Elsevier

Leukemia Research

Volume 27, Issue 3, March 2003, Pages 205-214
Leukemia Research

Salvage therapy in refractory acute myeloid leukemia: prediction of outcome based on analysis of prognostic factors

https://doi.org/10.1016/S0145-2126(02)00089-9Get rights and content

Abstract

Acute myeloid leukemia (AML) non-responsive to initial chemotherapy is generally of poor prognosis. High-dose cytarabine (HD-AraC) has been proposed as salvage therapy in combination with amsacrine. The aim of the current study was first to assess the toxicity and the efficacy of such a combination therapy, and secondly to determine prognostic factors allowing to predict whom patients could benefit of such a treatment.

Out of 91, 45 patients referred to our institution have been treated by HD-AraC (3 g/m2/12 h from day 1 to 4) combined with amsacrine (90 mg/m2 per day from day 5 to 7) as a salvage regimen. Forty-five of the 91 patients (49%, 95% confidence interval (CI): 39–60%) achieved complete remission (CR). Thirty-five patients were refractory to the salvage therapy and 11 patients died from toxicity during aplasia. Median disease-free survival (DFS) was 11.5 months (95% CI: 6–16 months). After CR achievement, 26 patients received consolidation therapy according to the protocol in which they were included. Nineteen patients with an HLA-identical sibling donor underwent allogeneic bone marrow transplantation. At time of analysis, 27 of the 45 patients (60%) who achieved CR have relapsed. Median overall survival (OS) was 7.5 months (95% CI: 6–15 months). There was 12 long survivors (13%). In univariate analysis, initial karyotype was the main prognostic factor as well as in terms of CR achievement (P=0.002) than in terms of DFS (P=0.01) or OS (P=0.009). CR achievement was negatively influenced by higher WHO performance status index (P=0.006), higher LDH level (P=0.02), and higher CD34 expression by leukemic cells (P=0.03) at diagnosis, and presence of circulating blasts (P=0.001), platelet count <80×109/l (P=0.0001), and polymorphonuclear (PMN) percentage <30% (P=0.01) at time of starting salvage therapy. DFS was negatively influenced by secondary AML (P=0.01), weight loss ≥5% (P=0.03), and higher white blood cell (WBC) count (P=0.03) at time of diagnosis. Age over 60 years (P=0.002), prior history of toxic exposure (P=0.01), higher CD34 expression (P=0.02), weight loss ≥5% (P=0.006), and WHO performance status index ≥2 (P=0.01) at diagnosis, and platelet count <80×109/l (P=0.02) at time of salvage therapy were the main prognostic factors associated with shorter OS. In multivariate analysis, karyotype grouping at diagnosis (P=0.006) and blood count before salvage therapy (P=0.001) were of prognostic value for CR achievement. Karyotype remained of prognostic value for DFS and OS (P=0.007 and <0.0001, respectively).

We conclude that HD-AraC combined with amsacrine was as a useful salvage regimen in AML non-responding to a first intensive course of chemotherapy. Using objective parameters of proven significance (karyotypic grouping and blood count before salvage), we devised a prognostic system of immediate clinical utility for prognostic stratification and risk-adapted therapeutic choices. Patients with favorable risk cytogenetics and those with intermediate risk cytogenetics and favorable blood count (PMN ≥30%, no circulating blasts, and platelet count ≥80×109/l) before salvage therapy had a similar outcome than those achieving CR after only one course of chemotherapy. All other patients displayed a poor outcome. This suggests their orientation at an earlier time to alternate therapeutic programs based on investigational drugs.

