Selenium (Se) is an essential trace element that occurs in both organic and inorganic forms. The organic form is found predominantly in grains, fish, meat, poultry, eggs, and dairy products and enters the food chain via plant consumption. There is marked geographic variability of Se in food related to local soil content. Se is also widely available in over-the-counter supplements and multivitamins. Se is widely distributed in body tissues and is an important constituent of many antioxidant enzymes.

Many epidemiologic observations suggest that Se acts to protect against cancer. Both case-control and randomized placebo-controlled trials in humans also suggest that Se can decrease the risk of getting prostate cancer (19). As a typical example, in a case-control study of 445 men, serum Se levels were inversely proportional to the risk of prostate cancer, with an odds ratio of 0.71 (95% CI 0.39?1.26, comparing highest to lowest quartiles) (20).

The strongest evidence for a protective effect of Se comes from the Nutritional Prevention of Cancer Trial, a randomized placebo-controlled study of oral selenized yeast in patients with non-melanoma skin cancer (21). In that trial, 1312 participants took the equivalent of 200 μg yeast per day. With a mean follow-up of 4.5 years, the incidence of prostate cancer was reduced in the Se arm by two thirds compared to placebo. In an important update to this trial with an additional 25 months of follow-up (mean 7.45 years), Se supplementation continues to show a marked reduction of the incidence of prostate cancer, with a hazard ratio of 0.48 (95% CI 0.29?0.87) (22). As in the initial analysis, the effect was strongest for those with a PSA <4 ng/ml and those with the lowest serum Se levels at study entry. The findings of this trial, in which the incidence of prostate cancer was a secondary endpoint, prompted the use of Se in the large-scale Selenium and Vitamin E Cancer Prevention Trial (SELECT) (see below).

Se inhibits tumorigenesis in a variety of experimental models, and a number of potential mechanisms have been proposed for its antitumorigenic effects. Accumulating evidence suggests that Se works by inhibiting important early steps in carcinogenesis (19).

Vitamin E is a family of naturally occurring, essential, fat-soluble vitamin compounds that functions as the major lipid-soluble antioxidant in cell membranes. The most active form of vitamin E is α-tocopherol, which is also among the most abundant. It is widely distributed in nature and is the predominant form in human tissues. α-tocopherol may influence the development of cancer through several mechanisms. α-tocopheryl succinate (VES), a derivative of vitamin E, is known to modulate prostate cancer cell growth. Recent work suggests that VES causes G1 cell cycle arrest by decreasing expression of the cell cycle regulatory proteins cyclin D1, D3, and E, and cdk2 and 4 (23). Thompson & Wilding (24) have demonstrated that the chromanol moiety (PMCol) of vitamin E has antiandrogen activity.

One large-scale randomized placebo-controlled trial, the Alpha-Tocopherol Beta-Carotene Cancer Prevention Trial (ATBC), supports the role of vitamin E in the prevention of prostate cancer (25). ATBC was a double-blind trial of α-tocopherol (50 mg synthetic dl-α-tocopheryl acetate daily) and β-carotene (20 mg daily), alone or in combination, among 29,133 male smokers 50?69 years old at entry. The primary endpoints were lung cancer incidence and mortality. ATBC found a statistically significant 32% reduction in prostate cancer incidence and a 41% lower mortality in those receiving α-tocopherol. Another randomized double-blind placebo-controlled lung cancer prevention trial, the Beta-Carotene and Retinol Efficacy Trial (CARET), lends support to the epidemiologic evidence that α-tocopherol may prevent prostate cancer. Analysis of serum micronutrients in CARET participants has demonstrated that low serum levels of α-tocopherol were associated with a higher risk of prostate cancer (26). Unlike some other studies, CARET found no association between prostate cancer risk and serum γ-tocopherol levels.

The accumulated epidemiologic and biological evidence that Se and vitamin E may prevent prostate cancer led to the design and launch of the Selenium and Vitamin E Cancer Prevention Trial (SELECT) (27). SELECT is a phase III, randomized, double-blind, placebo-controlled, population-based clinical trial sponsored by the National Cancer Institute. It was designed to test the efficacy of Se and vitamin E, alone and in combination, in the prevention of prostate cancer (Figure 2). The study has a 2×2 factorial design with a target accrual of 32,400. Eligibility criteria include age ≥50 years for African-Americans or ≥55 years for Caucasians, a DRE not suspicious for cancer, serum PSA ≤4 ng/ml, and normal blood pressure. Randomization will be equally distributed among four study arms (Se + placebo, vitamin E + placebo, Se + vitamin E, and placebo + placebo). Study duration is planned for 12 years, with a minimum of 7 and maximum of 12 years of intervention depending on the time of randomization. The study supplements consist of 200 μg l-selenomethionine, 400 mg racemic α-tocopheryl, and an optional multivitamin containing no Se or vitamin E.

