Journal of the National Cancer Institute Advance Access originally published online on July 24, 2007
JNCI Journal of the National Cancer Institute 2007 99(15):1200-1209; doi:10.1093/jnci/djm065
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press 2007.
ARTICLES |
Prospective Study of Fruit and Vegetable Intake and Risk of Prostate Cancer
on behalf of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial
Affiliations for authors: Research Unit, Division of Preventive Oncology, Cancer Care Ontario, Toronto, ON, Canada (VAK); Department of Epidemiology and Public Health, Yale University School of Medicine, New Haven, CT (VAK, STM); Public Health Sciences Division, Fred Hutchinson Cancer Research Center, Seattle, WA (UP); Department of Epidemiology, University of Washington, Seattle, WA (UP); Divisions of Cancer Epidemiology and Genetics (NC, RBH) and Cancer Control and Population Sciences (AFS), National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MD; Josephine Ford Cancer Center, Henry Ford Health System, Detroit, MI (CCJ)
Correspondence to: Richard B. Hayes, PhD, EPN 8114, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD 20892 (e-mail: hayesr{at}mail.nih.gov).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsParticipants and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Background: Several epidemiologic studies have reported associations between fruit and vegetable intake and reduced risk of prostate cancer, but the findings are inconsistent and data on clinically relevant advanced prostate cancer are limited.
Methods: We evaluated the association between prostate cancer risk and intake of fruits and vegetables in 1338 patients with prostate cancer among 29361 men (average follow-up = 4.2 years) in the screening arm of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial. Participants completed both a general risk factor and a 137-item food-frequency questionnaire at baseline. Cox proportional hazards models were used to estimate relative risks (RRs) and 95% confidence intervals (CIs). All statistical tests were two-sided.
Results: Vegetable and fruit consumption was not related to prostate cancer risk overall; however, risk of extraprostatic prostate cancer (stage III or IV tumors) decreased with increasing vegetable intake (RR = 0.41, 95% CI = 0.22 to 0.74, for high versus low intake; Ptrend = .01). This association was mainly explained by intake of cruciferous vegetables (RR = 0.60, 95% CI = 0.36 to 0.98, for high versus low intake; Ptrend = .02), in particular, broccoli (RR = 0.55, 95% CI = 0.34 to 0.89, for >1 serving per week versus <1 serving per month; Ptrend = .02) and cauliflower (RR = 0.48, 95% CI = 0.25 to 0.89 for >1 serving per week versus <1 serving per month; Ptrend = .03). We found some evidence that risk of aggressive prostate cancer decreased with increasing spinach consumption, but the findings were not consistently statistically significant when restricted to extraprostatic disease.
Conclusion: High intake of cruciferous vegetables, including broccoli and cauliflower, may be associated with reduced risk of aggressive prostate cancer, particularly extraprostatic disease.
Prior knowledge Several epidemiologic studies have found that fruit and vegetable intake are associated with reduced risk of prostate cancer, but the findings have been inconsistent and data on advanced prostate cancer are limited. Study design Prospective study of men in the screening arm of a long-term randomized screening trial. Contribution Fruit and vegetable intake was not related to the overall risk of prostate cancer. A decreased risk of extraprostatic prostate cancer (stage III or IV tumors) was associated with increased intake of vegetables, mainly cruciferous vegetables, including broccoli and cauliflower. Implications High intake of cruciferous vegetables, especially broccoli and cauliflower, may be associated with a reduced risk of aggressive prostate cancer. Limitations Individuals with high intakes of fruits and vegetables generally have lower rates of smoking, higher levels of physical activity, and a more healthy lifestyle than those with low intakes. These associations could confound the prostate cancer association.
|
Epidemiologic evidence indicates association between high fruit and vegetable intake and a reduced risk for many cancers (1), and there are a number of biologically plausible reasons for this observation. Fruit and vegetable constituents, including fiber, micronutrients (such as vitamins C and E and folate), and phytochemicals (such as carotenoids, phenolics, isoflavanoids, isothiocyanates, and indoles) have biologic activity as anticarcinogens (2). These substances induce detoxification enzymes, scavenge oxidative agents, inhibit malignant transformation, stimulate immune function, and regulate the cell cycle (2,3). Cruciferous vegetables, in particular, may play a role in prostate cancer risk (4), specifically because of their high content of glucosinolate-derived compounds. These compounds protect prostate cancer cells from DNA damage (5), induce apoptosis (6) and BRCA1 and BRCA2 expression in prostate cancer cells (7), and inhibit prostate cancer cell proliferation (8).
Prospective studies of an association between prostate cancer risk and fruit and vegetable intake have shown either nonstatistically significant inverse associations (9–11) or no associations (12–15), although there have been indications of potential benefit of cruciferous vegetables (10,16). None of the prospective studies reported statistically significant inverse associations with increasing intake of fruits (9–15,17–20). However, these studies were not conducted in populations uniformly screened with prostate-specific antigen (PSA) and only one included control for history of PSA tests (16). Moreover, only a few of these cohorts represented large-scale initiatives (10,14–16) (i.e., they ascertained 
600 patients with prostate cancer), several had relatively crude assessment of diet (9,17,19,21), many did not consider specific fruit and vegetable subgroups (11,17,19–21), only one prospective study reported risk with respect to total fruit and vegetable intake in relation to organ-contained and extraprostatic cancer (15), and one additional study reported disease-stratified findings for cruciferous vegetable intake (16).
