SEARCH

SEARCH BY CITATION

Keywords:

  • prostate cancer;
  • prevention;
  • antioxidants;
  • selenium;
  • lycopene;
  • soy;
  • vitamins

Abstract

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES

Prostate cancer is the most common male malignancy and the second or third leading cause of cancer death among men in the West. The descriptive epidemiology of prostate cancer suggests that it is a preventable disease. Prevention has the theoretical advantage of not only saving lives, but also reduce the morbidity of radical prostate cancer therapy. This article reviews the past, present, and future of prostate cancer prevention. In particular, the evidence and scientific data of a variety of prevention strategies are reviewed. Strategies reviewed include dietary fat reduction and supplementation with vitamins D and E, and selenium. Dietary intake of soy, green tea, and tomato-rich products (lycopene) are also reviewed. Data regarding pharmacological intervention with cyclo-oxygenease inhibitors, antiestrogens, and in particular 5-alpha reductase inhibitors are reviewed. The results of the Prostate Cancer Prevention Trial including the controversy surrounding higher-grade cancers among men randomized to finasteride are also summarized. Finally, a variety of trial designs as well as a roster of current phase 2 trials are presented. Probably no cancer is being investigated more thoroughly in the context of prevention as prostate cancer in 2007. Definitive answers to pivotal phase 3 trials will be available in the coming 2 to 7 years. Cancer 2007. © 2007 American Cancer Society.


Prostate Cancer Demographics

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES

Prostate cancer is the most common human visceral malignancy and the third most common cause of cancer-related deaths among men in the Westernized world.1 In many ways, prostate cancer is a natural part of the male aging process in that development of microscopic amounts of cancer is ubiquitous. Autopsy studies show that men in the fourth decade of life have a one-third risk of harboring small carcinomas.2 By age 60, this number reaches approximately 60%. Studies suggest that the chance of having occult cancers at autopsy is similar across the globe, despite wide variations in death rates.3 The lifetime risk of being diagnosed with prostate cancer now approaches 1 in 7 and may hit 25% over the next decade in North America.4 Death rates from prostate cancer vary across the globe, with Westernized nations having the highest risk of incidence and death and Asian nations having the lowest.5 The overall chance of death from prostate cancer, even among Westernized nations that have not historically treated the disease for cure, is 3.5% to 4%.6 Given the large discordance between histologic incidence and death, there is great potential for overdetection and overtreatment. This is even more relevant as treatment-related morbidities associated with prostate cancer treatment can impact urinary function, sexual function, and quality of life.7 Disease prevention thus offers an attractive paradigm for addressing this important public health problem.

Descriptive Epidemiology

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES

The descriptive epidemiology of prostate cancer provides insights into the role of the environment in disease ontogeny. Prostate cancer has a wide global variation, with high-risk nations such as the US displaying a 10-fold higher disease-specific mortality than low-risk nations such as China and Japan.5 Despite these wide variations in disease incidence and mortality, autopsy studies of aging men suggest an equal rate of histologic prostate cancer, regardless of country of residence. However, volume, grade, and number of malignant foci tend to be less within prostates of men from the Pacific Rim.8 The relative contribution of genetics versus environment is raised by the global variation data. Migrating populations from areas of low risk to high risk help sort out this question. Data among both Japanese and Chinese Americans suggest that immigrants gain an increased risk within 11 years of migrating to America and that their offspring have virtually similar rates of prostate cancer as Caucasian Americans.9 This suggests that it is primarily the environment, and not some innate genetic resistance among Asians, that is responsible for prostate carcinogenesis. These data are confirmed by twin studies assessing the impact of heredity on prostate cancers among identical twins in Scandinavia.10 Given that prostate cancers start histologically among men in their 30s, these data also suggest that early prostate cancer events such as the development of high-grade prostatic intraepithelial neoplasia (HGPIN) are likely similar, regardless of environmental factors, and that intraprostatic progression to clinically detectable disease and metastases are likely driven by unfavorable environmental conditions such as diet or androgenic effects.8, 9 It must also be recognized that if an effective agent exists that can prevent prostate cancer the same agent should be studied as a complementary therapeutic strategy among men with clinical prostate cancer. Unless we intervene with men in their early 20s, prevention in the context of prostate cancer refers to a slowing of the growth of existing prostate cancer cells so that they never harm the host. If an agent can slow the growth of existing cancer cells, it remains plausible that they may be effective as an adjunct to surgery, radiation, or chemotherapy.1 For the purpose of this review, we will be referring to primary prevention as the prevention of ‘clinically’ detectable prostate cancers and not the prevention of the very first prostate cancer cell. The advantage of focusing prevention efforts on slowing the growth of existing microscopic cells is 3-fold. First, by using this approach we can effectively intervene among men in their 40s and 50s, when they are likely to comply with a chemoprevention strategy. Second, the timelines along the ontogeny of these cancers is so long (decades) that an agent with the capacity to slow this process would be of tremendous benefit. For example, if it takes 20 years from first cell to clinically detectable cancer, then slowing this process by 50% would extend it to 30 years. Finally, basic science efforts at arresting prostate cancer cells, which are easier to culture and transplant, are valid as opposed to efforts assessing the transformation from benign to malignant cell.

