Advanced prostate cancer - Nature

Prostate Cancer and Prostatic Diseases (1999) 2, 21±26 ß 1999 Stockton Press All rights reserved 1365±7852/99 $12.00

Advanced prostate cancer

http://www.stockton-press.co.uk/pcan

AM Ziada and ED Crawford Division of Urology, Campus Box C319, 4200 East Ninth Avenue, Denver, Colorado, 80262, USA

Both the de®nition and therapy of advanced prostate cancer is challenging. The advanced prostate cancer session at ``The 8th International Prostate Cancer Update'' had discussions which tried to answer the questions of management of these patients who either present with advanced disease or fail any form of therapy for clinically con®ned prostate cancer. This article provides an overview of therapeutic options: monotherapy and minimal androgen blockade options versus maximal androgen blockade, adjuvant therapy, intermittent therapy and timing of therapy as well as chemotherapy when all these measures fail. The impact of these therapies on progression as well as quality of life is reviewed.

Keywords: prostate cancer; hormonal therapy; LHRH agonists; antiandrogens; 5a-reductase inhibitors; chemotherapy

Introduction: de®ning advanced disease Prostate cancer is the most frequently diagnosed cancer in the US and the second most common cause of death in Americans.1 Despite all efforts of screening and early diagnosis of prostate cancer in recent years, a large proportion of these patients, with localized disease, progress. The concept of advanced prostate cancer is changing. The majority of patients with advanced disease now present with rising PSA following de®nitive therapy. The classic presentation of advanced disease in which patients ®rst presented with distant bony metastasis or soft tissue involvement is now decreasing. A new spectrum was suggested for stage D prostate cancer (Table 1) by Crawford and Blumenstein.2 The timing, duration and long term side effects of androgen deprivation are controversial issues. Since hormonal therapy has been proved effective in 19413 several therapies have been advocated involving androgen blockade, whether monotherapy or total.

Hormonal monotherapy options For decades, bilateral orchiectomy was the treatment of choice. Side effects were psychological problems and irreversibility. Estrogen therapy has been considered more favorable but was abandoned due to increased risk of cardiovascular complications. The VACURG study demonstrated the toxicity of estrogen therapy.4 Correspondence: Dr ED Crawford, Urologic Oncology Professor Division of Urology, Campus Box C319, 4200 East Ninth Avenue, Denver, Colorado 80262, USA.

The search for novel therapies for advanced prostate cancer was promoted by several studies. Geller5 in 1978 demonstrated the importance of adrenal androgens and postulated the importance of dihydrotestosterone (DHT). Another report by Bracci6 in 1977 showed that cyproterone acetate in combination with orchiectomy improved success rates in treating prostate cancer. LHRH agonists were developed for medical castration; and antiandrogens were synthesized to block the peripheral androgenic effect. The concept of combined androgen blockade was described by Labrie7 who noted signi®cant improvement in response and survival rates when combining LHRH agonists with antiandrogens.

LHRH agonists LHRH agonists came into use in the 1980s and produced a very similar effect to orchiectomy. They have been demonstrated to lengthen survival and time to progression.8 They are as effective as orchiectomy and estrogens without the serious side effects involved with the latter. LHRH agonists have no effect on adrenal androgens. LHRH agonists offer better quality of life. Side effects include the initial elevated testosterone level before being shut down, what is known as the ¯are phenomenon caused by LH and FSH surge. This ¯are phenomenon has not been proved to have any effect on survival. Use of an antiandrogen with the initialization of LHRH agonists has been prescribed to prevent such ¯are reaction. In contrast, LHRH antagonists have an advantage of immediately suppressing androgen production.

