Finasteride as a Chemopreventive Agent in Prostate Cancer: Impact of the Prostate Cancer Prevention Trial on Urologic Practice

Manlio A. Goetzl; Jeffrey M. Holzbeierlein

Nat Clin Pract Urol. 2006;3(8):422-429. 

Summary and Introduction

Summary

Prostate cancer chemoprevention involves the use of natural and/or synthetic agents that inhibit or reverse the development of precancerous lesions or delay progression of these lesions to invasive disease. The recent completion of the first Phase III trial for prostate cancer prevention, the Prostate Cancer Prevention Trial (PCPT) using the drug finasteride, has provided the urologic community with the first evidence that a chemopreventive agent can reduce the risk of developing prostate cancer. The enthusiasm for the clear relative risk reduction in the finasteride arm of the trial has been tempered by the observation that the incidence of high-grade tumors was higher in men receiving finasteride compared to those on placebo. A question remains about whether the observed higher incidence in high-grade tumors is real or whether it is related to a pathologic or sampling artifact. The PCPT has instigated a great deal of debate, resulting in the larger urologic community being reluctant to recommend the widespread use of finasteride as a chemopreventive agent. This review summarizes the PCPT, analyzes its controversial results, and describes future prostate cancer chemoprevention studies.

Introduction

Although advances have been made in the diagnosis and treatment of prostate cancer, the disease remains the third leading cancer killer of men in the US in 2006. The substantial mortality and morbidity associated with prostate cancer and its treatments have led some medical professionals to shift attention away from diagnosis and treatment and instead focus on prevention.[1] Prostate cancer chemoprevention involves the use of natural or synthetic agents that inhibit precancerous lesions from developing or delay the progression of these lesions.[2] Chemotherapeutic prevention studies are an attractive option in prostate cancer given the high incidence of disease, the known hormonal influence on the manifestations of the disease, and the long latency period between the initiation of a premalignant phase and the progression to invasive disease.[3] The goal of these studies is the reduction in prostate cancer incidence.

The completion of the first Phase III trial for prostate cancer prevention, the Prostate Cancer Prevention Trial (PCPT) using the drug finasteride (Proscar®, Merck, Whitehouse Station, NJ), has provided the urologic community with the first evidence that a chemopreventive agent can reduce the risk of the development of prostate cancer.[4] What is most interesting about the findings of this study is that no other cancer agent has been shown to reduce the incidence of developing disease in a relatively low-risk population. This fact is significant because it represents the majority of men who develop prostate cancer. The results of the PCPT, however, also provoked a number of questions. In this article, we review the PCPT, discuss some of the questions provoked by the study's findings, and describe future studies relating to prostate cancer chemoprevention.

Hormonal Therapy for Prostate Cancer

The observation that prostate cancer is an endocrine-dependent tumor was first made in the 1940s when Huggins and Hodges showed that castration was effective in reducing the symptoms of metastatic cancer.[5] Since that time, we have improved our understanding of the role that testosterone and dihydrotestosterone (DHT) play in the development of both the normal and the malignant prostate gland. After synthesis in the Leydig cells of the testes, free testosterone enters the prostate gland where it is converted to DHT.[6] This reaction is catalyzed by the membrane-bound enzyme, steroid 5-alpha reductase (SRD5A). Two isoenzymes have been described, SRD5A1 and SRD5A2, with SRD5A2 being the predominant enzyme in the prostate gland.[7] Once formed, DHT binds to the intracytoplasmic androgen receptor, and the receptor–androgen complex is translocated to the cell nucleus.[8] This complex binds to DNA and leads to increased protein synthesis and eventually cellular proliferation.

