Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster display session

5489 - Effect of sequence on outcome of prostate cancer patients: retrospective study of a French cohort


10 Sep 2017


Poster display session


Prostate Cancer


Reza Elaidi


Annals of Oncology (2017) 28 (suppl_5): v269-v294. 10.1093/annonc/mdx370


R. Elaidi1, Y. Belhadj1, S. Oudard2

Author affiliations

  • 1 Medical Oncology - Hôpital Européen Georges Pompidou, ARTIC, 75015 - Paris/FR
  • 2 Immunothérapie Et Traitement Antiangiogénique En Pathologie cancérologique, Hopital European George Pompidou, 75015 - Paris/FR


Abstract 5489


several drugs are approved in prostate cancer (PC), both in localized and metastatic setting. Challenge of daily practice is the sequencing of available agents for optimal disease management. Trying to extract actionable information from the overall history of disease for each patient remains a difficult task but could provide new insights for better sequencing. This retrospective analysis aimed to follow-up patients included in the Rising-PSA phase 3 clinical trial (R-PSA-CP-03) until death or last contact.


we retrospectively analyzed therapies received by pts included in R-PSA at the HEGP hospital (Paris, France). Drugs were coded in 8 categories: LH: LHRH modulators, AA: anti-androgens, AA2: new generation AA, DC: docetaxel, CZ: cabazitaxel, EX: blinded experimental drugs, P: therapeutic pause, PCB: placebo(experimental). Sequence rank, therapy duration and their interaction was estimated using both a conditional repeated events model (CREM) and a multi-state model (MSM) based on Markov process stratified on disease setting. Covariables included in the models were: age and Gleason score at inclusion time.


152 pts included between 01/2003 and 09/2007 were followed > 10years. Metastatic progression: 70(46%). Death: 31(20%). Median age(y): 64(51-80)), Gleason ≥8: 47(31%). Median (range) number of sequences received: M0=8(1-15) & M1=6(1-19) including pauses. Number of times each therapy was used whatever the sequence (%M0/%M1): LH(48/10), AA(6/6), AA2(0/22), DC(10/10), CZ (0/10), EX(1/6), PCB(0/2), P(35/34). Upon CREM, the overall model fitted perfectly well the time on therapies and their sequence (robust estimation: p 


to our knowledge, this is the first attempt to model the entire course of PC taking into account both therapies and sequence. Given the complexity of our model, these results should be validated with further studies and methods.

Clinical trial identification


Legal entity responsible for the study





All authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.