Chapter  4:  Relative Effectiveness and Cost-Effectiveness of Methods of Androgen Suppression in the Treatment of Advanced Prostate Cancer: Evidence Report/Technology Assessment Number 4

Comparison of Monotherapies

• 1

  What is the effectiveness of treatment with an LHRH agonist compared to orchiectomy or diethylstilbestrol (DES)?

• 2

  What is the effectiveness of treatment with an antiandrogen compared to orchiectomy or DES?

• 3

  Is there any difference in effectiveness among the LHRH agonists?

• 4

  Is there any difference in effectiveness among the antiandrogens?

• 5

  How do the alternative methods of monotherapy compare with respect to adverse effects and quality of life?

The following agents were included in the assessment: LHRH agonists (leuprolide, goserelin, buserelin); nonsteroidal antiandrogens (flutamide, nilutamide, bicalutamide); and the steroidal antiandrogen, cyproterone.

Each agent was compared to orchiectomy or diethylstilbestrol (DES, 3 mg/d), which was confirmed to be equivalent to orchiectomy as a comparator. No trials compared the effectiveness of two LHRH agonists or two antiandrogens as monotherapies.

Combined Androgen Blockade

• 1

  Does combined androgen blockade improve outcomes compared to monotherapy using orchiectomy or an LHRH agonist?

• 2

  Does combined androgen blockade benefit particular subpopulations of patients?

• 3

  How do combined androgen blockade and monotherapy compare with respect to adverse effects and quality of life?

The evidence was examined in the aggregate (i.e., combined androgen blockade using any antiandrogen) and by class of antiandrogen (nonsteroidal antiandrogens or cyproterone). In addition, the class of nonsteroidal antiandrogens was examined by agent.

Immediate Compared to Deferred Androgen Suppression

Three distinct patient populations and treatment settings were considered. The evidence for each was analyzed separately because it is unknown whether findings from one population and setting generalize to another. All available trials used monotherapy; none used combined androgen blockade.

• 1

  In men who have previously undergone definitive therapy for initially localized prostate cancer, does androgen suppression initiated at prostate-specific antigen (PSA) rise improve outcomes compared to androgen suppression deferred until clinical progression?

• 2

  In men who are newly diagnosed with locally advanced or asymptomatic metastatic prostate cancer, does primary androgen suppression initiated at diagnosis improve outcomes compared to androgen suppression deferred until clinical progression?

• 3

  In men who have locally advanced or asymptomatic metastatic prostate cancer and who undergo radiotherapy, does adjuvant androgen suppression initiated with radiotherapy, and continued for several years or more, improve outcomes compared to radiotherapy alone followed by androgen suppression at clinical progression?

The population of interest to this report is men with advanced prostate cancer, including: (1) regional or disseminated metastases (D1 or D2; N+/M0 or M1); (2) minimally advanced disease (C; T3-4/N0 or NX/M0); and, only for immediate versus deferred therapy, (3) rising PSA, or other signs of progression after definitive therapy for early stage disease.

Outcomes of interest are: overall, cancer-specific, and progression-free survival; time to treatment failure; adverse effects; and quality of life. Where available, data on patient preferences are included. We also looked for treatment outcomes that were analyzed by patient subgroups based on prognostic factors such as tumor grade, extent of disease, and performance status.

Comparison of Monotherapies

Combined Androgen Blockade

Immediate Compared to Deferred Androgen Suppression

Cost-Effectiveness Analysis

• DES and orchiectomy are used infrequently in the United States, as regimens that are more expensive have grown in popularity. The meta-analysis suggests and the cost-effectiveness analysis confirms that the extra benefit from these new therapies is small, even after differential toxicities are accounted for. However, the results are very sensitive to the quality of life associated with orchiectomy. For patients whose quality of life would diminish substantially if they underwent orchiectomy, the use of LHRH agonists or nonsteroidal antiandrogens may represent reasonable alternatives.

• Combined androgen blockade is expensive. At a cost-effectiveness threshold of $100,000/ quality-adjusted life year (QALY), combined androgen blockade with an LHRH agonist must increase efficacy by 20 percent compared to orchiectomy before this drug combination is considered cost effective. Combined androgen blockade with an orchiectomy must increase efficacy by 10 percent. This value is within the 95 percent confidence limits of the meta-analysis.

• Our model suggests that for patients with locally advanced or asymptomatic metastatic cancer at the time of diagnosis, initiating antiandrogen therapy early, when patients enjoy a good quality of life, will result in higher costs and no added benefit, or possible harm, compared to deferring therapy. Our analysis did not address radiotherapy with adjuvant antiandrogen therapy for men who have locally advanced prostate cancer.

• Our findings change little when uncertain values entered into the model are varied over wide ranges in one-way sensitivity analyses.

1. The relative effectiveness of the available methods for monotherapy: orchiectomy, luteinizing hormone-releasing hormone (LHRH) agonists, and antiandrogens

Monotherapy uses a single drug or a surgical procedure, orchiectomy, for androgen suppression. The following drug classes and agents are included in the assessment: the LHRH agonists (leuprolide, goserelin, and buserelin), the nonsteroidal antiandrogens (flutamide, nilutamide, and bicalutamide), and the steroidal antiandrogen, cyproterone. Each agent was compared to orchiectomy or diethylstilbestrol (DES, 3 mg/d), which was confirmed by our meta-analysis to be equivalent to orchiectomy as a comparator. No trials directly compared the effectiveness of two LHRH agonists or two antiandrogens as monotherapies.

2. The effectiveness of combined androgen blockade compared to monotherapy

For combined androgen blockade, an antiandrogen is used with orchiectomy or an LHRH agonist in order to inhibit simultaneously production of and cellular responses to androgens. The evidence comparing monotherapy to combined androgen blockade was examined in the aggregate (i.e., using any antiandrogen) and by class of antiandrogen (nonsteroidal antiandrogens or cyproterone). In addition, the class of nonsteroidal antiandrogens was examined by agent.

3. The effectiveness of immediate compared to deferred androgen suppression

Three distinct patient populations and treatment settings were considered: (a) androgen suppression initiated at PSA rise in men who have previously undergone definitive therapy for initially localized prostate cancer, (b) androgen suppression as primary therapy in men who are newly diagnosed with locally advanced or asymptomatic metastatic prostate cancer, and (c) long-term androgen suppression initiated at radiotherapy in men with locally advanced or asymptomatic metastatic prostate cancer. For each population and setting, immediate androgen suppression was compared to androgen suppression deferred until signs or symptoms of clinical progression.

The population of interest to this report is men with advanced prostate cancer, including: (1) regional (D1; N+/M0) or disseminated (D2; M1) metastases; (2) locally advanced disease (C; T3-4/N0 or NX/M0); and, only for the third key question, (3) rising PSA, or other signs of progression after definitive therapy for clinically localized disease (A or B; T1-2/N0). Thus, the use of neoadjuvant (short-term) androgen suppression in candidates for definitive treatment of clinically localized disease is outside the scope of this report.

Outcomes of interest are: overall, cancer-specific, and progression-free survival; time to treatment failure; adverse effects; and quality of life. Where available, data on patient preferences are included. We also looked for treatment outcomes that were analyzed by patient subgroups based on prognostic factors such as tumor grade, extent of disease, and performance status. Two supplementary analyses were conducted for each key question: (1) meta-analysis of overall survival at 2 years (questions 1 and 2) and 5 years (questions 2 and 3); and (2) cost-effectiveness analysis.

This introductory chapter provides background information to aid the reader in understanding the clinical context and prior research on the key questions addressed in this evidence report. The following topics are addressed in the introduction:

• Prevalence of prostate cancer and costs of treatment.

• Prostate cancer staging systems.

• Treatment of prostate cancer.

• Description of androgen suppression methods considered in this report.

• Clinical research issues for each key question.

Clinically Localized Disease

Patients with clinically localized prostate cancer, i.e., disease that appears to be confined within the prostatic capsule, are offered definitive therapy (surgery or radiation) or watchful waiting. The American Urological Association Guidelines state that patient preference should guide the decision, as there is presently insufficient evidence to recommend treatment over watchful waiting or to recommend surgery or radiotherapy as the treatment of choice (Middleton, Thompson, Austenfeld et al., 1995). Options for definitive therapy for clinically localized prostate cancer are surgery (radical prostatectomy) or radiation therapy. Radiation may be delivered by external beam or by interstitial brachytherapy, a newer modality. Evidence comparing the long-term outcomes of brachytherapy and external beam radiotherapy is not yet available (Blue Cross and Blue Shield Association, 1997).

Longitudinal data from the Physician Health Study suggested that, for patients with an aggressive tumor, median survival from the initial rise in PSA levels is 13 years (Gann, Hennekens, and Stampfer, 1995). Based on this observation, treatment is frequently recommended for patients who have clinically localized disease and who are in good health, with 10 or more years of life expectancy (Balducci, Pow-Sang, Friedland et al., 1997). However, it is presently unknown whether immediate treatment of clinically localized disease improves survival over that in similar populations managed by watchful waiting and treatment upon progression (Albertsen, 1996; Albertsen, Fryback, Storer et al., 1995; Johansson, Holmberg, Johansson et al., 1997).

At surgery, a substantial proportion of patients is discovered to have more extensive disease: as much as 60 percent of those with clinical T2b lesions (Oesterling, Fuks, Lee et al., 1997). There has been considerable effort to improve techniques of predicting which patients actually have more extensive disease, so that unnecessary surgery can be avoided. A recent nomogram combines information from clinical stage, PSA, and Gleason biopsy grade (Partin, Kattan, Subong et al., 1997). In patients with PSA levels > 10, preoperative bone scan is used to detect metastases.

Because more extensive disease is so frequently discovered at surgery in patients initially diagnosed with clinically localized disease, neoadjuvant therapy has been attempted. In this approach, a short course of androgen suppression is administered for several months before either radical prostatectomy or definitive radiation therapy is performed (Garnick, 1997; Wang and Fair, 1997). When used with radiation therapy, the androgen suppression continues until the course of radiation is completed. The major objective of neoadjuvant androgen suppression is to downstage the tumor by reducing its volume so that it is confined within the prostatic capsule.

Data from pilot studies of neoadjuvant androgen suppression prior to prostatectomy suggest that tumors that appear to be locally advanced by clinical criteria infrequently are converted to organ-confined tumors (Wang and Fair, 1997). On the other hand, several studies reported that the frequency of finding extracapsular extension or positive margins at surgery was reduced in those with clinically localized disease given neoadjuvant androgen suppression (for review, see: Garnick, 1997; Wang and Fair, 1997). However, at present, data comparing long-term survival of patients given neoadjuvant androgen suppression prior to prostatectomy with survival after surgery alone have not been reported from randomized controlled trials.

Data from randomized trials that compare neoadjuvant androgen suppression followed by radiotherapy with radiotherapy alone also are limited (Zietman, Prince, Nakfoor et al., 1997). Interim analysis of results at 6 years of followup from one ongoing trial showed that neoadjuvant androgen suppression significantly improved local control and decreased the development of metastases when compared with radiation therapy alone (Pilepich, Winter, Roach et al., 1998). However, the difference in overall survival between treatment arms did not reach statistical significance.

Some surgeons suggest that shrinking the prostate prior to surgery also may reduce blood loss and other complications of surgery. Radiation therapists propose that neoadjuvant androgen suppression permits more selective targeting of radiation to the prostate, thus preventing collateral damage to surrounding tissues (primarily the bowel and bladder). Reports from several studies show that the prostatic volume is reduced by neoadjuvant androgen suppression (for review, see: Garnick, 1997; Wang and Fair, 1997). This does permit the use of a smaller target volume in treatment planning for patients treated with radiation. However, at present, no data are available from randomized comparative trials of neoadjuvant androgen suppression that demonstrate either reduced blood loss during prostatectomy or reduced collateral damage from radiation therapy.

Locally Advanced Disease

The objectives of treatment for locally advanced prostate cancer (stage C or T3-4/N0 or Nx/M0) include eradication (or at least control) of the primary tumor, prevention or delay of disease progression and the morbidity and symptoms it causes, and increasing the duration of survival (Corn and Hanks, 1997; Frydenburg and Oesterling, 1997; Lee, Hanks, and Schultheiss, 1996; Lerner, Blute, and Zincke, 1996; Oesterling, Fuks, Lee et al., 1997; Perez, 1997). At present, conclusive data from prospective, controlled, randomized studies are lacking to compare the modalities (and their combinations) for treatment of locally advanced prostate cancer. Patient selection bias and other confounding factors preclude reliable conclusions on optimal therapies based on data from single-arm studies of radiotherapy (with or without androgen suppression), surgery (with or without adjuvant radiotherapy or adjuvant androgen suppression), or androgen suppression alone.

The treatment of choice is likely to differ based on prognostic factors such as Gleason score and the degree of extracapsular extension: penetration of the capsule with no tumor at the margins (stage C1), margin involvement (stage C2), or seminal vesicle invasion (stage C3) (Frydenberg and Oesterling, 1997). However, treatment selection usually occurs before the patient's substage is known reliably, since the diagnosis requires pathologic examination on a surgical specimen. For those with higher grade and/or more extensive tumors, a single modality using either radiation therapy, surgery, or androgen suppression alone is generally considered inadequate treatment and not usually recommended (Corn and Hanks, 1997; Frydenburg and Oesterling, 1997; Lee, Hanks, and Schultheiss, 1996; Lerner, Blute, and Zincke, 1996; Oesterling, Fuks, Lee et al., 1997; Perez, 1997). Nevertheless, until data are available from randomized controlled trials, patient preferences regarding the balance of risks from progression and adverse effects will continue to play a major role in treatment selection.

The most frequently recommended treatment for locally advanced disease is external beam radiation, which may be supplemented with adjuvant androgen suppression for several years or more (Corn and Hanks, 1997; Lee, Hanks, and Schultheiss, 1996; Oesterling, Fuks, Lee et al., 1997; Perez, 1997; Pilepich, 1996). Radiation therapy alone is not curative if there are clinically silent regional or distant metastases or if it does not fully eradicate the primary tumor. Nevertheless, successful local control can delay progression and may increase survival duration (Bolla, Bartelink, Gibbons et al., 1997).

For patients with locally advanced disease, radiation is delivered to the tumor by means of an external beam either alone or combined with a boost dose of interstitial brachytherapy (Corn and Hanks, 1997; Lee, Hanks, and Schultheiss, 1996; Perez, 1997). Brachytherapy alone is not usually recommended for these patients (Prestidge, Prete, Buchholz et al., 1998). Evidence is not yet available from randomized trials comparing long-term survival after external beam radiotherapy with or without a brachytherapy boost. Evidence comparing the outcomes of treatment for locally advanced disease using radiation therapy alone with those using radiation combined with adjuvant androgen suppression for several years or more is reviewed in this assessment as part of Key Question 3.

Surgery alone also is unlikely to cure patients with prostate cancer that is judged to be T3 or T4 by clinical criteria. Nevertheless, prostatectomy followed by adjuvant radiation therapy or adjuvant androgen suppression also may be offered to these patients if they are otherwise healthy and have a life expectancy of at least 10 years (Frydenberg and Oesterling, 1997; Lerner, Blute, and Zincke, 1996). The observation that up to 20 percent of those with stage C or T3 tumors by clinical criteria are found to have organ-confined disease by pathologic examination after surgery often is cited as one justification for this approach. In addition, surgery with adjuvant radiation or adjuvant androgen suppression can achieve the clinical objectives of local tumor control. It also permits more definitive evaluation of lymph node involvement and lymph node dissection, if necessary. For patients who select surgical treatment of locally advanced disease, there are no data from randomized comparative trials to determine whether adjuvant therapy should use radiation or androgen suppression.

Salvage Therapy for Local Relapse or Residual Disease

Patients who relapse after treatment with radical prostatectomy for initially localized disease and those discovered to have positive margins at surgery may be offered radiation therapy to the tumor bed for salvage (Corn and Hanks, 1997; Perez, 1997; Zeitman, 1996). Similarly, those who relapse after radiation therapy for clinically localized disease and those whose tumors are not fully eradicated by primary irradiation may be offered radical prostatectomy for salvage (Reiter, Scardino, and Miles, 1996). Androgen suppression is the treatment of choice for those who are not suitable for, decline, or fail salvage therapy. Some clinicians also may recommend the addition of androgen suppression to salvage therapy.

Before the availability of PSA monitoring, clinical signs or symptoms of disease progression were required to identify those who failed definitive treatment for localized disease; presently, biochemical monitoring permits earlier detection. Measurable PSA levels after radical prostatectomy may indicate local treatment failure, metastatic disease, or both (Stamey and Kabalin, 1989). Many of these patients may remain free of symptoms and/or detectable metastases for extended periods of time (Frazier, Robertson, Humphrey et al., 1993). Elevated or increasing PSA levels after radiation therapy also commonly signal local and/or distant treatment failure, but, again, the disease may not be clinically symptomatic (Kuban, El-Mahdi, and Schellhammer, 1995a). A critical question for both groups failing definitive treatment is the optimal timing for initiating androgen suppression.

Metastatic Disease

The management of metastatic prostate cancer, disease that has spread to regional lymph nodes or beyond, takes into consideration age, comorbidities, clinical symptoms, and whether distant metastases or only regional lymph node involvement is present. Androgen suppression is the mainstay of treatment for metastatic prostate cancer (Balducci, Pow-Sang, Friedland et al., 1997). The treatment goals in these patients include preventing or delaying disease progression, preventing or delaying the development of symptoms or their worsening, and increasing the duration of survival.

For many years, when it was thought that the primary source of androgen (90 percent to 95 percent) was testicular, bilateral orchiectomy was the gold standard for effective ablation. A medical approach to obtaining castration levels of serum testosterone was effectively achieved with the administration of DES (Byar and Corle, 1988; Veterans Administration Cooperative Urological Research Group, 1967) or other estrogens, but the risk of cardiovascular complications has tempered their use. Newer drugs, luteinizing hormone-releasing hormone agonists, have been substituted; and orchiectomy may still be used, usually in a modified, subcapsular approach. But castration, whether achieved by surgical or medical means, does not suppress production of the adrenal androgens.

It is now well known that the 5 percent to 10 percent of circulating androgens that originate in the adrenal glands are controlled by the hypothalamus through the anterior pituitary (McLeod, Crawford, and DeAntoni, 1997). Whether the adrenal androgens are important in the growth of prostate cancer and a necessary target of medical suppression has only been hypothesized. The use of combined androgen blockade, which adds antiandrogen therapy to medical or surgical castration, has been advocated as standard therapy for advanced prostate cancer (Caubet, Tosteson, Dong et al., 1997; Debruyne and Dijkman, 1995; Labrie, Dupont, Belanger et al., 1982; Thrasher and Crawford, 1996) but remains the subject of controversy.

Other trials have investigated the intermittent use of androgen suppression therapy, in which treatment is given in response to PSA fluctuations. For example, therapy is administered with rising levels of PSA and withdrawn when PSA levels decrease (Goldenberg, Bruchovsky, Gleave et al., 1995; Klotz, Sogani, and Block, 1997). Obviously, this approach is possible only with medical approaches to androgen suppression. A potential advantage is recovery of sexual function and avoidance of the other adverse consequences of reduced serum testosterone when treatment is temporarily discontinued. There are concerns, however, that disease relapse and/or progression may occur more frequently than with continuous treatment. Long-term data on the outcomes of intermittent androgen suppression also are not yet available (Garnick, 1997; Klotz, Sogani, and Block, 1997; Moul, 1998; Newling, 1997).

Hormone-Refractory Disease

Prostate cancer that does not respond to or that relapses after primary androgen suppression treatment is particularly problematic. Patients who develop hormone-refractory clinical disease may respond to second-line androgen suppression therapies (Ritter, Messing, Shanahan et al., 1992; Small and Vogelzang, 1997). Various methods for secondary hormonal therapy are available, but responses rarely last longer than 6 months and most patients die within 1 year (Mahler, Akaza, Boccardo et al., 1997). Patients unresponsive to androgen suppression may be referred to clinical trials investigating chemotherapy regimens.

Results of various chemotherapy trials have been mixed (Rubben, Dijkman, Ferrari et al., 1997). A recent international committee has concluded that "complete response cannot be achieved by [single-agent] chemotherapy, that a cytostatic drug of first choice is not available, and that a comparative study with quality of life as an end point is urgently needed" (Rubben, Dijkman, Ferrari et al., 1997). Trials of combination chemotherapy are ongoing, as are studies assessing the efficacy of combining cytotoxic therapy and hormonal treatment (Rubben, Dijkman, Ferrari et al., 1997). In one such randomized study, palliation was achieved more frequently, and certain parameters of health-related quality of life were better in patients randomized to mitoxantrone plus prednisone than in those randomized to prednisone alone (Tannock, Osoba, Stockler et al.,1996). However, there was no difference in overall survival.

Production and Action of Endogenous Androgens

Androstenedione and testosterone (and its active metabolite, dihydrotestosterone [DHT]) are produced in the testes, whereas androstenedione and dehydroepiandrosterone (and its sulfate), which are relatively weak androgens, are produced by the adrenal glands (Denis, 1993; Huben and Perrapato, 1991; Vogelzang and Kennealey, 1992). The adrenal androgens undergo further conversion to testosterone and DHT in the peripheral tissues and in the prostate (Denis, 1993). About 95 percent of the circulating plasma testosterone is produced from cholesterol in the Leydig cells of the testes (Huben and Perrapato, 1991; Vogelzang and Kennealey, 1992; Wojciechowski, Carter, Skoutakis et al., 1986). The testes contribute, on average, 5 to 10 mg of testosterone to the daily androgen production, whereas the adrenal cortex contributes 0.4 mg of androgens per day, under the influence of adrenocorticotropin (Denis, 1993).

Approximately 95 percent of testosterone and dihydrotestosterone circulate bound to carrier proteins such as plasma beta-globulin, sex hormone-binding globulin (SHBG, including testosterone-binding globulin [TBG]), and, to a lesser extent, plasma albumin (Huben and Perrapato, 1991; Vogelzang and Kennealey, 1992). Androgens can circulate in the blood as "free" active unbound hormones; however, when bound to the carrier proteins, the androgens are inactive and cannot interact with tissue androgen receptors; approximately 3 percent of the total circulating testosterone circulate in the "free," unbound form (Vogelzang and Kennealey, 1992). Plasma testosterone concentrations in men aged 20 to 60 years are relatively constant, after which there is a steady decrease; this decrease is partly due to the increase in circulating SHBG, which decreases the concentration of free testosterone (Huben and Perrapato, 1991).

Testosterone enters cells by passive diffusion (Vogelzang and Kennealey, 1992; Wojciechowski, Carter, Skoutakis et al., 1986). Once inside the cell, testosterone is converted by the cytoplasmic membrane-bound enzyme 5-alpha-reductase to DHT, its active metabolite; dihydrotestosterone is the most active intracellular form of androgen (Huben and Perrapato, 1991; Vogelzang and Kennealey, 1992). Dihydrotestosterone binds to intracellular receptor proteins, forming a receptor-steroid complex that initiates transcription and protein production after the complex is transported to the nucleus. The proteins produced as a result of this process are believed to affect the growth and function of cells (e.g., prostatic cells, hormone-dependent cancer cells) (Denis, 1993; Vogelzang and Kennealey, 1992; Wojciechowski, Carter, Skoutakis et al., 1986).

The production of androgens in the Leydig cells of the testes is influenced by pituitary luteinizing hormone (LH), the production of which is regulated by the pulsatile release of hypothalamic luteinizing hormone-releasing hormone (Denis, 1993; Huben and Perrapato, 1991; Wojciechowski, Carter, Skoutakis et al., 1986). It is this pulsatile release that exerts the physiologic response of LH production (Denis, 1993; Wojciechowski, Carter, Skoutakis et al., 1986). LH regulates the rate-limiting step of the synthesis of testosterone from cholesterol in the Leydig cells (Wojciechowski, Carter, Skoutakis et al., 1986). This "hypothalamic-pituitary-gonadal" axis of testosterone production allows for interventions at several different points in the axis (Huben and Perrapato, 1991).

Orchiectomy

Bilateral orchiectomy is a relatively safe, albeit irreversible, procedure with low morbidity and associated cost (Vogelzang and Kennealey, 1992; Wojciechowski, Carter, Skoutakis et al., 1986). The surgical approach can be subcapsular or may involve total orchiectomy. The subcapsular approach preserves the scrotum and thus allows cosmetic replacement with prostheses.

Following surgery, circulating concentrations of testosterone fall to castrate levels (i.e., decreasing the total plasma concentration of testosterone to 90 to 95 percent of baseline) within 3 hours (Huben and Perrapato, 1991; Maatman, Gupta, and Montie, 1985; Vogelzang and Kennealey, 1992), providing a rapid and permanent means of androgen suppression. The rapid decline in circulating concentrations of testosterone was once an important advantage of orchiectomy when many patients had disseminated metastases at initial diagnosis and some were at risk for spinal cord compression. Currently, it is very rare to encounter newly diagnosed patients with such an advanced stage of disease.

Because orchiectomy permanently eliminates testicular production of androgens, there are no questions regarding compliance with therapy as there may be with medical approaches to androgen suppression. However, irreversibility can be a disadvantage for orchiectomy if current trials of intermittent androgen suppression show that its long-term efficacy is comparable to that of continuous androgen suppression. Orchiectomy also is poorly accepted by patients because of its psychologic impact (Vogelzang and Kennealey, 1992; Wojciechowski, Carter, Skoutakis et al., 1986). Nevertheless, some patients may prefer a one-time surgical treatment to repeated daily, monthly, or quarterly injections. Furthermore, there is inadequate information for patients to use in making treatment decisions that reliably compares the anticipated effects of different options for androgen suppression with the realities of life during and after treatment.

