KRAS mutational status is associated with clinical response in patients with metastatic colorectal cancer receiving panitumumab monotherapy

Abstract

B114

Background: Panitumumab is a fully human monoclonal antibody targeting the epidermal growth factor receptor (EGFr) and is indicated in patients (pts) with refractory metastatic colorectal cancer (mCRC) (Van Cutsem 2007). Since anti-EGFr therapy does not benefit all pts with mCRC, identifying biomarkers of responsiveness has become a forefront clinical effort as it would allow physicians to target therapy to those pts most likely to benefit. We evaluated the association of KRAS mutational status with clinical efficacy to panitumumab in pt samples from 4 large mCRC studies of safety and efficacy with panitumumab monotherapy.
Methods: Tumor sections from 66/709 pts from three phase 2 studies and one phase 3 extension study were consented for genetic sequencing, had response data, and were available for analysis of KRAS gene mutations. Genomic DNA was isolated from FFPE tumor sections (pretreatment). PCR was performed on KRAS (exons 2 and 3) to determine the prevalence of activating mutations. For exons 2 and 3, more than 30 independent colonies from each cloning procedure were sequenced and resolved on a Genetic Analyzer. Best objective response was assessed using RECIST (2 studies) or WHO criteria (2 studies) at prespecified timepoints; the three phase 2 studies were assessed by blinded central review and the extension study was assessed by local review.
Results: Of the 66 pts, 41 (62%) pts had wild-type (WT) KRAS and 25 (38%) pts harbored a KRAS mutation (codon 12 & 13). In the mutant KRAS group, 6 pts had SD (24%) and 19 pts had PD (76%) as their best OR; there were no responses. In the WT KRAS population, the PR rate was 12% (95% CI: 2 to 22), the SD rate was 54% (95% CI: 38 to 69), and the PD rate was 34% (95% CI: 20 to 49). The association between KRAS mutation status and lack of response to panitumumab was statistically significant (Fisher’s exact test, p = 0.003). From a Cox PH model, the hazard ratio for WT:mutant KRAS was 0.6 (95% CI: 0.34 to 0.95) for PFS and 0.5 (95% CI: 0.29 to 0.91) for OS.
Conclusion: Our findings suggest that KRAS mutational status may be a predictive marker of clinical benefit with panitumumab in mCRC pts. Pts with activating KRAS mutations in exon 2 may be less likely to respond to treatment with panitumumab monotherapy. These findings warrant further investigation in a controlled clinical trial to determine the predictive value of KRAS mutational status for response to panitumumab treatment.