Probability of remission with reinduction with 7+3 versus high-dose cytarabine: analysis of SWOG trial S1203

Abstract

In patients with acute myeloid leukemia (AML), the goal of intensive induction chemotherapy is to induce a morphologic complete remission (with or without count recovery, CR or CRi respectively), which is strongly correlated with overall survival (OS) [1, 2]. Intensive chemotherapy is defined as the standard 7+3 regimen or regimens using cytarabine at daily doses of ≥1 g/m², (such as FLAG-ida, CLAG-M, or high-dose cytarabine combined with idarubicin known as IA). A common clinical question is whether patients whose disease has not achieved a CR after the first course of induction should receive a second identical course or change to a different, often investigational, regimen [3]. Because results with the latter are by definition unknown, the decision must rest on the expected results with a second course of 7+3 or a high-dose cytarabine-containing regimen. Analyzing SWOG trials S8600, S9031, S9333, and S0106, conducted in the 1980s,1990s, and 2000s, we previously reported a CR rate of 44% among 242 patients given a second course of 7+3 after failure of a first, compared to a CR rate of 48% among the 987 patients who received a first course [4]. Notably, in the most recent of these studies (S0106), though CR was still achieved in a high percentage of patients, survival was considerably longer if CR was observed after the first course [5]. In contrast to the similar rate of CR after the first or second course of 7+3 in SWOG trials, a single-center analysis by Ravandi et al. noted a CR rate of 76% with an initial high-dose cytarabine-containing course, but reported a rate of 32% in 129 patients receiving a second similar course [6]. SWOG Cancer Network trial S1203 randomized patients with untreated AML who were 60 years of age or younger among 7+3, IA, and IA + vorinostat, thereby allowing a comparison of rates of CR and CRi after 1 and 2 cycles of induction therapy in a group of patients who met the same eligibility criteria and underwent randomization. This report analyzes data from patients randomized to the 7+3 and IA arms.

The primary outcome of S1203 has been previously reported [7]. All patients had untreated AML and were \(\le\)60 years of age. Patients randomized to 7+3 (n = 261) received 100 mg/m² cytarabine on days 1–7 by continuous infusion and 90 mg/m² daunorubicin on days 1–3. Patients randomized to IA (n = 261) received 1500 mg/m² cytarabine on days 1–4 by continuous infusion and 12 mg/m² Idarubicin on days 1–3. CR and CRi were defined per contemporary consensus criteria (CR: ANC \(\ge\) 1000/mcl, platelet count \(\ge\)100,000/mcl, <5% bone marrow blasts; CRi same as CR but either ANC or platelet criteria not met). Patients who did not achieve morphologic CR or CRi after the first cycle of induction were eligible for a second cycle. Patients randomized to 7+3 were to have a marrow examination at approximately day 14 and if residual blasts were seen, reinduction was to be initiated. Patients randomized to IA were to have a bone marrow examination at approximately day 28 and if residual blasts were seen, reinduction was to be initiated. Re-induction with 7+3 called for the same dose and schedule of cytarabine with a decrease in the dose of daunorubicin to 45 mg/m² on days 1–3; reinduction with IA was identical to induction in terms of schedule and dosing. The institutional review boards of the participating institutions approved all protocols, and patients were treated according to the Declaration of Helsinki.