2025 Volume 71 Issue 4 Pages 277-282
Chronic myeloid leukemia (CML) is a myeloproliferative neoplasm driven by the BCR::ABL tyrosine kinase. Tyrosine kinase inhibitors (TKIs) have significantly improved the outcome of CML patients. Dasatinib, a second-generation TKI, is highly effective but associated with off-target effects, including pulmonary toxicities. While pleural effusion induced by dasatinib has been linked to therapeutic efficacy, its role remains controversial. Severe pulmonary complications, such as diffuse alveolar hemorrhage (DAH), can lead to treatment failure and increased mortality. We report a 72-year-old man with de novo blast-phase CML on clopidogrel who developed respiratory failure due to DAH 16 days after initiating dasatinib and prednisolone as induction therapy. Immediate steroid pulse therapy with methylprednisolone (1,000 mg/day for three days) was administered, and both dasatinib and clopidogrel were discontinued. Maintenance prednisolone (1 mg/kg/day) was then tapered by 10 mg per week. The patient’s symptoms and radiographic findings improved without recurrence during tapering. This case highlights the importance of early recognition and management of severe complications like DAH in patients receiving dasatinib. Careful monitoring is essential to mitigate the risk of life-threatening respiratory failure and optimize CML treatment outcomes.
Chronic myeloid leukemia (CML) is a myeloproliferative neoplasm driven by the BCR::ABL tyrosine kinase1). The development of tyrosine kinase inhibitors (TKIs) has dramatically improved CML treatment and prognosis1,2). Patients achieving a deep molecular response can expect a life expectancy comparable to individuals without CML3). However, the optimal treatment strategy for de novo blast phase CML (BP-CML), particularly in elderly patients ineligible for hematopoietic stem cell transplantation, remains undefined and represents a pressing challenge4). For lymphoid crisis in BP-CML, induction therapy following the principles of Philadelphia chromosome-positive (Ph+) acute lymphoblastic leukemia (ALL) is administered, often involving TKI-based regimens.
Dasatinib, a second-generation TKI, has shown high efficacy in newly diagnosed chronic phase (CP)-CML and Ph+ ALL5). It is increasingly preferred over imatinib due to superior BCR::ABL inhibition and central nervous system penetration6). However, dasatinib, as a multi-kinase inhibitor, has off-target effects including pulmonary toxicities such as interstitial lung disease, pleural effusion, and pulmonary artery hypertension7-11). In contrast, diffuse alveolar hemorrhage (DAH) caused by dasatinib is exceptionally rare, with only suspected cases reported12). DAH, characterized by bleeding into alveolar spaces, can lead to severe respiratory failure requiring prompt diagnosis and treatment13).
We present a rare case of dasatinib-induced DAH in a patient with BP-CML driven by major BCR::ABL. The patient’s symptoms resolved following steroid pulse therapy and dasatinib withdrawal, and disease management was successfully continued with alternative TKIs. This case underscores the importance of recognizing and managing this rare but severe adverse effect.
A 72-year-old man was referred to our hospital due to elevated white blood cell (WBC) counts. The patient had a history of type 2 diabetes mellitus and sick sinus syndrome with paroxysmal atrial fibrillation, for which a permanent pacemaker had been implanted. Cardiac function was stable, with no signs of heart failure, such as dyspnea or edema. The patient was on long-term medication, including clopidogrel, olmesartan, rosuvastatin, and canagliflozin, and had no history of smoking, chemotherapy, or radiation therapy. Acute leukemia was suspected based on detailed examinations conducted at our hospital, and a plan was made for inpatient evaluation and treatment. A physical examination revealed no abnormal breath sounds, palpable lymphadenopathy, or splenomegaly. Laboratory tests showed a marked increase in WBC count to 111.1×103/µL, with an increased number of blast cells. A differential WBC count revealed that 69% blast cells, 2% promyelocytes, 4% myelocytes, 2% metamyelocytes, 1% lymphocytes, 6% monocytes, and 3% eosinophils, indicating an increase in immature granulocytes, and eosinophils. The hemoglobin level was 13.3 g/dL, and the platelet count was 70 × 103/μL, indicating a mild decrease in hemoglobin and platelet count. Additionally, lactate dehydrogenase was markedly elevated to 2,135 U/L, and Wilms tumor 1 mRNA was detected at 2,900 copies/µg RNA in peripheral blood (PB). To determine the type of leukemia, bone marrow (BM) aspiration was performed. BM aspirates showed an elevated nucleated cell count of 477,500/µL, with 74.6% blast cells and no evidence of myelodysplastic changes (Figure 1A). The blasts were positive for myeloperoxidase (MPO) staining and exhibited lymphoblast-like morphology. In flow cytometry, the blasts simultaneously expressed myeloid markers MPO, CD13, and CD33, as well as B-cell lineage markers CD10, CD19, CD20, and TdT, and were also positive for CD34, CD38, and CD66c. Therefore, the involvement of BCR::ABL was suspected. Major BCR::ABL was detected using real-time PCR from BM cells, and BCR::ABL translocation was confirmed in neutrophils from PB using fluorescence in situ hybridization. Chromosomal analysis of BM cells revealed a translocation between chromosomes 9 and 22, known as the Ph (20/20 examined cells). Two subclones were present, one with del(6) and add(14)(q32) in 17/20 cells, and the other with additional chromosomal abnormalities, +8, +9, and +22 in 3/20 cells. Based on these findings, the patient was diagnosed with lymphoid crisis of BP-CML.
