Purpose: To investigate the prognostic value of KRAS in a large cohort of patients undergoing liver resection for colorectal liver metastases (CRLM).
Methods: Between 2003 and 2013, 334 patients underwent hepatic resection for CRLM at Johns Hopkins Hospital and met the inclusion criteria. Clinicopathologic characteristics, perioperative details, and outcomes were stratified by KRAS status-mutant KRAS (mtKRAS) versus wild-type KRAS (wtKRAS)-and analyzed.
Results: mtKRAS was identified in 115 (34.4 %) patients. At a median follow-up of 28.2 months, recurrence was observed in 59 (51.3 %) patients with mtKRAS and 117 (53.4 %) patients with wtKRAS (P = 0.79); there was no difference in the pattern of recurrence (liver: mtKRAS 39.0 % vs. wtKRAS 52.1 %; lung: mtKRAS 55.6 % vs. wtKRAS 64.3 %; both P > 0.05). Although 5-year log-rank overall survival (OS) was comparable among mtKRAS (41.6 %) vs. wtKRAS (48.5 %), on multivariable Cox survival analysis and after adjusting for known predictors of OS mtKRAS was associated with worse OS (hazard ratio 1.65; 95 % confidence interval 1.07-2.54; P = 0.02). Among patients who experienced a recurrence, 5-year OS was worse among those patients who had mtKRAS (mtKRAS 28.1 % vs. wtKRAS 44.5 %; P = 0.004). After controlling for tumor factors and receipt of chemotherapy, mtKRAS status remained independently associated with a worse outcome among patients who experienced recurrence (hazard ratio 2.07; 95 % confidence interval 1.31-3.27; P = 0.002).
Conclusions: mtKRAS was noted in one-third of patients with CRLM. Although KRAS status did not affect the pattern of recurrence and recurrence-free survival, mtKRAS was an independent predictor of worse OS. The effect was more pronounced among patients who experienced a recurrence after resection of CRLM.