Prevention is an important strategy for limiting prostate cancer morbidity and mortality. Two major types of prevention are primary (reduction of incident cases) and tertiary (inhibition of disease progression and recurrence). Pharmacological and dietary interventions have potential functions in both the primary and tertiary prevention of prostate cancer. Five-α reductase inhibitors (5-ARIs) reduce the incidence of prostate cancer in both general and higher-risk populations and are currently under study for tertiary prevention in active surveillance and biochemical recurrence patients. Selenium, vitamin E, and vitamin C do not prevent incident prostate cancer in the general population; however, other promising diet-based interventions are currently under study for tertiary prevention. We recommend consideration of 5-ARIs for prostate cancer prevention in (1) asymptomatic men with a PSA 3.0 ng ml–1 who are undergoing or anticipate undergoing PSA screening for early detection of prostate cancer and (2) asymptomatic men with PSA 2.5 and 10 ng ml–1 and an earlier prostate biopsy negative for cancer. Men should be informed of the potential risks of 5-ARI therapy. Currently, there is neither clinical evidence to support the use of 5-ARIs for tertiary prevention in active surveillance or biochemical recurrence populations, nor micronutrients for prostate cancer prevention of any type.
Prostate cancer chemoprevention
Prostate cancer, the most commonly diagnosed non-cutaneous cancer among US men, represents an important target for disease prevention for several reasons.1, 2 First, the majority of patients now present with localized, low-risk cancers that are potentially amenable to prevention. Second, despite technological and technical refinements, the most common therapies—surgery and radiation—are associated with substantial detriments to quality of life that persist for years after treatment.3, 4 Finally, recent data suggest that many low-risk prostate cancers are over-treated: the European randomized study of screening for prostate cancer study observed that prevention of one prostate cancer death necessitated screening of 1068 men with PSA and definitive treatment of an additional 48 cases of prostate cancer.5
Cancer chemoprevention is the use of natural, synthetic, or biologic substances to prevent or suppress the development of cancer. This review summarizes Level 1 evidence and ongoing trials of prostate cancer chemoprevention and suggests a set of practical guidelines for the clinician to follow.
General principles of disease prevention
There are three basic types of prevention: primary, secondary, and tertiary (Figure 1). Primary focuses on prevention of incident disease in at-risk populations, secondary on attenuating the severity of prevalent disease through early detection and intervention, and tertiary on halting disease progression and recurrence in patients. All three types of prevention have prominent functions in the management of prostate cancer. However, because secondary prevention involves population screening with PSA rather than chemoprevention, this review discusses only primary and tertiary prevention.
There are three broad categories of chemoprevention for prostate cancer: hormonal, dietary, and anti-inflammatory. Most studies have focused on hormonal interventions, which manipulate sex steroid hormone pathways, and dietary interventions, which alter the balance of nutritional intake.
Inflammation is believed to be an important component in the genesis of prostate cancer.6 Aspirin,7 non-steroidal anti-inflammatory medications8, and statins9 have been hypothesized to reduce the incidence of prostate cancer. However, although there are several preclinical and epidemiologic studies supporting a protective effective of anti-inflammatory therapies for prostate cancer, as yet there have been no clinical trials. Thus, anti-inflammatory medications will remain beyond the scope of this review.
Risk stratification and defining the target population
In discussing prostate cancer prevention, it is important to consider differential risks for incident, recurrent, and progressive prostate cancer and to stratify men appropriately. These differences help define target populations and are crucial to designing practical, realistic, and cost-effective prevention strategies. For example, lower-risk populations—in which the prevalence and incidence of prostate cancer are relatively low, and thus the numbers needed to treat to prevent a single case of prostate cancer are relatively high—present particular challenges that may not be present in higher-risk populations. On the other hand, higher-risk populations may harbor a higher prevalence of more aggressive phenotypes that are less susceptible to prevention strategies.
General population: the Prostate Cancer Prevention Trial, Selenium and Vitamin E Cancer Prevention Trial, and Physicians Health Study II studies
Several large randomized-clinical trials (RCTs) have focused on primary prevention of prostate cancer in the general population: men without prostate cancer who, based on study selection criteria, were at relatively low risk for prostate cancer diagnosis. The prostate cancer prevention trial (PCPT) tested the hypothesis that finasteride would prevent prostate cancer through hormonal manipulation. Finasteride is a selective type two five-α reductase inhibitor (5-ARI) that decreases levels of circulating and intraprostatic dihydrotestosterone (DHT) and is FDA approved for the treatment of symptomatic BPH. Compared with testosterone, DHT has a greater binding affinity for the androgen receptor and results in 2- to 10-fold greater translational activity.10 Finasteride-induced reductions in DHT concentrations result in decreased activation of the androgen receptor, decreased prostate volume, and increased prostatic cell apoptosis.11
In the PCPT, 18 882 men were randomized to receive either finasteride or placebo over a 9-year period. Eligibility criteria included age 55 years, PSA 3.0 ng ml–1, and a normal digital rectal exam (DRE). Men underwent annual screening with PSA and DRE: a PSA >4 ng ml–1 or abnormal DRE prompted biopsy. At study completion, all patients not earlier diagnosed with prostate cancer underwent an end-of-study prostate biopsy.