In the early 1990s, finasteride, a 4-aza steroid compound and type 2 5-alpha reductase inhibitor, was shown to profoundly affect the size of the prostate gland in animals. George et al.[9] administered finasteride to noncastrated rats for 7 days and found that the drug caused a 55% decrease in prostate weight and an 87% decrease in 5-alpha reductase enzyme activity. Laroque et al.[10] treated mature dogs with increasing doses of finasteride for either 27 or 53 weeks and found that prostate shrinkage was the result of a decrease in both glandular and fibromuscular components of the prostate gland. After promising results in animal studies, finasteride was first used in patients with benign prostatic hyperplasia (BPH) and was found to markedly suppress serum DHT levels and to substantially decrease prostate volume.[11] Furthermore, early studies suggested that the suppression of DHT might inhibit the carcinogenic transformation of prostate cells.[12] These early studies provided the clinical rationale for conducting the PCPT.

Overview of the PCPT

The PCPT, sponsored by the National Cancer Institute and the National Institutes of Health (NIH), was started in October 1993. It was a prospective, randomized clinical trial undertaken to evaluate whether finasteride (5 mg/day) significantly reduced the period prevalence of prostate cancer in initially healthy men compared to a group of men taking a placebo.[4] The period prevalence relates to the pathologically proven presence of prostate cancer at the end of the trial. The inclusion criteria for the trial specified that the men had to be over 55 years of age with a normal digital rectal examination (DRE) and a PSA level of less than 3.0 ng/ml. A negative biopsy was not required for inclusion in the study. The investigators planned a trial lasting 7 years. Annually, the study participants in both groups underwent a DRE and PSA level measurement.

A centralized safety monitoring committee ensured that men taking finasteride had a recommendation rate for biopsy that was approximately equal to men in the control arm. In order to meet this goal, the investigators initially doubled the PSA value of the finasteride-treated men and used a PSA cutoff of over 4.0 ng/ml for biopsy, using information from the Proscar Long-Term Efficacy and Safety Study (PLESS) trial.[13] Thus, the investigators used the adjusted PSA value for the finasteride group and the unadjusted value in the control group. In order to maintain equal numbers of biopsies in both arms, the investigators had to increase the factor adjustment to 2.3 in the finasteride arm at the beginning of each man's fourth year on the study. For-cause biopsies were performed for a PSA level, either adjusted or unadjusted, of over 4.0 ng/ml or for any abnormal DRE. If the biopsy was positive, the man was removed from the study. Any subject in either group who, at the end of 7 years, had a negative biopsy or had no predetermined indication for a biopsy was offered an end-of-study biopsy.

Over 18,800 men were randomized between January 1994 and May 1997 to the finasteride group (9,423 men) or the placebo group (9,457 men). 9,060 were included in the final analysis on the basis of a biopsy, either for-cause or at the end of the study. Prostate cancer was detected in 803 of 4,368 men (18.4%) taking finasteride as opposed to 1,147 of the 4,692 men (24.4%) in the placebo group (P <0.001). This was a 24.8% relative risk reduction in prevalence of prostate cancer during the trial (  ).

Table 1.  Results from the Prostate Cancer Prevention Trial Showing the Numbers of Men Who Underwent Biopsy in Each Group and the Numbers Subsequently Diagnosed With Prostate Cancer

Type of biopsy n with confirmed prostate cancer in the finasteride arm (n men biopsied) n with confirmed prostate cancer in the placebo arm (n men biopsied) Level of significance
Total biopsies 803 (4368) 1147 (4692) <0.001a
For-cause biopsies 435 (1639) 571 (1934) 0.05
End-of-study biopsies 368 (3652) 576 (3820) <0.001

 

aThere was a relative risk reduction (95% CI) of 24.8% (18.6–30.6%) in the men in the finasteride treatment group compared to placebo.

When the results of the biopsies were examined, the investigators found a higher proportion of high-grade tumors (Gleason score >7) in the finasteride group (6.4%) compared with the controls (5.1%) (P = 0.005; Figure 1). More specifically, in patients who underwent a for-cause biopsy, 47.8% of the finasteride group had high-grade cancer versus 29.4% of controls. In men with an indication for an in-study biopsy, those who had taken finasteride had a 1.6-fold increased risk of having high-grade prostate cancer compared to men taking the placebo.