Table 1. Roster of Major Prevention Trials by Agent and Design
Study typePopulationSponsorAgentSample size (approximate)Estimated completion
  1. SWOG indicates Southwest Oncology Group; NCIC, National Cancer Institute of Canada.

General riskHealthySWOGSelenium & Vitamin E (SELECT)32,4002012
  MerckRefocoxib 8000Cancelled
  SWOGFinasteride19,000Completed
Higher riskElevated PSA-negative biopsyGlaxo Smith KlineDutasteride 82002010
PreneoplasticHigh-grade prostatic intraepithelial neoplasiaGTXToremifene 12002010
  SWOGSelenium 7002011
  NCICSoy, Vitamin E, selenium 3252008

Prostate Cancer Prevention: Past

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES

Relatively little effort has been made historically at preventing prostate cancer. The concept is relatively new, with pivotal studies not even contemplated until the late 1980s. Although fated without the intent to prevent disease, it has been historically said that eunuchs do not develop prostate cancer.9 In 1960, 26 eunuchs (average age 72 years) from the Qing dynasty were still alive in Beijing. They were castrated between ages 10 and 26 years of age. Examination of their prostates revealed an impalpable prostate in 81% and a small fusiform urethral swelling in the remainder.11 Biopsies were not performed, however. Although modern eunuch cultures still exist today, there has been no systematic evaluation of their prostate glands.12

Aside from the unique scenario among eunuchs, it must be recognized that several investigators have conducted observational epidemiologic studies regarding prostate cancer causation and a variety of exposures including: endocrine factors (androgen and others), sexually transmitted diseases, baldness, and occupational exposure. These studies were largely inconsistent and revealed marginal risk estimates.1, 5, 9, 12

Prostate Cancer Prevention: Present

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES

Background

Over the past 2 decades increased awareness of prostate cancer coupled with improved techniques for disease detection dramatically increased the number of men being treated for early-stage disease.13 Although the radical treatment of early-stage disease has improved overall survival, many men are still being treated unnecessarily, with concomitant adverse effects on sexual and urinary function.14 A fresh look at prostate cancer prevention was thus viewed as a way to deal with this problem on a population level. Many of the candidate agents or strategies to prevent prostate cancer stemmed from case-control and cohort epidemiological studies performed over the past 30 years. This ‘public-health’ approach has been successful in reducing the risk of cardiovascular morbidity and death. Other agents have stemmed from endocrinological principles or serendipitous findings derived from large-scale intervention trials.

Endocrine agents

Finasteride

Androgens play an important role in prostate cancer. Androgen suppression therapy can cause regression of established prostate cancer and eunuchs do not develop prostate cancer.11, 15 It has thus long been hypothesized that agents with a capacity to modulate androgen signaling could potentially prevent prostate cancer. Dihydrotestosterone, the 5-alpha reduced metabolite of testosterone, is the major effector hormone on prostatic epithelium. Finasteride, a compound that blocks the type II isoenzyme of 5-alpha reductase, induces prostatic volume reduction and has been used to treat symptomatic benign prostatic hyperplasia (BPH) for years.16 The known pharmacological effects of finasteride led the National Cancer Institute (NCI) to conceive, design, and launch the Prostate Cancer Prevention Trial (PCPT).