Advanced prostate cancer AM Ziada and ED Crawford

22

Table 1 Suggested classi®cation for stage D prostate cancer D1 D1.5 D2 D2.5 D3 D3S D4

2

Pelvic lymph node involvement Rising PSA after failed local therapy Metastatic disease in bone and/or other organs Rising PSA after nadir level Hormone refractory prostate cancer Hormonally sensitive Hormonally insensitive

Antiandrogens The antiandrogens currently available include cyproterone acetate (steroidal), ¯utamide, nilutamide and bicalutamide (non-steroidal). These agents block androgenic receptors both peripherally as well as in the hypothalamic pituitary access, resulting in LH and subsequently testosterone increase. Antiandrogens have not shown survival bene®t when compared to castration for advanced prostate cancer.9,10 Currently, antiandrogens are used in combination with either orchiectomy or with LHRH for combined androgen blockade. Nonsteroidal antiandrogens have minimal side effects on libido and sexual potency. The more common side effects include breast tenderness and swellings, which occur as a result of LH release. Also, increased testosterone levels result in some aromatization to estrogens. Libido and potency are preserved in 80 ± 100% of patients.11,12 Monotherapy with antiandrogen is presently investigated in prospective randomized trials. Adjuvant treatment with ¯utamide after radical prostatectomy compared to prostatectomy alone showed initial results which seem to indicate progression free survival bene®t at 4 y (90% vs 70% in the control group). In advanced metastatic disease, androgen monotherapy compared to LHRH analog, no clinical difference could be demonstrated. Antiandrogen monotherapy may play an increasing role in the future for patients with locally advanced disease who are not candidates for either surgery or radiotherapy. Dr Schulmann stated that in spite of combined androgen blockade being the gold standard for advanced prostate cancer, antiandrogen monotherapy may play a role in selected indications. These include T4 disease with minimal distant metastasis, T3 patients who are not candidates for either surgery or radiotherapy, adjuvant to radical prostatectomy for patients at high risk (margin positive, Gleason > 7, high initial PSA), early prostate cancer (T1c ± T2) in patients above 70 y or sexual priority for the patient.

Adjuvant antiandrogen therapy after radical prostatectomy Wirth13 presented a randomized controlled multicenter study on adjuvant antiandrogenic treatment after radical prostatectomy in stage C prostate cancer. A total of 356 patients were recruited in 25 centers in Germany and Austria. The study aimed at evaluating whether adjuvant antiandrogenic treatment can prolong progression free survival and=or survival in patients with locally

Figure 1 Tumor grade distribution in adjuvant antiandrogenic study after radical prostatectomy courtesy of Dr Wirth.

advanced prostate cancer (stage C=pT3N0). Patients were randomized to either receiving ¯utamide (250 mg tid) following surgery vs no adjuvant treatment. Progression free survival rates were 97% and 90% at 2 and 4 y respectively for the group treated with ¯utamide vs 84% and 68% for the control group. Side effects were gynecomastia in 21% and GI disturbances in 10%, leading to 21% of patients in the treatment group discontinuing therapy. These are preliminary results and longer follow up is still going on (Figure 1).

Antiandrogen withdrawal syndrome This phenomenon was ®rst reported by Kelly and Scher14 in 1993 in patients treated with combined androgen blockade using an LHRH agonist and ¯utamide. They observed that stoppage of the antiandrogen at the time of disease progression resulted in remission and PSA decline which continued for an average of ®ve months. Both steroidal and nonsteroidal antiandrogens have also shown the same effect. Although several hypothesized mechanisms have been offered to explain this syndrome, none has been con®rmed in patients treated with antiandrogen monotherapy.

Combined androgen blockade Combined androgen blockade (CAB) is the elimination of all antiandrogens by surgical or medical castration combined with antiandrogen.15 Several studies have investigated CAB, but only two have shown a statistically signi®cant survival advantage over monotherapy.16,17 CAB is most effective in patients with minimal disease and may have a subjective bene®t in those with symptomatic disease. The theory of CAB results from the hypothesis by Labrie.18 They determined that after elimination of testicular androgens, the average concentration of dihydrotestosterone (DHT) in the prostate remains at about 40% of that found in non-castrated men. Labrie reported a

