A Cell-Based Screening Method for Resistance of Bcr-Abl Positive Leukemia Identifies the Mutation Pattern for an Alternative Abl Kinase Inhibitor.

The increasing impact of targeted cancer treatment demands strategies to identify and evaluate resistance mechanisms toward kinase inhibitors prior to their therapeutic application. Point mutations within the Bcr-Abl kinase domain constitute the major mechanism of resistance toward imatinib mesylate in Philadelphia-positive (Ph+) leukemia. Using Bcr-Abl-transformed Ba/F3 cells, we established a cell-based screening strategy for the prediction of specific kinase mutations that cause resistance toward kinase inhibitors. With imatinib at clinically relevant concentrations, we generated 368 resistant Ba/F3 sublines that were derived from resistant colonies. Thirty-two different single point mutations within the kinase domain of Bcr-Abl were identified in twenty-five per cent (liquid culture conditions) and seventy-two per cent (solid culture conditions) of these lines at known and novel positions. Using imatinib, the pattern and relative frequency of mutations reflected matters observed in patients with imatinib resistance. We then applied this screen to the pyrido-pyrimidine PD166326 (PD16), an investigational Abl kinase inihibitor. Compared to imatinib, we observed a five to seven times lower frequency of resistant colonies with equipotent concentrations of PD16. In addition, PD16 produced a distinct pattern of Bcr-Abl mutations. P-loop, A-loop and the known imatinib contact site T315 were affected with both inhibitors, whereas C-helix and SH2 contact sites were affected in imatinib resistant colonies exclusively. In contrast to imatinib, where kinase domain mutations were still widely distributed over the kinase domain even at at 4μM, mutations observed with PD16 at a concentration of 100nM narrowed to the exchange at position T315 to iseulicine. We did not detect mutations outside the kinase domain. Some resistant sublines displayed increased Bcr-Abl activity. Mutations that were derived from the screen were cloned and examined for the extent of cross-resistance to both inhibitors. The majority of mutations were effectively suppressed by PD16 at 50–500nM. In contrast, only few mutations were inhibited by imatinib at 5–10μM. However, exchanges at position F317 mediated resistance toward PD16, but were inhibited by standard concentrations of imatinib. Since this cell-based system produced results that are clinically significant, it may be used to predict resistance mutations in Bcr-Abl and other oncogenic kinases like cKit, EGFR, FIP1L1-PDGFRalpha or FLT3 towards clinically applicated and investigational drugs. Thus, this robust and simple screening strategy provides a rational basis for combinatorial and sequential treatment strategies.