In our study, we examined whether fruit and vegetable intake is associated with a reduced risk of prostate cancer among approximately 30000 men, including more than 1300 patients with prostate cancer. Our study, among participants in the screened arm of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, avoids the risk of differential screening, which can bias epidemiologic studies of diet and prostate cancer (22). We gave special attention to specific fruit and vegetable subgroups and different tumor subtypes. Moreover, we converted questionnaire responses into the number of servings of each of the Food Guide Pyramid’s food groups by using a standardized metric, as described (23). This pyramid servings method is an innovation in analyses of associations between diet and disease, in that exposure assessment is enhanced by disaggregating vegetables and fruit eaten in composite foods into the appropriate food groups and subgroups; the standardized assessment of food item or group intake allows comparability of data between studies. This method also allows the findings to be communicated to the public in the context of US Dietary Guidelines (24), which are consistent with recommendations of the 5-A-Day Program of the federal government (25).
|
Participants and Methods |
|---|
|
TopAbstractContext and CaveatsParticipants and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Study Setting
The PLCO Cancer Screening Trial is a multisite study that was designed to investigate the effectiveness of early detection for prostate, lung, colorectal, and ovarian cancers and to identify early markers and etiologic determinants of cancer (26). The trial enrolled participants from November 1, 1993, to June 30, 2001, and included men and women recruited from the general population by use of direct mailings, advertisements, and other means (27,28). Screening procedures were carried out at centers in Birmingham, AL; Denver, CO; Detroit, MI; Honolulu, HI; Marshfield, WI; Minneapolis, MN; Pittsburgh, PA; Salt Lake City, UT; St Louis, MO; and Washington, DC. Men who were aged 55–74 years were eligible for the trial if they had no history of prostate, colon, or lung cancer; were not under treatment for any cancer (excluding nonmelanoma skin cancer); had not had surgical removal of the prostate, a lung, or the colon; had not taken finasteride in the past 6 months; had not had more than one PSA test in the previous 3 years; and were not participating in another screening or cancer prevention trial. The study was approved by the institutional review boards of the US National Cancer Institute and the 10 screening centers. Study participants provided written informed consent.
Men who were randomly assigned to the screening arm of the trial underwent screening for prostate cancer early detection by serum PSA testing (at entry and annually for 5 years) and digital rectal examination (at entry and annually for 3 years). Men in the screening arm also received flexible sigmoidoscopy and chest x-rays for detection of colorectal and lung cancers, respectively. Men who had a positive PSA test result (i.e., a PSA level > 4 ng/mL) or a digital rectal examination that was suspicious for prostate cancer were referred to their medical care providers for diagnostic evaluations for prostate cancer. Trial participants were requested to provide information about recent diagnoses of cancer through annual mailed endpoint follow-up questionnaires.
For participants with suspected or reported prostate cancer, medical and pathology records related to the diagnostic follow-up were obtained from medical care providers to confirm the diagnosis. For deaths, death certificates and related medical and pathology records were obtained and abstracted by trained medical abstractors, who also performed systematic quality control reviews on data for approximately 100 patients with prostate cancer per year. Clinical stage groups were assigned on the basis of clinical (57% of tumors) or clinical and surgical (43% of tumors) assessments of the extent of tumor involvement by use of the tumor–node–metastasis stage of disease classification (29). Gleason scores were assigned according to the highest summary values reported for biopsy and prostatectomy results.
Study Population
Analyses were restricted to men randomly assigned to the screening arm of the trial to eliminate confounding by differential screening and because men in the control arm completed an alternate baseline questionnaire. Of the 38352 men in the screening arm, we excluded men reporting a history (before study entrance) of cancer (other than nonmelanoma skin cancer, n = 1001); men who did not have an initial PSA test or digital rectal examination (n = 2530); men who received an initial screening examination but with whom there was no subsequent contact (n = 1045); men who did not complete a baseline risk factor questionnaire (n = 903); and men who did not provide a dietary questionnaire (n = 6604; 83% response rate), missed more than seven items on the food-frequency part of the questionnaire (n = 250), or who reported energy intake in the top or bottom 1% of the reported energy intake distribution (corresponding to >5573 or <781 kcal/day, respectively; n = 634). We also excluded men whose initial screening examination after randomization occurred after October 1, 2001, the censor date for this analysis (n = 155). After exclusions, the analytic cohort comprised 29361 men (some participants were in more than one exclusion category), of whom 90.7% were white, 4.0% were Asian/Pacific Islanders, 3.3% were African American, 1.8% were Hispanic, and 0.2% were American Indians/Alaskan natives. Follow-up continued for up to 8 years (average follow-up = 4.2 years), during which 1338 men (4.5%) were diagnosed with prostate cancer (470 men were diagnosed in the first year and 868 were diagnosed thereafter), including 520 (38.9%) with aggressive disease (i.e., stage III or IV prostate cancer or Gleason score of 7).
Procedures
At study entry (baseline), participants provided the following information by questionnaire: age, race, education level, height, weight, adult occupation, smoking history, family medical history (including family history of prostate cancer), personal medical history (including use of selected medications), and physical activity level. Dietary information was collected through a self-administered food-frequency questionnaire (available at http://www3.cancer.gov/prevention/plco/DQX.pdf) that obtained information on 137 food items (including 23 individual fruit and 37 vegetable items) to assess the participant's usual diet for the previous year, including portion size (small, medium, or large) for 77 items, and information on nutrient supplement use. Except for juices, fruits and vegetables were among the foods for which portion size was not asked. Nine mutually exclusive response categories were provided for the frequency of intake, with the choices ranging from "never or less than once per month" to "two or more times per day" for food items and up to "six or more times per day" for beverages. Gram weights per portion size were assigned by use of data from the two 24-hour recalls administered in the 1994–1996 Continuing Survey of Food Intake by Individuals (CSFII), a nationally representative survey (30). For mixed dishes and juices, cut points between small and medium portions and between medium and large portions correspond to the 25th and 75th percentiles, respectively, for portion sizes reported by male CSFII participants who were aged 51 years or older (31). For those foods for which portion size was not asked, mean portions by sex were assigned.
Questionnaire responses were converted to the number of pyramid servings of each of the Food Guide Pyramid's food groups (24) by use of the Pyramid Servings Database corresponding to the 1994–1996 CSFII (30), as previously described (32). Briefly, the data files provide the numbers of servings from each Food Guide Pyramid group per 100 g of the food, for each of more than 5000 foods, including mixed dishes; recipes for mixed dishes were used to apportion each item into constituent foods (23). Pyramid definitions were used to describe serving sizes: a serving of fruit is defined as a medium-sized whole fruit, three-fourths of a cup (i.e., 6 ounces or 178 mL) of fruit juice, or one-half cup (i.e., 119 mL) of cut-up fruit. A serving of vegetables is defined as one cup (i.e., 237 mL) of leafy vegetables, one-half cup of other vegetables, or three-fourths cup of vegetable juice (24). As examples, one food-frequency questionnaire–reported portion of broccoli translates into 1.36 pyramid servings, a serving of raw spinach translates to 0.92 pyramid servings, and two slices of pizza (medium) translates to 0.42 pyramid vegetable servings. Vegetables included cooked dried beans and potatoes. Daily, total, and subgroup-specific fruit and vegetable intakes were derived by summing the daily pyramid servings across all relevant food items.