Diethylstilbestrol

Estrogens are one of the older methods of androgen suppression for advanced prostate cancer. Estrogens block the production of androgens by inhibiting pituitary LH secretion by blocking testosterone binding sites in the hypothalamus, resulting in a decrease in LHRH (Denis, 1993; Huben and Perrapato, 1991; Vogelzang and Kennealey, 1992). They also appear to inhibit directly testosterone synthesis in the testes as well as increasing the amount of circulating SHBG, increasing its binding capacity, thereby decreasing the unbound or "free" fraction of circulating androgens (Denis, 1993; Huben and Perrapato, 1991). Estrogen therapy, is seldom used currently for the treatment of advanced prostate cancer because of the relative risks, most notably cardiovascular, associated with such therapy (Physician Data Query Database, 1998).

A variety of estrogens have been used in the treatment of advanced prostate cancer, but for this report, only DES is of interest. DES was used in nearly all randomized trials that compared LHRH agonists or antiandrogens with estrogen therapy and thus serves, along with orchiectomy, as a common comparator of the relative effectiveness of the newer androgen suppression agents.

Diethylstilbestrol is a synthetic nonsteroidal estrogen (McEvoy, 1997). It can occur as DES base, as the diphosphate ester, or as the disodium salt of the diphosphate ester. One milligram of DES is approximately equivalent to 1.6 mg of DES diphosphate on a molecular basis (McEvoy, 1997).

Diethylstilbestrol has been used in the palliative treatment of advanced prostate cancer (McEvoy, 1997). At daily oral doses of 3 mg or more, DES reduces circulating testosterone concentrations to castrate levels within 7 to 21 days (Vogelzang and Kennealey, 1992).

DES is administered orally as the base or as the diphosphate ester or as an intravenous injection as the disodium salt of the diphosphate ester (McEvoy, 1997). Although DES at 1 mg/d has been prescribed, this level of administration decreases plasma testosterone concentration to castrate levels in only about 70 percent of patients (Huben and Perrapato, 1991). Nevertheless, data from a randomized trial showed that DES treatment at 1 mg/d was equivalent to castration with respect to survival and progression of metastatic disease (Newling, Pavone Macaluse, Smith et al., 1992). Most trials that compared an LHRH agonist or antiandrogen to DES utilized a dose of 3 mg/d. This dosage is preferred to the 5 mg/d dose used in some earlier trials (e.g., Veterans Administration Cooperative Urological Research Group, 1967), since it may provide the best balance of optimal androgen suppression with reduced cardiovascular toxicity (Atala and Amin, 1991; Huben and Perrapato, 1991; McEvoy, 1997; Physician Data Query Database, 1998).

Luteinizing Hormone-Releasing Hormone Agonists

The LHRH agonists were synthesized following determination of the natural gonadotropin-releasing hormone (GnRH) decapeptide structure in 1973 (Huben and Perrapato, 1991; (Wojciechowski, Carter, Skoutakis et al., 1986); modifications in the peptide sequence have resulted in compounds with much longer half-lives than the naturally occurring compound (Huben and Perrapato, 1991). GnRH is a collective term that includes both follicle-stimulating hormone (FSH)-releasing hormone (FSH-RH) and luteinizing hormone-releasing hormone (LHRH) (Wojciechowski, Carter, Skoutakis et al., 1986).

The effects of LHRH agonists on androgen production are thought to be due primarily to "downregulation," or desensitization of pituitary LHRH receptors, causing a suppression in luteinizing hormone release and a resultant decrease in testosterone production (Huben and Perrapato, 1991; Wojciechowski, Carter, Skoutakis et al., 1986). The long-acting analogs can occupy the pituitary LHRH receptors, without changing the affinity of those receptors for LHRH (Wojciechowski, Carter, Skoutakis et al., 1986). It is generally accepted that the pulsatile release of LHRH from the hypothalamus results in its physiologic effect (i.e., LH production), whereas maintenance of a sustained, high concentration of LHRH results in downregulation of the receptors (Wojciechowski, Carter, Skoutakis et al., 1986). The resultant decrease in testosterone production decreases both prostatic and testicular weight (Wojciechowski, Carter, Skoutakis et al., 1986). This type of pharmacologic therapy is sometimes termed "chemical" or "medical" castration, as it essentially results in the same physiologic state as that following surgical orchiectomy (Atala and Amin, 1991)

Serum testosterone may actually increase by as much as 50 percent above baseline when therapy with an LHRH agonist is first initiated (TAP Pharmaceuticals, Inc., 1997a, 1997b, 1997c). Since testosterone production is stimulated, patients may experience a tumor "flare," with a worsening of disease symptoms (e.g., bone pain, dysuria, hematuria, weakness) for the first few days to weeks. As therapy continues, however, serum testosterone will decrease to castrate concentrations (Huben and Perrapato, 1991; Physician Data Query Database, 1998; TAP Pharmaceuticals, Inc., 1997a, 1997b, 1997c; Wojciechowski, Carter, Skoutakis et al., 1986). Concomitant short-term therapy with antiandrogens or estrogens can prevent tumor flare (Physician Data Query Database, 1998). LHRH agonist therapy should be initiated with great caution in patients with impending spinal cord compression due to vertebral metastatic disease or early ureteral obstruction due to tumor involvement of the base of the bladder because of the possibility of tumor flare (TAP Pharmaceuticals, Inc., 1997a, 1997b, 1997c). Some clinicians maintain that LHRH therapy is contraindicated in such patients (Huben and Perrapato, 1991). However, the consequences of tumor flare may no longer be clinically significant, since patients rarely begin treatment with an LHRH agonist at this late stage of disease. LHRH agonists also can cause loss of libido, hot flushes, and impotence (Physician Data Query Database, 1998).

Because these agents are peptides, they are poorly absorbed and are only weakly active following oral administration; therefore, they have been administered as subcutaneous or intramuscular injections, intranasal sprays or drops, or percutaneous and intramuscular depots (Vogelzang and Kennealey, 1992).

Table 3. LHRH Agonists

Table 3. LHRH Agonists

Generic Name (Brand Name); Manufacturer	Use in Prostate Cancer a	Dose and Administration
Leuprolide acetate (Lupron®); TAP Pharmaceuticals	"palliative treatment of advanced prostatic cancer when orchiectomy or estrogen administration are either not indicated or unacceptable to the patient"	For the treatment of advanced prostate cancer, leuprolide is administered as an intramuscular depot injection (using a 22-gauge needle) in a dose of 7.5 mg per month (TAP Pharmaceuticals, Inc., 1997a, 1997b, 1997c). Three suspension strengths are available, allowing for administration once per month (7.5 mg); once every 3 months (22.5 mg); or once every 4 months (30 mg) (TAP Pharmaceuticals, Inc., 1997a,( 1997b, (1997c). The manufacturer states that Lupron Depot® must be administered under the supervision of a physician (TAP Pharmaceuticals, Inc., 1997a,( 1997b, (1997c). Lupron also may be administered as an aqueous subcutaneous injection in a dosage of 1 mg/d; these injections may be self-administered by patients (TAP Pharmaceuticals, Inc., 1996). In clinical trials, leuprolide was generally administered subcutaneously in daily doses of 1 mg, although doses of up to 20 mg/d also had been evaluated in early clinical trials (TAP Pharmaceuticals, Inc., 1997a,( 1997b, (1997c; Wojciechowski, Carter, Skoutakis et al., 1986.
Goserelin acetate (Zoladex®); Zeneca Pharmaceuticals	"palliative treatment of advanced carcinoma of the prostate when orchiectomy or estrogen administration are either not indicated or unacceptable to the patient"	For the treatment of advanced prostate cancer, goserelin is administered as a subcutaneous injection (with a 16-gauge needle) into the upper abdominal wall (Zeneca Pharmaceuticals, 1996a, 1996b). It is given at a dosage of 3.6 mg every 28 days or 10.8 mg every 12 weeks, depending on the strength of the formulation used (Zeneca Pharmaceuticals, 1996a, 1996b). The manufacturer states that Zoladex® must be administered under the supervision of a physician (Zeneca Pharmaceuticals, 1996a, 1996b).
Buserelin (Suprefact®, Hoechst Marion Roussel Canada, not commercially available in the U.S.)	previously untreated patients with stage C and D prostate cancer, either as monotherapy, or in combination with surgical orchiectomy or antiandrogen therapy	Buserelin is administered either intranasally or via subcutaneous injection (Brogden, Buckley, Ward, 1990). For the treatment of advanced prostate cancer, patients generally receive 3-7 days of high-dose therapy to achieve rapid androgen suppression (e.g., subcutaneous injection of 1500 µg daily), followed by daily maintenance dosages of 1200 µg intranasally, or 200 µ g subcutaneously (Brogden, Buckley, and Ward, 1990).

a

Note: material in quotation marks comes directly from the FDA-approved package insert

Antiandrogens

Antiandrogens block the stimulating effect of testosterone on prostate cell growth (Vogelzang and Kennealey, 1992). Antiandrogens can inhibit binding of endogenous androgens to the receptors present in target tissues and/or inhibit androgen uptake (Vogelzang and Kennealey, 1992; Schering Corporation, 1996). Interaction with the androgen receptor results in an unstable antiandrogen-receptor complex that is transient and does not result in androgen-dependent gene transcription and protein synthesis (Soloway and Matzkin, 1993). The relative binding affinity of the antiandrogens for the androgen receptor decreases with time, indicating that the antiandrogen-receptor complex dissociates more readily than the testosterone- or dihydrotestosterone-receptor complex (Soloway and Matzkin, 1993).

Antiandrogens can be divided into two classes-steroidal and nonsteroidal, based primarily on whether the compound exerts hormone-like effects (including estrogenic, progestational, or androgenic activity) (Huben and Perrapato, 1991). The nonsteroidal antiandrogens are generally "pure" antiandrogens, meaning that they have no androgenic steroidal effects (Brogden and Clissold, 1989; Vogelzang and Kennealey, 1992). These were developed to circumvent the steroidal adverse effects associated with the steroidal antiandrogens (e.g., cyproterone acetate, megestrol) (Vogelzang and Kennealey, 1992). Because they block both central (i.e., diencephalic) and peripheral androgen receptors when administered to nonorchiectomized rats, these drugs cause a slow and gradual increase in plasma testosterone due to the increase in LHRH secretion (Soloway and Matzkin, 1993). Increases in LH, testosterone, and estrogen (estradiol) have been reported in patients receiving nilutamide or bicalutamide as monotherapy (Soloway and Matzkin, 1993; Zeneca Pharmaceuticals, 1997). This has led to a clinical reluctance to use the nonsteroidal antiandrogens as monotherapy. In addition, limited clinical study of monotherapy in humans has produced equivocal results (Soloway and Matzkin, 1993).

Table 4. Nonsteroidal Antiandrogens

Table 4. Nonsteroidal Antiandrogens

Generic Name (Brand Name); Manufacturer	Use in Prostate Cancer a	Dose and Administration
Flutamide (Eulexin®); Schering Laboratories	"combination therapy with an LHRH agonist in the treatment of advanced (stage D2) prostate cancer" or "combined with an LHRH agonist and radiation therapy when used for the management of locally confined (stages B2 and C) disease"	250 mg orally 3 times daily (Schering Corporation, 1996; Huben and Perrapato, 1991). For stages B2-C prostate cancer, treatment with flutamide plus an LHRH agonist should begin 8 weeks prior to beginning radiation therapy and should continue during radiation therapy (Schering Corporation, 1996). For stage D2 metastatic cancer, flutamide plus LHRH therapy should be initiated immediately and continued until signs of disease progression (Schering Corporation, 1996).
Bicalutamide (Casodex®); Zeneca Pharmaceuticals	"for use in combination therapy with an LHRH agonist for the treatment of Stage D2 metastatic carcinoma of the prostate"	daily dosages of 50 mg orally (Zeneca Pharmaceuticals, 1997; Kolvenbag, Blackledge, and Gotting-Smith, 1998); although daily doses of up to 200 mg have been evaluated, especially as monotherapy (Kolvenbag, Blackledge, and Gotting-Smith, 1998; Zeneca Pharmaceuticals, 1997).
Nilutamide (Nilandron®); Hoechst Marion Roussel)	"in combination with surgical castration for the treatment of metastatic prostate cancer (Stage D2)"	300 mg (six 50-mg capsules) orally once daily for the first 30 days of therapy; thereafter, the dosage is 150 mg (three 50-mg capsules) once daily (Hoechst Marion Roussel, Inc., 1996). The manufacturer states that for maximum benefit, nilutamide therapy must begin on the same day as or on the day after orchiectomy (Hoechst Marion Roussel, Inc., 1996).

a

Note: material in quotation marks comes directly from the FDA-approved package insert.

Table 5. Steroidal Antiandrogens

Table 5. Steroidal Antiandrogens

Generic Name (Brand Name); Manufacturer	Use in Prostate Cancer	Dose and Administration
Cyproterone acetate (Androcur®); Schering Argentina, not commercially available in the U.S.	as monotherapy or in combination with LHRH agonists for the treatment of advanced prostate cancer; the drug also has been used in combination with surgical orchiectomy or may be used during the first few weeks of LHRH therapy to prevent tumor flare; or may be used in conjunction with surgical or medical orchiectomy to prevent hot flushes	Oral dosages of 200-300 mg/d have been used in clinical trials (Barradell and Faulds, 1994; McLeod, 1993; Schroder, 1993; Soloway and Matzkin, 1993). When used to prevent tumor "flare" associated with LHRH therapy, cyproterone is administered in an oral dosage of 200 mg/d for 7 days before initiating LHRH therapy (Barradell and Faulds, 1994).

Although other steroids with antiandrogenic effects (e.g., medroxyprogesterone acetate, megestrol acetate) have been used in clinical trials as treatment for advanced prostate cancer, none of them is currently used as primary therapy in the United States. Furthermore, none has been used in combination with an LHRH agonist or orchiectomy for combined androgen blockade. Consequently, these drugs and agents used for secondary hormonal therapy (e.g., aminoglutethimide, ketoconazole, spironolactone) are outside the scope of this evidence report.

Relative Effectiveness of Monotherapy Methods

The pioneering work of Huggins (Huggins and Hodges, 1941; Huggins, Stevens, and Hodges, 1941); (Herbst, 1941; (Herbst, 1942), and their colleagues established the clinical benefit of androgen ablation as therapy for men with advanced prostate cancer. During much of the time since these initial discoveries, the goal of androgen ablation has been to delay clinical progression and to palliate symptoms of metastatic disease. For all but the last two decades, surgery (i.e., orchiectomy) and estrogens were the only options available for androgen ablation.

Early placebo-controlled randomized trials conducted by the Veterans Administration Cooperative Urological Research Group (VACURG) compared orchiectomy with the synthetic estrogen, diethylstilbestrol (DES), and with their combination (Byar and Corle, 1988; Jordan, Blackard, and Byar, 1977; Veterans Administration Cooperative Urological Research Group, 1967). This landmark study demonstrated that 5 mg/d of DES reduced the number of deaths due to prostate cancer when compared with placebo or when added to orchiectomy. However, this dose of DES also increased the number of deaths due to cardiovascular disease and, thus, provided no improvement in overall survival. Furthermore, the excess deaths due to cardiovascular toxicity occurred mostly in the first year of treatment with DES, arguing against any benefit from early treatment. A second VACURG placebo-controlled randomized trial compared three dosages of DES and found a similar tradeoff between fewer deaths due to prostate cancer and more early deaths due to cardiovascular toxicity when the highest dose (5 mg/d) was compared with an intermediate dose (1 mg/d).

A newer approach to androgen ablation was developed in the late 1970s. This relied on the use of luteinizing hormone-releasing hormone agonists (e.g., leuprolide, goserelin, or buserelin) to suppress androgen production by the testes (Chrisp and Goa, 1991; Chrisp and Sorkin, 1991; Conn and Crowley, 1991; Plosker and Brogden, 1994; Roila, 1989). Avoiding the adverse psychologic effects of orchiectomy and, thus, increasing patients' acceptance of treatment was proposed as a major advantage of medical castration using LHRH agonists. Many randomized trials have compared the outcomes of treatment with LHRH agonists (e.g., leuprolide, goserelin, or buserelin) with either orchiectomy or DES. However, no studies directly compared one of the LHRH agonists with another drug of the same class when used for monotherapy.

The clinical studies of LHRH agonists also showed that similar endocrine physiologic sequelae of reduced testosterone concentrations resulted from surgical or long-term medical castration. These often may include impotence, loss of libido, gynecomastia, breast pain and tenderness, hot flushes, and/or loss of muscle mass. Additionally, concerns have been expressed regarding the possibility of osteoporosis as a long-term consequence of reduced testosterone concentrations (Daniell, 1997). Consequently, there was interest in developing drugs that prevent the intracellular actions of androgens by blocking their binding to the androgen receptor without reducing testosterone concentrations. Collectively termed "antiandrogens," these may be steroid (e.g., cyproterone, megestrol, or medroxyprogesterone acetates) or nonsteroidal (e.g., flutamide, bicalutamide, or nilutamide) in chemical structure. Antiandrogens are a fourth alternative to orchiectomy, DES, or LHRH agonists for treatment of advanced prostate cancer.

Based on results from clinical and laboratory animal studies, monotherapy with cyproterone acetate appears to be more effective than monotherapy with either megestrol acetate or medroxyprogesterone acetate (Goldenberg and Bruchovsky, 1997; Nicholson and Waxman, 1997; Schroder, 1993). Consequently, cyproterone acetate is the only steroidal antiandrogen that continues to be used alone (albeit outside the United States) for primary hormonal therapy of advanced prostate cancer. However, steroidal antiandrogens have progestational and/or gonadotropic effects in addition to their inhibitory effects at the androgen receptor. In contrast, the nonsteroidal antiandrogens are more selective in their actions at the intracellular androgen receptor, without any gonadotropic or progestational effects. Most of the comparative trials investigating antiandrogens as monotherapy for prostate cancer have utilized either DES or orchiectomy as the control treatments. Several of the antiandrogens also have been compared with an LHRH agonist or with the patients' choice of surgical or medical castration. No trials have directly compared the nonsteroidal and steroidal antiandrogens.

Orchiectomy and DES are the most commonly available comparators for all other monotherapies. Although DES is no longer widely used for hormonal suppression, trials comparing DES to orchiectomy were included in this report to establish the equivalence of DES and orchiectomy when used as the control arm to compare other monotherapies. Our initial meta-analysis established that the hazard ratio for survival after treatment with DES relative to survival after orchiectomy was 0.9726 (95 percent confidence interval 0.756 to 1.252). If DES were not included as a comparator, the evidence basis for this assessment would be less robust. The review would be reduced from 22 trials to 16 trials, and the number of patients would be reduced from more than 5,300 patients to somewhat fewer than 4,500 patients.

To compare the effectiveness of the methods available for monotherapy as primary treatment for advanced prostate cancer, this evidence report will address the following questions:

• 1

  What is the effectiveness of treatment with an LHRH agonist compared to orchiectomy or DES?

• 2

  What is the effectiveness of treatment with an antiandrogen compared to orchiectomy or DES?

• 3

  Is there any difference in effectiveness among the LHRH agonists?

• 4

  Is there any difference in effectiveness among the antiandrogens?

• 5

  How do the alternatives available for monotherapy compare with respect to adverse effects?

• 6

  How do the alternatives available for monotherapy compare with respect to their effects on quality of life and with respect to patient preferences?

Combined Androgen Blockade Compared to Monotherapy

More than 50 years ago, (Huggins and Scott,1945) reported clinical responses to bilateral adrenalectomy in four patients whose cancers had relapsed after orchiectomy. However, three of the four died soon after surgery because methods then available to replace adrenocortical steroid hormones were inadequate (Newling, 1997 ; Thrasher and Crawford, 1996). It is now well accepted that 5 percent to 10 percent of the DHT found in the prostate derives from circulating androgens (other than testosterone) that are produced by the adrenal gland. Neither orchiectomy nor treatment with an LHRH agonist prevents production of adrenal androgens. This is the theoretical rationale for combined androgen blockade. However, it is still controversial whether preventing the stimulation of prostate cancer cells by DHT formed from adrenal androgens adds clinical benefit beyond that of blocking testicular production of testosterone.

Despite the availability of better methods for hormone replacement after adrenalectomy, surgery for ablation of adrenal androgens is no longer used in men with prostate cancer. The antiandrogens are drugs that bind relatively tightly to the androgen receptor. By occupying the binding site, they prevent binding of the endogenous androgens. However, the antiandrogens do not permit the occupied receptor to carry out its normal repertoire of physiologic responses. Combined androgen blockade thus involves the use of either surgery or an LHRH agonist to prevent testicular production of testosterone, together with an antiandrogen to block the action of adrenal androgens.

An initial series of reports from nonrandomized studies generated considerable enthusiasm for this approach, particularly for combining an LHRH agonist with an antiandrogen, which avoided surgery (for reviews, see Hussain and Crawford, 1997; Labrie, Belanger, Cusan et al., 1996; Robertson, Roberson, Padilla et al., 1996; Thrasher and Crawford, 1996). The earliest reported results focused primarily on prevention of the flare reaction that occurred during the first few days of treatment with an LHRH agonist (Labrie, Dupont, Belanger et al., 1982). Subsequently, additional nonrandomized pilot studies reported that men given combined androgen blockade survived longer than historical controls. These observations led to randomized controlled trials to test the hypothesis that men with advanced prostate cancer survived longer after combined androgen blockade than after monotherapy with an LHRH agonist.

The National Cancer Institute (NCI)-sponsored Intergroup trial (INT 0036) was one of the earliest and largest randomized studies to test this hypothesis and reported positive results (Crawford, Eisenberger, McLeod et al., 1989). More than 600 men with stage D2 prostate cancer were treated with either leuprolide plus placebo or leuprolide plus flutamide. Median survival was 6 months longer in the combined treatment arm. Longer followup showed that an additional 3 percent of these patients were alive at 5 years, when compared with those in the control arm.

One concern in generalizing from the results of the Intergroup study (INT 0036) was whether combined androgen blockade only improved survival when compared to the use of an LHRH agonist alone. There was some doubt as to whether LHRH agonists were clinically as effective as orchiectomy, although there were comparable declines in serum testosterone concentration by one month after initiation of treatment. A related concern was whether the flare reaction, which is caused by the use of an LHRH agonist, might also bias the comparison. Consequently, additional trials were conducted to compare survival after orchiectomy with or without an antiandrogen, and to compare survival after orchiectomy alone with survival after an LHRH agonist plus an antiandrogen.

A variety of monotherapies and combinations have been tested. These include either orchiectomy or one of the LHRH agonists (leuprolide, goserelin, or buserelin) used alone in the control arm and used with a nonsteroidal antiandrogen (flutamide, nilutamide, or bicalutamide) or a steroidal antiandrogen (cyproterone) in the other arm. Although results from several of these trials agreed with those of the Intergroup study (INT 0036), most did not find a statistically significant difference in survival between treatment arms. Whether combined androgen blockade improves survival relative to monotherapy remains a matter of controversy.

The NCI also sponsored a second Intergroup trial (INT 0105) as a further contribution towards resolving this controversy. This trial, led by the Southwest Oncology Group, compared orchiectomy alone to orchiectomy plus a nonsteroidal antiandrogen and thus eliminates tumor flare as a potential source of bias. It is the largest trial of combined androgen blockade thus far, with sufficient statistical power to detect a 25 percent improvement in overall survival. Furthermore, the trial design included prospective stratification for major prognostic factors (extent of disease and performance status) and an adequate number of patients in the minimal disease subgroup to permit a test of the hypothesis that patients with minimal disease benefit more from combined androgen blockade than from monotherapy. The results of this trial at a median followup of 50 months have recently been reported (Eisenberger, Blumenstein, Crawford et al., 1998) and are included in this assessment.

Reviews and overviews that addressed the controversy over combined androgen blockade reached differing conclusions. One viewpoint, that the benefits of combined androgen blockade were unproven, was based on the large number of trials with no statistically significant difference between arms compared with the few with a significant difference (e.g., Fitzpatrick, 1996; Goldenberg and Bruchovsky, 1997; Schroder, 1995). Related concerns were the additional economic costs and adverse effects of combination therapy. The case for combined therapy was based on the size of the positive trials and issues related to the design of the negative trials (e.g., Debruyne, 1996; Iversen, 1997; Labrie, Belanger, Cusan et al., 1996; Schellhammer, 1996; Thrasher and Crawford, 1996). These issues included:

• Whether sample sizes were adequate to detect small differences in survival;

• Whether duration of followup was adequate;

• Whether consistent criteria were used to evaluate response or progression;

• Whether patients with locally advanced disease should have been aggregated with those who had metastatic disease; and

• Whether patients should be stratified prospectively for important prognostic factors.

The Prostate Cancer Trialists' Collaborative Group (PCTCG) addressed this controversy by a meta-analysis that combined results from 22 randomized trials with a total of 5,710 patients (Prostate Cancer Trialists' Collaborative Group, 1995). The PCTCG solicited updated individual patient data from all published and unpublished randomized trials that compared any monotherapy with any combination regimen, attempting to exclude trials with confounding factors that could bias the results. For each trial, the number of deaths in the combination arm was compared with the number that would have been expected on the basis of the average survival for all randomized patients using the logrank statistic. The PCTCG analysis used an intention-to-treat approach and had a median followup for all patients of 40 months with 57 percent of patients no longer alive. Crude death rates were 58 percent for monotherapy and 56 percent for combined therapy, with actuarial estimates of survival at 5 years of 22.8 percent and 26.2 percent, respectively. The difference in survival of 3.4 percent (95 percent CI 0 to 7 percent) was not statistically significant.

