- Email: [email protected]
Finasteride and flutamide therapy in patients with advanced prostate cancer: response to subsequent castration and long-term follow-up
ADULT UROLOGY
FINASTERIDE AND FLUTAMIDE THERAPY IN PATIENTS WITH ADVANCED PROSTATE CANCER: RESPONSE TO SUBSEQUENT CASTRATION AND LONG-TERM FOLLOW-UP WILLIAM K. OH, JUDITH MANOLA, LAURIE BITTMANN, ADAM BRUFSKY, IRVING D. KAPLAN, MATTHEW R. SMITH, DONALD S. KAUFMAN, AND PHILIP W. KANTOFF
ABSTRACT Objectives. To report the efficacy of castration after progression on finasteride and flutamide. Standard androgen deprivation strategies for prostate cancer typically lead to castrate levels of testosterone. One alternative is the use of finasteride and flutamide. Methods. A Phase II trial evaluated the combination of finasteride (5 mg/day) and flutamide (250 mg three times daily) in patients with rising prostate-specific antigen levels after local treatment for prostate cancer or with newly discovered metastatic disease. Patients were followed up for subsequent events, including castration-free, androgen-independent prostate cancer (AIPC)-free, and overall survival. Results. With a median follow-up of 88 months, 5 patients (25%) continued on finasteride and flutamide, and 12 had stopped this combination and subsequently underwent medical or surgical castration. No patients experienced a flutamide withdrawal effect. All patients experienced more than a 50% decline in prostate-specific antigen after castration (mean 89%). The median protocol treatment failure-free survival was 29.9 months, the median castration-free survival was 37 months, and the median AIPC-free survival was 48.6 months. At 5 years, the overall survival rate was 65% (95% confidence interval 47% to 90%); 29% were alive and have not required castration, and 35% were alive and free of AIPC. Conclusions. Finasteride and flutamide have a durable effect in suppressing prostate-specific antigen progression in some men with advanced prostate cancer. Furthermore, castration induces secondary responses that may be of shorter duration than if started initially, although the overall period of hormonally responsive prostate cancer is more than 4 years. UROLOGY 62: 99–104, 2003. © 2003 Elsevier Inc.
A
ndrogen deprivation therapies (ADTs) such as luteinizing hormone-releasing hormone agonists and bilateral orchiectomy are associated with a growing list of previously unrecognized toxicities. Each ADT induces castrate levels of testosterThis study was supported by a grant from the C. Brendan Noonan, Jr. Charitable Foundation to W.K. Oh. From the Lank Center for Genitourinary Oncology, Department of Medical Oncology, and Department of Biostatistics, Dana-Farber Cancer Institute, Boston, Massachusetts; Division of Hematology/Oncology, Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; Department of Radiation Oncology, Beth Israel Deaconess Medical Center; and Department of Medical Oncology, Massachusetts General Hospital, Boston, Massachusetts Reprint requests: William K. Oh, M.D., Lank Center for Genitourinary Oncology, Department of Medical Oncology, DanaFarber Cancer Institute and Harvard Medical School, 44 Binney Street, Boston, MA 02115 Submitted: November 25, 2002, accepted (with revisions): February 11, 2003 © 2003 ELSEVIER INC. ALL RIGHTS RESERVED
one, causing side effects such as hot flashes, loss of libido, and osteoporosis. As a result, patients are seeking alternative hormonal treatments for advanced prostate cancer,1–3 including intermittent therapy4 and antiandrogen monotherapy.5 Antiandrogens may have less toxicity than ADT, for instance with libido and physical capacity.6 – 8 However, randomized trials suggest that antiandrogen monotherapy is inferior to standard ADT in controlling cancer, especially in patients with metastases.6 –9 Finasteride has minimal activity by itself in the treatment of advanced prostate cancer,10 but combination studies suggest that finasteride and flutamide is active and may preserve sexual function.11–14 In 1994, a Phase II study was initiated of finasteride and flutamide as initial hormonal treatment of patients with newly diagnosed metastatic prostate cancer or a rising prostate-specific antigen (PSA) level after local therapy.11 Results suggested 0090-4295/03/$30.00 doi:10.1016/S0090-4295(03)00145-6 99
that (a) the combination was active, with a mean PSA decline of 94%, (b) finasteride provided additional suppression of dihydrotestosterone and a further diminution of PSA after flutamide alone, and (c) erectile function was intact in 55% of men with normal baseline function. Our purpose was to provide long-term data on the efficacy of finasteride and flutamide, including freedom from castration and androgen-independent prostate cancer. MATERIAL AND METHODS ELIGIBILITY CRITERIA All patients had histologically confirmed adenocarcinoma of the prostate. Evidence of advanced disease was demonstrated by (a) positive bone scan and/or computed tomography of bi-dimensionally measurable lymph nodes or soft-tissue metastases; (b) PSA value of 100 ng/mL or greater; and/or (c) failure after radiotherapy (RT) to the prostate, defined as PSA values of 10 ng/mL or greater 1 year after RT and/or a rising trend in PSA values, as indicated by three abnormal values, with the second 50% greater than the first and the third 50% greater than the second after initiation of RT, and/or a confirmed tripling of PSA values from an originally abnormal value. Patients had Eastern Cooperative Oncology Group performance status of 0 to 2. Patients who had received prior ADT, those with abnormal liver function (bilirubin or serum glutamic-oxaloacetic transaminase levels more than two times normal), and those with untreated spinal cord compression were ineligible. The Institutional Review Board of the DanaFarber Cancer Institute approved the study. All patients provided informed consent.
TABLE I. Patient characteristics Characteristic Age at study entry (yr) Median Range Biopsy Gleason score Median Range Prior therapy RT alone RP ⫹ RT None PSA at study entry (ng/mL) Mean Median Range Disease stage at study entry PSA failure after RT D1 D2
63 42–76 7 5–8 12 (60) 3 (15) 5 (25) 94.6 40.2 12.2–723.4 13 (65) 4 (20) 3 (15)
KEY: RT ⫽ radiotherapy; RP ⫽ radical prostatectomy; PSA ⫽ prostate-specific antigen. Numbers in parentheses are percentages.
