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

3904 - Conditional survival probability in patients with resected oesophageal adenocarcinoma receiving neoadjuvant chemotherapy


08 Oct 2016


Poster Display


Donna Graham


Annals of Oncology (2016) 27 (6): 207-242. 10.1093/annonc/mdw371


D.M. Graham1, F.J. Bannon2, F. Noble3, R. O'Neill4, J.K. Blayney5, T.J. Underwood3, R.D. Kennedy1, R.C. Fitzgerald6, R.C. Turkington1

Author affiliations

  • 1 Centre For Cancer Research And Cell Biology, Queen's University Belfast, BT7 1NN - Belfast/GB
  • 2 Centre For Public Health, Queen's University Belfast, Belfast/GB
  • 3 Department Of Surgery, University Hospital Southampton NHS Foundation Trust, Southampton/GB
  • 4 Clinical Surgery, Edinburgh Cancer Research UK-University of Edinburgh, Edinburgh/GB
  • 5 Department Of Bioinformatics, Centre For Cancer Research And Cell Biology, Queen's University Belfast, BT7 1NN - Belfast/GB
  • 6 Hutchison/mrc Cancer Unit, University of Cambridge, Cambridge/GB


Abstract 3904


Traditional recurrence and survival figures are based upon factors determined at baseline and become less relevant for patients over time. Conditional survival (CS) estimates future prognosis based upon survival to a specific time point since treatment. We analysed CS and conditional recurrence-free survival (CRFS) data for patients in the United Kingdom undergoing surgery and neoadjuvant chemotherapy for gastro-oesophageal junction (GOJ) or oesophageal adenocarcinoma (OAC).


378 patients with GOJ/OAC treated with neoadjuvant chemotherapy and surgical resection from 2003-2012 were identified. Clinicopathological and survival data was collected as part of the Oesophageal Cancer Clinical and Molecular Stratification (OCCAMS) consortium. A multivariable parametric survival model was used to analyse factors associated with overall survival and recurrence from time of surgery.


The cohort includes 305 males (80.7%) with median age 65 (range 28-83) years. 5-year RFS conditional on recurrence-free years to date increased over time (see table). For those with stage T3/4, moderately-poorly differentiated tumours with lymphovascular invasion, 5-year disease-specific survival (DSS) improved from 7.7% for those with node-positive, R1 disease to 45.4% conditional on 3 years post-treatment survival; compared with 55.6% actuarial 5-year DSS increasing to 86.7% conditional on 3 years post-treatment survival for patients with N0 R0 disease. Age, sex, year of surgery, and Siewert classification had no association with recurrence or mortality rate.

Recurrence-free survival (RFS) by prognostic factors

Poor Prognostic factor 5-year RFS (%) 5-year RFS conditional on 1 year (%) 5-year RFS conditional on 2 years (%) 5-year RFS conditional on 3 years (%)
T-stage: 3 or 4 Nodes positive Margins involved
1 1 1 7.2 13.2 28.7 49.5
1 1 0 35.4 45.1 61.2 75.8
1 0 1 25.6 35.1 52.4 69.5
1 0 0 58.5 66.2 77.6 86.7
0 1 1 20.0 29.1 46.7 65.0
0 1 0 53.1 61.5 74.1 84.4
0 0 1 43.5 52.7 67.4 80.0
0 0 0 72.0 77.7 85.6 91.6


CRFS and CS provide a more dynamic estimation of future recurrence risk and survival among patients who have accrued survival time, especially in patients with high-risk features, including positive resection margins. Margin and node positivity govern early relapse events but as the time from surgery increases these factors become less relevant.

Clinical trial identification


Legal entity responsible for the study

Queen's University Belfast


Queen's University Belfast


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