Data Analysis
Person-years were calculated from the date of the baseline PSA screening for prostate cancer at study entry to the date of the most recently completed endpoint follow-up questionnaire or the date of prostate cancer diagnosis, death, or October 1, 2001, whichever came first. During the study period of November 1, 1993, to October 1, 2001, 9% of the cohort died or were lost to follow-up. Because the PLCO study is an ongoing randomized clinical trial, incidence rates are not presented in this article. To evaluate the risk of prostate cancer, we used Cox proportional hazards regression analysis, with age as the underlying time metric (33) to generate unadjusted and multivariable-adjusted relative risks (RRs) and 95% confidence intervals (CIs). The proportionality assumption was evaluated graphically by inspecting the log–log plots for the exposure variables under study and by testing for evidence of a statistical interaction with time. To consider risk for more clinically important prostate cancer, we also evaluated risks with respect to "aggressive" prostate cancer (stage III or IV tumors or tumors with a Gleason score of 7) and to extraprostatic cancer (stage III or IV tumors only). All P values were from two-sided statistical tests.
Missing values were imputed from the mean (for continuous variables) or mode (for categorical variables) of the known values (
1% of the data were missing in each instance). The two exceptions to this rule were the participants with missing physical activity information (<1% of total participants), who were assigned to the category of "no or low physical activity," and the participants with missing diabetes status (2.7% of total participants), who were assigned to the "no disease" category. Nonresponse to a food item was considered to indicate nonconsumption of the item. Models that included and excluded participants with missing information yielded similar values for the association between fruit and vegetable intake and prostate cancer risk.
For the analysis of prostate cancer risk, we categorized total fruit, total vegetable, and food group intakes into equal-sized quintiles of average daily intake. Individual fruit and vegetable items had discrete intake distributions, and category cut points for these items were chosen from reasonable ranges in number of servings. Multivariable analyses were adjusted for suspected prostate cancer risk factors, including age (by modeling age as the underlying time metric; continuous), total energy intake (quintiles; 0–1620, 1621–2019, 2020–2424, 2425–2984, or >2984 kcal/day), race (white, black, Asian/Pacific Islander, or other), study center, family history of prostate cancer (yes or no), body mass index (<25, 25 to <30, or 30 kg/m2), smoking status (never, current, former, or pipe or cigar only), physical activity (i.e., hours spent in vigorous activity per week; none, <1, 1, 2, 3, or 4), supplemental vitamin E intake (0, 0–30 IU/day, >30–400 IU/day, >400 IU/day, or past use), total fat intake (quintiles; 0–62, 63–72, 73–79, 80–88, or >88 g/day), red meat intake (quintiles; 0–44, 45–69, 70–98, 99–146, or >146 g/day), diabetes (yes or no), aspirin use (never, <1 pill per day, or 1 pill per day), and total number of prostate cancer screening examinations during the follow-up period (as a time-dependent variable; continuous). Nutrient values were adjusted for energy intake by using the residual method (34). Results were not statistically significantly altered by additional adjustment for total fruit or vegetable intake (as appropriate) and tomato intake or for history of PSA tests before study enrollment.
To study whether the association between the dietary exposure and the risk of prostate cancer differed statistically significantly between the first year of follow-up (which may have included a greater proportion of prevalent cases of prostate cancer) and the subsequent years of follow-up, we defined a time-dependent covariate as the product of time (1 versus >1 year after the start of follow-up) and dietary exposure of interest, with testing of the resulting coefficient(s) by a –2-log likelihood statistic; there was no evidence of violation of the proportionality assumption. We also conducted sensitivity analyses by excluding case patients diagnosed within the first year.
Tests for trend in relative risks were conducted by assigning the median value for each category and treating this variable as continuous by use of a Wald chi-square statistic. Spearman's correlation coefficients were calculated to estimate the correlation between fruit and vegetable intake and between intakes of specific vegetables.
|
Results |
|---|
|
TopAbstractContext and CaveatsParticipants and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Among 29361 men, 1338 (4.5%) were diagnosed with prostate cancer, including 520 (38.9%) with aggressive disease (i.e., stage III or IV prostate cancer or Gleason score of
7). Participants who had a PSA test in the 3 years before study enrollment reported greater fruit intake than those who did not (Table 1). Greater fruit and vegetable intakes were associated with greater physical activity, as well as greater energy, lycopene, fish, and supplemental vitamin E intake. Fruit and vegetable intakes were inversely associated with tobacco use and red meat consumption, and greater fruit intake was associated with lower fat and greater calcium intake. African Americans and Asians/Pacific Islanders were less likely than whites to have high vegetable intake. Fruit and vegetable intakes were positively correlated (r = .37). The range of reported fruit and vegetable intake among participants in the PLCO was wide, with threefold and sixfold differences in intake, respectively, between the highest and lowest quintile medians of vegetables and fruits (Table 2).
|
|
Fruit intake, including its subcategories, was not associated with the risk of prostate cancer, overall or of aggressive or extraprostatic disease (Table 2). Vegetable intake was not related to prostate cancer overall; however, risk of extraprostatic disease was statistically significantly lower in men who consumed more of these foods than men who consumed less (RRs for increasing quintiles of intake = 1.00 [referent]; 0.78, 95% CI = 0.51 to 1.21; 0.72, 95% CI = 0.45 to 1.15; 0.88, 95% CI = 0.55 to 1.39; 0.41, 95% CI = 0.22 to 0.74; Ptrend = .01), particularly cruciferous (RR = 0.60, 95% CI = 0.36 to 0.98, for high versus low intake; Ptrend = .02) and possibly all dark green (RR = 0.70, 95% CI = 0.42 to 1.16, for high versus low intake; Ptrend = .09) vegetables. Excluding cases diagnosed within the first year yielded similar findings for all vegetables (RR = 0.47, 95% CI = 0.21 to 1.09, for high versus low intake) and cruciferous vegetables (RR = 0.77, 95% CI = 0.39 to 1.53, for high versus low intake). No associations were noted for tomatoes, as previously reported in greater detail (35). Results were not markedly changed when we excluded potatoes and/or other starchy vegetables from the vegetable index.