Potential limitations of the PCTCG meta-analysis have been widely discussed in the literature (e.g., Hussain and Crawford, 1997; Iversen, 1997; Schellhammer, 1996; Thrasher and Crawford, 1996). A threshold issue was whether trials using a nonsteroidal antiandrogen, which selectively blocks the androgen receptor, should be combined with those using a steroidal antiandrogen, which also have progestational and other hormonal effects. Although the PCTCG analysis found no statistically significant heterogeneity of results between these two groups of trials, the validity of pooling data across these trials remained a concern. The trials with cyproterone for combined androgen blockade were consistently negative whereas the trials with nonsteroidal antiandrogens were generally positive, even when the difference was not statistically significant.

A second issue with respect to the PCTCG meta-analysis was whether the results from trials that did not use short-term antiandrogens to block the flare reaction in the control arm should be combined with those that controlled flare or those that used orchiectomy. Trials with orchiectomy as the control did not need to prevent flare, since it only occurs with an LHRH agonist. The following were among the other issues raised:

• Survival was the only outcome collected and subjected to meta-anyalyses;

• Trials that were not mature with respect to followup were included;

• Data were not available from three large trials (including the SWOG INT-0105 trial with 1,371 patients);

• Trials that enrolled patients with and without metastases were included; and

• Analyses of outcomes for subgroups with minimal disease or other good prognostic features were not possible.

A second meta-analysis, restricted to randomized trials that used nonsteroidal antiandrogens in the combination regimen, was reported by Caubet and coworkers, (Caubet, Tostenson, Dong et al. 1997). This meta-analysis excluded studies published only as abstracts, those that did not publish data on survival, and one trial that used short-term cyproterone to control tumor flare in controls receiving an LHRH agonist alone. The analysis utilized two different methods to estimate the log hazard ratio for survival from each trial. Although 13 trials met the inclusion criteria for the meta-analysis, only 9 included the necessary statistical summaries in the published reports to permit estimation of the relative risk by at least one of the two methods. The meta-analyses used a random effects model to generate pooled estimates of the relative risk for survival from each of the two methods used to estimate the log hazard ratios. Both methods of analysis yielded statistically significant relative risks that favored combined androgen blockade (0.78, 95 percent CI 0.67 to 0.90; 0.84, 95 percent CI 0.76 to 0.93). There was also a statistically significant increase in time to progression (relative risk, 0.74; 95 percent CI 0.63 to 0.86) that favored combined androgen blockade.

Note also that a third meta-analysis is less commonly cited in the literature and now may be largely of historical interest because its scope was limited to only one of the nonsteroidal antiandrogens (Bertagna, De Gery, Hutcher et al., 1994). This meta-analysis, which included 7 trials and 1,056 patients, was restricted to double-blind studies that compared orchiectomy with orchiectomy plus nilutamide. The analysis reported that the odds of death were reduced in the combination arm relative to the monotherapy arm but that the difference was not statistically significant. The odds of disease progression were significantly reduced (odds ratio, 0.84; p=0.05).

As of this writing, a planned second cycle for updating the PCTCG meta-analysis is nearing completion. This analysis will be an important contribution to the literature on combined androgen blockade because of the updated data it contains. It includes data from additional trials that were not sufficiently mature at the time of the first analysis (most notably the SWOG INT 0105 study), as well as longer followup from all of the trials in the previous report.

To compare the effectiveness of combined androgen blockade with that of monotherapy as primary treatment for advanced prostate cancer, this evidence report will address the following questions:

• 1

  Does combined androgen blockade improve outcomes compared to monotherapy using orchiectomy or an LHRH agonist?

• 2

  Does combined androgen blockade benefit particular subpopulations of patients?

• 3

  How do combined androgen blockade and monotherapy compare with respect to adverse effects?

• 4

  How do combined androgen blockade and monotherapy compare with respect to their effects on quality of life?

One analysis will aggregate evidence for combined androgen blockade using any antiandrogen. A second analysis will aggregate evidence for each class of antiandrogen (nonsteroidal or steroidal antiandrogen) separately. To the extent permitted by the evidence, additional analyses will examine the class of nonsteroidal antiandrogens separately by agent (flutamide, bicalutamide, and nilutamide).

Immediate Compared to Deferred Androgen Suppression

Early placebo-controlled randomized trials conducted by the Veterans Administration Cooperative Urological Research Group found that excess deaths due to cardiovascular toxicity occurred mostly in the first year of therapy with DES, arguing against any benefit from early treatment (Byar and Corle, 1988; Jordan, Blackard, and Byar, 1977; Veterans Administration Cooperative Urological Research Group, 1967). Although orchiectomy was not associated with such toxic effects, there was reluctance to perform this surgery until absolutely necessary. The availability of the LHRH agonists, which avoided the cardiovascular risks associated with DES, stirred interest in the potential benefits of earlier androgen suppression. This renewed interest in the hypothesis that earlier treatment might improve survival and other outcomes for patients with advanced prostate cancer.

For asymptomatic patients, an important question is whether androgen suppression should begin early and which clinical observations should trigger the decision to start treatment (Crawford, Fourcade, Iversen et al., 1997; Mazeman and Bertrand, 1996; Schroder, 1989). Put simply, the alternatives are to initiate androgen suppression as soon as it is known that the disease has spread beyond the prostatic capsule, generally referred to as immediate therapy, or to defer therapy until progression is detected. In practice, however, these alternatives vary for patients at different points along the natural course of the disease.

One patient population of interest is those who are treated with definitive therapy (surgery or radiation) for early stage disease and then either are not cured or relapse. Presently, the most likely finding in this group of patients would be either a persistent elevation or a renewed increase in PSA levels. Their alternatives are to begin hormonal therapy immediately after the persistent or renewed PSA elevation is detected or to defer until additional progression is documented. Evidence of progression might include, for example, detection of nodal involvement or metastatic lesions by imaging techniques. Note also that the PSA threshold to trigger the decision for immediate therapy may vary.

A second patient population of interest is those with locally advanced disease at the time of initial diagnosis but without nodal involvement or distant metastases. The alternatives for these patients include radiotherapy with or without androgen suppression, surgery with adjuvant radiotherapy or adjuvant androgen suppression, or androgen suppression alone. Conclusive data are lacking to determine which of these choices provides optimal treatment outcomes, both for the group as a whole and for subgroups defined by the extent of extracapsular invasion and tumor grade (Corn and Hanks, 1997; Frydenburg and Oesterling, 1997; Lee, Hanks, and Schultheiss, 1996; Lerner, Blute, and Zincke, 1996; Oesterling, Fuks, Lee et al., 1997; Perez, 1997).

For those who select radiotherapy for locally advanced disease, it is of interest to determine if adjuvant androgen suppression that begins with radiotherapy and continues for several years or more improves outcomes when compared with androgen suppression that is deferred until additional progression. However, studies that restrict their focus to a direct comparison of these two alternatives leave open the possibility that immediate androgen suppression without radiotherapy may be adequate treatment for locally advanced disease. Furthermore, the benefits of radiotherapy combined with adjuvant androgen suppression may be restricted to patients with higher grade tumors and/or more extensive extracapsular extension.

An additional comparison of immediate to deferred treatment applies to those who select androgen suppression alone or after surgery for locally advanced disease. Except for the minority who select surgical treatment, many of these patients may continue to have elevated PSA levels following the diagnosis of locally advanced disease. For them, possible triggers for beginning deferred therapy might be the development of symptoms, the appearance of nodal or distant metastases, or a higher threshold for PSA levels.

The third population of interest incluYQZFEIz7y diagnosed patients with metastases that are not causing pain or other symptoms. Treatment alternatives for these patients are hormonal therapy immediately upon diagnosis or deferred until symptoms appear. The percentage of patients initially diagnosed at this stage of disease is declining in the United States. This decline may be due to the widespread adoption of screening programs using PSA measurement and/or digital rectal examination. Nevertheless, there were several small nonrandomized studies in patients with stage D1 disease that was detected by pathologic examination after radical prostatectomy (for review, see Mazeman and Bertrand, 1996). Data from these studies suggested a survival benefit from immediate treatment and led to randomized controlled trials to compare the outcomes of immediate and deferred therapy (Medical Research Council Prostate Cancer Working Party Investigators Group, 1997).

A fourth population of interest includes patients with apparently localized disease but at risk for extra-prostatic extension of the tumor based on tumor volume and/or PSA levels. These patients may be considered for brief early treatment with androgen suppression, usually referred to as neoadjuvant therapy. The underlying hypothesis of this approach is that it may shrink the tumor within the confines of the prostatic capsule and thus increase the possibility of cure with definitive therapy. As described previously, studies have been reported on the use of neoadjuvant androgen suppression prior to prostatectomy (e.g., Schulman, 1994; Van Poppel, Ameye, Oyen et al., 1992; Voges, Mottrie, Stockle et al., 1994; for review see Wang and Fair, 1997) or radiotherapy (e.g., Pilepich, Krall, al-Sarraf et al., 1995; Radiation Therapy Oncology Group, 1994; for review, see Zietman, Prince, Nakfoor et al., 1997). However, these patients have not been diagnosed with advanced prostate cancer and thus are outside the scope of the present systematic review.

In patients with locally advanced disease or asymptomatic metastatic cancer, the rationale for immediate hormonal therapy includes the following (Crawford, Fourcade, Iversen et al., 1997):

• A smaller tumor volume may be more easily treated than a larger tumor volume.

• Complications of the tumor, such as pathologic fractures or ureteral obstruction, may be prevented.

• Local progression may be prevented, and the frequency of transurethral resection for relief of obstruction may be reduced.

• Subgroup analyses from some trials of combined androgen blockade suggest patients with minimal disease may derive more benefit than those with extensive disease.

On the other hand, there also are observations and concerns that tend to support deferred therapy. These include:

• Laboratory research suggests early use of androgen suppression may lead to more rapid development of hormone-resistant disease (for brief review, see Mazeman and Bertrand, 1996).

• The longer duration of androgen suppression therapy when patients are treated immediately may substantially reduce the quality of life because they suffer the adverse effects of treatment for longer periods.

The last concern has lead to the concept of intermittent androgen suppression, in which treatment is discontinued once PSA levels decline below a set threshold and then reinstituted if or when PSA levels rise once again (Goldenberg, Bruchovsky, Gleave et al., 1995; Klotz, Sogani, and Block, 1997). This approach allows patients some time free of therapy and may provide a better quality of life. A concern that has been raised, however, is that intermittent therapy may not delay progression or improve survival as effectively as continuous therapy. Randomized trials comparing intermittent androgen suppression with continuous treatment are under way, but no data are available as yet (Garnick, 1997; Klotz, Sogani, Block et al., 1997; Moul, 1998; Newling, 1997).

To compare the effectiveness of immediate versus deferred androgen suppression as primary treatment for advanced prostate cancer, this evidence report will address the following questions, each in a distinct patient population:

• 1

  Patients: Men who have previously undergone definitive therapy (radical prostatectomy or radiation therapy) for initially localized prostate cancer.
  Question: Does initiation of androgen suppression immediately upon a rising PSA improve outcomes compared to initiation of androgen suppression that is deferred until signs or symptoms of clinical progression?

• 2

  Patients: Men who are newly diagnosed with locally advanced or asymptomatic metastatic prostate cancer and are undergoing primary therapy with androgen suppression.
  Question: Does primary androgen suppression initiated immediately at diagnosis improve outcomes compared to androgen suppression deferred until signs or symptoms of clinical progression?

• 3

  Patients: Men who have locally advanced or asymptomatic metastatic prostate cancer and are undergoing radiation therapy.
  Question: Does adjuvant androgen suppression initiated with radiotherapy, and continued for several years or more, improve outcomes compared to radiotherapy alone followed by androgen suppression initiated at signs or symptoms of clinical progression?

These distinct patient populations and treatment settings are considered separately because it is unknown whether findings from one population generalize to another population.

Types of Studies

This report includes studies that focus on the comparison of outcomes of different androgen suppression monotherapies, the comparison of outcomes of monotherapies versus combined androgen blockade, and the comparison of outcomes of immediate versus deferred initiation of androgen suppression. The primary criterion for selection of studies, with respect to efficacy outcomes, required that studies be designed as randomized controlled trials. For adverse events data, randomized trials were included, and nonrandomized phase II studies were included if they reported the frequency of patients withdrawing from therapy due to adverse events. For quality of life data, all reports that were identified by the search strategy were included.

Types of Participants

Trials that compared different options for monotherapy, or that compared monotherapy with combined androgen blockade, were included if they enrolled men with advanced prostate cancer who were not previously treated with hormonal therapy for prostate cancer. For this review, advanced prostate cancer is defined to include the following two groups:

• 1

  Those with disseminated and/or symptomatic metastases (defined as stage D1/D2, N+ or M1 disease); and

• 2

  Those with asymptomatic or minimally advanced disease (defined as stage C or T3-4NXM0 disease).

For trials that compared immediate versus deferred initiation of androgen suppression only, a third patient group was included:

• 3

  Those with a rising prostate-specific antigen (PSA), or other signs of progression, after surgery or radiotherapy for early stage disease.

Wherever possible, data for subgroups were analyzed separately; however, reports of outcome by subgroups were sparse in this literature. In addition, for purposes of the meta-analyses, sensitivity analyses were performed that restricted inclusion to studies that provided separate data for stage D2 patients. No other patient subgroup was represented by sufficient data to justify performing separate sensitivity analyses.

Trials that compared immediate with deferred treatment were included if they enrolled men in groups "2" and "3," as above.

Types of Interventions

Types of Outcome Measures

Trials were included if they reported at least one of the following outcomes, each of which were compared and analyzed separately:

• overall survival

• cancer-specific survival

• progression-free survival

• time to hormone refractory status

• time to treatment failure

• adverse effects of treatment

• quality of life

• patient preferences or satisfaction

Determining Study Eligibility

Procedures were followed with the intention of systematically screening references to include all studies relevant to this literature review. The title and abstract of each reference retrieved by the search initially were evaluated against the inclusion and exclusion criteria by one of two reviewers. Each reviewer was responsible for 50 percent of the retrieved citations. The reviewer sorted references into three categories: "include," "exclude," and "uncertain." All references sorted into the "include" category by either reviewer were slated for retrieval. All references sorted into the "exclude" category by one reviewer were evaluated by the other reviewer. Any reference sorted as "exclude" which the alternate reviewer sorted as "include" was slated for retrieval. All references sorted as "uncertain" were reconsidered by both reviewers with a bias toward being inclusive. After articles were retrieved, each reviewer evaluated all articles against the inclusion and exclusion criteria. Disagreements were resolved by consensus of the two reviewers.

Data Abstraction

Two reviewers independently collected data for each eligible study, recording it with electronic database software. The data elements that were abstracted, when available in published reports, are detailed in the data abstraction form (Appendix IV). Data elements were grouped into the following broad categories: trial identifiers, study methods (including enrollment and withdrawal numbers), patient characteristics, outcomes, and comments.

Among the data elements sought for these outcomes were the proportion of patients surviving at 1 year, 2 years, 5 years, and 10 years. The median survival durations were also sought for each outcome. When an article did not report these elements specifically, the reviewers estimated them from curves that were available in the published reports. After each reviewer completed data abstraction, the data were compared and disagreements were resolved by consensus of the two reviewers.

Assessing Study Quality

The reviewers assessed the quality of methods and reporting, with the intention of documenting the quality of individual trials and determining whether studies could be grouped into categories by grade of methodologic quality. Quality assessments included the following elements: method of random sequence generation, blinding of the randomization process during patient recruitment, blinding of the investigator and patient to treatment allocation prior to breaking the randomization code (except when the patient underwent surgical castration), reporting of withdrawals and handling of them in the analysis, use of intent-to-treat analysis, reporting of power analysis, whether compliance with treatment was monitored, and description of treatment protocols, including concurrent treatments.

Table 6. Decision Rules for Reviewer Quality Assessments (more...)

Table 6. Decision Rules for Reviewer Quality Assessments

Assessment	Decision Rule
Randomization method, adequate	The order of group assignments is generated in a truly random fashion: sequence of assignments is obtained from random number table, computer program, or coin flips (but not by coin flip when patient presents for randomization)
Randomization method, inadequate	Allocation by day of week, hospital number, date of birth, date of admission, alternate allocation
Randomization method, unclear	No details are provided about randomization method
Concealment of allocation method, adequate	Patients are assessed as meeting eligibility criteria and enrolled before assignment to treatment; the patient and investigator are unaware of next group assignment; use of consecutively numbered, sealed opaque envelopes; prenumbered or coded identical containers that are administered serially to patients; onsite computer system that contains assignments in a locked unreadable computer file; assignment by telephone call to central office that is unaware of patient characteristics, use of prepackaged drugs. After assignment has been revealed, individuals are unable to alter assignment or decision about eligibility
Concealment of allocation method, inadequate	The patient and/or investigator could alter assessment of eligibility or presentation for assignment based on prior knowledge of next group assignment
Concealment of allocation method, unclear	No details are provided about concealment of allocation method
Blinding not applicable	Assess as "yes" if one of the interventions is orchiectomy and comparisons involve only monotherapies
Blinding unclear	Assess as "yes" only if it is possible to blind patients (none of the monotherapies is orchiectomy, or the interventions are different drugs) and the article does not specify whether patients or investigators were blind to intervention
Withdrawals documented	Article provides full accounting of enrollment, including those randomized, those not randomized (with reasons for nonrandomization); those randomized and completed; and those randomized and not completed (with reasons for noncompletion)
Withdrawals partially documented	Article provides partial accounting of enrollment, with omissions or discrepancies from any of the following: those randomized, those not randomized; those randomized and completed; and those randomized and not completed
Withdrawals not documented	Article provides only the number of patients who were analyzed; no details are provided about original number of patients enrolled or withdrawals
Intent-to-treat analysis, all outcomes	All eligible patients randomized to each treatment were analyzed for all outcomes of interest
Intent-to-treat analysis, some outcomes	All eligible patients randomized to each treatment were analyzed for some of the outcomes of interest
Intent-to-treat analysis, no outcomes	Analysis of all eligible patients randomized was not performed for any of the outcomes of interest
Higher quality study	The study was double-blinded or blinding was not applicable and intent-to-treat analysis was performed on the outcome of interest
Units of time	Outcomes were measured in months. To convert from years, multiply by 12. To convert from weeks, divide by 4.35. To convert from days, divide by 30.44

During data abstraction, the reviewers discovered that information was commonly lacking about many of the above methodologic elements, making it impossible to make a definitive determination about many elements of methodologic quality. After consultation with members of the Technical Advisory Group, a simple method of evaluating study quality was adopted, using the two elements that were most often possible to determine from articles: blinding of intervention and use of intent-to-treat analysis. Studies of higher quality were those that used double-blinding or those where blinding was not applicable and those studies where an intent-to-treat analysis was performed on the outcome of interest. This method of classifying higher quality studies was used in the meta-analyses for the purposes of performing sensitivity analyses. Sensitivity analyses for those studies that were assessed as being of higher quality were performed after meta-analyses were performed on all eligible studies for a particular outcome.

Adverse Events

Abstraction and analysis of data on adverse events present particular difficulties. The difficulties encountered in this project are representative of the general problem of the limitations of clinical trials as a source of data on adverse events. One well-recognized problem is that some adverse events may be so infrequent that clinical trials are not large enough to capture events that may be of concern when the treatment is used in the general population of patients. A second problem is inconsistency in which adverse events are reported and how they are measured. Efforts to improve standards in reporting of randomized trials have emphasized the need for more thorough and systematic reporting of the spectrum of effects for an intervention (Mosteller, Gilber, and McPeek, 1980).

The literature on androgen suppression for the treatment of prostate cancer lacks a consistent framework for reporting adverse events. Studies vary substantially with respect to which adverse events are reported and how these are measured. Reports are often ambiguous with respect to whether failure to report on a specific adverse event means that the event did not occur or that data on the event was not collected. The methods for collecting, pooling, and reporting adverse events data used in this report attempt to systematize, to the extent feasible, a body of evidence that has serious limitations.

There are three types of evidence relating to adverse events: adverse events by category (e.g., cardiovascular, endocrine), adverse events leading to withdrawal from therapy, and adverse events reported by degree of severity. It is generally recognized that each of these methods of androgen suppression has characteristic adverse event profiles, in which certain categories of adverse events are more prominent or of greater clinical importance. Taken together, the primary categories of adverse events for all androgen suppression methods include the following: cardiovascular, endocrine, gastrointestinal, hepatic, and ophthalmologic. This review focused on these categories of adverse events.

Evidence on the severity of adverse events is likely to be of most use to patients and clinicians for the purposes of choosing interventions. However, there are very few data of this type. Only three studies have reported data on adverse events by degree of severity, which is insufficient to compare interventions. Therefore, adverse events leading to withdrawal from therapy serve as indicator of severity.

Supplemental Analysis: Meta-analysis

Meta-analysis describes a set of analytic techniques that allows for synthesis of data across a number of similarly designed studies. One of the rationales for performing meta-analysis is to increase the power by which to detect an effect when individual studies do not obtain statistically significant results. An excellent approach to the meta-analysis of trials in prostate cancer was provided in the analysis by (Caubet, Tosteson, Dong, and coworkers,1997). This report used that general approach with some additional guidance from the paper of (Whitehead and Whitehead,1991). One interesting methodologic problem resulted from the variety of monotherapies available and the fact that they are not all compared against the same control. A general methodology for handling this problem was given by (Hasselblad,1998) and was applied to this particular problem.

Survival rates for prostate cancer are poor, which implies a large value for the hazard rate (the rate of death across time). Although these values may not be exactly constant over time, in general, it was assumed that the hazard is constant. The actual survival curve often is not available and, as a result, those assumptions cannot be checked.

Based on the above assumptions, the object was to obtain estimates of the hazard rate for each arm of each study or to obtain the estimate of the proportional hazards term and its standard error. For studies that do not provide this, estimates from other statistics may be obtained. (Caubet, Tosteson, Dong, and coworkers,1997) suggested that the log-hazard ratio Beta can be estimated from the chi-squared value of the log-rank test:
(1)
where e is the number of events. In those cases where Kaplan-Meier curves are given, it is generally possible to estimate the individual hazards. Finally, in those cases where the survival at a given point in time, S, is given as S = X/N (where X is the number surviving and N is the number of subjects followed, the hazard can be estimated as:
(2)
where t is the time at which the survival was measured.

The hazard rates can be combined using a fixed or random effects model. The fixed effects model assumes that the hazard rates are the same in each study. In this case, the measures are often combined using an inverse variance weighting formula. This assumption is probably unreasonable given the variety of studies. The populations, initial stage, treatment, and length of followup are slightly different in each study.

The model used for the meta-analyses in this document is a random effects model (see, e.g., DerSimonian and Laird, 1986, and Hedges and Olkin, 1985). Random effects models differ from fixed effects models in that a measure of the variation between studies is included in computation of the total uncertainty used to compute weights for each estimate. The random effects mean and variance are used to compute confidence intervals in a manner that corresponds exactly to the fixed effects case.

Meta-analyses were performed where appropriate using random effects models. Sensitivity analyses were used to test for heterogeneity of methods (including the effect of including studies of lower methodologic quality), participants, and interventions. For the meta-analysis of combined androgen blockade, sensitivity analysis was also used to test whether studies that reported 5-year survival differed from studies reporting 2-year survival. An initial analysis was performed to determine whether the results of castration and administration of diethylstilbestrol are comparable and thus whether it is valid to pool studies in which the control arms used either of these monotherapies. Separate analyses will also be used to compare the available monotherapies, compare monotherapy with combined androgen blockade, and compare the outcomes of immediate androgen suppression with those of deferred treatment.

Supplemental Analysis: Cost-Effectiveness Analysis

The cost-effectiveness of androgen suppression strategies for patients with advanced prostate cancer also was evaluated for this report. The decision analysis model incorporates benefits and harms of the interventions, captures a broad range of costs, and accounts for quality of life effects. The model is conducted from a societal perspective, with all costs and benefits expressed in present value terms with a rate of time discount of 3 percent, in accordance with the guidelines of the Panel on Cost-Effectiveness in Health and Medicine ( Gold, Seigel, Russell et al., 1996 ). The analysis was conducted with DATA software version 3.0.16 (TreeAge). Further details about the cost-effectiveness analysis methods will be described in the cost-effectiveness analysis chapter.

Interventions

Ten randomized trials, including 1,908 patients, compared an LHRH agonist with either orchiectomy (comparison codes 2.1, 2.2, 2.3), DES (comparison codes 3.1, 3.2, 3.3), or the choice of orchiectomy or DES (comparison codes 4.1, 4.2, 4.3). There was some minor variability among trials in the doses, schedules of administration, and dosage forms of the LHRH agonists used (daily injection versus depot preparations given monthly, every 3 months, or every 4 months). Nevertheless, in all instances, previous clinical studies that compared doses or dosage forms showed that all the regimens utilized in the 10 trials reliably reduced serum testosterone concentrations to castrate levels by 1 to 3 weeks after treatment was initiated and maintained those levels while treatment continued. Thus, it is unlikely that any differences in dose, schedule, or dosage form might have biased the results of any trial or affected the comparability of results among trials.

Of the 10 trials of LHRH agonists, only 1 investigated leuprolide; this trial compared leuprolide to DES (comparison code 3.1) in 199 randomized patients (The Leuprolide Study Group, 1984). Five studies investigated goserelin: two trials compared goserelin with orchiectomy (comparison code 2.2) in a total of 608 randomized patients (Kaisary, Tyrrell, Peeling et al., 1991; Vogelzang, Chodak, Soloway et al., 1995) and three trials compared goserelin with DES (comparison code 3.2) in a total of 580 patients (Citrin, Resnick, Guinan et al., 1991; Moffat, 1990; Waymont, Lynch, Dunn et al., 1992 ). Four studies investigated buserelin: three trials compared buserelin with orchiectomy (comparison code 2.3) in a combined total of 354 patients (Bruun and Frimodt-Moller for the Danish Buserelin Study Group, 1996; de Voogt, Klijn, Studer et al., 1990; Koutsilieris and Tolis, 1985), and one trial compared buserelin with the choice of orchiectomy or DES (comparison code 4.3) in 167 patients (Huben and Murphy, 1988). All six trials with DES in the control arm used the same dose: 3 mg/d.

A total of 13 randomized trials were identified that included 3,840 patients and compared an antiandrogen with either orchiectomy (comparison codes 5.1 through 5.4), DES (comparison codes 6.1 through 6.4), an LHRH agonist (comparison code 8.1), or the choice of orchiectomy or an LHRH agonist (comparison codes 7.1 through 7.4). Of these, three trials studied the nonsteroidal antiandrogen flutamide: one trial compared flutamide with orchiectomy (comparison code 5.1) in 104 patients (Boccon-Gibod, Fournier, Bottet et al., 1997), and two trials compared flutamide with DES (comparison code 6.1) in a total of 132 patients (Chang, Yeap, Davis et al., 1996; Lund and Rasmussen, 1988). All three trials used the same dose of flutamide, 250 mg three times daily, and both trials with DES in the control arm utilized 3 mg/d.

Five trials investigated the nonsteroidal antiandrogen bicalutamide: one study compared it directly with orchiectomy (comparison code 5.3) in 376 patients (Iversen, Tveter, and Varenhorst, 1996), and four trials compared it with the patient's choice of orchiectomy or an LHRH agonist (comparison code 7.3) in a total of 2,105 patients (Chodak, Sharifi, Kasimis et al., 1995; Iversen and the International Casodex Investigators, 1994; Iversen, Tyrrell, Kaisary et al., 1998; Kaisary, Tyrell, Beacock et al., 1995). Of these five trials, three utilized 50 mg/d of bicalutamide (Chodak, Sharifi, Kasimis et al., 1995; Iversen, Tveter, and Varenhorst, 1996; Kaisary, Tyrell, Beacock et al., 1995) and two utilized 150 mg/d (Iversen and the International Casodex Investigators, 1994; Iversen, Tyrrell, Kaisary et al., 1998). This difference may make it difficult to compare results reported from these trials. No randomized trials investigated the nonsteroidal antiandrogen nilutamide as a monotherapy for advanced prostate cancer.

Five trials investigated the steroidal antiandrogen cyproterone: two studies compared cyproterone with orchiectomy (comparison code 5.4) in 175 randomized patients (Ostri, Bonnesen, Nilsson et al., 1991; Peeling and members of the British Prostate Group, 1984); three studies compared cyproterone with DES (comparison code 6.4) in 432 randomized patients (Moffat, 1990; Pavone-Macaluso, de Voogt, Viggiano et al., 1986; Peeling, 1984); and two studies compared cyproterone with goserelin (comparison code 8.1) in 689 randomized patients (Moffat, 1990; Thorpe, Azmatullah, Fellows et al., 1996). Note that two of these five studies (Moffat, 1990; Peeling, 1984) were three-arm trials and thus each provided data for two of the three comparisons of cyproterone. The dosage of cyproterone varied in these trials, with one utilizing 100 mg/d (Ostri, Bonnesen, Nilsson et al., 1991), one using 250 mg/d (Pavone-Macaluso, de Voogt, Viggiano et al., 1986), and the other three using 300 mg/d (Moffat, 1990; Peeling, 1984; Thorpe, Azmatullah, Fellows et al., 1996). All three trials with DES utilized a dose of 3 mg/d. Nevertheless, the differences in the doses of cyproterone may make it difficult to compare the results reported from these trials.

Three trials directly compared orchiectomy with DES (comparison code 1.0) in a total of 1,302 randomized patients. Of these, the VACURG trial (Blackard, Byar, and Jordan, 1973; Veterans Administration Cooperative Urology Research Group, 1967) utilized a dose of 5 mg/d of DES, whereas two studies (Peeling, 1984; Robinson, Smith, Richards et al., 1995) utilized a dose of 1 mg/d of DES.

Note that the studies with more than two monotherapy arms (Moffat, 1990; Peeling, 1984) are counted more than once in the above listing of trials. Note also that the VACURG trial comparing orchiectomy with DES reported most (but not all) outcomes separately for patients diagnosed in stage III or stage IV. To accommodate the outcomes reported from the VACURG trial, the Evidence Tables (see Appendix I) include three separate listings for the VACURG trial: one for stage III patients (Veterans Administration Cooperative Urology Research Group, 1967), one for stage IV patients (Blackard, Byar, and Jordan, 1973), and one for the combined group (Jordan, Blackard, and Byar, 1977).

Patient Populations

The populations of the trials that compared monotherapies overwhelmingly consisted of patients with metastatic disease, largely staged as either D2 or M1. Only 13 of the 24 trials required histologic confirmation of the diagnosis; an additional 4 accepted cytology for confirmation; and the remainder did not specify a requirement or a method for confirmation. All the trials were restricted to patients who had not undergone previous hormonal therapy for prostate cancer. The mean or median ages of the patients enrolled in these trials ranged from 65 to 75 years.

Seventeen of the 24 trials limited enrollment to patients with metastatic disease. Of these, 16 included only patients with metastases to bone, soft tissue, or extrapelvic lymph nodes (stages D2/M1). One trial also included patients with metastases only to regional lymph nodes but did not provide the distribution of patients by stage within the treatment arms (Peeling, 1984).

Five trials included patients with either locally advanced (stage C/M0) or metastatic disease (Koutsilieris and Tolis, 1985; Lund and Rasmussen, 1988; Moffat, 1990; Pavone-Macaluso, de Voogt, Viggiano et al., 1986; Waymont, Lynch, Dunn et al., 1992). In these five trials, the percentage of patients with locally advanced disease ranged from 15 to 48 percent but was generally balanced across the two treatment arms. The VACURG trial also included patients with either locally advanced (stage III) or metastatic disease but reported nearly all outcomes separately by stage. Finally, one study included only patients with stage C/M0 disease (Iversen, Tyrrell, Kaisary et al., 1998).

Within the individual trials, patient and tumor characteristics at study entry were generally well balanced between arms. The following were among the prognostic factors evaluated: age (19 studies), tumor grade (12 studies), presence or absence of bone pain (14 studies), and performance status (13 studies). In one small study (n=40), there were slightly more patients with poorly differentiated tumors in the arm treated with DES (45 percent) than in the arm treated with flutamide (35 percent) (Lund and Rasmussen, 1988). In another small study (n=29), 43 percent of those treated with buserelin and 100 percent of those treated with orchiectomy had bone pain from metastatic disease at study entry (Koutsilieris and Tolis, 1985). The distribution of patients with bone pain at study entry was imbalanced to a lesser degree in a second study (n=92) of flutamide (56 percent) versus DES (43 percent) (Chang, Yeap, Davis et al., 1996). No other imbalances between study arms that might have confounded the results were reported for these prognostic factors.

Only three studies each reported the distribution of patients within arms with respect to the duration of disease (Chang, Yeap, Davis et al., 1996; The Leuprolide Study Group, 1984; Waymont, Lynch, Dunn et al., 1992), PSA levels at study entry (Boccon-Gibod, Fournier, Bottet et al., 1997; Chodak, Sharifi, Kasimis et al., 1995; Iversen, Tyrrell, Kaisary et al., 1998), or race/ethnic group (Citrin, Resnick, Guinan et al., 1991; Iversen, Tyrrell, Kaisary et al., 1998; Vogelzang, Chodak, Soloway et al., 1995). In one of these studies, there was a significant imbalance (p=0.04) with respect to the percentage of African-American patients in the arm treated with goserelin (36 percent) compared with the arm treated with orchiectomy (24 percent) (Vogelzang, Chodak, Soloway et al., 1995). No other imbalances between study arms were reported.

Only three trials reported one or more of the primary outcomes separately by stage of disease (Iversen and the International Casodex Investigators, 1994; Iversen, Tyrrell, Kaisary et al., 1998; Pavone-Macaluso, de Voogt, Viggiano et al., 1986; Veterans Administration Cooperative Urology Research Group, 1967/ Blackard, Byar, and Jordan, 1973). None of these trials reported actuarial analyses that compared treatment arms with respect to any primary outcomes for subgroups defined by prognostic factors other than stage of disease.

Outcomes

Nearly all (21 of 24) of the studies comparing monotherapies that met the inclusion criteria for this report provided some data on the overall survival of randomized patients. However, the duration of followup was limited, and only six studies reported on survival at 5 years after treatment. Fifteen trials reported outcomes related to disease progression. Four of these studies reported either the actuarial or crude rate of progression-free survival, and the remaining 11 reported time to progression. Only one trial (the VACURG study) reported cancer-specific survival. Eleven trials provided data on time to treatment failure. Nearly all the trials provided some data on the adverse effects of treatment. However, the specific adverse outcomes that were reported varied substantially among the trials. Two randomized controlled trials and an additional nonrandomized study reported on quality of life.

Quality of Study Design and Conduct

Efficacy Outcomes

Adverse Events

Using the decision rules described in the Methodology section, the data were pooled across studies that used the same class of intervention. Classes of intervention were orchiectomy, DES, cyproterone, LHRH agonists, and nonsteroidal antiandrogens. Since DES is no longer widely used for treatment, adverse events associated with DES are not discussed in this review of evidence. Trials using DES were included in this evidence report to provide a common comparator of efficacy, in addition to orchiectomy, for other monotherapies. As described in the Methodology section, we report two types of evidence on adverse events: adverse events by category and adverse events leading to withdrawal from therapy.

The categories of adverse events of interest are: cardiovascular, endocrine, gastrointestinal, hepatic, and ophthalmologic. However, there were insufficient reports of adverse events within the hepatic and ophthalmologic categories. These data were pooled across studies that used the same class of intervention. Classes of intervention are: orchiectomy, LHRH agonists, nonsteroidal antiandrogens, and cyproterone.

Due to the limitations in the evidence, estimates of specific events by category reported here should be viewed with caution. The evidence on adverse events leading to withdrawal from therapy is more reliable. Within the intervention classes of LHRH agonists and nonsteroidal antiandrogens, adverse events leading to withdrawal from therapy are reported by specific agent. These data are summarized in Evidence Tables I.8; I.9; I.10 in Appendix I.

The only study that directly compared two LHRH agonists was a trial comparing different combination regimens without a monotherapy control arm (Schellhammer, Sharifi, Block et al., 1997). Therefore, this study is included in Part II: Combined Androgen Blockade rather than in this section. However, it should be noted that a recent abstract from this trial (Sarosdy, Schellhammer, Sharifi et al., 1998b) provides the only data that compare depot preparations of leuprolide and goserelin for local adverse effects at the injection site. Local pain, reactions, hypersensitivity, or development of a mass at the site of depot injection occurred very infrequently. Furthermore, there were no differences in frequency between patients given leuprolide and those given goserelin.

Quality of Life

Only two randomized trials used a standardized and validated instrument (Cleary, Morrissey, and Oster, 1995) to measure quality of life (Chodak, Sharifi, Kasimis, et al., 1995; Iversen, Tyrrell, Kaisary et al., 1998; comparison 7.3). Both compared bicalutamide with the choice of surgical or chemical castration. Treatment with bicalutamide was reported to improve sexual interest and physical capacity from that measured at enrollment by more than was observed for those randomized to surgical or chemical castration (p<0.01). A few other reports discussed quality of life in relation to the frequency of various disease symptoms and adverse effects of therapy but did not attempt to measure quality of life.

The only other data that compare the effects of monotherapies on quality of life in men with advanced prostate cancer is from a nonrandomized study of 115 men who selected treatment with goserelin and 32 men who selected orchiectomy (Cassileth, Soloway, Vogelzang et al., 1992). Quality of life, as measured by the Functional Living Index: Cancer scale, improved from baseline at both 3 and 6 months in those who selected goserelin (p=0.0001) but did not change from baseline in those who selected orchiectomy (p=0.54 at 6 months). These investigators also reported an improvement in psychosocial measures for those who selected goserelin but no improvement for those who selected orchiectomy. These data must be interpreted cautiously, however, because the study was not randomized and the factors that influenced patient choice of therapy may have biased the results.

Although it is not strictly a measure of quality of life, patient preferences when they were offered a choice between treatments may provide useful information. Two of the four studies on bicalutamide that offered those in the control arm a choice of orchiectomy or an LHRH agonist (comparison 7.3) reported the number who chose each option. Of 285 patients offered this choice in the two trials (Iversen, Tyrrell, Kaisary et al., 1998; Kaisary, Tyrrell, Beacock, et al., 1995), only 87 (30 percent) chose orchiectomy. However, no direct evidence compares patients treated with orchiectomy with those given an LHRH agonist with respect to quality of life. Thus, it is not known whether patient perceptions when contemplating this choice of therapies are borne out by the realities of life after treatment.

Meta-Analysis

Table 7. Combined Estimates of Treatments for Prostate (more...)

Table 7. Combined Estimates of Treatments for Prostate Cancer - Hazard Ratios Relative to Orchiectomy

Treatment Class	Agent	Hazard Ratio	95% CI Lower	95% CI Upper
Estrogens	diethylstilbestrol	0.9927	0.762	1.294
LHRH gonists	leuprolide	1.0994	0.207	5.835
	goserelin	1.1172	0.898	1.390
	buserelin	1.1315	0.533	2.404
Nonsteroidal antiandrogens	flutamide	1.9583	0.369	10.394
	bicalutamide	1.2027	0.973	1.487
Steroidal antiandrogens	cyproterone	1.2005	0.592	2.433

Table 8. Combined Estimates of Treatment Classes for (more...)

Table 8. Combined Estimates of Treatment Classes for Prostate Cancer - Hazard Ratios Relative to Orchiectomy

Treatment Class	Hazard Ratio	95% CI Lower	95% CI Upper
Diethylstilbestrol a	0.9835	0.764	1.267
Antiandrogens	1.2194	0.995	1.494
Nonsteroidal antiandrogens	1.2158	0.988	1.496
LHRH agonists a	1.1262	0.915	1.386

a

Results for diethylstilbestrol and LHRH agonists taken from comparisons with antiandrogens in Tables 8-10. Findings for these treatment classes compared with nonsteroidal antiandrogens were similar to the above data.

Table 9. Combined Estimates of Treatment Classes for (more...)

Table 9. Combined Estimates of Treatment Classes for Prostate Cancer-Hazard Ratios Relative to Orchiectomy for Studies of Subjects With Stage D2 Cancer

Treatment Class	Hazard Ratio	95% CI Lower	95% CI Upper
Diethylstilbestrol	0.7788	0.568	1.068
Steroidal Antiandrogens	1.1336	0.988	1.300
Nonsteroidal antiandrogens	1.2492	0.985	1.585
LHRH agonists	1.0467	0.869	1.261

Table 10. Combined Estimates of Treatment Classes for (more...)

Table 10. Combined Estimates of Treatment Classes for Prostate Cancer-Hazard Ratios Relative to Orchiectomy for Studies Determined to be of High Quality

Treatment Class	Hazard Ratio	95% CI Lower	95% CI Upper
Diethylstilbestrol	0.9338	0.701	1.244
Steroidal Antiandrogens	1.2390	0.992	1.548
Nonsteroidal antiandrogens	1.2386	0.992	1.547
LHRH agonists	1.1494	0.921	1.435

Figure 1. Survival at 2 Years, Monotherapies

   Figure 1. Survival at 2 Years, Monotherapies

Figure 1

Figure 2. Survival at 2 Years, Monotherapies, by Category (more...)

   Figure 2. Survival at 2 Years, Monotherapies, by Category

Figure 2

This reduced analysis assumed that the various treatments that were grouped were actually equivalent. To formally test that hypothesis, we calculated the likelihood ratio chi-squared value with 4 degrees of freedom and found it to be 1.268 (p=0.742), indicating that the treatments were not different from each other.

The above analysis was repeated for those studies that restricted the subjects to stage D2. This eliminated six comparisons of monotherapies from four studies (Iversen, Tyrrell, Kaisary et al., 1998; Pavone-Macaluso, de Voogt, Viggiano et al., 1986; Veterans Administration Cooperative Urology Research Group, 1967; Blackard, Byar, and Jordan, 1973; Waymont, Lynch, Dunn et al., 1992).

Figure 3. Survival at 2 Years, Monotherapies, by Category, (more...)

   Figure 3. Survival at 2 Years, Monotherapies, by Category, Stage D2 Patients

Figure 3

Note that the results are virtually identical to the analysis of all stages except that the confidence bands are slightly wider.

Figure 4. Survival at 2 Years, Monotherapies, by Category, (more...)

   Figure 4. Survival at 2 Years, Monotherapies, by Category, High Quality Studies

Figure 4

Again, note that the results are virtually identical to the analysis of all stages except that the confidence bands are slightly wider.

Summary

Twenty-four randomized controlled trials, including more than 6,600 patients, met the study selection criteria for this report. These trials compared the following monotherapies for androgen suppression in men with advanced prostate cancer:

• DES versus orchiectomy

• LHRH agonist versus DES or orchiectomy

• Nonsteroidal antiandrogen versus DES or orchiectomy or an LHRH agonist

• The steroidal antiandrogen cyproterone versus DES or orchiectomy or an LHRH agonist

No study directly compared one LHRH agonist to another as monotherapies.

Although DES is no longer widely used for hormonal suppression, trials comparing DES with orchiectomy were included in this report to establish the equivalence of DES and orchiectomy when either is used as the control arm to compare other monotherapies. If DES were not included as a comparator, the evidence basis for assessing the relative efficacy of LHRH agonists and the antiandrogens would be less robust. The review would be reduced from 22 trials, including more than 5,300 patients, to 16 trials, including somewhat less than 4,500 patients.

The population studied was overwhelmingly patients with metastatic disease, largely staged as either D2 or M1. Of the 24 trials, 21 reported either median survival or survival at 2 years after enrollment, or both. Only six trials reported survival at 5 years, and two others reported survival at 4 years.

A meta-analysis combined data from 20 trials on overall survival at 2 years. Meta-analysis was performed using a random effects model. Each drug used for androgen suppression was compared to orchiectomy. A hazard ratio of 1.0 indicates that patients treated with the therapy of interest and patients treated with orchiectomy had an equal chance of death from any cause within 2 years of treatment.

The meta-analysis included two sensitivity analyses. One restricted the analysis to subjects with stage D2/M1 disease; the second was restricted to higher quality trials. A trial was classified as higher quality when it was double-blinded (not required when orchiectomy was one of the treatments) and it reported outcomes based on an intention-to-treat analysis. The sensitivity analyses yielded results consistent with the overall analysis but with wider confidence intervals.

These trials did not provide sufficient information to stratify results by known prognostic factors. Nor did the trials consistently report outcomes other than overall survival, such as cancer-specific or progression-free survival. No trial measured quality of life by a standard instrument.

Conclusions

Interventions

Twenty-three of the 27 trials comparing monotherapy with combined androgen blockade were two-arm trials. One of these trials used a 2X2 factorial design but is classified here as a two-arm trial because results of the two monotherapy arms and the two combined androgen blockade arms were pooled after 48 weeks of followup (Schulze, Kaldenhoff, and Senge, 1988). The overwhelming majority of the two-arm studies compared orchiectomy, leuprolide, or goserelin with the same monotherapy plus either flutamide or nilutamide.

Four of the 27 trials were three-arm trials. Three of these trials compared combined androgen blockade using cyproterone as the antiandrogen to two different monotherapy arms (de Voogt, Klijn, Studer et al., 1990; Robinson, Smith, Richards et al., 1995; Thorpe, Azmatullah, Fellows et al., 1996). The fourth compared monotherapy to two combined androgen blockade arms, each with a different dose of nilutamide (Brisset, Boccon-Gibod, Botto et al., 1987).

The 28th trial (n=813) compared four regimens for combined androgen blockade: goserelin with flutamide, goserelin with bicalutamide, leuprolide with flutamide, and leuprolide with bicalutamide (Schellhammer, Sharifi, Block et al., 1997). Nearly all reports from this trial pooled data from each of two arms to compare flutamide with bicalutamide for combined androgen blockade. However, a recent report from this study included separate analyses of outcomes for each of the four arms (Sarosdy, Schellhammer, Sharifi et al., 1998a).

Among the 27 trials comparing combined androgen blockade with monotherapy, the monotherapy was orchiectomy in 14 studies (n=4,156), an LHRH agonist in 12 studies (n=3,733), and orchiectomy or an LHRH agonist in 1 study (Schulze, Kaldenhoff, and Senge, 1988). Only 2 (n=588) of the 14 trials that used orchiectomy as monotherapy combined an LHRH agonist with an antiandrogen in the other arm (Denis, Carnelro de Moura, Bono et al., 1993; Iversen, Christensen, Friis et al., 1990).

Drug treatments were continued until death, disease progression, or withdrawal due to adverse effects in all studies. The 12 trials that compared an LHRH agonist with and without an antiandrogen used the same dose of LHRH agonist in both arms. All used doses, dosage forms, and schedules of administration known to reduce serum testosterone to castrate levels within 2 to 3 weeks after initiation of treatment and to maintain this level while therapy continued. Thus, it is unlikely that variations in the results of individual trials might be attributable to differences in the treatment regimens used for LHRH agonists.

Only 4 of the 12 trials that used an LHRH agonist for monotherapy also used an initial brief treatment with an antiandrogen to control the tumor flare reaction (Bono, DiSilverio, Robustelli della Cuna et al., 1998; de Voogt, Klijn, Studer et al., 1990; Ferrari, Castagnetti, Ferrari et al., 1993, 1996). Thus, the other eight trials may be biased against the monotherapy arm because an antiandrogen was not used to control flare. The flare reaction does not occur when orchiectomy is used as monotherapy. Therefore the 14 trials in which the control arm was orchiectomy alone may provide a better comparison of combined androgen blockade with monotherapy.

All trials with flutamide as the antiandrogen used the same dose (250 mg 3 times per day). All but one of the trials with nilutamide as the antiandrogen used a dose of 300 mg/d. The single exception (Dijkman,Janknegt, De Reijke et al.,1997) and one of three arms in a second trial (Brisset, Boccon-Gibod, Botto et al., 1987) used 150 mg/d. The trials using cyproterone as the antiandrogen varied in dosage, ranging from 100 to 300 mg/d.

Patient Populations

The overwhelming majority of patients in trials of combined androgen blockade had metastatic disease, largely stage D2. Reports from 18 of the 28 trials specified that histologic confirmation of the diagnosis was required. All trials were restricted to patients undergoing primary hormonal therapy. The mean or median age of these patients was between 65 and 75 years. On the whole, these studies were well balanced with regard to stage and, where reported, on the distribution of other prognostic factors.

A substantial number of studies reported the distribution of prognostic factors other than stage, such as tumor grade or presence of symptoms from metastases. However, only four trials provided data that compared treatment arms for one or more primary outcomes after stratifying by prognostic group (Crawford, Eisenberger, McLeod et al., 1989; Denis, Carnelro de Moura, Bono et al., 1993; Eisenberger, Blumenstein, Crawford et al., 1998; Iversen, Christensen, Friis et al., 1990). Two others stated that they completed comparative analyses on subgroups, but did not publish the relevant data (de Voogt, Klijn, Studer et al., 1990; Robinson, Smith, Richards et al., 1995). The lack of reporting of results stratified by prognostic groups is a notable deficiency of this body of evidence.

Enrollment was limited to patients with metastatic disease in 20 trials. Of these, 12 trials enrolled only those with metastases to the bone, soft tissue, or extrapelvic lymph nodes (stages D2 or M1). In the eight trials that also enrolled patients with metastases limited to the pelvic lymph nodes (stages N1 to N3/M0 or D1), the percentage of those not staged as M1/D2 ranged from 3 percent (Namer, Toubol, Caty et al., 1990) to 30 percent (Ferrari, Castagnetti, Ferrari et al., 1996). However, the arms within each of these trials were well balanced for stage of disease.

Of the remaining trials, seven included patients with no evidence of metastases (stages C or T3-T4/N0/M0). The percentage of patients without metastases ranged from 7 percent (Iversen, Christensen, Friis et al., 1990) to 45 percent (Boccardo, Pace, Rubagotti et al., 1993), but the arms within each of these trials also were well balanced for disease stage. The final trial (Schulze, Kaldenhoff, and Senge, 1988) did not provide information on the stage distribution of the randomized patients.

Fourteen trials compared treatment groups for the degree of differentiation of the primary tumor (i.e., tumor grade). Of these, only one (Dijkman, Janknegt, De Reijke et al., 1997) did not include any patients with well--differentiated tumors. Among the others, the percentage of patients in each arm with poorly differentiated, high-grade tumors ranged from less than 20 percent in three trials (Di Silverio, Serio, D'Eramo et al., 1990; Thorpe, Azmatullah, Fellows et al., 1996; Tyrrell, Altwein, Klippel et al., 1991) to 50 percent to 60 percent in three others (Iverson, Christensen, Friis et al., 1990; Namer, Toubol, Caty et al., 1990; Williams, Asopa, Abel et al., 1990). The treatment arms were imbalanced with respect to the percentage of patients with poorly differentiated tumors in only three studies. The imbalance favored combined androgen blockade (53 percent versus 34 percent) in one trial (Namer, Toubol, Caty et al., 1990), whereas it favored the control arm (50 percent versus 63 percent in each study) in the others (Iversen, Christensen, Friis et al., 1990; Williams, Asopa, Abel et al., 1990).

Twenty trials compared treatment groups for the percentage of patients who were symptomatic at study entry. For this report, symptomatic was defined as the presence of pain from skeletal metastases. The proportion of symptomatic patients ranged from less than 30 percent (Denis, Carnelro de Moura, Bono et al., 1993) to almost 80 percent (Crawford, Eisenberger, McLeod et al., 1989). The treatment arms were imbalanced with respect to bone pain at study entry in two studies. The imbalance favored the control arm (42 percent versus 61 percent) in one trial (Brisset, Boccon-Gibod, Botto et al., 1987) and favored the combined androgen blockade arm (64 percent versus 42 percent) in the other (Knonagel, Bolle, Hering et al., 1989).

Twenty-five studies compared treatment arms for the distribution by age; they were uniformly well balanced. Twenty trials compared treatment arms for the number of patients with impaired performance status. In three trials, more patients were impaired in the combined androgen blockade arms (Brisset, Boccon-Gibod, Botto et al., 1987; Fourcade, Cariou, Coloby et al., 1990; Knonagel, Bolle, Hering et al., 1989). Eight studies limited enrollment to those who were newly diagnosed, and five also accepted previously treated patients. Of the five, only two compared arms with respect to the duration of disease, and both were well balanced (Dijkman, Janknegt, De Reijke et al., 1997; Zalcberg, Raghaven, Marshall et al., 1996). Published reports from the remaining 15 trials were silent with respect to this characteristic.

Only six studies compared arms with respect to racial/ethnic distributions and all were well balanced (Crawford, Eisenberger, McLeod et al., 1989; Crawford, Kasimis, Gandara et al., 1990; Dijkman, Janknegt, De Reijke et al., 1997; Eisenberger, Blumenstein, Crawford et al., 1998; Schellhamer, Sharifi, Block et al.,1997; Tyrrell, Altwein, Klippel et al., 1991). Only two studies reported PSA levels at entry (Bono, DiSilverio, Robustelli della Cuna et al., 1998; Dijkman, Janknegt, De Reijke et al., 1997); both were well balanced. No other imbalances between study arms that might have confounded the results were reported.

Outcomes

Efficacy outcomes summarized in the Results section include overall survival, cancer-specific survival, progression-free survival and/or time to progression, and time to treatment failure. Nearly all the trials provided some data on the adverse effects of treatment. However, the specific adverse outcomes that were reported varied markedly among the trials.

Overall survival after monotherapy was compared with survival after combined androgen blockade in 21 studies. The 21 trials included 23 separate comparisons of a monotherapy with a combination regimen and a total of 6,871 patients. An additional trial that compared combined androgen blockade regimens without a monotherapy control group also reported overall survival (Schellhammer, Sharifi, Block et al., 1997).

Overall survival was obtained at the following intervals: 1 year (19 trials), 2 years (20 trials), and 5 years (10 trials). The published Kaplan-Meier curves were used to estimate nearly all of these data. Median overall survival was obtained from 18 trials. These data were provided explicitly in 12 studies and were estimated from the published Kaplan-Meier curves in 6.

Cancer-specific survival was reported in five trials. Time to treatment failure was reported in six trials. Some measure of disease progression was reported in 21 trials. Five trials reported progression-free survival, 1 trial reported the crude progression-free rate without an actuarial analysis, and 15 trials reported time to disease progression. Increase in the serum level of PSA was rarely used as evidence of disease progression.

Nearly all the trials provided some data on the adverse effects of treatment. However, the specific adverse outcomes that were reported varied markedly among the trials. In addition, one randomized trial, a feasibility study of a second trial, and a cross-sectional survey reported on quality of life.

Quality of Study

Efficacy Outcomes

Overall Survival

Data on survival were reported by 15 trials, including 5,636 patients, that used nonsteroidal antiandrogens for combined androgen blockade and from 6 trials, including 1,235 patients, that used cyproterone in the combination arms (total n=6,871). Overall, 18 of the 21 trials, including 5,485 patients, found no statistically significant difference in survival between monotherapy and combined androgen blockade (Evidence Table II.2, Appendix II). This group includes the trial recently completed by the Southwest Oncology Group (Eisenberger, Blumenstein, Crawford et al., 1998), which compared orchiectomy plus placebo with orchiectomy plus flutamide and is the largest single trial comparing monotherapy with combined androgen blockade (n=1,382). Median survival was 29.9 months in the control arm and 33.5 months in the combination arm.

Only three trials, including 1,386 patients, reported a statistically significant difference in survival between the monotherapy and combined androgen blockade arms (Crawford, Eisenberger, McLeod et al., 1989; Denis, Carnelro de Moura, Bono et al., 1993; Dijkman, Janknegt, De Reijke et al., 1997). All three trials favored combined androgen blockade, and each utilized a nonsteroidal antiandrogen in the combination arms. There was no survival benefit from combined androgen blockade in any of the six trials that used cyproterone in the combination arm.

Table 11. Survival Data from Studies With Significant Differences (more...)

Table 11. Survival Data from Studies With Significant Differences

Study	Treatment Arms	N	Median (months)	2 Years (%)	5 Years (%)	P Value
Dijkman et al., 1997	orchiectomy orch+nilut	232 225	23.6 27.3 =3.7	51 60 =9	18 27 =9	0.0326
Crawford et al., 1989	leuprolide leup+ flut	300 303	29.0 35.0 =6	59 66 =7	23 26 =3	0.035
Denis et al. 1993	orchiectomy goser+flut	163 163	27 34 =7	54 61 =8	20 28 =8	0.004

Table 11 summarizes pertinent findings from the three trials with significant differences between treatment arms. Dijkman, Janknegt, De Reijke, and coworkers (1997) compared orchiectomy plus placebo with orchiectomy plus nilutamide. Median survival was 3.7 months longer in the combination arm and overall survival was 9 percent greater at both 2 years and 5 years. Crawford, Eisenberger, McLeod, and coworkers (1989) compared leuprolide plus placebo with leuprolide plus flutamide. Median survival was 6 months longer in the combination arm and survival was 7 percent greater at 2 years and 3 percent greater at 5 years. Denis, Carnelro de Moura, Bono, and coworkers (1993) compared orchiectomy with goserelin plus flutamide. Median survival was 7 months longer in the combination arm and survival was 7 percent greater at 2 years and 8 percent greater at 5 years.

Of the 18 trials that did not reach statistical significance, 3 reported trends that favored combined androgen blockade. For this report, a trend is defined as Kaplan--Meier curves that differ with 0.05<p<0.20. The recent SWOG/ECOG trial (INT 0105; n=1,382) compared orchiectomy plus placebo to orchiectomy plus flutamide (Eisenberger, Blumenstein, Crawford et al., 1998). Median survival was 29.9 months versus 33.5 months and survival was 62 percent versus 63 percent at 2 years and 28 percent versus 32 percent at 5 years (p=0.14). Beland, Elhilali, Fradet, and coworkers (1990) compared orchiectomy plus placebo with orchiectomy plus nilutamide (n=194). Median survival was 18.9 months versus 24.3 months and survival at 2 years was 38 percent versus 49 percent (p=0.137). Tyrrell, Altwein, Klippel, and coworkers (1991) compared goserelin alone to goserelin plus flutamide (n=569). Median survival was 37.7 months versus 42.4 months and survival was 65 percent versus 72 percent at 2 years and 34 percent versus 42 percent at 5 years (p=0.14). Only one trial found a slight nonsignificant trend that favored monotherapy (Di Silverio, Serio, D'Eramo et al., 1990). This study (n=315) compared goserelin alone to goserelin plus cyproterone; median survival was 30.1 months versus 23.8 months and survival at 2 years was 63 percent versus 49 percent (p=0.26).

Combined Androgen Blockade Compared with Monotherapy in Prognostic Subpopulations

Only four trials compared treatment arms for one or more primary outcomes after stratifying by prognostic group. The first NCI Intergroup study (INT 0036) compared leuprolide alone to leuprolide plus flutamide and reported significantly longer overall survival in the combination arm. Patients in this trial were divided into good or poor prognostic groups based on the extent of disease (limited versus extensive) and ECOG performance status (0-2 versus 3) (Crawford, Eisenberger, McLeod et al., 1989; Eisenberger, Crawford, Wolf et al., 1994). There were large differences in median survival (42 months versus 61 months) and progression-free survival (19 months versus 48 months) that favored the arm given combined androgen blockade in the group with the best prognosis (n=82). Substantially smaller differences between arms were reported for the groups with a poorer prognosis (n=522). However, note that only a small percentage of patients (13.6 percent) met the criteria for the best prognostic group. Because of the small sample size, the investigators did not test these differences for statistical significance.

The second Intergroup study (INT 0105) stratified patients by minimal versus extensive disease and by good versus poor (>3, ECOG scale) performance status, with dynamically balanced randomization with respect to the stratification factors (Eisenberger, Blumenstein, Crawford et al., 1998). Furthermore, the investigators allowed an accrual overrun in order to increase the trial's statistical power in the minimal disease subgroup. Although less than 5 percent of patients in either arm had poor performance status, they reported separate actuarial analyses of overall and progression-free survival for the minimal and extensive disease subgroups of each study arm. There was no statistically significant difference in the median overall (52.1 months) or progression-free (48.1 months) survival for patients in the minimal-disease subgroup treated with orchiectomy plus flutamide when compared with those in the same subgroup treated with orchiectomy plus placebo (51.0 months and 46.2 months, respectively). Thus, this large, adequately powered trial did not confirm the hypothesis regarding a potentially enhanced benefit of combined androgen blockade in a subset of patients with minimal disease.

(Iversen, Christensen, Friis, and coworkers,1990) compared goserelin alone to goserelin plus flutamide and did not report a statistically significant benefit from combined androgen blockade for either overall or cancer-specific survival or for time to progression. These investigators also reported cancer-specific and progression-free survival separately in subgroups with minimal (n=73) and extensive (n=189) disease (Iversen, Rasmussen, Klarskov et al., 1993). There were no statistically significant differences between treatment arms for either outcome in either prognostic subgroup. Again, though, only 28 percent of the patients were in the subgroup with minimal disease.

The European Organization for Research and Treatment of Cancer (EORTC) trial 30853 also compared orchiectomy alone to goserelin plus flutamide and found significantly longer survival in the combination arm (Denis, Carnelro de Moura, Bono et al., 1993). The hazard ratio for death (orchiectomy to combination therapy) was 0.67 for those with WHO performance status 0 (n=108) and 0.79 for those with WHO performance status 2 (n=70). Similarly, the hazard ratios were 0.60 for those with fewer than 5 skeletal lesions on bone scan (n=74) versus 0.84 for those with more than 15 lesions (n=86). However, no confidence intervals were reported for these data, and there were few patients in each subgroup.

Two additional trials reported that they compared the effects of monotherapy and combined androgen blockade for good and poor prognostic groups based on formulas that combined factors shown to have independent prognostic value in proportional hazards analyses (de Voogt, Klijn, Studer et al., 1990; Robinson, Smith, Richards et al., 1995). Neither study included Kaplan-Meier curves for any of the efficacy outcomes that compared treatments for specific prognostic groups. However, both stated that there was no significant effect of combined androgen blockade in either the good or poor prognostic groups.

Comparisons of Different Regimens for Combined Androgen Blockade

The Casodex Combination Study (Schellhammer, Sharifi, Block et al., 1996a, 1996b, 1997) was the only trial that directly compared different regimens for combination therapy. This trial utilized a 2X2 factorial design, with leuprolide and goserelin as the LHRH agonists and flutamide and bicalutamide as the nonsteroidal antiandrogens. Patients in each of the four arms received one of the four possible combinations. Eleven of the 13 reports published by these investigators pooled data on outcomes for all patients given flutamide and for all patients given bicalutamide.

The median overall survival was longer in the group that received an LHRH agonist plus bicalutamide (41 months; n=404) than in the group that received an LHRH agonist plus flutamide (34 months; n=409). However, the hazard ratio for mortality did not achieve statistical significance (0.87; 95 percent CI 0.72 to 1.05; p=0.15). There also was no statistically significant difference between the groups in median times to progression (22 months versus 18 months; hazard ratio, 0.93; 95 percent CI 0.79 to 1.10; p=0.41). At a median followup of 49 weeks, the hazard ratio for time to treatment failure was significantly in favor of those given an LHRH agonist plus bicalutamide (42 percent versus 53 percent; hazard ratio 0.749; p=0.005). After additional followup to a median of 95 weeks, there was only a strong trend in favor of fewer treatment failures among those given an LHRH agonist plus bicalutamide (hazard ratio 0.87; 95 percent CI 0.74 to 1.03; p=0.10).

A recent abstract from this study reported data separately for each of the four study arms (Sarosdy, Schellhammer, Sharifi et al., 1998a). There were no statistically significant differences in survival between the goserelin plus bicalutamide (n=268) and goserelin plus flutamide (n=272) groups or between the leuprolide plus bicalutamide (n=136) and goserelin plus bicalutamide (n=268) groups. There was a significant difference in survival that favored the leuprolide plus bicalutamide group (n=136) over the leuprolide plus flutamide group (n=137) (survival at 4 years: 48 percent versus 32 percent; p=0.008). There also was a significant difference that favored the goserelin plus flutamide group (n=272) over the leuprolide plus flutamide group (n=137) (survival at 4 years: 40 percent versus 32 percent, p=0.047). Of the four combinations, survival was greatest for leuprolide plus bicalutamide and shortest for leuprolide plus flutamide. However, the p values reported were not corrected for multiple comparisons, and the investigators point out that the trial did not have adequate statistical power to support definitive comparisons between the four regimens they tested.

An additional set of exploratory analyses from the Casodex Combination trial (Sarosdy, Schellhammer, Sharifi et al., 1998a, 1998b) pooled data for all patients given goserelin plus a nonsteroidal antiandrogen (n=540) and compared these data to those pooled for all patients given leuprolide plus a nonsteroidal antiandrogen (n=273). The hazard ratio for time to progression was 0.99 (95 percent CI 0.84 to 1.18; p=0.92). The hazard ratio for overall survival was 0.91 (95 percent CI 0.75 to 1.11; p=0.34). Local adverse effects at the site of depot injection were infrequent and occurred no more frequently with one of the LHRH agonists than with the other.

Adverse Events

Using the decision rules described in the Methodology section of this report, the data were pooled across studies that used the same class of intervention. We report two types of evidence on adverse events: adverse events by category and adverse events leading to withdrawal from therapy. Because of the limitations in the evidence, estimates of specific events by category reported here should be viewed with caution. The evidence on adverse events leading to withdrawal from therapy is more reliable. The categories of adverse events of interest are: cardiovascular, endocrine, gastrointestinal, hepatic, and ophthalmologic. These data were pooled across studies that used the same class of intervention. Classes of intervention are: orchiectomy or LHRH agonist plus nonsteroidal antiandrogen and orchiectomy or LHRH agonist plus cyproterone. Not included in the tables, but discussed below, are data from the Casodex Combination trial that compared four regimens of combined androgen blockade and reported on local adverse effects from injection of depot dosage forms for LHRH agonists (Sarosdy, Schellhammer, Sharifi et al., 1998a, 1998b).

The second type of evidence was on adverse events leading to withdrawal from therapy for each of the specific antiandrogens used in combination with chemical or surgical castration. These data are summarized in Evidence Tables II.11-13 (Appendix II).

Quality of Life

A substudy of the recently completed SWOG/ECOG trial (INT 0105) is the only source of data from a randomized controlled trial that compares patients treated with a monotherapy to patients treated with combined androgen blockade with respect to formal measures for quality of life (Moinpour, Savage, Lovato et al., 1997; Moinpour, Savage, Troxel et al., 1998). Data were collected at randomization and at 1, 3, and 6 months after the start of treatment on three treatment-related symptoms (diarrhea, gas pain, and body image), on physical functioning, and on emotional functioning. Patients in this substudy who were treated with orchiectomy plus flutamide reported significantly more diarrhea at 3 months (p<0.001), and worse emotional functioning at 3 and 6 months (p<0.003), than did those given orchiectomy plus placebo. There also were nonsignificant trends toward more problems among those in the combination arm with the following: physical functioning, fatigue, abdominal gas, overall pain, and body image.

A cross-sectional survey used three separate instruments to measure quality of life in patients receiving an LHRH agonist plus flutamide (Albertsen, Aaronson, Muller et al., 1997). However, this study only compared patients who progressed while on combined androgen blockade to those who did not progress. No patients treated with a monotherapy were included. A feasibility study conducted by the EORTC (as part of protocol 30853) included an optional assessment of quality of life at entry and at each followup visit using a validated questionnaire (da Silva, 1993; da Silva and Aaronson, 1988). However, only 63 of the 327 patients randomized to either orchiectomy or goserelin plus flutamide completed the questionnaire at one or more clinic visits (da Silva, Fossa, Aaronson et al., 1996; da Silva, Reis, Costa et al., 1993). Consequently, it was not possible to compare the two treatment arms.

It is noteworthy, however, that the EORTC feasibility study demonstrated marked discrepancies between patients' self-assessments and treating physicians' evaluations of quality-of-life parameters. Other investigators have reported similar discrepancies (Litwin, Lubeck, Henning et al., 1998). Consequently, it is unlikely that a report of more frequent improvement and/or delayed deterioration in subjective parameters assessed by physicians after orchiectomy plus flutamide than after orchiectomy plus placebo (Dijkman, Fernandez del Moral, Debruyne et al., 1995) can be interpreted as evidence of improved quality of life. Also noteworthy is the marked discrepancy between the patients' perceptions collected by the SWOG investigators (Moinpour, Savage, Lovato et al., 1997; Moinpour, Savage, Troxel et al., 1998) and physicians' subjective evaluations reported by Dijkman, Fernandez del Moral, Debruyne et al. (1995) for the identical comparison of treatments.

A few other reports discussed quality of life in relation to the frequency of various disease symptoms and adverse effects of therapy but did not formally attempt to measure quality of life.

Summary

Twenty-seven randomized controlled trials, including 7,987 patients, compared the outcomes of monotherapy with the outcomes of combined androgen blockade. Twenty of these trials used a nonsteroidal antiandrogen in the combined androgen blockade arm; 12 trials used flutamide, 8 used nilutamide, and none used bicalutamide. In seven trials, combined androgen blockade was accomplished with the nonsteroidal antiandrogen agent cyproterone. The monotherapy arm was orchiectomy in 14 trials, an LHRH agonist in 12 trials, and either orchiectomy or an LHRH agonist in 1 trial. In addition, a 28th trial compared four regimens for combined androgen blockade.

Twenty-one trials compared survival after monotherapy to survival after combined androgen blockade (n=6,871). Of these, 20 trials reported overall survival at 2 years, and 10 of these trials also reported survival at 5 years. Twenty-one trials reported some measure related to disease progression: either progression-free survival (5 trials) or time to progression (16 trials). Only five studies reported cancer-specific survival and only six trials reported time to treatment failure.

Conclusions

Interventions

The trial conducted by the Medical Research Council (MRC) Working Party Investigators Group randomized 938 patients (Medical Research Council Working Party Investigators Group, 1997). Immediate therapy was initiated within 6 weeks of study entry, using either orchiectomy or an LHRH agonist with a short-term course of antiandrogen to control tumor flare. Deferred therapy was initiated when clinically significant progression occurred, based on the treating physician's usual practice. Deferred therapy was initiated for symptoms of local progression almost as frequently as for metastatic symptoms.

One concern about the MRC trial involves whether patients in the deferred therapy arm were allowed to progress too far before initiation of hormonal therapy (Walsh, 1997). This concern is raised by evidence that 3 percent of patients experienced pathologic fracture or spinal cord compression before initiation of deferred therapy. If patients were allowed to progress too far, it creates the potential for bias that would inflate differences in outcomes between immediate and deferred therapy.

The Veterans Administration Cooperative Urological Research Group (VACURG) studies were placebo-controlled trials designed to examine the efficacy of orchiectomy, DES at various doses, and orchiectomy, alone or in combination with DES ( Blackard, Byar, and Jordan 1973 ; Byar and Corle, 1988 ; Christensen, Aagaard, and Madsen 1990 ; Hurst and Byar, 1973 ; Jordan, Blackard, and Byar, 1977 ; Veterans Administration Cooperative Urological Research Group, 1967 ). Hormonal therapies tested in the VACURG trials and found to be ineffective (e.g., DES 0.2 mg) or excessively toxic (e.g., DES 5 mg) are not reviewed here. In terms of comparing immediate to deferred androgen suppressive therapy, the first relevant comparison in the VACURG studies concerns the outcomes of patients who received immediate orchiectomy versus those of placebo-treated patients who received therapy at the first sign of disease progression; this comparison is referred to as "VACURG 1" in the tables (Blackard, Byar, and Jordan 1973; Veterans Administration Cooperative Urological Research Group, 1967). The second relevant VACURG comparison concerns outcomes of patients receiving immediate DES therapy (1 mg) versus those of placebo-treated patients who received therapy at the first sign of disease progression; this comparison is referred to as "VACURG 2" in the tables (Byar and Corle, 1988; Christensen, Aagaard, and Madsen 1990).

In the VACURG studies, placebo group patients received active therapy when the treating physician judged that a change of therapy was necessary for the welfare of the patient. Thus, patients in the placebo arm could be considered as having received deferred hormonal therapy. The most common reason for a change in therapy was that the patient was "too ill" to continue with the assigned treatment. In VACURG 1, after 9 years of followup, approximately 67 percent of stage III placebo group patients had received an active therapy, while 100 percent of stage IV placebo group patients changed treatment by this time. In the VACURG 2 trial, information is lacking about the degree to which stage III placebo patients underwent active treatment; however, 40 percent of stage IV patients who received placebo never underwent active hormonal therapy. Given that high proportion of untreated stage IV patients and uncertainty about subsequent treatment of stage III patients, the outcome comparisons of immediate and deferred groups in VACURG 2 must be viewed cautiously.

In each trial, patients in the deferred therapy arms were treated with a wider variety of interventions compared to patients in the immediate therapy arms. In the VACURG 1 trial (Blackard, Byar, and Jordan 1973; Veterans Administration Cooperative Urological Research Group, 1967), placebo patients who progressed mainly received DES or DES plus orchiectomy. The VACURG 2 study (Byar and Corle, 1988; Christensen, Aagaard, and Madsen 1990) did not report subsequent treatment of patients in the placebo arm. In the MRC study (Medical Research Council Working Party Investigators Group, 1997), 72 percent of the patients who had deferred therapy received orchiectomy and 24 percent received an LHRH agonist.

Populations

The MRC and VACURG trials enrolled newly diagnosed, previously untreated patients with histologically confirmed locally advanced or metastatic disease. The MRC trial enrolled patients with "local disease considered too advanced for curative treatment (T2-T4)" or asymptomatic M1 disease (Medical Research Council Working Party Investigators Group, 1997). The VACURG trials, which used an earlier staging system, enrolled patients with stage III or IV disease. Stage III was defined as local extension beyond the prostatic capsule, whereas stage IV was marked by elevated prostatic acid phosphatase or demonstrated metastases.

The VACURG trials have not provided data regarding the comparability of patients in different treatment arms. The MRC paper noted that the randomization process used an algorithm to limit chance differences among groups in age, T stage, and M stage.

Outcomes

Efficacy outcomes, which were measured at 5 years, included overall survival, cancer-specific survival, and time to progression. All three trials reported overall survival; two of the three reported cancer-specific survival; and 1 reported time to progression. In general, reporting of adverse events is sparse in these studies. The VACURG trials reported on cardiovascular deaths related to DES. No studies reporting on quality of life were identified.

An important limitation in most of these studies is incomplete followup of all patients. In the VACURG 1 study, only 44 percent of the patients in the placebo arm had progressed and had changed to an active treatment. There is no information from the VACURG 2 trial regarding the proportion of placebo patients who progressed or what treatments they underwent after progression. In the MRC trial, a more complete proportion of patients in the deferred arm (75 percent) had progressed and received hormonal therapy.

Quality of Study Design and Conduct

All studies were randomized controlled trials. Overall study quality was assessed as described in the Methods section. Studies that were double-blinded and used an intent-to-treat analysis were classified as higher quality. Blinding was considered to be not applicable when the treatment was orchiectomy.

Evidence Table III.1 (in Appendix III) shows whether studies were blinded, whether intent-to-treat analyses were used, and whether withdrawals were documented. Of the three trials, two were assessed as being of higher quality, for purposes of performing sensitivity analyses. However, we judged the total number of studies too small to perform sensitivity analyses, and no sensitivity analysis is included in this meta-analysis.

Interventions

The two largest trials compared radiotherapy plus adjuvant goserelin to radiotherapy alone. The third trial used DES as the adjuvant hormonal therapy, and the fourth trial used orchiectomy.

