Risk of Human Ovarian Cancer Is Related to Dietary Intake of Selected Nutrients,

[Guest guest]

http://www.healthy.net/asp/templates/news.asp?Id=7248

 

Risk of Human Ovarian Cancer Is Related to Dietary Intake of Selected Nutrients,

Phytochemicals And Food Groups1

 

 

Printed With Permission HealthWorld Online (http://www.healthy.net)

 

 

 

 

 

 

Originally Published:20030601.

The literature concerning the relationship between dietary components and

ovarian cancer is limited. In general, reduced risks of ovarian cancer have been

associated with higher intakes of dietary fiber, vitamin A, [beta]-carotene,

fruits and vegetables (1-12), and fruit and vegetable intakes during adolescence

(13), whereas increased risks have been associated with higher intakes of total

fat and animal fat, cholesterol and lactose intakes, meat, eggs and whole milk

(8,11,14-17). The mechanisms by which diet could affect the etiology of ovarian

cancer include modulation of endogenous hormones (18-24), antioxidant activity

or other anticarcinogenic mechanisms (25).

Recent research has identified food components (phytochemicals) that may have

important anticarcinogenic activities. These include phytoestrogens,

phytosterols and flavonoids. Phytoestrogens, including isoflavones (primarily

from soy) and lignans (fruits, vegetables, grains and seeds), have been shown to

possess estrogenic as well as antiestrogenic activity (26). Phytosterols are

plant compounds found in food oils that are structurally similar to cholesterol.

These compounds could affect endogenous hormone levels through alterations in

bile acid metabolism and estrogen reabsorption, or through competition with

cholesterol as a substrate for steroid hormone synthesis (27). Flavonoids (found

in fruits and vegetables), such as quercetin, have been shown to possess potent

antioxidant, antibacterial, antithrombotic, antiinflammatory and

anticarcinogenic properties (28,29).

To our knowledge, the relationship between phytochemical intake and ovarian

cancer has not been investigated previously. The aim of this study was to

investigate the relationship of selected nutrients, other food components

including phytochemicals, and food groups with risk of ovarian cancer in a case

-control study conducted in western New York.

SUBJECTS AND METHODS

The present study utilized data obtained as part of a series of case-control

studies of diet and cancer of the breast, endometrium, ovary and prostate in

western New York (1986-1991). The methods were described in detail elsewhere

(30-32). The study protocol was approved by the Institutional Review Board of

the University at Buffalo and informed consent was obtained from all

participants. Briefly, women in this study included 124 incident, primary,

histologically confirmed cases of ovarian carcinoma identified in women between

the ages of 40 and 85 y. The mean number of months between diagnosis and

interview for cases was 2.59 (median = 2.00); 93% were interviewed within 6 mo

of diagnosis. Controls from the complete case-control series (breast,

endometrium and ovary; n = 696) were frequency matched to cases on age (+ or - 5

y) and county of residence and were randomly selected from driver's license

lists for women <65 y of age and from Health Care Finance Administration lists

for

women > or =65 y of age.

Data were collected during a detailed in-person interview by trained nurse

interviewers and included questions on diet, reproductive history, family

history of cancer, medical history, health habits such as cigarette smoking and

physical activity, and other lifestyle and occupational factors. Diet in the

12-mo period 2 y before the interview was queried with an extensive

food-frequency questionnaire (FFQ) that included questions for portion size for

each food, seasonality of use and food preparation methods. Although the FFQ

used in this study was not validated in this population, it is comparable in

structure to widely used and extensively validated instruments such as those

used by the National Cancer Institute and Harvard University's Nurse's Health

Study, and thus is likely to possess comparable validity.

Nutrient intakes from foods and beverages were calculated using food composition

data from the USDA and published food composition tables (33-35). Vitamin

supplement intake was not assessed. [beta]-Sitosterol, campesterol,

stigmasterol, total phytosterols and the flavonoids, quercetin and kaempferol,

were calculated using published food composition data compiled by Pillow et al.

(36). Dietary intakes of total lignan precursors were calculated as the sum of

secoisolariciresinol and matairesinol using the method of de Kleijn et al. (37).

Isoflavone intake was not assessed because soy foods were not queried.

Statistical analyses. Analyses were conducted using SPSS for Windows (Chicago,

IL), and all statistical tests were considered significant at P < 0.05.

Descriptive characteristics of cases and controls were compared with Student's t

tests for continuous variables and [chi]^sup 2^ for categorical variables. Odds

ratios (OR) and 95% CI for risk of ovarian cancer associated with the following

nondietary risk factors was estimated with unconditional logistic regression:

BMI [kg/m^sup 2^] (>27.5 vs. <27.5), cigarette smoking (never, former, current),

pregnancy (never, ever), fertility problems (no, yes, never attempted

pregnancy), total months menstruating (median cutpoint), oral contraceptive use

(never, ever) and hormone replacement therapy (never, ever).

