941P - Impact of salvage therapy with APC8015F on the overall survival (OS) benefit achieved with sipuleucel-T in three phase 3 studies of metastatic castr...

Date 29 September 2012
Event ESMO Congress 2012
Session Poster presentation I
Topics Drug Development
Prostate Cancer
Presenter Daniel George
Authors D.J. George1, L.G. Gomella2, T. Devries3, J.B. Whitmore3, M.W. Frohlich4, C. Nabhan5
  • 1Medicine, Duke University, 27710 - Durham/US
  • 2Urology, Thomas Jefferson University, Philadelphia/US
  • 3Biometrics, Dendreon, 92101 - Seattle/US
  • 4Dendreon Corporation, US-98121 - Seattle/US
  • 5Hematology And Medical Oncolgoy, Advocate Lutheran General Hospital, Park Ridge/US



Sipuleucel-T is an autologous cellular immunotherapy approved for asymptomatic or minimally symptomatic mCRPC. Three Phase 3 sipuleucel-T trials (D9901, D9902A, IMPACT) allowed control patients (pts) to receive salvage treatment with an autologous product derived from previously frozen cells (APC8015F). We previously reported that APC8015F demonstrated no deleterious effect. Control pts who received APC8015F had more favorable baseline prognostic features; however, after adjusting for these imbalances, APC8015F appears to have prolonged OS in control pts, potentially reducing the observed sipuleucel-T treatment effect. This exploratory integrated analysis estimates the OS benefit conferred by sipuleucel-T, adjusting for APC8015F in control pts.


A rank-preserving structural failure time (RPSFT) model was used to estimate control arm OS if treatment with APC8015F had not occurred. This allows estimation of the treatment effect of sipuleucel-T, adjusting for salvage effect.


Median OS from randomization in the three pooled trials was 25.4 months with sipuleucel-T (n = 488) and 21.5 months with control (n = 249). Of the control arm pts, 165 (66.3%) subsequently received APC8015F. Median OS from randomization in the control population was 23.6 months for pts receiving APC8015F and 12.7 months for those who did not. Using the RPSFT model, and assuming APC8015F was as effective as sipuleucel-T, the estimate of median OS for control pts was 17.3 months, representing an 8.1 month median increase in OS with sipuleucel-T. Results from extensions of the RPSFT model, where APC8015F is assumed to have less treatment effect than sipuleucel-T, will be presented.


These analyses estimate a median OS benefit for sipuleucel-T between 3.9 and 8.1 months, assuming that APC8015F had either no efficacy or comparable efficacy to sipuleucel-T, respectively. These results suggest a possible greater treatment effect of sipuleucel-T than was reported in the three Phase 3 studies. Future studies should account for potential crossover treatment bias as this may diminish estimates of OS benefit.


D.J. George: Consultant, Investigator, or Lecturer for the following companies: Astellas, Bayer, BMS, Dendreon, Exelixis, Genentech/Roche, GSK, Ipsen, Jansen, Medivation, Novartis, Pfizer, Sanofi, Viamet,

L.G. Gomella: Consultant/Advisor and Clinical Trials for Dendreon,

T. Devries: Employee and Stockholder of Dendreon.

J.B. Whitmore: Employee and Stockholder of Dendreon.

M.W. Frohlich: Employee and Stockholder of Dendreon.

C. Nabhan: I am on the speakers bureau of Dendreon and have received honoraria for speaking engagements.