Ponatinib-based combination therapies including asciminib and venetoclax were effective on patient-derived xenograft models of chronic myeloid leukemia in blast Phase

Abstract Introduction:Blast phase of chronic myeloid leukemia (CML-BP) represents advanced and aggressive phase of CML, marked by high genetic heterogeneity, compromised efficacy of BCR::ABL1-targeted tyrosine kinase inhibitors (TKIs), and poor clinical outcomes. Emerging evidence indicates that combination therapies may offer more effective treatment options in this setting. This European Treatment and Outcome Study (EUTOS 2024) work aims to evaluate the efficacy of allosteric BCR::ABL1 inhibitor asciminib, 2nd generation TKI nilotinib, 3rd generation TKI ponatinib, BCL2 inhibitor venetoclax, and their combinations in patient-derived xenograft (PDX) models of CML-BP. Methods:Leukemic cells from CML-BP patients were analyzed for chromosomal abnormalities and mutations in BCR::ABL1 kinase domain (KD) and 61 other leukemia-associated genes. Ex vivo drug sensitivity was conducted. CML blasts were transplanted into immunodeficient mice to establish PDX models. Animals (n=5–8 per group) were randomized into control or treatment arms and received daily oral treatment for 7 days: ponatinib (25 mg/kg b.w.), nilotinib (40 mg/kg b.w.), asciminib (30 mg/kg b.w.), venetoclax (50 mg/kg b.w.), or their combinations. Tumor growth and event-free survival (EFS) were assessed. The experimental design was approved by the institutional Animal Care and Use Committee and the Ministry of Education Youth and Sports of the Czech Republic. Results: The PDX model BC-CML2, derived from a 70-year-old male patient with myeloid CML-BP following rapid progression on imatinib, exhibited a complex karyotype with additional Philadelphia chromosomes but no detectable somatic mutations. In line with ex vivo findings, the model was poorly responsive to nilotinib, asciminib and venetoclax monotherapies but responded well to ponatinib. Combination therapies—asciminib+venetoclax and especially ponatinib+venetoclax—significantly reduced tumor growth and prolonged EFS compared to controls and monotherapies (p<0.001). Conversely, combination of nilotinib and asciminib have limited efficacy against tumor growth and have no survival benefit. The BC-CML3 model originated from a 58-year-old female patient with B-lymphoid CML-BP in 3rd relapse on ponatinib. Blasts carried BCR::ABL1 compound mutations (T315I+E255K and T315I+E459K), a RUNX1 mutation (R201Q), and a complex polyclonal karyotype. This model was resistant to venetoclax and nilotinib alone and showed only modest responses to asciminib and ponatinib monotherapies. However, combinations of asciminib+ponatinib and ponatinib+venetoclax delayed disease progression and significantly improved EFS compared to controls (p<0.001) and single agents (p<0.01 for asciminib+ponatinib; p<0.001 for ponatinib+venetoclax). The BC-CML4 model was established from a 65-year-old male patient with myeloid CML-BP relapsing after four lines of TKI therapy including ponatinib. Mutational profiling revealed polyclonal BCR::ABL1 mutations (E255K and E255V), along with NRAS (G12D) and GATA2 (R307Q) mutations, and a complex karyotype. Ex vivo testing showed resistance to venetoclax. Ponatinib and its combination with venetoclax significantly delayed tumor growth and extended EFS (p<0.001). Conclusions: Preclinical testing in CML-BP PDX models confirmed the enhanced efficacy of combination therapies. Across all models, dual targeting with ponatinib and venetoclax was most effective. In the BC-CML3 model, harboring compound BCR::ABL1 mutations linked to ponatinib resistance, the asciminib+ponatinib combination showed notable activity, consistent with previous findings in cell line-derived xenograft (CDX) models and in vitro testing. Support:EUTOS 2024, MH CZ – DRO (IHBT, 00023736)