Dasatinib (BMS-354825) is a second-generation BCR-ABL inhibitor active against several imatinib-insensitive BCR-ABL mutant forms. We have treated in the phase II program with dasatinib a total of forty-five Ph+ patients (pts) who were resistant to or intolerant of imatinib. At the time of writing, twenty-one pts (1 pt with CML in chronic phase, 2 pts with CML in accelerated phase, 3 pts with CML in myeloid blast crisis, 15 pts with CML in lymphoid blast crisis or with Ph+ ALL) have failed to respond to or relapsed on dasatinib therapy. In order to assess which pre-existent or emerging ABLkinase domain (KD) mutations are challenging for dasatinib clinical efficacy, we retrospectively analyzed ABL KD sequences before the start of treatment and every month thereafter, until dasatinib discontinuation. Mutation monitoring was done by D-HPLC, followed by sequencing in D-HPLC-positive cases. Eight pts had primary resistance to dasatinib, defined as failure to achieve any hematologic response (HR). In all cases, a T315I or a F317L mutation was already detectable before the onset of treatment or became detectable after one month. The mutations persisted up to the time of disease progression, which occurred at a median of 1.5 months (range, 1-4) from dasatinib start. Thirteen pts had acquired resistance to dasatinib, defined as loss of HR during treatment. Relapse occurred after a median of 7.5 months (range, 3-15) from dasatinib start. Mutation analysis performed before the onset of treatment showed that five of these pts had a wild-type ABL sequence, while the remaining eight pts had evidence of various imatinib-resistant KD mutations (G250E, Y253H, E255K, D276G, M351T, L387M). At the time of relapse, however, most of the original mutant clones had disappeared, whereas mutations at residues 315 and/or 317 had invariably emerged in all but one pt. This pt was found to have developed a novel K356R mutation which is now under characterization. Interestingly, unreported amino acid substitutions (T315A and F317I) were found in two cases at the time of or immediately before relapse. Our results indicate that residues 315 and 317 are mutation hotspots in dasatinib-resistant pts, according to the experimental observation that they are both involved in inhibitor binding. They also provide a proof of principle that novel, inhibitor-specific mutant variants (i.e., T315A, F317I, K356R) may be selected, and raise some concerns about the limitations of single-agent treatment in the long term disease control of advanced phase-CML or Ph+ ALL pts. Supported by European LeukemiaNet, AIL, AIRC, FIRB and PRIN projects.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA