Nutritional Strategies for Cancer Prevention

Converging evidence indicates that diet contributes to the etiology of about 1/3 of fatal cases of human cancer in the developed countries. Although incomplete, the existing knowledge provides the foundation for an effective policy of primary cancer prevention through diet modification. A prudent diet that reduces cancer should be high in vegetables, fruits, legumes and cereals, and low in red meat, salt and saturated fat of animal origin. Added lipids should be of plant origin and, among them, olive oil has a safety record of several thousand years. Obesity should be avoided mainly by increasing physical activity. Regular physical activity during childhood and adolescence may also slow down excessive growth, as reflected in attained height, and have beneficial consequences on several cancer types. Concerning alcoholic beverages, a strategy that could preserve most of the health benefits of alcohol intake and could reduce the alcohol-related cancer mortality would include: for everyone, reduction of heavy alcohol consumption; for smokers limitation to alcoholic beverages that are taken with meals (mainly wine); for nonsmoking men encouragement of moderate drinking (up to 2 glasses per day); for nonsmoking women encouragement of light drinking (one glass per day), unless they are at high risk of developing breast cancer. J Epidemiol,1996 ; 6 : S111-S115.


Strategies for Cancer Prevention Dimitrios Trichopoulos1 and Pagona Lagiou2
Converging evidence indicates that diet contributes to the etiology of about 1/3 of fatal cases of human cancer in the developed countries.Although incomplete, the existing knowledge provides the foundation for an effective policy of primary cancer prevention through diet modification.A prudent diet that reduces cancer should be high in vegetables, fruits, legumes and cereals, and low in red meat, salt and saturated fat of animal origin.Added lipids should be of plant origin and, among them, olive oil has a safety record of several thousand years.Obesity should be avoided mainly by increasing physical activity.Regular physical activity during childhood and adolescence may also slow down excessive growth, as reflected in attained height, and have beneficial consequences on several cancer types.Concerning alcoholic beverages, a strategy that could preserve most of the health benefits of alcohol intake and could reduce the alcohol-related cancer mortality would include: for everyone, reduction of heavy alcohol consumption; for smokers limitation to alcoholic beverages that are taken with meals (mainly wine); for nonsmoking men encouragement of moderate drinking (up to 2 glasses per day); for nonsmoking women encouragement of light drinking (one glass per day), unless they are at high risk of developing breast cancer.
nutrition, diet, cancer prevention, cancer etiology, risk factors A consensus has gradually developed that diet contributes to the etiology of close to one third of fatal cases of human cancer, at least in the developed world.Five lines of evidence support this conclusion: i) experimental investigations in laboratory animals have documented the role of quantitative and qualitative aspects of nutrition in the etiology of several forms of cancer in several animal species; ii) studies in migrants highlighted the role of microenvironment, including diet, in the etiology of such variable forms of human cancer as those of the large bowel, the stomach, the breast and the prostate; iii) ecological correlations of several types of cancer on the one hand and various aspects of nutrition on the other have demonstrated striking concomitant variability; iv) analytic epidemiologic studies have traced the nutritional origin of several common types of human cancer; and v) metabolic and molecular inves-tigations have provided powerful corroborating evidence.The emerging central question is whether the existing knowledge is sufficient to allow the formulation of an effective nutritional strategy for cancer prevention in humans.

STRATEGIES FOR CANCER PREVENTION
Prevention of cancer may be accomplished through primary or secondary prevention strategies.The objective of primary prevention is to prohibit the damage of susceptible molecules in the human body and is implemented in two ways: i) by avoidance or abatement of a carcinogenic exposure as, for example, by non smoking, or switching from high to low tar cigarettes, and ii) by fortifying the body's defenses through vaccination, dietary modification, chemoprevention, or other National Nutrition Center, Athens School of Public Health, Athens, Greece.Primary prevention is superior to secondary prevention and clinical therapy, even when the three approaches are equally effective in reducing mortality.This is because primary prevention is not accompanied by the anguish that characterizes the diagnosis and treatment of even presymptomatic cancer, and because failure of primary prevention leaves intact the two other defense lines.

OVERVIEW OF THE NUTRITIONAL ETIOLOGY OF CANCER
The effect of diet on cancer occurrence is poorly understood in mechanistic terms, but several patterns have emerged (Table 1).Vegetable and fruit consumption provide some protection against several types of human cancer.Red meat is closely linked to the occurrence of colorectal cancer, and saturated fat of animal origin to the occurrence of prostate cancer and probably colorectal cancer.Excess energy intake in early life, possibly reflecting reduced physical activity, may be responsible for the repeatedly demonstrated positive association between Table 1.Risk implications for major forms of cancer by consumption of foods in major groups, intake of energy-generating nutrients, intentional dietary exposure to selected non-nutrients, and nutrition-related indicators.
height on the one hand and risk for breast cancer and possibly other cancer types on the other.Physical activity in adult life reduces the risk of colorectal cancer, possibly breast cancer, and conceivably other cancers as well.Obesity in adult life, due to either excess energy intake or reduced expenditure, is an important cause of endometrial cancer, an established cause of postmenopausal breast cancer and a likely cause of cancers of the kidney and gallbladder.There is also evidence that intrauterine growth may be positively associated with some cancer types in the offspring, including breast and prostate cancer.
Among non-nutrient food additives that have been in use, salt appears to be an important contributor to stomach cancer in some populations.Moreover, intake of salty fish very early in life has been linked to the occurrence of nasopharyngeal cancer in Southeast Asia.The intake of very hot drinks has been shown to increase the risk of esophageal cancer in Central Asia and South America.
Consumption of large quantities of alcoholic beverages, particularly in conjunction with tobacco smoking, sharply increases the risk of cancer in the upper respiratory and digestive tract, whereas alcoholic cirrhosis frequently leads to liver cancer.Converging data suggest that intake of smaller quantities of alcohol may be linked to the occurrence of breast and colorectal cancer.

