Z Gastroenterol 2010; 48 - P178
DOI: 10.1055/s-0030-1263622

KRAS- and BRAF-Mutations and their clinical correlates in a prospective, multi-center, diagnostic clinical trial „molecular signatures in colorectal cancer“ (MSKK)

P Pechanska 1, R Mantke 2, R Pauli 2, K Ridwelski 3, K Hellwig 3, M Pross 4, C Radke 4, B Unger 5, H Lobeck 5, J Pertschy 6, H Kosmehl 7, T Manger 8, C Boltze 8, T Buthut 9, V Henn 9, B Hinzmann 1, I Klaman 1, HP Adams 1, D Nürnberg 9, A Rosenthal 1
  • 1Signature Diagnostics AG, Potsdam, Germany
  • 2Städt. Klinikum Brandenburg, Brandenburg an der Havel, Germany
  • 3Klinikum Magdeburg, Magdeburg, Germany
  • 4DRK Kliniken Berlin Köpenick, Berlin, Germany
  • 5Klinikum Ernst von Bergmann, Potsdam, Germany
  • 6Kath. Krankenhaus St. Johann Nepomuk, Erfurt, Germany
  • 7Helios Klinikum Erfurt, Erfurt, Germany
  • 8SRH Wald-Klinikum Gera, Gera, Germany
  • 9Ruppiner Kliniken, Neuruppin, Germany

Aims: KRAS mutation status is already used for tailoring therapy in colorectal cancer patients. Therefore, we investigated the relation of KRAS- and BRAF-status (mutated or wild type (WT)) with patient and tumor characteristics in the framework of a multi-center prospective study (MSKK) aiming at development of predictive and prognostic biomarkers.

Methods: We have analyzed KRAS- and BRAF-mutation status in macro-dissected paraffin-embedded and snap-frozen tumor samples from 581 patients (52% male; 48% female, age: 36yrs –93yrs (median 70yrs), UICC-stages: I 21.2%, II 32.9%, III 35.6%, IV 10.3%), tumor location: 75.1% colon and 24.9% rectum) recruited into the study in 16 centers between 2005 and 2008. Patients and samples represent a random sample of a population study excluding patients undergoing neo-adjuvant therapy in rectal cancer. KRAS mutation status was determined by the 8 most frequent substitutions, and BRAF mutation status was assessed by substitution (1799 T>A).

Results: KRAS-mutation status was assessable in 98%, BRAF-mutation status could be determined in 99% of samples. Mutations of KRAS (WT=63.7%) and BRAF (WT=90.0%) exclude each other (p=<0.0001). Results of multivariate logistic regression modeling indicate: KRAS-mutation is not related to gender, age, tumor location, UICC-Status, grading, or venous vessel invasion. However, there is a weak relation to lymphatic vessel invasion (odds ratio (OR): 1.657 (95% confidence interval (CI): 1.163–2.361)). In contrast, BRAF-mutations are linked to gender (female OR: 4.577 (2.429–8.6234)), age (>60yrs OR: 4.583 (1.359–15.454)), UICC stage (>=III OR: 2.065 (1.145–3.724)), and tumor location (colon OR: 19.894 (2.701–146.5161)).

Conclusion: In a population based study, we could show that there is almost no relation between KRAS-mutation status and patient or tumor characteristics, BRAF-mutation status is clearly linked to tumor and patient. Thus, KRAS mutation status may have independent prognostic impact; this does not hold for BRAF mutation status.