STATISTICAL ANALYSIS Summary statistics are provided as the mean, median, and range. The method of Kaplan and Meier was used to estimate survival functions.15 Greenwood estimates of the standard error were used to construct 95% confidence intervals.16
RESULTS TREATMENT PLAN Details have been previously reported. Patients were initially followed up monthly. Treatment was initiated with flutamide (Eulexin, Schering Oncology) 250 mg orally three times daily. PSA was monitored weekly until a first nadir value was obtained, which was defined as the first of three weekly values, for which each value was within 90% of the previous value. Finasteride (Proscar, Merck), 5 mg orally once daily, was then added. PSA was followed up weekly until a second nadir value was reached. Patients continued to take both finasteride and flutamide and were seen every 3 months until progression. Bone and/or computed tomography scans were repeated every 6 months. 11
TREATMENT FAILURE AND SUBSEQUENT THERAPY Treatment failure was defined as the appearance of new lesions on bone or computed tomography scan or a PSA value greater than 150% of the second nadir value noted on two separate determinations 2 weeks apart. At treatment failure, both finasteride and flutamide were discontinued for a period of 2 weeks and a repeat PSA test was obtained. If the PSA value decreased, measurements were repeated every 2 weeks until a nadir value of PSA was achieved, as defined previously. PSA was then measured every month until progression. Patients with an increasing PSA value after withdrawal of flutamide and finasteride were offered luteinizing hormone-releasing hormone agonists or orchiectomy at the discretion of their treating physician. Patients were then followed up every 3 months with examinations and PSA tests. 100
Twenty patients were enrolled between May 1994 and April 1995. Table I summarizes their baseline characteristics. Hormone levels have been previously reported.11 Table II outlines the toxicity of this regimen, which was mild to moderate. The most common side effects were gynecomastia and mild elevations in liver function tests. Table III outlines the long-term sexual function during this trial. Fewer questionnaires were completed at the time patients were removed from the study. Nonetheless, it appeared that patients experienced a continued decline of sexual function, although retaining this remained a high priority. Two patients continue progression free while receiving protocol therapy with finasteride and flutamide. After ending the protocol therapy, 7 patients continued to receive finasteride and flutamide off study. In each case, the patient officially had progression by the stringent PSA criteria of the protocol (ie, a confirmed 50% increase greater than the second nadir value), but none had evidence of clinical progression or significant toxicity from therapy. Thus, each of these 7 patients resumed finasteride and flutamide off study. One patient died of an unrelated cerebrovascular accident while on protocol. Another patient died of unrelated congestive heart UROLOGY 62 (1), 2003
TABLE II. Toxicity by type and maximal grade Grade Toxicity
1
2
3
4
5
Alopecia Anemia Anxiety Arthralgia Bone pain Depression Diarrhea Dry mouth Dyspepsia Dysuria Fatigue Fecal urgency Flatulence Gynecomastia Hematuria Hot flushes Hypercholesterolemia Hyperglycemia Hypertension Infection (without neutropenia) Leukopenia Lightheadedness/dizziness Liver test abnormalities Lymphopenia Memory loss Muscle weakness Nausea Rash Rectal bleeding Secondary malignancy Testicular pain Tinnitus Thrombocytopenia Cerebrovascular event Urinary frequency Urinary incontinence
1 6 0 4 0 4 5 1 2 1 3 1 2 8 0 1 2 0 0 1 2 3 9 5 0 0 1 1 2 0 2 1 4 0 2 2
0 0 2 0 0 1 1 0 3 0 1 0 0 4 2 0 0 1 2 1 0 0 2 0 0 1 0 1 0 0 0 0 0 0 0 0
0 0 0 0 1 0 1 0 0 0 0 0 0 5 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
failure while taking finasteride and flutamide off study. A third patient developed a second primary cancer of the lung. Twelve patients were subsequently treated with castration, 11 with luteinizing hormone-releasing hormone agonist therapy, and 1 with bilateral orchiectomy. No patients experienced a flutamide withdrawal response after discontinuation of finasteride and flutamide. The median PSA level at the time of castration was 27 ng/mL (range 5 to 820). All 12 patients (100%) experienced a 50% or greater PSA decline. The mean PSA decline was 89% (range 62% to 100%). Of the 12 patients, 11 subsequently failed castration and had androgen-independent prostate cancer (AIPC) at last follow-up. The follow-up time was defined as the time from study entry to the date the patient was last known UROLOGY 62 (1), 2003
to be alive. Patients who died were censored at the date of death. The median follow-up was 88 months. The survival time was defined as the time from study entry to death. Patients who were still alive were censored at the date last known to be alive. Eight patients have died. Two patients died of progressive prostate cancer, two of second primary cancers, three of cardiovascular events, and one of unknown causes. The median survival time had not yet been reached. The 5-year survival rate was 65% (95% confidence interval 47% to 90%). The protocol treatment failure-free survival was defined as the time from study entry until the time off study for progression or until death from any cause. Patients who were alive without progression on protocol therapy were censored at the last date they were known to be progression free. The median protocol treatment failure-free survival was 29.9 months (95% confidence interval 23.8 to 60.9 months). The castration-free survival was defined as the time from study entry until the beginning of medical or surgical castration or until death from any cause. Patients alive without castration were censored at the date they were last known to be stable receiving current therapy. The median castration-free survival was 37 months. The lower limit of the 95% confidence interval was 29.5 months, and the upper limit was not estimable. At 5 years, 29% of patients were alive and had not received castration (95% confidence interval 15% to 58%). AIPC-free survival was defined as the time from study entry to the time when a secondary hormonal treatment was added to androgen-deprivation