The impact of karyotype on remission rates in adult patients with de novo acute myeloid leukemia receiving high-dose cytarabine-based induction chemotherapy.

One hundred and twenty-eight patients aged 15 years or over (median 34) with de novo acute myeloid leukemia (AML) received 2- or 3-drug induction chemotherapy comprising 5 days of daily high-dose cytarabine (2 g/m2 q12h) and etoposide (100 mg/m2), without (n=62, 1985-90, protocol BF11) or with (n=66, 1990-97, protocol BF12) daily 5 mg/m2 anthracycline (61 idarubicin, 5 mitoxantrone). Twelve patients with t(15;17) were not included. Evaluable karyotypes were obtained in 110 (86%): 30 (27%) favorable, 60 (55%) intermediate, and 20 (18%) adverse. Three patients dying during chemotherapy were inevaluable. Eighty-four (67%) patients remitted with one cycle, and the overall complete remission (CR) rate was 72%. CR rates were comparable for patients with and without evaluable karyotypes. CR rates with BF11 (64% after one cycle; 72% overall) and BF12 (70% after one cycle; 72% overall) were comparable (P=.4 and 1.0 respectively). CR rates after one cycle (86%, 61% and 55%; P=.03) as well as overall CR rates (90%, 69% and 55%; P=.02) were significantly different for patients with favorable, intermediate and adverse karyotypes respectively. In Cox analysis, the karyotype was the only factor found to influence CR independently. We conclude that the karyotype of the leukemic clone is the most important determinant of response to high-dose cytarabine-based induction chemotherapy in AML. The addition of idarubicin to high-dose cytarabine and etoposide does not appear to improve CR rates. A different treatment strategy may be needed to improve CR rates for patients with non-favorable karyotypes.