Advanced prostate cancer AM Ziada and ED Crawford

96% response rate, with 29.2% complete response rate and long term survival in 37 patients with advanced prostate cancer who were treated with the combined androgen blockade. The drawback of this study is that there was no control arm.15 The superiority of CAB was documented by two major randomized studies.16,19 Dijkman20 also reported on a large double blind trial done on 457 patients randomized to placebo or nilutamide following orchiectomy. This study was designed to determine the long term ef®cacy and tolerability of nilutamide combined with orchiectomy in patients with advanced prostate cancer. The study showed signi®cant bene®t to the addition of nilutamide (a non-steroidal antiandrogen) over placebo in the interval to progression. At 8.5 y of follow-up, statistically signi®cant improvements in cancer survival, overall survival and interval to progression were demonstrated. Early normalization of PSA was shown to predict an improved long term response in terms of interval to disease progression. In spite of the fact that treatment of advanced prostate cancer is only palliative, quality of life issues bene®t from delayed progression (Figure 2). The ®rst meta-analysis of CAB used data from seven randomized double blind studies comparing orchiectomy plus nilutamide with orchiectomy plus placebo on 1056 patients with advanced prostate cancer. The group receiving CAB had a signi®cant reduction in disease progression. There was also a 10% reduction in the risk of death, but this was not statistically signi®cant.21 Other metaanalyses showed slight survival advantage to CAB.22 The most recent meta-analysis study by the Prostate Cancer Trialists' Collaborative Group (published recently) included 22 trials with a total of 5710 patients with advanced prostate cancer. Some trials used medical castration whereas others used bilateral orchiectomy as part of CAB. Overall, mortality was 56.3% in those receiving CAB vs 58.4% in those receiving medical or surgical castration alone. Five year survival estimates were 26.2% with CAB and 22.8% with castration alone. According to this meta-analysis, there is no statistically signi®cant survival advantage to CAB.23

Figure 2 Cancer survival in patients on nilutamide and ochiectomy vs patients on placebo and orchiectomy courtesy of Dr Dijkman.

Surgical or medical castration alone may be more appealing with regard to minimizing morbidity and cost, however, with the available data the superiority of such regimen to CAB cannot be proved.

Intermittent hormonal therapy The concept of intermittent hormonal therapy was ®rst described by Klotz and colleagues.24 This treatment concept suggests that therapy should continue until PSA nadir is reached. Following this, treatment should be stopped until PSA starts to rise. In the non-treatment period, serum testosterone levels would return to normal range within eight weeks as well as libido and sexual potency in patients who reported normal sexual function prior to therapy. The idea is to improve quality of life, reduce cost and delay development of hormone resistance. Also, men receiving continuous hormonal therapy are at increased risk for osteoporosis and anemia, in addition to the sexual dysfunction. Evidence does not suggest dramatic survival improvement, but at the same time no harm was reported. Laboratory experiments showed that time to hormone independent cancer may be extended using this approach.25 One suggestion is that the repetitive cycles of hormonal therapy may serve to maintain the ability of prostate cancer cells to undergo apoptosis. A lot of questions remain unanswered regarding intermittent hormonal therapy. One question is: when is the best time to restart therapy? What should be the best regimen of therapy? What should our marker be? Would intermittent therapy would offer any equivalence or advantage over continuous therapy as regards survival and cancer control? Hopefully, these questions will be answered by an ongoing SWOG trial of over 1500 patients to evaluate intermittent vs standard hormonal therapy.

5-a reductase inhibition Minimal androgen blockade using ®nasteride lowers PSA level, but has minimal effect on the disease.26 However, the combination of ®nasteride with ¯utamide has been shown by Fleshner and Trachtenberg27 to be effective in men with T3N1 prostate cancer. A very important advantage is preservation of potency. Laboratory studies have shown that the ®nasteride=¯utamide combination is superior to ¯utamide monotherapy.28 A study from the University of Colorado using a combination of low dose ¯utamide (125 mg bid) and ®nasteride (5 mg bid) in patients who present with relapse after radical prostatectomy or radiation therapy, showed consistent control of prostate cancer evidenced by lowering of PSA levels.29,30 However, further follow up is still needed. A growing number of animal studies suggest that 5alpha reductase inhibitors may have a role in prostate cancer treatment. Studies by Bologna31,32 working on prostate cancer cell lines (PC3, DU145, LnCap) demonstrated dose dependent inhibition of cell growth by ®nasteride. Clinical studies on the use of 5a reductase inhibitors have been reported either with ®nasteride alone or in combination with ¯utamide. In 1995 Fleshner and

23

Advanced prostate cancer AM Ziada and ED Crawford

24

Overall survival was signi®cantly prolonged in the patients treated early. Laboratory studies have shown that early hormonal therapy does not result in earlier androgen resistance.38 The value of early androgen deprivation lies in prolongation of disease free survival and hence improvement of the quality of life.37

Chemotherapy with mitoxantrone and prednisone Figure 3 PSA change from baseline=3 year data courtesy of Dr. Gormley.