The PCPT was designed as a phase 3 placebo-controlled study with an a priori defined endpoint of preventing histologic prostate cancer. Eligible men for PCPT were older than 55 years of age with a prostate-specific antigen (PSA) level less than 3 ng/mL and a normal digital rectal examination (DRE). All men had an American Urological Association (AUA) symptom score less than 20 and no major comorbidities. Men were allocated to either placebo or 5 mg finasteride daily.17 The PCPT was designed as a 7-year intervention study among 18,882 American men. All men were screened with annual PSA and DRE and intended to have a prostate biopsy after the 7-year observation period.17 If during the screening period PSA significantly rose or DRE became abnormal a biopsy was done. These biopsies were referred to as ‘for-cause’ biopsies. Because finasteride reduces PSA levels, all PSA results were reported from a central site as an adjusted value among men randomized to finasteride. PSA adjustment was performed to minimize bias toward more ‘for-cause’ biopsies among men randomized to placebo, as PSA elevations usually trigger a biopsy. At the start of the trial a doubling of PSA was used based on data from patients with BPH.18 By the trials end a factor of 2.3 was used based on a need to balance ‘for cause’ biopsies in both cohorts.17

The PCPT was prematurely stopped, as the primary objective was achieved. Among men randomized to placebo, 24.4% developed cancer over the study period. This compared with 18.4% for men randomized to finasteride. The relative and absolute risk reductions were 24.8% and 6%, respectively. Both ‘for cause’ and end of study biopsies demonstrating cancer were reduced by finasteride, with a greater effect on end of study biopsies. Survival was the same in both arms.17

Side effects noted in this trial were marginal decreases (approximately 6%) in libido and erectile function among men assigned finasteride. A small proportion of men also developed breast-associated symptoms such as mastodynia or gynecomastia. No cases of male breast cancer were noted in the study. There was a benefit with respect to urinary complaints such as prostatitis development, urinary retention, and lower urinary tract symptoms (LUTS) among men who received finasteride.

Histologic grade has long been considered the most significant predictor of outcome among men with localized disease.19 The grading system put forward by Gleason has become the most commonly used scale.20 In recent years, scores less than 6 are considered low-grade cancers, 7 intermediate grade, and 8 to 10 high grade.21 In the PCPT trial, examination of the histologic grades of the cancers noted among men on placebo and finasteride yielded surprising results. The predominant prevented cancers were of low histologic grade (Gleason 6 or less). Tumors of high grade, however, were more common among men who had ‘for-cause’ biopsies and were allocated finasteride. This observation is particularly concerning as not only was the proportion of high-grade tumors higher (as one might expect if there were differential effects on the lower-grade tumors) but the absolute numbers of high-grade cancer were higher. This observation raises the possibility that finasteride could actually induce grade progression. In total, there were 43 more high-grade cancers found among men randomized to finasteride (relative risk [RR] 1.67, P < .05). The observation of more high-grade cancers among men randomized to finasteride has led to many investigative studies that have tried to explain these unexpected findings. It should be noted that unlike other models of drug-induced carcinogenesis (such as tamoxifen-induced endometrial cancers), the number of high-grade cancers did not increase over time (ie, no exposure-response effect). Four major hypotheses have been put forward to explain the grade differences seen with finasteride treatment.

The first theory is that the grade progression is a true phenomenon. It is plausible that alteration of the androgen milieu could induce genetic instability that would lead to a more aggressive phenotype. The 3 remaining hypotheses suggest that the grade differences are because of systemic biases inherent in study design. These include pathological interpretation artifact, volume-grade artifact, and an artifact secondary to PSA adjustment.

