Despite the efficacy and safety of dasatinib treatment for chronic-phase chronic myeloid leukemia (CML-CP), adverse effects such as pleural effusion (PE) are still a serious concern. We determined the clinical significance of PE incidence using patient data derived from the D-First clinical study. In the present study, chest radiography and quantification of specific lymphocyte subsets were performed routinely after initiation of dasatinib treatment. Among 52 patients with newly diagnosed CML-CP, 17 (33%) developed PE within 18 months after initial dasatinib administration, but all cases were moderate (Grade 1, 10 patients; Grade 2, 7 patients). CD56+ lymphocyte counts at 1 month correlated significantly with the incidence of PE, whereas lymphocytosis did not. The major molecular response (MMR) rate at 3 months (although not at later times) was significantly higher in PE-positive patients than PE-negative patients (59% versus 24%, respectively; P=0.013). Deep molecular response rates did not differ significantly between the PE groups at any time point during the observation period. Our results suggest that an immune-mediated mechanism involving natural killer cells underlies the development of PE in patients receiving dasatinib for 18 months. This mechanism likely promotes transient tumor regression in patients newly diagnosed with CML-CP.