FLAG (fludarabine, cytarabine, G-CSF) as a second line therapy for acute lymphoblastic leukemia with myeloid antigen expression: in vitro and in vivo effects.

Thirteen consecutive adult patients with primary refractory (n = 5) or relapsed (n = 8) acute lymphoblastic leukemia (ALL) were treated by an induction schedule (FLAG) consisting of Fludarabine (30 mg/sqm/d) plus high dose Cytarabine (HD-ara-C: 2 g/sqm/d) (d 1-5) and G-CSF (from d O to polymorphonuclear recovery). Patients achieving complete remission (CR) were administered a second FLAG course as consolidation and were then submitted to an individualized program of post-remission therapy, depending on the patient's age and performance status. CR was achieved in 8/12 evaluable cases (67%). The median CR duration was 22.5 w. CR attainment was significantly related to the co-expression of lymphoid and myeloid antigens. ALL/My+ patients achieved CR in 6/6 evaluable cases vs. 2/6 for ALL/My+. In vitro 3H ara-C incorporation into cellular DNA resulted significantly increased by Fludarabine (in 7/9 tested cases) and, furthermore, by the association of Fludarabine G-CSF in 5 evaluable ALL/My+ cases; in contrast, no effect of G-CSF addition to Fludarabine was observed in 4 ALL/My. Myelosuppression was observed in all patients: the median time to neutrophils > 0.5 x 10(9)/1 was 16.3 d (range 13-22) and 16.2 d (range 9-29) to platelets > 20 x 10(9)/1. Nonhematological toxicity was minimal. In conclusion, FLAG is an active and tolerable combination in refractory ALL, particularly in cases with myeloid antigen expression where G-CSF appears to improve efficacy, probably increasing ara-C incorporation into the DNA of leukemic cells.