The primary endpoint for SELECT is the clinical incidence of prostate cancer. Prostate biopsy will be performed at the discretion of study physicians according to local community standards based on abnormalities in DRE or elevations in serum PSA. Secondary endpoints will include prostate cancer?free survival, all-cause mortality, and the incidence and mortality of other cancers and diseases potentially affected by the chronic use of Se and vitamin E. Other trial objectives include periodic assessments of quality of life, serum micronutrient levels, and prostate cancer risk, as well as studies of the impact of various biological and genetic markers on the risk for prostate cancer. The study design will permit detection of a 25% reduction in the incidence of prostate cancer associated with Se or vitamin E alone, with an additional 25% reduction for the combination of both agents. Since neither oral Se nor vitamin E are known to affect serum PSA, no PSA adjustments are planned. SELECT reached full accrual of 32,400 men in April 2004. Initial data analysis is anticipated in 2006 and complete results in 2013.

Interest in vitamin D as a preventive agent for prostate cancer comes from several epidemiologic observations: 1.  Men who live in northern latitudes with less exposure to sunlight-derived UV (which activates vitamin D in the skin) have a higher mortality rate from prostate cancer.

2.  Prostate cancer occurs more frequently in older men. Vitamin D deficiency is more common in older people because of less UV exposure and age-related declines in the hydroxylases responsible for synthesis of active vitamin D.

3.  African-Americans, whose skin melanin blocks UV radiation and inhibits activation of vitamin D, have the highest worldwide prostate cancer incidence and mortality rates.

4.  Dietary intake of dairy products rich in calcium, which depresses serum levels of vitamin D, is associated with a higher risk of prostate cancer.

5.  Native Japanese, whose diet is rich in vitamin D derived from fish, have a low incidence of prostate cancer.

In addition, prostate cancer cells express vitamin D receptor, and several studies have demonstrated an antiproliferative effect of vitamin D on prostate cancer cell lines (see Reference 28 for a summary of these data). In support of the UV hypothesis, one case-control study has suggested a protective effect for individuals with more highly pigmented skin (29). Additional data on the potential role of calcium and dairy intake are provided by analysis of the Cancer Prevention Study II Nutrition Cohort, a prospective cohort of 65,321 elderly U.S. men (30). Participants completed a detailed questionnaire on diet, medical history, and lifestyle at enrollment in 1992?1993. Overall dairy intake was not associated with a higher risk of prostate cancer, but a modest increase in risk of cancer was observed for high levels of calcium intake (RR = 1.2, 95% CI 1.0?1.6) and for high levels of calcium intake by diet + supplements (≥2000 versus <700 mg/day; RR = 1.6, 95% CI 1.1?2.3). The results support the hypothesis that calcium intake very high above the daily recommendation may modestly increase risk. Microarray gene expression studies demonstrate that active vitamin D exerts its antiproliferative effect predominantly by inducing cell cycle arrest (31). Use of vitamin D analogs in humans has been limited by their hypercalcemic effects, but newer analogs with more tolerable toxicity are being tested in phase I and II trials.

Nonsteroidal anti-inflammatory drugs (NSAIDs) function by nonselective inhibition of both COX-1 and COX-2, isoforms of cyclooxegenase that convert arachidonic acid to prostaglandins. COX-1 is constitutively expressed and mediates preservation of renal blood flow and function, platelet aggregation and hemostasis, and cytoprotection of the gastrointestinal mucosa. COX-2 is an inducible enzyme that mediates acute and chronic inflammation, pain, and cellular repair mechanisms. Inhibition of COX-2 expression by NSAIDs and more selective drugs blocks its proinflammatory effects and may underlie an important anticancer mechanism. Increased expression of COX-2 is known to correlate with increased angiogenesis, decreased apoptosis, increased tumor invasiveness, and immunosuppression in various tumors (32). Prostate cancers express more COX-2 than does benign prostatic epithelium, and several epidemiologic studies have noted an inverse association of prostate cancer and NSAID use (33). COX-2 inhibtion induces apoptosis in several prostate cancer cell lines (32). Recent work demonstrates that specific inhibition of COX-2 by celecoxib and nimesulide reduces the expression of several androgen-inducible genes, repressed androgen receptor (AR)?mediated activation of PSA and hK2 promoter activity, and repressed AR protein expression (34). Another study has demonstrated selective expression of COX-2 in high-grade prostatic intraepithelial neoplasia (PIN) in LPB-Tag transgenic mice, suggesting its involvement early in carcinogenesis (35). These results support the hypothesis that inhibition of COX-2 may be an effective preventive strategy. ViP, an industry-sponsored large-scale trial of rofecoxib in men with elevated PSA, closed without reaching accrual when this drug was withdrawn from the market because of cardiovascular toxicity concerns.

Interest in SERMs as preventive agents is stimulated by an apparent role of estrogens in the pathogenesis of prostate cancer. Both prostate stroma and epithelial cells express estrogen receptor, and estrogens promote prostatic growth (36). Epidemiologic evidence also supports involvement of estrogen in prostate cancer: Age-related prostatic disease parallels increases in serum estrogen levels, and the incidence of prostate cancer is low in cultures with diets rich in phytoestrogens (37). SERMs possess both agonistic and antagonistic estrogen-like activity and have been shown to repress prostate cancer growth in several transgenic mouse models. In the transgenic adenocarcinoma of the mouse prostate (TRAMP) model, toremifene reduces the incidence of high-grade PIN and cancer in an estrogen-dependent, androgen-independent manner (38). This agent is currently under study in clinical trials in men with high-grade PIN on biopsy.