Trachtenberg27 reported on 39 patients PSA declines of 89% from baseline, with almost 80% of their patients remaining potent. In patients with detectable PSA levels following radical prostatectomy33 signi®cant delay in PSA rise was seen in one year with continued suppression detectable to two years (Figure 3). An interesting potential application of 5-a reductase inhibitor is related to prostate cancer prevention. This can be achieved through removing DHT, an important promoter of prostate cancer with little toxicity.34 In summary, 5a reductase enzyme plays an important role in prostate physiology. The role of 5a reductase inhibitors in prostate cancer remains to be de®ned, and needs further work and studies. For the time being, the potential use in prostate cancer is in combination therapy for advanced disease, adjuvant monotherapy following radical prostatectomy, and chemoprevention of prostate cancer.

Timing of hormonal therapy: early vs deferred It is generally agreed that symptomatic men with metastatic disease should receive hormonal therapy. The controversy arises in those patients with stable disease and no symptoms. Hormonal deprivation timing in this group of patients is debatable. Deferred hormonal therapy was generally the rule after early studies based on the VACURG studies.35 The idea of delayed treatment was encouraged as a result of this study because survival was not prolonged as a result of early hormonal manipulation. Also, the use of early hormonal therapy was believed to alter the nature of metastatic lesions. An update of the VACURG study by Byar in 198836 determined that disease progression from stage C to stage D was decreased from 50 ± 10% with DES therapy. Crawford and associates16 also showed a bene®t of early hormonal therapy in patients with distant metastasis. The MRC study37 published in 1997 was a randomized study of 938 patients with locally advanced or asymptomatic metastatic prostate cancer. Patients either received immediate treatment with orchiectomy or LHRH agonist vs the same treatment deferred till symptoms occurred. Development of extraskeletal metastases, pathologic bone fractures, spinal compression and ureteric obstruction were twice as common in the deferred treatment group.

Development of advanced hormone refractory prostate cancer (HRPC) is marked by bone pain, pathological fractures, spinal compression symptoms, weight loss, fatigue and anorexia. All of this occurs with failure of patients to respond to hormonal therapy and rise of PSA levels. The average lead time is 20 weeks from rise of PSA to development of bone pain. The goal of management of patients with HRPC is symptomatic relief, improving quality of life, and improving progression free as well as overall survival. The average survival for HRPC patients used to range from 6 ± 9 months. Recent data from the Canadian group39 and from the Cancer and Leukemia Group B (CALGB)40 suggest an average survival of 11 months for this group of patients. Trials of megace41 and liarazole42 also suggest median survivals between 11 and 12 months for HRPC patients. PSA levels will vary between patients. The general rule is that PSA doubling time is approximately two months. However, each tumor is unique in its' behavior, as is each patient's physiological and psychological response. The ®rst step in managing a patient with HRPC is to evaluate the extent of disease. Patients with mildly rising PSA and otherwise normal parameters are observed. De®nitive management starts with withdrawal of antiandrogens if the patient was taking any. Antiandrogen withdrawal leads to decreased cellular proliferation. The exact mechanism is not understood, but may be the result of some mutation of the androgen receptor. Second and third line hormonal agents may be tried with some success. Eventually, hormonal manipulations fail to produce any response. Some studies, however, have tried ketoconazole and hydrocortisone in addition to antiandrogen withdrawal.43 Rapidly rising levels of PSA and LDH may suggest visceral metastasis. These levels suggest the development of true HRPC. In this group of patients, chemotherapy should be tried. In the meeting, Dr Vogelzang presented chemotherapy with mitoxantrone and prednisone. Mitoxantrone is a synthetic anthracenedione antineoplastic agent. The mechanism of action is through intercalating DNA by a hydrogen binding mechanism, induction of DNA-protein cross links thus inhibiting protein synthesis and DNA replication. In addition, mitoxantrone binds to topoisomerase II, an enzyme that permits unraveling DNA and hence inhibits DNA replication. Overall, all these actions induce apoptosis. Mitoxantrone plus a glucocorticoid was approved for the treatment of HRPC in September 1996 by the FDA. The Canadian phase III trial39 showed reduction in pain or analgesics used in 38% of patients on mitoxan-