In our study, the dark green vegetable group was comprised predominately of broccoli (52%), spinach (23%), mixed foods (e.g., beef stew or pot pie with vegetables or lasagna) (17%), and mustard/turnip greens (7%). Cruciferous vegetables included broccoli (36%); cole slaw, cabbage, and sauerkraut (26%); cauliflower (20%); Brussels sprouts (12%); and mustard or turnip greens (5%). Each of the specific crucifera, with the exception of mustard or turnip greens and Brussels sprouts, was associated with decreased risk, with the strongest inverse associations being observed for extraprostatic cancer with greater broccoli (RR = 0.55, 95% CI = 0.34 to 0.89, for >1 serving per week versus <1 serving per month; Ptrend = .02) and cauliflower (RR = 0.48, 95% CI = 0.25 to 0.89 for >1 serving per week versus <1 serving per month; Ptrend = .03) intake (Table 3). When patients with low-stage (I or II) prostate cancer with a Gleason sum of 7 (n = 258) were excluded from the aggressive category, inverse associations with high total, dark green, and cruciferous vegetable intakes were still not as strong as those for extraprostatic disease alone.
|
Risks for aggressive prostate cancer tended to be lower in association with higher intake of spinach, a component of the dark green vegetable group (RR = 0.63, 95% CI = 0.38 to 1.02, for
2 servings per week versus <1 serving per month; Ptrend = .03), compared with lower intake (Table 3), with similar results for raw (RRs for increasing categories = 1.00 [referent]; 0.87, 95% CI = 0.63 to 1.21; 0.78, 95% CI = 0.45 to 1.33; 0.68, 95% CI = 0.36 to 1.28; Ptrend = .11) and cooked (RRs for increasing categories = 1.00 [referent]; 0.77, 95% CI = 0.58 to 1.03; 1.06, 95% CI = 0.72 to 1.56; 0.39, 95% CI = 0.16 to 0.94; Ptrend = .04) spinach. We found no associations of prostate cancer with intake of deep yellow vegetables (namely, squash, carrots, and sweet potato) (Table 2) or of beans, tofu and soybeans, onions, or garlic (Table 4), either overall or by disease subgroup (data not shown).
|
|
Discussion |
|---|
|
TopAbstractContext and CaveatsParticipants and MethodsResultsDiscussionFundingReferencesNotes |
|---|
In this large cohort study, total fruit and total vegetable consumption was not related to overall risk of prostate cancer; however, risk of extraprostatic disease was statistically significantly lower in association with greater vegetable intake, particularly for cruciferous and possibly dark green vegetables. Among the crucifera, broccoli and cauliflower showed a strong inverse association with aggressive and extraprostatic prostate cancer. We also noted a statistically significant inverse trend with increasing spinach consumption that was restricted to aggressive prostate cancer.
Our findings are consistent with six cohort studies (9–15) that found no relationship between vegetable intake and overall risk of prostate cancer, although in three of these cohorts (9–11) weak inverse associations were noted, and nine of 13 case–control studies reported reduced risks, four of which reached statistical significance (1). Previous cohort reports have not explicitly related total vegetable intake to risk for extraprostatic prostate cancer. Two case–control studies and one cohort study assessed the associations with aggressive disease. Kolonel et al. (36) reported an overall inverse association for vegetable intake that became more pronounced when restricted to aggressive disease, and Hayes et al. (37) and Stram et al. (15) found no association.
Prospective results for associations between cruciferous vegetable intake and prostate cancer risk have been inconsistent, with two cohorts (10,16) showing inverse associations—only one (16) in men who had PSA screening—and three (9,14,15) showing no association. Four case–control studies (36,38–40) reported inverse findings for cruciferous vegetables, with marked reductions in risk for high versus low cruciferous vegetable intake reported by Kolonel et al. (36) (RR = 0.61, 95% CI = 0.42 to 0.88, for risk of aggressive disease), Cohen et al. (39) (RR = 0.59, 95% CI = 0.39 to 0.90), and Joseph et al. (40) (RR = 0.58, 95% CI = 0.38 to 0.89). Results among the PSA-screened subgroup in the large study of Giovannucci et al. (16) (which was restricted to men who had had at least one PSA test in the previous 6 years) are likely more valid than those from the main analyses—in which no associations were found—because they are less subject to detection bias and confounding (22). The differential findings underlie the importance of control for screening and indicate that major strengths of our study are the screening setting and careful control for precise number of PSA tests.
Only three studies have examined cruciferous vegetable intake and extraprostatic or aggressive disease: this study and the case–control study of Kolonel et al. (36) provided evidence that intake of cruciferous vegetables was associated with decreased risk of such disease, whereas that of Giovannucci et al. (16) did not. Dark green vegetable intake was associated with a nonstatistically significantly decreased risk of advanced disease in one case–control study (15) and an increased risk in one cohort study (36); neither study considered spinach specifically. Our study is, to our knowledge, the first to report an inverse association between spinach consumption and risk of advanced prostate cancer.
Epidemiologic studies have also variously identified legumes and/or soy foods (19,20,36,38,41–45), green–yellow vegetables (46,47), carrots (36,41,48), tomatoes (13,20,38,49,50), and allium vegetables (50,51) as being inversely associated with prostate cancer risk. In contrast, we did not note substantial associations for these foods. However, the mean intake of tofu or soybeans was extremely low in our study population (on average, one half serving per month), and we cannot exclude the possibility of effects at higher levels of intake. Similarly, the limited range of intake of garlic and onions (components of the allium group) in our study could contribute to the null findings for this group.
