Abstract
Salvage treatment for refractory metastatic colorectal cancer (mCRC) has yet to be identified. We aimed to evaluate the efficacy of a salvage lenvatinib-based regimen for refractory mCRC. In total, 371 patients were categorized into lenvatinib-based and non-lenvatinib-based groups. In the lenvatinib-based group, patients who received lenvatinib at a dosage of 10 mg/day were categorized into lenvatinib/chemotherapy and lenvatinib/immunotherapy subgroups. We reported overall survival (OS) and progression-free survival (PFS) using the Kaplan-Meier method. OS1 was used to measure the time from disease progression after TAS-102 and regorafenib treatment to death, while OS2 was used to measure the time from TAS-102 or regorafenib treatment to death. Propensity score matching analysis was employed to compare the characteristics between the lenvatinib-based and non-lenvatinib-based groups. Next-generation sequencing (NGS) information was analyzed using R software. The lenvatinib-based group exhibited longer OS than did the non-lenvatinib-based group (OS1, 11.4 vs. 3.7 months; OS2, 27.2 vs. 8.2 months). The disease control rate (DCR) and objective response rate (ORR) of the lenvatinib-based regimens were 69.4% and 6.1%, respectively. Lenvatinib/chemotherapy and lenvatinib/immunotherapy had similar PFS, OS, DCR, and ORR. The adverse effects were manageable. After propensity score matching, the lenvatinib-based group continued to exhibit significantly longer OS1 and OS2 than did the non-lenvatinib-based group. NGS analysis revealed that GNAS and KRAS alterations were associated with a worse treatment response and prolonged survival, respectively. In conclusion, a moderate-dose salvage lenvatinib-based regimen demonstrated promising clinical activity and tolerability in treating refractory mCRC.