• Open Access

Combined androgen blockade for prostate cancer: Review of efficacy, safety and cost-effectiveness

Authors


To whom correspondence should be addressed.
E-mail: akazah@med.rcast.u-tokyo.ac.jp

Abstract

A standard treatment for advanced prostate cancer is androgen deprivation by surgical or medical castration. In theory, however, combined androgen blockade (CAB) with an antiandrogen plus castration should be more effective because castration alone does not completely eliminate androgens in the prostate. Therefore, a number of randomized clinical trials (RCT) were conducted in the 1990s to investigate the efficacy of CAB with an antiandrogen (nilutamide or flutamide) plus castration; however, there were both positive and negative results for the efficacy of CAB. The lack of data on safety, quality of life (QOL) and cost-effectiveness has been a hindrance to the adoption of CAB for the treatment of prostate cancer. Nevertheless, discussion on CAB for the treatment of prostate cancer has continued for over 20 years, which suggests that there remains some hope for this regimen. In the 2000s, clinical research on CAB with the antiandrogen bicalutamide commenced. CAB using this new antiandrogen was found to prolong overall survival (OS) in patients with prostate cancer, with favorable safety profiles and cost-effectiveness, without deteriorating QOL. In this article, we discuss the feasibility of CAB with bicalutamide for the treatment of prostate cancer by reviewing the theoretical background of CAB and then the results of RCT conducted in the 1990s when the usefulness of CAB was assessed. (Cancer Sci 2011; 102: 51–56)

Theoretical Background of CAB

Huggins et al.(1) first reported the effects of hormonal therapy on metastatic prostate cancer about 70 years ago. Since then, several studies have examined the efficacy of hormonal therapy for this disease. However, results from these earlier studies were probably not optimal because few drugs were available and the disease was mostly advanced before the prostate-specific antigen (PSA) test was available for screening. Therefore, these data with older drugs could have disguised the benefits of CAB described in a later section. It has been reported that prostate cancer is androgen-dependent in the majority of cases and that 95% of androgens are testosterone of testicular origin.(2) These androgens are thought to promote the growth of cancer cells by binding to androgen receptors (AR) in prostate cancer cells. Therefore, the first-line treatment for advanced prostate cancer has been androgen deprivation by medical castration with luteinizing hormone-releasing hormone agonist (LHRH-A) or by surgical castration with bilateral orchiectomy.

However, it has been shown that dehydroepiandrosterone (DHEA) and androstenedione are also converted to androgens in prostate cancer cells after secretion from the adrenal glands.(3) Therefore, it is expected that blockade of androgen of adrenal origin by an antiandrogen, which inhibits the binding of androgen to AR, combined with castration could lead to more effective inhibition of prostate cancer.(4) Thus, since the end of the 1970s, CAB has been investigated as a potential treatment for prostate cancer. It has been reported that nonsteroidal antiandrogens show favorable efficacy profiles and are well-tolerated, and that antiandrogens with a higher AR affinity have stronger androgen-suppressive effects.(5–7)

Usefulness of CAB Versus Castration Alone

In the 1990s, approximately 30 were conducted to investigate the efficacy and safety of CAB compared with castration alone.

Crawford et al.(8) studied the efficacy of CAB in 603 patients with stage D2 prostate cancer after being randomized to treatment with leuprorelin (a LHRH-A) plus placebo or flutamide. They showed that the progression-free survival (PFS) was significantly prolonged in the flutamide group compared with the placebo group at 16.5 vs 13.9 months (= 0.039). Additionally, the median survival was 35.6 months in the flutamide group and 28.3 months in the placebo group, which was significantly better in the flutamide group (= 0.035). Although tolerability was similar in both groups, the incidence of moderate diarrhea was significantly greater in the flutamide group (< 0.001).