The study ended 15 months before its intended completion date because of a significantly decreased incidence of prostate cancer in patients taking finasteride.12 The prevalence of prostate cancer was 18.4% for finasteride versus 24.4% for placebo: a 25% decreased risk for finasteride. Initially, it was reported that those in the finasteride group with cancer had an increased prevalence of higher-grade tumors (Gleason >6). Further analyzes of the PCPT data, however, suggested that the increased detection of high-grade tumors in the finasteride arm resulted from diagnostic biases introduced by finasteride-induced prostate volume reductions rather than alterations in tumor characteristics.13 Men on finasteride experienced significantly more loss of libido, gynocomastia, erectile dysfunction, and reduced volume of ejaculate compared with those on placebo,12 though the clinical significance of these findings remains a matter of debate.14
The PCPT is an important study in that it showed a significantly reduced risk in biopsy-diagnosed prostate cancer in the treatment arm. Still, the clinical significance of these findings has been questioned. It is estimated that 71 healthy men would need to be treated with finasteride for 7 years to prevent one case of prostate cancer.15 Of note, the American Urologic Association and the American Society of Clinical Oncology have jointly released clinical practice guidelines on the use of 5-ARIs for prostate cancer prevention. The guidelines were based on a systematic, quantitative meta-analysis of 15 randomized-controlled trials. They concluded that asymptomatic men with PSA 3.0 ng ml–1 who are undergoing screening for prostate cancer may benefit from a discussion about the benefits and risks of finasteride for prostate cancer prevention.15
The selenium and vitamin E cancer prevention trial (SELECT) was a randomized, phase 3, placebo-controlled study that was the largest cancer prevention trial ever undertaken.16 Selenium is an essential trace nutrient that human beings ingest through plant consumption. Laboratory studies have indicated that it alters androgen receptor signaling modulation, induces apoptosis, and inhibits growth in prostate cancer cells.17, 18 Numerous epidemiological studies have observed reductions in prostate cancer risk ranging from 26 to 52%, depending on the population under study and the amount of selenium ingested.19 Vitamin E is a collective term for 4 tocopherols and 4 tocotrienols. The main dietary sources of vitamin E are vegetable oils and α-tocopherol, which is found in dietary supplements and is thought to be the most biologically active form20 Preclinical data have suggested that vitamin E potentially inhibits carcinogenic pathways in prostate cancer.21 In addition, epidemiological data have shown significantly decreased risk of incident or advanced prostate cancer with vitamin E intake in the α-tocopherol, β-carotene trial,22 the Health Professionals Follow-up study,23 the prostate, lung, colorectal and ovarian cancer screening trial,24 and the NIH-AARP diet and health study.25
The primary end point of SELECT was incident prostate cancer. Secondary end points included overall survival, cardiovascular events, and diabetes. Eligibility criteria included age >50 years (African Americans) or >55 years (all others), a serum PSA 4 ng ml–1, and normal DRE findings. SELECT opened in 2001 and finished accrual of 35 533 participants in 2004. Participants were randomized to selenium alone (200 μg per day from L-selenomethionine), vitamin E alone (400 IU per day of all-rac-α-tocopherol acetate), both, or placebo for a minimum of 7 years.
Although the original study design planned for an 11-year duration, at the recommendation of an independent data safety and monitoring committee, SELECT ended 4 years early because no evidence of benefit from vitamin E, selenium, or the combination of the two was convincingly demonstrated (P<0.0001), and there was no possibility of a benefit with additional follow-up. Moreover, there were non-significant increased risks of prostate cancer and diabetes for vitamin E alone and selenium alone, respectively. Compared with placebo, the hazard ratios for prostate cancer were 1.13 (99% confidence interval [CI], 0.95–1.35; 95% CI, 0.99–1.29; P=0.06) in the vitamin E-only group, 1.05 (99% CI, 0.88–1.25; 95% CI, 0.91–1.20; P=0.52) in the selenium + vitamin E group, and 1.04 (99% CI, 0.87–1.24; 95% CI, 0.90–1.18; P=0.62) in the selenium-only group.
