Meta-Analysis of Hepatic Arterial Infusion for Unresectable Liver Metastases From Colorectal Cancer: The End of an Era?

PURPOSE: The treatment of unresectable liver-confined metastatic disease from colorectal cancer (CRC) is a challenging issue. Although locoregional treatments such as hepatic arterial infusion (HAI) claim the advantage of delivering higher doses of anticancer agents directly into the affected organ, the benefit in terms of overall survival (OS) is unclear. We quantitatively summarized the results of randomized controlled trials (RCT) comparing HAI with systemic chemotherapy (SCT). METHODS: To date, 10 RCTs have been published, for a total of 1,277 patients enrolled. For tumor response rates, relative risks (RR) and their 95% CIs were obtained from raw data; for OS, hazard ratios (HRs) and their 95% CIs were extrapolated from the Kaplan-Meier survival curves. RESULTS: HAI regimens were based on floxuridine (FUDR) in nine of 10 RCTs, whereas in one RCT, fluorouracil (FU) + leucovorin was used. SCT consisted of FUDR, FU, FU + leucovorin, or a miscellany of FU and best supportive care in three, one, four, and two studies, respectively. Pooling the data, tumor response rate was 42.9% and 18.4% for HAI and SCT, respectively (RR = 2.26; 95% CI, 1.80 to 2.84; P < .0001). Mean weighted median OS times were 15.9 and 12.4 months for HAI and SCT, respectively; the meta-risk of death was not statistically different between the two study groups (HR = 0.90; 95% CI, 0.76 to 1.07; P = .24). CONCLUSION: Currently available evidence does not support the clinical or investigational use of fluoropyrimidine-based HAI alone for the treatment of patients with unresectable CRC liver metastases, at least as a first-line therapy.