Purpose: To evaluate Kirsten rat sarcoma viral oncogene homolog (KRAS) mutation status as a prognostic factor for survival after yttrium-90 ((90)Y) radioembolization for colorectal cancer (CRC) liver metastases.
Materials and methods: Consecutive patients with unresectable CRC liver metastases and documented KRAS mutation status who were treated with (90)Y radioembolization during the period 2007-2014 were investigated. Patient demographics, disease characteristics, therapy regimens, and overall survival (OS) from first (90)Y radioembolization were compared between patients with KRAS wild-type (wt) and mutant status. Kaplan-Meier estimation and Cox regression were used for survival analysis and to assess independent prognostic factors for OS.
Results: Of 186 patients, 104 underwent KRAS mutation analysis before (90)Y radioembolization, with 45 (43.3%) identified as mutant. The wt and mutant groups were similar in demographics, liver status, overall performance status, and tumor characteristics (all P > .05). Mean time from liver metastasis to (90)Y radioembolization was greater in patients with KRAS wt status (P = .033). A greater percentage of wt patients received anti-epidermal growth factor receptor therapies before (90)Y radioembolization (66.1% vs 8.9%; P < .001). Median OS from first (90)Y radioembolization was significantly greater in KRAS wt patients (9.5 mo vs 4.8 mo; P = .041). Univariate analysis identified Child-Pugh class, carcinoembryonic antigen (CEA), chemotherapy after (90)Y radioembolization, KRAS status, and treatment-induced toxicity as prognostic factors for OS. Multivariate Cox regression analysis demonstrated Child-Pugh class, CEA, and KRAS status to be independent prognostic factors for OS, even when correcting for the effect of chemotherapy after (90)Y radioembolization.
Conclusions: Patients with CRC and KRAS wt may derive greater survival benefit from (90)Y radioembolization therapy than patients with KRAS mutant.
Copyright © 2015 SIR. Published by Elsevier Inc. All rights reserved.