Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

3.0008 Biologically active nonnutrients may modulate the

initiation or progression of carcinogenesis at vari-

ous stages.

Carcinogens and Promoters

0009 The Swiss toxicologists Lutz and Schlatter analyzed

the contribution of various dietary factors to the risk

of human cancer. They compared human exposure to

known carcinogens with the activity of these sub-

stances in animal models, and concluded that only a

small proportion of the human cancer cases attrib-

uted to diet by Doll and Peto were likely to be caused

by foodborne carcinogens. Certainly there is very

little evidence that food additives and pesticide

residues impose any carcinogenic risks on the general

public in industrialized societies. Food additives are

rigorously screened to eliminate any risk of carcino-

genic effects, and synthetic pesticide residues are no

more likely to be mutagenic than natural plant con-

stituents. In contrast, food supplies in developing

countries are often at risk of contamination with

carcinogenic toxins derived from fungi. Aflatoxin, a

product of the fungus Aspergillus flavus, frequently

contaminates groundnuts stored in hot climates.

Dietary exposure to aflatoxin is associated with in-

creased risk of hepatocellular carcinoma, a form of

liver cancer. Food supplies in western countries are

screened for the presence of aflatoxins, but socioeco-

nomic conditions make this difficult to achieve in

developing countries where locally produced ground-

nuts are a dietary staple. Another potent class of

carcinogens known to be present in foods are the

nitrosamines. The best established example of a prob-

able link between these compounds and disease is

the high risk of nasopharyngeal cancers in southern

Chinese populations consuming high levels of salted

fish. In Japan, a high intake of various pickled and

highly salted vegetable and meat products is believ-

ed to be one cause of the very high incidence of gastric

cancer.

0010 Epidemiological evidence derived from both case-

control and cohort studies suggests that regular con-

sumption of more than about 80 g of red meat

(beef, pork, or lamb) per day leads to an approxi-

mately 2.5-fold increase in the risk of colorectal

cancer. Around half of these studies show a statistic-

ally significant effect. The strongest evidence of an

adverse effect comes from North America. However

at the time of writing, the European Prospective In-

vestigation of Cancer and Nutrition (EPIC) has begun

to publish its first findings, and preliminary reports

suggest a moderate adverse effect of red meat con-

sumption on colorectal cancer risk, and a somewhat

stronger effect of processed meat products. Cooked

meat contains carcinogens, particularly heterocyclic

amines, but at rather low levels. A second hypothesis

is that heavy meat consumption increases the forma-

tion of nitrosamines by colonic bacteria. It is not

entirely clear why processed meat should entail a

higher risk, but one possibility is that the iron-

containing protein heme is the risk factor, and

that processed meats contain a higher concentration.

In any case meat consumption is only one among

many nutritional factors that influence colorectal

carcinogenesis.

0011Alcohol is not a genotoxic carcinogen but there is

good evidence that even moderate consumption in-

creases the risk of breast cancer. Heavy consumption

is an important cause of oropharyngeal, esophageal,

laryngeal, and hepatic cancers. The risk is dose-

related but also strongly linked to smoking.

Nutritional Factors

0012A large proportion of human diet-related cancers

appear to be due to the effects of overconsumption

of energy. Evidence that restricting energy intake

would inhibit tumor growth in experimental animals

first began to appear in the early part of the twentieth

century. More recent studies have confirmed the rela-

tionship, and shown that increased physical exercise,

leading to greater energy expenditure, has a similar

effect. There are several plausible mechanisms of

action, including a suppression of endogenous DNA

damage caused by reactive oxygen species, enhanced

DNA repair, and suppression of tumor growth, medi-

ated by insulin and other endocrine factors. Direct

evidence that similar mechanisms are at work in

humans is difficult to achieve, but restricted energy

intakes may well contribute to the reduced risk of

colorectal cancer observed in agrarian communities

of Asia and Africa. It is difficult to see how deliberate

restriction of energy intake could be made to work as

a cancer prevention strategy in the west, but regular

exercise and avoidance of obesity are likely to be

beneficial. An understanding of the mechanisms in-

volved may eventually lead to other biological strat-

egies for suppression of carcinogenesis.

0013It has long been thought that dietary fat may have

an effect on the risk of cancer, independently of any

effect of energy intake. International comparisons

suggest positive links between fat intake and breast

and colon cancer, and there is evidence that total fat

intake is positively related to induction of tumors by

chemical carcinogens in animals. However, the evi-

dence is much less convincing when comparisons are

made within populations, and there is no clear scien-

tific consensus that total fat intake is an important

risk factor. There is stronger experimental evidence to

show that the fatty acid composition of the diet is

806 CANCER/Diet in Cancer Prevention

important. Polyunsaturated fatty acids (PUFA) of the

n-6 series tend to promote mammary carcinogenesis

in experimental animals, whereas n-3 PUFA such as

eicosapentaenoic acid, which is found in fish oils,

tend to suppress mammary and colorectal carcino-

genesis. Epidemiological evidence has begun to

emerge for protective effects of fish consumption

against cancer of the colon, breast, and prostate, but

these findings require validation.

0014 Micronutrient status is another aspect of general

nutrition that has received a great deal of attention

in relation to several types of cancer. Small quantities

of many minerals and organic micronutrients are

essential to normal growth and function. Substances

for which a specific deficiency condition has been

identified are classified as nutrients, but many have

biological effects beyond their classical function.

