59P - Genomic alterations in circulating tumor DNA (ctDNA) are associated with clinical outcomes in treatment-naive metastatic castration-resistant prost...

Date 10 October 2016
Event ESMO 2016 Congress
Session Poster display
Topics Prostate Cancer
Translational Research
Basic Principles in the Management and Treatment (of cancer)
Presenter Alexander Wyatt
Citation Annals of Oncology (2016) 27 (6): 15-42. 10.1093/annonc/mdw363
Authors A.W. Wyatt1, M. Annala2, K. Beja1, S. Parimi3, G. Vandekerkhove1, E. Warner1, M. Zulfiqar4, D. Finch5, C. Oja6, J. Vergidis7, M. Nykter2, M.E. Gleave1, K. Chi3
  • 1Vancouver Prostate Centre, University of British Columbia, V6H 3Z6 - Vancouver/CA
  • 2Institute Of Biosciences And Medical Technology, University of Tampere, Tampere/FI
  • 3Medical Oncology, British Columbia Cancer Agency, V5Z 4E6 - Vancouver/CA
  • 4Abbotsford Centre, British Columbia Cancer Agency, Abbotsford/CA
  • 5Centre For The Southern Interior, British Columbia Cancer Agency, Kelowna/CA
  • 6Fraser Valley Cancer Centre, British Columbia Cancer Agency, Surrey/CA
  • 7Vancouver Island Centre, British Columbia Cancer Agency, Victoria/CA



There are no established genomic biomarkers to predict response to the AR-targeted therapies enzalutamide and abiraterone, partly due to the impracticality of sampling tissue in a bone-predominant metastatic disease. Cell-free DNA (cfDNA) is a promising “liquid biopsy” approach to genomic characterization of mCRPC.


We performed deep targeted sequencing of 72 mCRPC-related genes in baseline cfDNA from 62 chemotherapy-naïve mCRPC patients enrolled in an ongoing randomized phase II trial of abiraterone vs enzalutamide (NCT02125357). Genomic alterations in cfDNA were examined for association with clinical variables including time on treatment.


Evidence for ctDNA was detected in 38 of 62 (61.3%) cfDNA samples at baseline with 27 samples harbouring ≥1 mutation(s) with an allele fraction above 1%, and 28 samples containing copy number alterations affecting at ≥2 genes. Patients with confirmed ctDNA displayed a trend towards higher cfDNA yield (p = 0.07), higher baseline PSA (p = 0.14), and shorter time on treatment (p = 0.12). AR amplification was found in 17 patients and associated with shorter time on treatment (median 200 vs 420 days, p = 3.5 x 10−4). AR mutations were found in 6 patients, exhibited mutual exclusivity with amplifications, and correlated with prior non-steroidal anti-androgen therapy. TP53 and BRCA2 inactivating alterations were detected in 17 and 9 patients respectively and were associated with shorter time on treatment (median 236 vs 420 days, p = 1.1 x 10−4; median 120 vs 342 days, p = 5.6 x 10−5, respectively). The associations for AR, TP53 and BRCA2 remained significant after adjusting for patient age, baseline PSA and ctDNA presence (p 


In this preliminary analysis, the majority of patients with treatment-naïve mCRPC had detectable ctDNA and the presence of certain genomic aberrations was associated with shorter duration of therapy with abiraterone or enzalutamide. CtDNA holds great potential as a minimally-invasive biomarker to identify chemotherapy-naïve mCRPC patients that have poor therapeutic outcomes.

Clinical trial identification


Legal entity responsible for the study

Dr. Alexander Wyatt and Dr. Kim Chi


Research grants from the Canadian Cancer Society and Prostate Cancer Canada; Academic support from the BC Cancer Foundation and the Vancouver Prostate Centre (UBC); Industry grants from Janssen and Astellas.


S. Parimi: Honoraria from Astellas for a journal club presentation (February 2016) and an advertisement board sponsored by Janssen (April 2016). D. Finch: Has received honoraria for participation on an ad boards for Janssen and Astellas. Was a PI on the enzalutamide Affirm clinical trial. J. Vergidis: Has received honoraria from Astellas and am a member of an advisory board also for Astellas. K. Chi: Has received honorarium and research funding from Janssen and Astellas. All other authors have declared no conflicts of interest.