Response to Dasatinib after Imatinib Failure According to Type of Preexisting BCR-ABL Mutations.

Dasatinib (BMS-354825) is a novel, oral, multi-targeted kinase inhibitor of BCR-ABL and SRC kinases with preclinical activity against 20/21 imatinib resistant BCR-ABL mutations and clinical phase I/II efficacy in patients with chronic myelogenous leukemia (CML) and BCR-ABL positive acute lymphoblastic leukemia (ALL). We sought to establish a relationship between type of preexisting BCR-ABL mutations associated with imatinib resistance and efficacy of dasatinib in patients (pts) with CML and ALL. We have investigated 872 peripheral blood samples from 394 pts (53% male, median age 60 yrs, range 17–85) who had been enrolled in international phase II studies investigating the activity of 70mg dasatinib BID after imatinib failure (chronic phase, CP, n=198; accelerated phase, AP, n=78; myeloid blast crisis, MyBC, n=53; lymphoid blast crisis, LyBC, or ALL, n=65).

Screening for BCR-ABL mutations was performed by D-HPLC combined with DNA sequencing. During follow up, pts were monitored in 3-monthly intervals by RQ-PCR for BCR-ABL mRNA transcripts and by mutation analysis to determine the quantitative course of the preexisting mutation or the emergence of new mutations. Hematologic and cytogenetic response data have been collected sequentially for a median of 8 months (range, 2–11) after start of therapy. Prior to dasatinib, 46 different BCR-ABL mutations involving 36 amino acids were detected in 202/394 pts (51%). 162 pts showed one, 33 pts two, 6 pts three, and 1 pt four mutations. Mutations were observed in 84 pts in CP (42%), 47 pts in AP (60%), 23 pts in MyBC (43%), and 48 pts in LyBC and ALL (74%). In patients with mutations, hematologic response was 91% in CP, 62% in AP, 41% in MyBC, and 34% in LyBC/ALL (p<0.01), major and complete cytogenetic response did not differ significantly (47% and 34% in CP, 35% and 27% in AP, 33% and 28% in MyBC, 53% and 51% in LyBC/ALL). Major cytogenetic response rates were comparable in pts bearing no mutations (44%), mutations within the P-Loop (43%), SH2 domain (47%), activation loop (56%), or other sites (49%), but not for T315I (0%, p<0.001). Sorting individual pts by the underlying mutation and cellular IC50 values of dasatinib revealed clearly higher hematologic and cytogenetic response rates in those with lower IC50 values. In line with the virtual insensitivity to dasatinib in vitro, none of the 17 pts carrying a T315I mutation showed any hematologic response. Two distinct patterns of response were observed:

1. a parallel decrease of the BCR-ABL load and the mutated clone or

2. a decrease of the BCR-ABL load followed by a decrease of the mutated clone after a delay of up to 4–6 months.

Up until now 5 pts developed new mutations associated with dasatinib resistance (T315I, n=2, AP and MyBC pts; F317L+F486S, n=2, LyBC and MyBC pts; E507G, n=1, CP pt). We conclude that dasatinib is capable of inducing hematologic and cytogenetic remissions in a significant proportion of imatinib resistant pts harboring BCR-ABL mutations, except T315I. Response dynamics depend on the individual type of mutation which may be a basis for individual dose adaptation according to the mutation pattern.