We considered the following dietary components: total energy (kJ), carbohydrates

(g), protein (g), fat (g), saturated fat (g), monounsaturated fat (g),

polyunsaturated fat (g), cholesterol (mg), alcohol (g), dietary fiber (g),

vitamin C (mg), folate ([mu]g), total carotenoids (mg), and the phytochemicals

[beta]-sitosterol (mg), campesterol (mg), stigmasterol (mg), total phytosterols

(mg), total lignan precursors (secoisolariciresinol and matairesinol) ([mu]g),

quercetin ([mu]g), and kaempferol ([mu]g). Finally, risk of ovarian cancer

associated with monthly grams of intake of fruits, vegetables, grain products,

red meat, poultry, fats, snacks and dairy were estimated. Each dietary variable

was categorized into quintiles on the basis of the distribution in the controls,

and risk in each quintile estimated referent to the lowest quintile of intake.

For these analyses, the OR and 95% CI were calculated with unconditional

logistic regression adjusting for age, education, total months

menstruating, difficulty becoming pregnant, oral contraceptive use, menopausal

status and total energy intake.

RESULTS

Compared with controls, cases were older, had less education, had a greater

number of total months menstruated, were more likely to be postmenopausal, more

likely to have had difficulty becoming pregnant, and, consequently, they had

somewhat lower parity than controls (Table 1). Although not significant, cases

were somewhat less likely to have ever been pregnant (P = 0.08). Cases also were

less likely to have ever used oral contraceptives (P = 0.09), but more likely to

have used hormone replacement therapy (P = 0.37) and less likely to be current

smokers (P = 0.48). No differences were observed between cases and controls for

BMI, age at menarche or age at first birth.

Among nutrient and food group intakes, cases consumed significantly less total

carotenoids and more snacks than did controls (Table 2). Compared with controls,

cases also tended to have slightly higher mean intakes of energy (P = 0.19),

carbohydrate (P = 0.20), protein (P = 0.46), total and specific fats (P = 0.12,

0.12, 0.27, 0.38, total, saturated, monounsaturated, and polyunsaturated fats,

respectively), and cholesterol (P = 0.24), and somewhat lower mean intakes of

alcohol (P = 0.32), dietary fiber (P = 0.73), folate (P = 0.69) and total lignan

precursors (P = 0.09).

The primary food sources for the phytosterols included black tea, peanuts and

peanut butter, and margarine (data not shown). For the flavonoids, quercetin and

kaempferol, the major food sources included tea, tomato sauce, vegetable soup,

onions, apples and applesauce, broccoli and strawberries. Top contributors to

the lignan precursors, secoisolariciresinol and matairesinol, included

decaffeinated and caffeinated coffee, carrots, cucumbers and strawberries.

Consistent with previous literature, reduced ovarian cancer risks were observed

for women who had ever been pregnant compared with those who had never been

pregnant (OR 0.64, 95% CI, 0.39-1.06, P = 0.081) and for ever use of oral

contraceptives compared with never use (OR 0.72, 95% CI, 0.49-1.06, P = 0.093)

(Table 3). Similarly, increased risks were observed for women reporting problems

becoming pregnant (OR 1.61, 95% CI, 1.03-2.51), or who had never attempted

pregnancy (OR 3.07, 95% CI, 1.29-7.35) and for longer duration of exposure to

menstruation. (OR 1.77, 95% CI, 1.19-2.63). No associations with risk were

observed for ovarian cancer and BMI, cigarette smoking or hormone replacement

therapy.

We observed a substantial reduction in risk associated with the highest vs.

lowest quintile of dietary fiber intake (OR 0.43, 95% CI, 0.20-0.94) and total

carotenoids (OR 0.33, 95% CI, 0.16-0.68) (Table 4). Risk was not associated with

intakes of other nutrients or food groups.

Among the phytochemicals examined (Table 5), reduced ovarian cancer risks were

observed for the highest vs. lowest intakes of stigmasterol (OR 0.42, 95% CI,

0.20-0.87) and total lignan precursors (OR 0.43, 95% CI, 0.21-0.85). Risk did

not appear to be related to intakes of the other phytosterols, total

phytosterols or the flavonoids quercetin or kaempferol. Adjustment for possible

confounding nutrients such as dietary fiber and folate had no effect on these

estimates (data not shown)

For summary examinations of diet, we also estimated risk of ovarian cancer

associated with monthly intakes of selected food groups (Table 6). Consistent

with the finding of reduced risks associated with dietary fiber and carotenoid

intakes, we observed a substantial decrease in risk for women in the highest vs.

lowest quintile of vegetable intake (OR 0.47, 95% CI, 0.23-0.97). Reduced risks

were also observed for the highest intakes of poultry (OR 0.45, 95% CI,

0.22-0.92), but risk did not appear to be related to fruits, grain products, red

meat, fats, snacks or dairy products.

DISCUSSION

Although ovarian cancer is a relatively uncommon neoplasm, accounting for an

estimated 4% of cancer cases among women in 2002, mortality from this cancer is

high (38). Ovarian cancer is not easy to detect, making early diagnosis

difficult. Therefore, identification of modifiable factors contributing to its

etiology is important in reducing the incidence of this highly fatal neoplasm.

Previously identified risk factors have been related to hormonal factors.

