Outcome of patients with chronic myeloid leukemia in lymphoid blastic phase and Philadelphia chromosome–positive acute lymphoblastic leukemia treated with hyper-CVAD and dasatinib

Abstract

BACKGROUND: Dasatinib monotherapy has demonstrated modest clinical activity in chronic myeloid leukemia in lymphoid blastic phase (CML-LBP). The outcome of Philadelphia chromosome (Ph)–positive acute lymphoblastic leukemia (ALL) has dramatically improved with hyperfractionated cyclophosphamide, vincristine sulfate, doxorubicin hydrochloride, and dexamethasone (hyper-CVAD) in combination with tyrosine kinase inhibitors (TKIs). METHODS: The authors reviewed 85 patients (23 with CML-LBP and 62 with newly diagnosed Ph-positive ALL) who received hyper-CVAD plus dasatinib. RESULTS: In the CML-LBP cohort, 19 had prior chronic myeloid leukemia as chronic phase (n = 17; 74%), accelerated phase (n = 1; 4%), or myeloid blastic phase (n = 1; 4%); 4 (17%) presented with de novo CML-LBP. The BCR-ABL1 transcript was p210 in 22 patients (96%) and p190 in 1 patient (4%). In the Ph-positive ALL cohort, p210 and p190 transcripts were detected in 13 patients (21%) and 48 patients (77%), respectively. Patients with CML-LBP were less likely to achieve deep molecular remission than patients with Ph-positive ALL: the major molecular response (MMR) rates were 70% and 95%, respectively (P =.007), and the complete molecular response (CMR) rates were 55% and 74%, respectively (P =.16). Survival outcomes were similar for CML-LBP and Ph-positive ALL: the 5-year overall survival (OS) rates were 59% and 48%, respectively (P =.97). Allogeneic stem cell transplantation was associated with a better outcome in CML-LBP (5-year OS rate, 88% vs 57%; P =.04). In Ph-positive ALL, the outcome was driven by deeper molecular remission: the 5-year OS rates were 63% and 25% with CMR and MMR, respectively (P =.002). CONCLUSIONS: The outcome of CML-LBP has improved with hyper-CVAD plus dasatinib therapy with survival comparable to that of Ph-positive ALL. Further improvement may be achieved with the use of novel TKIs and targeted agents.