Objective： To investigate the dynamic change of absolute natural killer cell counts（ANKC） and its prognostic value for the treatment response in the newly diagnosed patients with chronic myeloid leukemia（CML） before and after 3 months and 12 months of tyrosine kinase inhibitors（TKI） treatment. Methods：All 80 patients with chronic-phase CML（CML-CP） diagnosed from 2014 to 2019 were treated with first-generation TKI（imatinib） or second-generation TKI（nilotinib, dasatinib）. According to the treatment response，they were divided into two groups：the optimal response group and the warning/treatment failure group，aiming to analyze the influence of ANKC and related clinical data on prognosis. The receiver operating characteristic（ROC）curve was used to optimize the cut-off values of clinical indicator. Results：After 3 months and 12 months of treatment，the efficacy of patients with ANKC≥0.14×109 /L at the first diagnosis was better than those with lower ANKC（P < 0.05）. Following 3 months of TKI treatment，ANKC in patients with optimal response was significantly lower than that at initial diagnosis（P < 0.001），whereas ANKC in patients without optimal response was significantly higher（P=0.017）. Dynamic analysis was performed on the correlation between the changes in ANKC and the efficacy in 13 patients receiving dasatinib，and we found that in the patients with optimal response，ANKC after 3 months of treatment was significantly lower than that at the initial diagnosis（P=0.025），whereas ANKC after 12 months of treatment was significantly higher than that at the initial diagnosis（P=0.015）. Multivariate analysis showed that ANKC，white blood cells count and second-generation TKI were independent factors for treatment response（P < 0.05）. Conclusion：ANKC in CML-CP patients is an effective prognostic factor for TKI treatment response，meanwhile patients with ANKC≥0.14×109/L at diagnosis can achieve an optimal response.