Boccardo et al.(9) randomized 373 patients with stage C or D prostate cancer to treatment with goserelin (a LHRH-A) alone, or CAB with goserelin plus flutamide. At a median follow up of 24 months no significant differences were observed in response rates, PFS or overall survival (OS) between the groups. Although the median time to progression (TTP) was 18 months in the goserelin group and 24 months in the CAB group, the difference was not statistically significant (= 0.09). In addition, the time to normalization of serum prostatic acid phosphatase concentrations and the time to relief of bone pain were shorter in the CAB group than the goserelin alone group, whereas the incidence of adverse reactions such as diarrhea and increased blood transaminases was significantly higher in the flutamide group than the goserelin alone group.

Eisenberger et al.(10) randomized 1387 patients with metastatic prostate cancer to bilateral orchiectomy plus placebo or the antiandrogen flutamide. The median follow-up time was 49.2 months in the placebo group and 50.1 months in the flutamide group. The OS did not differ significantly between the two groups (= 0.14) and there was no significant reduction in the risk of death in the flutamide group compared with the placebo group (hazard ratio [HR] = 0.91; 95% confidence interval [CI] = 0.81–1.01). Although the incidence of toxicity associated with both treatments was generally low, grade 2 or higher diarrhea and anemia were significantly more frequently observed in the flutamide group than the placebo group (= 0.002 and = 0.024, respectively).

Dijkman et al.(11) reported the clinical results of 457 patients with stage D2 prostate cancer randomized to treatment with the antiandrogen nilutamide or placebo following orchiectomy; follow up was approximately 8.5 years. The proportion of patients who achieved normalization of the PSA level after 3 months of treatment was significantly higher in the nilutamide group than the placebo group (< 0.001). Additionally, TTP was significantly prolonged in the nilutamide group at 21.2 months versus only 14.7 months in the placebo group (= 0.002). Moreover, median cancer-specific survival was significantly prolonged in the nilutamide group at 37.0 months versus only 29.8 months in the placebo group (= 0.013).

These four studies were the most notable RCT conducted in the 1990s, and they showed both positive and negative efficacy results for CAB using nilutamide and flutamide for the treatment of prostate cancer. Therefore, it remains unclear and questionable as to whether the strategy of CAB is superior to castration alone.

To help clarify these data, the Prostate Cancer Trialists’ Collaborative Group (PCTCG)(12) conducted a meta-analysis of 27 RCT that involved a total of 8275 patients with advanced prostate cancer. The efficacy of surgical or medical castration alone was compared with that of CAB using a steroidal antiandrogen (cyproterone acetate) or a non-steroidal antiandrogen (flutamide or nilutamide). The 5-year survival rate was 25.4% in the CAB group and 23.6% in the castration alone group, which was not significantly different (log-rank 2= 0.11). However, when the data were stratified by the type of antiandrogen, the 5-year survival rate in patients receiving CAB with a non-steroidal antiandrogen (flutamide or nilutamide) was found to be significantly superior to that of patients receiving castration alone: 27.6%vs 24.7%, respectively (log-rank 2= 0.005) (Fig. 1). It should be noted that the non-steroidal antiandrogens used in the analysis were flutamide and nilutamide, because bicalutamide – currently the leading antiandrogen – was not available at that time.

Figure 1.

 Mortality results from randomized trials of maximal androgen blockade (MAB) versus androgen suppression (AS) alone in advanced prostate cancer. The black squares indicate the ratio of daily death rates (MAB vs AS), as calculated from the log rank statistics, and the horizontal line gives the corresponding 99% confidence interval (CI). The area of the square is proportional to the amount of information it represents. Ratios less than 1.0 favor MAB and ratios more than 1.0 favor AS alone. A diamond is used to denote the corresponding result and its 95% CI for the total of all trials (and for subtotals). *Adjusted = for balance, control patients in three-way trials count half or twice in subtotals and in the final totals of deaths and patients. O–E, Observed minus Expected. Adapted from the Prostate Cancer Trialists' Collaborative Group, with permission.(12)