The Radiation Therapy Oncology Group (RTOG) Protocol 85-31 (Pilepich, Caplan, Byhardt et al., 1997) randomized 977 patients to radiotherapy plus adjuvant goserelin (begun during the last week of radiation therapy and continued indefinitely or until progression) or to radiotherapy alone followed by observation and goserelin deferred until local failure, regional failure, or metastasis was observed. The EORTC trial (Bolla, Gonzalez, Warde et al., 1997) randomized 415 patients to radiotherapy alone or radiotherapy plus adjuvant goserelin begun on the first day of radiotherapy and continued for 3 years. Zagars, Johnson, von Eschenback and coworkers (1988) randomized 82 patients to radiotherapy alone or radiotherapy plus 2 or 5 mg DES begun upon completion of radiation therapy and continued indefinitely or until withdrawal. Granfors, Modig, Damber and coworkers (1998) randomized 91 patients to radiation therapy alone or radiation therapy plus orchiectomy, performed 4 to 5 weeks before radiation therapy began.

Populations

Within each of the four studies, patients in the two arms were well-balanced with respect to prognostic factors. Both the RTOG 85-31 trial ( Pilepich, Caplan, Byhardt et al., 1997) and the EORTC trial ( Bolla, Gonzalez, Warde et al., 1997) enrolled a mix of patients with locally advanced disease (T3/T4) and patients with localized disease who were at high risk for disease progression.

RTOG 85-31 (Pilepich, Caplan, Byhardt et al., 1997) enrolled patients with histologically confirmed T-3 disease or T1-T2 disease with regional lymph node involvement. Patients with bulky primary lesions were excluded (except those with extrapelvic nodal disease). The EORTC trial (Bolla, Gonzalez, Warde et al., 1997) enrolled patients without prior treatment for prostate cancer who had histologically confirmed T3-T4,N0-X disease or T1-T2,N0 grade 3 disease. All patients in the Zagars, Johnson, von Eschenback and coworkers (1988) study had previously untreated stage C disease. Granfors, Modig, Damber and coworkers (1998) selected patients with T1-4, pathologically confirmed N0-3, M0 patients. Patients with negative lymph nodes were excluded if they had early stage disease or had well or moderately differentiated tumor.

Only the EORTC trial (Bolla, Gonzalez, Warde et al., 1997) included patients who had been previously treated. Some patients were included who had undergone radical prostatectomy if they had positive surgical margins or seminal vesicle involvement.

Outcomes

Efficacy outcomes, which were measured at 5 years, included overall survival, cancer-specific survival, and time to progression/disease-free survival. The review of data on time to progression and disease-free survival include information on local control. All four trials reported overall survival; one of the four reported cancer-specific survival; all four trials reported disease-free survival; and two trials reported local control.

Adverse events related to hormonal therapy as an adjuvant to radiation therapy were reported in one trial. No studies reporting on quality of life were identified.

Lack of complete followup is shown in the two largest studies of hormonal therapy as an adjuvant to radiation therapy. The RTOG trial (Pilepich, Caplan, Byhardt et al., 1997) and the EORTC trial (Bolla, Gonzalez, Warde et al., 1997) indicated that 36 percent and 38 percent of patients in the radiation therapy alone arms had progressed and underwent hormonal therapy. The Zagars trial (Zagars, Johnson, von Eschenback et al., 1988) had more complete followup (68 percent of radiation alone patients had progressed), but this trial included only 80 patients. Granfors, Modig, Damber and coworkers (1998) had followup on progression for 61 percent of patients who underwent radiation therapy.

Quality of Study Design and Conduct

All studies were randomized controlled trials. Overall study quality was assessed as described in the Methods section. Studies that were double-blinded and used an intent-to-treat analysis were classified as higher quality. Blinding was considered to be not applicable when the treatment was orchiectomy.

Evidence Table III.1 (in Appendix III) shows whether studies were blinded, whether intent-to-treat analyses were used, and whether withdrawals were documented. Of the four trials, none was assessed as being of higher quality for purposes of performing sensitivity analyses. However, we judged the total number of studies too small to perform sensitivity analyses, and no sensitivity analyses are included in the meta-analysis.

Overall Survival

Two of the three trials found a statistically significant difference in survival overall in favor of immediate therapy (Evidence Table III.2, Appendix III). The VACURG 2 study reported 5-year overall survival of 42 percent for patients who initially received DES 1 mg, compared to 30 percent for patients who were initially assigned to placebo (30 percent). The MRC trial reported that overall survival at 5 years in the immediate therapy group was 41 percent, compared with 37 percent in the deferred therapy group (p=0.02).

A subgroup analysis of the MRC trials found that the advantage for immediate therapy was observed only in M0 patients, who represented 53.5 percent of all patients. No difference between immediate and deferred therapy was found for patients with M1 disease (18.5 percent of patients) or MX disease (28 percent of patients).

The third trial, VACURG 1, found no statistically significant differences between placebo (deferred therapy) and immediate treatment with orchiectomy.

Cancer-Specific Survival

Of two studies that reported cancer-specific survival, one found a statistically significant difference favoring immediate hormonal therapy (Evidence Table III.3, Appendix III). The MRC trial reported 59 percent cancer-specific survival at 5 years in the immediate therapy group and 46 percent in the deferred therapy group (p<0.001). VACURG 1 showed no difference between immediate orchiectomy and placebo.

Time to Progression

Outcomes related to disease progression were reported only for the MRC trial, which found a statistically significant difference in favor of immediate therapy (Evidence Table III.4, Appendix III). Among patients without metastatic disease (about 54 percent of all patients), the time to distant progression was longer in the immediate group. At 5 years, 63 percent of immediate therapy patients were free of distant metastases, compared to 43 percent of deferred therapy patients (p<0.001).

Local progression was also slower in the immediate group, as measured by the number of patients requiring transurethral resection (TURP). Other indicators of progression showed an advantage for immediate therapy, including the frequency of spinal cord compression, ureteral obstruction, and development of extraskeletal metastases.

Adverse Events

Reporting of adverse events data is sparse in these studies. The VACURG studies provide data only on cardiovascular deaths, which were similar in the immediate and deferred treatment arms. The MRC trial did not report data on adverse events related to hormonal therapy.

In general, adverse events would be expected to be those characteristic of the monotherapies. However, patients who undergo immediate treatment will have a longer duration of treatment, during which they experience these adverse events. In the MRC trial, duration of hormonal suppression was 18 months longer for the immediate treatment arm than for the deferred treatment arm.

Overall Survival

Two trials and subgroup analysis of a third trial found a significant improvement in overall survival at 5 years for patients who received immediate adjuvant hormonal therapy (Evidence Table III.2, Appendix III). The EORTC trial (Bolla, Gonzalez, Warde et al., 1997) reported 79 percent survival at 5 years for the group receiving immediate adjuvant hormonal therapy, compared to 62 percent for the deferred therapy group (p=0.001). Granfors, Modig, Damber and coworkers (1998) reported 83 percent survival at 5 years for immediate hormonal therapy, compared with 69 percent for deferred hormonal therapy (p=0.02).

In addition, a subgroup analysis from the RTOG trial on patients with Gleason scores of 8 to 10 found improved survival for the immediate adjuvant therapy group (66 percent at 5 years) over deferred hormonal therapy (55 percent, p=0.03). Granfors, Modig, Damber and coworkers (1998) also performed a subgroup analysis, which showed that the difference was significant in patients with node positive disease, but not in patients with node negative disease.

The RTOG 85-31 trial (Pilepich, Caplan, Byhardt et al., 1997), considering all patients, found no difference between the immediate therapy arm (75 percent at 5 years) versus the received deferred therapy arm (71 percent). The Zagars trial (Zagars, Johnson, von Eschenback et al., 1988) also found no differences between the immediate and deferred therapy arms.

Cancer-Specific Survival

Only the Granfors trial (Granfors, Modig, Damber et al., 1998) reported on cancer-specific survival (Evidence Table III.3, Appendix III). The difference between groups in cancer-specific survival approached statistical significance (p=0.056), favoring the group that received immediate hormonal therapy.

Time to Progression/Disease-Free Survival

Granfors, Modig, Damber and coworkers (1998) reported that the immediate adjuvant hormonal therapy group had significantly better progression-free survival (86 percent at 5 years), compared to the deferred therapy group (73 percent, p=0.005) (Evidence Table III.4, Appendix III). Immediate adjuvant hormonal therapy improved disease-free survival in all three studies that reported that outcome. In the RTOG 85-31 study (Pilepich, Caplan, Byhardt et al., 1997), patients who underwent immediate adjuvant therapy had significantly better disease-free survival at 5 years (60 percent) than did patients who received deferred therapy (44 percent, p<0.0001). The EORTC trial (Bolla, Gonzalez, Warde et al., 1997) reported a similar finding (85 percent versus 48 percent, p<0.001). The difference between groups in the Zagars study (Zagars, Johnson, von Eschenback et al., 1988) was significant only at p=0.05 (66 percent versus 51 percent), possibly due to the study's small sample size.

Related outcomes showed similar advantages for immediate adjuvant therapy, including better rates of local control (RTOG 85-31 and EORTC), a slower rate of progression to distant metastases (RTOG 85-31), and better rates of disease-free survival that accounted for PSA values (RTOG 85-31)

Adverse Events

The RTOG 85-31 trial (Pilepich, Caplan, Byhardt et al., 1997), the Zagars study (Zagars, Johnson, von Eschenback et al., 1988) and the Granfors trial (Granfors, Modig, Damber et al., 1998) reported no data on adverse events related to use of hormonal therapy. The EORTC trial (Bolla, Gonzalez, Warde et al., 1997) reported a significantly higher incidence of grade 1 to 3 incontinence among immediate adjuvant therapy patients (29 percent) than in deferred therapy patients (16 percent, p=0.002).

In general, adverse events related to hormonal therapy would be expected to be those characteristic of the monotherapies. These reports do not provide information about the duration of therapy for either immediate or deferred arms. However, patients who undergo immediate treatment will have a longer duration of treatment during which they experience these adverse events.

Summary

This section compares the outcomes of immediate and deferred androgen suppression in three patient populations. These constitute distinct patient populations and treatment settings. They were considered separately because it is unknown whether findings from one population generalize to another population.

Trials are in progress to compare androgen suppression initiated at PSA rise to androgen suppression deferred until clinical signs or symptoms of progression among patients who have undergone definitive therapy (radical prostatectomy or radiation therapy) for initially localized prostate cancer. No published results are available yet from these trials.

Three trials including a total of 1,209 patients compared primary androgen suppression initiated at diagnosis with primary androgen suppression deferred until clinical progression of disease. The population for these trials consisted of patients with locally advanced (T3/T4) or metastatic (M1) disease, both symptomatic and asymptomatic.

Four trials, including a total of 1,529 patients, compared adjuvant androgen suppression, initiated with radiotherapy and continued for several years or more, to radiotherapy alone followed by androgen suppression at clinical progression. The population for these studies was mainly patients with locally advanced disease, but also included patients with local disease (T1/T2) at high risk for progression in the two larger studies.

Conclusions

Perspective of the Analysis

For whom is a CE analysis conducted? The answer to this question determines the perspective of the analysis. Perspective describes the point of view from which the study is conducted-for example, it might be the patient, the provider, the payer, or the developer of a medical technology. The perspective of the analysis determines which costs are included. An analysis conducted from the perspective of an individual patient might incorporate only the costs that the patient bears directly, such as a 20 percent copayment. An analysis conducted from the perspective of a government program like Medicare would include Medicare payments but ignore deductibles and copayments and other expenses patients bear directly (out-of-pocket payments), as well as payments made on behalf of patients by family members or private supplemental insurance. Neither of these perspectives incorporates all of the resources used to treat the patient. The most widely used and best accepted perspective for conducting CE analysis is the societal perspective, which includes all costs. ²

Specifying the Alternatives

Every CE analysis implicitly or explicitly compares at least two alternatives. The first alternative, and usually the one that motivates the analysis, is the intervention under study. Because cost-effectiveness is relative, the cost-effectiveness of any intervention depends on the alternative to which it is compared. One possible alternative is to "do nothing"; another is standard therapy or an older, but well-studied therapy. "Do nothing" is only an appropriate alternative if it is used commonly or is otherwise justifiable. In the current study we have not included a "do nothing" alternative because androgen suppression is universally regarded as beneficial for patients with advanced prostate cancer. We have included older therapies (such as DES) that have fallen out of favor but are well studied.

After the intervention and the alternative(s) have been selected, the numerator of the cost-effectiveness ratio is calculated by subtracting the cost of the alternative from the cost of the intervention; thus the numerator is the difference in costs between the two interventions, not simply the cost of the intervention under primary consideration. Similarly, the denominator of the cost-effectiveness ratio is the difference between the health effects that result from using the intervention and the health effects with the alternative. In short, a cost-effectiveness ratio is always incremental: the numerator consists of the amount by which health care costs of the patient receiving the intervention exceed the costs of a patient receiving the alternative (incremental costs), and the denominator consists of the amount by which the health outcome expected if a patient is treated with the intervention exceeds the health outcome expected under the alternative (incremental effectiveness).

Measuring Health Effects

To be useful, the measure of health effects or outcomes in a CE analysis must accurately quantify the impact of the interventions on all aspects of health that patients value. Occasionally CE analyses use specialized, highly circumscribed measures of health effects. For example, the cost-effectiveness ratio might be reported as the cost per positive test or the cost per unit change in blood pressure. Such special-purpose effectiveness measures are of limited value unless they are directly linked to "final outcomes" like life expectancy. The effectiveness measure should be sufficiently global to make it possible to compare different kinds of health interventions and sufficiently sensitive to capture changes in health that lead to an appreciable improvement in a patient's sense of well-being.

The most widely used global health outcome measure has been life expectancy, which is readily measured, easily interpreted, and immediately comparable across a wide range of interventions. As a measure of health outcome, though, life expectancy has a significant weakness: it is not designed to detect changes in the quality of life. Thus it fails to quantify the benefits of interventions whose major purpose is pain relief or improvement in functional status. Potential improvement of quality of life is an important reason to treat advanced prostate cancer, which can be highly symptomatic.

The outcome measure that is usually considered most desirable, even though it has not been applied as often as life expectancy, is quality-adjusted life years (QALYs) or, as it is sometimes called, quality-adjusted life expectancy. As its name suggests, this measure is analogous to life expectancy. In calculating life expectancy, each year that an individual lives longer contributes an additional year to life expectancy. In calculating QALYs, each year of additional life does not necessarily add an equal amount. Life years marred by functional limitations, pain, and other burdens associated with illness receive less weight than years in good health. In calculating life expectancy, each additional year receives a weight of 1, but in calculating QALYs, each additional year receives a preference weight of between 0 (for years when health is so bad that it is considered no better than death) and 1 (usually defined as best health imaginable).

Interventions can raise QALYs either by lengthening life or by improving its quality. Similarly, an intervention that lengthens life produces more QALYs if it maintains or improves quality of life than if it adds years of life that are impaired by significant morbidity.

Technical Note

Life expectancy for the general population, adjusted for age, sex, and race, is usually calculated from life tables. Equivalent information for clinically defined groups, such as patients with heart disease, can often be obtained from observational studies or randomized clinical trials. Life expectancy for a 50-year-old woman is the sum of the probabilities that she is alive at age 51, age 52, age 53, and so on. In general, the formula for life expectancy is the sum of the probabilities that an individual will be alive at each age (denoted by i) in the future, up to the maximum life span, or

where F_i is the probability that the person who is now at the "current age" will still be alive at age i.

Quality-adjusted life years are calculated in the same way as life expectancy, with the exceptions that QALYs give less weight to years of life that will occur in the distant future than to years of life that will occur soon (time discounting), and they give less credit to years of life with substantial morbidity than to years of perfect health. To calculate QALYs, it is necessary first to measure "utility" or "preference weights." The preference weight for the health characterizing age i, denoted by qi, can range in value from 0 (for the worst state of health, usually assumed to be death or its equivalent) to 1 (corresponding to "perfect" health). Each such term is the expected value of quality adjustments for all possible states of health at age i. To illustrate the calculation, imagine that individuals alive at age 60 could be in one of only two possible states of health: perfect health, (qh = 1), occurring with probability 0.5, or suffering from heart disease (qd = 0.8), also occurring with probability 0.5. Then q ₆₀, the expected value of the preference weight corresponding to being alive at age 60, is (0.5 X 1) + (0.5 X 0.8) = 0.9. After estimating the value of q_i for each age i, it is possible to calculate the expected number of QALYs according to the formula:

where is a time discount factor whose value is between 0 and 1. As in the formula for life expectancy, F_i is the probability that the person is still alive at age i. If = 1, two years of life in which q_i = 0.33 contribute the same number of QALYs as one year in which q_i = 0.66. If there is no time discounting ( = 1) and if each year of life has perfect health, or quality adjustment is ignored (q_i = 1 for every value of i), this formula reduces to the formula for life expectancy.

Measuring Costs

Most CE analyses use direct costs, or the costs of preventing, diagnosing, and treating health conditions, as the principal measure of resources consumed as a result of choosing a particular health intervention. Indirect costs, or the value of lost wages due to morbidity or premature death, are ordinarily excluded (Gold, Seigel, Russell et al., 1996). Costs include the immediate costs of an intervention (and its alternative) as well as future costs generated by the treatment along with savings due to disease prevented. ³ Thus a full accounting of the costs of a hypertension screening program, for use in a CE analysis, would include the costs of administering the screening test(s), costs of additional diagnostic procedures and treatment administered to individuals found to have hypertension, costs of treatment complications; subtracted from these costs would be the savings resulting from preventing strokes, myocardial infarction, and other complications of hypertension.

In most circumstances, the specific measure of costs that is appropriate to use in CE analysis is the difference between the marginal cost of applying the intervention and the marginal cost of the alternative. Marginal cost refers to the cost of producing an added (small) quantity of a good or service. Usually marginal costs cannot be obtained directly from available accounting data, or from data on payments, reimbursements, or charges for health care. Consequently, CE studies use a wide variety of cost figures. ⁴ For example, available cost measures are often designed to capture average costs, which may not be appropriate because they include both marginal costs and a fraction of the fixed costs of producing health services. There is no consensus about the best practically achievable measure of costs to use. In markets that exhibit competitive behavior, such as competing health plans bidding for hospital or physician services in a large metropolitan market, the price paid for services may approximate marginal costs.

Discounting

Most CE analyses place less weight on future costs and health effects than on immediate effects. The rationale for placing less weight on dollar costs that will be incurred in the future is familiar to anyone who has ever borrowed money or received interest on savings; a dollar today is worth more than a dollar tomorrow. If you borrow $5,000 dollars from a bank today, promising to repay the $5,000 dollars (the principal) and the accumulated interest in 7 years, by the end of the loan period, you might owe twice as much as you borrowed. Similarly, if you know that you will have a large expense in the future (retirement or a child's college costs), you can add to your savings today and let the money accumulate over time. When the time comes to spend the money, you will have the amount you saved as well as accumulated interest on the savings; thus future dollars are appropriately discounted relative to current dollars.

Future health effects are similarly discounted, although the precise rate of discount to use is controversial. In this analysis, we follow the recommendations of the Panel on Cost-Effectiveness in Health and Medicine that health effects be discounted at the same rate as costs (Gold, Seigel, Russell et al., 1996). When resources are being spent to increase health, failing to discount health effects that occur in the future or using a different rate of time discount for health effects than for monetary costs leads to inconsistent and paradoxical policy implications. Note here that we are focusing on societal rather than individual discount rates.

What discount rate should be used? For the base case analysis, we followed the recommendations of the Panel on Cost-Effectiveness in Health and Medicine (Gold, Seigel, Russell et al., 1996) and applied a discount rate of 3 percent to both costs and health effects. This value was estimated as the most appropriate for costs when data on real economic growth and the real consumption rate of interest was taken into account. As outlined above, we applied the same rate to health effects.

Sensitivity Analysis

Aspects of nearly every CE analysis are subject to uncertainty. The magnitude of a treatment effect may not be known with precision, the cost of a test or a drug may be unknown or may vary substantially, and the frequency of an uncommon but severe long-term adverse effect may be highly uncertain. Cost-effectiveness analyses usually include a sensitivity analysis to learn whether the results of the study are sensitive to (i.e., change substantially with) variation in the values of any of several uncertain numbers that are included in the analysis. For example, the frequency of an adverse effect of treatment may have little impact on the cost-effectiveness ratio of a screening test, particularly if the screening test seldom leads to use of the treatment; hence, even if the rate at which adverse effects occur may not be known with certainty, within a plausible range of values adverse effects have little influence on the principal conclusions of the analysis. This use of the term sensitivity is distinct from the sensitivity of a diagnostic test, or the likelihood that the test result will be positive in a patient with the disease. Sensitivity analyses can be used to gauge the likely validity of a CE analysis and to point to areas in which additional data that are more precise are needed.

Objectives of the Cost-Effectiveness Analysis

The CE analysis seeks to address the following questions:

• What is the most cost-effective drug within each class of antiandrogens? What is the most cost-effective androgen suppression therapy, when outcomes are measured in terms of life expectancy? What is the most cost-effective androgen suppression therapy, when outcomes are measured in quality-adjusted life years? How do the cost-effectiveness estimates change when assumptions about the efficacy of antiandrogen strategies change? Are the cost-effectiveness estimates sensitive to variation in the values of other uncertain numbers that are included in the analysis? For patients presenting with regional metastatic prostate cancer, what is the cost-effectiveness of early (when prostate cancer is diagnosed) compared to late (when disease progression occurs or symptomatic distant metastases develop) antiandrogen therapy? How might the cost-effectiveness estimates change if clinical strategies incorporating biochemical markers are included?

Markov Modeling

Figure 13. A Schematic of the Markov Model

   Figure 13. A Schematic of the Markov Model

At baseline, patients have local progression of disease. Death from other causes can occur in any cycle. Patients receiving androgen suppression therapy can have adverse effects that are fatal, severe (with discontinuation of treatment), or minor. Patients can progress to developing distant metabases, which are initially asymptomatic. Patients can then develop symptomatic disease and death from prostate cancer. At all stages, patients can also develop local obstructive symptoms.

Table 20. Summary of Model Assumptions

Table 20. Summary of Model Assumptions

All Models
Base Case
Alternative Efficacy Assumptions a
Timing of Antiandrogen Therapy
Incorporating Biochemical Markers
Choice of initial therapy (radiotherapy or radical prostatectomy) does not affect the course of disease after recurrence, particularly response to hormonal therapy Short-term neoadjuvant therapy does not affect subsequent response to hormonal therapy Patients initially have hormone-sensitive disease No patients with metastatic prostate cancer are cured Patients using antiandrogens will eventually develop hormone resistant disease Transitions between health states occur at a constant rate in the model (except for the transition to death from causes other than prostate cancer, which is age dependant) Local obstruction can occur more than once Each episode of obstruction is an independent event Patients discontinue antiandrogens if they develop severe adverse effects Patients discontinue antiandrogens if they experience disease progression Second line therapy is with ketoconazole Patients using combined androgen blockade (CAB) can switch nonsteroidal antiandrogens if they experience severe adverse effects Adverse effects can be fatal, severe, and dose limiting or minor
The transition from local recurrence to distant asymptomatic disease is 0.116/patient-year (hormone-responsive) (Kuban et al., 1989, 1995b) The transition from distant asymptomatic to distant symptomatic is 0.399/patient-year if prostate cancer is hormone sensitive and receives treatment and 1.2/patient-year if prostate cancer is hormone resistant (regardless of hormone treatment) (Byar, 1973; Byar and Corle, 1988; Cowen et al., 1994; Crawford et al., 1989) The transition from symptomatic distant metastases to death is 0.524/patient-year (all hormone-resistant) (Byar, 1973; Byar and Corle, 1988; Cowen et al., 1994) Local obstruction occurs at a rate of 0.022/patient-year (Fleming, Wassen, Albertson et al., 1993) 20% of patients respond to ketoconazole for an average duration of 3 months (Mahler and Denis, 1995; Small and Vogelzang, 1997) Ketoconazole, when effective, results in a reduced prostate-specific antigen (PSA), which is associated with a 67% reduction in the transition rate for the state in which ketoconazole is started (Smith and Pienta, 1997) Patients using CAB (but not monotherapies) benefit from hormone withdrawal with a reduced PSA and an associated decreased rate of progression (Smith and Pienta, 1997) 35% of patients respond to hormone withdrawal for an average duration of 3.5 months, and no benefit after 12 months (Kelly et al., 1997; Small and Vogelzang, 1997) Minor adverse effects are equally common with all strategies and occur in 55% of patients (literature review) All interventions are equally efficacious-that is, they are associated with similar rates of disease progression (meta-analysis)
Meta-analysis estimates of relative hazards for survival are also estimates of the relative hazard of disease progression when cancer is hormone sensitive Strategies are not equally efficacious Relative hazards for disease progression are: (meta-analysis)
DES	0.9726
Orchiectomy	1 (reference)
NSAA	1.2190
LHRH	1.1124
NSAA + orchiectomy	0.979
NSAA + LHRH	0.943
the base case has stage C disease and enters the model at diagnosis initiating antiandrogen therapy at the time of diagnosis, when asymptomatic distant metastases develop, or when symptomatic distant metastases develop does not increase survival (meta-analysis) antiandrogen therapy prolongs the time spent in the disease state in which it is started
Patients enter the model at their first biochemical failure (e.g., rise of PSA) instead of being identified as having "asymptomatic distant metastases," patients are now identified as having a second failure of PSA the rate of developing a second biochemical failure is 0.14/patient-year (Zietman et al., 1996) the rate of developing symptomatic distant metastases after a second biochemical failure is 0.05/patient-year, yielding an increased time in the model of 2.2 years (Lillis and Thompson, 1996)

a

Results based on previous meta-analysis results. Revised results are slightly different (relative hazard for NSAA+ orchiectomy and for NSAA + LHRH of 0.977 and 0.945, respectively) but do not significantly change the results of the cost-effectiveness analysis.

