The selective use of AMSA following high‐dose cytarabine in patients with acute myeloid leukaemia in relapse: a Leukemia Intergroup Study

Richard A. Larson, Roger S. Day, Nozar Azarnia, John M. Bennett, George Browman, Jack Goldberg, Arlan Gottlieb, Hans Grunwald, Kenneth Miller, Azra Raza, Ralph Vogler, Elliot Winton, Harvey D. Preisler

Research output: Contribution to journalArticlepeer-review

13 Scopus citations


Summary. This clinical trial was designed to evaluate the role of high‐dose cytarabine (ara‐C) in the treatment of adults with acute myeloid leukaemia (AML) in first relapse. We also tested the hypothesis that the selective use of AMSA (100 mg/m2/d on days 7, 8 and 9) would increase the complete remission (CR) rate when leukaemia cells remained in the bone marrow immediately following 6 d of Ara‐C (2–3 g/m2/12 h) alone. Of 155 patients evaluable for response, 115 (74%) experienced marked cytoreduction by day 6 and received no further induction chemotherapy; 53 (45%) of these patients achieved CR after one course and 45 (38%) had resistant disease. The 36 patients (23%) with inadequate cytoreduction after the 6 d of ara‐C alone were randomly assigned either to no further chemotherapy (21 patients) or to 3 d of AMSA (15 patients). The CR rates after one course were 14% and 53%, respectively (P= 0001). and the fractions with resistant disease were 76% and 40%, respectively. The fractional reduction of leukaemia cells in the day 6 bone marrow aspirate specimen (P<0.0001) and the reduction in the leukaemia cell mass measured in the day 6 marrow biopsy (P= 0.001) were the strongest predictors for achieving CR versus having residual disease in univariate analyses. The median duration of remission was 5 months, but seven patients (10%) remain in CR after 30–92+ months. Among the 140 patients who received only the 6 d of ara‐C, the pretreatment albumin (P= 0.002) and lactate dehydrogenase (P= 0.01) levels were the strongest predictors of response in univariate analyses, but only the albumin remained significant (P= 0.01) in a stepwise logistic regression analysis. Those patients with albumin >4.0 mg/dl and LDH <125% of normal had a 71% CR rate, and only 16% had resistant disease. Among patients lacking both of these favourable characteristics, only 38% had a CR and 39% had resistant disease. Thus, pretreatment characteristics and rapid cytoreduction in the day 6 bone marrow sample identified a favourable subset of patients with AML in first relapse, some of whom responded quite well to 6 d of ara‐C alone and have had long disease‐free remissions.

Original languageEnglish
Pages (from-to)337-346
Number of pages10
JournalBritish Journal of Haematology
Issue number2
StatePublished - Oct 1992
Externally publishedYes


Dive into the research topics of 'The selective use of AMSA following high‐dose cytarabine in patients with acute myeloid leukaemia in relapse: a Leukemia Intergroup Study'. Together they form a unique fingerprint.

Cite this