Chronic Myelogenous Leukemia Patients with Highly Nilotinib-Resistant BCR-ABL Mutations Demonstrate Similar Efficacy to Dasatinib, but Have a Higher Likelihood of Developing New Mutations in the Event of Clinical Resistance

Assess the outcome of imatinib- and/or nilotinib-failure chronic myelogenous leukemia (CML) patients with pre-existing, highly nilotinib-resistant mutations (Y253H, E255K/V and F359V/C), during dasatinib therapy.

Sixty-one CML patients who failed imatinib (n=44) or imatinib and nilotinib (n=17), including 20 in chronic phase (CP), 16 in accelerated phase (AP) and 25 in blast phase (BP), were switched to dasatinib, except for patients with highly dasatinib-resistant mutations (T315I/A, F317L/V/I/C and V299L). Patients were grouped into 3 cohorts: no mutation (n=22), no-specific mutation (any mutation except highly nilotinib-resistant mutations; n=15), or specific mutation (Y253H, E255K/V or F359V/C; n=24), according to baseline BCR-ABL mutation status. Cumulative incidence frequencies of hematological response (HR), complete hematological response (CHR), major cytogenetic response (MCR), complete cytogenetic response (CCR) and major molecular response (MMR) were compared. Probabilities of clinical resistance to dasatinib (according to European LeukemiaNet recommendations), progression-free survival (PFS) and overall survival (OS), and dynamics of BCR-ABL mutations during dasatinib therapy were also assessed.

Prior nilotinib was associated with a higher proportion of specific mutation versus no mutation (50.0% vs 13.6%, P=0.001) and non-specific mutation (50.0% vs 13.3%, P=0.02). Patients harboring specific mutation also had a higher frequency of advanced phase CML before dasatinib treatment than patients with no mutation (83.3% vs 50.0%, P=0.008), and a similar frequency to those with non-specific mutation (83.3% vs 66.7%, P=0.266). Response, PFS and OS rates, and follow-up duration were similar among the 3 cohorts by baseline mutation status, as shown in Table. Probability of clinical resistance to dasatinib in patients with specific mutation was not significantly different to those with no mutation (54.2% vs 77.3%, P=0.100) or non-specific mutation (54.2% vs 33.3%, P=0.204) despite that there was a significant difference between the cohort with no mutation and non-specific mutation (77.3% vs 33.3%, P=0.008). Among 32 patients who developed clinical resistance to dasatinib and were assessed for BCR-ABL mutation status, newly detectable mutations were identified in 14 patients and most frequently occurred in those already harboring specific mutation versus those with no mutation (76.9% vs 26.7%, P=0.008) and non-specific mutation (76.9% vs 0%, P=0.015) at baseline. T315I (9/14, 64.3%) was the most common newly acquired BCR-ABL mutation. Harboring specific mutation compared with other mutation states at baseline (76.9% vs 16.7%, P=0.001), and developing hematological resistance rather than cytogenetic resistance during dasatinib therapy (65.0% vs 8.1%, P=0.002), were identified as risk factors associated with the development of new mutations.

Imatinib- and/or nilotinib-failure patients with highly nilotinib-resistant BCR-ABL mutations demonstrate similar efficacy to dasatinib as those with no or any other mutation. However, patients with highly nilotinib-resistant mutations had a higher likelihood of developing new mutations at the time of clinical resistance. More extensive studies are needed to assess the significance of various BCR-ABL mutations prior to and during tyrosine kinase inhibitor therapy for CML.

No relevant conflicts of interest to declare.