When the Gleason system was first described, the intent was for it to be used among men with normal androgen status. It has been long understood that grading cancers among men on luteinizing hormone release hormone (LHRH) agonists results in a bias toward higher grade assignment because of changes in pathological architecture.22 The histologic effects of finasteride on architecture were much less subtle and not well studied before PCPT. This was particularly relevant as the intervention in PCPT was for 7 years. Civantos et al.23 published a small experience before PCPT, detailing the changes recognized histologically among men on finasteride. In that study it was felt that the changes induced with finasteride were similar to those seen after LHRH agonist treatment.23 Other long-term studies contradicted these findings.24 A recent study was performed where pathologists blindly reviewed biopsies among men on LHRH agonists, finasteride (at least 6 months), and no therapy. In this study, no consistent features of finasteride therapy could be identified.25 It is thus felt that the likelihood of this bias as an explanation for the PCPT findings is small.

The volume-grade artifact was described in the study by Kulkarni et al.26 Their study examined the associations of biopsy grade, true histologic grade (determined at radical prostatectomy), and prostate size among men receiving sextant prostate biopsies (similar technique as used in the vast majority of PCPT participants). It was discovered that there was no association between prostate size (as determined by diagnostic transrectal ultrasound) and histologic grade at prostatectomy, but there was an association between biopsy grade and prostate size among the same patients. It thus appears that among men with smaller prostates, foci of preexisting high-grade acini are easier to find at biopsy. This observation is relevant as finasteride shrinks the prostate gland by approximately 30%.27 It thus appears that simply shrinking the prostate makes the determination of true histologic grade more likely at biopsy.

As mentioned above, when PSA values were interpreted among men on finasteride a multiplier was used to balance the number of for-cause biopsies. Post-hoc analyses of the receiver operator characteristic (ROC) curves of PSA for the prediction of high-grade prostate cancers at biopsy reveal that they are significantly shifted among men on finasteride.28 In other words, PSA reductions among men on finasteride with high-grade cancers are less than men with low-grade cancers. This bias leads to more for-cause biopsies showing high-grade disease among men on finasteride. This bias, although interesting, should not have a large effect on total study outcome, as these cancers should eventually be found by end-of-study biopsies. It would appear that biases not previously recognized offer the most plausible explanation for the discrepancy in grade between the 2 treatment arms in PCPT.

Unger et al.29 published an analysis of the PCPT results on a population-wide level. In their analysis, even with the assumption that the high-grade cancer findings were true, over 260,000 life-years would be saved per year in the US. An analysis by Lotan et al.30 showed similar results. There are few costing studies, however in the article by Zeliadt et al.31 the cost and benefits of finasteride were felt to only make economic sense if the price of finasteride was cut in half. Interestingly, finasteride will be available in the US as a generic product over the next year. At the time of publishing this article, the uptake of the use of finasteride among the general public for prostate cancer prevention has been modest because of concerns about the biological significance of the tumors being prevented and lingering concerns about the grade progression issue. Over the next few years we may witness the use of this agent among high-risk men, such as those with a strong family history of prostate cancer.

Dutasteride

Dutasteride is a newer 5-alpha reductase inhibitor (5ARI). Like finasteride, it inhibits the type II receptor, but unlike finasteride it also inhibits the type I receptor.32 The role of the type I receptor in prostate carcinogenesis continues to be refined. It has long been known that the type I receptor has a minimal role in the evolution of BPH. The evidence for dutasteride as a potentially better prevention agent stems from a noted 50% reduction in ‘for cause’ biopsy cancers in studies of men with symptomatic BPH.33 A similar reduction was never noted in prior trials of finasteride. Furthermore, a variety of basic science studies suggest that 5AR type I expression at the RNA and protein level seems to be increased in prostate cancers and that this expression increases as one migrates along the spectrum of more advanced disease.33, 34 In a preoperative study, administration of dutasteride for 4 months caused a 40% reduction in tumor volume compared with control.35

Dutasteride is currently being assessed in a large phase 3 trial called REDUCE (reduction by dutasteride of prostate cancer events).33 This trial will be assessing the role of dutasteride among 8250 patients between 50 and 75 years of age with an elevated PSA level and a negative biopsy. A repeat biopsy will be performed at 2 and 4 years to determine efficacy. This study will report within the next few years and should have significant impact, as it will assess the role of 5ARIs in a more clinically relevant setting for urologists. It will also allow for a second look at the grading issue seen in PCPT.