Figure 1.

 

Distribution of Gleason scores in men diagnosed with prostate cancer in the finasteride and placebo arms in the Prostate Cancer Prevention Trial. This figure shows how the Gleason scores are distributed amongst the 803 men in the finasteride group and the 1,147 men on placebo who were diagnosed with prostate cancer. More men in the placebo group were diagnosed with cancer compared with the treatment group. Most men with prostate cancer in both groups had tumors of Gleason grade 5–6. The placebo group had a higher number of low Gleason score tumors. Tumors of higher Gleason grade (>7) were more common in the finasteride group. an = 4368, n prostate cancer cases = 803; bn = 4692, n prostate cancer cases = 1,147.

With regard to side effects, sexual dysfunction was more common in the finasteride group as decreased libido, decreased ejaculate volume and erectile dysfunction are known adverse effects associated with the drug. Lower urinary tract symptoms and urinary retention were more common in the placebo group, as one would expect given the fact that finasteride reduces the size and thus the obstructive effects of an enlarged prostate. Notably, five men in each group died from prostate cancer.

Discussion of the PCPT Findings

The 24.8% relative risk reduction in prostate cancer seen in the finasteride arm prompted the early termination of the study on February 21, 2003, 15 months before its anticipated completion. While the trial achieved its primary end point more than a full year before its completion, the study's findings have raised many questions, thus delaying the full implementation of its conclusions.

High-grade cancers were significantly more likely to be found in men taking finasteride compared to those on placebo. This is an important finding, because the biopsy Gleason grade provides an assessment of the biologic aggressiveness of the cancer and, thus, has a clear impact on patient outcome.[14] Tumors with a high-grade component (either a Gleason 4 or 5) have been found to portend a worse prognosis after either radical prostatectomy or radiation therapy.[15] The results, therefore, seemed to suggest that androgen deprivation, specifically decreased levels of intracellular DHT, promoted more biologically aggressive tumors.

In patients treated with luteinizing hormone-releasing hormone (LHRH) agonists, histologic changes can occur which have characteristics similar to high-grade disease.[16,17] LHRH agonists have been shown to alter the histologic features of prostate cells (loss of glandular architecture, cytoplasmic vacuolization, etc.) making the cells appear more aggressive and poorly differentiated.[17] Prostate cells that would be typically Gleason 3 might have the appearance of cells more characteristic of a Gleason 5 tumor.[18] Some pathologists advise against the assignment of a postprostatectomy Gleason score in patients treated with neoadjuvant LHRH therapy because it is not reliable.[19] These studies, however, were on patients treated with LHRH agonists, which reduce both cellular testosterone and DHT.

Few studies have evaluated the effect of finasteride, with its decrease in DHT but its relative increase in testosterone, on the morphology of prostate cancer cells and the impact of the drug on the Gleason score. In a small study, Civantos et al.[16] concluded that there was a risk of overgrading the Gleason score in patients who had been treated with finasteride. Bostwick et al.[20] concluded that the Gleason grade should not be used to classify prostate cancer in patients that had been taking finasteride, as it might lead to an overestimation of the biological potential of high-grade cancer. A recent study by Rubin et al.[21] reached a different conclusion, however. The study examined 56 patients with prostate cancer who had taken finasteride for at least 6 months and compared them with a matched control group of 56 patients, and a group consisting of 44 patients who received a 3-month course of a LHRH agonist. They found no consistent histologic effect in the finasteride group compared with the LHRH agonist group. Furthermore, the hormonal effect was even observed in the control group, which was comprised of patients who had never undergone hormonal manipulation. Further studies are needed to fully delineate the cellular changes that finasteride induces in prostate cancer cells and also to clarify the applicability of the Gleason score in grading finasteride-treated glands.

