Abstract
Background
Many studies have suggested that nutritional factors may affect prostate cancer development. The aim of our study was to evaluate the relationship between dietary habits and prostate cancer detection.
Methods
We studied 917 patients who planned to have transrectal ultrasonography–guided prostatic biopsy based on an elevated serum prostate-specific antigen (PSA) level, a rising serum PSA level or an abnormal digital rectal examination. Before receiving the results of their biopsy, all patients answered a self-administered food frequency questionnaire. In combination with pathology data we performed univariable and multivariable logistic regression analyses for the predictors of cancer and its aggressiveness.
Results
Prostate cancer was found in 42% (386/917) of patients. The mean patient age was 64.5 (standard deviation [SD] 8.3) years and the mean serum PSA level for prostate cancer and benign cases, respectively, was 13.4 (SD 28.2) μg/L and 7.3 (SD 4.9) μg/L. Multivariable analysis revealed that a meat diet (e.g., red meat, ham, sausages) was associated with an increased risk of prostate cancer (odds ratio [OR] 2.91, 95% confidence interval [CI] 1.55–4.87, p = 0.027) and a fish diet was associated with less prostate cancer (OR 0.54, 95% CI 0.32–0.89, p = 0.017). Aggressive tumours were defined by Gleason score (≥ 7), serum PSA level (≥ 10 μg/L) and the number of positive cancer cores (≥ 3). None of the tested dietary components were found to be associated with prostate cancer aggressivity.
Conclusion
Fish diets appear to be associated with less risk of prostate cancer detection, and meat diets appear to be associated with a 3-fold increased risk of prostate cancer. These observations add to the growing body of evidence suggesting a relationship between diet and prostate cancer risk.
Introduction
The incidence and mortality rates of prostate cancer vary widely among countries. The lowest prostate cancer incidence and mortality rates are observed in the Far East and on the Indian subcontinent, and the highest rates occur in western Europe, Australia and North America, with up to a 30-fold variation between highest and lowest rates.1,2 Interestingly, there are differences within the same ethnic groups, such as Japanese living in the United States, who have a 4–5 times higher incidence of prostate cancer than those living in Japan.3 Environmental and lifestyle factors, including dietary habits, are suggested as determinants. Total energy intake,4,5 fat intake, especially of animal sources,5–7 and dairy products and calcium8,9 have been associated with a positive risk.10 Among fruits and vegetables with suggested protective effects are tomatoes11 and yellow-orange and cruciferous vegetables,12 although other studies13,14 have not shown this association. A fish diet, with its unique marine omega-3 polyunsaturated fatty acids, has been suggested by some studies15,16 to protect against prostate cancer although others17 have found no association. We studied the association between different dietary items and the detection of prostate cancer in a cohort of high-risk men.
Methods
Our study included 1356 patients who underwent transrectal ultrasonography–guided prostatic biopsy at the McGill Prostate Cancer Detection Clinic in Montréal, Que., between August 2003 and November 2006. The indication for biopsy was based on an abnormally elevated serum prostate-specific antigen (PSA) level, a rising serum PSA level or a suspicious digital rectal examination. All men were asked to respond to a self-administered food frequency questionnaire (Appendix 1) before the procedure. Of these patients, 917 (68%) agreed and completed the questionnaire, which measured the number of servings per week among 12 food groups. The questionnaire also included demographic data (i.e., age, marital status, ethnicity, occupation and level of education), smoking habits, alcohol intake, family history of prostate cancer, medical history of general diseases (e.g., diabetes mellitus, liver diseases, heart diseases and arthritis) and local genitourinary conditions (e.g., cystitis, prostatitis, sexually transmitted infections and vasectomy). Transrectal ultrasonography–guided prostatic biopsy was performed via the standard approach as described by Tanguay and colleagues18 with a median of 10 biopsy cores (range 6–14). All biopsies were examined by the same uropathologist (L.B.).
The dietary intakes were divided into quintiles based on distributions. To assess recall bias we compared 50 (5.5%) patients who had repeated the questionnaire in their follow-up visits after 3 to 12 months. There was no statistically significant difference between the questionnaire answers in both sessions (p = 0.11) using the χ2 test. We performed univariable and multivariable logistic regression analyses between the highest and lowest quintile of each food group and the detection of prostate cancer, and we estimated the odds ratio (OR) of prostate cancer and 95% confidence interval (CI) for food groups with adjustments for age, ethnicity, level of education, family history of prostate cancer, smoking and alcohol consumption, sexually transmitted infections, cystitis and prostatitis. We compared prostate cancer cases (n = 386, 42.1%) and nonsignificant pathology (n = 268, 29.2%) after exclusion of cases with high-grade prostatic intraepithelial neoplasia and atypical small acinar proliferation. We defined aggressive prostate cancer by a Gleason score of 7 or greater, a serum PSA level of 10 μg/L or greater, or 3 or more malignant cores. All statistics were performed using Stata 9.2 (StataCorp).
Results
The mean age was 64.5 (standard deviation [SD] 8.3) years, and the mean serum PSA level for prostate cancer and benign cases was 13.4 (SD 28.2) μg/L and 7.3 (SD 4.9) μg/L, respectively. Table 1 demonstrates the baseline characteristics of both cases and controls with regard to age, ethnicity, level of education, family history of prostate cancer for father or brothers, smoking, alcohol consumption, history of vasectomy, sexually transmitted infections, cystitis and prostatitis. Both groups were similar except for statistically significant differences in the age between cases and controls (OR 3.71, 95% CI 2.59–5.30, p < 0.001). The rate of cancer appeared to be higher in African patients and lower in Asian patients but the difference was statistically insignificant.
Table 1.
Characteristics of study patients
| Characteristics | No. (and %) of cases | No. (and %) of controls | Odds ratio (95% CI) | p value |
|---|---|---|---|---|
| Age, yr | ||||
| 60–64 | 77 (20.0) | 72 (26.9) | 1.06 (0.77–1.47) | |
| 65–69 | 96 (24.9) | 68 (25.4) | 1.41 (1.03–1.92) | |
| ≥ 70 | 141 (36.5) | 38 (14.2) | 3.71 (2.59–5.30) | < 0.001 |
| Ethnicity* | ||||
| White | 212 (55.0) | 151 (56.3) | 1.00† | |
| Black | 13 (3.4) | 5 (1.9) | 2.60 (0.92–7.29) | |
| Asian | 16 (4.2) | 22 (8.2) | 0.72 (0.38–1.38) | |
| Hispanic | 9 (2.3) | 9 (3.4) | 1.00 (0.39–2.51) | |
| Other | 97 (25.9) | 66 (24.6) | 0.63 (0.49–0.81) | 0.250 |
| Level of education, yr* | ||||
| < 7 | 60 (15.5) | 32 (12.0) | 1.00† | |
| 7–11 | 132 (34.2) | 76 (28.4) | 1.71 (1.11–2.65) | |
| ≥ 12 | 152 (39.4) | 138 (51.5) | 1.10 (0.87–1.38) | 0.060 |
| Family history of prostate cancer | 23 (6.0) | 24 (9.0) | 0.95 (0.54–1.69) | 0.150 |
| Smoking habit, index,* cigarette packs/yr | ||||
| Never smoke | 129 (33.4) | 102 (38.1) | 1.00† | |
| < 200 | 35 (9.1) | 30 (11.2) | 1.16 (0.71–1.89) | |
| 200 to < 400 | 33 (8.6) | 23 (8.6) | 1.43 (0.84–2.44) | |
| ≥ 400 | 94 (24.4) | 69 (25.8) | 1.36 (0.99–1.85) | 0.600 |
| Alcohol, servings/wk* | ||||
| No alcohol | 41 (10.6) | 33 (12.3) | 1.00† | |
| 1 | 39 (10.1) | 39 (14.6) | 1.00 (0.64–1.55) | |
| 2–3 | 52 (13.5) | 30 (11.2) | 1.73 (1.10–2.71) | |
| 4–6 | 31 (8.0) | 29 (10.8) | 1.06 (0.64–1.77) | |
| ≥ 7 | 45 (11.7) | 33 (12.3) | 1.63 (1.06–2.52) | 0.330 |
| Vasectomy‡ | 70 (18.1) | 45 (16.8) | 1.09 (0.65–1.78) | 0.340 |
| STIs‡ | 101 (26.2) | 65 (24.3) | 1.13 (0.59–1.96) | 0.230 |
| Cystitis and/or prostatitis‡ | 58 (15.0) | 44 (16.4) | 1.23 (0.83–2.12) | 0.520 |
CI = confidence interval; STI = sexually transmitted infection.