Vitamin C (ascorbic acid) is one example. Ascorbic

acid is essential for the prevention of scurvy in

humans but it is also a strong antioxidant and numer-

ous studies have shown that a relatively low level of

ascorbate in the serum is a risk factor for several

forms of cancer. However the evidence that supple-

mentation with vitamin C reduces the risk of cancer is

much less convincing. It is possible that plasma vita-

min C is a biomarker for a high intake of fruit and

vegetables. Similar arguments apply to a number of

other antioxidant nutrients, including b-carotene,

but strong protective effects against several types of

cancer have been demonstrated in a human interven-

tion study with selenium.

0015 Low intakes of folate, one of the B-group vitamins,

have been linked to increased risk of cancers of the

colon, pancreas, lung, and breast. Individuals with

one or more common inherited abnormalities of

methylenetetrahydrofolate reductase, one of the prin-

cipal enzymes involved in folate metabolism, may be

at higher risk. The relationship is biologically plaus-

ible because folate is essential for normal synthesis of

DNA, and for other aspects of cell proliferation.

Folate deficiency causes misincorporation of nucleo-

tides and DNA breaks similar to those caused by

ionizing radiation.

0016 Of all human cancers, squamous cell carcinoma of

the esophagus seems the most likely to be caused, at

least partially, by one or more deficiencies of known

micronutrients. Populations characterized by a very

high incidence of the disease often consume restricted

diets, with low intakes of protein and many micro-

nutrients, including riboflavin, niacin, vitamins C and

E, b-carotene, and retinal. It seems probable that in

high-risk communities, specific nutrient deficiencies

act synergistically with other factors, such as specific

foodborne carcinogens or the irritating effects of very

hot or abrasive foods and drink.

Dietary fiber

0017The dietary fiber hypothesis has complex origins, but

its best-known advocate was the colonial medical offi-

cer Dennis Burkitt, who sought to explain the rarity of

colorectal carcinoma and other bowel disorders

amongst rural Africans. Burkitt proposed that the vari-

ous nondigestible polysaccharides of plant cell walls,

collectively termed dietary fiber, are protective against

cancer of the large bowel. His original hypothesis was

based largely on the concept of fecal bulk. Working

from clinical observations, Burkitt deduced that popu-

lations consuming traditional rural diets, rich in vege-

tables and cereal foods, produced frequent bulky

stools. The increased risk of colorectal cancer in west-

ern populations eating low-fiber diets was attributed to

a low volume of feces and infrequent bowel move-

ments, causing prolonged exposure of the colonic epi-

thelial cells to mutagens. The mildly laxative effects of

dietary fiber are now well recognized, and numerous

human intervention trials have confirmed that dietary

supplements containing wheat bran or other types of

dietary fiber can increase fecal volume and reduce tran-

sit times. Reports from such bodies as the World

Cancer Research Fund, the European Cancer Preven-

tion Organization, and the UK Department of Health

have agreed that the evidence for a protective effect of

dietary fiber against colorectal cancer is at least moder-

ately convincing. The fecal-bulking hypothesis pro-

vides the principal rationale for the current dietary

reference values for nonstarch polysaccharides in the

UK, which recommend that adults should consume an

average of 18 g of fiber per day.

0018Apart from the simple dilution effect of increased

fecal bulk, the nonfermentable particulate compon-

ents of plant cell walls may sequester bile acids, thus

reducing their concentration in the aqueous phase of

the fecal stream. This hypothesis is consistent with the

results of human intervention studies with high-fiber

low-fat diets, which have been shown to reduce fecal

bile acids and slow the appearance of large adenomas.

0019In the three decades or so that have elapsed since the

original work of Burkitt, great progress has been made

in our understanding of the gradual transition from a

normal colonic epithelium to a malignant tumor.

Apoptosis is thought to provide a defense mechanism

against the appearance and survival of cells carrying

unrepaired mutations, and it is probably also import-

ant at a later stage in the adenoma–carcinoma sequence

because the speed at which tumors grow depends upon

their relative rates of cell production and death. The

colonic microflora is rich in bacteria that can break

down carbohydrates to yield the short-chain fatty

acids acetate, propionate, and butyrate, and these are

absorbed and metabolized by human tissues. Butyrate

CANCER/Diet in Cancer Prevention 807

is utilized preferentially by the colonic mucosa, and

besides supplying metabolic energy, it suppresses pro-

liferation, increases differentiation, and induces apop-

tosis of tumor cells in vitro. Thus the principal role of

fermentable carbohydrate may be to regulate mucosal

cell turnover and death by providing a constant en-

dogenous supply of butyrate.

0020 The dietary fiber hypothesis has never been dis-

proved. Some large North American cohort studies

have failed to confirm an inverse relationship between

fiber intake and colorectal cancer, but it is doubtful

whether the highest levels of fiber consumption in

such studies were great enough to provide the protect-

ive effects seen in international comparisons. Prelimin-

ary reports from the EPIC study do provide some

evidence of beneficial effects of dietary fiber in Europe,

where the range of fiber intakes may be greater.

0021 The colon is perhaps the most obvious site for fiber

to exert anticarcinogenic effects, but there is evidence

that it protects against other cancers. One analysis of

a series of case-control studies on breast cancer and

diet has suggested that a 20-g increase in daily fiber

intake was associated with a statistically significant

reduction in risk of breast cancer of about 15%. Diets

containing this rather high level of fiber may be asso-

ciated with low levels of body fat, reduced plasma

estrogen levels, and improved insulin sensitivity, all

of which might inhibit the growth of hormone-

dependent tumors. Furthermore, fiber-rich plant

foods contain a variety of other biologically active

plant cell constituents, including antioxidants and

phytoestrogens, which probably exert a variety of

anticarcinogenic effects in their own right.

Phytochemicals

0022 The strongest association between dietary factors and

cancer in industrialized societies is a protective effect

of high intakes of fruit and vegetables. The phenom-

enon extends to most carcinomas (cancers of epithe-

lial tissues) but is not consistently observed for

leukemia, nor for cancers of lymphoid or neural

tissues. Fruits and vegetables are rich in a variety of

important micronutrients, including antioxidants and

folate, and they probably contribute to the protective

effects of these foods (Table 2). However, studies with

experimental animals have shown that plant tissues

also contain a variety of nonnutrient phytochemicals

that can act as blocking agents, which interfere with

the mutagenic effects of carcinogens at target cells,

or suppressing agents, which slow or reverse the de-

velopment of precancerous lesions or tumors, after

the initial mutagenic effect of the carcinogen has

occurred. Some compounds can act in both ways.

0023 Although much of the epidemiological data is rela-

tively nonspecific in that it simply indicates that a

variety of vegetables and fruits are protective against

an equally diverse range of cancers, there is also a

growing body of evidence relating specific plant

foods, and hence specific types of chemical com-

pound, to particular disease processes. Evidence of

this type forms the starting point for focused mechan-

istic studies. Some examples of potentially anticarci-

nogenic phytochemicals and their principal sources

are given in Table 1. A thorough discussion of these

compounds cannot be given here but one group, the

glucosinolates, will be discussed in some detail.

0024Many cohort and case-control studies indicate

that a high consumption of brassica vegetables is

associated with a decreased risk of carcinomas of

the lung, stomach, colon, and rectum. Probably the

most important common characteristic of cabbages,

broccoli, cauliflower, and Brussels sprouts is that they

all contain glucosinolates, a large family of sulfur

compounds linked to glucose, and bearing a side

chain derived from various amino acids. Following

tissue damage caused by pests, harvesting, food pro-

cessing, or chewing, glucosinolates are broken down

by the endogenous enzyme myrosinase. Hydrolysis of

the glucosidic bond releases glucose and an unstable

intermediate, which spontaneously degrades to form

a variety of breakdown products, the most important

of which are the isothiocyanates. These hot and bitter

compounds, commonly termed ‘‘mustard oils,’’ pro-

vide much of the flavor of brassica vegetables.

0025Isothiocyanates such as phenethyl isothiocyanate,

benzyl isothiocyanate, and sulforaphane modify the

balance of phase I and II xenobiotic-metabolizing

enzymes that are expressed in liver, and in epithelial

cells including those of the colon. Phase I enzymes

such as the cytochrome p450 family metabolize pro-

carcinogens, often converting them to highly carcino-

genic products in the process. Phase II enzymes such

as the glutathione transferase family (GST) convert

these products to inactive, water-soluble conjugates,

readily excreted in urine. In one animal study in

which a carcinogenic nitrosamine and phenethyl iso-

thiocyanate (a constituent of watercress) were admin-

istered simultaneously, lung tumors were induced in

tbl0002Table 2 Some examples of potentially anticarcinogenic phyto-

chemicals

Class of compound Dietary source

Isothiocyanates Brassica vegetables and condiments

Flavonoids Apples, onions, broccoli, tea infusions,

chocolate

Isoflavonoids Soy beans, soy products

Carotenoids Tomatoes, carrots and other colored

vegetables

Protease inhibitors Legumes

808 CANCER/Diet in Cancer Prevention

70% of the control rats given only the carcinogen, but

only 5% of those cotreated with isothiocyanate.

0026 One problem with such animal models is that the

concentrations of both carcinogens and anticarcino-

gens used are often unrealistically high. However

there is also evidence from epidemiological studies

to show that this work has real implications for

human health. In a recent study by London et al. the

excretion of isothiocyanates in urine was used as a

marker of brassica vegetable consumption in a large

cohort of Chinese men. The presence of inherited

defects in two phase II enzymes, GSTM1 and

GSTT1, was recorded at the same time as the urinary

analysis, and the cohort was followed up for 10 years.

Individuals with detectable levels of isothiocyanate

metabolites in urine were found to be at a reduced

risk of cancer, but the effect was only observable in

subjects who lacked GSTM1 or GSTT1, and

strongest in those with deletion of both enzymes.

The GST enzymes play a major role in the detoxifi-

cation of carcinogens, but they also break down the

isothiocyanates themselves. These observations pro-

vide evidence of a very complex interplay between

environmental factors, genotype, and risk of cancer.

The protective mechanisms may include suppression

of the carcinogenic process by induction of apoptosis.

Like butyrate, glucosinolate breakdown products

block the cell cycle in vitro, and induce apoptosis in

the rat colon after treatment with a chemical carcino-

gen. It is interesting to note that a protective effect of

broccoli consumption against precancerous adeno-

matous polyps of the colon has been observed in

individuals with the GSTM1 null genotype.

Conclusion

0027 Clearly there is still a huge disparity between the

proportion of human cancers that are theoretically

avoidable, and the small number of well-established

dietary strategies for actually reducing the risk of

cancer in practice. By far the most important cause

of human cancer is cigarette smoke. Its avoidance is

the surest way for individuals to minimize their risk

of cancer. Nevertheless, there are subtle interactions

between dietary components, genetic profile, and ex-

posure to carcinogens that shape the individual’s risk

of disease to an important degree. Understanding

such mechanisms in greater depth is the key to har-

nessing the protective effects of diet more effectively

in the future. The current rapid developments in gen-

etics and molecular biology will do much to facilitate

this process. In the meantime, the best advice avail-

able is to eat large quantities of fruit and vegetables.