therapy because of progression. Patients who were alive without failing castration were censored at the time they were last known to be progression free. Patients who died of any cause (including second primary cancer) or who failed castration were considered events at the date that any additional therapy was added to their primary ADT or the date of death, whichever came first. Using this definition, 6 patients were alive and free of AIPC. The median AIPC-free survival was 48.6 months. The lower limit on the 95% confidence interval was 41.8 months, and the upper limit was not estimable. At 5 years, 35% of patients were alive and free of AIPC (95% confidence interval 19% to 64%). Figure 1 illustrates the Kaplan-Meier plot of protocol treatment failure-free, castration-free, AIPCfree, and overall survival. COMMENT Patients are being treated with ADTs earlier in the course of their disease, often in the setting of a rising PSA alone. As a result, patients are seeking alternative means of treating recurrent prostate 101
TABLE III. Sexual function questionnaire Baseline (n) Questionnaires completed Erections past 4 wk No Yes Trouble achieving erection? No Yes No intercourse in past 4 wk Trouble achieving erection during intercourse? No Yes No intercourse in past 4 wk Missing Lack sexual interest? NA or missing Little or no problem Somewhat or very much a problem Unable to relax and enjoy sex? NA or missing Little or no problem Somewhat or very much a problem Difficulty becoming sexually aroused? NA or missing Little or no problem Somewhat or very much a problem Difficulty having an orgasm? NA or missing Little or no problem Somewhat or very much a problem Difficulty getting an erection? NA or missing Little or no problem Somewhat or very much a problem How much do you care about having an active sex life? A lot Some A little Not at all Missing
Second Nadir (n)
Off Study (n)
20
19
9
3 (15) 17 (85)
5 (26) 14 (74)
4 (44) 5 (56)
11 (55) 5 (25) 4 (20)
7 (37) 7 (37) 5 (26)
1 (11) 5 (56) 3 (33)
11 (58) 5 (26) 3 (16) 1
10 (53) 4 (21) 5 (26) 0
2 (22) 4 (44) 3 (33) 0
2 (10) 15 (79) 3 (16)
1 (6) 15 (83) 2 (11)
2 (22) 4 (44) 3 (33)
2 (10) 16 (80) 2 (10)
4 (23) 11 (61) 4 (22)
4 (44) 1 (11) 4 (44)
0 17 (85) 3 (15)
2 (12) 12 (66) 5 (28)
1 (11) 5 (55) 3 (33)
1 (5) 15 (75) 4 (20)
3 (17) 9 (50) 7 (39)
1 (11) 2 (22) 6 (66)
2 (10) 11 (55) 7 (35)
1 (6) 10 (56) 8 (45)
2 (22) 2 (22) 5 (55)
15 (75) 4 (20) 1 (5) 0
13 (72) 4 (22) 1 (6) 0 1
5 (56) 3 (33) 0 1 (11)
Numbers in parentheses are percentages.
cancer that minimizes exposure to the untoward side effects of castration. This is increasing as the toxicity of long-term ADT is better appreciated. Long-term follow-up of patients treated with finasteride and flutamide has demonstrated that this combination may delay the initiation of castration. Remarkably, 25% of patients remained free of castration 7 years after the start of the study. The overall survival in this cohort is encouraging, with a 5-year survival rate of 65%, although this may reflect the fact that two thirds of the patients enrolled had no overt evidence of metastases. Toxicity was also modest in this trial, with gynecomastia and 102
liver test abnormalities most common. In addition, sexual function appeared to worsen as patients continued treatment. The median time before protocol treatment failed was 30 months, but another 7 months elapsed before patients received ADT. The median time before AIPC developed was another 12 months, suggesting that castration was less effective in the salvage setting than if used initially. Nonetheless, that the median time to AIPC was more than 4 years suggests that the use of finasteride plus flutamide did not compromise the overall duration of “hormonal control” of prostate cancer. UROLOGY 62 (1), 2003
FIGURE 1. Protocol treatment failure-free, castrationfree, AIPC-free, and overall survival.
It is notable that no patients experienced a flutamide withdrawal response. First described 10 years ago, such responses were seen after discontinuation of flutamide in combination with castration.17 The mechanism of this effect is unclear, although mutations in the androgen receptor gene may lead to paradoxical stimulation of growth by antiandrogens.18 Discontinuation of an antiandrogen may induce cancer regression. The presence of testosterone in patients treated with antiandrogens alone likely removes a selective pressure for prostate cancer cells expressing mutant androgen receptors, although it is also possible that simultaneous withdrawal of finasteride is confounding an antiandrogen withdrawal effect. Few studies have evaluated the subsequent response to castration after progression during antiandrogen monotherapy or the combination of finasteride and flutamide. Three studies evaluated the response to secondary castration in patients treated with bicalutamide or finasteride plus flutamide.19 –21 Kasimis et al.19 treated 54 patients with metastatic prostate cancer with bicalutamide 50 mg/day as initial hormonal therapy. The time to treatment failure was only 47 weeks. However, after castration, patients survived another 71 weeks, so survival after both therapies was 2.3 years. The investigators suggested that this interval is comparable to historical survival after castration alone, and postulate whether sequential therapy with antiandrogens followed by castration results in a similar survival to castration alone. Tan et al.20 described 28 patients treated with bilateral orchiectomy after failure of antiandrogen monotherapy and reported a 50% or greater PSA decline in 40% of all patients and 60% (9 of 15 patients) of those previously taking flutamide. Ornstein et al.21 described 18 patients, 10 of whom received finasteride alone and 8 finasteride plus flutamide. Because finasteride alone has only UROLOGY 62 (1), 2003