Consistent with our results, none of the 10 prospective studies showed inverse associations between total fruit intake and risk of prostate cancer (reviewed in 1,9–14,17–20); case–control studies also noted absence of protective associations (1,52). We found moderately strongly increased relative risks across levels of fruit juice intake (apple, orange, or other); however, the changes were not linear (Ptrend = .38). Fruit juices represent a source of citrus fruit, and others (38,42) have reported increased risk of prostate cancer associated with high citrus fruit consumption. The importance of the finding regarding fruit juice intake is, however, uncertain.
Consumption of cruciferous vegetables and spinach could (if the association is assumed to be causal) confer protection through various mechanisms. Cruciferous vegetables are especially rich sources of glucosinolates; certain hydrolytic products of which including isothiocyanates, indoles, and sulforaphane, have anticarcinogenic effects (6,7,53,54). These effects include induction of phase II detoxification enzymes, which protect cells from DNA-damaging agents (5,55); inadequate levels of these enzymes is consistent with the loss of expression of the phase II enzyme glutathoine S-transferase-
, which is a characteristic of prostate cancer (56,57). Antioxidant activity is another plausible mechanism for protection (2,58). Broccoli and spinach have particularly high antioxidant and antiproliferative activities (59), and because oxidative damage may be higher in metastatic prostate cancer than in primary prostate cancer (60), these activities could account for our findings of a benefit of cruciferous and dark green vegetables that is largely restricted to advanced prostate cancer. Broccoli and spinach are also high in free phenolics (e.g., flavonoids) (59), which, in addition to having antioxidant effects (61), have been shown to stimulate the immune system, decrease the level of DNA adducts, and inhibit the mutagenicity of aromatic and heterocyclic amines (62,63).
Our study has limitations that are inherent to all observational studies. It is well established that individuals who have diets rich in fruits and vegetables generally have lower rates of smoking, higher levels of physical exercise, and a more healthy lifestyle overall, and this association could confound the prostate cancer association. Although adjustment for a wide range of potentially confounding factors did not alter the inverse associations that we observed, the possibility of residual confounding still remains because a high intake of fruits and vegetables might also be an indicator of unmeasurable behavioral and psychosocial factors that we are not able to adjust for in analyses. Arguing against the potential for residual confounding, however, is the specificity of the inverse association with vegetables but not fruit. Also, we were limited in our ability to evaluate results by ethnic or racial subgroups because of small numbers and concerns regarding comparability of dietary assessment across these groups.
An important strength of our study is the ability to control for screening for prostate cancer. Health behaviors, such as PSA screening, are likely to be associated with fruit and vegetable intake (22), and PSA screening is strongly associated with prostate cancer (64). The nature of this randomized screening trial, however, eliminates differential screening practices because all participants underwent screening according to the same prescribed regimen. In addition, we assessed the potential for detection bias by creating a time-dependent variable representing total number of prostate screening examinations; in this way, at any given time point, only participants who had had the same number of screening examinations (assessed since baseline) and, thus, the same opportunity for cancer detection were being compared. There is also the concern that the etiology of PSA-detected early-stage disease might differ from aggressive prostate disease and that epidemiologic studies conducted in the PSA era will be enriched with clinically latent prostate cancer and thus fail to replicate positive findings from earlier studies (65). We addressed this possibility by stratifying by aggressive versus nonaggressive disease status.
The total fruit and vegetable intake values in our study population were higher than reports from national surveys and other epidemiologic studies. By applying CSFII 1994–1996 sex- and age-specific intake values (66) to the distribution of our population, we would have expected the mean fruit intake to be approximately 1.9 servings per day and the mean vegetable intake to be approximately 3.9 servings per day. The observed values were 3.3 and 5.4, respectively. Because all men were participating in a cancer screening trial, they were better educated (67) and might be more health conscious than men in the general population. The men in our study population may, however, have overreported their intake, perhaps because of greater awareness of the importance of fruit and vegetable intake, particularly since the National Cancer Institute introduced the 5-A-Day Program in 1991 (25). Also, it is possible that our exposure assessment may be subject to less nondifferential misclassification than other epidemiologic studies, none of which report including mixed dishes in their fruit and vegetable indices. Some of the heterogeneity observed across studies with respect to quantifying dietary intake may be caused by the inherent measurement error in dietary assessment instruments and in food-frequency questionnaires, in particular (68). Because of the prospective nature of our data collection, we would not expect intakes to be biased by disease status, and so nondifferential error would not account for our inverse findings.
In conclusion, these findings indicate that intakes of cruciferous and dark green vegetables, especially broccoli and cauliflower, are associated with a decreased risk of aggressive, particularly extraprostatic, prostate cancer. Aggressive prostate cancer is biologically virulent and associated with poor prognosis (69). Therefore, if the association that we observed is ultimately found to be causal, a possible means to reduce the burden of this disease may be primary prevention through increased consumption of broccoli, cauliflower, and possibly spinach.
|
Funding |
|---|
|
TopAbstractContext and CaveatsParticipants and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Division of Cancer Prevention and Intramural Research Program, National Cancer Institute, National Institutes of Health, U. S. Department of Health and Human Services.
|
NOTES |
|---|
|
TopAbstractContext and CaveatsParticipants and MethodsResultsDiscussionFundingReferencesNotes |
|---|
The authors thank Drs Christine Berg and Philip Prorok, Division of Cancer Prevention, National Cancer Institute; the Screening Center investigators and staff of the PLCO Cancer Screening Trial; Mr. Tom Riley and staff, Information Management Services, Inc; Ms Barbara O’Brien and staff, Westat, Inc. Most importantly, we acknowledge the study participants for their contributions to making this study possible.
The leading and corresponding authors, as well as several co-authors, were or are employed by the study sponsor and were directly involved in the design of the study; the collection, analysis, and interpretation of the data; the writing of the manuscript; and the decision to submit the manuscript for publication to the journal.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsParticipants and MethodsResultsDiscussionFundingReferencesNotes |
|---|
(1) Vainio H, Bianchini F. Fruit and vegetables. In: IARC handbooks of cancer prevention (2003) Lyon (France): IARC Press. 376.