The current recommendation in many countries is to

eat five portions (80 g) of leafy vegetables and fruits

per day from a variety of sources. This is already well

in excess of the average intake in the UK, but even

higher quantities may be of additional benefit. The

bulk of such a diet, coupled with quantities of com-

plex carbohydrates sufficient to provide around 18 g

of dietary fiber per day, will ensure a relatively low

energy density. If combined with regular exercise, this

should help to maintain body weight within the desir-

able range. The most recent evidence suggests that

moderating the consumption of red and processed

meat products in favor of oily fish and poultry may

also help to reduce the risk of bowel cancer. Alcohol

consumption should be maintained within current

Department of Health guidelines. There is no convin-

cing evidence that very large doses of any particular

vitamin are protective against cancer, but in view of

the evidence linking suboptimal micronutrient status

to some forms of carcinoma, consumption of a daily

multivitamin supplement may be prudent.

See also: Aflatoxins; Alcohol: Properties and

Determination; Ascorbic Acid: Properties and

Determination; Cancer: Carcinogens in the Food Chain;

Carcinogens: Carcinogenic Substances in Food:

Mechanisms; Carcinogenicity Tests; Folic Acid:

Properties and Determination; Nitrosamines; Functional

Foods; Fatty Acids: Tr a n s -fatty Acids: Health Effects

Further Reading

Ames BN and Gold LS (1998) The prevention of cancer.

Drug Metabolism Reviews 30: 201–223.

Doll R (1996) Nature and nurture: possibilities for cancer

control. Carcinogenesis 17: 177–184.

Doll R and Peto R (1981) The causes of cancer: quantitative

estimates of avoidable risks of cancer in the United

States today. Journal of the National Cancer Institute

66: 1191–1308.

Johnson IT and Fenwick GR (2000) Dietary Anticarcino-

gens and Antimutagens: Chemical and Biological

Aspects. Cambridge: Royal Society of Chemistry.

Johnson IT, Williamson GM and Musk SRR (1994) Anti-

carcinogenic factors in plant foods: a new class of nutri-

ents? Nutrition Research Reviews 7: 175–204.

London S, Yuan JM, Chung FL et al. (2000) Isothiocya-

nates, glutathione S-transferase M1 and T1 polymorph-

isms, and lung-cancer risk: a prospective study of men in

Shanghai, China. Lancet 356: 724–729.

Lutz WK (1999) Carcinogens in the diet vs. overnutrition.

Individual dietary habits, malnutrition and genetic sus-

ceptibility modify carcinogenic potency and cancer risk.

Mutation Research Reviews 443: 251–258.

Lutz WK and Schlatter J (1992) Chemical carcinogens and

overnutrition in diet-related cancer. Carcinogenesis 13:

2211–2216.

Willett WC (2001) Diet and cancer: one view at the start of

the millennium. Cancer Epidemiology, Biomarkers and

Prevention 10: 3–8.

CANCER/Diet in Cancer Prevention 809

Diet in Cancer Treatment

J W T Dickerson, University of Surrey, Guildford,

Surrey, UK

Introduction

0001 The nutritional requirements of persons with cancer,

and hence the diet required to meet them, may be

affected by the body’s response to the disease itself, a

condition known as ‘cancer cachexia,’ and by the

methods used to treat the disease. Assessment of the

factors involved and their interrelationships can help

towards the adequate nutrition of cancer patients and

the maintenance of their maximum possible quality of

life. Furthermore, it is now becoming evident that some

food components may be able to modify the cachectic

process, whilst others may have a direct effect on some

kinds of cancer cells. This latter effect raises the possi-

bility that diets may be used in the future to treat, or at

least to control the growth of some kinds of cancer.

Cancer Cachexia

0002 Weight loss occurs in many but not all patients with

cancer. Thus, whilst patients with cancer of the gas-

trointestinal tract may become progressively wasted,

those with breast cancer may maintain or even in-

crease their body weight. Variability in weight change

both within the same patient and between different

patients has been reported in patients with lung

cancer. In some patients, weight loss may be the first

indication of the presence of a tumor. This is particu-

larly so in patients with gastrointestinal cancer in

whom protein-energy malnutrition (PEM) may de-

velop somewhat insidiously owing to a decrease in

food intake, though this may be denied by the patient.

0003 In healthy individuals the body adapts to starva-

tion by conserving protein and reducing energy

expenditure. In many patients with cancer these adap-

tations do not adequately occur and a wide spectrum of

changes in nutrient metabolism occur, similar to those

found in the so-called ‘metabolic response to trauma.’

Some of these changes are mediated by cytokines re-

leased by macrophages. At least four of these have been

described: tumor necrosis factor-a, interleukins-1 and -

6 (IL-1 and IL-6 respectively) and interferon-g.Inadd-

ition, tumors themselves elaborate catabolic factors

which cause direct breakdown of host lipid and skeletal

muscle protein stores with the production of acute-

phase proteins, such as C-reactive protein. The cach-

exia syndrome is characterized by anorexia, early sati-

ety, changes in taste perception, weight loss, weakness,

anemia, and edema.

Anorexia

0004A failure of appetite is the most important cause of a

low food intake. The etiology of anorexia is, however,

often complex with a number of interacting etio-

logical factors, rather than one single factor.

0005The factors responsible for anorexia in the cancer

patient may be psychological, physical, and metabolic.

Moreover, the anorexia may be one of three types:

0006Transient anorexia is caused by emotional distress

such as during a diagnostic work-up, or when

recurrence or metastases are diagnosed, or when

pain or discouragement is experienced.