Thus, although there was a theoretical justification for using CAB for prostate cancer, the survival benefit observed compared with castration alone in RCT and meta-analyses in the 1990s was small. In addition, the data did not demonstrate the superiority of CAB versus castration alone with respect to safety profiles, QOL and cost-effectiveness. Therefore, there was insufficient evidence to recommend CAB strongly at that time. Indeed, the 2004 American Society of Clinical Oncology (ASCO) Recommendations for the Initial Hormonal Management of Androgen-Sensitive Metastatic, Recurrent or Progressive Prostate Cancer(13) evaluated these strategies in terms of benefit, harm and cost, and concluded that “a small survival advantage was likely with CAB over castration alone, although the benefit must be balanced against great toxicity and extraordinarily poor cost-effectiveness.”

Evaluation of CAB with Bicalutamide

As mentioned, data on CAB from the 1990s have not established its superiority over castration alone, with regard to efficacy, safety, QOL or cost-effectiveness. By the late 1990s, CAB with bicalutamide had been frequently evaluated to assess the usefulness of CAB.

Efficacy.  Schellhammer et al.(14) studied 813 patients with stage D2 prostate cancer after randomization to treatment with either LHRH-A plus bicalutamide or flutamide. They showed that the median TTP was 97 weeks in the bicalutamide group and 77 weeks in the flutamide group, and that the median survival was 180 weeks in the bicalutamide group and 148 weeks in the flutamide group. Although bicalutamide prolonged both TTP and median survival, the differences between the two groups were not significant (= 0.41 and = 0.15, respectively) (Fig. 2). Re-analysis by Klotz et al.(15) found that CAB with 50 mg bicalutamide reduced the risk of death by 20% compared with castration alone (Fig. 3), through careful evaluation of disparate trials which is called “delta-method”. However, there are several limitations to cross-study comparisons and the lack of data from an RCT that has compared bicalutamide-containing CAB versus castration alone leaves room for discussion of the benefits of this combination therapy.

Figure 2.

 Kaplan–Meier distributions of survival time comparing CAB with bicalutamide and CAB with flutamide. LHRH-A, luteinizing hormone-releasing hormone agonist. Adapted from Schellhammer et al., with permission.(14)

Figure 3.

 Hazard ratios (HR) for OS. CI, confidence interval. Adapted from Klotz et al., with permission.(15)

A phase III randomized, double-blind, placebo-controlled trial(16,17) was conducted with 205 Japanese patients with stage C or D prostate cancer, who were randomized to either CAB with LHRH-A plus bicalutamide or LHRH-A monotherapy. Over a median observation period of 2.4 years, there was significant prolongation of TTP and the time to treatment failure (TTTF) in the CAB group compared with the LHRH-A monotherapy group (< 0.001). Therefore, further follow up was conducted to investigate the survival outcome (Fig. 4).(18) After a median follow up of 5.2 years(19) a significant OS advantage was observed with CAB versus LHRH-A monotherapy (HR = 0.78; 95% CI = 0.60–0.99; = 0.0498) (Fig. 5). Indeed, the 5-year OS rate estimated by the Kaplan–Meier method was 75.3% in the CAB group and 63.4% in the LHRH-A monotherapy group. A subgroup analysis of OS by disease stage(19) revealed that the survival rate of stage C or D1 patients was significantly higher in the CAB group than in the LHRH-A monotherapy group (= 0.0041). However, in stage D2 patients no significant difference was observed in the survival rate between the study arms (= 0.8335). In addition, no significant differences were observed between the groups in relation to cause-specific survival. This suggests that CAB is more effective in prostate cancer patients with early stage disease, such as C and D1. The proportion of patients who achieved a PSA nadir of ≤1 ng/mL was significantly different between the groups, with 81.4% of patients in the CAB group achieving that end-point versus only 33.7% in the LHRH-A monotherapy group (< 0.001).(19) The investigators found in an exploratory analysis that achieving a PSA nadir of ≤1 ng/mL was a significant prognostic factor for improved OS. These data support the findings of Klotz et al.(15) that CAB with bicalutamide improved survival by approximately 20% compared with surgical or chemical castration alone.

