High Dose (HD-AraC) vs Standard Dose Cytosine Arabinoside (SD-AraC) during Induction in Acute Myelogenous Leukemia (AML): Impact on Stem Cell Mobilization after Consolidation and on Autologous Transplantation (Second Report of the EORTC-Leukemia Group (LG) - GIMEMA AML-12 Trial.

The AML-12 randomized phase III trial of EORTC-LG and GIMEMA assessed the efficacy and toxicity of HD-AraC (3 g/m²/12 hrs for 4 days) with daunorubicin (50 mg/sqm for 3 days) and etoposide (50 mg/sqm for 5 days) vs SD-AraC (100 mg/sqm for 10 days) with the same drugs. Patients (pts) in complete remission (CR) received consolidation (Co) consisting of AraC (500 mg/sqm/12 hrs for 6 days) and daunorubicin. Subsequently an allogeneic (allo-SCT) or autologous stem cell transplantation (auto-SCT) was planned according to donor availability and age. A 2nd randomization was performed after Co in pts without a donor: auto-SCT followed or not by low dose IL-2. From 9 1999 till 7 2006, 1675 AML pts (APL excluded), age <61 years, from 67 centers (23 EORTC-LG and 44 GIMEMA) entered. Currently 1571 pts have been randomized for induction and 429 pts post-Co. During the induction toxicity was similar in the 2 arms except for conjunctivitis: 6% (HD-AraC) vs 0% (SD-AraC). HD-AraC given in the induction had no impact on the organ toxicity during Co but platelet recovery (> 50x10⁹/l) was longer (median 4.0 vs 3.3 weeks; P=0.01). Among 886 pts randomized until 7 2005 by EORTC centers and 6 large GIMEMA centers, median follow-up of 2.5 years, 815 were evaluable for response. Out of 643 pts who reached CR, 57 went off study (toxicity, early progression). Among the remaining 586 who received Co, 37 could not be evaluated (early death/relapse, too early) and 549 were still CR after Co: 297 pts had no donor/no sibling, 197 had a donor and 55 were not typed. In these 3 groups the present estimates of the SCT rates are: 63% (auto-SCT), 71% (allo-SCT) and 69% (auto-SCT), resp. The 2.5-yr DFS rates (SE%) were 45% (3%), 61% (4%), and 67% (7%), resp. In pts < 50 yrs, 216 pts had no donor/no sibling, 135 had a donor and 14 have not been typed. For the first 2 groups, the 2.5 yr DFS rates (SE%) were 50% (4%) vs 68% (4.5%), hazard ratio=0.64, 95% CI (0.44, 0.93), P=0.02. In pts without a donor/a sibling successful mobilization of blood stem cells (b-SC) after Co was in HD-Ara-C vs SD-Ara-C arm 53 vs 69%, of failure/postponement 37 vs 24%, and other 9.5 vs 7%. The rate of auto-SCT was similar (65 vs 64%), but harvest of BM cells was more often required in the HD-AraC group (15 vs 4.5%). Pts with an insufficient/delayed b-SC harvest had a longer (P<0.001) platelet recovery (> 50x10⁹/l) after Co than those with a successful harvest: median = 6.7 vs 3.3 wks. Among 393 pts with information on cytogenetics, 14% had good risk, 50% normal, 23% other and 13% poor risk (-5/5q-, -7/7q-, complex). The 2.5-year EFS (no CR, time to relapse or death in CR) rates (SE%) were 68% (7%), 42% (4%), 32% (5%) and 14% (5%), resp. So far: toxicity of HD-Ara-C was acceptable; in those who received HD-AraC in induction platelet recovery after Co was longer and the rate of successful b-SC collection was lower; SCT rates are high and similar in the 2 randomized arms; pts <50 yrs with a donor do have a longer DFS; pts with poor risk cytogenetics continue to have a poor prognosis.