Androgen deprivation therapy: past, present and future


Professor F Schröder, Erasmus MC University Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. e-mail:


Since Huggins and Hodges demonstrated the responsiveness of prostate cancer to androgen deprivation therapy (ADT), androgen-suppressing strategies have formed the cornerstone of management of advanced prostate cancer. Approaches to ADT have included orchidectomy, oestrogens, luteinizing hormone-releasing hormone (LHRH) agonists, anti-androgens and more recently the gonadotrophin-releasing hormone antagonists. The most extensively studied antagonist, degarelix, avoids the testosterone surge and clinical flare associated with LHRH agonists, offering more rapid PSA and testosterone suppression, improved testosterone control and improved PSA progression-free survival compared with agonists. The clinical profile of degarelix appears to make it a particularly suitable therapeutic option for certain subgroups of patients, including those with metastatic disease, high baseline PSA (>20 ng/mL) and highly symptomatic disease. As well as forming the mainstay of treatment for advanced prostate cancer, ADT is increasingly used in earlier disease stages. While data from clinical trials support the use of ADT neoadjuvant/adjuvant to radiotherapy for locally advanced or high-risk localized prostate cancer, it remains to be established whether specific ADT classes/agents provide particular benefits in this clinical setting.


androgen deprivation therapy


adjuvant hormonal therapy


cancer-specific survival


disease-free survival


European Association of Urology


Eastern Cooperative Oncology Group


European Organisation for Research and Treatment of Cancer


hazard ratio


luteinizing hormone


neoadjuvant hormonal therapy


overall survival


prostate cancer


progression-free survival


radical prostatectomy




Radiation Therapy Oncology Group


serum alkaline phosphatase


Trans-Tasman Radiation Oncology Group


In 1941, Huggins et al. demonstrated the favourable impact of androgen deprivation therapy (ADT) via orchidectomy or oestrogens on metastatic prostate cancer (PCa) [1,2]. Following that discovery, orchidectomy became the gold standard for ADT in advanced PCa [3] (Table 1). Oestrogens (most commonly diethylstilboestrol) were used as a medical alternative to surgical castration until the late 1960s and 1970s [17]. However, while oestrogens provided effective testosterone suppression, their use has been limited due to an increased risk of cardiovascular adverse events. Indeed, studies by the Veterans Administration Cooperative Urological Research Group showed an increased risk of cardiovascular mortality with diethylstilboestrol 5 mg [4,18].

Table 1.  Summary of the relative advantages and disadvantages of ADT classes [3–16]Thumbnail image of

In the late 1960s and early 1970s, researchers also investigated anti-androgenic compounds. Anti-androgens compete with androgens for binding sites on androgen receptors in the prostate cell nucleus, promoting apoptosis and inhibiting the growth of PCa [8]. Over the years, investigators have examined the possible use of these agents as an alternative to castration in advanced PCa. Evidence from a meta-analysis using data from eight trials involving 2717 patients suggested that non-steroidal anti-androgen monotherapy was associated with lower overall survival (OS) than orchidectomy [6]. Bicalutamide is the most extensively studied non-steroidal anti-androgen. Survival analyses of two large prospective randomized trials including a total of 1435 patients showed that anti-androgen monotherapy with bicalutamide was less effective than castration in patients with M1 disease [19] but in locally advanced M0 patients (n= 480) no significant difference in OS was noted [5]. Thus, European Association of Urology (EAU) guidelines consider non-steroidal anti-androgen monotherapy (eg bicalutamide) as an alternative to castration in patients with locally advanced disease [3]. Common side effects with non-steroidal agents include gynaecomastia and breast pain. Steroidal anti-androgens are associated with loss of libido and erectile dysfunction, as well as cardiovascular toxicity and hepatotoxicity [8]. Questions persist on the benefit of using anti-androgens in combination with other forms of ADT. In a meta-analysis of 27 randomized trials including 8275 patients with metastatic or locally advanced PCa, adding an anti-androgen to androgen suppression improved 5-year OS by only 2% [20].