Figure 4.

 Study design comparing CAB with bical-utamide and LHRH-A monotherapy.(17) Adapted from Hinotsu et al., with permission.(18)

Figure 5.

 Kaplan–Meier curve of overall survival comparing CAB with bicalutamide and LHRH-A monotherapy.

Although de Leval et al.(20) and Sato et al.(21) have reported data suggesting a role for intermittent therapy, no consensus about methodology and efficacy has yet been reached. Therefore, intermittent CAB therapy for advanced prostate cancer should be investigated only in clinical research.

Safety and QOL.  Usami et al.(17) compared safety outcomes in the previously mentioned RCT between the CAB and LHRH-A monotherapy groups after 2.4 years of follow up. They found that the dropout rate due to adverse drug reactions (ADR) was 8.8% in the CAB group and 10.9% in the LHRH-A monotherapy group (95% CI = 6.4–10.7). In addition, tolerability profile, overall ADR (66.7% for CAB and 65.3% for LHRH monotherapy) and adverse events (93.1% for both groups) was similar in both groups. Arai et al.(22) compared QOL between the CAB group and the LHRH-A monotherapy group using the Functional Assessment of Cancer Therapy-Prostate (FACT-P) questionnaire. They reported that there was no decrease in overall QOL in the CAB group, but rather the CAB group had more rapid and greater improvements in “emotional well-being” and “prostate cancer-specific issues” domain scores in FACT-P compared with the LHRH-A monotherapy group. Additionally, they showed that CAB improved micturition disorder-related QOL – a factor that greatly contributed to an improvement in the “prostate cancer-specific issues” domain.

With regard to the effect of CAB on cardiovascular risk, three scientific societies – the American Heart Association, American Cancer Society and American Urological Association – have jointly proposed guidelines on hormonal therapy for prostate cancer and cardiovascular risk.(23) The guidelines include a statement that “at present, it is appropriate to consider that androgen deprivation may be associated with cardiovascular events and cardiovascular death,” based on research reports that androgen deprivation for prostate cancer causes weight gain, a decrease in insulin sensitivity and lipid metabolism abnormalities. In light of these data, the guidelines recommend monitoring blood pressure lipid and blood glucose levels before starting androgen deprivation therapy and within 3–6 months after the start of therapy. In addition, for patients on long-term androgen deprivation therapy, the guidelines recommend monitoring lipid and blood glucose levels at least once a year.

In Japan, the number of patients with cardiovascular disease is lower than in Western countries. Even though the consumption of fat and the average total serum cholesterol level have increased in Japan, rates of mortality and morbidity from myocardial infarction remain the lowest of all developed countries.(24) These observations are supported by findings from the WHO multinational monitoring of trends and determinants in cardiovascular disease (The WHO MONICA project),(25) which monitors trends in cardiovascular diseases and related risk factors.

Akaza et al.(26) have reported that the expected life years of patients with localized or locally advanced prostate cancer after hormonal therapy or surgical castration are similar to those of the general population. The results of a search of a large database, introduced in the review article by Akaza et al.(27) (Table 1), also support this finding. However, at this time, because no definite conclusion has been reached about the relationship between androgen deprivation therapy and cardiovascular risk, it might be preferable to monitor patients with cardiovascular complications on a regular basis.

Table 1.   Cardiovascular deaths among leuprorelin-treated patients compared with a similar-sized general Japanese population cohort in the years 2001–2006 (data from Japan study group of Prostate Cancer [J-Cap])
YearObserved number of leuprorelin-treated patientsCV deaths in leuprorelin-treated patientsEstimated CV mortality rate/Japanese general population cohort
  1. CV, cardiovascular. Adapted from Akaza et al., with permission.(27)

200180024.0/800
2001166659.1/1666
200325151615.1/2515
20042243914.5/2243
200518351012.7/1835
20061470711.0/1470

No RCT has been conducted on the fracture risk with CAB; however, Bolla et al.(28) have reported a fracture risk with hormonal therapy after external beam radiotherapy. In their study, 415 patients with advanced prostate cancer received the LHRH-A goserelin for 3 years after external beam radiotherapy and pathological fracture was observed in only two patients.