Toremifene

The third studied agent for prostate cancer prevention using manipulation of the hormonal axis is estrogen receptor modulation. It has long been known that estrogens are a vital component in the induction of prostate cancer among laboratory animals.36 Toremifene is a selective estrogen receptor modifier (SERM) approved for the treatment of breast cancer at a 60 mg/day dosage. Preclinical work using this compound in a transgenic mouse model demonstrated an ability to prevent prostate cancer and high-grade prostatic intraepithelial neoplasia (HGPIN).37 A phase 2b study was initiated among 514 men with HGPIN such that men were randomized to placebo, 20, 40, or 60 mg of toremifene. At study end the 40 and 60 mg dose did not prevent invasive carcinoma; however, the 20 mg dose had a beneficial effect compared with placebo (24.4% vs 31.2%, P < .05). The scientific explanation of the lower dose of this compound may relate to its selective inhibition of the alpha subtype or of the estrogen receptor at this dose.38 Overall adverse effects and drug-related side effects were similar to placebo.38 These promising results have led to a definitive phase 3 trial of the 20 mg dose of toremifene that has now completed accrual. The final results should be available within 24 months.

Dietary fat intake

There is strong epidemiologic evidence linking dietary fat and prostate cancer. Case-control, ecologic, and cohort studies have consistently demonstrated this association.39 There are 4 prevailing theories that link dietary fat consumption with prostate cancer.

Pesticide exposure

Since the late 1950s there has been a significant rise in the use of pesticides in the farming industry. Researchers have attempted to link the rise in many cancer rates to indirect exposure to pesticides through diet. Most pesticides are highly fat soluble, providing an excellent medium for storage and transfer within the food chain. In addition, the molecular structure of pesticides mimics that of steroidal hormones, which is relevant because the prostate is steroid-responsive. Schrader and Cooke40 have shown that the organochlorine pesticides can act as antagonists and agonists of androgens. Tessier and Matsumura41 showed increased proliferation as well as up-regulation of the erbB2 oncogene of LNCaP cells when exposed to a variety of pesticides. These toxic effects have been supported by other animal model studies.42

It is logical to assume that humans who work with pesticides would be at increased risk for prostate cancer if pesticides do represent significant carcinogens. The majority of case-control and cohort studies have shown a positive association between pesticide exposure and risk of prostate cancer (as well as other malignancies). Settimi et al.43, 44 noted a 2.5-fold increased risk of prostate cancer among farmers or men exposed to pesticides in an agricultural setting. Data from Flemming et al.45, 46 and Janssens et al.47 showed similar results.

Elevated androgens

Dietary fat has been postulated to increase androgen levels. Littman et al.48 and Hill et al.49 reported that urinary androgen levels decreased in African and Caucasian males who reduced their fat intake. The same authors also noted lower serum testosterone levels among vegetarians. In a controlled study, Howie and Schultz50 found that omnivorous Seventh-Day Adventists (SDA) had higher testosterone and estrogen levels than vegetarian SDAs. Bishop et al.51 analyzed androgen levels among twin pairs and found higher testosterone levels in the co-twin who consumed more fat. More recently, Dorgan et al.52 performed a randomized crossover study of high and low fat dietary interventions. In that study, total and free testosterone levels were increased when men consumed a higher fat diet. Postprandial androgen levels do change in response to diet.53 Habito and Ball53 noted that men had significant reductions in androgen levels when consuming lower-fat meals. Other investigators have challenged the fat-androgen theory, arguing that measurement of hormonal levels in a haphazard fashion is associated with considerable intraassay, interassay, and diurnal variation.54 Pusateri et al.55 noted no difference in serum androgen levels in young SDA men on different dietary regimes. Hsing et al.56 noted lower dihydrotestosterone levels in men who subsequently developed prostate cancer compared with controls.

Specific fatty acids (linolenic acid)

Prostate cancer cell proliferation may be influenced by the dietary intake of certain essential fatty acids. Linolenic acid is a fatty acid (FA) that has been extensively investigated. It is an omega-3 FA, which has demonstrated both inhibitory and promotional effects on prostate cancer.57, 58 Alpha-linolenic acid is primarily derived from plant sources such as flax, corn, and perilla oil. It is also derived from meat and dairy products. Case-control59 and stored sera cohort studies60 have demonstrated a positive association between alpha-linolenic acid and prostate cancer. They also demonstrated a negative association, with an increased ratio of linolenic to alpha-linolenic acid.49, 61 Other studies have confirmed these associations.50, 62, 63 Although the data are still not plentiful, most investigators would agree that specific fatty acids such as linolenic acid warrant further investigation with respect to their role in prostate cancer causation.