Controversy remains about whether the increase in proportion of high-grade Gleason tumors seen in the finasteride treatment arm was real or artifactual.[22] When looking at the numbers of for-cause biopsies, 1,934 men in the placebo group were biopsied as opposed to 1,639 in the finasteride group. Since it has been shown that there is a significant positive linear correlation between PSA value and Gleason score (the higher the PSA level, the higher the Gleason score), we would expect a higher biopsy rate in the finasteride group if there were truly more high grade tumors.[23] If these tumors were truly high grade, they would cause an increase in PSA level that would increase the chance of a for-cause biopsy. The for-cause biopsy rate, however, was higher in the placebo group, raising the possibility that men in the placebo group also harbored high-grade tumors that might have simply been missed on biopsy.

Yet another argument against the induction of higher grade tumors in the finasteride arm is the lack of an increase in higher grade tumors over time. The cumulative proportional incidence of newly diagnosed high-grade tumors in the finasteride arm increased at a similar rate to the controls (Figure 2). The fitted lines in both groups have the same slope or rate of increase. If finasteride induced higher grade tumors, one would expect the incidence in the finasteride arm to increase at a greater rate, particularly as the trial progressed and the prostate cells were exposed to finasteride for a longer duration. The finasteride curve theoretically would have diverged from the placebo curve over time; however, this was not the case.[24]

Figure 2.

 

Cumulative incidence of high-grade prostate cancer diagnosed from either for-cause biopsies or biopsies taken after an interim procedure. The number at risk is the number of surviving men still being followed-up who were free from prostate cancer, and the number of events includes all cases of high-grade prostate cancer (Gleason grade 7–10) detected on a biopsy performed for-cause after an interim procedure (i.e. transurethral resection of the prostate). The cumulative proportion incidence or newly diagnosed high-grade tumors in the finasteride arm increased at a similar rate to the controls. The fitted lines in both groups have the same slope or rate of increase. If finasteride induced higher grade tumors, one would expect the incidence in the finasteride arm to increase at a greater rate, particularly as the trial progressed and the prostate cells were exposed to finasteride for a longer duration. The finasteride curve does not markedly diverge from the placebo curve over time. Permission obtained from Massachusetts Medical Society © Thompson IM et al. (2003) New Engl J Med 349: 215–224.

The proportion of high-grade tumors in the finasteride group might also be a volume artifact from transrectal ultrasound of the prostate (TRUS) guided biopsy as the median prostate volume of men in the finasteride group was smaller than controls. In a recent study, Kulkarni et al.[25] evaluated the risk factors in their institution for Gleason 4 or 5 grade cancer in their biopsies. Risk factors for high-grade disease included older age (P = 0.01), presence of hypoechoic lesions on TRUS (P <0.001) and prostate volume using TRUS (P = 0.008). Prostate volume was negatively correlated with Gleason 4 or higher grade cancer in biopsy specimens. Any high-grade disease would have an increased chance of being sampled in smaller prostate glands, as long as the volume of each needle core and the number of cores taken are constant. In the PCPT, the median prostate volume among men in the finasteride group was 25.5 cm3 versus 33.6 cm3 in controls (a 24.1% reduction in the treatment arm). The reduction in prostate volume would have increased the likelihood that cancer, and particularly high-grade disease, if present, would have been sampled on biopsy. Interestingly, Kulkarni et al.[25] found that when the prostate was fully examined after prostatectomy, TRUS-determined volume became insignificant, raising the possibility that high-grade disease in the PCPT control group was simply missed when their relatively larger prostates were biopsied using the same size needle core.[25] An extensive evaluation of biopsy and prostatectomy specimens in the PCPT is currently under review, and it remains to be seen if the increase in high-grade tumors remains true when the finasteride-treated men have their prostates fully sampled after prostatectomy.[26]

