Resistance Profiling of BCR-ABL Compound Mutations Linked to Tyrosine Kinase Inhibitor Therapy Failure in Chronic Myeloid Leukemia

Abstract 1416

Among patients with chronic myeloid leukemia (CML) who develop resistance to ABL tyrosine kinase inhibitor (TKI) therapy, the most common explanation is acquisition of point mutations within the BCR-ABL kinase domain that interfere with drug binding. The second-generation ABL TKIs nilotinib and dasatinib have proven effective against most imatinib-resistant mutants, with the critical exception of BCR-ABL^T315I. Furthermore, sequential treatment with ABL TKIs can select for BCR-ABL compound mutations (two or more point mutations within the same BCR-ABL molecule) that are resistant to multiple inhibitors. With third-generation ABL^T315I TKIs such as ponatinib (AP24534) and DCC-2036 currently in clinical development, containment of single point mutation-based resistance appears tenable. However, the ability of available inhibitors to address resistance due to the growing number of reported BCR-ABL compound mutations has not been investigated. Here, we generated stable Ba/F3 cell lines expressing a panel of clinically-observed BCR-ABL compound mutants (including two novel mutants which we identified from CML clinical resistance specimens, BCR-ABL^G250E/V299L and BCR-ABL^T315I/H396R) as well as cells expressing each constituent mutation. Cells were plated in graded concentrations of imatinib, nilotinib, dasatinib, ponatinib, and DCC-2036, incubated for 72 hours, and cellular proliferation was analyzed by standard methanethiosulfonate-based assay. We found that, consistent with our previous cell-based mutagenesis screens for resistance (Blood 2006, 108: 2332–8; Cancer Cell 2009, 16: 401–12; Cancer Research 2011, 71: 3189–95), all five tested inhibitors demonstrated unique but partially overlapping resistance profiles relative to the panel of compound mutants tested. As expected, all mutants harboring a T315I component were insensitive to imatinib, nilotinib, and dasatinib, whereas ponatinib and DCC-2036 showed varying levels of efficacy against such mutants. The most resistant, clinically reported compound mutants were BCR-ABL^G250E/T315I, BCR-ABL^E255K/T315I, and BCR-ABL^E255V/T315I, all of which conferred markedly increased resistance to ponatinib and DCC-2036 relative to cells expressing each constituent mutation alone (Table 1). For example, the most resistant compound mutant, BCR-ABL^E255V/T315I, conferred high-level resistance to ponatinib and DCC-2036 (IC₅₀: 425 and 1272 nM, respectively), while cells expressing either BCR-ABL^E255V or BCR-ABL^T315I were inhibited at clinically relevant concentrations of ponatinib (IC₅₀: 33 and 18 nM, respectively) or DCC-2036 (IC₅₀: 659 and 149 nM, respectively). Additional structural, biochemical, and signaling pathway analyses addressing the striking differences in sensitivity of BCR-ABL compound mutants versus their constituent mutations are being pursued and will be presented. Overall, our findings demonstrate that BCR-ABL compound mutations confer varying degrees of resistance to currently available ABL TKIs, necessitating rational treatment selection to optimize outcomes. Patients harboring highly multi-drug-resistant compound mutants such as BCR-ABL^E255V/T315I may have limited therapeutic options available, warranting investigation into combination therapies involving ABL^T315I TKIs. Furthermore, studies of compound mutation-based resistance mechanisms in CML advance the possibility of maximum disease control in a greater proportion of patients and have important implications for anticipating resistance and designing treatment strategies in similar, more rapid clinical resistance scenarios such as those encountered with EGFR or ALK mutations in non-small-cell lung cancer.