Does not sum to total because of missing values.
Reference category.
Categorized variables.
Multivariable analysis of food groups as shown in Table 2 revealed that a meat diet (e.g., red meat, ham, sausages) was associated with a higher risk of prostate cancer (OR 2.91, 95% CI 1.55–4.87, p = 0.027) and a fish diet was associated with less prostate cancer (OR 0.54, 95% CI 0.32–0.89, p = 0.017) after adjustment with suspected confounders (e.g., age, ethnicity, family history of prostate cancer).
Table 2.
Multivariable odds ratios of food groups and the risk of prostate cancer*
| Quintiles† |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1
|
2
|
3
|
4
|
5
|
|||||||
| Food | Servings/ wk | Risk of prostate cancer, OR‡ | Servings/ wk | Risk of prostate cancer, OR (95% CI) | Servings/wk | Risk of prostate cancer, OR (95% CI) | Servings/wk | Risk of prostate cancer, OR (95% CI) | Servings/wk | Risk of prostate cancer, OR (95% CI) | p value |
| Dairy products | 2 | 1.00 | 4 | 1.44 (0.99–2.18) | 6 | 1.26 (0.94–2.21) | 7 | 1.35 (0.84–1.91) | NA | 1.21 (0.74–2.01) | 0.240 |
| Red meat, ham, sausages | 1 | 1.00 | 2 | 1.55 (0.85–1.69) | 4 | 1.97 (0.74–2.73) | 5 | 2.31 (1.32–2.46) | NA | 2.91 (1.55–4.87) | 0.027 |
| Chicken | 1 | 1.00 | 2 | 0.89 (0.52–1.48) | 3 | 1.02 (0.56–1.84) | 4 | 1.26 (0.58–2.75) | NA | 0.67 (0.28–1.53) | 0.340 |
| Fish | 1 | 1.00 | 2 | 1.23 (0.67–1.45) | 3 | 0.97 (0.95–1.56) | 4 | 0.77 (0.26–0.91) | NA | 0.54 (0.32–0.89) | 0.017 |
| Bread, pasta, rice, maize | 2 | 1.00 | 4 | 1.70 (0.86–3.66) | 6 | 1.81 (0.84–3.86) | 7 | 1.67 (0.91–3.05) | NA | 2.22 (0.87–5.66) | 0.090 |
| Potatoes | 1 | 1.00 | 2 | 1.09 (0.60–1.98) | 3 | 0.65 (0.34–1.25) | 4 | 0.63 (0.30–1.31) | NA | 0.57 (031–1.08) | 0.080 |
| Any green vegetables | 2 | 1.00 | 4 | 1.29 (0.63–2.60) | 6 | 1.66 (0.81–3.40) | 7 | 1.82 (0.87–3.82) | NA | 1.60 (0.56–4.57) | 0.370 |
| Tomatoes (e.g., fresh, sauce) | 2 | 1.00 | 4 | 0.95 (0.56–1.63) | 6 | 0.72 (0.39–1.32) | NA | 1.08 (0.54–2.16) | NA | NA | 0.810 |
| Pulses (e.g., beans, peas) | 2 | 1.00 | 4 | 0.98 (0.50–1.91) | 5 | 1.20 (0.59–2.43) | NA | 0.78 (0.34–1.77) | NA | NA | 0.550 |
| Fruits (e.g., fresh, juices) | 2 | 1.00 | 4 | 1.59 (0.98–2.57) | 6 | 1.15 (0.72–1.83) | 7 | 1.49 (1.17–1.91) | NA | 1.03 (0.63–1.66) | 0.590 |
| Soy products | 2 | 1.00 | 4 | 1.22 (0.60–2.53) | NA | 0.98 (0.45–2.14) | NA | NA | NA | NA | 0.430 |
| Cakes, dessert, ice cream | 1 | 1.00 | 3 | 1.34 (0.92–1.97) | 5 | 1.41 (1.03–1.93) | NA | 1.08 (0.75–1.56) | NA | NA | 0.470 |
CI = confidence interval; NA = data is missing or unavailable; OR = odds ratio.