Similarly, the physicians health study II was a randomized, phase 3, placebo-controlled trial of vitamins C and E.26 Vitamin C (ascorbic acid) is a water-soluble anti-oxidant and acts as a free-radical scavenger. Vitamin C may potentially inhibit malignant transformation through prevention of DNA adduct formation and reduction of in vivo formation of carcinogenic N-nitroso compounds.27, 28 The primary outcome of physicians health study II was incident prostate cancer. More than 14 000 male physicians 50 years were randomized to vitamin E (400 IU synthetic α-tocopherol) every other day, vitamin C (500 mg synthetic ascorbic acid) every day, or placebo. Neither vitamin C nor vitamin E supplementation reduced the risk of prostate or total cancer over an 8-year study period.
The outcomes of physicians health study II and SELECT have dampened enthusiasm for prostate cancer prevention using micronutrient or vitamin supplements. Potential explanations for the lack of efficacy of these supplements include incorrect dosages, imperfect study designs, flawed interpretations of earlier observational evidence and—perhaps most importantly—absence of a clinically significant physiological effect.
Men at higher risk for prostate cancer diagnosis: the Reduction by Dutasteride of Prostate Cancer Event study and toremifene
Despite the promising findings of the PCPT, null results for primary prevention studies of micronutrients in men at relatively low risk of incident prostate cancer highlight the potential pitfalls of prevention studies in populations with a low prevalence of the disease of interest. Other studies have targeted higher-risk populations. The reduction by dutasteride of prostate cancer events (REDUCE) trial was designed to determine whether the 5-ARI dutasteride would reduce prostate cancer incidence in higher-risk patients. There are two well-characterized isoforms of 5-ARI.29, 30 Type 1 is expressed predominantly in extra-prostatic tissue, and its function in human physiology remains to be defined. Type 2 is expressed in high quantities in the prostate.31 A third isoform has been described; its significance remains unknown.32 Finasteride is a selective 5-ARI and blocks only the 5-ARI isoform 2; dutasteride is non-selective and blocks isoforms 1 and 2. Although type 2 is the predominant form in the prostate, type 1 potentially has a more substantial function in the development of prostate cancer than was earlier thought.10, 33, 34, 35 As dutasteride inhibits both isoforms, it results in greater reductions in serum and intraprostatic DHT levels than finasteride.36, 37, 38, 39
REDUCE was a 4-year, multi-institutional, RCT of 0.5 mg daily oral dutasteride versus placebo in 8000 men. Inclusion criteria included PSA 2.5–10 ng ml–1 for men 50–60 years or 3.0–10 ng ml–1 for men 60–75 years; prostate volume <80 ml; American Urologic Association symptom score <25; and a prostate biopsy negative for cancer, high-grade prostatic intraepithelial neoplasia (HGPIN), and atypical small acinar proliferation within the earlier 6 months. Patients received PSA screening every 6 months with 10 core prostate biopsies for cause (at any time based on the clinical judgment of the treating physician) and at years 2 and 4. Preliminary results indicated that dutasteride resulted in a 23% reduced risk of biopsy-detectable prostate cancer compared with placebo: 29% of patients taking placebo were found to have prostate cancer compared with 22.5% on dutasteride. Importantly, the risk of developing high-grade tumors was similar between groups.29, 30
REDUCE has several important implications with respect to the use of 5-ARIs to prevent prostate cancer. First, it confirmed the findings of the PCPT: 5-ARIs can reduce the risk of biopsy-detectable disease. Second, it validated the post hoc PCPT analyzes showing that 5-ARI prophylaxis was not associated with increased risk of high-grade disease. Third, it showed the validity of using 5-ARI prophylaxis in a patient population at higher risk of prostate cancer diagnosis.