Advanced prostate cancer AM Ziada and ED Crawford

trone and prednisone vs 21% in the group receiving prednisone alone. There was no statistically signi®cant survival difference between the groups. Importantly, the duration of palliative response was longer in the group receiving mitoxantrone and prednisone (median 43 weeks) than in the group receiving prednisone alone (median 18 weeks; P < 0.0001). There was also a higher probability of PSA reduction in the patients receiving mitoxantrone. The combination of mitoxantrone and prednisone was generally well tolerated. Side effects included severe nausea and vomiting in 0.5% of the cycles, nine neutropenic fevers in 796 cycles (130 patients), and cardiac side effects in ®ve patients, two of whom had previous congestive heart failure (Figures 4,5). In summary, the Canadian study showed that the combination of mitoxantrone and prednisone was superior to prednisone alone in pain control as well as the duration of control. This was associated with an improved quality of life. The combination was well tolerated. There was no difference in overall survival,

but the performance status as well as overall well being had strong associations with survival duration. Another phase III study was the CALGB study 9182.40 The preliminary results indicated PSA decline > 75% in 14% with mitoxantrone plus hydrocortisone vs 7% with hydrocotisone alone. Impact on pain favored the combination therapy; however, the results were not statistically signi®cant. The time to progression was improved. As with the Canadian trial, there was no statistically signi®cant difference in survival. When comparing both trials (Table 2), the most important ®ndings would be that the combination of mitoxantrone plus prednisone is associated with a higher probability of palliation compared with prednisone alone. Duration of palliation is also associated with better quality of life. Moreover, the use of mitoxantrone plus hydrocortisone is associated with improved pain control. There were no differences in survival identi®ed.

Conclusions The unanswered questions in the hormonal therapy of prostate cancer include: How do we minimize side effects? How can we preserve sexual potency? Early vs delayed treatment? The choice of optimal hormonal therapy for carcinoma of the prostate depends not only on the desired progression-free and overall survival, but also on the patient's quality of life, cost of treatment and treatment toxicity. However, the goal with advanced prostate cancer is not cure but rather palliation, especially with hormone refractory disease. Some evidence shows that CAB prolongs progression-free and overall survival. The use of

Figure 4 Mitoxantrone phase III Canadian trial: duration of palliative response. Courtesy of Dr. Vogelzang.

Figure 5 Mitoxantrone phase III Canadian trial: PSA response. Courtesy of Dr Vogelzang.

Figure 6

Table 2 Mitoxantrone phase III Canadian trial and CALGB study 9182: Outcome data compared39,40 courtesy of Dr Vogelzang

Primary Endpoint Secondary Endpoints

Canadian Trial

CALGB Study 9182

Mitoxantrone ‡ Prednisone ‡ Prednisone (n ˆ 80) (n ˆ 81) Palliation 29% 12% Duration of palliation (wks) 33 8 Improved time to progression (24 vs 10 wks.) (P ˆ 0.0001) No difference in survival (48 vs 46 wks.)

Mitoxantrone ‡ Hydrocortisone vs Hydrocortisone alone No survival advantage Improved time to progression (31 vs 17 wks.) (P ˆ 0.065) Mitoxantrone associated with improved pain control Improved obj. Resp. Rate 8.9 vs 1.6% (P ˆ 0.01)

25

Advanced prostate cancer AM Ziada and ED Crawford

26

chemotherapy in advanced HRPC (in which all measures of hormonal manipulation failed) plays an important role in improving quality of life for this patient group.