0007Iatrogenic anorexia is related to treatment by

surgery, the result of radiation sickness, or a side-

effect of chemotherapy.

0008Pathological anorexia is related to the disease,

particularly gastrointestinal or bronchial cancers,

or part of the anorexia–cachexia syndrome of

advanced disease.

The importance of psychological factors may de-

crease with the progress of the disease as physical

factors become more prominent; nevertheless, they

may make a substantial contribution to appetite fail-

ure. Feelings of worthlessness, loss of self-esteem,

pessimism, guilt, and suicidal ideas may be experi-

enced. Anorexia and cachexia may be the physical

result of the patient’s own beliefs and attitudes

towards the disease and its treatment. Patients may

become quite unable to overcome a sense of fear and

hopelessness, leading to withdrawal.

0009We know little about the pathology of disturbed

appetite control in cancer patients. It is part of

the cachexia syndrome and it seems certain that

the release of cytokines and associated acute-phase

proteins is involved but the mechanism of their action

is not clear. In hospitalized cancer patients a low food

intake has been found to be significantly associated

with a lower quality of life, but whether by cause or

effect is not known. Consuming insufficient food to

meet energy requirements leads to weight loss and a

low body mass index (BMI). In the community,

cancer patients with a BMI below 20 kg m

2

have

been found to have higher rates of consultation with

their doctor, higher rates of prescription, and higher

death rates during the follow-up period compared

with those with a BMI of 20–<25kgm

2

Taste Changes

0010There is strong evidence that taste changes and asso-

ciated alterations in food preferences occur during the

development of neoplastic disease. These can occur

irrespective of the location of the primary tumor or

810 CANCER/Diet in Cancer Treatment

the therapeutic management of the patient. Measure-

ments of taste detection thresholds (TDTs) have

shown the threshold for bitterness to be reduced,

that is, that patients could taste bitterness at lower

concentrations than their noncancerous counterparts.

Conversely, patients with cancer had elevated thresh-

olds for sucrose and could only taste sucrose (sweet)

at higher concentrations than noncancerous subjects.

The changes in TDT were more common in patients

with weight loss and in early reports there was a

suggestion that they decreased as the energy intake

increased. The metabolic cause of the taste changes is

not at present clear but studies in end-stage cancer

patients with low thresholds to bitter have been found

to have an ongoing metabolic response with higher

levels of the acute-phase reactant protein, C-reactive

protein, when compared with an age-matched group.

Moreover, patients with the lowest bitter thresholds

had the highest plasma levels of the cytokine, tumor

necrosis factor-a. Heightened recognition of house-

hold odors has also been reported in patients with

end-stage disease.

0011 Attempts to correlate changes in TDT with plasma

vitamin A (retinol) and plasma zinc in patients with

cancer are complicated by the interpretation of such

levels. Plasma retinol levels depend on the availability

of retinol-binding-protein (RBP), the levels of which

are susceptible to nutritional status, and plasma zinc

concentrations are an uncertain measure of tissue zinc

status. Magnesium status has been found to correlate

with salt detection thresholds.

0012 Taste changes can have important consequences for

the feeding of cancer patients. Reduced threshold for

bitter taste (as determined with urea) results in beef

and pork becoming less acceptable, poultry and fish

rather intermediate, and cheese and eggs pleasurable.

Taste abnormalities vary between patients and this

is one aspect of the condition which needs to be

individually assessed.

0013 Taste changes may result in aversions, but aver-

sions may also develop in patients who do not have

disturbances in taste. There is some evidence that

food aversions may be associated, or learned, by the

association of certain foods with side-effects of drug

treatment. Whatever the mechanisms by which they

develop, aversions to foods may be persistent, lasting

for several months, and ignoring them will have a

definite effect on food intake.

Energy Expenditure

0014 The resting energy expenditure (REE) of cancer

patients has been reported to be increased, decreased,

or the same as that of control subjects. Some of this

variation is due to the subjects’ different kinds of

cancer, the cross-sectional nature of the studies, and

to the different body composition of the subjects. The

resting metabolic rate (RMR) of a group of lung

cancer patients has been found to be 108% of pre-

dicted values at diagnosis, with a range of 96–151%.

In those who experienced a partial or complete remis-

sion, maintenance of body weight was associated

with a significant decrease in RMR. In contrast, pa-

tients who did not respond to treatment and who had

lost weight had an unchanged RMR. It seemed that

the elevated RMR in those with active disease was a

response mediated by cytokines and acute-phase pro-

teins rather than to the metabolic activity of the small

tumors themselves.

0015However, RMR is only part of the total energy

expenditure (TEE) and changes in physical activity

can exert a greater influence on TEE. It is a matter

of common observation that sick people reduce their

physical activity and this could counterbalance the

increase in RMR. This, indeed, seemed to be so in

patients with lung cancer, and this meant that the

energy requirement of lung cancer patients is unlikely

to be higher than their preillness levels.

Glucose Metabolism

0016Altered glucose metabolism in the patient who is

losing weight is the result of glucose intolerance,

changed b-cell function, and decreased sensitivity

and responsiveness to insulin. The causes of these

changes are not clear but it seems likely that decreased

food intake, protein loss, and hormone changes con-

tribute, as do specific characteristics of the tumor.

Increased glucose turnover is a composite of in-

creased Cori cycle activity (80%), glucose oxidation

(15%) and lipid synthesis, and ketone formation

(5%). (See Glucose: Function and Metabolism.)