minor effects on PSA progression,10 it is not surprising that there was a high response rate to castration. However, 6 of 8 patients treated with finasteride and flutamide had undetectable PSA levels after castration, with a median duration of more than 20 months. This suggests that secondary castration after finasteride and flutamide can suppress PSA for an extended duration. Antiandrogen monotherapy is associated with inferior survival compared with castration in metastatic disease.6,8,9,22 However, these trials did not require secondary castration in the antiandrogen monotherapy arm. For instance, in a trial of flutamide versus diethylstilbestrol, Chang et al.9 reported superior survival in the diethylstilbestrol arm (43.2 versus 28.5 months, P ⫽ 0.04). However, only 47% of patients treated with flutamide underwent salvage orchiectomy. Three trials comparing bicalutamide (50 to 150 mg/day) to castration similarly recommended secondary therapy with castration in the antiandrogen arm, but none mandated subsequent castration.6,8,22 It is, therefore, possible that in patients with metastatic disease, secondary castration might improve the survival of patients treated with antiandrogens alone. In patients with nonmetastatic disease, survival with antiandrogen monotherapy does not appear to be different than with castration, but the added value of secondary castration in these patients has not been assessed.7,8 The ability to generalize these results to a larger population is limited by the small number of patients and variability in treatments used after discontinuing protocol therapy. Also, selection bias is an issue when interpreting progression-free and overall survival data in a nonrandomized trial. Finally, defining more uniform populations would have made these data more valuable. Nonetheless, we would argue that these data are valuable for two reasons: the long duration of follow-up and a dearth of data on salvage response to androgen deprivation after antiandrogen treatment. The Cancer and Leukemia Group B recently completed a multicenter Phase II trial of finasteride and flutamide in men with a rising PSA level after prostate surgery or RT without metastatic disease (CALGB 9782). Considering the possible quality-of-life benefit to patients without metastases, delaying castration may represent a reasonable and safe strategy. CONCLUSIONS Finasteride and flutamide as initial hormonal therapy is associated with a significant response duration in men with advanced prostate cancer. Castration appears to have a shorter duration of response than if started initially, but the overall 103
period of hormonally sensitive prostate cancer is more than 4 years. A randomized trial comparing this combination to standard ADT is needed, but castration should be required in all patients at the time of progression taking finasteride and flutamide. REFERENCES 1. Basaria S, Lieb J, Tang AM, et al: Long-term effects of androgen deprivation therapy in prostate cancer patients. Clin Endocrinol 56: 779 –786, 2002. 2. Berruti A, Dogliotti L, Terrone C, et al: Changes in bone mineral density, lean body mass and fat content as measured by dual energy x-ray absorptiometry in patients with prostate cancer without apparent bone metastases given androgen deprivation therapy. J Urol 167: 2361–2367, 2002. 3. Ross RW, and Small EJ: Osteoporosis in men treated with androgen deprivation therapy for prostate cancer. J Urol 167: 1952–1956, 2002. 4. Grossfeld GD, Chaudhary UB, Reese DM, et al: Intermittent androgen deprivation: update of cycling characteristics in patients without clinically apparent metastatic prostate cancer. Urology 58: 240 –245, 2001. 5. Kolvenbag GJ, Iversen P, and Newling DW: Antiandrogen monotherapy: a new form of treatment for patients with prostate cancer. Urology 58: 16 –23, 2001. 6. Bales GT, and Chodak GW: A controlled trial of bicalutamide versus castration in patients with advanced prostate cancer. Urology 47: 38 – 43, discussion 48 –53, 1996. 7. Iversen P, Tyrrell CJ, Kaisary AV, et al: Bicalutamide monotherapy compared with castration in patients with nonmetastatic locally advanced prostate cancer: 6.3 years of followup. J Urol 164: 1579 –1582, 2000. 8. Boccardo F, Rubagotti A, Barichello M, et al: Bicalutamide monotherapy versus flutamide plus goserelin in prostate cancer patients: results of an Italian Prostate Cancer Project study. J Clin Oncol 17: 2027–2038, 1999. 9. Chang A, Yeap B, Davis T, et al: Double-blind, randomized study of primary hormonal treatment of stage D2 prostate carcinoma: flutamide versus diethylstilbestrol. J Clin Oncol 14: 2250 –2257, 1996.
104
10. Presti JC Jr, Fair WR, Andriole G, et al: Multicenter, randomized, double-blind, placebo controlled study to investigate the effect of finasteride (MK-906) on stage D prostate cancer. J Urol 148: 1201–1204, 1992. 11. Brufsky A, Fontaine-Rothe P, Berlane K, et al: Finasteride and flutamide as potency-sparing androgen-ablative therapy for advanced adenocarcinoma of the prostate. Urology 49: 913–920, 1997. 12. Fleshner NE, and Fair WR: Anti-androgenic effects of combination finasteride plus flutamide in patients with prostatic carcinoma. Br J Urol 78: 907–910, 1996. 13. Fleshner NE, and Trachtenberg J: Combination finasteride and flutamide in advanced carcinoma of the prostate: effective therapy with minimal side effects. J Urol 154: 1642– 1646, 1995. 14. Ornstein DK, Rao GS, Johnson B, et al: Combined finasteride and flutamide therapy in men with advanced prostate cancer. Urology 48: 901–905, 1996. 15. Kaplan EL, and Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457–481, 1958. 16. Collett D: Modelling Survival Data in Medical Research. London, Chapman and Hall, 1994. 17. Kelly WK: Endocrine withdrawal syndrome and its relevance to the management of hormone refractory prostate cancer. Eur Urol 34(suppl 3): 18 –23, 1998. 18. Culig Z, Klocker H, Bartsch G, et al: Androgen receptor mutations in carcinoma of the prostate: significance for endocrine therapy. Am J Pharmacogenom 1: 241–249, 2001. 19. Kasimis B, Wilding G, Kreis W, et al: Survival of patients who had salvage castration after failure on bicalutamide monotherapy for stage (D2) prostate cancer. Cancer Invest 18: 602–608, 2000. 20. Tan A, Tuckey J, and Rice M: Orchidectomy following failure of antiandrogen monotherapy in patients with metastatic prostate cancer. Eur Urol 40: 130 –134, 2001. 21. Ornstein DK, Smith DS, and Andriole GL: Biochemical response to testicular androgen ablation among patients with prostate cancer for whom flutamide and/or finasteride therapy failed. Urology 52: 1094 –1097, 1998. 22. Tyrrell CJ, Kaisary AV, Iversen P, et al: A randomised comparison of “Casodex” (bicalutamide) 150 mg monotherapy versus castration in the treatment of metastatic and locally advanced prostate cancer. Eur Urol 33: 447–456, 1998.
