Adding Celecoxib With or Without Zoledronic Acid for Hormone-Naïve Prostate Cancer: Long-Term Survival Results From an Adaptive, Multiarm, Multistage, Platform, Randomized Controlled Trial

Abstract

Purpose Systemic Therapy for Advanced or Metastatic Prostate Cancer: Evaluation of Drug Efficacy is a randomized controlled trial using a multiarm, multistage, platform design. It recruits men with high-risk, locally advanced or metastatic prostate cancer who were initiating long-term hormone therapy. We report survival data for two celecoxib (Cel)-containing comparisons, which stopped accrual early at interim analysis on the basis of failure-free survival. Patients and Methods Standard of care (SOC) was hormone therapy continuously (metastatic) or for ≥ 2 years (nonmetastatic); prostate (± pelvic node) radiotherapy was encouraged for men without metastases. Cel 400 mg was administered twice a day for 1 year. Zoledronic acid (ZA) 4 mg was administered for six 3-weekly cycles, then 4-weekly for 2 years. Stratified random assignment allocated patients 2:1:1 to SOC (control), SOC + Cel, or SOC + ZA + Cel. The primary outcome measure was all-cause mortality. Results were analyzed with Cox proportional hazards and flexible parametric models adjusted for stratification factors. Results A total of 1,245 men were randomly assigned (Oct 2005 to April 2011). Groups were balanced: median age, 65 years; 61% metastatic, 14% N+/X M0, 25% N0M0; 94% newly diagnosed; median prostate-specific antigen, 66 ng/mL. Median follow-up was 69 months. Grade 3 to 5 adverse events were seen in 36% SOC-only, 33% SOC + Cel, and 32% SOC + ZA + Cel patients. There were 303 control arm deaths (83% prostate cancer), and median survival was 66 months. Compared with SOC, the adjusted hazard ratio was 0.98 (95% CI, 0.80 to 1.20; P = .847; median survival, 70 months) for SOC + Cel and 0.86 (95% CI, 0.70 to 1.05; P =.130; median survival, 76 months) for SOC + ZA + Cel. Preplanned subgroup analyses in men with metastatic disease showed a hazard ratio of 0.78 (95% CI, 0.62 to 0.98; P = .033) for SOC + ZA + Cel. Conclusion These data show no overall evidence of improved survival with Cel. Preplanned subgroup analyses provide hypotheses for future studies.

Figures

Fig 1.

CONSORT flow diagram depicting the flow of patients who joined the STAMPEDE trial while these specific comparisons were open to recruitment. Further context is given in the Data Supplement. A, standard of care (SOC); AE, adverse event; Cel, celecoxib; D, SOC + Cel; F, ZA + SOC + Cel; ZA, zoledronic acid.

Fig 2.

Failure-free and overall survival, by research comparison. Kaplan-Meier plots showing time to event for the definitive primary outcome measure (overall survival) and intermediate primary outcome measure (failure-free survival). (A) Failure-free survival in the celecoxib comparison. (B) Overall survival in the celecoxib comparison. (C) Failure-free survival in the ZA + celecoxib comparison. (D) Overall survival in the ZA + Cel comparison. Cel, celecoxib; SOC, standard of care; ZA, zoledronic acid.

Fig 3.

Overall survival for SOC + Cel + ZA versus SOC in patients with metastatic disease. Kaplan-Meier plot showing overall survival for the ZA + Cel comparison in patients who presented with metastatic disease at random assignment. Cel, celecoxib; SOC, standard of care; ZA, zoledronic acid.

Fig 4.

Forest plots of treatment effect on survival within subgroups, by research comparison, showing assessment of consistency of the treatment effect on overall survival in preplanned subgroups for (A) the Cel comparison and (B) the ZA + Cel comparison. The number of deaths and the number of patients are shown by arm for each treatment level, together with an adjusted hazard ratio and a test for heterogeneity of the treatment effect. Cel, celecoxib; Mets, metastases; NSAID, nonsteroidal anti-inflammatory drug; PS, performance status; RT, radiotherapy; SOC, standard of care; ZA, zoledronic acid.

Fig 4.

Forest plots of treatment effect on survival within subgroups, by research comparison, showing assessment of consistency of the treatment effect on overall survival in preplanned subgroups for (A) the Cel comparison and (B) the ZA + Cel comparison. The number of deaths and the number of patients are shown by arm for each treatment level, together with an adjusted hazard ratio and a test for heterogeneity of the treatment effect. Cel, celecoxib; Mets, metastases; NSAID, nonsteroidal anti-inflammatory drug; PS, performance status; RT, radiotherapy; SOC, standard of care; ZA, zoledronic acid.