Protein Metabolism

0017Muscle wasting is clear evidence of altered protein

metabolism. Most studies on this subject have been

performed using experimental animals but insofar as

similar studies have been carried out in patients, they

have confirmed those in animals. Cancer increases

whole-body protein turnover, with an increased rate

of protein synthesis in the liver, but a decreased rate of

synthesis in skeletal muscle with a simultaneous in-

crease in skeletal muscle degradation. It is likely that a

number of factors affect protein metabolism in the

cancer patient, e.g., tumor specificity, the staging

of the tumor, age, sex, and previous nutritional

condition of the patient. (See Protein: Digestion and

Absorption of Protein and Nitrogen Balance; Syn-

thesis and Turnover.)

CANCER/Diet in Cancer Treatment 811

0018 The biggest challenge in the nutritional manage-

ment of a patient with cancer is to reverse the changes

in protein metabolism which result in the severe

muscle wasting so characteristic of cachexia.

0019 Hypoalbuminemia is common in cancer patients

and is often interpreted as evidence of malnutrition.

However, this is only so if the other causes have been

eliminated. These other causes are decreased albumin

synthesis, increased plasma volume, and passage of

protein-rich fluid into the gut, i.e., protein-losing

enteropathy.

Fat Metabolism

0020 Loss of subcutaneous fat is an obvious feature of

cancer cachexia. Patients with progressive disease

metabolize more fat than those with nonprogressive

disease. If a rise in plasma free fatty acids (FFA) can be

taken as an index of the mobilization of fat stores,

then a significant increase has been found to correlate

with the clinical activity of a tumor. However, these

findings have not been confirmed. There does seem to

be a failure of normal homeostatic mechanisms, pos-

sibly owing to insulin resistance, because glucose does

not depress FFA oxidation to the same degree in

cancer patients as it does in control subjects.

0021 Whatever their cause, loss of body fat, inappropri-

ate energy consumption, and loss of lean body mass

can only contribute to a negative energy balance and

continuing weight loss. (See Fats: Digestion, Absorp-

tion, and Transport; Fatty Acids: Metabolism.)

Vitamins and Minerals

0022 Overt vitamin-deficiency diseases are rare in western

societies but their absence is no guarantee of satisfac-

tory status with respect to any vitamin. Patients with

malignant disease are, indeed, likely to suffer some

degree of deficiency with respect to a number of

vitamins which may affect their continuing resistance

to the disease and determine their response to treat-

ment. Poor food intake and abnormal losses through

the alimentary and renal tracts, possibly coupled with

increased requirements which may be drug-induced,

may contribute to vitamin depletion. If such deficien-

cies are suspected during recovery from active treat-

ment, or during palliative care, patients should be

given appropriate supplements. However, patients re-

ceiving anticancer treatment are advised not to take

vitamin supplements, unless under medical super-

vision, as they may interfere with the action of the

drugs.

0023 Some vitamins, particularly those that are anti-

oxidants (vitamin C, b-carotene, vitamin A and a-

tocopherol) have been suggested, on the basis of

epidemiological studies in humans and experimental

studies in animals, to give protection against the de-

velopment of some kinds of tumors, particularly lung

cancer. However, large-scale trials with supplements

of these substances have so far yielded adverse and

contradictory results and it is considered at present

that there is no place for supplements of these vita-

mins in the prevention or treatment of cancer.

0024Of the essential minerals investigated in relation to

cancer, selenium has attracted most attention, largely

because of its antioxidant properties. Though in

rodents dietary supplementation has been found to

inhibit colon, mammary gland, and stomach cancer,

there is as yet no evidence for such an effect in

humans. Iron transferrin and ferritin have been linked

to cancer risk and cancer cell growth, possibly

because many cancer cells require iron for growth.

The relevance of this to dietary management seems

unclear.

Diet in the Chemoprevention of Cancer

0025Epidemiological studies have provided evidence that

diets low in fat and high in vegetables, whole-grain

cereals and fruits are associated with lower risk of

colon, breast, prostate, and endometrial cancer. Al-

though such diets are rich in antioxidant vitamins, it

seems that their role in cancer prevention is due to the

presence in these foods of other substances with anti-

cancer properties. Foremost amongst those at present

being investigated are the phytoestrogens (coume-

stans, isoflavonoids, flavonoids, and lignans). A

number of these compounds have been identified in

fruits, vegetables, and whole grains commonly con-

sumed by humans and research is under way to deter-

mine the quantities present. Studies in humans,

animals, and cell culture systems have suggested a

number of possible mechanisms for their anticancer

action, including estrogenic and antiestrogenic effects,

induction of cancer cell differentiation, and suppres-

sion of angiogenesis (i.e., extension of existing blood

vessels).

0026Good sources of phytoestrogens are legumes (par-

ticularly soybeans), dark green leafy vegetables, nuts

(including peanuts), seeds (especially linseed), whole

grains and fresh fruits (including apples and bananas)

and dried fruits (especially dates, prunes, and raisins).

Antibreast-cancer diets including rich sources of

phytoestrogens have been proposed. Genistein, a

prominent isoflavonoid in soy products, may be

responsible for lowering breast cancer risk in Asian

women. Other plant constituents which are under

investigation for their anticancer properties include

indole-3-carbinol and glucosinolates. Indole-3-carbi-

nol is found in vegetables of the Brassica genus and

812 CANCER/Diet in Cancer Treatment

has been found to induce cell cycle arrest of breast

cancer cells. Glucosinolates are often responsible for

the bitter taste of condiments (e.g., in mustard and

horseradish) and are found in turnips, cabbage, and

Brussels sprouts. Some glucosinolates are effective

inhibitors of cancer induction in rodents by modify-

ing the metabolism of carcinogens. The use of these

compounds in humans as anticancer agents may be

limited by the fact that some of them are goitrogenic.

