Preoperative chemoradiation (CRT) with concurrent cetuximab, irinotecan and capecitabine in MRI-defined locally advanced rectal cancer (LARC) within the phase II EXCITE trial: Relationship of epidermal growth factor receptor (EGFR) pathway mutations to histological response and survival.


Year:

Session type:

Simon Gollins1, Nick West2, Phil Quirke2, Arthur Sun Myint3, Mark Saunders4, Shabbir Susnerwala5, David Sebag-Montefiore2, Sharadah Essapen6, Les Samuel7, Bruce Sizer8, Sandy Beare9, Emma Lawrie9, Mark Jitlal9
1North Wales Cancer Treatment Centre, Denbighshire, UK, 2Leeds Teaching Hospitals NHS Trust, Leeds, UK, 3Clatterbridge Cancer Centre, Bebington, UK, 4Christie NHS Foundation Trust, Manchester, UK, 5Rosemere Cancer Centre, Preston, UK, 6St Lukes Cancer Centre, Guildford, UK, 7Aberdeen Royal Infirmary, Aberdeen, UK, 8Essex County Hospital, Colchester, UK, 9Cancer Research UK & UCL Cancer Trials Centre, London, UK

Background

A preoperative cetuximab-containing CRT regimen in LARC was examined, including the influence of EGFR pathway mutations.

Method

Patients with MRI-defined LARC threatening/involving potential circumferential resection margin received pelvic radiotherapy (RT) 45Gy in 25 daily fractions with concurrent oral capecitabine (650mg/m2 twice-daily 5 days/week) plus IV cetuximab (400mg/m2 one week prior then weekly at 250mg/m2 weeks 1-5) plus IV irinotecan (weekly at 60mg/m2 weeks 1-4). Surgical resection was stipulated 8 weeks after CRT. EGFR pathway mutations were not assessed pre-treatment. DNA pyrosequencing was later performed on pre-treatment biopsies for mutation status of KRAS codons 12/13/61/146, NRAS codons 12/13/61, PIK3CA codons 542/545/546/1047, and the BRAF V600E hotspot.

Results

Between April 2009-October 2011 82 patients were recruited and 80 commenced RT. 76 patients underwent surgery with pathological complete response in 14(18%) and near-complete (microfoci) in 6(8%). 4 patients did not undergo surgery because of clinical complete response meaning 24(30%) had an excellent clinical or pathological response (ECPR). With median 26 month follow-up there were 3 pelvic failures, 15 metastatic failures and 13 deaths. 36-month progression-free survival (PFS, 21 events) was 65% (95%CI:51-76%) and overall survival 77%(95%CI:63-86%). Amongst the 24 ECPR patients 1(4%) progressed and none died. Amongst the 56 non-ECPR patients 20(36%) had a PFS event and 13(23%) died. Mutation status was available in 78 patients with KRAS mutation (34), BRAF(3), NRAS(3), PIK3CA(10) and pathway (any mutation)(45). There was no association between KRAS status (p=0.43) or pathway status (p=0.72) and ECPR. There was no association between KRAS status (36-month Restricted Mean Survival Time, p=0.99) or pathway status (unadjusted hazard ratio=0.79(95%CI:0.31-1.98; p=0.61)) and PFS.

Conclusion

The 30% ECPR rate and associated marked survival improvement were similar to our previoustrial (RICE) using a similar regimen without cetuximab1. EGFR pathway mutations did not predict histological response and were not prognostic for survival although follow-up and patient numbers are limited.