687P - Clinical utility of circulating tumor DNA (ctDNA) in resectable pancreatic ductal adenocarcinoma (PDAC)

Date 08 October 2016
Event ESMO 2016 Congress
Session Poster Display
Topics Pancreatic Cancer
Presenter Hui-li Wong
Citation Annals of Oncology (2016) 27 (6): 207-242. 10.1093/annonc/mdw371
Authors H. Wong1, K. Bushell2, J. Karasinska3, S. Arthur2, P. Pararajalingam2, R. Morin2, D.F. Schaeffer4, D.J. Renouf5
  • 1Division Of Medical Oncology, BC Cancer Agency, V5Z 4E6 - Vancouver/CA
  • 2Department Of Molecular Biology And Biochemistry, Simon Fraser University, Vancouver/CA
  • 3Pancreas Centre Bc, Vancouver, BC/CA
  • 4Division Of Anatomical Pathology, Vancouver General Hospital, Vancouver/CA
  • 5Division Of Medical Oncology, BC Cancer Agency, Vancouver/CA



There is considerable interest in investigating ctDNA as a non-invasive biomarker, with potential utility in screening, detecting minimal residual disease after curative resection and monitoring treatment response or resistance. Here we perform sequential ctDNA quantification in patients (pts) with resectable PDAC using a novel and highly sensitive multiplex technology to explore the clinical utility of ctDNA as a diagnostic and prognostic biomarker.


Banked plasma and tumor samples from 32 pts with resected PDAC were retrieved. Plasma samples were collected 0-28 days before, and 28-70 days after surgery. DNA was extracted using standard protocols and analyzed using the OnTarget system, which enriches for DNA molecules containing hot spot mutations prior to sequencing. A 96-plex panel that includes the most prevalent mutations in KRAS, PIK3CA and TP53 was used.


29 pts (91%) had at least 1 mutation detected by OnTarget in the tumor sample, most frequently in KRAS codon 12 (n = 26). 25 of 29 pts with a panel-detected tumor mutation had a pre-operative blood sample available, where a concordant mutation was detected in 8 pts (sensitivity 32%). There was no correlation between pre-operative ctDNA and tumor pathologic features. At median follow-up of 16.5 months, there was no difference in recurrence-free survival (RFS) between pts with and without detectable pre-operative ctDNA (p = 0.595). Of the 22 available post-operative blood samples, a concordant mutation was detected for 5 pts. RFS was significantly shorter in pts with concordant ctDNA detected after surgery (median 2.1 vs 11.6 months, p 


Pre-operative ctDNA has low sensitivity, suggesting limited utility in PDAC screening, and does not appear to be prognostic. RFS was significantly shorter in pts with detectable tumor-specific ctDNA post-operatively; this analysis is limited by small numbers and short follow-up. The significance of discordant plasma and tumor mutations is unknown.

Clinical trial identification

Legal entity responsible for the study

Pancreas Centre BC


Pancreas Centre BC


D.J. Renouf: Honorarium from Celgene Canada. All other authors have declared no conflicts of interest.