Effects of Antineoplastic Treatment on

Nutritional Status

Surgery

0027 Not only are cytokines produced as a result of the

presence of cancer in the body, they are also produced

as a result of surgery, and a decrease in muscle mass

characteristic of the cachectic process also occurs as

part of the neuroendocrine response to surgery. The

initial effect of the cytokines produced by macro-

phages is on the hypothalamus and this can cause

anorexia, even in a previously noncachectic patient.

The hypothalamus produces corticotropin-releasing

factor which stimulates the pituitary to secrete ad-

renocorticotropic hormone which causes the adrenals

to produce glucocorticoid hormones, and it is these

hormones that stimulate the breakdown of muscle

proteins. Some of the amino acids produced by

muscle proteolysis, principally alanine, are converted

to glucose by gluconeogenesis, whilst others are used

to synthesize acute-phase proteins, one of which, C-

reactive protein, is also produced during the cachectic

process caused by cancer. The glucose produced by

gluconeogenesis following surgery causes a rise in

blood glucose levels which has been described as the

‘diabetes of trauma.’

0028 Largely experimental studies have shown that the

production and suppression of cytokines can be

manipulated by nutritional means, particularly fats,

and this offers scope for their potential use in the

treatment of severely ill patients.

0029 The extent of these changes, their duration, and the

period of convalescence are directly proportional to

the severity of the surgery. This is important for the

patient with cancer in whom surgery may need to be

extensive in order to remove all cancerous tissue.

Moreover, surgery is the treatment of choice for

cancers of the alimentary tract and may have direct

nutritional consequences (Table 1) which will be

superimposed on the general ‘metabolic response to

injury’ experienced by all surgical patients. Nutri-

tional management of these patients requires careful

attention to specific problems, possibly over an

extended period of time.

Radiation Therapy

0030The purpose of radiation therapy is to kill cells and

the effects may be dramatic. As the cells are des-

troyed, toxic waste products are liberated, causing

an inflammatory response with consequent edema.

Cells may be lost from tissue surfaces which then

become ulcerated and clinically the patient shows

symptoms such as mucositis and stomatitis, nausea,

vomiting, cystitis, or diarrhea. Irradiation may have

both immediate and late effects on tissues and hence

on nutritional status. Normal and tumor cells are

affected in similar ways and cell death may be imme-

diate, delayed, or ‘natural’:

0031Immediate cell death is the result of irreversible

damage to DNA; this mechanism is responsible for

the immediate side-effects.

0032Delayed cell death occurs after mutation of DNA;

limited functions usually continue until the M

phase in the multiplication process, when the

cells are unable to divide; this mechanism is

responsible for late effects.

0033‘Natural’ cell death involves the formation of giant

(sterile) cells which function but cannot divide.

Some of the nutritional consequences of radiation ther-

apy may be severely disabling, even life-threatening,

tbl0001Table 1 Possible nutritional effects of gastrointestinal surgery

Surgicalprocedure Possibleeffects

Resection of the

oropharyngeal region

Chewing or swallowing difficulties

Prolonged dependence on tube

feeding

Esophagectomy Gastric stasis

Esophageal

reconstruction

Reduced acid secretion

(both these effects are due to cutting

the vagus nerve)

Fat malabsorption

Diarrhea

Gastric resection

(partial or total)

‘Dumping syndrome’

Malabsorption

Reduced acid secretion

Reduced absorption of vitamin B

Low blood glucose level

Small intestine:

duodenum

Reduced secretion of pancreatic juice

and bile

Jejunum Reduced efficiency of absorption

Ileum Reduced absorption of bile salts

Reduced absorption of vitamin B

Increased urinary excretion of oxalate

Massive resection Severe malabsorption

Malnutrition

Metabolic acidosis

Colon Fluid and electrolyte imbalance

Removal of pancreas Reduced secretion of pancreatic

enzymes

Reduced secretion of insulin –

diabetes mellitus

CANCER/Diet in Cancer Treatment 813

and occur in patients whose nutritional status is

already compromised. Food intake is substantially

reduced and malabsorption resulting from treatment

of tumors in or near the gastrointestinal tract may

be very difficult to control. The severity of these

problems means that aggressive, symptom-related

nutritional support (Table 2) is an essential adjunct

to radiation therapy in order to prevent deterioration

of nutritional status and to allow compensatory

hyperplasia of undamaged bowel.

Chemotherapy

0034 Anticancer drugs have profound cytotoxic effects and

since they cannot discriminate between ‘normal’ and

‘cancer’ cells they may seriously affect ‘normal’

tissues. Contributions to host malnutrition may be

by both direct and indirect mechanisms that cause

many symptoms. Current medical practice is to ad-

minister drugs cyclically in multiple combinations,

sometimes called ‘cocktails.’ This increases the possi-

bility of ‘killing’ cancer cells at different stages of cell

multiplication, but it can increase both the range and

severity of side-effects. Different anticancer drugs

cause a slightly different pattern of side-effects

(Table 2), but the general effect is to reduce food

intake and since this may occur over prolonged

periods it will lead to weight loss and debility.

0035Indirect effects of chemotherapy that may contrib-

ute to a reduced food intake include moniliasis

or candidiasis. These are particularly common in

patients with leukemia and lymphoma and affect the

oral cavity, pharynx, or esophagus, causing a sore,

painful mouth, odynophagia, and dysphagia.

0036Possible palliative treatment for some of these

problems is shown in Table 2.