Legumes play an important role in the traditional diets of Eastern countries, where prostate cancer incidence is low, but only a minor role in the West, where the incidence is highest worldwide. Soybeans are unique among the legumes because they are a concentrated source of isoflavones, which have weak estrogenic activity. Several studies have demonstrated a consistent anticancer effect of soy-based diets compared with controls in a variety of prostate cancer animal models (39). The major isoflavone components of soy inhibit benign and malignant prostatic epithelial cell growth, downregulate androgen-regulated genes, and reduce tumor growth in some animal models (40?42). Epidemiologic evidence also suggests that soy is an anticancer agent. Those who consume tofu five times per week have a lower risk of prostate cancer than those who consume it once per week (43). Japanese men excrete high levels of isoflavones in the urine, and urinary levels correlate with soybean-product intake (44). A larger study of 59 countries demonstrated that prostate cancer mortality is inversely associated with estimated consumption of cereals, nuts and oilseed, and fish, and that soy products are protective with an effect size per kilocalorie at least four times as large as that of any other dietary factor (45).

No large-scale clinical trials using soy or soy-based products as preventive or therapeutic agents in prostate cancer have been reported.

Lycopene is a red-orange carotenoid found primarily in tomatoes and tomato-derived products including tomato sauce, tomato paste, and ketchup, and in other red fruits and vegetables. Lycopene is a highly unsaturated acyclic isomer of β-carotene, is the predominant carotenoid in human plasma, and possesses potent antioxidant activity. There is mixed epidemiologic evidence that lycopene consumption is associated with a lower risk of prostate cancer (46).

Lycopene inhibits the growth of benign and malignant prostatic epithelial cells in vitro (47). However, when male rats were treated with N-methyl-N-nitrosourea and testosterone to induce prostate cancer, calorie restriction and tomato powder both exerted a protective effect, but pure lycopene did not (48). This observation suggests that tomato products contain compounds besides lycopene that modify prostate carcinogenesis, and that tomato phytochemicals and diet restriction may act by independent mechanisms. Reduced caloric consumption and a diet rich in tomato-based foods may be more beneficial than oral lycopene supplements in reducing the risk of prostate cancer in humans.

Two non-placebo-controlled prospective clinical trials have examined the effect of lycopene on known prostate cancer (49, 50). In the first trial, 26 men with clinically localized prostate cancer scheduled for radical prostatectomy were randomized to 15 mg lycopene PO bid for three weeks versus no lycopene preoperatively. Statistically significant reductions in serum PSA (18% drop versus 14% increase) and in the rate of positive margins (from 72% to 17%) were observed in the lycopene group. No differences were seen in various biological endpoints, including serum IGF-1, pr　eval　ence of high-grade PIN, and tumor expression of bcl-2, bax, or connexin 43. However, the sample size was small, and there were significant differences in tumor burden between the intervention and control groups as assessed by pretreatment stage and tumor grade. These limitations could account for the differences in pathological findings. In the second trial, 32 patients with localized disease scheduled for radical prostatectomy ate tomato sauce?based pasta dishes (equivalent to 30 mg of lycopene per day) for the three weeks before surgery. Serum and prostate lycopene concentrations were statistically significantly increased in the intervention group compared to subjects not taking lycopene.

Green tea has been suggested as a prostate cancer preventive based on epidemiologic observations of a low incidence of prostate cancer among native Asians with a high dietary intake. Previous work has focused on the effects of polyphenols contained in green tea, but the molecular mechanism of their action has not been elucidated. Prostate cancer cell culture experiments have demonstrated that the major polyphenolic constituent of green tea, (-)-epigallocatechin-3-gallate (EGCG), induces apoptosis, cell growth inhibition, and cyclin kinase inhibitor WAF-1/p21-mediated cell cycle dysregulation (51). Gene expression analysis found that EGCG treatment of LNCaP cells results in induction of growth-inhibitory genes that belong to the G-protein signaling network. In the TRAMP model, oral infusion of a polyphenolic fraction isolated from green tea at a dose achievable in humans (equivalent to six cups of green tea per day) significantly inhibits tumor development and metastasis (52). Additional work has shown that EGCG induces apoptosis by inhibiting fatty acid synthase (53). Taken together, the data indicate that EGCG induces apoptosis in human prostate carcinoma cells by shifting the balance between pro- and antiapoptotic proteins to favor apoptosis.

Because of its ubiquity and disease burden, prostate cancer is an attractive target for chemoprevention. Several large-scale, randomized, population-based studies are under way or completed, and evidence from the PCPT suggests that intervention with an effective agent can indeed lower the risk of developing this disease. It is likely that additional ongoing molecular studies will identify new molecular targets and agents that may also be effective.