References 1 Parker SL et al. Cancer Statistics 1997. CA Cancer J Clin 1997; 47: 5 ± 27. 2 Crawford ED, Blumenstein BA. Proposed substages for metastatic prostate cancer. Urology 1997; 50(6): 1027 ± 1028. 3 Huggins C, Hodes CV. Studies of prostatic cancer: effects of castrations, estrogen and androgen injections on serum phosphatases in metastatic carcinoma of prostate. Cancer Res 1941; 1: 293 ± 307. 4 Veterans' Administrative Cooperative Urological Research Group: Treatment and survival of patients with cancer of the prostate. Surg Gynecol Obstet 1967; 124: 1011 ± 1017. 5 Geller J et al. DHT concentrations in human prostatic cancer tissue. J Clin Endocrinol Metab 1978; 46: 440 ± 444. 6 Bracci U. Present procedures in the treatment of prostatic cancer In: Bracci U, diSilverio G (eds). Hormonal therapy in Prostatic Cancer. Publ Coffese: Palermo, 1977, pp 177 ± 192. 7 Labrie F et al. New hormonal therapy with an LHRH agonist and antiandrogen. Clin Invest Med 1982; 5: 267 ± 275. 8 Leuprolide Study Group: Leuprolide vs diethylstilbestrol for metastatic prostate cancer. N Engl J Med 1984; 311: 1281 ± 1286. 9 Delaere KP, VanThilio EL. Flutamide monotherapy as primary treatment in advanced prostatic carcinoma. Semin Oncol 1991; 18: 13 ± 18. 10 Tyrrel CJ. Casodex: a pure non-steroidal antiandrogen used as monotherapy in advanced prostate cancer. Prostate 1992; 4 (Suppl): 97 ± 104. 11 Sogani PC, Vagaiwala MR, Whitmore WF. Experience with ¯utamide in patients with advanced prostate cancer without prior endocrine therapy. Cancer 1984; 54: 744 ± 750. 12 Lund F, Rasmussen. Flutamide vs stilbestrol in the management of advanced prostatic cancer, a controlled prospective study. Br J Urol 1988; 61: 140 ± 142. 13 Wirth M et al. Adjuvant anitandrogenic treatment after radical prostatectomy in stage C prostate cancer ± randomized controlled multicenter trial. Abstract presented at 8th International Prostate Cancer Update, Beaver Creek, January 28 ± February 1, 1998. 14 Kelly WK, Scher HI. Prostate speci®c antigen decline after antiandrogen withdrawal. J Urol 1993; 149: 607 ± 609. 15 Labrie F et al. New approach in the treatment of prostate cancer: complete instead of partial withdrawal of androgens. Prostate 1983; 4: 579 ± 594. 16 Crawford ED et al. A controlled trial of leuprolide with and without ¯utamide in prostatic carcinoma. N Eng J Med 1984; 311: 1291. 17 Waxman J et al. Importance of early tumor exacerbation in patients with long acting analogues of gonadotrophin releasing hormone for advanced prostate cancer. Br Med J 1985; 291: 1387 ± 1388. 18 Labrie F. Endocrine therapy of prostate cancer: Optimal form and timing. J Clin Endocrinol Metab 1995; 80: 1066 ± 1071. 19 Denis LJ et al. Gosereline acetate and ¯utamide vs bilateral orchiectomy: A phase III EORTC trial (30853). Urology 1993; 42: 119 ± 130. 20 Dijkman GA, Janknegt RA, De Reijke TM, Debruyne FM. For the international anandron study group. Long term ef®cacy and safety of nilutamide plus castration in advanced prostate cancer, and the signi®cance of early prostate speci®c antigen normalization. J Urol 1997; 158: 160 ± 163. 21 Bertegna C et al. Ef®cacy of the combination of nilutamide plus orchictomy in patients with metastatic prostate cancer. A metaanalysis of seven randomized double blind trials (1056) patients. Br J Urol 1994; 73: 396 ± 402. 22 Denis L, Murphy JP. Overview of phase III trials on combined androgen treatement in patients with metastatic prostate cancer. Cancer 1993; 72: 3888 ± 3895.