UROLOGY 62 (1), 2003
FINASTERIDE AND FLUTAMIDE THERAPY IN PATIENTS WITH ADVANCED PROSTATE CANCER: RESPONSE TO SUBSEQUENT CASTRATION AND LONG-TERM FOLLOW-UP WILLIAM K. OH, JUDITH MANOLA, LAURIE BITTMANN, ADAM BRUFSKY, IRVING D. KAPLAN, MATTHEW R. SMITH, DONALD S. KAUFMAN, AND PHILIP W. KANTOFF
ABSTRACT Objectives. To report the efficacy of castration after progression on finasteride and flutamide. Standard androgen deprivation strategies for prostate cancer typically lead to castrate levels of testosterone. One alternative is the use of finasteride and flutamide. Methods. A Phase II trial evaluated the combination of finasteride (5 mg/day) and flutamide (250 mg three times daily) in patients with rising prostate-specific antigen levels after local treatment for prostate cancer or with newly discovered metastatic disease. Patients were followed up for subsequent events, including castration-free, androgen-independent prostate cancer (AIPC)-free, and overall survival. Results. With a median follow-up of 88 months, 5 patients (25%) continued on finasteride and flutamide, and 12 had stopped this combination and subsequently underwent medical or surgical castration. No patients experienced a flutamide withdrawal effect. All patients experienced more than a 50% decline in prostate-specific antigen after castration (mean 89%). The median protocol treatment failure-free survival was 29.9 months, the median castration-free survival was 37 months, and the median AIPC-free survival was 48.6 months. At 5 years, the overall survival rate was 65% (95% confidence interval 47% to 90%); 29% were alive and have not required castration, and 35% were alive and free of AIPC. Conclusions. Finasteride and flutamide have a durable effect in suppressing prostate-specific antigen progression in some men with advanced prostate cancer. Furthermore, castration induces secondary responses that may be of shorter duration than if started initially, although the overall period of hormonally responsive prostate cancer is more than 4 years. UROLOGY 62: 99–104, 2003. © 2003 Elsevier Inc.
A
ndrogen deprivation therapies (ADTs) such as luteinizing hormone-releasing hormone agonists and bilateral orchiectomy are associated with a growing list of previously unrecognized toxicities. Each ADT induces castrate levels of testosterThis study was supported by a grant from the C. Brendan Noonan, Jr. Charitable Foundation to W.K. Oh. From the Lank Center for Genitourinary Oncology, Department of Medical Oncology, and Department of Biostatistics, Dana-Farber Cancer Institute, Boston, Massachusetts; Division of Hematology/Oncology, Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; Department of Radiation Oncology, Beth Israel Deaconess Medical Center; and Department of Medical Oncology, Massachusetts General Hospital, Boston, Massachusetts Reprint requests: William K. Oh, M.D., Lank Center for Genitourinary Oncology, Department of Medical Oncology, DanaFarber Cancer Institute and Harvard Medical School, 44 Binney Street, Boston, MA 02115 Submitted: November 25, 2002, accepted (with revisions): February 11, 2003 © 2003 ELSEVIER INC. ALL RIGHTS RESERVED
one, causing side effects such as hot flashes, loss of libido, and osteoporosis. As a result, patients are seeking alternative hormonal treatments for advanced prostate cancer,1–3 including intermittent therapy4 and antiandrogen monotherapy.5 Antiandrogens may have less toxicity than ADT, for instance with libido and physical capacity.6 – 8 However, randomized trials suggest that antiandrogen monotherapy is inferior to standard ADT in controlling cancer, especially in patients with metastases.6 –9 Finasteride has minimal activity by itself in the treatment of advanced prostate cancer,10 but combination studies suggest that finasteride and flutamide is active and may preserve sexual function.11–14 In 1994, a Phase II study was initiated of finasteride and flutamide as initial hormonal treatment of patients with newly diagnosed metastatic prostate cancer or a rising prostate-specific antigen (PSA) level after local therapy.11 Results suggested 0090-4295/03/$30.00 doi:10.1016/S0090-4295(03)00145-6 99
that (a) the combination was active, with a mean PSA decline of 94%, (b) finasteride provided additional suppression of dihydrotestosterone and a further diminution of PSA after flutamide alone, and (c) erectile function was intact in 55% of men with normal baseline function. Our purpose was to provide long-term data on the efficacy of finasteride and flutamide, including freedom from castration and androgen-independent prostate cancer. MATERIAL AND METHODS ELIGIBILITY CRITERIA All patients had histologically confirmed adenocarcinoma of the prostate. Evidence of advanced disease was demonstrated by (a) positive bone scan and/or computed tomography of bi-dimensionally measurable lymph nodes or soft-tissue metastases; (b) PSA value of 100 ng/mL or greater; and/or (c) failure after radiotherapy (RT) to the prostate, defined as PSA values of 10 ng/mL or greater 1 year after RT and/or a rising trend in PSA values, as indicated by three abnormal values, with the second 50% greater than the first and the third 50% greater than the second after initiation of RT, and/or a confirmed tripling of PSA values from an originally abnormal value. Patients had Eastern Cooperative Oncology Group performance status of 0 to 2. Patients who had received prior ADT, those with abnormal liver function (bilirubin or serum glutamic-oxaloacetic transaminase levels more than two times normal), and those with untreated spinal cord compression were ineligible. The Institutional Review Board of the DanaFarber Cancer Institute approved the study. All patients provided informed consent.
TABLE I. Patient characteristics Characteristic Age at study entry (yr) Median Range Biopsy Gleason score Median Range Prior therapy RT alone RP ⫹ RT None PSA at study entry (ng/mL) Mean Median Range Disease stage at study entry PSA failure after RT D1 D2
63 42–76 7 5–8 12 (60) 3 (15) 5 (25) 94.6 40.2 12.2–723.4 13 (65) 4 (20) 3 (15)
KEY: RT ⫽ radiotherapy; RP ⫽ radical prostatectomy; PSA ⫽ prostate-specific antigen. Numbers in parentheses are percentages.