Oxidative stress

Oxidative stress refers to the genesis of reactive oxygen species (ROS), which interfere with a host of cellular mechanisms important in cell growth and regulation. ROS can cause mutations in DNA basepairs through intercalation. Wildtype p53 proteins placed in an oxidatively stressed environment can behave as if they emanated from a mutant phenotype.64 In addition, studies have shown that ROS can increase oncogene expression.65 Rao et al.66 demonstrated that prostatic tissue markers of oxidative stress are associated with prostate cancer.

Dietary fat has clearly been shown to increase known markers of oxidative stress and is a well-recognized pro-oxidant. Other factors that can increase oxidative stress in prostate tissue include smoking, sedentary lifestyle, aging, and androgens.8 In addition, in laboratory models of high-fat diet-induced prostate cancer, vitamin E and other antioxidants including lycopene and selenium can eliminate or blunt this effect.67 Given these strong observations, dietary fat may be acting in part as a pro-oxidant in the context or prostatic carcinogenesis. In 1 small intensive nutrition alteration study among men on watchful waiting, a small improvement in PSA was noted.68 Although the data remain intriguing, it is unlikely that a large-scale, well-powered trial will be performed.

Vitamin E (alpha tocopherol)

Vitamin E is a major antioxidant and has been touted to have many health benefits for a variety of chronic conditions. The association of vitamin E intake and prostate cancer arose from a non-a priori-driven result from a large cancer prevention trial. The alpha tocopherol beta carotene (ATBC) trial randomized 19,000 Finnish men in a 2-factorial fashion to either vitamin E, beta-carotene, both, or placebo. The primary endpoint of the study was lung cancer development. Quite surprisingly, at the end of 4 years there was a one-third reduced chance of prostate cancer among men randomly allocated vitamin E. Furthermore, a 41% reduction in prostate cancer deaths was noted at 6 years.69 Additional studies have suggested a benefit of vitamin E in reducing prostate cancer risk principally among men who smoke.70 This association seems quite consistent despite the finding that smoking per se is not associated with prostate cancer.67

Although no basic science base had existed before the ATBC report, a host of laboratory science has now examined the impact of vitamin E in the context of prostate cancer. These studies have demonstrated in vivo and in vitro effects of vitamin E on apoptosis, cell cycle arrest, and proliferation arrest in prostate cancer tumor model systems.67, 71 Recent data have questioned the safety of vitamin E at high doses greater than 400 IU per day and other antioxidants.72–74 Heart failure and all-cause mortality increases have been noted in recent publications. Although these trials were largely conducted among men with multiple comorbidities, some caution must be exercised in recommending high-dose vitamin E to men at risk for prostate cancer.

Selenium

Selenium is a trace micronutrient that is important in cellular host defenses to oxidative stress. Most of North America is a low-selenium area, with minimal levels noted in the soil where most local produce is grown. A large body of epidemiologic evidence suggests that humans who live in low-selenium areas have a higher risk of cancer.75 The association of selenium and prostate cancer risk was serendipitous in that it was derived from a non-a priori endpoint of a cancer prevention study. The selenium intervention trial by Clarke et al.75 looked at 1312 men with nonmelanoma skin cancer. This cohort was randomized to selenium (200 μg per day) or placebo. Surprisingly, after 10 years there was a 49% lower chance of developing clinically diagnosed prostate cancer among the men randomized to selenium.

Similar to the vitamin E history, the results of this phase 3 trial led to additional studies suggesting that selenium possessed significant anticancer properties in vivo and in vitro, including induction of apoptosis, DNA repair augmentation, and cell cycle arrest in human prostate cancer cell lines.76, 77 Different forms of selenium such as methylselenic acid and selenomethionine may have different properties as well.78

Selenium is well tolerated at doses of up to 200 μg. A toxic dose results in brittle nails, hair, and halitosis.77 Long-term follow up of the Clarke et al. trial revealed an increased incidence of secondary skin cancers among patients randomized to selenium,79 so this agent should be avoided in patients with a history of basal and squamous cell carcinomas of the skin.