(2) Steinmetz KA, Potter JD. Vegetables, fruit, and cancer. II. Mechanisms. Cancer Causes Control (1991) 2:427–42.[CrossRef][Web of Science][Medline]
(3) Liu RH. Potential synergy of phytochemicals in cancer prevention: mechanism of action. J Nutr (2004) 134:3479S–85S.
(4) Kristal AR, Lampe JW. Brassica vegetables and prostate cancer risk: a review of the epidemiological evidence. Nutr Cancer (2002) 42:1–9.[CrossRef][Web of Science][Medline]
(5) Brooks JD, Paton VG, Vidanes G. Potent induction of phase 2 enzymes in human prostate cells by sulforaphane. Cancer Epidemiol Biomarkers Prev (2001) 10:949–54.
(6) Singh SV, Srivastava SK, Choi S, Lew KL, Antosiewicz Xiao D, et al. Sulforaphane-induced cell death in human prostate cancer cells is initiated by reactive oxygen species. J Biol Chem (2005) 280:19911–24.
(7) Fan S, Meng Q, Auborn K, Carter T, Rosen EM. BRCA1 and BRCA2 as molecular targets for phytochemicals indole-3-carbinol and genistein in breast and prostate cancer cells. Br J Cancer (2006) 94:407–26.[CrossRef][Web of Science][Medline]
(8) Xiao D, Srivastava SK, Lew KL, Zeng Y, Hershberger P, Johnson CS, et al. Allyl isothiocyanate, a constituent of cruciferous vegetables, inhibits proliferation of human prostate cancer cells by causing G2/M arrest and inducing apoptosis. Carcinogenesis (2003) 24:891–7.
(9) Hsing AW, McLaughlin JK, Schuman LM, Bjelke E, Gridley G, Wacholder S, et al. Diet, tobacco use, and fatal prostate cancer: results from the Lutheran Brotherhood Cohort Study. Cancer Res (1990) 50:6836–40.
(10) Schuurman AG, Goldbohm RA, Dorant E, van den Brandt PA. Vegetable and fruit consumption and prostate cancer risk: a cohort study in The Netherlands. Cancer Epidemiol Biomarkers Prev (1998) 7:673–80.[Abstract]
(11) Chan JM, Pietinen P, Virtanen M, Malila N, Tangrea J, Albanes D, et al. Diet and prostate cancer risk in a cohort of smokers, with a specific focus on calcium and phosphorus (Finland). Cancer Causes Control (2000) 11:859–67.[CrossRef][Web of Science][Medline]
(12) Shibata A, Paganini-Hill A, Ross RK, Henderson BE. Intake of vegetables, fruits, beta-carotene, vitamin C and vitamin supplements and cancer incidence among the elderly: a prospective study. Br J Cancer (1992) 66:673–9.[Web of Science][Medline]
(13) Giovannucci E, Ascherio A, Rimm EB, Stampfer MJ, Colditz GA, Willett WC. Intake of carotenoids and retinol in relation to risk of prostate cancer. J Natl Cancer Inst (1995) 87:1767–76.
(14) Key TJ, Allen N, Appleby P, Overvad K, Tjonneland A, Miller A, et al. Fruits and vegetables and prostate cancer: no association among 1,104 cases in a prospective study of 130,544 men in the European Prospective Investigation into Cancer and Nutrition (EPIC). Int J Cancer (2004) 109:119–24.[CrossRef][Web of Science][Medline]
(15) Stram DO, Hankin JH, Wilkens LR, Park S, Henderson BE, Nomura AM, et al. Prostate cancer incidence and intake of fruits, vegetables and related micronutrients: the multiethnic cohort study* (United States). Cancer Causes Control (2006) 17:1193–207.[CrossRef][Web of Science][Medline]
(16) Giovannucci E, Rimm EB, Liu Y, Stampfer MJ, Willett WC. A prospective study of cruciferous vegetables and prostate cancer. Cancer Epidemiol Biomarkers Prev (2003) 12:1403–9.
(17) Appleby PN, Key TJ, Burr ML, Thorogood M. Mortality and fresh fruit consumption. IARC Sci Publ (2002) 156:131–3.[Medline]
(18) Le Marchand L, Hankin JH, Kolonel LN, Wilkens LR. Vegetable and fruit consumption in relation to prostate cancer risk in Hawaii: a reevaluation of the effect of dietary beta-carotene. Am J Epidemiol (1991) 133:215–9.
(19) Severson RK, Nomura AM, Grove JS, Stemmermann GN. A prospective study of demographics, diet, and prostate cancer among men of Japanese ancestry in Hawaii. Cancer Res (1989) 49:1857–60.
(20) Mills PK, Beeson WL, Phillips RL, Fraser GE. Cohort study of diet, lifestyle, and prostate cancer in Adventist men. Cancer (1989) 64:598–604.[CrossRef][Web of Science][Medline]
(21) Le Marchand L, Kolonel LN, Wilkens LR, Myers BC, Hirohata T. Animal fat consumption and prostate cancer: a prospective study in Hawaii. Epidemiology (1994) 5:276–82.[Web of Science][Medline]
(22) Kristal AR, Stanford JL. Cruciferous vegetables and prostate cancer risk: confounding by PSA screening. Cancer Epidemiol Biomarkers Prev (2004) 13:1265.
(23) Cook AJ, Friday JE. Pyramid servings database for USDA survey food codes. Version 2.0. (2004) Washington (DC): Agricultural Research Service, US Department of Agriculture.
(24) Shaw A, Fulton L, Davis C, Hogbin M. Using the Food Guide Pyramid: a resource for nutrition educators (2005) Washington (DC): Center for Nutrition Policy and Promotion, US Department of Agriculture.
(25) National Cancer Institute. Five a day for better health (2004) Bethesda (MD). Available at: http://www.5aday.gov. [Last accessed: June 29, 2007.].