Clinical Relevance of the PCPT Results

The decrease in prevalence of prostate cancer seen in men in the finasteride arm of the PCPT primarily relates to Gleason 6 cancer, because the prevalence of high-grade tumors was higher in the experimental arm. In the finasteride group, 51.3% of tumors were Gleason score 6 versus 61.6 % in the control group. Are the Gleason 6 tumors that were prevented in the finasteride arm clinically significant (i.e. do they have the ability to become symptomatic and or at the worst, lead to death)? Clearly, the urologic community views Gleason score 6 cancers as clinically significant. Fowler et al.[27] performed a study in which they asked both urologists and radiation oncologists the following question: "A prostate cancer patient of yours is 65, in good health, and has a negative DRE result and no evidence of extraprostatic spread. Which primary treatment would you be most likely to recommend for localized prostate cancer if this patient has the following biopsy results?" In patients with Gleason score 5–6 tumors and a PSA level of 10–20 ng/ml, 85% of urologists would recommend radical prostatectomy; when the PSA drops to 5 ng/ml, 93% would recommend surgery. Although these are recommendations, urologists clearly believe that surgical management for this potentially aggressive tumor offers these patients the best prognosis for complete cure.

The risks of prostate cancer mortality from a tumor with a Gleason score of 6 can be very significant. In a contemporary study of a cohort of patients who died of prostate cancer, Thompson et al.[28] found that a third of patients had tumors of Gleason 6 or less. These data impact on the discussion of treatment options that doctors have with patients who are newly diagnosed with Gleason 6 tumors. In a study of the large CaPSure database, 95% of men who presented to either an academic or community urologic practice with newly diagnosed Gleason 6 tumors chose to be treated immediately.[29] These treatments, either surgery, radiation or brachytherapy, are associated with significant costs to society and patient morbidity. If finasteride is able to prevent the burden of diagnosis and the adverse effects of treatment of these moderate-risk tumors, then the effect should be viewed as significant, both from a patient and a public health perspective.[30]

The use of finasteride in the chemoprevention of prostate cancer also has significant economic consequences if we assume that men will take the pill daily from the age of 55 years onwards. This could be an issue, because the cost of years of treatment for a large proportion of men who might ultimately not benefit from treatment could be high.[31] In their cost-effectiveness model, Zeliadt et al.[31] assumed that 60% of men 55 years and older would be eligible for therapy, and 50% of those eligible would take finasteride. They then calculated that the annual medical care expenditures would increase by $2.2 billion for men aged 55–64 and $1.2 billion for men 65 years and older, even after accounting for savings from the reduction in prostate cancer and BPH incidence. If finasteride was shown to prevent high-grade disease as well as low-grade disease, however, then treatment might be more cost-effective. It should be noted that the patent on finasteride has expired, and a generic version might lessen the financial impact of using the drug in chemoprevention.

The optimal dose and dosing schedule of finasteride in chemoprevention remains undetermined. The PCPT used the standard 5 mg daily dose, as this is what had been approved for the treatment of symptomatic BPH. If a smaller dosage or less frequent dosing schedule is shown to provide the same benefit, then this might also lessen the societal cost of chemoprevention.

Future Studies

The PCPT has instigated a great deal of debate resulting in the limited use of finasteride for chemoprevention in the urologic community. The reluctance to initiate the widespread use of finasteride was predominantly a result of the finding that the drug increased the risk of developing high-grade cancer. Clearly, further studies are needed to re-examine the histologic characteristics of the prostate specimens from patients in the PCPT. What will be most interesting is to follow the patients long term in order to determine if those who took finasteride, developed prostate cancer, and were treated, have outcomes that are similar, worse, or better than men in the control group. As finasteride was associated with some significant adverse effects in the trial (primarily sexual dysfunction) quality of life analyses will be very important to examine as well.