Green tea

Green tea has long been of interest as a preventative agent because of its large consumption in Asia, an area where prostate cancer incidence and mortality is low. Green tea contains catechins, including epigallocatechin (EGCG), the best-studied constituent. In vitro studies suggest that this agent can induce cell cycle arrest, inhibit insulin-like growth factor-1 synthesis, and induce apoptosis in a variety of prostate cancer cell lines.80–82 In vivo studies have also been promising in both xenograft and transgenic models.82, 83 Most promising to date, however, are the recent data from Betuzzi et al.,84 who performed a small, randomized trial of green tea catechin tablets in 60 men with HGPIN. At the 1-year biopsy, 9 patients randomized to placebo versus 1 randomized to green tea tablets demonstrated invasive cancer. One small trial from Canada was negative among men with advanced disease.85 Clearly, confirmatory trials are warranted.

Soy

The low incidence of prostate cancer in Asia has led to an interest in the study of soy, which is highly consumed in those nations. Soy is rich in isoflavones, which induce cell cycle arrest and inhibit proliferation in a variety of prostate cancer tumor model systems.86, 87 Soy is also rich in vitamin E and has some selectivity for inhibiting the alpha subtype of the estrogen receptor, which has been linked to prostate carcinogenesis.87 A variety of pilot studies are under way, some reporting beneficial effects on endocrine markers including trials as adjuvants to surgery or watchful waiting.88 A phase 3 Canadian trial among men with HGPIN has completed accrual.

Lycopene

Lycopene is an antioxidant that is found primarily in tomato and tomato-based products in the North American diet. Processed forms of tomato products such as paste, soup, sauce, and juice are particularly rich in lycopene because of better bioavailability. Men who consume tomato-rich diets are at lower risk of developing cancer in general, including prostate cancer. Giovanucci89 demonstrated that men who consume more than 10 servings of tomatoes per week have a one-third lower risk of developing prostate cancer. There was also a dissociative benefit, favoring protection from advanced and metastatic disease. This suggests that lycopene may inhibit disease progression. Numerous in vivo and in vitro studies have demonstrated anticancer properties of lycopene, including significant impact on the transgenic mouse adenocarcinoma (TRAMP) model.90, 91 No trials are currently assessing this agent in a phase 3 setting with relevant clinical endpoints.

Nonsteroidal antiinflammatory drugs

Preclinical evidence suggests that the cyclo-oxygenase (COX) pathways may play an important role in prostate carcinogenesis. COX inhibitors have proapoptotic properties in prostate cancer cell lines and in vivo studies.92 Furthermore, non-a priori analyses from clinical trials, health record cohorts, and case-control studies support this concept.93, 94

There was much excitement about using these agents to prevent prostate cancer over the past decade. Refocoxib was under study in a large phase 3 setting for prostate cancer prevention; however, concerns about cardiovascular toxicity prematurely closed the study. It is thus unlikely that these agents will be tested again, given the side effect profile, unless the toxicity can be modified.

Vitamin D

Vitamin D is not technically a vitamin, but truly a steroid hormone. It is produced under the influence of the sun with further modification in the liver and kidney to active metabolic forms. It has been shown that particularly in the winter months, Canadians and Northern Americans seem to be quite deficient in vitamin D.95 Vitamin D has a variety of antiproliferative and proapoptotic effects in human prostate cancer cell lines as well as other cancer cell lines.96, 97 A variety of studies are under way to determine whether vitamin D can prevent prostate cancer.

Clinical trials

A host of study designs have been devised to test for efficacy of prevention agents/strategies in the context of prostate cancer (Table 1). Because of prostate cancers' long natural history, it is essentially impossible to design a trial with a survival endpoint, as it would take decades. It should be noted that trials of pharmacologic agents or tablets have a greater ease of compliance and thus completion compared with radical lifestyle change trials such as dietary fat reduction.68 Furthermore, issues of agent compliance, subject retention, minority participation, and long trial duration also represent unique challenges in the study of prostate cancer prevention.