Treatment with dasatinib (100 mg/day) was initiated seven days after starting prednisolone (PSL, 1 mg/kg/day)6). The treatment response was favorable, with a decrease in WBC count and blast cells (Figure 1C). However, 16 days after dasatinib initiation, the patient developed dyspnea, with a respiratory rate of 30 breaths/min and percutaneous oxygen saturation of 83% on ambient air. Coarse crackles were audible throughout the lung field. A chest radiograph revealed consolidation and diffuse ground-glass opacities (GGO) (Figure 2A). Chest computed tomography (CT) revealed diffuse alveolar GGO with a superimposed irregular reticular pattern, referred to as the crazy-paving pattern14) (Figure 2B). Laboratory findings showed no hematological or coagulation abnormalities. Although surfactant protein-D (SP-D) and Krebs von den Lungen-6 (KL-6) levels were within normal limits, surfactant protein-A (SP-A) was slightly elevated at 74.8 ng/mL. Brain natriuretic peptide was mildly elevated at 216 pg/mL. Autoimmune markers, including antinuclear antigen, anti-Sjögren syndrome antigen-A (SS-A) antibody, anti-SS-B antibody, and anti-neutrophil cytoplasmic antibodies were all negative. The transthoracic echocardiogram revealed an ejection fraction of 70% and no evidence of cardiac dysfunction, such as wall motion abnormalities or valvular defects. Bronchoscopy did not detect active alveolar hemorrhage, but hemoptysis was accumulated in the right main bronchus. Grossly bloody bronchoalveolar lavage fluid (BALF) from the right B5a demonstrated an increased number of total cells and neutrophils (total cell counts, 121.8 × 104/ml BALF;alveolar macrophages:5.5%;lymphocytes: 1.7%;neutrophils:92.7%;eosinophils: 0.0%; CD4/CD8 ratio:0.2). Additionally, the microbiological cultures, including acid-fast bacilli of BALF was negative. Hemosiderin-laden macrophages were also observed in BALF. Based on these findings, the patient was diagnosed with DAH caused by dasatinib.
Steroid pulse therapy with methylprednisolone (1,000 mg/day for three days) was initiated, and both dasatinib and clopidogrel were discontinued. Following pulse therapy, maintenance PSL therapy (1 mg/kg/day) was administered and tapered by 10 mg per week. His symptoms and chest radiograph findings gradually improved, and no recurrence was observed during the PSL tapering process (Figure 2C). A follow-up BM examination revealed normocellular BM without increased blast cells, consistent with CP-CML (Figure 1C). The BCR::ABL transcript level on the international scale (IS) was 4.52%, indicating that short-term dasatinib and PSL treatment were effective. CML treatment was resumed with bosutinib once the patient’s performance status improved. While bosutinib did not induce pulmonary symptoms, generalized edema developed, prompting a switch to asciminib. The patient has experienced no adverse effects since switching to asciminib. Approximately six months after restarting TKI therapy, the BCR::ABL transcript level on the IS in PB decreased to 0.34%, achieving an optimal molecular response. We plan to gradually increase the dose of asciminib to 80 mg/day while closely monitoring for the occurrence of adverse events.
