SEARCH

SEARCH BY CITATION

Keywords:

  • testosterone;
  • prostatic neoplasms;
  • radiotherapy;
  • androgen blockade

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Study Type – Therapy (case series)

Level of Evidence 4

What's known on the subject? and What does the study add?

The return of testosterone to normal levels following short-course androgen blockade in prostate cancer is variable. Factors associated with a longer time to recovery include older age and lower baseline testosterone level. In this study, we found that among men treated with 6 months of combined androgen blockade and radiation therapy, higher biopsy Gleason grade was associated with a shorter time to testosterone normalization.

OBJECTIVE

  • • 
    To determine whether the biopsy Gleason score is associated with duration of testosterone suppression following 6 months of combined androgen blockade (CAB) and radiation therapy (RT) in men with prostate cancer (PCa).

PATIENTS AND METHODS

  • • 
    The study cohort consisted of 221 men with PCa treated with RT and 6 months of CAB between 1996 and 2005.
  • • 
    We defined the duration of testosterone suppression as the time between the last day of CAB and the date the testosterone returned to ≥252 ng/dL. We used Cox regression multivariable analysis to relate biopsy Gleason score to duration of testosterone suppression following cessation of CAB.

RESULTS

  • • 
    A biopsy Gleason score of 8–10 had an adjusted hazard ratio (AHR) of 1.56 (95% confidence interval [CI] 1.04, 2.34; P= 0.03) for a shorter time to testosterone normalization relative to Gleason 6. Specifically, the 51 men with biopsy Gleason score of 8–10 had a median time to testosterone normalization of 17.0 months compared with 22.1 months and 23.8 months for those with biopsy Gleason ≤6 and 7, respectively.
  • • 
    Increasing age was significantly associated with a longer duration of testosterone suppression (AHR of 0.95 [95% CI 0.92, 0.97; P < 0.001]) as was a higher baseline PSA (AHR 0.82 [95% CI 0.69, 0.97; P= 0.02]).

CONCLUSION

  • • 
    A biopsy Gleason score of 8–10 was associated with a shorter period of testosterone suppression following 6 months of CAB and RT. These data are consistent with the hypothesis that a factor released from high-grade PCa cells may impact on testosterone production.

Abbreviations
AHR

adjusted hazard ratio

CAB

combined androgen blockade

IQR

inter-quartile range

LH

luteinizing hormone

LHRH

luteinizing hormone releasing hormone

PCa

prostate cancer

RT

radiation therapy

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

For men with intermediate to high risk prostate cancer, randomized trials have documented a survival advantage to the addition of 6 months of androgen suppression to external beam radiation over radiation alone establishing this regimen as an option for men with these higher risk tumours [1–3]. Following cessation of androgen suppression, the time to normalization of serum testosterone concentrations can be quite variable and prior analyses have demonstrated that a longer duration of testosterone suppression is associated with a decreased risk of prostate cancer specific mortality [4], an effect that may be modified by a man's baseline comorbid conditions [5].

Several groups have investigated factors associated with duration of androgen suppression following luteinizing hormone-releasing hormone (LHRH) agonist administration. Early observations of LHRH agonists in men with prostate cancer found significant heterogeneity in the duration of castrate testosterone concentrations following cessation of therapy [6–9]. While initial work in healthy younger men showed a relatively prompt return to baseline testosterone [10], in older men with prostate cancer, after a single 3 month depot injection, normalization of the testosterone frequently took 6 months or more [7,8]. Subsequent analyses found that increasing age [11–13], lower baseline testosterone concentration [14] and duration of induced testosterone suppression [9,12,13] were significantly associated with a prolonged time to testosterone recovery. Whether a tumour-specific factor could relate to the duration of testosterone suppression has not been well-studied. Early observations suggested an increase in serum testosterone following radical prostatectomy [15] with more limited data suggesting that this effect might be variable according to Gleason score [16]. How Gleason might impact on the recovery of testosterone following definitive treatment with short-course androgen suppression and radiation has not been reported.