Finally, one additional study of hormonal manipulation in higher-risk populations deserves mention: an RCT of the selective estrogen receptor modulator toremifene in men with biopsy-proven HGPIN. Estradiol is thought to contribute to the origination of HGPIN by inducing hyperplasia and dysplasia, though the mechanism has not been fully elucidated.40 Toremifene has mixed estrogen agonist and antagonist activities and is primarily used for the treatment and prevention of estrogen receptor-positive breast cancer. HGPIN is generally believed to be a precursor lesion to prostate cancer, and it was originally thought that isolated HGPIN on biopsy indicated an increased risk of prostate cancer diagnosis on subsequent biopsy.41 More recent studies have since refuted this principle.42, 43
Five hundred and fourteen patients with biopsy-proven HGPIN were randomized to 20, 40, or 60 mg toremifene, or placebo. Prostate biopsies were repeated at 6 months and 1 year. Although toremifene cohorts showed potential efficacy at all three dosages, only those in the 20 mg arm had a statistically significant decreased risk of cancer on repeat biopsy compared with placebo (24.4% versus 31.2%, P<0.05). There was no increased risk of serious adverse events or thromboembolic events in the toremifene cohorts. The authors estimated that toremifene would prevent 6.8 cancers annually per 100 men treated with toremifene. Although it is unclear why only the 20 mg arm showed a significant reduction in the risk of prostate cancer, the authors hypothesize that it was the greater selectivity and inhibition of the α subtype of the estrogen receptor, which stimulates prostate growth. Still, given current clinical recommendations for HGPIN, toremifene should not be considered for prostate cancer prevention at this time.42, 44
Tertiary prevention focuses on halting disease progression and recurrence in patients with prostate cancer. The widespread use of PSA screening has led to a stage migration in prostate cancer, with men being diagnosed with low-pathologic and low-clinical stage disease compared with historical cohorts.45 As a result, many tumors that were not earlier detected are now being diagnosed at early stages and are potentially amenable to tertiary prevention. In addition, treatment with intent to cure at younger ages allows for a longer window of time for potential disease recurrence or progression during a patient's lifetime.
For these reasons, hormonal and dietary prevention strategies are currently under study in prostate cancer patients. Importantly, studies of tertiary prevention are all ongoing: to date, there are no results from large RCTs on which to base clinical recommendations.
Reduction by Dutasteride of Clinical Progression Events in Expectant Management, Therapeutic Assessment of Rising PSAs, Avodart after Radical Therapy for Prostate Cancer and Men's Eating And Living studies
Reduction by dutasteride of clinical progression events in expectant management is an ongoing trial testing the hypothesis that the 5-ARI dutasteride will prevent disease progression in patients with low-risk prostate cancer on active surveillance. Three hundred men aged 50–80 years with biopsy-proven clinical T1c or T2a prostate cancer diagnosed within the earlier 6 months were randomized to receive dutasteride 0.5 mg per day or placebo for 3 years. At baseline, participants had Gleason sum 6 disease and a serum PSA 10 ng ml–1. Repeat biopsies will be performed at 1.5 and 3 years, and PSA results will be provided to physicians and participants. The primary study end point is time to disease progression (defined as time to treatment or pathologic progression). Of note, these investigators hypothesize that dutasteride may reduce anxiety of patients on active surveillance by decreasing PSA and potentially dampening PSA spikes. Reduction by dutasteride of clinical progression events in expectant management finished accrual in 2007 and is expected to complete follow-up in 2010.46
Similarly, the avodart after radical therapy for prostate cancer study is a 2-year RCT to test the hypothesis that dutasteride will reduce disease progression in patients with biochemical recurrence after primary treatment for prostate cancer. Biochemical (PSA-only) recurrence after radical prostatectomy or radiotherapy for localized prostate cancer occurs in 27–53% of patients within 10 years.47 Biochemical recurrence predates clinically detectable metastatic disease by an average of 8 years.48 In 1995, Andriole et al.49 reported that men with detectable PSA after radical prostatectomy treated with finasteride experienced delay in the rise of serum PSA and reduction in local and distant recurrences compared with a placebo cohort. It is hypothesized that dutasteride may be used for tertiary prevention of progression from PSA-detectable disease only after primary treatment to clinically detectable metastatic disease and thus prevent or delay the need for second-line treatment.50
Avodart after radical therapy for prostate cancer study is an ongoing European trial in which patients were initially treated for prostate cancer with intent to cure and subsequently experienced biochemical recurrence. Participants are stratified by earlier therapy and randomized to dutasteride daily or placebo. Patients who underwent earlier radical prostatectomy must experience three PSA rises from nadir, with each being 4 weeks apart, and a PSA between 0.4 and 10 ng ml–1 at enrollment. Patients who underwent earlier radiation therapy must wait at least 1 year from end of radiotherapy and then must have three increases in PSA from nadir, with each rise 4 weeks apart and an enrollment PSA of 2–20 ng ml–1. Exclusion criteria include men with rapid (<3 months) or prolonged (>24 months) PSA doubling time. It is estimated that 276 patients will enroll. Study end points include time to PSA doubling, time to disease progression, treatment response (PSA decrease or increase of 15% from baseline), changes in PSA and PSA doubling time, and changes in anxiety. PSA will be measured every 3 months and patients with evidence or signs of disease progression will be offered alternative treatment strategies. The results from this study are likely to be unavailable for several years.