Adjusted for age, ethnicity, level of education, family history of prostate cancer, smoking and alcohol consumption, sexually transmitted infections, cystitis and prostatitis.
Expressed as the upper cutoff point.
Reference category.
As regards the factors of aggressive tumours, none of the food groups were found to be associated with a higher Gleason score, a higher serum PSA level or a higher number of cancer cores in our cohort as shown in Table 3.
Table 3.
Multivariable odds ratios of food groups and the incidence of aggressive prostate cancer
| Quintiles*; odds ratio (95% confidence interval)
|
|||||||||
|---|---|---|---|---|---|---|---|---|---|
| Gleason score ≥ 7; n = 188, 48.70%
|
No. of cancer cores ≥ 3; n = 235, 60.88%
|
Serum PSA level ≥ 10 μg/L; n = 84, 21.76%
|
|||||||
| Food | Q1† | Q5 | p value | Q1† | Q5 | p value | Q1† | Q5 | p value |
| Dairy products | 1 | 1.16 (0.68–1.97) | 0.58 | 1 | 0.93 (0.54–1.60) | 0.81 | 1 | 0.76 (0.43–1.35) | 0.35 |
| Red meat, ham, sausages | 1 | 1.09 (0.65–2.45) | 0.47 | 1 | 0.77 (0.44–1.36) | 0.37 | 1 | 0.97 (0.54–1.76) | 0.94 |
| Chicken | 1 | 1.04 (0.59–1.84) | 0.59 | 1 | 0.96 (0.54–1.72) | 0.90 | 1 | 1.05 (0.54–2.04) | 0.87 |
| Fish | 1 | 1.44 (0.68–3.03) | 0.33 | 1 | 1.04 (0.65–1.33) | 0.12 | 1 | 0.91 (0.39–2.15) | 0.84 |
| Bread, pasta, rice, maize | 1 | 0.92 (0.55–1.53) | 0.76 | 1 | 1.06 (0.63–1.78) | 0.80 | 1 | 0.80 (0.46–1.39) | 0.43 |
| Potatoes | 1 | 1.17 (0.63–2.15) | 0.61 | 1 | 1.11 (0.60–2.05) | 0.60 | 1 | 1.49 (0.75–2.94) | 0.24 |
| Any green vegetables | 1 | 0.89 (0.65–2.14) | 0.27 | 1 | 1.09 (0.69–2.25) | 0.46 | 1 | 1.20 (0.62–2.32) | 0.58 |
| Tomatoes (e.g., fresh, sauce) | 1 | 1.27 (0.72–2.23) | 0.39 | 1 | 1.16 (0.66–2.02) | 0.60 | 1 | 1.25 (0.66–2.39) | 0.48 |
| Pulses (e.g., beans, peas) | 1 | 1.86 (0.86–3.99) | 0.11 | 1 | 2.89 (0.98–8.16) | 0.06 | 1 | 1.42 (0.61–3.41) | 0.39 |
| Fruits (e.g., fresh, juices) | 1 | 1.55 (0.91–2.62) | 0.10 | 1 | 1.28 (0.76–2.17) | 0.34 | 1 | 1.09 (0.93–3.78) | 0.09 |
| Soy products | 1 | 1.03 (0.88–1.21) | 0.66 | 1 | 0.90 (0.65–1.98) | 0.23 | 1 | 1.13 (0.76–1.89) | 0.11 |
| Cakes, desserts, ice cream | 1 | 0.95 (0.55–1.62) | 0.86 | 1 | 0.70 (0.41–1.22) | 0.22 | 1 | 1.78 (0.93–3.39) | 0.08 |
PSA = prostate-specific antigen.
Expressed as the lowest and highest quintiles.
Reference category.
