Objective To evaluate the clinical efficacy and safety of flumatinib in the treatment of chronic myelocytic leukemia (CML) by using propensity score matching (PSM) to balance baseline confounding factors.

Methods A retrospective cohort of 201 patients with CML treated between January 2019 and January 2025 was selected for this study. Patients were stratified into imatinib and flumatinib groups based on their treatment regimens. PSM was employed to control for confounding variables, enabling a comparative analysis of early treatment responses and long-term survival outcomes, including overall survival (OS), between the two tyrosine kinase inhibitor therapies, as well as further analysis of long-term prognosis.

Results Through PSM, a total of 61 patient pairs were successfully matched. In the flumatinib group, the rates of complete hematologic response (CHR) and early molecular response (EMR) at 3 months of first-line treatment were 96.72% and 83.61%, respectively, with the EMR rate being significantly higher than achieved with imatinib treatment (P < 0.001). For flumatinib, the rates of cumulative major molecular response (MMR) at 6, 9, and 12 months of first-line therapy were 34.43%, 50.82%, and 59.02%, respectively, all exceeding those observed with imatinib (P < 0.05). The molecular response 4 (MR⁴) rates for flumatinib at 9 and 12 months of first-line treatment were 19.67% and 27.87%, respectively, also higher than those for imatinib (P < 0.05). Additionally, the cumulative complete cytogenetic response (CCyR) rates for flumatinib at 3, 6, and 12 months of first-line treatment were 67.21%, 85.25%, and 91.28%, respectively, consistently outperforming imatinib (P < 0.05). Generalized estimating equation analysis revealed that patients in the flumatinib group had a significantly higher probability of achieving MMR than those in the imatinib group (Wald χ² = 16.280, P < 0.001, odds ratio OR = 2.130, 95% CI: 1.570-2.890). The time effect indicated that the probability of achieving MMR increased with the extension of treatment duration in both groups (Wald χ² = 48.720, P < 0.001). In addition, a statistically significant group-by-time interaction effect was observed (Wald χ² = 8.940, P = 0.003). During the treatment period, the highest incidence of diarrhea was observed in the flumatinib group (26.23%), while thrombocytopenia was most prevalent in the imatinib group (31.15%). Furthermore, the imatinib group exhibited higher incidences of rash and edema compared to the flumatinib group (P < 0.05). No significant differences were observed in the incidence of other adverse events between the groups (P > 0.05). The median follow-up duration was 43 (37, 52) months. By the end of the follow-up, a total of 7 patients had died, with 4 deaths (6.56% 4/61) in the imatinib group and 3 deaths (4.92% 3/61) in the flumatinib group. The comparison of OS between the two groups revealed no statistically significant difference (log-rank P = 0.955).

Conclusion Compared with imatinib, flumatinib demonstrates superior short-term efficacy and safety profile in the treatment of CML. Despite the lack of significant improvement in long-term OS, flumatinib can still serve as an effective first-line therapeutic option for patients with CML.