In this retrospective study, our objective was to determine whether the biopsy Gleason score was associated with the duration of testosterone suppression in a cohort of men with prostate cancer uniformly treated with 6 months of combined androgen blockade (CAB) and radiation (RT).

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The study cohort consisted of 221 men who received RT and 6 months of CAB between 1996 and 2005 at a single academic and associated suburban centre for higher risk prostate cancer. The cohort was identified from a registry including men from two prospective studies [1,17] who were included in this analysis given their uniform treatment with radiation and 6 months of CAB. Men were considered to have higher risk disease if they had at least one of the following features: PSA >10 ng/mL, Gleason score ≥7 or 2002 AJCC clinical T category ≥T2b [18]. Work-up included a prostate specific antigen (PSA) value, physical examination, bone scan and computerized tomographic or magnetic resonance imaging scan. All biopsy material was re-reviewed by a single pathologist with genitourinary expertize. The radiation techniques have been described previously [1,17] but generally comprized of a two-phase, three-dimensional based approach to the prostate and seminal vesicles to a total of 70.35 Gy over 36 fractions. CAB consisted of two, 3 month injections of an LHRH agonist (leuprolide acetate 22.5 mg every 3 months or goserelin 10.8 mg every 3 months) and a non-steroidal anti-androgen (flutamide 250 mg every 8 h or bicalutamide 50 mg daily). The anti-androgen was discontinued on day 85 after the second administration of the LHRH agonist.

This study was approved by the Dana-Farber Cancer Institute Institutional Review Board.

Men were seen in follow-up every 3 months for 2 years, every 6 months until 5 years and annually thereafter. Within 1 week of each follow-up visit, a PSA and total serum testosterone were measured. Serum testosterone was measured using the Bayer assay (Tarrytown, NY) with a lower limit of normal for men ≥18 years of age of 280 ng/dL. Because not all men had a baseline testosterone concentration, for the purposes of analysis, we defined the duration of testosterone suppression as the time between the last day of the 6 months of oral non-steroidal anti-androgen (day 85 following the second LHRH administration) and the date the testosterone concentration was ≥252 ng/dL, a threshold 10% below the lower bound of normal for the assay. Given the daily variation of approximately 10% for the assay, we selected this threshold to identify men who had returned to a normal level. Testosterone measurements were complete for every visit in all patients in the study cohort.

Descriptive statistics were used to characterize the study cohort. The median and associated 95% confidence intervals (CIs) for the duration of testosterone suppression were calculated. Men were censored at the last follow-up, at the time of re-initiation of androgen suppression therapy following clinical failure or at the time of death, whichever date was earlier. A Cox proportional hazards model was used to assess whether age (continuous), PSA (continuous), T-category (T3 or 4 vs T1 or 2), biopsy Gleason score (8–10 or 7 vs ≤6), and in a subset analysis, baseline testosterone concentration (continuous) were associated with the duration of testosterone suppression [19]. The PSA concentration was log transformed to allow it to conform to a normal distribution. The adjusted hazards ratio (AHR) for the duration of testosterone suppression with associated 95% CIs and two-sided P values were calculated for each covariate. The adequacy of the model fit was assessed using Schoenfeld residuals. The cumulative incidence of testosterone recovery was calculated and graphically displayed stratified by biopsy Gleason score (8–10 vs 7 or less). These estimates were compared using a k-sample P value and adjustments using a Bonferonni correction were made for multiple (n= 3) comparisons. SAS software (version 9.2; SAS Institute, Cary, NC) was used for all analyses.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

The median age of the cohort was 68.8 years and the median baseline PSA was 8.3 ng/mL (Table 1). More than half had a Gleason score of 7, of whom 54% had Gleason 3 + 4. Body mass index (BMI) was available for 120 men, of whom 27% had a BMI >30 kg m2. Baseline testosterone concentrations were available on 141 men with a median value of 437 ng/dL. The baseline testosterone was not statistically significantly lower in men with Gleason 8 disease compared with Gleason 7 or ≤6 (402 ng/dL, 430 ng/dL and 453 ng/dL, respectively). The median follow-up from the time of treatment for the cohort was 6.6 years (inter-quartile range [IQR] 4.3–8.0 years).