Discussion
In our study we examined the possible association of different dietary habits with prostate cancer detection. A statistically significant associated risk of prostate cancer was found with a meat-containing diet (OR 2.91) and a fish diet was found to be associated with less prostate cancer (OR 0.54). In our cohort there were 35% and 18% of cases present in the fourth and fifth quintiles of meat diet, respectively, and 25% and 14% of controls were present in the fourth and fifth quintiles, respectively. On the other hand, 11% and 8% of cases were present in the fourth and fifth quintiles of fish diet, respectively, and 14% and 10.5% of controls were present in the fourth and fifth quintiles, respectively. The fourth and fifth quintiles of meat diet presented 5 and greater than 5 servings per week, respectively, and the fourth and fifth quintiles of fish diet presented 4 and greater than 4 servings per week, respectively. The relationship between prostate cancer development and either total meat and fish or the specific fatty acids as marine omega-3 polyunsaturated fatty acids and omega-6 polyunsaturated fatty acids present in animal sources, was studied in several epidemiological and experimental studies.19,20 In vivo and in vitro studies suggest that omega-3 polyunsaturated fatty acids may protect against prostate cancer, whereas omega-6 polyunsaturated fatty acids stimulate malignant cell growth.21–23 Those studies may explain our observations of the protective effect of a high intake of fish and the positive association between a high intake of different types of meat and prostate cancer detection.
Although there exists some evidence of the protective effect of tomatoes and green vegetables against prostate cancer,13 we did not find any statistically significant association between those types of food with prostate cancer. Stram and coauthors14 performed a large multiethnic study on the effect of fruits, vegetables and micronutrients and the incidence of prostate cancer and, surprisingly, they also did not find any significant association. In our cohort we observed an association between potatoes and low rates of cancer, and bread, pasta, rice and maize trended toward an association with a higher rate of cancer. Although these associations approached statistical significance (p = 0.08 and p = 0.09, respectively) it is possible that significance could be achieved with a greater number of patients. In addition, since the main component of both food groups is starch, it is possible that the apparently protective effect of potatoes is due to other associated dietary habits.
With respect to aggressive prostate cancer we examined the association of the same dietary habits with a high Gleason score (≥ 7), higher number of cancer cores (≥ 3) or higher serum PSA level (≥ 10). These criteria were chosen as they are commonly used in the clinic to differentiate low-risk from higher risk cancers. Although there was a statistically insignificant association between a high intake of pulses and a high Gleason score and a higher number of cancer cores (p = 0.11 and p = 0.06, respectively), we did not find any statistically significant association between dietary habits and aggressive prostate cancer. It is possible that bias due to biopsy sampling limitations in the identification of aggressive cancer is another reason we did not find significant relationships.
Case–control studies are considered susceptible to information bias more than cohort studies are24 since cases may be more careful than controls in searching their memory and may infer from known or suspected causes of their disease. To avoid this bias we asked all patients to answer the questionnaire before they received the results of their biopsies. In addition, there are several methods to assess dietary habits. Since developed in the 1950s, self-administered food frequency questionnaires have been considered appropriate methods for dietary assessment in nutritional epidemiology studies since they measure average long-term habitual dietary intake.25 Self-administered food frequency questionnaires have been validated in many studies compared with 24-hour recalls, food diaries or records of varying length, as well as biomarkers or calculating total energy intake and comparing it with energy expenditure.26–30 We assessed our questionnaire for variability by asking some patients to answer it again after a variable period of time and we observed no significant difference.
The weaknesses of our study include the short and incomplete nature of the food frequency questionnaire without home confirmation, as well as the fact that the study population comprised men at high risk of prostate cancer who were referred to the centre for biopsy. As such, this study may not apply to the general population.
Conclusion
Fish-containing diets appear to be protective against prostate cancer, and meat diets are associated with a 3-fold increased risk of prostate cancer detection. None of the dietary groups were found to be associated with aggressive prostate cancer on biopsy. These observations add to the growing body of evidence suggesting a relationship between diet and prostate cancer risk.
Appendix 1. The self-administered food frequency questionnaire
Date of birth
Marital status
Ethnic background
Occupation
Level of education
- Family history of prostate cancer
- History of prostate cancer in biological father
- History of prostate cancer in biological brothers (how many?)
- History of prostate cancer in biological sons (how many?)
- Smoking
- Age of starting
- Age of stopping
- Number of cigarettes per day
- Dietary intake (how many servings per week?)
- Dairy products
- Red meat, ham, sausages
- Chicken
- Fish
- Bread, pasta, rice, maize
- Potatoes
- Any green vegetables
- Tomatoes (fresh, sauce)
- Pulses (beans, peas, etc.)
- Fruits (fresh, juices, etc.)