Figure 13

The Base Case

The model follows a hypothetical cohort of patients with advanced prostate cancer over the course of their remaining lifetimes. The base case is a 65-year-old man with localized recurrence of prostate cancer but no distant metastases who is treated with antiandrogen therapy. We do not model the clinical course of a patient with prostate cancer untreated with antiandrogen therapy, a clinically unacceptable scenario. ⁶ The model's time horizon is 20 years, at which point virtually all patients have died of prostate cancer or other causes. In this model, decisions to initiate or change therapy are based on clinical events only. Decisions based on biochemical progression (such as a rising prostate specific antigen) and optimal timing of antiandrogen therapy are analyzed separately.

The Health States and Transitions

Figure 14. State-cycle Diagram of the Markov Cycle

   Figure 14. State-cycle Diagram of the Markov Cycle

Shown are the basic states in the model. Patients start in the state "Local Recurrence" (State I). If they experience disease progression, they first develop "Asymptomatic Distant Metastases" (State II). Asymptomatic metastases then become "Symptomatic Distant Metastases" (State III); the patients then progress to death from prostate cancer (State IV). At any time, patients may also die of causes other than prostate cancer (State V). Patients are further classified into states depending on the medication taken (first-line, second-line, or none) and the presence or absence of an adverse effect. Transitions between states are shown in dark numbered circles: (1) from local recurrence to asymptomatic distant metastases, (2) remaining in the local recurrence state, (3) from asymptomatic distant to symptomatic distant metastases, (4) remaining in the asymptomatic distant metastases state, (5) from symptomatic distant metastases to death, (6) remaining in the symptomatic distant metastases state, and (7) from any state to death from other causes.

Figure 14

Figure 14

Events Prior to Entry into the Model

The base case pertains to a patient previously diagnosed with localized (clinical stage A or B) and moderately differentiated (Gleason stage 5 or 6) prostate cancer. The patient previously received definitive local treatment with either radiotherapy or radical prostatectomy. Although the choice of initial therapy may affect the chances of developing recurrence, the model builds from an assumption that the choice does not affect the course of disease after recurrence, particularly the response to hormonal therapy. We do not include patients who opted for watchful waiting when initially diagnosed. We further assume that any short-term use of neoadjuvant hormonal therapy during the initial treatment for prostate cancer does not affect responses to antiandrogen therapy for recurrent disease.

Strategies in the Model

We evaluate four antiandrogen strategies: bilateral orchiectomy, an LHRH agonist (leuprolide, goserelin, and buserelin), diethylstilbestrol (DES), and nonsteroidal antiandrogens (NSAA, flutamide, bicalutamide, and nilutamide). We first present an analysis of the choice of LHRH agonist and of nonsteroidal antiandrogen. We then present an analysis of choice between agents, including combined androgen blockade. Not included in this model is a strategy of intermittent androgen suppression.

Further Assumptions of the Natural History Model

We assume that patients would initially have hormone-sensitive disease, manifest by a response to androgen suppression. We based this assumption on the observation that mutations in the androgen receptor gene that confer "hormone resistance" are rare in prostate specimens from patients who have not received prior androgen therapy (Bubley and Balk, 1996). However, patients eventually develop hormone-resistant disease (Garnick, 1997). We assume that new metastatic disease defines the development of hormone resistance. Further, all deaths attributable to prostate cancer occur in patients whose disease is no longer amenable to hormone manipulation.

Details of Health States in the Model

Figure 14

Figure 14

Health states are further defined according to the antiandrogen therapy used (first-line, second-line, or none) and adverse effects from treatment (present or absent). Note that not all combinations exist; for example, patients stop active therapy when they develop asymptomatic distant metastases. A complete listing of the health states in the base case model follows:

• Local recurrence, active therapy

• Local recurrence, active therapy with minor adverse effects

• Local recurrence, second-line therapy

• Local recurrence, no therapy

• Asymptomatic distant metastases, second-line therapy

• Asymptomatic distant metastases, no therapy

• Symptomatic distant metastases, second-line therapy

• Symptomatic distant metastases, no therapy

• Dead from prostate cancer

• Dead from other causes

In addition, the following states are included when combined androgen blockade strategies are considered. We have included states defined by the use of an alternative active therapy and by drug withdrawal (see "Drug Sequence"):

• Local recurrence, alternative active therapy

• Local recurrence, alternative active therapy with minor adverse effects

• Local recurrence, drug withdrawal

• Asymptomatic distant metastases, alternative active therapy

• Asymptomatic distant metastases, alternative active therapy with minor adverse effects

• Asymptomatic distant metastases, drug withdrawal

• Symptomatic distant metastases, alternative active therapy

• Symptomatic distant metastases, alternative active therapy with minor adverse effects

• Symptomatic distant metastases, drug withdrawal

Note that two clinical events, the occurrence of severe adverse effects and the development of local obstructive symptoms, are not modeled as separate states. Rather, the costs and quality of life effects associated with these events are captured in the transitions between states.

Details of Transitions Between Health States

13

14

Table 21. Summary of Transition Rates Under Different Scenarios (more...)

Table 21. Summary of Transition Rates Under Different Scenarios

Base Case and Alternative Efficacy Assumptions
From health state	Local recurrence	Asymptomatic distant Metastases	Symptomatic distant metastases
To health state	Asymptomatic distant metastases (with treatment)	Symptomatic distant metastases (hormone-resistant)	Death (hormone-resistant)
Rate (number/patient-year)	0.116	1.2	0.524
Timing of Antiandrogen Therapy
Therapy started at diagnosis
From health state	Diagnosis with Stage C prostate cancer	Asymptomatic distant metastases	Symptomatic distant metastases
To health state	Asymptomatic distant metastases (with treatment)	Symptomatic distant metastases (hormone-resistant)	Death (hormone-resistant)
Rate (number/patient-year)	0.066	1.2	0.524
Therapy started when asymptomatic distant metastases develop
From health state	Diagnosis with Stage C prostate cancer	Asymptomatic distant metastases	Symptomatic distant metastases
To health state	Asymptomatic distant metastases (no treatment)	Symptomatic distant metastases (hormone-sensitive)	Death (hormone-resistant)
Rate (number/patient-year)	0.0875	0.399	0.524
Therapy started when asymptomatic distant metastases develop
From health state	Diagnosis with Stage C prostate cancer	Asymptomatic distant metastases	Symptomatic distant metastases
To health state	Asymptomatic distant metastases (no treatment)	Symptomatic distant metastases (no treatment)	Death (hormone-sensitive)
Rate (number/patient-year)	0.0875	1.2	0.260
Incorporating Biochemical Markers
From health state	First biochemical failure (PSA rise)	Second biochemical failure (PSA rise)	Symptomatic distant metastases
To health state	Second biochemical failure (hormone-sensitive)	Symptomatic distant metastases (hormone-resistant)	Death (hormone-sensitive)
Rate (number/patient-year)	0.14	0.05	0.524

Local Recurrence to Asymptomatic Distant Metastases

We first estimated the transition rate from local recurrence to asymptomatic distant metastases (Transition 1 in Figure 14). The base case estimate of this transition rate is derived from a study of recurrent prostate disease after definitive radiation therapy (Kuban, El-Mahdi, and Schellhammer 1989, 1995b). In this study, clinical failure was defined as progressive prostate induration, nodularity, increasing size or asymmetry, or obstructive symptoms. We estimate the transition rate to distant metastases to be approximately 0.116/patient-year for patients receiving antiandrogen therapy.

Data for a Markov model that represented progression from stage C to stage D1 and D1 to D2 disease came from another analysis that pooled natural history studies (Cowen, Chartrand, and Weitzel, 1994). Note that the definition of disease states in this model differs from those used in our model. Nevertheless, these estimates can provide a range for use in sensitivity analysis. We calculate the transition rate from stage C to stage D2 (including both patients using and not using antiandrogens) to be 0.192/patient-year (range 0.081 to 0.294). Restricting the analysis to studies in which patients received treatment, the rate is 0.066/patient-year. As expected, this rate is lower than the baseline estimate used in our model because it reflects the experience of patients with initial, rather than recurrent, disease. We used this estimate as the lower boundary for the range in sensitivity analyses. Note that we also used this estimate in our model analyzing the optimal timing of antiandrogen therapy (see Timing of Anitandrogen Therapy)

We assumed that patients who do not experience disease progression and do not die of causes other than prostate cancer would remain in the same health state in the next cycle (transition 2 in Figure 14).

Asymptomatic to Symptomatic Distant Metastases

We assumed that all patients who developed distant metastases would experience an initial period in which the metastases were asymptomatic (State II in Figure 14) before progressing to symptomatic distant metastases (State III in Figure 14). The occupancy time in this state is difficult to estimate. Most natural history studies have estimated a median overall survival after the development of metastases of 2 to 3 years (Robson and Dawson, 1996), but some followed patients from the development of asymptomatic metastases whereas others followed patients only after the metastases became symptomatic.

For the base case estimate of this transition probability (transition 3 in Figure 14), we used the results of a randomized controlled trial of antiandrogens in patients with stage D2 disease. We focused on the subgroup with "minimal" disease and good performance status (to approximate asymptomatic distant metastases) who were treated with a single antiandrogen (to approximate the base case) (Crawford, Eisenberger, MacLeod et al., 1989). Note, however, that we assumed that patients in our model who enter this state have hormone-resistant disease; thus, we expect their transition rate to be higher than patients in this clinical trial. To correct this estimate (and thus account for the development of hormone resistance), we assumed that the relative hazard comparing patients with hormone-resistant and hormone-sensitive disease was constant and adjusted the calculated transition rate for patients responding to treatment by this relative hazard.

After adjustment for the expected mortality due to other causes for the average patient in the clinical trial cohort (Black, Nease, and Welch, 1996), the estimated transition rate for the patients described above (who responded to therapy) was 0.399/patient-year. This estimate may be imprecise because it is based on a subgroup analysis of a single study. To account for this uncertainty, we used a wide range in the sensitivity analysis (0.200 to 0.800/year).

To calculate the relative hazard of disease progression comparing patients with hormone-resistant and hormone-sensitive disease, we estimated the progression rates from stage III to IV in the placebo and treatment arms of the first VACURG study (Byar, 1973; Byar and Corle, 1988). We made the simplifying assumption that a similar relative hazard applied to the hormone-resistant and hormone-sensitive states as defined in our model. The resulting relative hazard of disease progression for hormone-resistant disease was calculated as 3.0. This result was similar to the estimate when pooled rates for disease progression from stage C to D2 were compared in untreated and treated patients (2.8) (Cowen, Chartrand, and Weitzel, 1994). Thus, the final estimate for this transition rate used in the model was 1.2/patient-year (0.399 X 3.0). This relative hazard results in an estimated median prolongation of asymptomatic status of 14 months for patients with hormone-responsive disease, in accord with reported results.

We assumed that patients with asymptomatic distant metastases who do not experience disease progression and do not die from other causes remain in this health state (transition 4 in Figure 14).

Symptomatic Distant Metastases to Death

We estimated the rate of progression from symptomatic distant metastases to death (transition 5 in Figure 14) from natural history studies (Byar, 1973; Byar and Corle, 1998; Cowen, Chartrand, and Weitzel, 1994). We estimated this rate to be 0.524/patient-year for patients with hormone-resistant disease and used a range of 0.250 to 0.900 in the sensitivity analysis. We compared our estimates of mortality after a patient develops metastatic disease to survival curves from a meta-analysis of androgen blockade in prostate cancer and found a close fit (Prostate Cancer Trialists' Collaborative Group, 1995).

Death from Other Causes

At any time in the model, patients could die from causes other than prostate cancer (transition 7 in Figure 14). This transition could occur from any state. Note that these possible causes of death included dying from fatal adverse effects of antiandrogen therapy (see "Adverse Effects").

Events Occurring During Transitions

Figure 15. Details of Transitions Between States

   Figure 15. Details of Transitions Between States

Five illustrative states are shown: local recurrence, active therapy (corresponding to State I in Figure 14); asymptomatic distant metastases, second-line therapy (State II in Figure 14); local recurrence, second-line therapy (also State I in Figure 14); local recurrence, adverse effect (also State I in Figure 14); and dead from other causes (State V in Figure 14). Transitions between states depend on clinical events, as indicated in the boxes. Transition A corresponds to transition 1 in Figure 14; transitions B, C, D, and E correspond to transition 2 in Figure 14; and transition F corresponds to transition 7 in Figure 14.

We modeled two clinical events that do not define separate states but whose impact is captured in the transition between states (Figure 15). First, if a patient develops severe adverse effects necessitating discontinuation of medical antiandrogen therapy, he incurred a cost and a transient loss in quality of life. Further details about adverse effects modeling are presented below. Second, if a patient develops local obstructive symptoms, he also incurred a one-time added cost and a transient decrease in quality of life.

To clarify these possible transitions, we first present an example. Figure 15 details the possible transitions from the state "local recurrence, active therapy" for a patient treated with monotherapy. If patients experience disease progression, they transition into the state "asymptomatic distant metastases, second-line therapy" (transition A in Figure 15). Note that patients may experience local obstruction, a severe adverse effect, or a minor adverse effect in the same cycle as they experience disease progression. Note also that patients start second-line therapy after disease progression (see "Changes in Antiandrogen Therapy").

Patients may also experience local obstruction without developing distant metastases (transition B), possibly with an adverse effect. We assumed that patients would change medications after developing local obstructive symptoms. Similarly, if patients develop severe adverse effects from therapy, they start second-line therapy (transition C). Patients who experience only mild adverse effects continue with their medications; hence they move to the state "local recurrence, adverse effect" (transition D). If patients experience neither obstructive symptoms nor any adverse effects, they remain in their initial state (transition E). Finally, patients may die of causes other than prostate cancer (transition F). To understand the model fully, it may be useful to directly compare Figure 15 to Figure 14. Transition A corresponds to transition 1 in Figure 14. Transitions B, C, D, and E correspond to transition 2 in Figure 14 (indicating that several events may still occur even though patients remain in the same health condition), and transition F corresponds to transition 7 in Figure 14.

Note that we assumed that these were independent events. For example, a patient may move from the state "local recurrence, active therapy" to the state "asymptomatic distant metastases, no therapy" because two events have occurred-he has developed disease progression and he has experienced a severe adverse effect. To continue with the example, let us suppose that this patient also had local obstruction, as in transition A in Figure 15. The probability of this transition is the product of the probability of developing asymptomatic distant metastases, the probability of experiencing a severe adverse effect, and the probability of having an episode of local obstruction.

Changes in Antiandrogen Therapy

We assumed that patients would change therapy for either of two reasons: they experienced a severe, dose-limiting adverse effect, or clinical signs or symptoms indicated that their prostate cancer was no longer sensitive to hormonal intervention. In the model, patients switched therapies if they developed distant metastases, if asymptomatic distant metastases became symptomatic, or if they developed local obstructive symptoms.

Adverse Effects

The frequency, type, and severity of associated toxicities differ among treatment strategies. Our analysis distinguishes among fatal adverse effects (such as fulminant hepatic failure from nonsteroidal antiandrogens), severe adverse effects that require discontinuation of treatment (such as angina from diethylstilbestrol), and bothersome but tolerable minor adverse effects (such as hot flushes with all treatments). We assume that fatal adverse effects can occur at any time. We estimate that the rate of fatal hepatic failure with NSAAs is 0.0003/100 patient-year (Wysowski and Fourcroy, 1996). The rate of excess cardiac death from 5 mg DES is 0.03/patient-year and from 1 mg DES is 0.01/patient year (Byar, 1973; Byar and Corle, 1988).

Table 22. Frequency of Stopping Antiandrogen Medication (more...)

Table 22. Frequency of Stopping Antiandrogen Medication Due to Adverse Effects

Strategy	Proportion Discontinuing Therapy
DES 1 mg/d	14.3%
DES 3 mg/d	18.7%
Leuprolide 1 mg/d	0.0%
Goserelin 3.6 mg/month	2.0%
Goserelin 10.8 mg every 3 months	1.3%
Buserelin 0.4 mg/d	4.2%
Flutamide 750 mg/d	9.8%
Bicalutamide 50 mg/d	4.0%
Nilutamide 300 mg/d	6.8%

We assume that all the drugs can cause severe adverse effects, but they occur only in the first month of treatment. We assume that severe adverse effects are associated with a one time incremental cost and decrease in quality of life. Minor adverse effects, however, are associated with an incremental cost and loss of quality of life that last for the duration of drug use. Since DES is associated with thromboembolic and cardiovascular complications, we conduct a sensitivity analysis to test the assumption that severe adverse effects with DES are associated with higher costs and morbidity than other agents. In the base case, we assume that the risk of developing adverse effects with bicalutamide is independent of the risk of developing adverse effects to nilutamide. To estimate adverse effect rates, we pool the incidence of adverse effects across studies included in the meta-analysis that reported toxicities. The pooled estimates of drug withdrawal for each agent are shown in Table 22 .

We assume that minor adverse effects were associated with all agents, consisting primarily of effects of androgen deprivation such as hot flushes, and occurring with equal frequency (55 percent) in all strategies.

Efficacy of Antiandrogen Therapy

We assume that all interventions are as efficacious as orchiectomy in delaying disease progression when patients are on active therapy. We distinguish between efficacy and effectiveness. ⁸ We use efficacy in a narrow sense to indicate only how the various strategies influence disease transition rates (the impact of the therapies on the natural history of prostate cancer). In contrast, we use effectiveness as a broad term that incorporates not only effects of drugs on disease progression, but also effects on mortality from other causes (such as with DES), differing toxicities, and differing effects on quality-of-life rates (the impact of the therapies on the net health benefits).

The estimates of the efficacy of combined androgen blockade deserve special attention for several reasons. For example, these estimates are based on clinical trials in which this strategy may have been used much as monotherapies are typically used. However, combined androgen blockade including a nonsteroidal antiandrogen has two potential advantages over monotherapies that may not be captured in the meta-analysis: first, patients who are intolerant of one agent can start another, and second, a brief window exists after drug withdrawal during which the risk of disease progression is decreased. We repeated our analysis with and without incorporating these advantages.

We used several different methods to estimate the efficacy of combined androgen blockade. For the base case, we assumed that combined androgen blockade is not more effective than monotherapies because the current meta-analysis of 2-year survival (grouped by specific strategy) suggested that there was no significant difference between strategies. Nevertheless, the point estimates from the meta-analysis did suggest a trend toward decreased disease progression with combined androgen blockade. Therefore, we repeated our analysis using these efficacy estimates.

Cost Estimates

Because we use the societal perspective to estimate costs, all costs associated with the treatment strategy, regardless of who bears them, are considered. As in other contexts, the costs may vary across settings. For example, the price a pharmacy pays to acquire is one "cost," whereas the amount it charges an individual patient is another, and the price paid by a health plan or pharmacy benefits manager to a manufacturer is another. There may be geographic variation in each of these measures of drug costs. Although there is some ambiguity about which of these costs is the appropriate measure to use, and all may be at best rough proxies for the conceptually preferred measure of long-run marginal cost (Gold, Seigel, Russell et al., 1996), many CE analyses use the final prices charged for such a product or service. Nevertheless, variation in pricing means that the cost figures used in a CE analysis do not apply to every purchaser of the technology under study. Thus, it is particularly important to assess the consequences of cost variation for estimated cost-effectiveness.

In this analysis, we attempt to approximate the price paid by the health care system, payer, or patient as the cost of the new technologies. Although this might appear to be a "health care system" perspective, it is actually a societal perspective that includes costs borne by patients and others in addition to those of the payer or health system.

Table 23. Model Inputs

Table 23. Model Inputs

Variable	Base Case Value	Reference
Cost (annual) 9		Drug Topics Red Book, 1997
Estrogen
DES 1 mg/d	$36
DES 3 mg/d	$109
DES 5 mg/d	$96
Nonsteroidal antiandrogen
Flutamide 750 mg/d	$3,694
Nilutamide 150 mg/d	$2,842 10
Bicalutamide 50 mg/d	$3,890
LHRH agonist
Leuprolide 22.5 mg every 3 months	$6,282
Leuprolide 3.75 mg/month	$5,064
Leuprolide 1 mg/d	$22,356
Goserelin 10.8 mg every 3 months	$4,995
Goserelin 3.6 mg/month	$4,995
Second-line therapy
Prednisone 10 mg/d	$18
Ketoconazole 1200 mg/d	$7,043
Orchiectomy (per operation)	$3,360	Hillner et al., 1995
Cost of disease states (annual)
Local recurrence	$320	Hillner et al., 1995; Krahn et al., 1994; Taplin et al., 1995 Estimates
Distant, asymptomatic	$320
Distant, symptomatic	$410
Additional cost of living with a adverse effect 11	$30
Cost of preterminal care	$17,000
Quality-of-life adjustments
Local recurrent disease	0.92	Albertsen et al., 1997; Bennett et al., 1996; Fleming et al., 1993Krahn et al., 1994; Patrick and Erickson 1993; Expert opinion
Local recurrent disease with orchiectomy	0.92
Distant asymptomatic disease	0.9
Distant symptomatic disease (hormone responsive)	0.8
Distant symptomatic disease (hormone resistant)	0.4
Adjustment for living with minor adverse effect (multiplier) 12	0.85
Costs and quality-of-life adjustments incurred during transitions between states (per episode) 13
Cost of a severe adverse effect	$150	Krahn et al., 1994
Cost of local obstruction	$4,830	Coley et al., 1997b
Adjustment for local obstruction	- 0.10	Albertsen et al., 1997; Bennett et al., 1996; Fleming et al., 1993Krahn et al., 1994; Patrick and Erickson 1993
Adjustment for severe adverse effect	- 0.10

9

Costs of combined androgen blockade are calculated by adding together the therapies used.

10

Nilutamide is used at a dose of 300 mg/d for the first month of therapy.

11

If patients are living with a minor adverse effect, the cost of care for their health state is increased by $30.

12

If patients are living with a minor adverse effect, the quality of life for their disease state is multiplied by 0.85 (i.e., a 15% decrement).

13

These costs are incurred only once at the time that the event (a severe adverse effect or local obstruction) occurs.

Taplin et al., estimated that the 3-month cost of "continuing care" for prostate cancer patients with "local" disease was $1,277 and for "regional" disease was $1,356 (Taplin, Barlow, Urban et al., 1995). Costs were estimated at the Group Health Cooperative of Puget Sound and included materials, disposable goods, physician and other salaries, administrative overhead, inpatient costs, and hospice and visiting nurse services. However, these costs include treatment with antiandrogens, which we modeled as a separate cost. Assuming that the 3-month cost of antiandrogen therapy is approximately $900 to $1,250 (the costs of nilutamide and goserelin, respectively), the estimates of these treatment costs are similar to those reported by Krahn et al. (1994).

The other cost included during the transition between states was the cost of a local obstructive event. We estimated the cost of treating an obstruction from Coley and coworkers' estimate of the cost of transurethral prostatectomy for local obstruction, which was $4,830 (Coley, Barry, Fleming et al., 1997b) (see Table 23).

Quality-of-Life Adjustments

As noted previously, QALYs can be calculated by multiplying the time spent in different states by the utility or value assigned to those states and adding the products together. Although in theory measures like QALYs make it possible to compare treatments that have disparate effects (for example, one primarily affects survival, and the other affects morbidity), in practice the quality-of-life measures used to construct QALYs often fall short. Frequently quality-of-life measures are disease-specific and emphasize the sensitivity of the measurement instrument over the ability to generalize from its results to societal preferences.

Among the common symptoms of metastatic prostate cancer and its treatment are impotence, loss of sexual desire, appetite and weight loss, pain, fatigue, diminished social function, impaired mental health, and less frequently, nausea and vomiting, hot flashes, hair loss, headaches, diarrhea, and breast tenderness (Denis 1995; Garnick 1997; Lucas, Strijdom, Berk et al., 1995; Presant, Soloway, Klioze et al., 1987). A small number of studies have estimated how patients value different health states associated with advanced prostate cancer and with hormonal treatment for prostate cancer. Although the quality of life of patients with asymptomatic disease is similar to that of the general population, quality of life decreases with development of progressive disease. Causes of diminished quality of life include both the disease and its treatment. Although several studies have reported changes in disease-specific measures of quality of life, none of them estimated the decrease in quality of life associated with undergoing surgical or medical castration or the quality of life associated with palliative measures such as the administration of prednisone.

A commonly used global quality-of-life measure is the SF-36 questionnaire (Patrick and Erickson, 1993). A study using this instrument reported that patients with metastatic prostate cancer in remission had a quality of life comparable to the general quality of life in the United States for men of the same age (Albertsen, Aaronson, Muller et al., 1997). The SF-36, like other so-called "statistically weighted" measures, assigns the same weight or value to all items in the questionnaire. In contrast, measurements of "utility" or "preferences" for health states, also called preference weighted measures, use observed preferences for different health states to weight items differently. Although statistically weighted measures are usually considered inappropriate for cost-effectiveness analysis (Gold, Seigel, Russell et al., 1996), results derived with statistically weighted measures and preference weighted measures tend to follow the same trends (Patrick and Erickson, 1993).

Not surprisingly, physician estimates of patient's likely tradeoffs between time without symptoms and stable prostate cancer without treatment-related toxicity produced a lower utility estimate of 0.92. This means that the physicians believed that patients would equally value 1 year with stable or asymptomatic metastatic prostate cancer and 0.92 years of disease-free life (Bennett, Matchar, McCrory et al., 1996). Although this represents an important decrease in quality of life, it is less than that for moderate angina or the need to use a walking stick for ambulation (Nord, 1992). It is also important to note that it is lower than the estimates used for a similar state by The Prostate Patient Outcomes Research Team (Fleming, Wasson, Albertsen et al., 1993). In the following analysis, we use the best available data to estimate quality-of-life weights, preferring patients' preferences to that of proxies and preference measures to statistically weight measures. Where the less desirable measures are the only ones available or estimates are unreliable, we include a wide range for the values in sensitivity analyses.

