PURPOSE To evaluate the efficacy and safety of adjuvant hepatic arterial infusion (HAI) of oxaliplatin combined with intravenous (IV) fluorouracil/leucovorin (LV5FU2) after curative-intent surgery of ≥4 colorectal liver metastases (CRLM). METHODS Patients with an Eastern Cooperative Oncology Group performance status 0-1, who underwent resection or ablation of ≥4 CRLM after preoperative IV chemotherapy were enrolled. Patients were randomly assigned (1:1) to receive adjuvant oxaliplatin via HAI (HAI group, n = 50) or IV infusion (IV group, n = 49), both combined with IV LV5FU2 for at least 3 months. The primary end point was hepatic recurrence-free survival (h-RFS). Secondary end points included RFS, overall survival (OS), safety, and feasibility. RESULTS After a median follow-up of 59 months (IQR, 45-71), the median h-RFS was 25 (95% CI, 16 to 37) months in the HAI group versus 12 (95% CI, 8 to 19) months in the IV group (hazard ratio HR, 0.63 95% CI, 0.40 to 0.99; P = .047). Median RFS was 14 months (95% CI, 10 to 20) in the HAI group versus 9 months (95% CI, 7 to 11) in the IV group (HR, 0.63 95% CI, 0.41 to 0.97; P = .03). Median OS was 74 months (95% CI, 51 to not defined) in the HAI group versus 57 months (95% CI, 37 to 69) in the IV group (HR, 0.61 95% CI, 0.33 to 1.12; P = .11). Five-year OS was 62% in the HAI arm compared with 47% in the IV arm. Grade 3 to 4 adverse events occurred in 58% of HAI patients and 32% of IV patients ( P = .02). No treatment-related deaths were reported. Four or more cycles of adjuvant chemotherapy were delivered in 81% of patients in the HAI group and 78% in the IV group ( P = .75). CONCLUSION Adjuvant oxaliplatin HAI plus LV5FU2 improves h-RFS after curative-intent surgery of CRLM in high-risk patients, with an acceptable safety profile. These results support further evaluation in a phase III trial.
Gelli et al. (Wed,) studied this question.