The findings from another large trial currently underway will also more clearly define the role of 5-alpha reductase inhibitors in prostate cancer prevention. The REDUCE (The Reduction by Dutasteride of Prostate Cancer Events) trial is a randomized, double-blind, placebo-controlled trial currently enrolling patients in the US and Europe.[32] The investigators are seeking to enroll approximately 8,000 men aged 50–75 years and to randomize them to dutasteride (Avodart®, GlaxoSmithKline, Philadelphia, PA) 0.5 mg daily, or placebo, for 4 years. In some in vitro studies, dutasteride has been shown to be a more efficient SRD5A2 inhibitor than finasteride.[33] Dutasteride is also an inhibitor of SRD5A1 which predominates in prostate cancer cell lines and appears to be overexpressed in some prostate cancers.[32] Eligible men must have a PSA level of at least 2.5 ng/ml (ages 50–60 years) or at least 3.0 ng/ml (age >60 years) and they must have a negative 6–12 core biopsy (no evidence of adenocarcinoma or high-grade prostatic intraepithelial neoplasia within the 6 months before enrollment). Ten-core biopsies will be performed in all participants at 2 and 4 years. In addition to the study mandated biopsies at years two and four, for-cause biopsies can also be performed. The study end-point will be the incidence of biopsy-proven cancer.

Conclusion

The PCPT has shown finasteride to be an effective chemopreventive agent in low-grade prostate cancer. Enthusiasm for its widespread use has been tempered by the reported increase in high-grade cancers in low-risk men who had taken the drug. Recent studies seem to suggest that finasteride causes a detection bias in biopsy specimens, due mainly to its effect on prostate volume. Future information from both further analyses of the PCPT and from the results of the ongoing REDUCE trial will be helpful in clarifying the chemopreventive role of 5-alpha reductase inhibitors. In addition, questions such as economic viability, dose and dose schedule, and whether dutasteride is more effective than finasteride need to be answered before widespread use of these drugs as standard chemopreventive agents in prostate cancer.


CLICK HERE for subscription information about this journal.

Sidebar: Key Points

  • The recent completion of PCPT using the drug finasteride has provided the first evidence that a chemopreventive agent can reduce the risk of developing prostate cancer

  • High-grade cancers were significantly more likely to be found in the finasteride group compared to controls

  • The increase in high-grade cancers in the experimental arm might be related to a histologic or sampling artifact in the study

  • Further studies on the biopsy and prostatectomy specimens from men in the PCPT as well as the ongoing REDUCE trial should help to confirm whether 5-alpha reductase inhibitors promote higher-grade tumors; the results of these studies are needed before the use of 5-alpha reductase inhibitors can be widely accepted for prostate cancer chemoprevention

References

  1. Kelloff GJ et al. (2001) Agents, biomarkers, and cohorts for chemopreventive agent development in prostate cancer. Urology 57 (4 Suppl 1): 46–51

  2. Lieberman R (2003) Evolving strategies for prostate cancer chemoprevention trials. World J Urol 21: 3–8

  3. Djavan B et al. (2004) Chemotherapeutic prevention studies of prostate cancer. J Urol 171: S10–S14

  4. Thompson IM et al. (2003) The influence of finasteride on the development of prostate cancer. New Engl J Med 349: 213–222

  5. Huggins C et al. (1941) Studies on prostate cancer. I. The effect of castration, estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1: 293–297

  6. Hsing AW et al. (2002) Hormones and prostate cancer: current perspectives and future directions. Prostate 52: 213–225

  7. Thigpen AW et al. (1993) Tissue distribution and ontogeny of steroid 5 alpha-reductase isozyme expression. J Clin Invest 92: 903–910

  8. Andriole G et al. (2004) Dihydrotestosterone and the prostate: the scientific rationale for 5 alpha-reductase inhibitors in the treatment of benign prostatic hyperplasia. J Urol 172: 1399–1403

  9. George FW et al. (1991) Feed-forward control of prostate growth: dihydrotestosterone induces expression of its own biosynthetic enzyme, steroid 5 alpha-reductase. Proc Natl Acad Sci USA 88: 8044–8047

  10. Laroque PA et al. (1994) Effects of chronic oral administration of a selective 5 alpha-reductase inhibitor, finasteride, on the dog prostate. Prostate 24: 93–100