The therapeutic assessment of rising PSAs study is an ongoing RCT designed to determine the impact of dutasteride on patients with castrate-resistant prostate cancer on bicalutamide.51 The rationale behind this study is that low levels of circulating testosterone may still be converted to DHT by 5-α reductase, and this DHT has a greater affinity for the androgen receptor than bicalutamide. Thus, more complete blockage of the androgen receptor will theoretically be achieved if it is blocked directly with bicalutamide and concurrently intraprostatic DHT levels are suppressed with dutasteride. Approximately 150 subjects with PSA progression despite androgen deprivation therapy, with PSA between 2 and 20 ng ml–1 and serum testosterone <50 ng per 100 ml, and who have a negative bone scan within 8 weeks of enrollment and anticipated >2-year survival, are randomized to bicalutamide 50 mg once daily and either dutasteride 3.5 mg daily or placebo.
Finally, the men's eating and living study is a multi-center, 5-year, National Cancer Institute-funded RCT expected to open for accrual in 2010 through Cancer and Leukemia Group B. The first national trial of diet change for prostate cancer, men's eating and living will test the hypothesis that a dietary intervention consisting of high vegetable intake (primarily tomatoes, carotenoids, and crucifers) will decrease disease progression in patients with low-risk prostate cancer on active surveillance. Robust preclinical and epidemiological evidence suggest that diet may alter prostate cancer initiation and progression. Macronutrients and micronutrients associated with decreased prostate cancer risk include carotenoids (particularly lycopene), cruciferous vegetables, dietary fat, soy products, and omega fatty acids.52 A range of phytochemicals in fruits and vegetables could have effects on a metabolically active organ such as the prostate, and a number of plausible mechanisms have been proposed.53
Men's eating and living will recruit 440 participants with biopsy-proven prostate cancer, clinical stage T2a diagnosed within the earlier 24 months with 2 tissue cores positive for cancer, 50% of any one core positive for cancer, Gleason sum 6 (men 70 years) or biopsy Gleason score (3+4) =7 (men >70 years), and PSA 10 ng ml–1. The intervention will consist of centralized, telephone-based diet counseling that was earlier shown in a pilot study to be highly effective at changing dietary habits of prostate cancer patients.54, 55 The primary outcome variable will be disease progression as determined by PSA and pathology on repeat biopsy; secondary outcome variables will include incidence of active treatment, quality of life, and anxiety related to prostate cancer. The results are not expected before 2015.
Prostate cancer prevention: challenges and opportunities
As these agents diffuse into clinical practice, several issues remain unresolved. First, the economic costs of adopting a widespread prevention strategy remain undefined. Analyzes of PCPT and surveillance, epidemiology, and end results data show that, in terms of dollars per life year saved, finasteride is not a cost-effective strategy for prostate cancer prevention in a population of lower-risk men.56, 57 Further study, including the potential for 5-ARIs to prevent incident BPH, is required.
Second, although a preponderance of clinical evidence currently supports the interpretation that 5-ARIs do not increase the risk of high-grade disease or produce other serious side effects, the potential repercussions of chronic administration, spanning decades, remain unknown. It is also not known how discontinuation of 5-ARIs after many years may influence cancer biology or prostate physiology.
Third, since the target population comprises many millions of men, it is unclear who should lead the development and application of prevention strategies. Any successful implementation of a prostate cancer prevention program will likely necessitate a coordinated partnership of multiple disciplines including urology, primary care, epidemiology, and health policy. Still, although many questions remain unanswered, the potential for these agents to attenuate the population effects of prostate cancer represents a tremendous opportunity to serve the public health.
Prostate cancer prevention: clinical recommendations
In summary, based on available evidence and published clinical guidelines, we recommend the following:
5-ARIs should be considered for prostate cancer prevention in: (1) asymptomatic men with a PSA 3.0 ng ml–1 who are undergoing or anticipate undergoing PSA screening for early detection of prostate cancer and (2) asymptomatic men with PSA 2.5 and 10 ng ml–1 and an earlier prostate biopsy negative for cancer.
There is no definitive clinical evidence to support the use of 5-ARIs to inhibit cancer progression in active surveillance or biochemical recurrence patients, or to increase the sensitivity of DRE, PSA, and prostate biopsy for detecting progression in active surveillance patients. 5-ARIs may be considered in active surveillance patients for the treatment of BPH/lower urinary tract symptoms.
Men should be informed of the potential risks of 5-ARIs, although studies to date indicate that the risk of high-grade cancer is unverified.
Owing to the lack of efficacy and the potential for harmful side effects, micronutrients—including selenium, vitamin E, and vitamin C—should not be used for prostate cancer prevention at this time.
About this article
BMC Medicine (2011)