- Soy products
- Cakes, dessert, ice cream
- Alcohol
- The history of the following diseases and its duration
- Infection of the bladder
- Infection of the prostate
- Cancer of the prostate
- Diabetes
- Liver disease
- Arthritis
- Heart disease
- Hypertension
- Vasectomy
- Previous prostate biopsy
- Sexually transmitted diseases (genital warts, gonorrhea, syphilis)
Footnotes
Competing interests: None declared.
See related article on page 516
This article has been peer reviewed.
References
- 1.Parkin D, Muir C, Whelan W, et al. Cancer incidence in five countries. Lyon (FR): International Agency for Research on Cancer; 1992. [Google Scholar]
- 2.American Cancer Society. Cancer facts and figures, 2003. Atlanta: The Society; 2003. [Google Scholar]
- 3.Parkin DM, Whelan SL, Ferlay J, et al. Cancer incidence in five countries. VII. Lyon (FR): IARC Scientific Publications; 1997. [Google Scholar]
- 4.Hsieh LJ, Carter H, Landis P, et al. Association of energy intake with prostate cancer in a long-term aging study: Baltimore Longitudinal Study of Aging (United States) Urology. 2003;61:297–301. doi: 10.1016/s0090-4295(02)02120-9. [DOI] [PubMed] [Google Scholar]
- 5.Whittemore AS, Kolonel LN, Wu AH, et al. Prostate cancer in relation to diet, physical activity, and body size in blacks, whites, and Asians in the United States and Canada. J Natl Cancer Inst. 1995;87:652–61. doi: 10.1093/jnci/87.9.652. [DOI] [PubMed] [Google Scholar]
- 6.Giovannucci E, Rimm EB, Colditz G, et al. A prospective study of dietary fat and risk of prostate cancer. J Natl Cancer Inst. 1993;85:1571–9. doi: 10.1093/jnci/85.19.1571. [DOI] [PubMed] [Google Scholar]
- 7.Le Marchand L, Kolonel LN, Wilkens LR, et al. Animal fat consumption and prostate cancer: a prospective study in Hawaii. Epidemiology. 1994;5:276–82. doi: 10.1097/00001648-199405000-00004. [DOI] [PubMed] [Google Scholar]
- 8.Chan JM, Giovannucci EL. Dairy products, calcium, and vitamin D and risk of prostate cancer. Epidemiol Rev. 2001;23:87–92. doi: 10.1093/oxfordjournals.epirev.a000800. [DOI] [PubMed] [Google Scholar]
- 9.Chan JM, Stampfer MJ, Ma J, et al. Dairy products, calcium, and prostate cancer risk in the Physicians’ Health Study. Am J Clin Nutr. 2001;74:549–54. doi: 10.1093/ajcn/74.4.549. [DOI] [PubMed] [Google Scholar]
- 10.Kolonel LN. Fat, meat, and prostate cancer. Epidemiol Rev. 2001;23:72–81. doi: 10.1093/oxfordjournals.epirev.a000798. [DOI] [PubMed] [Google Scholar]
- 11.Giovannucci E, Rimm EB, Liu Y, et al. A prospective study of tomato products, lycopene, and prostate cancer risk. J Natl Cancer Inst. 2002;94:391–8. doi: 10.1093/jnci/94.5.391. [DOI] [PubMed] [Google Scholar]
- 12.Kolonel LN, Hankin J, Whittemore A, et al. Vegetables, fruits, legumes and prostate cancer: a multiethnic case–control study. Cancer Epidemiol Biomarkers Prev. 2000;9:795–804. [PubMed] [Google Scholar]
- 13.Schuurman AG, Goldbohm R, Dorant E, et al. Vegetable and fruit consumption and prostate cancer risk: a cohort study in The Netherlands. Cancer Epidemiol Biomarkers Prev. 1998;7:673–80. [PubMed] [Google Scholar]
- 14.Stram DO, Hankin JH, Wilkens LR, 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. doi: 10.1007/s10552-006-0064-0. [DOI] [PubMed] [Google Scholar]
- 15.Augustsson K, Michaud D, Rimm E, et al. A prospective study of intake of fish and marine fatty acids and prostate cancer. Cancer Epidemiol Biomarkers Prev. 2003;12:64–7. [PubMed] [Google Scholar]
