Background and objectives: While the prognostic role of KRAS status after resection of CRLM has been previously explored, the importance of KRAS status relative to the receipt of preoperative chemotherapy remains largely unknown.
Methods: A total of 430 patients who underwent curative-intent surgery for CRLM between 2000 and 2015 and who had available KRAS genotype data were identified. Data regarding KRAS mutation status, receipt of preoperative chemotherapy, and overall survival (OS) were assessed using univariable and multivariable analyses.
Results: Median patient age was 58 years (IQR, 50.4-66.4 years). A total of 258 patients (60.0%) received preoperative chemotherapy, while 172 (40.0%) had upfront surgery. Median and 5-year OS in the entire cohort was 65.1 months and 53.2%, respectively. KRAS mutation was associated with a worse 5-year OS compared with wild-type tumors (HR 1.41; P = 0.042). After stratifying by the receipt of preoperative chemotherapy, the prognostic value of KRAS mutation only persisted among patients who had received preoperative chemotherapy (HR 1.67; P = 0.012). In contrast, KRAS mutation status had no impact on OS among patients who had not received preoperative chemotherapy (P = 0.597).
Conclusions: KRAS mutation status was an independent predictor of OS among patients undergoing liver resection of CRLM. However, after stratifying by receipt of preoperative chemotherapy, KRAS was informative relative to prognosis only among patients who received preoperative chemotherapy. J. Surg. Oncol. 2016;114:361-367. © 2016 Wiley Periodicals, Inc.
Keywords: CRLM; KRAS mutations; preoperative chemotherapy.
© 2016 Wiley Periodicals, Inc.