There is a consistent decrease in quality of life with increasingly severe disease. After separating SF-36 results for patients in remission into minimal and extensive disease based on the number and location of bone metastases, patients with minimal disease in remission had significantly better quality of life in two dimensions, social functioning and mental health, compared to patients with extensive disease in remission (Albertsen, Aaronson, Muller et al., 1997). Similarly, the development of gastrointestinal adverse effects due to flutamide was associated with a utility of 0.83 (Bennett, Matchar, McCrory et al., 1996), despite the caveat that these adverse effects were not severe enough to warrant discontinuation of treatment.

Utility scores decline as prostate cancer advances. Early progression of metastatic cancer was associated with a utility of 0.83, the same as that associated with stable metastatic disease with gastrointestinal adverse effects due to flutamide (Bennett, Matchar, McCrory et al., 1996). Late progression of disease was associated with a much greater decrease in utility, to 0.42 (Bennett, Matchar, McCrory et al., 1996). Patients with progressive metastatic prostate cancer, defined as continuing increase in PSA levels of 4 ng/ml, had a lower overall level of quality of life compared to a comparable population based on SF-36 results. Patients with progressive disease had significantly worse (p<0.05) scores on four of eight dimensions including bodily pain, vitality, social functioning, and mental health compared to all patients in remission. Patients with progressive disease also had significantly worse (p<0.05) scores on the bodily pain dimension compared to patients with extensive disease in remission (Albertsen, Aaronson, Muller et al., 1997). In this study, patients in remission and with progressive disease had received treatment with an LHRH agonist and flutamide, although the requirements for inclusion base on duration of treatment varied across the two groups. Unfortunately, it is unclear to what extent these decrements in quality of life result from adverse effects of different treatment strategies.

The decrease in quality of life accompanying disease progression is consistent with extrapolations from the Quality of Well-Being (QWB) Index, a preference weighted measure of health-related quality of life. For example, if disease in remission causes primarily impairments in social and physical activity, the QWB would assign a 6.1 percent reduction in quality of life compared to good health. More severe symptom groupings characteristic of symptomatic advanced disease, such as general tiredness or weight loss, urinary dysfunction, gastrointestinal upset, and hot spells, are associated with decrements of 24.4 percent to 29.2 percent in the QWB Index (Kaplan and Anderson, 1988; Patrick and Erickson, 1993.)

We make several assumptions regarding quality of life. First, we assume that orchiectomy is not associated with a lower quality of life than antiandrogen drug therapy. Because the utility attached to orchiectomy may vary greatly among patients with prostate cancer, we tested a wide range of alternative values for this value in sensitivity analyses. Second, we assume that all patients respond initially to antiandrogen therapy. Third, we assume that there is no decrement in quality of life for patients who use antiandrogens and do not experience adverse effects; thus patients who stop medications do not experience an improved quality of life. Fourth, we assume that the quality of life with symptomatic distant metastases was relatively high while receiving antiandrogen therapy (including second-line therapy and hormone withdrawal) but fell to a low level once all hormonal manipulations were exhausted.

Timing of Antiandrogen Therapy

Figure 14

Figure 14

Figure 14

Incorporating Biochemical Markers

Figure 14

Choice of Agent

We first identify the most economically attractive agent of the LHRH agonists and the nonsteroidal antiandrogens. The results of the meta-analysis suggest that the efficacy of the major antiandrogen treatment strategies is similar. Furthermore, the pooled adverse effect results suggest that all of the LHRH agonists have similar adverse effect profiles. In addition, the nonsteroidal antiandrogens have similar adverse effect profiles. We therefore used a cost-minimization analysis to identify the least expensive agent in each group.

We first consider the LHRH agonists. Not including administration costs, leuprolide is less expensive when administered every month rather than every 3 months ($5,064 versus $6,282 in annual drug costs). Even assuming a $25 charge per administration, the more frequent schedule is more advantageous ($5,364 versus $6,382). Buserelin is not yet commercially available in the United States. Compared to the least costly leuprolide regimen ($5,064), goserelin is marginally less expensive ($4,995). Because less frequent dosing results in lower costs for administration, the most economically advantageous LHRH agonist is goserelin administered every 3 months.

We next consider the nonsteroidal antiandrogens. Although monotherapy with a nonsteroidal antiandrogen is not considered the standard of care in the United States, this strategy has been extensively studied and we include it for completeness. Flutamide is marginally less expensive than bicalutamide ($3,694 versus $3,890), both of which are more expensive than nilutamide ($2,842). The cost of nilutamide in the first year is increased because nilutamide is given in higher doses during the first month of therapy (300 mg/d), but is still the least expensive NSAA ($3,079). Thus, the most economically advantageous nonsteroidal antiandrogen is nilutamide.

Comparison of Monotherapies

Figure 16. Lifetime Costs and Effectiveness of Each (more...)

   Figure 16. Lifetime Costs and Effectiveness of Each Strategy

Each dot represents the discounted lifetime costs and effectiveness of each therapy. Effectiveness is presented before and after adjusting for quality of life (top and bottom respectively). The inverse of the slope connecting any two lines yields the marginal cost-effectiveness ratio of the two strategies. Note that the y-axis in this and the remaining figures do not start at zero. Marginal cost-effectiveness is for comparison to immediately preceding nondominated alternative.

Strategy	Cost	Effectiveness (life years)	Marginal Cost-Effectiveness
DES	$3,553	5.96
Orchiectomy	$6,957	6.52	$6,106 / QALY
NSAA	$15,688	6.32	Eliminated by strict dominance
NSAA + orchiectomy	$20,701	6.49	Eliminated by strict dominance
LHRH agonist	$27,013	6.50	Eliminated by strict dominance
NSAA + LHRH agonist	$40,342	6.48	Eliminated by strict dominance

Figure 16. Lifetime Costs and Effectiveness of Each (more...)

   Figure 16. Lifetime Costs and Effectiveness of Each Strategy

Each dot represents the discounted lifetime costs and effectiveness of each therapy. Effectiveness is presented before and after adjusting for quality of life (top and bottom respectively). The inverse of the slope connecting any two lines yields the marginal cost-effectiveness ratio of the two strategies. Note that the y-axis in this and the remaining figures do not start at zero. Marginal cost-effectiveness is for comparison to immediately preceding nondominated alternative.

Strategy	Cost	Effectiveness (QALYs)	Marginal Cost-Effectiveness
DES	$3,553	4.64
Orchiectomy	$6,957	5.10	$7,453 / QALY
NSAA	$15,688	4.92	Eliminated by strict dominance
NSAA + orchiectomy	$20,701	5.05	Eliminated by strict dominance
LHRH agonist	$27,013	5.08	Eliminated by strict dominance
NSAA + LHRH agonist	$40,342	5.03	Eliminated by strict dominance

Figure 16

Figure 16

In contrast to the very similar survival estimates, the lifetime costs of the strategies differ markedly. The lowest lifetime cost, about $4,100, is associated with DES therapy. The cost of orchiectomy is $7,500, whereas the cost associated with monotherapy with an NSAA is $18,000 and for treatment with an LHRH agonist is $30,900. After a time discount rate of 3 percent is applied, these estimates are $3,600, $7,000, $15,700, and $27,000, respectively.

Figure 16

DES is the least expensive option. Orchiectomy is both slightly more effective and more expensive. The incremental cost-effectiveness of orchiectomy relative to DES is $6,100 per life year gained and $7,500 per QALY. Monotherapy with LHRH agonists or NSAA is dominated by other strategies both before and after quality-of-life adjustments are applied. Thus, orchiectomy is the most cost-effective strategy.

Combined Androgen Blockade

We compared the effects with combined androgen blockade (CAB) to monotherapies. We compared two regimens for which we had cost and toxicity data-nilutamide plus goserelin and nilutamide plus orchiectomy. The model predicts that both regimens are associated with undiscounted crude survivals of 7.5 years, discounted survival of 6.5 years, and discounted quality-adjusted survivals of 5.0 quality-adjusted life years respectively. The undiscounted lifetime costs of each therapy are $46,200 and $23,200 and the discounted lifetime costs are $40,300 and $20,700, respectively. Of course, if CAB is no more efficacious than monotherapies, then CAB cannot be more cost-effective (they are dominated strategies). If CAB is slightly more effective, it is no longer dominated by other strategies but the cost-effectiveness ratios are very high (see "Sensitivity Analysis"). These estimates include potential advantages to using combined androgen blockade due to increased therapeutic choices and benefits from drug withdrawal.

Alternative Efficacy Assumptions

Figure 17. Costs and Outcomes With All Monotherapies (more...)

   Figure 17. Costs and Outcomes With All Monotherapies and Combined Androgen Blockade

Assumptions about efficacy reflected the point estimates from the meta-analysis.
DES	$3,567	4.68
Orchiectomy	$6,957	5.10	$8,092 / QALY
NSAA	$14,382	4.61	Eliminated by strict dominance
NSAA + orchiectomy	$20,857	5.08	Eliminated by strict dominance
LHRH agonist	$25,709	4.89	Eliminated by strict dominance
NSAA + LHRH agonist	$41,443	5.13	$1,109,581 / QALY

Figure 17

Combined androgen blockade with nilutamide and orchiectomy has a discounted lifetime cost of $20,900 and a quality-adjusted survival of 5.08 QALYs. This strategy is eliminated by strict dominance by orchiectomy. Combined androgen blockade with nilutamide and goserelin is the most expensive therapy ($41,400 discounted lifetime costs) but is also associated with the highest quality-adjusted life expectancy (5.13 QALYs). Compared to the alternative CAB regimen, the cost-effectiveness of this strategy is $413,300 per QALY. Compared to orchiectomy, the cost-effectiveness of this CAB strategy is $1,110,000 per QALY.

We repeated the analysis assuming greater efficacy for both combined androgen blockade regimens. We first assumed that the relative risk for disease progression was 0.86, based on the meta-analysis results of 5-year survival when all CAB strategies were grouped together. Under these assumptions, the cost-effectiveness ratio of NSAA plus orchiectomy relative to orchiectomy alone was $74,800/QALY. The cost-effectiveness ratio of NSAA plus LHRH agonist relative to orchiectomy was $196,700/QALY. At a relative risk of 0.78 (the lower limit of the 95 percent confidence interval from the same meta-analysis result), the cost-effectiveness of NSAA plus orchiectomy relative to orchiectomy alone was $43,400/QALY and of NSAA plus LHRH agonist relative to orchiectomy was $110,900/QALY.

Sensitivity Analysis

We performed extensive sensitivity analysis to test the robustness of our findings and identify important areas of uncertainty. The relative rankings of strategies were not sensitive to assumptions about the natural history or cost of care of prostate cancer. Similarly, the cost-effectiveness of orchiectomy relative to DES remained less than $30,000 per QALY throughout these assumptions. Even with high cost estimates for orchiectomy, the cost-effectiveness ratio of orchiectomy relative to DES remained favorable.

Two assumptions were important in determining the cost-effectiveness of orchiectomy relative to DES. First, if the excess cardiovascular mortality from DES is less than we assumed, the associated quality-adjusted survival improved although it was always less than that associated with orchiectomy. If there was no excess mortality associated with DES, then the incremental cost-effectiveness of orchiectomy is $13,100/QALY.

Figure 18. Sensitivity to Quality of Life of Orchiectomy (more...)

   Figure 18. Sensitivity to Quality of Life of Orchiectomy

The cost-effectiveness of orchiectomy relative to LHRH agonists is plotted against the quality-of-life adjustment associated with orchiectomy. If the quality of life associated with orchiectomy is less than 0.88, the marginal cost-effectiveness ratio (comparing orchiectomy to LHRH agonists) is below $100,000/QALY in favor of LHRH agonists. In the base case, orchiectomy had a similar quality of life weight as DES (0.92).

Figure 18

Figure 19. Sensitivity Analysis on the Efficacy of (more...)

   Figure 19. Sensitivity Analysis on the Efficacy of Combined Androgen Blockade

Panel A

Figure 19. Sensitivity Analysis on the Efficacy of (more...)

   Figure 19. Sensitivity Analysis on the Efficacy of Combined Androgen Blockade

Panel B
The cost effectiveness of combined androgen blockade (CAB) relative to orchiectomy at different assumptions about the efficacy of CAB. Two CAB strategies are shown: a nonsteroidal antiandrogen with an LHRH agonist (Panel A) and a nonsteroidal antiandrogen with orchiectomy (Panel B). Efficacy is presented as the relative hazard of disease progression compared to orchiectomy. The shaded areas represent the confidence intervals from the current meta-analysis and the arrows indicate the point estimates.

Figure 19

Figure 19

We also conducted a sensitivity analysis on the quality of life associated with a adverse effect of treatment that was not dose-limiting because patients may vary considerably in how they rate the associated adverse effects. In our base case, we assigned a quality-of-life weight of 0.55 to this state. If adverse effects are judged as severely bothersome (assigned a much lower quality-of-life weight), orchiectomy becomes less favorable because it is irreversible. The incremental cost-effectiveness of orchiectomy relative to DES is $35,200 per QALY and of flutamide relative to orchiectomy is $361,100 per QALY. Conversely, if adverse effects are judged to be very minor (assigned a much higher quality-of-life weight), orchiectomy becomes considerably more favorable with an incremental cost-effectiveness relative to DES of $6,500 per QALY, with all other strategies eliminated by strict dominance.

The cost-effectiveness estimates were less sensitive to several other values assigned to baseline variables. The model was also insensitive to the cost of orchiectomy-even if orchiectomy cost as much as $7,000 (Krahn, Mahoney, Eckman et al., 1994) the cost-effectiveness relative to DES would be $15,400 per QALY and all other strategies would be dominated. As well, the model was insensitive to the cost of a local obstruction. If all obstructions were associated with costs of $300, the cost-effectiveness of orchiectomy relative to DES was $7,400 per QALY. The model was also relatively insensitive to assumptions about the age of the man with prostate cancer at the start of the analysis. For 50-year-old men, the cost-effectiveness of orchiectomy relative to DES is $5,800 per QALY, whereas for 75-year-old men the corresponding estimate is $11,000 per QALY. All other strategies are dominated in all age groups.

The relative cost-effectiveness of antiandrogen strategies for patients who have symptomatic distant metastases at presentation changes only slightly. DES is still the least expensive option with the worst quality-adjusted life expectancy. Orchiectomy is associated with an incremental cost-effectiveness of $31,500 /QALY. All other strategies are dominated.

Timing of Antiandrogen Therapy

Figure 20. Optimal Timing of Antiandrogen Therapy

   Figure 20. Optimal Timing of Antiandrogen Therapy

Strategy	Cost	Effectiveness (QALYs)	Marginal Cost-Effectiveness
Symptomatic Distant Metastases	$5,217	6.97
Asymptomatic Distant Metastases	$5,626	6.79	Eliminated by Strict Dominance
Local Recurrence	$7,355	6.19	Eliminated by Strict Dominance

Figure 20. Optimal Timing of Antiandrogen Therapy

   Figure 20. Optimal Timing of Antiandrogen Therapy

Costs and quality-adjusted survival when orchiectomy is performed at the time of diagnosis with local spread, when asymptomatic distant metastases develop, or when symptomatic distant metastases develop. Outcomes for orchiectomy (top) and combined androgen blockade with a nonsteroidal antiandrogen and an LHRH agonist (bottom) are shown.

Strategy	Cost	Effectiveness (QALYs)	Marginal Cost-Effectiveness
Symptomatic Distant Metastases	$14,710	6.94
Asymptomatic Distant Metastases	$12,169	6.76	$14,302 /QALY
Local Recurrence	$51,024	5.94	Eliminated by Strict Dominance

Table 24. Costs and Effectiveness of Strategies According (more...)

Table 24. Costs and Effectiveness of Strategies According to Time of Initiation

Strategy	Timing of Initiation	Cost	Effectiveness (QALYs)
DES	Symptomatic distant metastases	$3,135	4.95
	Asymptomatic distant metastases	$3,205	4.83
	Local Recurrence	$3,309	4.39
Orchiectomy	Symptomatic distant metastases	$5,217	6.97
	Asymptomatic distant metastases	$5,626	6.79
	Local recurrence	$7,355	6.19
NSAA	Symptomatic distant metastases	$6,524	6.56
	Asymptomatic distant metastases	$6,311	6.43
	Local recurrence	$19,359	5.80
LHRH	Symptomatic distant metastases	$12,101	5.75
	Asymptomatic distant metastases	$10,501	5.49
	Local recurrence	$26,963	4.84
NSAA + Orchiectomy	Symptomatic distant metastases	$9,833	6.94
	Asymptomatic distant metastases	$8,716	6.76
	Local recurrence	$24,682	5.95
NSAA + LHRH	Symptomatic distant metastases	$14,710	6.94
	Asymptomatic distant metastases	$12,169	6.76
	Local recurrence	$51,024	5.94

Figure 20

Figure 20

Incorporating Biochemical Markers

Figure 21. Incorporating Biochemical Markers

   Figure 21. Incorporating Biochemical Markers

Costs and effectiveness when decisions to change therapy are based on biochemical markers (PSA) as well as clinical outcomes. See text for details of assumptions.

Strategy	Cost	Effectiveness (QALYs)	Marginal Cost-Effectiveness
DES	$4,896	7.04
Orchiectomy	$8,417	7.54	$7,057/QALY
NSAA	$11,629	7.49	Eliminated by strict dominance
NSAA + orchiectomy	$15,919	7.50	Eliminated by strict dominance
LHRH agonist	$19,242	7.54	Eliminated by strict dominance
NSAA + LHRH agonist	$26,724	7.50	Eliminated by strict dominance

Figure 21

Discussion

Our evaluation of the cost-effectiveness of androgen suppression therapy for men with advanced prostate cancer incorporated several therapeutic strategies, considering not only a wide range of treatments but also issues such as the optimum time to initiate therapy. It considered how alternative treatments affect longevity, quality of life, and costs and examined how recent advances in clinical care might alter the principal findings. The results suggest that neither the most expensive nor the least expensive treatment options are good values. DES, the least costly therapy, is associated with the shortest life expectancy and the fewest quality-adjusted life years. Since the true excess mortality associated with low doses is unknown, DES may be considered as an appropriate agent for future studies. Failing this, alternative approaches will be preferred.

Orchiectomy represents a highly cost-effective alternative to DES. Under most of the assumptions we explored, its cost-effectiveness ratio is less than $30,000/QALY when compared with DES, a range that is usually considered to represent a very good value. Monotherapy with a nonsteroidal antiandrogen or an LHRH agonist costs significantly more and produces minimal additional benefit. Combined androgen blockade may yield slight additional benefit, but at an even greater increase in cost. DES and orchiectomy are used infrequently in the United States, as regimens that are more expensive have grown in popularity. The meta-analysis suggests, and the cost-effectiveness analysis confirms, that the extra benefit from these new therapies is small, even after accounting for differential toxicities. However, this conclusion is based on the assumption that orchiectomy is not associated with a worse quality of life than other antiandrogen therapies. Yet preferences toward orchiectomy may vary greatly among patients. Unlike medications, it is irreversible, and some men appear to believe that the treatment is unacceptable. For such individuals, orchiectomy would not produce the gain in QALYs that we estimate occurs on average. For other men, the psychological drawbacks of orchiectomy are minimal, and the treatment will be far more attractive. Medical therapy itself has significant drawbacks, such as increased office visits, pain with injectable medications, and associated psychological consequences (such as the constant reminder that patients have cancer). We did not include these factors in our analysis, but an increase in costs or a decrease in quality of life would strengthen our base case results. Nevertheless, the objective evidence supporting a decreased quality of life associated with orchiectomy is scant. Although many patients may be clear about how they value this health state, we present these results in detail to emphasize the uncertainty about the average patient's values and their importance to the f inal cost-effectiveness estimates.

The most expensive strategy is combined androgen blockade. Its efficacy is controversial; the meta-analysis suggests that it is not significantly more efficacious than orchiectomy. Note that we refer here only to the average benefit; a minority of patients may still experience a large benefit. Focusing on this average benefit and setting the cost-effectiveness threshold at $100,000/QALY, combined androgen blockade with a nonsteroidal antiandrogen and an LHRH agonist must decrease the risk of disease progression by 20 percent compared to orchiectomy before this strategy is considered cost-effective, a value within the range of estimates from the meta-analysis. Alternative combined androgen blockade strategies (such as orchiectomy and a NSAA) are less expensive and hence require less efficacy to be cost-effective. For these therapies, cost-effectiveness ratios are less than $100,000/QALY if the relative hazard of disease progression is less than 0.9, also a value within the range of estimates from the meta-analysis.

Our analysis of LHRH agonists and NSAAs used a cost-minimization approach that builds on the assumption that the efficacy and toxicities of drugs within a certain classification are identical. Although this assumption appears reasonable, the toxicities of NSAAs have not been rigorously evaluated. Flutamide, nilutamide, and bicalutamide may actually have dissimilar long-term adverse effects. Further toxicity data would clarify this choice and are a subject for future research.

Our model suggests that for patients diagnosed with locally progressive cancer at the time of diagnosis, initiating antiandrogen therapy early, when patients enjoy a good quality of life, will result in higher costs and no additional benefits and possible harm compared to deferring therapy. In the absence of a survival benefit, starting antiandrogen therapy when patients have advanced disease (whether orchiectomy or medications) is of greatest benefit to the patient. Indeed, our analysis implies that quality of life improves more when antiandrogen therapy is initiated at a later stage. A small empirical study found that patients with asymptomatic prostate cancer who did not receive hormonal therapy had similar or better quality of life than patients who received hormonal therapy (Herr, Kornblith, and Ofman, 1993). If this result is generally valid, the cost advantages of delaying prostate cancer would be strongly backed by clinical benefit. Furthermore, because we modeled the effects of late therapy as a delay in the occurrence of very poor health (rather than as an increase in quality of life), we may have underestimated the benefits of late therapy. If such a bias exists, starting antiandrogens when distant metastases develop would be even more economically attractive.

Our analysis of the optimal timing of antiandrogen therapy focuses only on patients who present with metastatic cancer at the time of diagnosis, but not patients who use antiandrogens as adjunctive therapy at the time of definitive treatment for localized prostate cancer. The meta-analysis indicates that for the latter patients, a survival benefit from early initiation of antiandrogens exists. Whether this approach is cost-effective is a topic for future study.

A previous model of antiandrogens has been published and updated (Bennett, Matchar, McCrory et al., 1996; Hillner, McLeod, Crawford et al., 1995). This analysis examined only the incremental cost-effectiveness of combined androgen blockade with orchiectomy plus flutamide compared to orchiectomy alone. The cost-effectiveness estimates were roughly of the same order when similar assumptions about this combined androgen blockade strategy were compared. If the efficacy of this strategy was 25 percent as assumed in the original previous cost-effectiveness analysis (relative hazard 0.75), that model estimated a cost-effectiveness ratio of $25,300 per QALY and corresponding estimate from our study is $37,600 per QALY. If the efficacy of this strategy was 9 percent, the previous cost-effectiveness estimate was $60,900 per QALY and the current estimate is $138,200 per QALY. Differences between estimates exist because the previous model used a different definition of health states (classifying asymptomatic distant metastases and symptomatic distant metastases as a single health state), did not include all features of the current model (for example, local obstruction), and used some different variables in the base case analysis (for example, a discount rate of 5 percent).

Our model has several limitations, including the need to simplify a number of assumptions. Nevertheless, its findings change little when uncertain values entered in the model are varied over wide ranges in one-way sensitivity analyses. Our model also did not include therapies unapproved for use in the United States. Lastly, our model does not indicate how individual patients, clinicians, or health plans might value the benefits or costs in the model differently.

Improvements in monitoring the stage and extent of prostate cancer can lead to earlier detection of metastatic disease (Smith and Pienta, 1997; Zietman, Dallow, McManus et al., 1996). Such advances might lead to the earlier initiation of antiandrogen therapy. Current practice often uses the PSA as an indicator of advancing disease, a strategy that we were unable to evaluate fully because no published evidence establishes the efficacy of this practice (we also did not evaluate the use of PSA as a prognostic indicator). However, redefining the states and transitions in our model to approximate the effects of such an "early detection" strategy incorporating PSA monitoring led to virtually no incremental improvement in outcomes. Consequently, our results are likely to continue to apply to evolving monitoring strategies. Without further advances in treatment, improved monitoring is likely to have little effect on the cost-effectiveness of treatment strategies for advanced prostate cancer.

Cost-effectiveness analysis can inform health policy decisions by indicating the most efficient use of scarce resources but does not substitute for the full decisionmaking process, which must account for factors beyond economics. Furthermore, cost-effectiveness analyses can identify key factors that may be important for decisionmakers. For example, the current analysis indicates that if there is substantial heterogeneity in patient preferences toward the effects of orchiectomy and alternative therapies, it will be important to individualize therapy. Other individual preferences such as attitude to antiandrogen adverse effects should also be considered. Appropriate application of cost-effectiveness results would include a careful assessment of each individual's values with respect to surgical and medical castration. For patients whose quality of life would diminish substantially if they underwent orchiectomy, the use of LHRH agonists or NSAA may represent reasonable alternatives, even though the average patient's quality of life may not fall substantially with orchiectomy.