23 Prostate Cancer Trialists' Collaborative Group. Maximum androgen blockade in advanced prostate cancer. An overview of 22 randomized trials with 3283 deaths in 5710 patients. Lancet 1995; 346: 265 ± 269. 24 Klotz LH et al. Intermittent endocrine therapy for advanced prostate cancer. Cancer 1986; 58: 2546 ± 2550. 25 Glave M, Bruchovsky N, Bowden M. Intermittent androgen supression prolongs time to androgen independent proscession in the LNCaP prostate tumor model. J Urol 1994; 151: 457A. 26 Presti JC Jr et al. Multicenter randomized double blind placebo controlled study to investigate the effect of ®nasteride (MK-906) on stage D prostate cancer. J Urol 1992; 148: 1201 ± 1204. 27 Fleshner NE, Trachtenberg J. Combination ®nasteride and ¯utatmide in advanced prostate carcinoma of the prostate: effective therapy with minimal side effects. J Urol 1995; 154: 1642 ± 1646. 28 Fleshner NE, Trachtenberg J. Sequential androgen blockade: a biological study in the inhibition of prostate growth. J Urol 1992; 148: 1928 ± 1931. 29 Crawford ED. Editorial comment. J Urol 1995; 154: 1648. 30 Ziada A, MacKenzie S, Stenner J, Crawford ED. Androgen deprivation therapy using ®nasteride and low dose ¯utamide to treat PSA failure following therapy for clinically localized adenocarcinoma of the prostate. Abstract presented at 8th International Prostate Cancer Update, Beaver Creek, January 28 ± February 1, 1998. 31 Bologna M et al. Antiandrogens and 5a reductase inhibition of the proliferation rate in PC3 and DU145 human prostatic cancer cell lines. Curr Ther Res 1992; 51(6): 799 ± 813. 32 Bologna M et al. Finasteride dose dependently reduces the proliferation rate of the LnCap human prostatic cancer cell line in vitro. Urology. September 1995. 33 Andriole G et al. Treatment with ®nasteride following radical prostatectomy for prostate cancer. Urology 1995; 45: 491 ± 497. 34 Gormley GJ. Chemoprevention strategies for prostate cancer: the role of 5a reductase inhibitors. Workshop Conference on Chemoprevention of Premalignant and Early Malignant Lesions of the Prostate. J Cell Biochem 1992; 16H (Suppl.): 113 ± 118. 35 Byar DP. The Veterans Administration Cooperative Urological Research Group's studies of cancer of the prostate. Cancer 1973; 32: 1126 ± 1130. 36 Byar DP, Corle DK. Hormone therapy for prostate cancer; results of The Veterans Administration Cooperative Urological Research Group studies. Natl Cancer Inst Monogr 1988; 7: 165 ± 170. 37 The Medical Research Council Prostate Cancer Working Party Investigators Group. Immediate vs deferred treatment for advanced prostate cancer. Initial results of the Medical Research Council trial. Br J Urol 1997; 79: 235 ± 246. 38 Isaacs JT. The timing of androgen ablation therapy and=or chemotherapy in the treatment of prostate cancer. Prostate 1984; 5: 1 ± 7. 39 Tannock IF et al. Chemotherapy with mitoxantrone plus prednisone or prednisone alone of symptomatic hormone resistant prostate cancer. A Canadian randomized trial with palliative endpoints. J Clin Oncol 1996; 14: 1756 ± 1764. 40 Kantoff PW et al. Hydrocortisone (HC) with or without mitoxantrone (M) in patients with hormone refractory prostate cancer (HRPC): Preliminary results from a prospective randomized Cancer and Leukemia Group B study (9182) comparing chemotherapy to best supportive care. Proc Am Soc Clin Oncol 1996; 15: 25. 41 Dawson NA et al. A randomized study comparing standard vs moderately high dose megsterol acetate in advanced prostate cancer: A Cancer and Leukemia Group B (9181) study. Proc Amer Soc Clin Oncol 1995; 14: 236. 42 Smith J et al. Effects of liarazole on prostate speci®c antigen levels in patients with relapsed stage D prostate cancer. Proc Am Soc Clin Oncol 1996; 15: 250. 43 Small EJ, Dawson NA, Kantoff PW, Vogelzang NJ. Cancer and Leukemia Group B trials for advanced prostate cancer. Sem in Oncol 1996; 23(6): 28 ± 31.