A phase III randomized trial of intermittent versus continuous androgen suppression for PSA progression after radical therapy (NCIC CTG PR.7/SWOG JPR.7/CTSU JPR.7/ UK Intercontinental Trial CRUKE/01/013)

Reporter: Lara Bonner Millar, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 5, 2011

Presenter: JM Crook
Presenter's Institution: British Columbia Cancer Agency, Kelowna, BC, Canada


  • Androgen-deprivation therapy (ADT) is the standard treatment for metastatic and recurrent prostate cancer, and is used frequently as adjuvant treatment for high-risk localized prostate cancer. Continuous hormonal therapy has been the norm for advanced disease.
    • The majority of patients with metastatic prostate cancer respond to hormone therapy, with a median response duration between 14 and 30 months.
    • Patients with biochemically (according to level of prostate-specific antigen [PSA]) recurrent disease have a more prolonged response duration (>60 months).
    • Despite a primary response rate of 80% to 90% with hormonal ablation, almost all patients advance to a state of androgen independence as manifested by PSA progression, and/or clinical tumor progression at local and/or distant sites. Once a patient becomes hormone-refractory, the median survival is 1-2 years.
  • Intermittent androgen suppression, known colloquially as "hormone holiday" was first introduced clinically in the 1980s to help minimize the side effects of anti-androgen therapy, which include vasomotor side effects, increased weight and cholesterol levels, reduced sexual function, increased fracture risk, lower energy levels, and reduced hemoglobin levels.
  • Because of the long natural history of prostate cancer and the widespread use of long-term hormone therapy, attention to the side effects of therapy and quality of life become paramount. Strategies to improve the duration of hormone response and minimize the side effects of ADT have been pursued.
  • In men with PSA recurrence after radical radiotherapy, intermittent androgen suppression has been suggested by phase II trials to improve quality of life (QoL) but effects on survival were unknown. In this study, intermittent androgen suppression (IAS) for PSA recurrence after radical radiotherapy (RT) was compared to continuous androgen deprivation (CAD) in an open label, non-inferiority-based, randomized trial.


  • This was an international study involving Canadian, US, and European participation with subjects enrolled from 1999-2005.
  • The definition of non-inferiority was defined as a less than 8% difference in 7-year survival. The study was designed such that 800 events (deaths) were needed for 80% power. Pre-planned interim analysis was done when 400 events were recorded.
  • Eligible patients were at least 1 year status post radical or salvage RT, rising PSA > 3.0 ng/ml and PSA greater than post-RT nadir, had no evidence of distant metastases at the time of enrollment, testosterone > 15 ng/mL, and ECOG PS of 0-1. They were allowed to have had up to 1 year of prior neo/adjuvant androgen deprivation therapy (ADT) for localized prostate cancer.
  • The randomization to IAS vs CAD was 1:1. Intermittent androgen suppression (IAS) was a cyclic therapy consisting of active treatment periods followed by observation periods. During these observation or off-treatment intervals (OTIs), patients were closely monitored for disease control; evidence of progression triggered the initiation of the next active treatment cycle.
  • Stratification factors were time since RT (1-3 vs. >3 years), initial PSA (<15 vs. >15), and prior radical prostatectomy or ADT. IAS was delivered for 8 months in each cycle, and included use of a leuteinizing-hormone-releasing hormone (LHRH) analog along with one month of an oral anti-androgen. Restart of IAS was indicated when PSA reached >10 ng/ml during an observation period.
  • The primary endpoint was OS. Secondary endpoints were QoL, time to hormone refractory state, cholesterol/HDL/LDL, length of non-treatment periods, testosterone and potency recovery. The independent Data and Safety Monitoring Committee recommended halting the trial after a planned interim analysis demonstrated that a pre-specified stopping boundary for non-inferiority was crossed.


  • Of 1386 patients, 690 patients were randomized to IAS ; 696 to CAD. The median age was 74 years. 11% had prior prostatectomy. The arms were well-balanced as to the numbers found to ineligible, those who were lost to follow-up, and number of deaths. The median follow up was 6.9 years.
  • Intermittent androgen suppression patients completed a median of two 8 month cycles. (range 1-9 cycles) and had reduced hot flashes but no differences in other AEs, including myocardial events or osteoporotic fractures. For patients completing more than one cycle, the time off between cycles decreased with time. For example, on average, there was a 20 month interval in between cycle 1 and cycle 2, a 13 month interval between cycle 2 and 3, and a 10 month interval between cycle 3 and 4.
  • The use of an LHRH analog was cut by two-thirds in the IAS arm, with the median duration of LHRH use of 14.3 months vs. 43.9 months in the continuous androgen deprivation arm.
  • 35% of IAS cases had full testosterone recovery.
  • Cross sectional QOL analysis shows a range of benefits for IAS at varying times. Using a 10 point change from baseline score as clinically meaningful, IAS patients had better QoL in physical function (p < 0.01), fatigue (p < 0.01), urinary problems (p = 0.01), hot flashes (p < 0.01), desire for sexual activity (p < 0.01) and erectile function (p < 0.01).
  • Over 800 patients were still alive at the close of the study. 524 patients died (268 IAS vs. 256 CAD). Of patients for whom the cause of death was known, 37% had died of prostate cancer, 22% had another primary tumor, and 32% died from other causes.
  • Median OS was 8.8 vs. 9.1 years on IAS and CAD arms respectively (HR 1.02, 95%CI = 0.86-1.21; p for non-inferiority [HR IAS vs. CAD ? 1.25] = 0.009).
  • The IAS arm had more disease-related (122 vs. 97) and fewer unrelated (134 vs. 146) deaths. Time to hormone refractory state was statistically significantly improved on the IAS arm (HR 0.80, 95CI 0.67-0.98, p=0.024).

Author's Conclusions

  • IAS is non-inferior to CAD with respect to OS. IAS improved the time to castration resistance (although there is an inherent bias in favor of the IAS arm for this endpoint).
  • There were 9% more prostate cancer deaths in the IAS arm and 8% more other deaths on the CAD arm. Thus, the increase in prostate cancer deaths in the IAS group was balanced by a relative decrease in deaths from other causes.
  • The authors conclude that IAS should therefore be the standard of care for patients with biochemical failure after primary or salvage radiotherapy.

Clinical Implications

  • Some people anticipated that good randomized clinical trials would show a clear survival benefit for intermittent ADT compared to continuous ADT. Though this was not demonstrated in this non-inferiority study, IAS compared to CAD may still be a useful therapy in terms of side effects, quality of life, and cost.
  • One of the other major findings of the study is that drug use is substantially reduced using the intermittent approach. This could result in considerable cost savings.
  • It is also debatable as to whether an 8% overall survival difference between the two arms should constitute non-inferiority; some may argue that this number is too high. For patients with PSA recurrence, the timing of ADT is still a matter of question.
    • Both immediate and deferred hormonal therapy are considered acceptable standards of care.
    • Some have argued that one way to delay castration-resistance is to delay castration. It is unclear if IAS improves longevity in asymptomatic patients. The optimal timing of hormonal therapy in this patient population is the subject of a trial through the Ontario Cooperative Oncology Group.


Malone S, Perry G, Eapen L, et al. Mature Results of the Ottawa Phase II Study of Intermittent Androgen-Suppression Therapy in Prostate Cancer: Clinical Predictors of Outcome, Int J Rad Oncol Biol Phys. 68(3):699-706; 2007.