Table 1. Baseline characteristics for the study cohort
Patient characteristicNumber of patients (%) n= 221
  • *

    n= 141 for the baseline testosterone.

Age, years 
 Median (IQR)68.8 (46.7–73.4)
 <6565 (29)
 65–6959 (27)
 ≥7097 (44)
PSA, ng/mL 
 Median (IQR)8.3 (5.6–14)
Gleason score 
 ≤653 (24)
 7117 (53)
 8–1051 (23)
Clinical T-category 
 T1c97 (44)
 T2a44 (20)
 T2b41 (19)
 T2c19 (9)
 T3a11 (5)
 T3b8 (4)
Baseline testosterone*, ng/dL
 Median (IQR)437 (324–554)

The median duration of testosterone suppression was 22.2 months (IQR 19.9–24.3) for the entire cohort and 20 men (9%) did not achieve a normal testosterone during follow-up. None of these 20 men died during follow-up and all had a Gleason score ≤7. On multivariate analysis (Table 2) with a model including baseline PSA, biopsy Gleason score, T category and age, factors associated with a shorter time to testosterone normalization included biopsy Gleason 8–10 compared with Gleason ≤6 (AHR 1.56; 95% CI 1.04, 2.34; P= 0.03), younger age (AHR 0.95; 95% CI 0.92, 0.97; P < 0.001), and a lower baseline PSA (AHR 0.82; 95% CI 0.69, 0.97; P= 0.02). The median duration of testosterone suppression for men with Gleason 8–10 was 17.0 months (95% CI 13.4, 24.2 months) compared with 23.8 months (95% CI 20.8, 27.8 months) and 22.1 months (95% CI 17.4, 24.6 months) for those with Gleason 7 and ≤6, respectively. There was no statistically significant difference in the duration of testosterone suppression between men with Gleason 7 and those with Gleason ≤6. As shown in Fig. 1, cumulative incidence estimates of time to testosterone normalization for men with Gleason score 8–10 cancer was significantly shorter than men with biopsy Gleason score 7 or less disease (P= 0.006).

Table 2. Multivariate Cox regression model for time to normalization of testosterone
ParameterAdjusted hazard ratio (95% CI) P
  1. *log transformed; †Gleason 6 as reference; ‡T1–2 as reference.

PSA*0.82 (0.69, 0.97)0.02
Gleason 70.73 (0.52, 1.04)0.08
Gleason ≥81.56 (1.04, 2.34)0.03
T3 or T41.25 (0.73, 2.12)0.42
Age0.95 (0.92, 0.97)<0.001
image

Figure 1. Cumulative incidence plot for the time to rebound of testosterone to within 10% of the lower limit of normal for Gleason score ≤7 vs 8–10 following 6 months of total androgen blockade and radiation (P= 0.006). Numbers below the x-axis denote the number of men at risk at each time point for the two groups

Download figure to PowerPoint

In exploratory analyses, baseline testosterone was not significantly related to the time to normalization among the 141 with baseline testosterone measured (AHR 1.05; 95% CI 0.95, 1.17; P= 0.35). Including baseline testosterone did not alter the significant relationship between Gleason 8 and age with the duration of testosterone suppression. Similarly, amongst men with BMI recorded, only age and Gleason 8 were significant predictors of the duration of testosterone suppression. Stratifying by age, Gleason 8–10 remained a significant predictor of duration of testosterone suppression in men <65 years old (AHR 3.25; 95% CI 1.25, 8.46; P= 0.02) but not among men 65–70 years or 70 years and older. There was no significant statistical interaction between age and Gleason score and the time to testosterone recovery.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

Testosterone is the central hormone in the growth and maintenance of the prostate and it has long been known to play an essential role in prostate cancer [20]. Testosterone is synthesized in the Leydig cells of the testes and subsequently converted to the more active form of dihydrotestosterone by 5α-reductase within the prostate. Its synthesis is predominantly under the control of the pituitary gonadotropin luteinizing hormone (LH) with high concentrations of testosterone feeding back on the hypothalamus and eventually the pituitary to reduce LH secretion.

