compound in this category is vitamin A. The mechanism by which vitamin A and its synthetic analogs, the retinoids, bring about inhibi- tion has not been established. Pending the emergence of definitive data on the mechanism of action, it may be possible to apply some of the existing technology to detect other dietary constituents that inhibit by the same means. Inhibitors of arachidonic acid metabolism have been found to inhibit carcinogenesis when administered during the postinitiation stages. Detection of dietary constituents that inhibit components of this cascade might aid in the identification of compounds that can inhibit carcinogenesis after exposure to carcinogenic agents. GENERAL RESEARCH RE CO!IMENDAT IONS Because the data accumulated thus far suggest that further study of dietary inhibitors may be fruitful, the recommendations in this chapter are more detailed than in some other chapters.
Vitamins, Minerals, and Nonnutritive Inhibitors of Carcinogenesis 43 · Research should be conducted to de ter~i ne the labora t ory con- ditions under which the occurrence of neoplasia can be prevented by the microconstituents discussed in this chapter, i.e., vitamins A, C, and E and the carotenes; selenium and possibly some other trace min- erals; and nonnutritive inhibitors of carcinogenesis. · The mechanisms by which these microconstituents prevent car- cinogenesis should be determined. · Efforts should be made to establish dose-response curves for microconstituents (e.g., vitamin A, carotenes, and selenium) that inhibit carcinogenesis. 0 The metabolism of vitamins A, C, E, carotenes, and possibly some trace elements should be fully elucidated. · Studies should be undertaken to determine if there are biochemi- cal markers that are indicative of long-term nutritional status of humans with regard to these microconstituents. Investigations should be expanded to include a determination of the amounts of these microconstituents in various foods. These should be followed by analytical epidemiological studies (case-control or cohort studies) to determine the effects exerted by various levels of microconstituent intake on the occurrence of cancers in humans. · Consideration should be given to studying the incidence of specific cancers in populations consuming large doses of vitamin supplements. · When justified by sufficiently definitive data from experimental and/or epidemiological investigations, intervention studies with these microconstituents or with foods rich in these substances should be considered (see Chapter 4~. O Techniques should be developed for detecting dietary compounds that have the capacity to inhibit carcinogenesis. These should be applied systematically to identify such inhibitors, and when new ones are discovered, they should be subjected to each of the recommendations mentioned above. · Experiments should be conducted in animals to evaluate the effects of dietary microconstituents on later stages of carcinogenesis. SPECIFIC RESEARCH RECOMMENDATIONS Chapters 4 and 5 contain general suggestions for improving both epi- demiological and laboratory methods. Recommendations pertaining spe- cifically to microconstituents and inhibitors are presented below.
44 DIET, NUTRITION, AND CANCER: DIRECTIONS FOR RESEARCH Vitami n A and Carotene s · Case-control or cohort studies are needed to distinguish between the effects of retinal and those of ,8-carotene as they pertain to cancer ri sk, since the dietary sources of these two nutrients are dif- f Brent . · Studies to examine the potential risk-reducing effect of ''vita- min A" and the retinoids would be worthwhile. For example, studies could be conducted on certain high-risk groups, such as asbestos workers who smoke. Intervention trials on this subject may be produc- t ive. · Experiments in several different animal models should be con- ducted to determine the inhibitory effect of different doses of vita- min A. · Studies to determine the mechanists) of inhibition for vitamin A and the retinoids should be pursued. · Additional experiments should be conducted in animals to deter- mine if ~ -carotene and other carotene s have the capacity to inhibit carcinogenesis and, if so, to identify the conditions under which such inhibition occurs and the mechanism by which it occurs. · Improved techniques for evaluating vitamin A and carotene levels in human tissues would be helpful, as would be studies of the metabo- lism of vitamin A and carotenes. Vitamin C ~ Ascorbic Acid) · The effects of vitamin C on chemically induced neoplasia of the large bowel should be studied further in laboratory animals, especially in the guinea pig. · Further studies should be pursued to determine if vitamin C has broadly applicable inhibitory effects on neoplastic manifestations of cells in culture. o Epidemiological studies should be conducted to assess more directly the possible inhibitory effect s of vitamin C on the induction of gastric cancer and possibly cancer of other sites in the gastrointes- tinal tract, e.g., the colorectum. Vi tamin E o S tudies should be conducted in laboratory animals to determine the effects of vitamin E on chemically induced neoplasia of the large bowel and the breast.
Vitamins, Minerals, and Nonnutritive Inhibitors of Carcinogenesis 45 o Epidemiological studies should be conducted to examine the relationship of vitamin E to cancer risk in humans. Selenium and Other Trace Elements Selenium · Basic studies should be conducted to determine the effect of selenium on immune reactions, its interactions with DNA, its interac- tions with heavy metals, and the mechanisms by which it protects tissues against attack by free radicals. · The stage of carcinogenesis at which selenium is most effective should be determined. · A complete dose-response curve for the protective effect of selenium against chemical and viral carcinogens should be established, and the influence of dietary fats, heavy metals, and vitamin E on such curves should be quantified. · Better methods to analyze the selenium content of foods should be developed, and the results of the subsequent analyses should be con- solidated into a data bank and validated. This effort must be under- taken for each different geochemical region. O Valid methods should be developed to assess the nutritional status of humans with regard to selenium. This would require that standard reference materials (e.g., foods, blood, and urine) be developed for selenium. · Cohort studies based on biochemical assays of selenium are needed to corroborate observations from correlation studies. Epidemiological studies of selenium and cancer will need to take into account the interactions of selenium with other dietary constituents (e.g., vitamin C, copper, and zinc), since these interactions can affect the bioavaila- bility of selenium. · Studies in humans should be conducted under close medical super- vision to monitor the metabolic effects of different forms of selenium supplements used in moderate amounts. · Ultimately, consideration should be given to intervention studies in countries known to have inadequate selenium intakes, e.g., China, New Zealand, and Finland. Molybdenum, Zinc, Iron, Iodine, and Arsenic o Initially, research on molybdenum, zinc' iron> and iodine should be directed toward confirming or disproving their carcinogenicity when administered orally.
46 DIET, NUTRITION, AND CANCER: DIRECTIONS FOR RESEARCH · If justified by findings from the initial research, basic research should be pursued to elucidate the mechanisms by which molyb- denuqn, iron, iodine, and zinc might affect carcinogenesis. · Analytical methods should be developed to determine snore accu- rately the molybdenum, iron, iodine, and zinc content of foods. · Valid 'methods should be developed to assess the nutritional status of individuals in regard to molybdenum, zinc, iron, iodine, and arsenic. Analytical epide~iological studies should be conducted to examine the possible carcinogenic effects of exposure to the low levels of zinc, iodine, and arsenic that are present in the average diet. The relationship between exposure to iodine and thyroid cancer should be examined in case-control studies that clearly separate the follicular and papillary histological types of tumors. Nonnutritive Inhibitors of Carcinogenesis · Studies should be conducted to detect dietary constituents that have the capacity to prevent the occurrence of neoplasia. · Studies should be conducted in animals to determine the condi- tions under which these compounds will inhibit carcinogenesis and to identify characteristics that are related to their potential for pre- venting neoplasia. Their mechani sms of inhibition and their adverse effects, if any, also need to be identified. 1 · Short-term studies should be conducted in humans to identify the protective responses elicited by these compounds. Epidemiologica1 studie s, including intervention trial s when appropriate, should be conducted to determine if consumption of foods containing high concentrations of these compounds results in a lower incidence of cancer ~ see also Chapter 4) .