Nutritional Management of Cancer Patients

0037The feeding of cancer patients should be seen as part

of their total care. Provision of food for these patients

requires the following:

0038Understanding. Carers and providers should

appreciate the clinical problems and difficulties

experienced by individual patients so that the

‘diet,’ whether oral, enteral, or parenteral, is suited

to the patients’ needs, including the effects of the

anticancer treatment.

0039Flexibility. Dietary provision must be flexible,

with menus to cater for a wide range of needs

with respect both to the nature of the food and

the timing of its availability.

0040Psychological importance. The provider should be

sensitive to the psychological condition of patients

and the importance of food to them. This involves

tbl0002 Table 2 Effects of radiation and drug treatment of cancer and their palliative treatment

Treatment Undesirable effects Possible palliative treatment

Radiation

Mouth and throat Nausea Small frequent meals; reduced fat

Vomiting Total parenteral nutrition (TPN)

Anorexia Small meals, served attractively

Severe anorexia Tube feeding

Loss of taste Emphasis on aroma of food

Dental deterioration Reduced sugar intake

Sore mouth Bland soft foods; tube feeding

Lack of saliva Avoidance of dry foods

Swallowing and

chewing difficulties

Soft, semisolid small meals at correct

temperature; between-meal supplements

Upper abdomen Nausea and vomiting As above

Sense of fullness Small frequent meals; reduced fat

Reduced secretion of digestive enzymes Soft, moist, easily digested foods

Lower abdomen Intestinal cramping Soft, bland, low-residue diet

Diarrhea – dehydration Low-residue foods, high fluid intake

Malabsorption Low-fat (?), lactose-free (?), elemental diet; TPN

Drugs

General effects Nausea, vomiting, anorexia, diarrhea As above

Particular effects

Nitrogen mustard Metallic taste in mouth Elimination of any food accentuating this taste; individual tolerance

Cyterabine Abdominal pain Bland low-residue diet

Hydroxyurea Intermittent constipation Prune juice, large fluid intake (at least 3 l day

1

)

Vinca alkaloids Intermittent constipation As above

Abdominal pain As above

Adapted from Dickerson JWT and Williams CM (1988) Nutrition and cancer. In: Dickerson JWT and Lee HA (eds) Nutrition in the Clinical Management of

Disease, 2nd edn, pp. 350–373. London: Edward Arnold.

814 CANCER/Diet in Cancer Treatment

the way in which food is provided, as well as its

nature. Food can be a ‘messenger’ between the

carer and the patient.

Oral food is always the first choice and considerable

skill may be needed to obtain an appropriate intake.

Detailed suggestions for dealing with specific prob-

lems are given in some of the works quoted in the

Further Reading section. Practical suggestions for

increasing energy and protein intakes are as follows:

0041 Plan a definite eating schedule and then adhere to

it. Do not omit meals or between-meal snacks.

0042 Add cereal with banana or other fruit, sugar, and

cream to the breakfast menu.

0043 Butter toast or bread when it is hot because more

butter can be used.

0044 Use jam, marmalade, or cheese with toast or

bread.

0045 Add cream to milk beverages.

0046 Add skim milk powder to milk, soups, pudding,

mashed potatoes, etc.

0047 Add icecream or whipped cream to desserts (cake

and icecream, apple pie with whipped cream, etc.).

0048 Use mayonnaise, salad dressing, butter, or mar-

garine with sandwiches, salads, and vegetables.

0049 Serve thickened gravy with meat and potatoes.

10.

0050 Have bacon, ham, or sausages with eggs at

breakfast.

11.

0051 Add eggs to beverages or recipes to increase the

protein.

12.

0052 Use milk instead of water to prepared canned

condensed cream soups.

Oral food can be supplemented with ‘sip’ feeds. These

may be homemade, milk- and egg-based, or commer-

cially available complete feeds such as Ensure, Clin-

ifeed, or Fortisip. If elemental feeds such as Flexical

or Vivonex are needed due to malabsorption, they

should never be given orally because of their taste.

Patients with inadequate oral intake should be fed

through a fine-bore tube (1–2 mm in diameter)

inserted nasogastrically, or directly into the stomach

or intestine. Feeds used as sip feeds should be intro-

duced from a reservoir either by gravity or with a

pump. If the gastrointestinal tract cannot be used, or

if the patient cannot tolerate a tube, complete nutri-

tion can be given parenterally through a central vein.

Parenteral feeds (for TPN) contain energy (usually fat

and glucose), amino acids, vitamins, and minerals.

Given under expert supervision, their use in cancer

patients before surgery has been found to reduce

complications and in some reports to reduce the

length of hospital stay. Usually, the greatest benefit

has been in the most malnourished individuals. TPN

may also widen the scope for radiation and drug

treatment. Indeed, the survival of some cancer pa-

tients may depend on this kind of treatment. A poten-

tial problem with TPN is that it results in a decrease in

the barrier function of the gut, but this can be pre-

vented by giving some food enterally.

Conclusions

0053The feeding of cancer patients is part of their total care

at all stages of the disease and its treatment. It is at

different times ‘supportive,’‘adjunctive,’ and ‘defini-

tive.’ It is supportive as part of the preparation for

definitive treatment. It is adjunctive as a part of the

overall therapy. In some individuals it becomes defini-

tive as the therapy that permits survival. In the future, it

is envisaged that diets containing foods with anticancer

properties may be used to treat specific tumors.

See also: Fats: Digestion, Absorption, and Transport;

Fatty Acids: Metabolism; Glucose: Function and

Metabolism; Protein: Digestion and Absorption of Protein

and Nitrogen Balance; Synthesis and Turnover