LHRH agonists offer another reversible alternative to surgical castration and in the 1980s began to supersede oestrogens and orchidectomy [21] to become the mainstay of ADT for advanced PCa [3]. In a meta-analysis of data from 10 trials (n= 1908), OS with LHRH agonists was found to be equivalent to orchidectomy or diethylstilboestrol in advanced disease [6]. Moreover, ADT with LHRH agonists is now also having an increasingly important role in earlier stages of PCa. In patients with high metastatic risk, immediate androgen suppression with an LHRH agonist given during and for 3 years after external radiotherapy (RT) has been shown to improve 10-year disease-free survival (DFS) and OS [22]. Thus, in patients with locally advanced or high-risk localized disease, the addition of neoadjuvant and adjuvant ADT is now considered the standard of care for men treated with radical RT [16]. However, LHRH agonist therapy is associated with several drawbacks. The initial stimulation of LHRH receptors causes a testosterone surge, delaying the achievement of castrate testosterone levels for around 2–4 weeks. The surge may also be associated with clinical flare effects in advanced disease (eg bone pain, spinal cord compression, ureteral obstruction and possibly death) [23,24]. Furthermore, some patients fail to achieve castration levels of testosterone. Tombal and Berges [12] reported that up to 12% of patients do not reach a testosterone level <50 ng/dL and up to 37% fail to achieve levels <20 ng/dL. Also, breakthrough testosterone increases of >50 ng/dL have been reported to occur in around 2–25% of patients after long-term hormone therapy [25,26]. Breakthrough increases of 20–50 ng/dL were reported in 32% of patients receiving an LHRH agonist (with or without an anti-androgen) [27]. Importantly, time to progression to androgen-independent disease was significantly shorter in those with testosterone breakthrough increases >32 ng/dL compared with those without breakthrough [26].


The flare phenomenon associated with LHRH agonists helped to drive the search for alternative agents. Consequently, the new millennium saw the development of GnRH antagonists. These agents bind directly to and block GnRH receptors, without causing the initial testosterone surge and flare associated with agonists [13].


Phase III data showed that abarelix, the first GnRH antagonist approved for use in PCa, was as effective in achieving castration as GnRH agonists with or without anti-androgens and was not associated with a testosterone surge [28–30]. For example, in a European phase III trial, 99% of patients receiving abarelix and 100% of those receiving goserelin/bicalutamide had achieved castration levels of testosterone by day 84 [30]. However, with regard to long-term efficacy, data from the abarelix package insert, based on two randomized trials, warned that in some patients effectiveness of testosterone suppression diminishes with continued dosing and that effectiveness beyond 12 months had not been established [31]. Moreover, after 24 weeks, waning of castration rates appeared to occur more frequently with abarelix than with active controls [32]. Furthermore, escape from castration was found to occur more frequently with abarelix (22%) than with the comparator agonist/anti-androgen (8%) in the phase III European trial, and also time to escape was significantly shorter with abarelix [33].The clinical use of abarelix was also limited by a risk of immediate-onset histamine-mediated hypersensitivity reactions and in 2005 abarelix was voluntarily withdrawn from the US market [34].


The third-generation GnRH antagonist degarelix is currently the most extensively studied and widely available antagonist approved for the treatment of advanced PCa. Degarelix has been the subject of a comprehensive and on-going clinical development programme, including comparative studies with the GnRH agonist leuprolide.


In a large pivotal, 1-year, randomized, open-label study involving 610 patients with PCa (CS21), degarelix was as effective as leuprolide in suppressing testosterone to castrate levels [11]. However, unlike leuprolide, there was no initial testosterone surge or subsequent microsurges with degarelix. Moreover, the reduction in testosterone and PSA levels was significantly faster with degarelix. Some studies have shown that faster PSA reduction may be associated with longer time to progression to hormone-refractory disease and improved OS [14,35]. In addition, PSA progression-free survival (PFS), which is indicative of time to castration-resistant disease [15], was significantly longer for degarelix vs leuprolide (P= 0.05) (Figure 1) [36]. Delaying progression to castrate-resistant disease is clearly desirable as it also delays the initiation of second-line therapies. As well as PSA, degarelix has also shown beneficial effects on bone markers. Thus, in patients with metastatic PCa or with baseline PSA levels ≥50 ng/mL, reductions in serum alkaline phosphatase (S-ALP; a marker of bone formation) were shown to be significantly greater for degarelix than leuprolide [37].

Figure 1.

PSA progression-free survival probability – ie probability of freedom from PSA recurrence or death. Reprinted from Tombal B, Miller K, Boccon-Gibod L et al. Additional analysis of the secondary end point of biochemical recurrence rate in a Phase 3 trial (CS21) comparing degarelix 80 mg vs leuprolide in prostate cancer patients segmented by baseline characteristics. Eur Urol 2010;57:836–842 [36] with permission from Elsevier.