Clinical course of the patient during chemotherapy and management of DAH
(A) Initial bone marrow examination showed an increased percentage of blast cells, consistent with a diagnosis of acute lymphoblastic leukemia. (B) Reduction of blast cells was observed following the withdrawal of bosutinib. (C) The complete blood counts, percentage of blast cells, BCR::ABL transcript levels on the international scale (IS), and therapeutic agents administered during the patient’s treatment course are illustrated.
Chest imaging findings
(A) Chest radiograph taken 16 days after initiating dasatinib shows diffuse ground-glass opacities and consolidation. (B) Chest CT reveals bilateral alveolar ground-glass opacities with a superimposed irregular reticular pattern, characteristic of a crazy-paving pattern. (C) Post-steroid pulse therapy, the chest radiograph demonstrates significant improvement.
Laboratory data of the patient at the onset of diffuse alveolar hemorrhage.
PT:prothrombin time, PT-INR:prothrombin time international normalized ratio, APTT:activated partial thromboplastin time, PR3-ANCA:proteinase3 anti-neutrophil cytoplasmic antibody, MPO-ANCA: myeloperoxidase anti-neutrophil cytoplasmic antibody.
Dasatinib is commonly used for de novo BP-CML in lymphoid crisis based on clinical trial data6). In this case, dasatinib triggered respiratory failure due to DAH. While the patient had been on long-term clopidogrel therapy for cardiac complications, the timeline indicates dasatinib as the primary trigger. Dasatinib targets multiple kinases, including the SRC kinase family, but its off-target effects, such as pleural effusion and DAH, remain significant concerns7). Although pleural effusion has been linked to high therapeutic efficacy, this association is increasingly questioned15). Severe complications like DAH pose a significant risk of treatment-related mortality. Randomized trials have not shown dasatinib to be superior to imatinib-containing regimens, and the risk factors for dasatinib-induced DAH remain unclear, underscoring the importance of avoiding this complication whenever possible16).
DAH involves bleeding into alveolar spaces and can arise from various causes, including vasculitis, collagen diseases, coagulopathy, and drug effects11). Dasatinib-induced pulmonary vascular damage has been linked to endoplasmic reticulum stress, mitochondrial reactive oxygen species production, and inhibition of the platelet-derived growth factor receptor8,9,11). In this case, autoimmune markers and acute heart failure indicators were negative, and bronchoscopy and BALF excluded infections. While clopidogrel can cause bleeding, DAH from long-term anti-platelet monotherapy is rare and usually occurs shortly after drug initiation13,17-20). Given the patient’s 10-year history of clopidogrel use without complications and the immediate onset of DAH following dasatinib initiation, dasatinib was identified as the primary cause. Steroid pulse therapy was initiated due to the severity of symptoms, alongside discontinuation of dasatinib and clopidogrel, which led to significant improvement. After tapering steroids, bosutinib was initiated for CML management due to the risk of cross-intolerance among TKIs16). The patient achieved hematological remission without recurrence of DAH. However, bilateral lower limb edema developed two months after starting bosutinib. Subsequently, asciminib, a third-generation TKI targeting the ABL myristoyl pocket with reduced off-target effects, was prescribed2,21). The patient tolerated asciminib well and achieved an optimal response without adverse effects. Although the disease achieved a second CP, evidence supporting the use of asciminib in treating de novo BP-CML remains limited, warranting careful monitoring of the patient’s progress.
Additionally, as far as we could identify, there are only two reported cases in which DAH was suspected to be associated with dasatinib. One case involved a patient with CP-CML12) and the disease phase was not specified in the second case22). Given these limited cases, including the present case, it can be inferred that dasatinib-induced DAH is not exclusively observed in BP-CML but can also occur in CP-CML as a potential complication. The detailed mechanism remains unclear, and further investigation is required.
In conclusion, this report highlights a rare but severe complication of dasatinib-induced DAH. While TKIs remain essential for CML treatment, awareness of severe adverse events is crucial. This case underscores the importance of careful monitoring and individualized management to optimize CML treatment outcomes.
Takayuki Ikezoe:Honoraria, Asahi Kasei, Nippon Shinyaku, Chugai, Sanofi, Alexion and Pfizer.
We obtained informed consent from the patient for the publication of this work.