These studies suggest that the effects of hormonal therapy on bone metabolism and cardiovascular risk in patients with prostate cancer do not outweigh the benefits of hormonal therapy. Therefore, it is recommended that health care providers discuss both the risks and benefits of CAB with bicalutamide with patients prior to the selection of a treatment approach.

Cost effectiveness.  To address the issue of cost-effectiveness, Nishimura et al.(29) constructed a Markov model that examined the prognosis of untreated prostate cancer and estimated the cost-effectiveness of CAB in comparison with that of LHRH-A monotherapy. The model showed that the expected costs of CAB and LHRH-A monotherapy were 5.24 and 3.66 million yen (approximately US$55 851 and US$39 010), respectively, with expected survival durations of 7.45 and 6.44 years, respectively. The incremental cost-effectiveness ratio (ICER) for CAB compared with LHRH-A monotherapy was 1.56 million yen/life-year-saved (approximately US$16 627), and was lower than the ICER threshold set in the study (6 million yen/life-year-saved [approximately US$63 952]), which demonstrates that the cost-effectiveness of CAB is superior to that of LHRH-A monotherapy. Similar results were found in a study conducted by Penson et al.(30)

CAB Conclusion

As described in this review, CAB for prostate cancer has been investigated in many RCT and in meta-analyses since the 1990s. However, its role in the treatment of prostate cancer has been debated due to the contradictory results from these trials. Therefore, the 2004 clinical practice guidelines issued by ASCO concluded that CAB confers a statistically significant but questionable clinical improvement in survival over orchiectomy or LHRH-A monotherapy. However, more recently, the efficacy and safety of CAB have been investigated using the antiandrogen bicalutamide. Data from these trials suggest that CAB with bicalutamide significantly prolongs survival without deteriorating safety and QOL(16–19,22) in Japanese patients with prostate cancer. These data might be supported by the results that Fukagai et al.(31) reported in 2006, which suggested racial differences between Japanese and Caucasians as a factor for the differences in clinical outcomes after hormonal therapy. Therefore, after evaluating these data,(15,16) ASCO revised its recommendations(32) for hormonal therapy for prostate cancer in 2007 as follows: “Given that the bicalutamide CAB has minimal, if any, additional toxicity over castrate therapies alone and is significantly cheaper than the newer systemic therapies, until the results of a trial designed to address the potential survival benefit is available, patients should be made aware of the findings described herein, and bicalutamide CAB should be considered.” Additionally, the cost-effectiveness of CAB with bicalutamide has been shown to be excellent, when considering its overall survival benefit.(29,30) Thus, recent data have shown that CAB, a strategy which has been debated for many years, is a viable treatment option for prostate cancer when bicalutamide is used as the antiandrogen for CAB.

Discussion

In conclusion, CAB with bicalutamide is gaining support due to the publication of the ASCO 2007 guidelines and the Akaza et al.(19) trial. In addition, new hormonal drugs are being developed, including an LHRH antagonist (degarelix), selective CYP17 inhibitors (abiraterone and TAK700) and MDV3100, which is said to be a second-generation oral antiandrogen. In addition, in castration-resistant prostate cancer, which has recently been gaining attention, Cheng et al.(33) have suggested the involvement of AR in the growth of cancer cells. Therefore, the role of hormonal therapy using drugs with pharmacological effects via androgens is gaining attention for the treatment of prostate cancer. It remains to be seen how CAB will compare with castration alone and other potential therapies when these new drugs become available.

Disclosure Statement

The authors have no conflict of interest.

Ancillary