Long-term efficacy

Long-term treatment with degarelix is being evaluated in a 5-year extension trial (CS21A). Patients who completed the pivotal 1-year trial (CS21), either continued on degarelix at the same monthly dose or crossed over from leuprolide to degarelix. An interim analysis, with a median follow-up of 27.5 months, showed that the efficacy of degarelix was maintained in the long term [15]. Thus, PSA PFS hazard rates for those continuing to receive degarelix 240/80 mg in CS21A were similar (0.14 events/year) to those in the first year in the pivotal trial (0.11 events/year; P= 0.464). For patients who crossed from leuprolide to degarelix, PSA PFS hazard rates improved significantly, from 0.20 to 0.08 events/year (P= 0.003). The extension trial also showed that effective suppression of testosterone and PSA was maintained for >3 years in patients continuing to receive degarelix 240/80 mg and in those who crossed over from leuprolide to degarelix. During the first year in CS21, luteinizing hormone (LH) was suppressed at low levels in degarelix and leuprolide groups and this level of suppression was maintained in both groups during the first 3 months of CS21A. Follicle-stimulating hormone (FSH) suppression was greater with degarelix vs leuprolide in the first year during CS21. However, in the first 3 months of CS21A, while FSH suppression was maintained with continuing degarelix, further suppression was noted in the group crossing over from leuprolide to degarelix [15].


The overall incidence of adverse events was similar for degarelix and leuprolide in the pivotal CS21 comparative study [11]. Although patients receiving degarelix experienced a higher incidence of injection-site reactions and chills compared with leuprolide, they experienced a lower incidence of urinary tract infection and musculoskeletal and connective tissue events. No immediate-onset systemic allergic reactions were reported with degarelix. In the long-term extension trial, the overall incidence of adverse events was similar in those who received continuous degarelix and those crossing over from leuprolide after 1 year, and decreased throughout the 4 years of follow-up in both treatment groups [15]. At year 4, the incidence of individual adverse events was low, with no major between-group differences. Crossing over from leuprolide to degarelix was associated with more injection-site reactions in year 2. However, in patients who crossed from leuprolide to degarelix, the incidence of these effects decreased in years 3 and 4 to reach levels similar to those of patients who had received continuous degarelix. Patients who crossed from leuprolide to degarelix experienced an improved musculoskeletal adverse event rate after crossover, becoming similar to that in patients continuing on degarelix.

Degarelix received European regulatory approval in 2009 for advanced hormone-dependent PCa, based on robust efficacy and tolerability data from clinical studies where it displayed fast, profound and sustained testosterone suppression without the systemic allergic reactions associated with abarelix [38]. GnRH antagonists now offer an established alternative to LHRH agonists for ADT in PCa.


In certain subgroups of patients (eg those with later-stage disease), treatment differences between GnRH antagonists and agonists are more evident as event rates are likely to be higher over shorter time periods. Thus, data in CS21 and CS21A trials were analysed for possible treatment differences in patient subgroups with more advanced disease [15,36,37].


Analysis of this subgroup revealed differences in PSA profiles for degarelix 240/80 mg and leuprolide. As well as fewer PSA failures with degarelix vs leuprolide in patients with metastatic disease during the CS21 trial, there was also an initial PSA increase in leuprolide patients which did not occur with degarelix 240/80 mg [36]. In addition, the proportion of patients achieving PSA <4 ng/mL over the study period was higher with degarelix 240/80 mg than with leuprolide. Patients with metastatic disease also revealed differences in effects on bone markers. Elevated S-ALP and bone-specific ALP have been associated with progression of skeletal metastases in PCa [39,40] and may be significant predictors of early death [41–44]. In CS21, all patients with metastatic disease had high baseline S-ALP levels, reflecting the presence of skeletal metastases. However, patients receiving degarelix experienced a faster and more profound control of S-ALP than those receiving leuprolide. Furthermore, patients receiving degarelix maintained S-ALP suppression throughout the study, without the late increases in S-ALP (which might suggest therapy failure) noted with leuprolide. Thus, degarelix appears to offer prolonged control of skeletal metastases compared with GnRH agonists [37].


Patients with baseline PSA >20 ng/mL are at highest risk of PSA failure and in the 1-year degarelix trial (CS21), PSA failure occurred exclusively in this patient subgroup [36]. Moreover, in this subgroup, those receiving degarelix had a significantly longer time to PSA failure compared with those receiving leuprolide (P= 0.04; log-rank). In the extension study, a significant improvement in PSA PFS was observed after crossover from leuprolide to degarelix in patients with baseline PSA >20 ng/mL [15]. Thus, the PSA PFS hazard rate decreased significantly from 0.38 events annually in year 1 to 0.19 events annually after crossover in patients on leuprolide (chi-square test; P= 0.031). The corresponding hazard rates for degarelix were 0.23 during the first year and 0.23 events annually in subsequent years (P= 0.988). In addition, the time for 25% of patients to experience PSA failure or death, was significantly longer (514 days) for degarelix vs leuprolide (303 days, P= 0.01; using data for degarelix beyond 1 year from CS21A) [38] (Figure 2).