In this study, we found that the time to recovery to near normal testosterone concentrations following 6 months of a LHRH agonist and an anti-androgen plus definitive radiation was significantly associated with age, baseline PSA and biopsy Gleason score with Gleason 8–10 tumours having a significantly shorter time to normalization. In exploratory analyses in smaller subgroups of this study, baseline testosterone concentration and BMI were unrelated to the duration of testosterone suppression. Further, the relationship of high Gleason score with shortened time to testosterone normalization appeared to be most pronounced in men younger than 65 years of age.

To our knowledge, this is the first report showing a relationship between Gleason score and duration of testosterone suppression for men treated with primary radiation and 6 months of CAB. Pickles et al. [12] showed that increasing age and longer duration of induced androgen suppression were associated with a longer time to testosterone recovery in men treated with primary androgen deprivation and radiation but it does not appear that Gleason score was included in their multivariate model. Several prior studies have included the baseline Gleason score in models predicting time to testosterone recovery following cessation of medical castration in the salvage setting after local therapy. In a cohort of 80 men with rising PSA after definitive local treatment enrolled in a randomized trial of 6 months of a LHRH agonist plus or minus thalidomide, Gulley et al. [14] found that the median time to recovery to a normal testosterone concentration was 16.6 weeks. They investigated the influence of Gleason score, PSA concentration, prior definitive therapy, prior androgen suppression, patient age and baseline testosterone concentration on time to normalization and found that only a low baseline testosterone concentration was significantly associated with duration of suppression. A subsequent, larger analysis of 129 men from this same study once again showed no association between time to recovery of normal testosterone concentration and Gleason score, PSA, prior androgen suppression and the local treatment utilized [21]. In another study addressing this question in the salvage setting, among 141 men enrolled on a phase II protocol of 2 years of androgen suppression and radiation following prostatectomy, after evaluating age, androgen suppression duration, baseline testosterone concentration, pretreatment PSA, radiation dose, pathologic T category, Gleason and prior androgen suppression, only age was significantly associated with time to recovery of normal testosterone concentration [22]. This lack of association between Gleason score and duration of testosterone suppression for men treated for salvage may have little bearing on the intact prostate setting as the bulk of disease is removed and prior androgen suppression may have altered the relationship between initial Gleason and time to testosterone normalization.

This study was not designed to determine the biologic mechanism for the observed finding. The uniform treatment of men in this cohort makes it unlikely that the observation is the result of differential treatment such as radiation to the testes [23–25]. It has been recently observed that as prostate cancers become castration resistant, they appear to be capable of producing sufficient local testosterone to be detectable within the gland [26]. While sufficient to bind to the androgen receptor in the prostate, this process is observed in the setting of castrate serum concentrations of testosterone and is therefore an unlikely explanation for our findings. There are several reports showing a modest increase in testosterone following radical prostatectomy [15,27] with more limited data suggesting that this effect is most prominent in those with higher Gleason grades [16]. These data suggest that there may be a factor released by high-grade prostate tumours which suppresses testosterone and by treating the disease, this inhibitory action is removed. While our data are consistent with this observation, additional work will be needed to characterize why testosterone recovers more quickly following treatment of higher-grade tumours.

The biology underlying the association between a higher baseline PSA and longer time to testosterone recovery is also unclear. Our data suggest that clinically this is a modest effect as we modelled the log-transformed PSA. Future studies will need to investigate further this relationship.

