Long-term results following treatment of newly-diagnosed acute myelogenous leukemia with continuous-infusion high-dose cytosine arabinoside.

The long-term results in 130 patients with newly diagnosed acute myelogenous leukemia treated with continuous infusion high-dose ara-C (1.5 gm/m2/day x 4 days, CIHDAC) were compared with those in 264 patients treated in previous studies with standard dose ara-C (70-90 mg/m2/d x 7 days) plus either adriamycin (Ad-OAP), or amsacrine (AMSA-OAP). All patients have been followed at least 5 years. Patients in first CR at 5 years (FCR5) treated on protocols prior to CIHDAC had only 5% chance of relapse (median subsequent follow-up of 9 years). Therefore, we considered patients in FCR5 potentially cured. The two groups were similar with respect to known prognostic factors and CR rates. Although remission duration and survival were shorter with CIHDAC than Ad-OAP/AMSA-OAP, the percent of patients potentially cured was similar (10 vs. 15%). Marked differences between regimens were seen in inv(16) and t(15;17) patients. CIHDAC was better for patients with inv(16) with more patients in FCR5 (80 vs. 38%), longer remission duration and survival, and lower incidence of CNS relapse (0 vs. 43%). The Ad-OAP/AMSA-OAP protocols were superior in patients with t(15;17). We also measured steady-state ara-CTP concentrations (ara-CTPss) in 54 CIHDAC-treated patients presenting with high-blast count. While there was no correlation between ara-CTPss and response duration, all five patients in FCR5 in whom ara-CTPss was measured had high concentrations. These data support the concept that patients with AML should be treated differently according to cytogenetics. Inv(16) patients should be treated with high-dose ara-C while t(15;17) should rely more on anthracycline exposure.