ECC 2015. News in GI Cancer: The ICECREAM Study

European Cancer Congress (ECC) 2015
Presented by the European Society for Medical Oncology

The AGITG ICECREAM Study: The Irinotecan Cetuximab Evaluation and Cetuximab Response Evaluation Amongst Patients with a G13D Mutation - analysis of outcomes in patients with refractory metastatic colorectal cancer harbouring the KRAS G13D mutation

E. Segelov, S. Thavaneswaran, P. Waring, J. Desai, K. Mann, E. Elez, L. Chantrill, N. Pavlakis, L. Nott, C. Underhill, M. Khasraw, H. Wasan, F. Ciardiello, M. Jefford, W. Joubert, A. Haydon, C. Karapetis, T. Price, K. Wilson, J. Shapiro

Background: Patients (pts) with metastatic colorectal cancer (mCRC) harbouring KRAS and now NRAS mutations, are precluded from treatment with EGFR-inhibitors (EGFR-I) due to lack of response. Mutations in KRAS are found in 40% of CRCs with the KRAS exon 2 mutation c.38G>A: pGly13Asp (G13D) accounting for ∼19%, with an absolute incidence in mCRC of 8%. The G13D mutation conveys sensitivity to EGFR-I in pre-clinical models and retrospective clinical reports have suggested treatment benefit with cetuximab (cet) similar to KRAS wild type (WT) in both refractory and earlier settings. However, numbers were small and the impact of co-administered chemotherapy difficult to isolate. The addition of irinotecan (iri) to EGFR-I increased response rate and delayed progression in KRAS unselected mCRC and may also potentiate response in pts with G13D mutated tumours.

Methods: ICECREAM is a randomised phase II trial assessing efficacy of cet v cet + iri (cet-iri) in mCRC pts stratified for a G13D mutation or no mutations in KRAS, NRAS, BRAF and PI3KCA (cohort still recruiting). Pts with ECOG PS 0–2 who were refractory to iri (progression within 6 months (m)) and intolerant or refractory to fluoropyrimidine and oxaliplatin were randomised 1:1 to cet 400mg/m2 IV loading then 250mg/m2 weekly +/− iri 180mg/m2 q2 weeks. The primary endpoint was progression free survival at 6 m (6m PFS); secondary endpoints were response rate, overall survival, toxicity and quality of life. Results for the G13D cohort are reported.

Results: 53 pts were recruited to the G13D arm between Nov 2012 and Dec 2014. Two pts were ineligible (not iri-refractory). Most pt characteristics were balanced between cet v cet-iri except for age (61 v 66 years); time since first metastatic diagnosis (19.1 v 28.1m) and time since last iri (2.8 v 4.8m). 6 m PFS was 10% (95% CI: 2–26%) for cet v 23% (95% CI: 9–40%) for cet-iri; HR 0.75 (0.42–1.33). Median time to progression was 2.5 v 2.6 m respectively. No responses were seen with cet, with stable disease (SD) in 58% pts. For cet-iri, 9% of pts achieved PR and 70% SD. Toxicities were higher with cet-iri; 11(44%) v 16(64%) of pts experiencing ≥one Grade 3/4 event.

Conclusion: This is the first prospective, molecularly defined, randomised trial of standard therapies in mCRC and exemplifies international collaboration to recruit rare molecular subtypes to clinical trials. The AGITG ICECREAM trial shows lack of activity of cet monotherapy in pts with G13D mutated mCRC, consistent with smaller series. However, the combination of cet-iri achieved objective responses and delayed progression of disease. The benefit of combination therapy may warrant further evaluation to ascertain if there is a true synergistic effect of iri with EGFR-I in pts with G13D mutated mCRC.


CARE Faculty Perspective:

Recent evidence from De Rook et al.(1) and Tejpar et al.(2) has suggested that the population of patients harbouring a KRAS G13D mutation may be sensitive to EGFR inhibition similar to KRAS wild type patients even in irinotecan refractory patients. However, this is a small proportion of patients (~5%) and it was felt that it would not be feasible to perform an RCT and so it was explored in an international open label phase II study looking at PFS in patients with G13D mutations to those who are wild type for KRAS, NRAS, BRAF and PI3KCA (quad WT) and who have either progressed or are intolerant to 5FU and oxaliplatin and who were within 6 months of progressing on irinotecan.

At this time, the latter group is still recruiting and so the trial reported on the data of the G13D cohort only which had randomized a total of 51 patients in a 1:1 fashion between single agent cetuximab and a combination of cetuximab + irinotecan. There was no significant difference between the groups for PFS at 6 months (23% vs. 10%, HR 0.74; p=NS) although there was a greater RR in the combination arm (9% vs. 0%) albeit at the expense of greater toxicity. From this, it was concluded that patients with G13D mutated mCRC do not benefit from cetuximab monotherapy. This is a very interesting trial and despite the negative results, the authors should be congratulated on their efforts and we will await the final analysis of the G13D mutated group to the quad WT group with interest.

- Dr. Robert El-Maraghi, on behalf of the CARE Oncology Faculty


1. De Rook, W et al. JAMA. 2010 Oct 27;304(16):1812-20.

2. Tejpar S et al. J Clin Oncol. 2012 Oct 10;30(29):3570-7


Related Abstracts of Interest