Figure 2.

The probability of freedom from PSA failure or death over time for patients with baseline PSA ≥20 ng/mL in the CS21 and CS21A trials. The magnified area of the graph shows the time for 25% of patients with baseline PSA ≥20 ng/mL to experience PSA failure or death (TTP25%). Reprinted from Boccon-Gibod L, van der Meulen E, Persson B-E. An update on the use of gonadotropin-releasing hormone antagonists in prostate cancer. Ther Adv Urol 2011; 3: 127–40 [38] with permission from Sage Publications. KM, Kaplan–Meier.


Studies are on-going to evaluate the effect of degarelix in patients requiring fast symptom relief or prostate volume reduction. For example, the effect of degarelix 240/80 mg on reduction of total prostate volume, relief of lower urinary tract symptoms (LUTS) and improvement of quality of life was compared with goserelin/bicalutamide in a 3-month trial in 182 men with a total prostate volume >30 mL (CS31 [45]). After 3 months of treatment, reduction in total prostate volume for degarelix (−37.2%) and goserelin/bicalutamide (−39.0%) was comparable and fulfilled the predefined criterion for non-inferiority. The reduction in International Prostate Symptom Score (IPSS) was greater with degarelix than with goserelin/bicalutamide, and differences between treatments reached statistical significance in patients with baseline IPSS ≥ 13 (moderate/severe LUTS) (−6.7 vs −4.0; P= 0.02). When examining an individual patient's benefit, the proportion of patients achieving clinically meaningful LUTS relief (ie IPSS decrease of ≥3 points from baseline) was also significantly higher in patients treated with degarelix at week 12 (61.0 vs 44.3%, P= 0.02). Both treatment groups achieved a statistically significant improvement in quality of life from baseline (P < 0.001).

This study shows that degarelix is beneficial for LUTS relief in symptomatic PCa patients and may offer an alternative option to standard LUTS treatment, particularly for men with more advanced/symptomatic disease.


Timing of the introduction of ADT in advanced PCa remains controversial. The underpowered European Organisation for Research and Treatment of Cancer (EORTC) study 30846, found similar OS or cancer-specific survival (CSS) between immediate and delayed endocrine therapy [46]; there was a 22% increase in the risk of death with delayed treatment although the difference was not statistically significant (non-inferiority was also not proved). However, the Eastern Cooperative Oncology Group (ECOG) study 3886 showed that immediate ADT was associated with a significant improvement in OS, CSS and PFS [47]. A review of studies from the pre-PSA era concluded that early ADT provided a relatively small benefit in OS but did not improve CSS although it significantly reduced disease progression and resulting complication rates [48]. In the PSA era, EORTC trial 30891 showed similar results, with a small benefit in OS, but no CSS benefit [49]. American Society of Clinical Oncology guidelines on initial hormonal treatment for androgen-sensitive, metastatic, recurrent or progressive PCa concluded that no specific recommendations could be made regarding the question of early vs deferred ADT until data from studies using modern medical diagnostic/biochemical tests and standardized follow-up schedules become available [50]. EAU guidelines consider that in advanced PCa, immediate ADT (at diagnosis) significantly reduces disease progression and the complication rate due to progression, compared with deferred ADT (at symptomatic progression) but that the survival benefit is at best marginal and not related to CSS [3].


While ADT remains the mainstay of the medical management of advanced/metastatic PCa, for which it is recommended as palliative treatment by current EAU guidelines [3,51], it is starting to play an increasing role in earlier stages of the disease as a neoadjuvant/adjuvant to RT.


The addition of neoadjuvant/adjuvant ADT is now considered the standard of care for high-risk localized (T3a or Gleason score 8–10 or PSA >20 ng/mL) or locally advanced disease treated with radical RT [16]. This multimodal approach is supported by robust data from randomized clinical trials [22,52–57] (Table 2).

Table 2.  Studies of ADT neoadjuvant and adjuvant to radiotherapy: 10-year efficacy outcomes
StudyPatientsTreatmentMedian follow-up (y)10-year efficacy outcomes
MortalitySurvivalLocal failure/ progression (%)DM (%)
Overall/all-cause (%)DSM (%)DFS (%)OS (%)
  1. *P≤ 0.05, **P≤ 0.01, ***P≤ 0.001 vs control group.

