Outcome of elderly patients with AML is usually poor. Increased incidence of high risk AML and intolerance against dose escalation are the main causes for treatment failure. Pharmacokinetic measurements indicated that intracellular Ara-CTP formation is saturated at much lower infusion rates than used in high dose AraC schedules, probably causing AraC accumulation in the plasma and increased toxicity. Therefore, the East German Study Group (OSHO) used intermediate doses of AraC delivered at the presumptive saturating infusion rate over a prolonged period of time. The same schedule was applied to younger (≤60 years of age, AML 96) and older (>60 years of age, AML 97) patients with AML. In the present evaluation, determining factors for complete remission rate and hematopoietic reconstitution were identified.

Methods: All 690 patients entered in the AML 96 (n=370) and AML 97 (n=320) study of the OSHO and treated with one or two courses of induction therapy (AraC 2 g/m2 iv on day 1,3,5,7 in combination with idarubicine 12 mg/m2 day 1–3 or mitoxantrone 10 mg/m2 iv day 1 to 3) followed by 2 consolidation courses were evaluated. The following baseline variables were studied in univariate analysis for their impact on CR rate: sex (M vs. F), age (≤ 60 years vs. >60 years; continuous variable), cytogenetics at diagnosis (normal, favourable, unfavourable and others), disease classification (de novo or secondary AML), WBC (<2/2–90/>90 x 109/l), FAB-classification (M0/M1/M2/M3/M4/M5/M6/M7), LDH (≤ 2 x ULN vs. > 2 ULN) and the use of G-CSF (yes or no). Factors significant in this analysis were included in a multivariate model (logistic regression). Hematopoietic reconstitution was defined as the first of two consecutive days with leucocytes >1000/μL and platelets >50000/μL.

Results: As shown in the multivariate analysis, the most important and highly significant parameter for CR rate was cytogenetics at diagnosis (p=10−11) followed by disease classification (de novo or secondary AML; p=0,001), WBC (p=0,003) and sex (p=0,018). In contrast, we could not demonstrate any significant influence of age (p=0,64), G-CSF (p=0,16) and FAB classification (p=0,38) on CR rate. Significant factors for the recovery of leukocytes were the use of G-CSF (p= 10−12), cytogenetics (p=10−4) and disease classification (de novo or secondary AML; p=0,04). AML classification (de novo or secondary AML; p=0,00002) and cytogenetics (p=0,001) but not age (p=0,17) were determinants for platelet recovery.

Conclusion: Cytogenetics at diagnosis and disease type (de novo vs. secondary) were the most important determinants for CR rate using intermediate dose AraC. We could not demonstrate any influence of age on CR rate and hematological recovery.

Author notes

Corresponding author