With the chemembolization of colorectal-cancer (CRC)-metastatic hepatic lesions by irinotecan-loaded microspheres, most researchers recommend slow embolizate delivery at the lobar-artery level to the entire liver parenchyma without obtaining visible stasis. An association has been reported between postoperatively visible embolizate stasis and lesion response to treatment. Possibly, in some cases, more selective administration might give greater benefit, particularly with previous systemic chemotherapy failure. Objective: Treatment response evaluation after chemoembolization of CRC-metastatic liver lesions with irinotecan-loaded microspheres, according to a hepatic-artery branch level of administration. Patients and methods: The analysis included 54 patients (24 females, 30 males) with large (median diameter > 5 cm) CRC-metastatic liver lesions, who underwent 196 chemoembolization procedures (mean 3.63 per patient) with irinotecan (100 mg)-loaded microspheres. Patients were divided into two groups according to initial embolizate-administration branch level: Group A (n = 26): at the segmental or subsegmental-vessel level; Group B (n = 28): at the lobar-branch level. Treatment response was assessed by computed-tomography (mRECIST criteria); overall survival (OS) and progression-free survival (PFS) were calculated using the Kaplan–Meier method and adverse effects were assessed according to the Common Terminology Criteria for Adverse Events (CTCAE; version 5.0). Results: There were statistically significant differences in the occurrence of partial response (PR): higher in Group A (42.3%) than Group B (17.9%) (p = 0.039) and occurrence of stable disease (SD): lower (p = 0.025) in Group A (11.5%) than Group B (39.4%). However, occurrence of disease progression (PD) was similar: Group A: 42.3%; Group B: 42.9% (p = 0.93). Patients in Group A presented with more favorable PFS (p = 0.029) and OS (p = 0.039) than Group B. Median survival times: Group A: 15.2 months; Group B: 13.1 months. There was no significant difference in complication incidence between groups (Group A: seven complications; Group B: six complications; p = 0.863). Conclusion: Superselective chemoembolizate administration to vessels supplying large CRC-metastatic liver lesions gave better response to treatment and extended patient survival time, without significantly increasing complication risk.