150PD - Discordant RAS status between primary and metastatic colorectal cancer and predicted pattern of metastases

Date 20 December 2015
Event ESMO Asia 2015 Congress
Session Gastrointestinal tumours
Topics Colon and Rectal Cancer
Translational Research
Basic Principles in the Management and Treatment (of cancer)
Presenter KIN SANG Lau
Citation Annals of Oncology (2015) 26 (suppl_9): 42-70. 10.1093/annonc/mdv523
Authors K.S. Lau1, K.O. Lam2, T.S. Choy3, W.H. Shek4, L.P. Chung4, T.W. Leung1
  • 1Department Of Clinical Oncology, Queen Mary Hospital, 00852 - Hong Kong/HK
  • 2Department Of Clinical Oncology, The University of Hong Kong, 00852 - Hong Kong/HK
  • 3Department Of Clinical Oncology, QUEEN MARY HOSPITAL, 00852 - Hong Kong/HK
  • 4Department Of Pathology, Queen Mary Hospital, 00852 - Hong Kong/HK



KRAS and NRAS are predictive biomarkers for efficacy of anti-EGFR therapy. Previous studies showed high KRAS concordance between primary tumors and paired metastases while data on NRAS were lacking. This study aims to examine the concordance of KRAS/ NRAS mutation between paired primary and metastatic colorectal tumors. The correlation between pattern of metastases and RAS status was also examined.


A total of 230 patients diagnosed with metastatic colorectal cancer in August 2013-December 2014 were included. Inclusion criteria were metastatic colorectal cancer new cases in the above period, both primary and metastatic tumor available for test (biopsy or resection), quality of tumor DNA suitable for RAS test. Microdissections, PCR and Sanger sequencing were performed in codon 12/13/61 to determine KRAS/NRAS mutations. Concordance and Cohen Kappa were calculated using SPSS v20.


Sixty-seven patients were studied, 67 primary and 82 paired metastatic tumors were examined, 14 patients had more than 1 metastatic tumor examined. The mean age was 62.7 years old (35-88), 44 were males (65.7%), 23 were females (34.3%). The frequency of KRAS, NRAS mutations, wildtype status in primary tumors were 38.8% (26), 4.5% (3) and 56.7% (38) respectively, while the frequency in metastatic tumors were 41.8% (28), 4.5% (3), 53.7% (36) respectively. 8 discordant RAS cases were found. A concordance of 88.1% (59/67) (kappa = 0.759) of RAS mutation was observed between primary and metastatic tumors while the concordance among metastatic sites in same patient was 100%. Among discordant cases, 3 had wildtype metastases, KRAS mutated primary; 5 had KRAS mutated metastases, wildtype primary. No NRAS mutation were discordant. All of 14 liver metastases-only group were concordant. Six of 9 lung metastases-only group were concordant, 39 of 44 multiple-metastases group were concordant between primary and metastatic RAS mutation. The RAS concordance was statistically different among the 3 groups (p= 0.05). Lung metastases-only was associated with a lower RAS concordance (p= 0.033).


The concordance of RAS mutation between primary and paired metastatic tumors was 88.1%. Lung metastases-only was associated with lower RAS concordance.

Clinical trial identification


All authors have declared no conflicts of interest.