This study is limited by the lack of complete hormonal data in all patients. For example, we do not have associated LH levels making it difficult to show concordance between LH and testosterone recovery post-treatment. Baseline testosterone concentrations were available on 141 of the 221 included men but on multivariate analysis, including this variable did not change the reported relationship between Gleason and testosterone recovery time. Neither could we identify any differential measurement of baseline testosterone based on Gleason score. It is notable that we did not find a ‘dose effect’ for Gleason as it relates to the time to testosterone normalization. Specifically, we found no differences in the time it took to observe normalization of testosterone between men with biopsy Gleason 6 and those with 7. This observation could suggest some biologic threshold above Gleason 7 which influences the testosterone concentration or may reflect a relative lack of power in this study to see smaller differences between Gleason 6 and 7. The sample size in this cohort is relatively modest and the observations made here will need confirmation.

The clinical implications for our finding have been previously discussed [4,5]. The duration of testosterone suppression in this setting is independently related to mortality and, as such, understanding the factors associated with recovery of testosterone is important. Studies remain to be performed to discern if the difference in time to serum testosterone recovery can be eradicated or reduced with the use of more complete blockade of the androgen-axis as with the inclusion of medications such as abiraterone acetate [28]. The complex interplay of the potential factors involved in this observation make it difficult to infer the underlying mechanism and a more full understanding will require dedicated investigations of baseline hormonal factors as well as the time course of normalization following treatment across multiple hormones.

In summary, we found that a biopsy Gleason score of 8–10 is associated with a shorter period of testosterone suppression following 6 months of CAB and RT. These data are consistent with the hypothesis that a factor released from high-grade PCa cells may impact on testosterone production.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES

NEM is supported by a Young Investigator award from the Prostate Cancer Foundation. PLN is supported by a generous gift from an anonymous donor.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENTS
  8. CONFLICT OF INTEREST
  9. REFERENCES
  • 1
    D'Amico AV, Manola J, Loffredo M, Renshaw AA, DellaCroce A, Kantoff PW. 6-month androgen suppression plus radiation therapy vs radiation therapy alone for patients with clinically localized prostate cancer: a randomized controlled trial. JAMA 2004; 292: 8217
  • 2
    Denham JW, Steigler A, Lamb DS et al. Short-term androgen deprivation and radiotherapy for locally advanced prostate cancer: results from the Trans-Tasman Radiation Oncology Group 96.01 randomised controlled trial. Lancet Oncol 2005; 6: 84150
  • 3
    Jones CU, Hunt D, McGowan DG et al. Radiotherapy and short-term androgen deprivation for localized prostate cancer. N Engl J Med 2011; 365: 10718
  • 4
    D'Amico AV, Renshaw AA, Loffredo B, Chen M-H. Duration of testosterone suppression and the risk of death from prostate cancer in men treated using radiation and 6 months of hormone therapy. Cancer 2007; 110: 17238
  • 5
    D'Amico AV, Chen MH, Renshaw AA, Loffredo M, Kantoff PW. Interval to testosterone recovery after hormonal therapy for prostate cancer and risk of death. Int J Radiat Oncol Biol Phys 2009; 75: 105
  • 6
    Goldenberg SL, Bruchovsky N, Gleave ME, Sullivan LD, Akakura K. Intermittent androgen suppression in the treatment of prostate cancer: a preliminary report. Urology 1995; 45: 83944; discussion 844–5
  • 7
    Oefelein MG. Time to normalization of serum testosterone after 3-month luteinizing hormone-releasing hormone agonist administered in the neoadjuvant setting: implications for dosing schedule and neoadjuvant study consideration. J Urol 1998; 160: 16858
  • 8
    Hall MC, Fritzsch RJ, Sagalowsky AI, Ahrens A, Petty B, Roehrborn CG. Prospective determination of the hormonal response after cessation of luteinizing hormone-releasing hormone agonist treatment in patients with prostate cancer. Urology 1999; 53: 898902; discussion 902–3
  • 9
    Nejat RJ, Rashid HH, Bagiella E, Katz AE, Benson MC. A prospective analysis of time to normalization of serum testosterone after withdrawal of androgen deprivation therapy. J Urol 2000; 164: 18914
  • 10
    Linde R, Doelle GC, Alexander N et al. Reversible inhibition of testicular steroidogenesis and spermatogenesis by a potent gonadotropin-releasing hormone agonist in normal men: an approach toward the development of a male contraceptive. N Engl J Med 1981; 305: 6637
  • 11
    Oefelein MG. Serum testosterone-based luteinizing hormone-releasing hormone agonist redosing schedule for chronic androgen ablation: a phase I assessment. Urology 1999; 54: 6949
  • 12
    Pickles T, Agranovich A, Berthelet E et al. Testosterone recovery following prolonged adjuvant androgen ablation for prostate carcinoma. Cancer 2002; 94: 3627
  • 13
    Kaku H, Saika T, Tsushima T et al. Time course of serum testosterone and luteinizing hormone levels after cessation of long-term luteinizing hormone-releasing hormone agonist treatment in patients with prostate cancer. Prostate 2006; 66: 43944
  • 14
    Gulley JL, Figg WD, Steinberg SM et al. A prospective analysis of the time to normalization of serum androgens following 6 months of androgen deprivation therapy in patients on a randomized phase III clinical trial using limited hormonal therapy. J Urol 2005; 173: 156771
  • 15
    Miller LR, Partin AW, Chan DW et al. Influence of radical prostatectomy on serum hormone levels. J Urol 1998; 160: 44953
  • 16
    Madersbacher S, Schatzl G, Bieglmayer C et al. Impact of radical prostatectomy and TURP on the hypothalamic-pituitary-gonadal hormone axis. Urology 2002; 60: 86974
  • 17
    Cohen MS, Hanley RS, Kurteva T et al. Comparing the Gleason prostate biopsy and Gleason prostatectomy grading system: the Lahey Clinic Medical Center experience and an international meta-analysis. Eur Urol 2008; 54: 37181
  • 18
    Greene FL, American Joint Committee on Cancer, American Cancer Society. AJCC Cancer Staging Manual, 6th edn. New York: Springer-Verlag, 2002
  • 19
    Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Truncated Data. New York: Springer, 1997
  • 20
    Huggins C, Hodges CV. Studies on prostatic cancer. I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1941; 1: 2937
  • 21
    Gulley JL, Aragon-Ching JB, Steinberg SM et al. Kinetics of serum androgen normalization and factors associated with testosterone reserve after limited androgen deprivation therapy for nonmetastatic prostate cancer. J Urol 2008; 180: 14327; discussion 1437
  • 22
    Yoon FH, Gardner SL, Danjoux C, Morton G, Cheung P, Choo R. Testosterone recovery after prolonged androgen suppression in patients with prostate cancer. J Urol 2008; 180: 143843; discussion 1443–4
  • 23
    King CR, Maxim PG, Hsu A, Kapp DS. Incidental testicular irradiation from prostate IMRT: it all adds up. Int J Radiat Oncol Biol Phys 2010; 77: 4849
  • 24
    Zagars GK, Pollack A. Serum testosterone levels after external beam radiation for clinically localized prostate cancer. Int J Radiat Oncol Biol Phys 1997; 39: 859
  • 25
    Pickles T, Graham P. What happens to testosterone after prostate radiation monotherapy and does it matter? J Urol 2002; 167: 244852
  • 26
    Locke JA, Guns ES, Lubik AA et al. Androgen levels increase by intratumoral de novo steroidogenesis during progression of castration-resistant prostate cancer. Cancer Res 2008; 68: 640715
  • 27
    Olsson M, Ekstrom L, Schulze J et al. Radical prostatectomy: influence on serum and urinary androgen levels. Prostate 2010; 70: 2005
  • 28
    de Bono JS, Logothetis CJ, Molina A et al. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med 2011; 364: 19952005