Duration of androgen deprivation therapy in high-risk prostate cancer: A randomized trial

Reporter: Jacob Shabason, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 10, 2013

Presenter: Abdenour Nabid, MD
Presenter's Affiliation: Centre Hospitalier Universitaire de Québec, Québec, QC, Canada
Abstract #: LBA4510


  • Radiotherapy and long-term androgen deprivation therapy (ADT) is a standard treatment for patients with high-risk prostate cancer.
  • Prior trials have shown improvements in overall survival in high-risk patients treated with 36 months of ADT compared to both no ADT (EORTC 22863) and 6 moths of ADT (EORTC 22961).
  • However, the optimal duration of ADT is not yet completely defined.
  • ADT had significant toxicity including sexual, musculoskeletal, metabolic and cardiovascular toxicity.
  • The authors performed a phase III, randomized trial comparing the outcomes of 36 vs.18 months of ADT with radiotherapy in patients with high- risk prostate cancer.

Materials and Methods

  • The trial included high-risk (T3-4, PSA >20 ng/ml or Gleason score >7), node negative and non-metastatic prostate cancer patients.
  • All patients received pelvic radiation (whole pelvis 44 Gy/4 ½ weeks, prostate 70 Gy/7 weeks) and were randomized to 36 (arm 1) vs.18 months (arm 2) of ADT (neo- adjuvant, concomitant, adjuvant).
  • ADT consisted of bicalutamide 50 mg for one month and goserelin 10.8 mg every three months for 36 vs. 18 months.
  • The primary endpoints were overall survival (OS), disease specific survival (DSS) and quality of life (QOL).
  • OS and DSS rates were compared between arms with Kaplan-Meier log rank test and Cox regression.
  • Secondary endpoints included disease free survival (DFS), biochemical failure (BCF) and site of first relapse.

Materials and Methods

  • From October 2000 to January 2008, 310 patients were randomized to 36 months of ADT and 320 to 18 months of ADT.
  • Patients' characteristics were well balanced between the two arms (median age 71 years, median PSA 16 ng/ml, median Gleason score 8 and most patients had T2-T3 disease).
  • At a median follow-up of 78 months, 80/310 patients (25.8%) in arm 1 and 85/320 (26.6%) in arm 2 had died (p=0.829).
  • 113 patients died of causes other than prostate cancer.
  • Overall and cancer-specific survival hazard ratios were 1.15 (0.85-1.56), p=0.366 and 1.07 (0.62-1.84), p=0.819, respectively.
  • Five year OS was 91.1% (87.9-94.3) vs. 86.1% (82.3-90.0), p=0.06 and 5 year DSS was 96.6% (94.5-98.7) vs. 95.3% (92.8-97.7), p=0.427.
  • Ten year OS was 61.9% (54.1-69.7) vs. 58.6% (49.8-67.4), p=0.275 and 10 year DSS was 84.1% (77.6-90.6) vs. 83.7% (76.3-91.1), p=0.819.
  • There were also no significant differences in rates of biochemical, regional or distant failures between the two arms.

Author's Conclusions

  • With a median follow up of 78 months this randomized study indicates that long-term ADT can be safely reduced from 36 to 18 months without compromising any outcomes in patients with high-risk prostate cancer.
  • This decrease will presumably decrease toxicity related to ADT and improve QOL, however the analysis on quality of life is still pending analysis.

Clinical Implications

  • A standard treatment approach of patients with high-risk prostate cancer is radiotherapy with 2-3 years of neoadjuvant, concomitant and adjuvant ADT.
  • Due to the fact that ADT has significant morbidity, there is much interest in attempting to decrease the duration of this therapy.
  • This trial indicates that the duration of ADT can safely be decreased from 36 to 18 months with similar outcomes.
  • This shortened therapy would presumably be better tolerated, however toxicity data from the trial is still pending.
  • In addition, this trial was started before the benefit of dose-escalated radiotherapy (of greater then 78 Gy) was established in prostate cancer and all patients were treated to 70 Gy.
  • With dose escalated radiotherapy the duration of ADT could potentially be safely shortened to less then 18 months and further research could be designed to address this question.