PREVENTION OF DIET RELATED CANCER
Although we know little about the beneficiary or harmful food constituents, we have a very good idea of what a prudent diet that reduces cancer should be.This diet should be high in vegetables, fruits, legumes and cereals; low in red meat and in salt; and low in animal origin saturated fat.Added lipids should be of plant origin and, among them, olive oil has a safety record of several thousand years.Obesity should be avoided, but the best way to accomplish this would be by increasing physical activity which, in itself, can reduce the incidence of colorectal cancer and, perhaps, other types of cancer as well.Regular physical activity during childhood and adolescence may also slow down excessive growth, as reflected in attained height, and have beneficial consequences on several cancer types.
Alcoholic beverages are integral components of the communal life of many societies and their intake reduces cardiovascular mortality.Yet alcohol interacts with tobacco in the causation of cancer in the upper respiratory and gastrointestinal tract, it can cause liver cirrhosis-mediated hepatocellular cancer and may also cause cancer of the breast and, possibly, the large bowel.A strategy that could preserve most of the health benefits of alcohol intake and could reduce the alcohol-related cancer mortality would include: for everyone, reduction of heavy alcohol consumption; for smokers who cannot quit smoking, limitation to alcoholic beverages that are taken with meals (mainly wine) and, thus, are less likely to damage the mucosa and facilitate the carcinogenic action of tobacco constituents; for nonsmoking men encouragement of moderate drinking (up to 2 glasses per day); for nonsmoking women at high risk for breast cancer avoidance of drinking, whereas for nonsmoking women at low risk for this disease or at high cardiovascular risk encouragement of light drinking (one glass per day); for every drinker encouragement of high consumption of green vegetables that contain folate, the deficit of which may increase the potential of alcohol to increase the risk of colorectal cancer.

QUANTITATIVE ESTIMATES
The Harvard Center for Cancer Prevention has estimated that diet in adult life, including obesity, is involved in the etiology of about 30% of cancer mortality in the developed countries, whereas sedentary life per se contributes to an additional 3% of this mortality.Food additives, mostly salt, and very hot drinks may contribute to a further 1%.Perinatal influences and, more important, excessive growth reflecting energy imbalance in early life could be a component cause in a substantial proportion, perhaps 5%, of frequently fatal cancers of the breast and prostate.Finally, alcoholic beverages have been estimated to contribute to about 3% of total cancer mortality in the developed world (Table 2).Diet, physical activity and obesity are quantitative variables and there are in practice no categories of zero exposure.Moreover, absolute measurements are difficult to obtain because most measurements depend on study-specific design, available databases and analytic approach.Usually, individuals are distributed in quintiles, i.e. in 20% -wide categories in the respective frequency distributions.Occasionally, a score is created to integrate the various components of diet.It does not appear unrealistic to expect that individuals in all quintiles except the best ones could be motivated to improve their diets and increase their physical activity so as to move to the adjacent "better" quintile.If this were to happen, both the dietrelated and the sedentary life-related cancer mortality would be reduced by about 1/4.Reduction of salt intake could, in theory, accelerate the declining incidence of stomach cancer.Lastly, the increasing popularity of wine and the decreasing popularity of spirits could bring about a reduction of alcohol-related cancer mortality by about 1/3 (Table 3).

THE ROLE OF CHEMOPREVENTION
In the context of nutritional intervention, chemoprevention relies on the identification of specific macro-or micronutrients and documentation of their effectiveness through randomized Table 3. Realistic objectives for reduction of cancer mortality from specific causes or groups of causes in developed countries during the next few decades, on the basis of existing knowledge and technology.
Source (4) trials.Although randomization represents the gold standard in empirical research and the specification of nutrients, rather than foods or dietary patterns, projects scientific sophistication, this approach is unlikely to be very helpful in the immediate future.There are three reasons behind this somber assessment: i) incrimination of nutrients on the basis of epidemiological evidence derived from studies ascertaining dietary intakes is constrained by the available nutritional databases which in turn rely on analytic methodologies of unequal reliability; ii) randomized trials do not last long enough to address nutritional influences surrounding the initiation and early promotion stages of carcinogenesis; and iii) contrary to chemotherapeutic clinical trials, chemoprevention field trials must be very large, since rates of clinical outcomes are usually substantially higher than incidence rates of virtually all types of cancer in most population groups. CONCLUSION Address for correspondence : dimitrios Trichopoulos, Harvard Center for Cancer Prevention and Department of Epidemiology Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115 U.S.A. and P.Lagiou means, Secondary prevention relies on presymptomatic disease detection and may be implemented through individual checkups or through population screening.

Table 2 .
Cancer mortality attributable to specific factors or groups of factors in the developed countries.