Some patients with differentiated thyroid cancer (DTC) may require an initial low dose (LD) of lenvatinib. However, few studies have investigated the efficacy of LD lenvatinib. We compared the efficacy and tolerability of lenvatinib at an initial LD to those of the standard initial dose of 24 mg in patients with DTC.In this cross-sectional study, records of patients with DTC treated with lenvatinib were retrospectively reviewed. Patients were divided into 2 groups based on the initial dose of lenvatinib: a full-dose (FD) group that received an initial dose of 24 mg/d and a LD group that received an initial dose of less than 24 mg/d. Categorical variables were compared with the Fisher exact test and continuous variables with Student t test. A progression-free survival (PFS) curve was constructed with the Kaplan-Meier method. A probability (P) value of < .05 was considered statistically significant.Thirty-six patients with DTC were treated with lenvatinib (30 in the FD group and 6 in the LD group). The response rates were 43% and 33% in the FD and LD groups, respectively. The median PFS duration was 696 [95% confidence interval (CI): 318-not available (NA)] days in the FD group. The median PFS of the LD group was not reached (95% CI: 124-NA) (P = .293). Treatment interruptions were required in 25 (83%) patients in the FD group and 4 (67%) in the LD group (P = .573). Dose reductions were required in 28 (93%) patients in the FD group and 4 (67%) in the LD group (P = .121). There were no significant differences in the incidences of common adverse events between the 2 groups.The LD group also required dose reduction and interruption frequently. Since these findings are only the short-term results of a limited number of cases, a large number of cases and long-term observations are needed to determine whether an initial LD is effective for patients with DTC in poor general condition.