  2. ADT, androgen deprivation therapy; AHT, adjuvant hormonal therapy; DM, distant metastases; DFS, disease-free survival; DSM, disease-specific mortality; EORTC, European Organisation for Research and Treatment of Cancer; HR, hazard ratio; NHT, neoadjuvant hormonal therapy; m, months; OS, overall survival; RT, radiotherapy; RTOG, Radiation Therapy Oncology Group; SPGC-7/SFUO-3, Scandinavian Prostate Cancer Group/Swedish Association for Urological Oncology; TROG, Trans-Tasman Radiation Oncology Group; y, years.

RTOG 86-10 (Roach et al. [52])Locally advanced PCa (T2-4, N0-1) (n= 456)Goserelin/flutamide 2 m before and 2 m during RT11.9–13.2 23**11.2***43 35**
RT alone (control)363.434 47
TROG 96.01 (Denham et al. [53])Locally advanced PCa (T2b, T2c, T3 and T4, N0 M0) (n= 802)Goserelin/flutamide (6 m) + RT10.629.2***11.4***  13.3*** 
Goserelin/flutamide (3 m) + RT36.718.915.7***
RT alone (control)42.522.028.2
EORTC 22863 (Bolla et al. [22])Locally advanced (T1–2 poorly differentiated and M0, T3-4, N0-1 M0) (n= 415)Goserelin (3 y) + RT9.1  47.7***58.1***  
RT alone (control)22.739.8
RTOG 85-31 (Pilepich et al. [54])T3 or N1 M0 (n= 977)Goserelin (indefinite/until progression) + RT7.6 16** 49**23***24***
RT alone (control)22393839
RTOG 92-02 (Horwitz et al. [55])T2c-4 or N0-2 PSA <150 ng/mL) (n= 1521)Goserelin/flutamide 4 m before/during RT + goserelin for 2 y11.3  22.5***53.912.3***14.8***
Goserelin/flutamide 4 m before/during RT (control)   13.251.622.222.8
SPGC-7/ SFUO-3 (Widmark et al. [56])Locally advanced PCa (T1b-2 grade 2–3; T3 N0 M0; PSA <70 ng/mL) (n= 875)Leuprolide/flutamide (3 m) + RT + continuous flutamide7.629.6**11.9***    
Leuprolide/flutamide (3 m) + continuous flutamide (control) 39.423.9
PR3/PR07 (Warde et al. [57])Locally advanced (T3/T4 or T2, PSA >40 ng/mL or T2, PSA >20 ng/mL and Gleason ≥8 and N0 /NX, M0) (N= 1205)LHRH agonist or orchidectomy + RT6.0 ADT + RT significantly reduced prostate cancer death risk vs ADT alone: HR 0.54*** ADT + RT significantly improved overall survival at 7 years vs ADT alone: HR 0.77*  
LHRH agonist or orchidectomy alone (control)

Neoadjuvant ADT

Neoadjuvant hormonal therapy (NHT) before definitive RT can produce, on average, a reduction of 25–30% in prostate size [58,59]. The first phase III study of NHT was the Radiation Therapy Oncology Group (RTOG 86-10) trial, which included 456 evaluable patients with locally advanced disease (T2–T4) [60]. Patients received goserelin/flutamide for 2 months prior to RT and then for 2 months during RT, or RT alone. NHT significantly decreased 5-year local progression (46% vs 71%, P < 0.001) and increased PFS with normal PSA levels vs RT alone (36% vs 15%, P < 0.001) [60]. Ten-year OS (43% vs 34%) and median survival (8.7 years vs 7.3 years) favoured NHT over RT alone although these differences did not achieve statistical significance [52]. However, other efficacy variables showed significant superiority in the NHT arm: disease-specific mortality (P < 0.01), distant metastasis (P= 0.006), DFS (P < 0.0001) and biochemical failure (P < 0.0001).

The Trans-Tasman Radiation Oncology Group (TROG) 96.01 trial compared NHT with goserelin/flutamide for 3 or 6 months (starting 2 and 5 months prior to RT, respectively) with RT alone in men with locally advanced PCa [53]. The study included 802 evaluable patients (∼84% high-risk). Analysis of 10-year data showed that NHT for 6 months significantly reduced all-cause mortality (hazard ratio [HR] 0.63, P < 0.0008), PCa-specific mortality (HR 0.49, P= 0.0008) and distant progression (HR 0.49, P= 0.001) vs RT alone; NHT for 3 months did not significantly improve these endpoints compared with RT alone. Also, both 3 and 6 months of NHT significantly reduced PSA progression and local progression and significantly improved event-free survival compared with RT alone [53].