  11. Stoner E (1990) The clinical development of a 5 alpha-reductase inhibitor, finasteride. J Steroid Biochem Mol Biol 37: 375–378

  12. Brawley OW et al. (1994) 5-alpha reductase inhibition and prostate cancer prevention. Cancer Epidemiol Biomarkers Prev 3: 177–182

  13. Andriole GL et al. (1998) Treatment with finasteride preserves usefulness of prostate-specific antigen in the detection of prostate cancer: results of a randomized, double-blind, placebo-controlled clinical trial. Urology 52: 195–201

  14. Gleason DF et al. (1974) Prediction of prognosis for prostatic adenocarcinoma by combined histological grading and clinical staging. J Urol 111: 58–64

  15. Zhou P et al. (2005) Predictors of prostate cancer-specific mortality after radical prostatectomy or radiation therapy. J Clin Oncol 23: 6992–6998

  16. Civantos F et al (1996) Histopathological effects of androgen deprivation in prostatic cancer. Semin Urol Oncol 14 (Suppl 2): 22–31

  17. Reuter VE (1997) Pathological changes in benign and malignant prostatic tissue following androgen deprivation therapy. Urology 49 (Suppl): 16–22

  18. Van de Voorde WM et al. (1994) Morphologic and immunohistochemical changes in prostate cancer after preoperative hormonal therapy. A comparative study of radical prostatectomies. Cancer 74: 3164–3175

  19. Algaba F et al. (1996) Assessment of prostate carcinoma in core needle biopsy—definition of minimal criteria for the diagnosis of cancer in biopsy material. Cancer 78: 376–381

  20. Bostwick DG et al. (2004) Does finasteride alter the pathology of the prostate and cancer grading? Clin Prostate Cancer 2: 228–235

  21. Rubin MA et al. (2005) Effects of long-term finasteride treatment on prostate cancer morphology and clinical outcome. Urology 66: 930–934

  22. Andriole G et al. (2005) The effects of 5α-reductase inhibitors on the natural history, detection and grading of prostate cancer: current state of knowledge. J Urol 174: 2098–2104

  23. Gao HW et al. (2005) Mass screening of prostate cancer in a Chinese population: the relationship between pathological features of prostate cancer and serum prostate specific antigen. Asian J Androl 7: 159–163

  24. Marberger M et al. (2003) The clinical implications of the Prostate Cancer Prevention Trial. BJU Int 92: 667–671

  25. Kulkarni GS et al. (2006) Evidence for a biopsy derived grade artifact among larger prostate glands. J Urol 175: 505–509

  26. Parnes HL et al. (2005) Prevention of hormone-related cancers: prostate cancer. J Clin Oncol 23: 368–377

  27. Fowler FJ et al. (2000) Comparison of recommendations by urologists and radiation oncologists for treatment of clinically localized prostate cancer. JAMA 283: 3217–3222

  28. Thompson KE et al. (2005) Prognostic features in men who died of prostate cancer. J Urol 174: 553–556.

  29. Harlan SR et al. (2003) Time trends and characteristics of men choosing watchful waiting for initial treatment of localized prostate cancer: Results from CaPSure. J Urol 170: 1804–1807

  30. Klein EA et al. (2005) Assessing benefit and risk in the prevention of prostate cancer: the Prostate Cancer Prevention Trial Revisited. J Clin Oncol 23: 7460–7466

  31. Zeliadt SB et al. (2005) Lifetime implications and cost-effectiveness of using finasteride to prevent prostate cancer. Am J Med 118: 850–857

  32. Andriole G et al. (2004) Chemoprevention of prostate cancer in men at high-risk: rationale and design of the Reduction by Dutasteride of Prostate Cancer Events (REDUCE) trial. J Urol 172: 1314–1317

  33. Makridakis N et al. (2005) Pharmacogenetic analysis of human steroid 5-alpha reductase type II: comparison of finasteride and dutasteride. J Mol Endocrinol 34: 617–623