Ten Year Follow up Analysis of the OSHO Phase III Trial (AML 96) Comparing Different Application Modes of AraC in Patients below 60 Years with Acute Myeloid Leukemia (AML): No Impact of AraCApplication Mode on Remission Rate, Toxicity, Disease-Free Survival or Overall Survival

Clinical trials on different cytarabine doses for treatment of AML provide evidence of a dose response effect, but also for increase toxicity after high dose AraC (HDAC). Pharmacokinetic measurements of cytarabine-triphosphate (AraC-CTP), which is the most relevant cytotoxic metabolite of AraC, have revealed its formation in leukemic cells to be saturated with infusion rates above 250 mg/m²/h, this being significantly lower than used in HDAC schedules.

Methods: Based on a pharmacological model and encouraging results of a phase II study we conducted a prospective randomized multicenter clinical trial comparing the effects of two different application modes of AraC in patients up to 60 years with untreated newly diagnosed AML. Patients were randomized to receive AraC at two different infusion rates (IR) during induction and consolidation treatment:

1. arm A/experimental: 1 × 2 g/m²/d AraC over 8 hours (IR 250 mg/m²/h)

2. arm B/standard: 2 × 1 g/m²/d AraC over 3 hours (IR 333 mg/m²/h).

Induction and first consolidation consisted of AraC (days 1, 3, 5, 7) in combination with an anthracycline (Idarubicine 12 mg/m² or Mitoxantrone 10 mg/m², days 1–3). The final dosage points (AraC day 7 and anthracycline day 3) were excluded from the second consolidation. The third consolidation consisted of either allogeneic or autologous stem cell transplantation or of chemotherapy identical to second consolidation.

Results: From 02/97 to 04/02 419 patients were enrolled in the study. The present analysis is based on 361 eligible and evaluable patients with a median follow up of 7 years. CR was reached in 249/361 (69%; 95%CI: 65%–74%) patients. No statistically significant differences were detected between arms A and B with regard to CR-rate (69% vs 69%) or early death rate (11% vs 8%). Hematological recovery of median white blood cell count (WBC) > 10⁹/l and median platelets (plt) > 50 × 10⁹/l revealed no difference between arms A and B after induction (WBC day 22 vs 22, p=0,68; plt day 25 vs 26, p=0,41) and consolidation (WBC day 28 vs 27, p=0,07; plt day 42 vs 40, p= 0,58). The event free survival (EFS) after 5 years is 0,25 ± 0,03 % for all patients with an overall survival of 0,31 ± 0,03 % after 5 years. For the purposes of analysis, the 83 transplant patients (23 allogeneic MRD, 14 allogeneic MUD and 46 autologous) were censored at time of transplant. No statistically significant difference between arms A and B in regard to EFS (0,25 ± 0,04 vs 0,25 ± 0,04, p=0,99), relapse incidence (0,63 ± 0,06 vs 0,60 ± 0,06, p=0,89), overall survival (0,32 ± 0,04 vs 0,30 ± 0,04, p=0,44) and therapy associated mortality (0,18 ± 0,04 vs 0,17 ± 0,03, p=0,95) were detectable after adjustment of prognostic factors. An analysis of risk factors by multivariate cox regression model confirmed cytogenetics at diagnosis to be the most important risk factor for CR rate (p<10⁻⁶) and for EFS (p<10⁻⁶). Other significant prognostic factors for EFS evaluated in the multivariate analysis were de novo vs secondary AML (p=0,0001), WBC (continuous) (p=0,001), LDH (>1–4 × vs other ULN) (p=0,008) and FAB classification (FAB M0,6,7 vs FAB M1,2,4,5) (p=0,0005). EFS after 5 years shows a significant correlation to cytogenetics (p<10⁻⁶) with 0,71±0,1, 0,27±0,05, 0,20±0,06 and 0,03±0,03 for favorable, normal, other and unfavorable cytogenetic karyotype, respectively.

Conclusion: We conclude that the application of AraC at the presumptive saturating infusion rate of 250 mg/m²/h results in comparable remission rates, toxicity, event free survival and overall survival as compared to the standard IR with 333 mg/m²/h.