STATISTICAL ANALYSIS Summary statistics are provided as the mean, median, and range. The method of Kaplan and Meier was used to estimate survival functions.15 Greenwood estimates of the standard error were used to construct 95% confidence intervals.16
RESULTS TREATMENT PLAN Details have been previously reported. Patients were initially followed up monthly. Treatment was initiated with flutamide (Eulexin, Schering Oncology) 250 mg orally three times daily. PSA was monitored weekly until a first nadir value was obtained, which was defined as the first of three weekly values, for which each value was within 90% of the previous value. Finasteride (Proscar, Merck), 5 mg orally once daily, was then added. PSA was followed up weekly until a second nadir value was reached. Patients continued to take both finasteride and flutamide and were seen every 3 months until progression. Bone and/or computed tomography scans were repeated every 6 months. 11
TREATMENT FAILURE AND SUBSEQUENT THERAPY Treatment failure was defined as the appearance of new lesions on bone or computed tomography scan or a PSA value greater than 150% of the second nadir value noted on two separate determinations 2 weeks apart. At treatment failure, both finasteride and flutamide were discontinued for a period of 2 weeks and a repeat PSA test was obtained. If the PSA value decreased, measurements were repeated every 2 weeks until a nadir value of PSA was achieved, as defined previously. PSA was then measured every month until progression. Patients with an increasing PSA value after withdrawal of flutamide and finasteride were offered luteinizing hormone-releasing hormone agonists or orchiectomy at the discretion of their treating physician. Patients were then followed up every 3 months with examinations and PSA tests. 100
Twenty patients were enrolled between May 1994 and April 1995. Table I summarizes their baseline characteristics. Hormone levels have been previously reported.11 Table II outlines the toxicity of this regimen, which was mild to moderate. The most common side effects were gynecomastia and mild elevations in liver function tests. Table III outlines the long-term sexual function during this trial. Fewer questionnaires were completed at the time patients were removed from the study. Nonetheless, it appeared that patients experienced a continued decline of sexual function, although retaining this remained a high priority. Two patients continue progression free while receiving protocol therapy with finasteride and flutamide. After ending the protocol therapy, 7 patients continued to receive finasteride and flutamide off study. In each case, the patient officially had progression by the stringent PSA criteria of the protocol (ie, a confirmed 50% increase greater than the second nadir value), but none had evidence of clinical progression or significant toxicity from therapy. Thus, each of these 7 patients resumed finasteride and flutamide off study. One patient died of an unrelated cerebrovascular accident while on protocol. Another patient died of unrelated congestive heart UROLOGY 62 (1), 2003
TABLE II. Toxicity by type and maximal grade Grade Toxicity
1
2
3
4
5
Alopecia Anemia Anxiety Arthralgia Bone pain Depression Diarrhea Dry mouth Dyspepsia Dysuria Fatigue Fecal urgency Flatulence Gynecomastia Hematuria Hot flushes Hypercholesterolemia Hyperglycemia Hypertension Infection (without neutropenia) Leukopenia Lightheadedness/dizziness Liver test abnormalities Lymphopenia Memory loss Muscle weakness Nausea Rash Rectal bleeding Secondary malignancy Testicular pain Tinnitus Thrombocytopenia Cerebrovascular event Urinary frequency Urinary incontinence
1 6 0 4 0 4 5 1 2 1 3 1 2 8 0 1 2 0 0 1 2 3 9 5 0 0 1 1 2 0 2 1 4 0 2 2
0 0 2 0 0 1 1 0 3 0 1 0 0 4 2 0 0 1 2 1 0 0 2 0 0 1 0 1 0 0 0 0 0 0 0 0
0 0 0 0 1 0 1 0 0 0 0 0 0 5 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
failure while taking finasteride and flutamide off study. A third patient developed a second primary cancer of the lung. Twelve patients were subsequently treated with castration, 11 with luteinizing hormone-releasing hormone agonist therapy, and 1 with bilateral orchiectomy. No patients experienced a flutamide withdrawal response after discontinuation of finasteride and flutamide. The median PSA level at the time of castration was 27 ng/mL (range 5 to 820). All 12 patients (100%) experienced a 50% or greater PSA decline. The mean PSA decline was 89% (range 62% to 100%). Of the 12 patients, 11 subsequently failed castration and had androgen-independent prostate cancer (AIPC) at last follow-up. The follow-up time was defined as the time from study entry to the date the patient was last known UROLOGY 62 (1), 2003
to be alive. Patients who died were censored at the date of death. The median follow-up was 88 months. The survival time was defined as the time from study entry to death. Patients who were still alive were censored at the date last known to be alive. Eight patients have died. Two patients died of progressive prostate cancer, two of second primary cancers, three of cardiovascular events, and one of unknown causes. The median survival time had not yet been reached. The 5-year survival rate was 65% (95% confidence interval 47% to 90%). The protocol treatment failure-free survival was defined as the time from study entry until the time off study for progression or until death from any cause. Patients who were alive without progression on protocol therapy were censored at the last date they were known to be progression free. The median protocol treatment failure-free survival was 29.9 months (95% confidence interval 23.8 to 60.9 months). The castration-free survival was defined as the time from study entry until the beginning of medical or surgical castration or until death from any cause. Patients alive without castration were censored at the date they were last known to be stable receiving current therapy. The median castration-free survival was 37 months. The lower limit of the 95% confidence interval was 29.5 months, and the upper limit was not estimable. At 5 years, 29% of patients were alive and had not received castration (95% confidence interval 15% to 58%). AIPC-free survival was defined as the time from study entry to the time when a secondary hormonal treatment was added to androgen-deprivation therapy