LBA-03 - Neoadjuvant chemotherapy for resectable oesophageal and junctional adenocarcinoma: results from the UK Medical Research Council randomised OEO5 tria...

Date 04 July 2015
Event WorldGI 2015
Session Oral and LBA abstracts
Topics Anticancer Agents
Oesophageal Cancer
Biological Therapy
Presenter David Cunningham
Citation Annals of Oncology (2015) 26 (suppl_4): 117-122. 10.1093/annonc/mdv262
Authors D. Cunningham1, J. Thompson2, R. Langley3, M. Nankivell1, J. Blazeby4, M. Griffin5, A. Crellin6, H. Grabsch7, A. Okines1, C. Goldstein1, S. Falk4, R. Krysztopik8, F. Coxon9, S. Pritchard10, R. Langer11, S. Stenning1, D. Alderson12
  • 1Royal Marsden, London/UK
  • 2Birmingham Heartlands Hospital, Birmingham/UK
  • 3MRC Clinical Trials Unit at UCL, London/UK
  • 4University of Bristol, Bristol/UK
  • 5Royal Victoria Infirmary, Newcastle/UK
  • 6St James' Institute of Oncology, Leeds/UK
  • 7Leeds Institute of Cancer Studies and Pathology, Leeds/UK
  • 8Royal United Hospital, Bath/UK
  • 9Northern Centre for Cancer Care, Newcastle/UK
  • 10Wythenshawe Hospital, Manchester/UK
  • 11University of Bern, Bern/CH
  • 12Queen Elizabeth Hospital, Birmingham/UK



Chemotherapy in addition to surgery improves outcomes in gastro-oesophageal cancer. Both the OEO2 (neoadjuvant) and MAGIC (peri-operative) trials showed statistically significant improvements in overall survival though only 55% of patients in the MAGIC trial received post-operative treatment. We investigated whether more neoadjuvant chemotherapy (4 cycles epirubicin/cisplatin /capecitabine (ECX)) compared to a standard approach (2 cycles of cisplatin/5-fluorouracil) would improve outcomes.


A multi-centre, randomised, phase III trial comparing 2 cycles of CF with 4 cycles of ECX followed by oesophagectomy with 2-field lymphadenectomy for lower oesophageal and junctional (Types I and II) adenocarcinoma. Primary outcome was overall survival (OS); 842 patients (677 deaths) would detect an increase in 3-year survival from 30% to 38% (or 37%) with 82% (or 70%) power with 2α = 5%. Deaths accrued more slowly than anticipated but the Independent Data Monitoring Committee considered the data sufficiently robust for release. Secondary outcomes include disease-free (DFS) and progression-free survival (PFS), pathological R0 resection rate, Mandard grade and quality of life (QoL).


From 2005-2011, 897 patients (451 CF, 446 ECX) from 72 UK centres were randomly allocated (1:1). Baseline characteristics were similar between the groups (overall, male 90%, median age 62 (IQR 56-67), staging included PET 60%, T3 N0 22%, T3 N1 65%). 96% CF received 2 cycles, 89% ECX > 3 cycles. Grade 3/4 toxicity was lower with CF (30% v 47% p < 0.001). Of those patients having a resection R0 rates were CF 60%, ECX 66% with a Mandard grade ≤3 achieved in CF 15% v ECX 32% with 3% and 11% achieving complete response. Post-operative complications were similar (CF 57%, ECX 62%) as were deaths at 30 (CF 2%, ECX 2%) and 90 days post-surgery (CF 4%, ECX 5%). PFS and DFS favoured ECX, hazard ratio (HR, 95% CI) PFS 0.86 (0.74-1.01), DFS 0.88 (0.75-1.03). HR for OS was 0.92 (0.79-1.08, p = 0.3017) based on 315 CF and 298 ECX deaths, with similar 3 year survival rates CF 39% (35-44%) vs ECX 42% (37-46%). Exploratory subgroup analyses suggested that N0 patients may benefit from ECX, HR for OS was 0.68 (0.47-0.97). There were no clinically important differences in QoL (global QoL and oesophageal cancer specific domains from the EORTC QLQ-C30 and QLQ-OES18 questionnaires), either pre-operatively or 3-months post-operatively.


There is some evidence of a benefit from the prolonged ECX regimen, in terms of PFS, DFS and tumour regression at resection, but this does not translate into a survival benefit. Ongoing translational work is aimed at identifying subsets of patients that might benefit from the triplet anthracycline containing regimen.