(26) Hayes RB, Sigurdson A, Moore L, Peters U, Huang WY, Pinsky P, et al. Methods for etiologic and early marker investigations in the PLCO trial. Mutat Res (2005) 592:147–54.[Web of Science][Medline]
(27) Prorok PC, Andriole GL, Bresalier RS, Buys SS, Chia D, Crawford ED, et al. Design of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Control Clin Trials (2000) 21:273S–309S.[CrossRef][Web of Science][Medline]
(28) Simpson NK, Johnson CC, Ogden SL, Gamito E, Trocky N, McGuire C, et al. Recruitment strategies in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial: the first six years. Control Clin Trials (2000) 21:356S–78S.[CrossRef][Web of Science][Medline]
(29) Fleming ID, Cooper JS, Henson DE, Hutter RVP, Kennedy BJ, Murphy GP, et al, eds. AJCC Cancer Staging Manual (1997) 5th ed. Philadelphia (PA): Lippincott-Raven.
(30) Tippett KS, Cypel YS, eds. Design and operation: the continuing survey of food intakes by individuals and the diet and health knowledge survey, 1994–96. In: Nationwide Food Surveys Report No.: 96-1 (1998) Washington (DC): Agricultural Research Service, US Department of Agriculture.
(31) Subar AF, Midthune D, Kulldorff M, Brown CC, Thompson FE, Kipnis V, et al. Evaluation of alternative approaches to assign nutrient values to food groups in food frequency questionnaires. Am J Epidemiol (2000) 152:279–86.
(32) Cleveland LE, Cook DA, Krebs-Smith SM, Friday J. Method for assessing food intakes in terms of servings based on food guidance. Am J Clin Nutr (1997) 65:1254S–63S.[Medline]
(33) Korn EL, Graubard BI, Midthune D. Time-to-event analysis of longitudinal follow-up of a survey: choice of the time-scale. Am J Epidemiol (1997) 145:72–80.
(34) Willett W, Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol (1986) 124:17–27.
(35) Kirsh VA, Mayne ST, Peters U, Chatterjee N, Leitzmann MF, Dixon LB, et al. A prospective study of lycopene and tomato product intake and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev (2006) 15:92–8.
(36) Kolonel LN, Hankin JH, Whittemore AS, Wu AH, Gallagher RP, Wilkens LR, et al. Vegetables, fruits, legumes and prostate cancer: a multiethnic case-control study. Cancer Epidemiol Biomarkers Prev (2000) 9:795–804.
(37) Hayes RB, Ziegler RG, Gridley G, Swanson C, Greenberg RS, Swanson GM, et al. Dietary factors and risks for prostate cancer among blacks and whites in the United States. Cancer Epidemiol Biomarkers Prev (1999) 8:25–34.
(38) Jain MG, Hislop GT, Howe GR, Ghadirian P. Plant foods, antioxidants, and prostate cancer risk: findings from case-control studies in Canada. Nutr Cancer (1999) 34:173–84.[CrossRef][Web of Science][Medline]
(39) Cohen JH, Kristal AR, Stanford JL. Fruit and vegetable intakes and prostate cancer risk. J Natl Cancer Inst (2000) 92:61–8.
(40) Joseph MA, Moysich KB, Freudenheim JL, Shields PG, Bowman ED, Zhang Y, et al. Cruciferous vegetables, genetic polymorphisms in glutathione S-transferases M1 and T1, and prostate cancer risk. Nutr Cancer (2004) 50:206–13.[CrossRef][Web of Science][Medline]
(41) Schuman LM, Mandel JS, Radke A, Seal U, Halberg F. Some selected features of the epidemiology of prostatic cancer: Minneapolis-St. Paul, Minnesota case control study, 1976–1979. In: Trends in cancer incidence: causes and practical implications—Magnus K, ed. (1982) Washington (DC): Hemisphere Publishing Corp. 345–54.
(42) Key TJ, Silcocks PB, Davey GK, Appleby PN, Bishop DT. A case-control study of diet and prostate cancer. Br J Cancer (1997) 76:678–87.[Web of Science][Medline]
(43) Schuurman AG, van den Brandt PA, Dorant E, Goldbohm RA. Animal products, calcium and protein and prostate cancer risk in The Netherlands Cohort Study. Br J Cancer (1999) 80:1107–13.[CrossRef][Web of Science][Medline]
(44) Jacobsen BK, Knutsen SF, Fraser GE. Does high soy milk intake reduce prostate cancer incidence? The Adventist Health Study (United States). Cancer Causes Control (1998) 9:553–7.[CrossRef][Web of Science][Medline]
(45) Lee MM, Gomez SL, Chang JS, Wey M, Wang RT, Hsing AW. Soy and isoflavone consumption in relation to prostate cancer risk in China. Cancer Epidemiol Biomarkers Prev (2003) 12:665–8.
(46) Hirayama T. Epidemiology of prostate cancer with special reference to the role of diet. Natl Cancer Inst Monogr (1979) (53):149–55.
(47) Ohno Y, Yoshida O, Oishi K, Okada K, Yamabe H, Schroeder FH. Dietary beta-carotene and cancer of the prostate: a case-control study in Kyoto, Japan. Cancer Res (1988) 48:1331–6.
(48) Walker AR, Walker BF, Tsotetsi NG, Sebitso C, Siwedi D, Walker AJ. Case-control study of prostate cancer in black patients in Soweto, South Africa. Br J Cancer (1992) 65:438–41.[Web of Science][Medline]
(49) Tzonou A, Signorello LB, Lagiou P, Wuu J, Trichopoulos D, Trichopoulou A. Diet and cancer of the prostate: a case-control study in Greece. Int J Cancer (1999) 80:704–8.[CrossRef][Web of Science][Medline]
(50) Hodge AM, English DR, McCredie MR, Severi G, Boyle P, Hopper JL, et al. Foods, nutrients and prostate cancer. Cancer Causes Control (2004) 15:11–20.[CrossRef][Web of Science][Medline]
(51) Hsing AW, Chokkalingam AP, Gao YT, Madigan MP, Deng J, Gridley G, et al. Allium vegetables and risk of prostate cancer: a population-based study. J Natl Cancer Inst (2002) 94:1648–51.
(52) Chan JM, Giovannucci EL. Vegetables, fruits, associated micronutrients, and risk of prostate cancer. Epidemiol Rev (2001) 23:82–6.