It is plausible that an agent may prevent low-risk cancers that pose little or no threat to life, while leaving the significant cancers untouched. However, there would be benefits even from an intervention such as this because the diagnosis of prostate cancer has several adverse consequences, including psychological effects, financial costs, treatment-associated morbidity, and risk of recurrence. Thus, a treatment that could prevent these events, even if they do not reduce mortality rates, would have perceived benefit. Three types of trials have been conducted.

General risk trials

These take place among men who are at average risk for prostate cancer. The PCPT would fit this class. These studies typically take 10,000 to 30,000 subjects and take many years to complete. Pharmaceutical companies would have little interest in these studies, as intellectual property issues would preclude it. The Selenium Vitamin E Chemoprevention Trial (SELECT) is another contemporary study that fits this category.98 This 2-factorial designed study will assess vitamin E (400 IU/day) and selenium (200 μg/day) with the aim of preventing prostate cancer. Unlike PCPT, end-of-study biopsies are not needed and usual clinical practice will determine endpoints. This study is a ‘real-world’ study and will have tremendous impact if positive.

Higher-risk trials

These trials are also large (2000–10,000 subjects) and use men considered at high risk for prostate cancer. The REDUCE study assessing dutasteride would fit this category.33 This trial will assess 8200 men with elevated PSA and no cancer despite a biopsy. Patients will be rebiopsied at 2 and 4 years. This study will be seen as a ‘real-world’ study as it uses subjects in urologists' offices and mirrors clinical practice. Whether these men are truly at ‘high risk’ is debatable, given the previously unknown prevalence of cancers among men with low PSA (PCPT).

Preneoplastic trials

These studies use men with HGPIN and randomize an agent versus placebo. Men are then rebiopsied at some time in the future. These studies typically take 300-1000 patients and take 1–3 years to complete. Toremifene, selenium (SWOG), and a National Cancer Institute of Canada (NCIC) study of combination vitamin E, selenium, and soy versus placebo are actual examples of these studies.

Prostate Cancer Prevention: Future

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES

A variety of new and existing pathways of prostate carcinogenesis will lead toward the development of novel agents that may prevent prostate cancer. Insulin resistance, obesity, and the potential use of metformin or the glitazones is a promising field of scientific endeavor.99, 100 The use of statins has also been recently linked to lower prostate cancer rates, raising the possibility of using antilipid-type drugs in future trials.101 And finally, some tantalizing data on novel viruses raises the hope for vaccines as a means to deal with this burdensome problem.102

The future for prostate cancer prevention will focus on using high-throughput molecular biology to understand the basic genetics of prostate cancer. For example, the recent discovery of a fusion gene product from the ETS family of genes with the androgen regulated TMPSR22 gene in a majority of prostate cancers will allow for a leap forward in prognostics and theragnostics.103 Better understanding of risk from genetic and environmental points of view will be necessary.

Once we can reliably identify men at high risk for a diagnosis of prostate cancer, we can then use metabolomics and identify the appropriate agent for effective chemoprevention. Large cohorts of genetic, serum, and tissue banks from existing large-scale prevention trials such as SELECT, NCIC, PCPT, and REDUCE will form the backbone for these analyses.

Conclusions

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES

Prostate cancer prevention has grown from a virtually unknown field 15 years ago into a major area of scientific and clinical investigation. There are now a host of phase 3 studies that have been completed and analyzed (ie, PCPT) or completed accrual. Thus, the next 5 years hold great promise in identifying effective strategies. Wide-scale adoption of these strategies will need to be balanced by cost and treatment-related adverse effects. Most would agree that the ‘threshold’ for adoption of prevention among healthy men will need to be higher than treating men with actual disease. Long-term efficacy and toxicity studies must continue. Finally, the promise of personalized medicine will hopefully drive prevention in the coming decades.

REFERENCES

  1. Top of page
  2. Abstract
  3. Prostate Cancer Demographics
  4. Descriptive Epidemiology
  5. Prostate Cancer Prevention: Past
  6. Prostate Cancer Prevention: Present
  7. Prostate Cancer Prevention: Future
  8. Conclusions
  9. REFERENCES