Adjuvant ADT

The benefit of adding ADT to RT is now well established in terms of improvement in local disease control, development of metastasis, DFS and OS. Available evidence supports 2–3 years of adjuvant ADT following RT [22,55], and recent data suggest the bulk of the benefit occurs in the first year [61]. Two trials have shown superior 10-year survival rates with adjuvant ADT. The EORTC 22863 trial evaluated 3 years of goserelin adjuvant to RT vs RT alone in patients with locally advanced, non-metastatic PCa [62]. After a median follow-up of 9.1 years, the 10-year OS rates were 39.8% in patients receiving RT alone and 58.1% in those allocated combined treatment (HR 0.60, P= 0.0004) [22]. Ten-year clinical DFS was also higher (47.7%) in the adjuvant group vs the RT alone group (22.7%; HR 0.42, P < 0.0001). Another trial (RTOG 85-31) in patients who had undergone definitive RT, examined long-term adjuvant goserelin (indefinite or until disease progression) in patients with locally advanced PCa and a high metastatic risk [54]. Adjuvant ADT produced a significantly greater OS rate at 10 years (49%) vs RT alone (39%; P= 0.002) and the survival advantage was greatest in men with Gleason score 8–10 (adjuvant arm, 39%; RT alone, 25%; P= 0.0046). The risk of local failure or distant metastasis was also reduced with adjuvant therapy in this trial.

A third trial (RTOG 92-02) in men with locally advanced PCa compared 4 months of ADT with goserelin/flutamide before and during RT, followed by either no additional therapy or 2 years of adjuvant goserelin [63]. At 10 years, there was no significant improvement in OS with long-term adjuvant ADT, except for the subgroup with Gleason scores of 8–10 (OS in the long-term ADT group was 45.1% vs 31.9%; P= 0.0061) [55]. At 10 years, long-term adjuvant ADT also achieved significantly improved DFS, local progression, distant metastasis and biochemical failure, vs short-term ADT.

ADT adjuvant to RT has also demonstrated superiority to ADT monotherapy for men with locally advanced or high-risk localized PCa. Thus, in the Scandinavian Prostate Cancer Group/Swedish Association for Urological Oncology study, men with locally advanced PCa received ADT (leuprolide/flutamide) for 3 months, followed by RT or no additional treatment while continuing ADT with flutamide. At 10 years, RT plus ADT was associated with significantly reduced overall mortality (relative risk: 0.68; P= 0.004) and PCa-specific mortality (relative risk: 0.44; P < 0.0001) and PSA recurrence (relative risk: 0.16; P < 0.0001; [56]). A separate Intergroup PR3/PR07 study compared lifelong ADT (bilateral orchidectomy or LHRH agonist) with or without RT in 1205 men with locally advanced disease [57]. At a median follow-up of 6.0 years, the addition of RT to ADT significantly improved survival (HR 0.77; P= 0.03) and reduced the disease-specific mortality risk (HR 0.54; P= 0.001) vs ADT alone.


The benefit of adding NHT or ADT adjuvant to RT in patients with intermediate-risk localized PCa (eg stage T2b-T2c, or PSA 10–20 ng/mL, or Gleason score 7) is equivocal [64]. These patients are heterogeneous and often grouped with high-risk patients in clinical studies, making results difficult to interpret. However, NHT to RT has been shown to significantly improve biochemical control [65] and OS [66] in trials enrolling mostly intermediate-risk patients (comprising ∼70−79% of the study population). In addition, in the RTOG 94-08 trial involving 1979 patients with stage T1b, T1c, T2a, or T2b PCa and a PSA ≤20 ng/mL, short-term ADT for 4 months before and during RT was associated with significantly decreased disease-specific mortality and increased OS vs RT alone; the benefit was mainly seen in intermediate-risk, but not low-risk, men (Figure 3) [67].

Figure 3.

Kaplan–Meier estimates of overall survival in patients with (a) low- and (b) intermediate-risk disease treated with ADT + RT vs RT alone. Reprinted from Jones CU, Hunt D, McGowan DG et al. Radiotherapy and short-term androgen deprivation for localized prostate cancer. New Engl J Med 2011; 365: 107–18 [67] with permission from The Publishing Division of the Massachusetts Medical Society. ADT, androgen deprivation therapy; RT, radiotherapy.


The optimal timing and duration of neoadjuvant/adjuvant ADT is still to be established. Current EAU guidelines consider that in locally advanced disease (T3-4 N0 M0), OS is improved by concomitant and adjuvant hormonal therapy for a total duration of 3 years [3]. In a prospective randomized trial (EORTC 22961) in >900 men with locally advanced disease, 6 months of ADT adjuvant to RT was found to be inferior to 3 years of ADT adjuvant to RT; 5-year overall mortality was numerically higher with short-term ADT than with long-term ADT [68]. However, a recent analysis of data from 3666 PCa patients treated with either combined ADT and RT or RT alone concluded that the bulk of the benefit of adjuvant ADT occurs in the first year (approximately 85% of that obtainable from a full 3 years, on average), thus questioning recommendations for 3 years of ADT vs 2 years [61].

