P-237 - K-RAS exon 2 mutations in advanced colorectal cancer: are they really so bad prognostic indicators? A phase II mono-institutional retrospective study

Date 04 July 2015
Event WorldGI 2015
Session Posters
Topics Biomarkers
Colon Cancer
Rectal Cancer
Presenter V. Dadduzio
Citation Annals of Oncology (2015) 26 (suppl_4): 1-100. 10.1093/annonc/mdv233
Authors V. Dadduzio1, A. Cassano2, E. Cerchiaro1, S. Rossi1, M. Basso1, C. Barone1, M. Martini3, T. Cenci4, S. Capodimonti4, A. Strippoli1, M. Larocca1, P. Lombardi1, M. Bensi1
  • 1Catholic University of Sacred Heart - Rome, Roma/IT
  • 2Medical Oncology - Catholic University of Sacred Heart - Rome, Rome/IT
  • 3Università Cattolica del Sacro Cuore - Anatomia Patologica, Rome/IT
  • 4Catholic University of Sacred Heart - Rome, Rome/IT



K-Ras exon 2 codon 12 and 13 mutations are widely accepted negative predictive factors for anti-EGFR therapies in advanced colorectal cancer (CRC), while their role as prognostic factors is still discussed. No data are available concerning specific mutations and their prognostic role. Larger studies suggest a negative impact on cancer specific survival, although the whole CRC population (Stage I-IV) has been considered, not addressing the issue of the metastatic setting (mCRC). We conducted a mono-institutional retrospective study to investigate the real life impact of mCRC codon 12 and 13 mutations in terms of overall survival (OS).


All patients with diagnosis of mCRC treated at our institution between 2008 and 2014 carrying K-RAS exon 2 mutations were included in this study. Patients with Duke's stage I-III and patients with known K-RAS exon 3/4, B-RAF or N-RAS mutations were excluded. Primary endpoint was to determine any significant differences in OS between codon 12 and 13 mutations. Secondary endpoints included PFS, OS according to treatment received, OS in liver-limited disease (LLD) according to codon mutation and surgery of metastases.


198 out of 562 mCRC patients analysed for K-RAS exon 2 carried a mutation (35,2%), n.156 at codon 12 (27,7%), n.42 at codon 13 (7,4%). Homogeneous distribution of LLD at diagnosis was seen in both populations (32,7% and 35,7% respectively for codon 12 and 13). Allocation of systemic treatment was comparable: first line therapy included FP (5FU or capecitabine) in association to oxaliplatin in 57% and 54.8% of patients, or in association to Irinotecan in 36.5% and 38.1%, respectively for codon 12 and codon 13 mutated patients. Anti-VEGF agent bevacizumab was administered in 66% and 78.6% of codon 12 and 13 mutated, respectively. Median OS reached no statistically significant differences: 32.8 months (codon 12) and 33.9 months (codon 13). PFS was also comparable, reaching 10.8 months in both populations (no significant differences between first line Oxa vs Iri, with or without bevacizumab). Among LLD patients, a very high proportion underwent hepatic surgery with radical purpose (41/65 – 63.1%), thus probably affecting high values of survival in these patients: in fact, median OS has not been reached after a median follow-up of 24.8 months (estimate 62.7 months after 25/66 events). Notably, OS in non-LLD patients was quite high too, reaching 31.0 months, with 28% of patients (37/132) undergone to surgery of metastases, mainly pulmonary (median OS 44.8 vs 24.3 months in resected vs unresected patients, p < 0.001).


Our study highlighted no significant differences in terms of OS between K-RAS codon 12 and codon 13 mutated mCRC patients. Nevertheless, to our knowledge, this is the first report showing a very prolonged overall survival for K-RAS mutated patients, even when LLD patients are excluded. mOS of our patients favourably compares with mOS of pan-RAS wild type patients of latest randomized studies. Considering the high proportion of patients in which surgery was attempted, we argue that therapeutic aggressiveness (both medical and surgical) could counterbalance the lack of therapeutic tools such as anti-EGFR agents.