because of progression. Patients who were alive without failing castration were censored at the time they were last known to be progression free. Patients who died of any cause (including second primary cancer) or who failed castration were considered events at the date that any additional therapy was added to their primary ADT or the date of death, whichever came first. Using this definition, 6 patients were alive and free of AIPC. The median AIPC-free survival was 48.6 months. The lower limit on the 95% confidence interval was 41.8 months, and the upper limit was not estimable. At 5 years, 35% of patients were alive and free of AIPC (95% confidence interval 19% to 64%). Figure 1 illustrates the Kaplan-Meier plot of protocol treatment failure-free, castration-free, AIPCfree, and overall survival. COMMENT Patients are being treated with ADTs earlier in the course of their disease, often in the setting of a rising PSA alone. As a result, patients are seeking alternative means of treating recurrent prostate 101
TABLE III. Sexual function questionnaire Baseline (n) Questionnaires completed Erections past 4 wk No Yes Trouble achieving erection? No Yes No intercourse in past 4 wk Trouble achieving erection during intercourse? No Yes No intercourse in past 4 wk Missing Lack sexual interest? NA or missing Little or no problem Somewhat or very much a problem Unable to relax and enjoy sex? NA or missing Little or no problem Somewhat or very much a problem Difficulty becoming sexually aroused? NA or missing Little or no problem Somewhat or very much a problem Difficulty having an orgasm? NA or missing Little or no problem Somewhat or very much a problem Difficulty getting an erection? NA or missing Little or no problem Somewhat or very much a problem How much do you care about having an active sex life? A lot Some A little Not at all Missing
Second Nadir (n)
Off Study (n)
20
19
9
3 (15) 17 (85)
5 (26) 14 (74)
4 (44) 5 (56)
11 (55) 5 (25) 4 (20)
7 (37) 7 (37) 5 (26)
1 (11) 5 (56) 3 (33)
11 (58) 5 (26) 3 (16) 1
10 (53) 4 (21) 5 (26) 0
2 (22) 4 (44) 3 (33) 0
2 (10) 15 (79) 3 (16)
1 (6) 15 (83) 2 (11)
2 (22) 4 (44) 3 (33)
2 (10) 16 (80) 2 (10)
4 (23) 11 (61) 4 (22)
4 (44) 1 (11) 4 (44)
0 17 (85) 3 (15)
2 (12) 12 (66) 5 (28)
1 (11) 5 (55) 3 (33)
1 (5) 15 (75) 4 (20)
3 (17) 9 (50) 7 (39)
1 (11) 2 (22) 6 (66)
2 (10) 11 (55) 7 (35)
1 (6) 10 (56) 8 (45)
2 (22) 2 (22) 5 (55)
15 (75) 4 (20) 1 (5) 0
13 (72) 4 (22) 1 (6) 0 1
5 (56) 3 (33) 0 1 (11)
Numbers in parentheses are percentages.
cancer that minimizes exposure to the untoward side effects of castration. This is increasing as the toxicity of long-term ADT is better appreciated. Long-term follow-up of patients treated with finasteride and flutamide has demonstrated that this combination may delay the initiation of castration. Remarkably, 25% of patients remained free of castration 7 years after the start of the study. The overall survival in this cohort is encouraging, with a 5-year survival rate of 65%, although this may reflect the fact that two thirds of the patients enrolled had no overt evidence of metastases. Toxicity was also modest in this trial, with gynecomastia and 102
liver test abnormalities most common. In addition, sexual function appeared to worsen as patients continued treatment. The median time before protocol treatment failed was 30 months, but another 7 months elapsed before patients received ADT. The median time before AIPC developed was another 12 months, suggesting that castration was less effective in the salvage setting than if used initially. Nonetheless, that the median time to AIPC was more than 4 years suggests that the use of finasteride plus flutamide did not compromise the overall duration of “hormonal control” of prostate cancer. UROLOGY 62 (1), 2003
FIGURE 1. Protocol treatment failure-free, castrationfree, AIPC-free, and overall survival.
It is notable that no patients experienced a flutamide withdrawal response. First described 10 years ago, such responses were seen after discontinuation of flutamide in combination with castration.17 The mechanism of this effect is unclear, although mutations in the androgen receptor gene may lead to paradoxical stimulation of growth by antiandrogens.18 Discontinuation of an antiandrogen may induce cancer regression. The presence of testosterone in patients treated with antiandrogens alone likely removes a selective pressure for prostate cancer cells expressing mutant androgen receptors, although it is also possible that simultaneous withdrawal of finasteride is confounding an antiandrogen withdrawal effect. Few studies have evaluated the subsequent response to castration after progression during antiandrogen monotherapy or the combination of finasteride and flutamide. Three studies evaluated the response to secondary castration in patients treated with bicalutamide or finasteride plus flutamide.19 –21 Kasimis et al.19 treated 54 patients with metastatic prostate cancer with bicalutamide 50 mg/day as initial hormonal therapy. The time to treatment failure was only 47 weeks. However, after castration, patients survived another 71 weeks, so survival after both therapies was 2.3 years. The investigators suggested that this interval is comparable to historical survival after castration alone, and postulate whether sequential therapy with antiandrogens followed by castration results in a similar survival to castration alone. Tan et al.20 described 28 patients treated with bilateral orchiectomy after failure of antiandrogen monotherapy and reported a 50% or greater PSA decline in 40% of all patients and 60% (9 of 15 patients) of those previously taking flutamide. Ornstein et al.21 described 18 patients, 10 of whom received finasteride alone and 8 finasteride plus flutamide. Because finasteride alone has only UROLOGY 62 (1), 2003