Reduced ovarian cancer risks have been associated with oral contraceptive use

and fewer menstrual cycles (consequent to greater parity), whereas increased

risks have been associated with difficulties becoming pregnant and hormone

replacement therapy (39,40). Consistent with the previous literature, we found

reductions in risks for women who had ever been pregnant vs. never pregnant and

for oral contraceptive use. We also observed increased risks among women

reporting difficulty becoming pregnant or who had never become pregnant.

Investigations of dietary factors in the etiology of ovarian cancer have been

limited. One of the mechanisms by which diet could affect ovarian cancer risk is

through modulation of endogenous hormone levels. Relationships between diet and

endogenous hormone levels have been fairly well established, with lower levels

of estrone and estradiol levels reported for women consuming vegetarian diets

compared with omnivorous diets (18,19,24). Furthermore, it has been demonstrated

that endogenous hormone levels can be modified through changes in dietary intake

(20-23). In this study we found significantly reduced risks associated with

poultry and vegetable intake and related nutrients, especially dietary fiber and

lignans. These findings are consistent with previous investigations (1-12) and

provide further evidence that a healthy diet may at least contribute to the

etiology of this neoplasm.

Interestingly, although we observed reductions in risk associated with higher

dietary fiber and lignan intakes, risk was not related to higher intakes of

grain products. Among grains, both dietary fiber and lignans are found

predominantly in the bran portion of whole grains, which is removed in the

refining process. In the western New York population, consumption of whole grain

foods is low, and the primary contributors to lignan intakes in these data were

coffee, carrots, cucumbers and strawberries, not grains. On the other hand, the

findings for dietary fiber and lignans are consistent with the observed reduced

risks associated with vegetable intake.

To our knowledge, this is the first study to investigate the effect of

phytochemical intakes on ovarian cancer risk. In these data, we observed

important reductions in risk associated with the highest quintiles of intake of

stigmasterol and total lignans. Phytosterols have been shown to affect

cholesterol metabolism, which could ultimately affect steroidogenesis (27).

Plant lignans are phytoestrogens metabolized in the mammalian gut to form

enterolactone and enterodiol, compounds with estrogenic as well as

antiestrogenic activities (41). Reduced risks of breast cancer, another

hormonally related cancer, have been associated with higher urinary excretion of

lignans, although risks associated with dietary lignan intake have been

inconsistent (42-44). Part of the observed effects of high lignan intakes might

be explained by higher intakes of plant foods in general; however, adjustment

for possibly confounding nutrients such as dietary fiber and folate had no

effect on our observed

estimates. Our results strengthen the hypothesis that the phytoestrogen lignans

may be important contributors to the effect of diet for this and other

hormone-related cancers.

On the other hand, we observed no association in risk with higher intakes of the

flavonoids quercetin and kaempferol in these data. The hormonal etiology of

ovarian cancer is fairly well established; flavonoids are potent antioxidants.

Although it is possible that oxidative mechanisms might be important in some

portion of the development or progression of ovarian neoplasms, it is likely

that the hormonal mechanisms are more pronounced. Conversely, we were limited in

our investigation of flavonoid intakes. Food composition data were available

only for quercetin and kaempferol; there are >400 flavonoids present in the food

supply (28). Although quercetin and kaempferol are strongly represented in plant

foods, it is possible that we are not sufficiently estimating the true effect of

flavonoids in these analyses.

An important limitation of our study is the small number of cases available for

analyses, possibly reducing the generalizability of our results. Although this

may have reduced the precision of our estimates, we observed both dietary and

nondietary risk factors in these data that were similar to those reported in

larger studies. Furthermore, even with categorization of dietary variables into

quintiles, our estimates were fairly stable, and the confidence limits not

overly wide. It is unlikely that a larger sample size would change our observed

associations, but rather would strengthen the precision of the estimates.

As is common to all case-control studies, there may have been biased reporting

of dietary intake and other data. The mean number of months between diagnosis

and interview was short (2.59) and diet was queried for the 12-mo period 2 y

before diagnosis. Furthermore, we observed both dietary and nondietary risk

factors in these data similar to those reported in the literature. It is

unlikely that bias was a serious problem in these analyses.

Food composition data for the phytochemicals examined in this study have been,

until more recently, fairly limited. It is possible that we are underestimating

true intakes of these compounds. However, for estimation of risk associated with

diet, ranking of individuals on intake is more important than estimation of

actual intake, especially given that we are interested in usual long-term diet.

The error in intake estimation is likely to be random, and not associated with

case-control status in these data.

Our study provides additional evidence that diet may play an etiologic role in

ovarian cancer, and is the first, to our knowledge, to report an association

with phytoestrogen intakes. The evidence toward a protective effect of a

plant-based diet on hormone-related cancers continues to accumulate. Our

continuing challenge will be in the design and implementation, of strategies to

encourage population-based changes toward this dietary pattern.

 

 

 

 

 

@

 

Alternative Medicine/Health-Vitamins, Herbs, Aminos, etc.

 

To , e-mail to:

alternative_medicine_forum-

 

Or, go to our group site at:

alternative_medicine_forum

 

 

 

SBC DSL - Now only $29.95 per month!