Introduction

Treatment outcome in young adult patients with newly diagnosed acute myeloid leukemia (AML) has substantially improved over the past decade. Complete remission (CR) rates range from 60 to 80% with long-term survival in about 50% of cases. Advances can be attributed to several factors related to supportive care, which has allowed the safer use of more intensive chemotherapy schedules, but also to improved treatment strategies such as the introduction of high-dose cytarabine (HD-AraC) as consolidation therapy [1], [2]. However, therapeutic failure remains a major concern. Indeed, AML patients who fail to achieve CR after an initial chemotherapy have a very poor prognosis and it is generally accepted that few cures can be achieved at this stage using chemotherapy alone. Failure generally occurred because of adverse biological determinants including multidrug resistance mechanisms [3]. Therapeutic options thus consist in salvage therapy consisting either in a second course identical to the first course of chemotherapy, or in a dose escalation, or a total change of therapy. Investigational regimens could find a place there.

Amsacrine [4′-(9-acridinylamino)methanesulfon-m-anisidine] is an acridine derivative that was originally synthesized by Cain et al. [4] in 1974 and that was found to have a potent antitumor activity against leukemic cells by a mechanism of intercalation. In the early 1980s, phase II studies reported 15–28% of CR achievement with amsacrine monotherapy in a patient population with refractory AML [5], [6], [7]. Several studies proved the interest of regimens using amsacrine in combination with high-dose cytarabine (HD-AraC) in refractory AML [8], [9], [10], [11].

In our study, we reviewed the data from 91 patients with refractory AML who received salvage therapy combining HD-AraC and amsacrine after failure of a first course of induction chemotherapy. Main goals of the study were to analyze the efficacy and the toxicity of this treatment schedule. Over the past years, a considerable amount of effort has been made to identify factors correlating with outcome of treatment. The other goals of this study were therefore to identify prognostic factors in order to predict the outcome of salvage therapy. Such a prediction could be useful for orienting patients earlier to a second induction chemotherapy or their inclusion into phase I or II investigational drug trials.

Section snippets

Patients

Between 1985 and 1994, a total of 403 patients, referred to our institution for newly diagnosed AML, were enrolled onto two successive therapeutic protocols: LYLAM 85 (259 patients of whom 180 patients aged 15–59 and 79 patients aged ≥60) [12] or LAM 90 (144 patients, all aged 15–59) [13]. Overall, 115 patients (29%) did not respond to the first course of chemotherapy. Among them, 24 patients, with altered general condition (WHO performance status >2) or persistence of active severe infection,

Patient characteristics

Main characteristics of the 91 patients are summarized in Table 2. There were 56 males and 35 females with a median age of 44 years (range, 16–75 years). Eleven patients aged 60–70 (12%) and seven patients aged more than 70 (8%). At diagnosis, the incidence of tumoral syndrome was 34%. Only one patient presented with central nervous system involvement. Morphological FAB classification was available in 87 cases. Among them, five patients enrolled into the LYLAM 85 trial displayed M3 subtype.

Discussion

This study raises different questions. First of all, it confirmed the relative efficacy of the combination of amsacrine with high-dose cytarabine in the treatment of refractory AML. Secondly, it points to factors of prognostic value that could be used to predict CR achievement after salvage therapy and therefore to influence treatment strategy after failure of a first course chemotherapy with the aim to move earlier high-risk patients towards alternative drugs or investigational trials.

Acknowledgements

E. Tavernier drafted the paper. Q.H. Le provided statistical expertise. M. Elhamri collected and assembled the data. X. Thomas analyzed the data, provided critical revision of the manuscript and gave final approval.

References (28)

  • J.D Hines et al.

    High-dose cytosine arabinoside and m-AMSA is effective therapy in relapsed acute non-lymphocytic leukemia

    J. Clin. Oncol.

    (1984)
  • Z.A Arlin et al.

    Treatment of acute leukemia in relapse with 4′-(9 acridinylamino)-methansulfon-m-anisidide (AMSA) in combination with cytosine and thioguanine

    Cancer Clin. Trials.

    (1981)
  • Vogler WR, Stein R, Velez-Garcia E, et al. New approaches in the treatment of acute myeloblastic leukemia: results of...
  • E Berman et al.

    Comparative trial of cytarabine and thioguanine in combination with amsacrine and daunorubicin in patients with untreated acute non-lymphocytic leukemia: results of the L-16M protocol

    Leukemia

    (1989)
  • Cited by (0)

    View full text