minor effects on PSA progression,10 it is not surprising that there was a high response rate to castration. However, 6 of 8 patients treated with finasteride and flutamide had undetectable PSA levels after castration, with a median duration of more than 20 months. This suggests that secondary castration after finasteride and flutamide can suppress PSA for an extended duration. Antiandrogen monotherapy is associated with inferior survival compared with castration in metastatic disease.6,8,9,22 However, these trials did not require secondary castration in the antiandrogen monotherapy arm. For instance, in a trial of flutamide versus diethylstilbestrol, Chang et al.9 reported superior survival in the diethylstilbestrol arm (43.2 versus 28.5 months, P ⫽ 0.04). However, only 47% of patients treated with flutamide underwent salvage orchiectomy. Three trials comparing bicalutamide (50 to 150 mg/day) to castration similarly recommended secondary therapy with castration in the antiandrogen arm, but none mandated subsequent castration.6,8,22 It is, therefore, possible that in patients with metastatic disease, secondary castration might improve the survival of patients treated with antiandrogens alone. In patients with nonmetastatic disease, survival with antiandrogen monotherapy does not appear to be different than with castration, but the added value of secondary castration in these patients has not been assessed.7,8 The ability to generalize these results to a larger population is limited by the small number of patients and variability in treatments used after discontinuing protocol therapy. Also, selection bias is an issue when interpreting progression-free and overall survival data in a nonrandomized trial. Finally, defining more uniform populations would have made these data more valuable. Nonetheless, we would argue that these data are valuable for two reasons: the long duration of follow-up and a dearth of data on salvage response to androgen deprivation after antiandrogen treatment. The Cancer and Leukemia Group B recently completed a multicenter Phase II trial of finasteride and flutamide in men with a rising PSA level after prostate surgery or RT without metastatic disease (CALGB 9782). Considering the possible quality-of-life benefit to patients without metastases, delaying castration may represent a reasonable and safe strategy. CONCLUSIONS Finasteride and flutamide as initial hormonal therapy is associated with a significant response duration in men with advanced prostate cancer. Castration appears to have a shorter duration of response than if started initially, but the overall 103
period of hormonally sensitive prostate cancer is more than 4 years. A randomized trial comparing this combination to standard ADT is needed, but castration should be required in all patients at the time of progression taking finasteride and flutamide. REFERENCES 1. Basaria S, Lieb J, Tang AM, et al: Long-term effects of androgen deprivation therapy in prostate cancer patients. Clin Endocrinol 56: 779 –786, 2002. 2. Berruti A, Dogliotti L, Terrone C, et al: Changes in bone mineral density, lean body mass and fat content as measured by dual energy x-ray absorptiometry in patients with prostate cancer without apparent bone metastases given androgen deprivation therapy. J Urol 167: 2361–2367, 2002. 3. Ross RW, and Small EJ: Osteoporosis in men treated with androgen deprivation therapy for prostate cancer. J Urol 167: 1952–1956, 2002. 4. Grossfeld GD, Chaudhary UB, Reese DM, et al: Intermittent androgen deprivation: update of cycling characteristics in patients without clinically apparent metastatic prostate cancer. Urology 58: 240 –245, 2001. 5. Kolvenbag GJ, Iversen P, and Newling DW: Antiandrogen monotherapy: a new form of treatment for patients with prostate cancer. Urology 58: 16 –23, 2001. 6. Bales GT, and Chodak GW: A controlled trial of bicalutamide versus castration in patients with advanced prostate cancer. Urology 47: 38 – 43, discussion 48 –53, 1996. 7. Iversen P, Tyrrell CJ, Kaisary AV, et al: Bicalutamide monotherapy compared with castration in patients with nonmetastatic locally advanced prostate cancer: 6.3 years of followup. J Urol 164: 1579 –1582, 2000. 8. Boccardo F, Rubagotti A, Barichello M, et al: Bicalutamide monotherapy versus flutamide plus goserelin in prostate cancer patients: results of an Italian Prostate Cancer Project study. J Clin Oncol 17: 2027–2038, 1999. 9. Chang A, Yeap B, Davis T, et al: Double-blind, randomized study of primary hormonal treatment of stage D2 prostate carcinoma: flutamide versus diethylstilbestrol. J Clin Oncol 14: 2250 –2257, 1996.
104
10. Presti JC Jr, Fair WR, Andriole G, et al: Multicenter, randomized, double-blind, placebo controlled study to investigate the effect of finasteride (MK-906) on stage D prostate cancer. J Urol 148: 1201–1204, 1992. 11. Brufsky A, Fontaine-Rothe P, Berlane K, et al: Finasteride and flutamide as potency-sparing androgen-ablative therapy for advanced adenocarcinoma of the prostate. Urology 49: 913–920, 1997. 12. Fleshner NE, and Fair WR: Anti-androgenic effects of combination finasteride plus flutamide in patients with prostatic carcinoma. Br J Urol 78: 907–910, 1996. 13. Fleshner NE, and Trachtenberg J: Combination finasteride and flutamide in advanced carcinoma of the prostate: effective therapy with minimal side effects. J Urol 154: 1642– 1646, 1995. 14. Ornstein DK, Rao GS, Johnson B, et al: Combined finasteride and flutamide therapy in men with advanced prostate cancer. Urology 48: 901–905, 1996. 15. Kaplan EL, and Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457–481, 1958. 16. Collett D: Modelling Survival Data in Medical Research. London, Chapman and Hall, 1994. 17. Kelly WK: Endocrine withdrawal syndrome and its relevance to the management of hormone refractory prostate cancer. Eur Urol 34(suppl 3): 18 –23, 1998. 18. Culig Z, Klocker H, Bartsch G, et al: Androgen receptor mutations in carcinoma of the prostate: significance for endocrine therapy. Am J Pharmacogenom 1: 241–249, 2001. 19. Kasimis B, Wilding G, Kreis W, et al: Survival of patients who had salvage castration after failure on bicalutamide monotherapy for stage (D2) prostate cancer. Cancer Invest 18: 602–608, 2000. 20. Tan A, Tuckey J, and Rice M: Orchidectomy following failure of antiandrogen monotherapy in patients with metastatic prostate cancer. Eur Urol 40: 130 –134, 2001. 21. Ornstein DK, Smith DS, and Andriole GL: Biochemical response to testicular androgen ablation among patients with prostate cancer for whom flutamide and/or finasteride therapy failed. Urology 52: 1094 –1097, 1998. 22. Tyrrell CJ, Kaisary AV, Iversen P, et al: A randomised comparison of “Casodex” (bicalutamide) 150 mg monotherapy versus castration in the treatment of metastatic and locally advanced prostate cancer. Eur Urol 33: 447–456, 1998.
UROLOGY 62 (1), 2003