At present, prognostic and predictive values for KRAS mutation in lung cancer patients remain to be controversial. Biases in disease stage, treatment regimens and small-scale patient population led to inconsistent results in previous studies.Methods
We collected 91 consecutive lung cancer patients with KRAS mutation detected in Guangdong Lung Cancer Institute from January 2007 to May 2010. And each patient with mutant KRAS was paired randomly to a patient with KRAS/EGFR wild-type and an EGFR mutant patient with the identical TNM stage. All patients were followed up for treatment regimes, tumor response, progression-free survival (PFS) and overall survival (OS).Results
The patients with KRAS mutation were associated with a significantly shorter OS compared with patient without KRAS mutation (P = 0.0005), and the same trend of disease-free survival was also found in R0 resected lung cancer patients (P = 0.039). Multivariate analyses proved the status of KRAS mutation was an independent prognostic factor (HR = 2.69, 95% CI: 1.91–3.80, P = 0.0005). No difference was found in PFS and tumor response between the first- or second-line EGFR-TKI-treated patients harboring KRAS mutation and KRAS/EGFR wild-type (the first line: P = 0.23, 0.82; the second line: P = 1.00, 0.48). In the first- and second-line chemotherapy, there was no significant difference for PFS among the three groups (P = 0.15, 0.52).Conclusions
KRAS mutation was a poor prognosis factor, but not an independent predictive factor for EGFR-TKI and chemotherapy in lung cancer patients. EGFR wild-type is major reason for resistant to EGFR-TKI in patients with KRAS mutation.