- 16.Terry PD, Rohan T, Wolk A. Intakes of fish and marine fatty acids and the risks of cancers of the breast and prostate and of other hormone-related cancers: a review of the epidemiologic evidence. Am J Clin Nutr. 2003;77:532–43. doi: 10.1093/ajcn/77.3.532. [DOI] [PubMed] [Google Scholar]
- 17.Schuurman AG, van den Brandt P, Dorant E, et al. Animal products, calcium and protein and prostate cancer risk in The Netherlands Cohort Study. Br J Cancer. 1999;80:1107–13. doi: 10.1038/sj.bjc.6690472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Tanguay S, Begin LR, Elhilali MM, et al. Comparative evaluation of total PSA, free/total PSA, and complexed PSA in prostate cancer detection. Urology. 2002;59:261–5. doi: 10.1016/s0090-4295(01)01497-2. [DOI] [PubMed] [Google Scholar]
- 19.MacLean CH, Newberry SJ, Mojica WA, et al. Effects of omega-3 fatty acids on cancer risk: a systematic review. JAMA. 2006;295:403–15. doi: 10.1001/jama.295.4.403. [DOI] [PubMed] [Google Scholar]
- 20.Ritch CR, Wan RL, Stephens LB, et al. Dietary fatty acids correlate with prostate cancer biopsy grade and volume in Jamaican men. J Urol. 2007;177:97–101. doi: 10.1016/j.juro.2006.08.105. [DOI] [PubMed] [Google Scholar]
- 21.Aronson WJ, Glaspy JA, Reddy ST, et al. Modulation of omega-3/omega-6 polyunsaturated ratios with dietary fish oils in men with prostate cancer. Urology. 2001;58:283–8. doi: 10.1016/s0090-4295(01)01116-5. [DOI] [PubMed] [Google Scholar]
- 22.Pandian SS, Eremin OE, McClinton S, et al. Fatty acids and prostate cancer: current status and future challenges. J R Coll Surg Edinb. 1999;44:352–61. [PubMed] [Google Scholar]
- 23.Ghosh J, Myers CE. Arachidonic acid stimulates prostate cancer cell growth: critical role of 5-lipoxygenase. Biochem Biophys Res Commun. 1997;235:418–23. doi: 10.1006/bbrc.1997.6799. [DOI] [PubMed] [Google Scholar]
- 24.Breslow NE, Day NE. The analysis of case control-studies. I. Lyon (FR): IARC Scientific Publications; 1980. Statistical methods in cancer research. [PubMed] [Google Scholar]
- 25.Willett W. Nutritional epidemiology. New York: Oxford University Press; 1998. [Google Scholar]
- 26.Jian L, Binns CC, Lee AH. Validity of a food-frequency questionnaire for elderly men in southeast China. Public Health Nutr. 2006;9:928–33. doi: 10.1017/phn2005919. [DOI] [PubMed] [Google Scholar]
- 27.Potischman N, Freudenheim JL. Biomarkers of nutritional exposure and nutritional status: an overview. J Nutr. 2003;133(Suppl 3):873S–4S. doi: 10.1093/jn/133.3.873S. [DOI] [PubMed] [Google Scholar]
- 28.Navarro A, Osella AR, Guerra V, et al. Reproducibility and validity of a food-frequency questionnaire in assessing dietary intakes and food habits in epidemiological cancer studies in Argentina. J Exp Clin Cancer Res. 2001;20:365–70. [PubMed] [Google Scholar]
- 29.Neuhouser ML, Kristal AR, McLerran D, et al. Validity of short food frequency questionnaires used in cancer chemoprevention trials: results from the prostate cancer prevention trial. Cancer Epidemiol Biomarkers Prev. 1999;8:721–5. [PubMed] [Google Scholar]
- 30.Boeing H, Bohlscheid-Thomas S, Voss S, et al. The relative validity of vitamin intakes derived from a food frequency questionnaire compared to 24-hour recalls and biological measurements: results from the EPIC pilot study in Germany. European Prospective Investigation into Cancer and Nutrition. Int J Epidemiol. 1997;26(Suppl 1):S82–90. doi: 10.1093/ije/26.suppl_1.s82. [DOI] [PubMed] [Google Scholar]