The optimal duration of neoadjuvant ADT (with RT) remains to be defined. RTOG 86-10 showed that therapy for 4 months (neoadjuvant for 2 months plus concomitant for 2 months) was beneficial [52], whereas in TROG 96.01, 6-month NHT decreased distant progression, PCa-specific mortality and all-cause mortality, whereas 3-month NHT had no effect on these variables, vs RT alone [53]. Another trial comparing 3 vs 8 months of neoadjuvant ADT in localized disease did not show a significant difference in biochemical DFS or OS between treatment arms (except for the high-risk group) [69]. It has been suggested that biochemical response to NHT, rather than its duration, may be more important in determining the benefit of combined hormonal and RT treatment in the neoadjuvant setting [70]. Further analysis of data from the trial comparing 8- vs 3-month NHT showed significantly higher biochemical DFS in patients who achieved an excellent biochemical response, indicated by pre-RT post-hormone therapy PSA ≤0.1 ng/mL. Consequently, it is possible that achievement of a rapid decrease in PSA in response to NHT may allow minimization of ADT duration and thus of any related adverse events.

In the absence of comparative studies of different forms of ADT given as NHT or adjuvant hormonal therapy (AHT), it is not clear if any particular class/agent provides particular benefits when used in this manner. Nevertheless, it seems likely that agents with fast symptom and volume control and rapid PSA control may be beneficial in this regard.


Surgical treatment of clinical stage T3 PCa has traditionally been discouraged; however, EAU guidelines now acknowledge that, while still controversial, it is increasingly apparent that surgery has a place in the treatment of locally advanced disease [3]. While 56–78% of patients primarily treated by surgery eventually require adjuvant or salvage RT or ADT [71,72], excellent 5-, 10- and 15-year OS and CSS rates have been reported (eg references 71–73]. However, RP for clinical T3 PCa requires sufficient surgical expertise to achieve acceptable levels of morbidity.

The rationale for using ADT neoadjuvant to RP is to decrease tumour size prior to surgery to improve surgical curability [74]. Evidence for the use of neoadjuvant/adjuvant ADT with RP is less compelling than that for RT. Current EAU guidelines consider that further studies are needed to investigate the application of ADT neo-adjuvant to RP and that while NHT does not provide a significant advantage in OS or DFS over RP alone it substantially improves local pathological variables [3]. A systematic review of clinical trials in localized or locally advanced PCa showed that NHT prior to RP significantly reduced positive margin rates (P < 0.00001), organ confinement (P < 0.0001) and lymph-node invasion (P < 0.02) vs RP alone, but had no effect on OS or DFS [74]. A review of AHT following RP found no effect on OS but improved disease-specific survival (P= 0.0004) in one study; DFS was also better with AHT (P < 0.00001) [75]. These findings are also reflected in EAU guidelines.

ECOG study 3886 in men with node-positive PCa who have undergone RP and pelvic lymphadenectomy showed that immediate ADT significantly improved overall, disease-specific, and PFS compared with the withholding of ADT until disease progression occurs [47]. ADT after RP and pelvic lymphadenectomy also significantly reduced progression and overall and disease-specific mortality.

The role of AHT with RT after RP is also being evaluated. In RTOG 9601, patients with rising PSA post-RP received RT alone or RT plus bicalutamide (for 24 months). Interim results showed significant improvements in freedom from PSA progression and development of metastatic disease in the bicalutamide group (there were too few deaths to allow evaluation of OS, the primary endpoint, at this stage in the study and longer-term follow-up is awaited) [76].

The RADICALS (Radiotherapy and Androgen Deprivation In Combination After Local Surgery; NCT00541047) trial, which is currently underway in the UK, Canada, Denmark and the Republic of Ireland, will evaluate optimal timing of RT and duration of hormone therapy after RP [77,78]. Thus, after RP, patients will be randomized to adjuvant or early salvage RT. Within each of these arms patients will receive either RT alone or RT plus 6 months or 2 years of adjuvant ADT.