(53) Bradlow HL, Sepkovic DW, Telang NT, Osborne MP. Multifunctional aspects of the action of indole-3-carbinol as an antitumor agent. Ann NY Acad Sci (1999) 889:204–13.[CrossRef][Web of Science][Medline]
(54) Chinni SR, Li Y, Upadhyay S, Koppolu PK, Sarkar FH. Indole-3-carbinol (I3C) induced cell growth inhibition, G1 cell cycle arrest and apoptosis in prostate cancer cells. Oncogene (2001) 20:2927–36.[CrossRef][Web of Science][Medline]
(55) Zhang Y, Talalay P, Cho CG, Posner GH. A major inducer of anticarcinogenic protective enzymes from broccoli: isolation and elucidation of structure. Proc Natl Acad Sci USA (1992) 89:2399–403.
(56) Brooks JD, Weinstein M, Lin X, Sun Y, Pin SS, Bova GS, et al. CG island methylation changes near the GSTP1 gene in prostatic intraepithelial neoplasia. Cancer Epidemiol Biomarkers Prev (1998) 7:531–6.
(57) Sakr WA, Haas GP, Cassin BF, Pontes JE, Crissman JD. The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients. J Urol (1993) 150:379–85.[Web of Science][Medline]
(58) Rock CL, Lampe JW, Patterson RE. Nutrition, genetics, and risks of cancer. Annu Rev Public Health (2000) 21:47–64.[CrossRef][Web of Science][Medline]
(59) Chu YF, Sun J, Wu X, Liu RH. Antioxidant and antiproliferative activities of common vegetables. J Agric Food Chem (2002) 50:6910–6.[CrossRef][Web of Science][Medline]
(60) Oberley TD, Zhong W, Szweda LI, Oberley LW. Localization of antioxidant enzymes and oxidative damage products in normal and malignant prostate epithelium. Prostate (2000) 44:144–55.[CrossRef][Web of Science][Medline]
(61) Pietta PG. Flavonoids as antioxidants. J Nat Prod (2000) 63:1035–42.[CrossRef][Medline]
(62) Sarkar FH, Li Y. Soy isoflavones and cancer prevention. Cancer Invest (2003) 21:744–57.[CrossRef][Web of Science][Medline]
(63) Nakahara K, Roy MK, Ono H, Maeda I, Ohnishi-Kameyama M, Yoshida M, et al. Prenylated flavanones isolated from flowers of Azadirachta indica (the neem tree) as antimutagenic constituents against heterocyclic amines. J Agric Food Chem (2003) 51:6456–60.[CrossRef][Web of Science][Medline]
(64) Stanford JL, Wicklund KG, McKnight B, Daling JR, Brawer MK. Vasectomy and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev (1999) 8:881–6.
(65) Platz EA, De Marzo AM, Giovannucci E. Prostate cancer association studies: pitfalls and solutions to cancer misclassification in the PSA era. J Cell Biochem (2004) 91:553–71.[CrossRef][Web of Science][Medline]
(66) US Department of Agriculture. Food and nutrient intakes by individuals in the United States, by sex and age, 1994–1996, Nationwide Food Surveys Report No. 96-2 (1998) Beltsville (MD): Agricultural Research Service. 197.
(67) Weissfeld JL, Schoen RE, Pinsky PF, Bresalier RS, Church T, Yurgalevitch S, et al. Flexible sigmoidoscopy in the PLCO cancer screening trial: results from the baseline screening examination of a randomized trial. J Natl Cancer Inst (2005) 97:989–97.
(68) Bingham SA. Measurement error in the assessment of interaction between dietary and genetic factors in cohort studies of cancer. World Rev Nutr Diet (2005) 94:144–51.[Medline]
(69) Gloeckler Ries LA, Reichman ME, Lewis DR, Hankey BF, Edwards BK. Cancer survival and incidence from the Surveillance, Epidemiology, and End Results (SEER) program. Oncologist (2003) 8:541–52.
Manuscript received February 15, 2007; revised May 31, 2007; accepted June 19, 2007.
Related Article in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 1137.
This article has been cited by other articles:
|
|
 |
|
  G Kallifatidis, V Rausch, B Baumann, A Apel, B M Beckermann, A Groth, J Mattern, Z Li, A Kolb, G Moldenhauer, et al. Sulforaphane targets pancreatic tumour-initiating cells by NF-{kappa}B-induced antiapoptotic signalling Gut, July 1, 2009; 58(7): 949 - 963. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. S. Khwaja, S. Wynne, I. Posey, and D. Djakiew 3,3'-Diindolylmethane Induction of p75NTR-Dependent Cell Death via the p38 Mitogen-Activated Protein Kinase Pathway in Prostate Cancer Cells Cancer Prevention Research, June 1, 2009; 2(6): 566 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Kristal and S. M. Lippman Nutritional Prevention of Cancer: New Directions for an Increasingly Complex Challenge J Natl Cancer Inst, March 18, 2009; 101(6): 363 - 365. [Full Text] [PDF]
|
|
|
|
|
|
S. T. Okino, D. Pookot, S. Basak, and R. Dahiya Toxic and Chemopreventive Ligands Preferentially Activate Distinct Aryl Hydrocarbon Receptor Pathways: Implications for Cancer Prevention Cancer Prevention Research, March 1, 2009; 2(3): 251 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Kim and O. Kwon Garlic intake and cancer risk: an analysis using the Food and Drug Administration's evidence-based review system for the scientific evaluation of health claims Am. J. Clinical Nutrition, January 1, 2009; 89(1): 257 - 264. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Smith, D. Sepkovic, H. L. Bradlow, and K. J. Auborn 3,3'-Diindolylmethane and Genistein Decrease the Adverse Effects of Estrogen in LNCaP and PC-3 Prostate Cancer Cells J. Nutr., December 1, 2008; 138(12): 2379 - 2385. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Tang, G. R. Zirpoli, K. Guru, K. B. Moysich, Y. Zhang, C. B. Ambrosone, and S. E. McCann Consumption of Raw Cruciferous Vegetables is Inversely Associated with Bladder Cancer Risk Cancer Epidemiol. Biomarkers Prev., April 1, 2008; 17(4): 938 - 944. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Another Reason to Eat Your Vegetables: Effect on Risk for Aggressive Prostate Cancer Journal Watch Oncology and Hematology, August 14, 2007; 2007(814): 6 - 6. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||