How can we explain the success of combining a potentially curative local treatment of the primary tumour with endocrine therapy? In trials combining RT with ADT, the improved outcome may be explained in part by the radiosensitizing effect of ADT but may also result from a systemic effect of ADT [79]. In the EORTC 22863 trial of adjuvant ADT in locally advanced PCa [62], there would appear to be a very high risk of subclinical metastatic disease, while other trials included patients with overt lymph node metastatic disease [47,54]. Indeed, the ECOG 3886 study, with RP as local treatment, showed a survival benefit for immediate ADT of similar magnitude to the benefit observed in the concomitant and adjuvant RT studies [47]. It is expected that this metastatic subset contributes most to the early deaths in these studies and it seems unlikely that an improved local control alone is able to influence this early mortality risk. The success of ADT plus local therapy in such patients may reflect that treatment of the primary tumour in the presence of metastatic disease may prevent further metastases coming from clonal populations in the primary, as suggested by Messing et al. [80]. This would appear to be supported by a recent study which confirmed that most, if not all, metastatic PCas have monoclonal origins [81]. The significance of the role of aggressive treatment of the primary combined with endocrine treatment in lymph-node positive disease was confirmed in a multicentre matched analysis which showed that adjuvant combination treatment with long-term hormonal therapy and RT after RP is associated with significantly better long-term survival compared with AHT alone in node-positive PCa [82]. Moreover, the survival benefit was achieved regardless of the extent of nodal invasion.

Finally, a recent systematic review of randomized studies of ADT in PCa which examined a possible interaction between local treatment and ADT concluded that the role of local treatment in locally advanced and lymph node metastatic disease is important and supports an interaction between local treatment and systemic ADT [79]. Indeed, if the primary tumour continues to produce metastases then the key treatment factor in this setting may be effective local treatment of the primary to prevent this, rather than the type of local therapy (RT or RP) utilized.


The phase III Southwest Oncology Group (SWOG) S9921 trial is investigating the potential role of AHT plus chemotherapy following RP [83]. Following RP, men with high-risk features at RP (n= 983) were assigned to goserelin and bicalutamide for 2 years, either alone or combined with mitoxantrone. Interim data for the AHT-alone control arm showed a 5-year biochemical-failure free survival rate of 92.5%, and a 5-year OS rate of 95.9%. The final primary treatment comparison results are awaited.


In the 70 years since Huggins and Hodges first demonstrated the favourable impact of ADT in PCa, it has continued to evolve as a therapeutic strategy. For many years, LHRH agonists were the standard of care for ADT in PCa. However, their use has been associated with suboptimal testosterone control which, together with unwanted surge and flare effects, may influence disease control. The more recent introduction of GnRH antagonists, now offers an alternative first-line approach to ADT in PCa. Thus, agents such as degarelix, the most extensively studied and widely available member of this therapeutic class, offer faster and more effective control of testosterone, rapid PSA control and improved PSA PFS (potentially prolonging the time to castrate-resistant disease) compared with agonists. Moreover, recent clinical data confirm the long-term efficacy and tolerability of this agent, with maintenance of testosterone and PSA suppression and control for >3 years. Moreover, degarelix data indicate that the pharmacological properties of GnRH antagonists translate into clinical advantages, which may be more pronounced in patients with metastatic disease, high baseline PSA (>20 ng/mL) and highly symptomatic patients.

In recent years, the role of ADT has expanded beyond advanced disease and it is now starting to play an increasing role in earlier disease stages. There is increased OS and PFS when ADT is used adjuvant to RT for locally advanced and high-risk localized PCa and this combined modality therapy is superior to RT or ADT alone. Indeed, neoadjuvant/adjuvant ADT is now considered as the standard of care for locally advanced or high-risk localized disease treated with radical RT. While it remains to be seen which ADT provides most benefit in this therapeutic approach, it is likely that agents that offer rapid control of symptoms, prostate volume and PSA control may be of particular benefit in this context. The rapid testosterone and PSA suppression achieved with GnRH antagonists may therefore suggest potential advantages for these agents in clinical settings that utilize neoadjuvant/adjuvant therapy; studies of these agents in this therapeutic context are on-going.


Medical writing assistance (funded by Ferring Pharmaceuticals) was provided by Thomas Lavelle of Bioscript Stirling Ltd.


F. Schröder: Consultant/Advisor: Ferring, Genprobe, GlaxoSmithKline.

E. D. Crawford: Grant/Research: Ferring, Genprobe. Consultant/Advisor/Speakers' Bureau: Ferring, GlaxoSmithKline, Indevus, Poinard, sanofi aventis, Soar Biodynamics.

K. Axcrona: Consultant/Advisor/Speaker: Astellas Pharma. Research grant: Ferring.

H. Payne: has attended and received honorarium for advisory boards and served as a consultant for AstraZeneca, Janssen, Johnson and Johnson, sanofi aventis, Takeda, Amgen, Ferring and Novartis.

T. E. Keane: Ferring Pharmaceuticals – Advisor/Speaker; Amgen – Advisor/Speaker; Endo Pharmaceuticals – Advisor/Speaker.