Too much ara-C? Not enough daunorubicin?

It seems strange, but after 40 years of experience using cytosine arabinoside (ara-C) and daunorubicin, leukemia doctors are still unsure as to how much of each drug they should use to treat their patients with acute myeloid leukemia (AML).

Most would agree that, as initial therapy for patients with AML, ara-C by continuous infusion, in combination with a topoisomerase II inhibitor (eg, daunorubicin, idarubicin, mitoxantrone) is a standard of care around the world. In the postremission setting, administration of repeated cycles of escalated doses of ara-C is widely accepted as the optimal nontransplant treatment strategy. However, despite a farrago of studies, there continues to be debate on 2 points: (1) what is the optimal topoisomerase II inhibitor to use during induction therapy and at what dose, and (2) how much ara-C should be administered after remission and to which patients? In this issue of Blood, the results of a trial conducted by the Japan Acute Leukemia Study Group (JALSG) adds additional complexity to the debate. In a study (published as 2 separate papers) of more than 1000 untreated adults with AML under the age of 65 years, high doses of daunorubicin were compared with standard doses of idarubicin during induction, and high-dose ara-C was compared with a standard dose, ara-C–based multiagent regimen in the postremission setting.^1,2 

In the 1970s daunorubicin at a dose of 45 mg/m² × 3 (cumulative dose 135 mg/m²) combined with infusional ara-C for 7 days (“7 + 3”) was established as an effective regimen in adults with untreated AML.³  Studies conducted in the 1990s and reaffirmed in subsequent later trials, established both idarubicin (12 mg/m² × 3 doses) and mitoxantrone (12 mg/m² × 3 dose) as equivalent, if not, superior, substitutes for daunorubicin.^4,5  More recently, a trial conducted by the Eastern Cooperative Oncology Group (ECOG) demonstrated that doubling the dose of daunorubicin to 90 mg/m² × 3 (cumulative dose 270 mg/m²) significantly improved survival in adults with AML under the age of 60 years.⁶  In the current study by the JALSG, daunorubicin 50 mg/m² × 5 (cumulative dose 250 mg/m²) was found to be equivalent but not superior to idarubicin 12 mg/m² × 3. Importantly, in both the ECOG and JALSG studies, the higher than standard doses of daunorubicin did not appear to increase the risk of infection or cardiomyopathy.

What conclusions can be made? Whether one should use high-dose daunorubicin (250-270 mg/m² cumulative dose) or either idarubicin, mitoxantrone (or even daunorubicin at 60 mg/m² × 3) is unanswered, but one point should be a clear: daunorubicin 45 mg/m² × 3 is no longer the standard of care in AML patients under the age of 65 years.

The rationale for the administration of high doses of ara-C is based on a multitude of clinical and pharmacodynamic studies conducted over the past 3 decades. The oft-cited Cancer and Leukemia Group B (CALGB) trial 8461 demonstrated that, in patients under the age of 60 years, repeated cycles of ara-C administered at a cumulative dose of 18 g/m² led to superior survival compared with lower cumulative doses of 2 g/m² or 500 mg/m².⁷  Further analysis of this trial, as well as others, showed that the benefit of the high-dose arm was observed, for the most part, in patients with a favorable karyotype; t(8;21) and Inv (16)—the core binding factor (CBF) leukemias.⁸  The JALSG study published in this issue reaffirms the effectiveness of high-dose ara-C compared with standard doses in CBF mutated AML patients, but shows a lack of benefit over standard-dose ara-C–based consolidation therapy in the intermediate and unfavorable karyotypic subsets of AML.

Are we giving too much ara-C to patients with AML? The higher incidence of prolonged myelosuppression and serious infections observed in patients treated on the high-dose ara-C arm of the JALSG study should mandate a clearly proven clinical benefit for high doses to justify the risk of increased toxicity. For patients with CBF AML, the benefit of high-dose ara-C has now been demonstrated in multiple studies and this approach should be considered the standard of care. However, do we have to deliver up to 18 g/m² to these patients or can we obtain similar efficacy with a dose somewhere between 2 and 18 g/m²? This is as yet unanswered. Also unanswered is the relevance of C-Kit mutations associated with the CBF AMLs. Do these patients benefit from the higher doses?

For AML patients with intermediate and poor-risk karyotypes, the benefit of postremission high-dose ara-C is less well established. A retrospective analysis by the CALGB showed that in AML patients with a normal karyotype, a significant disease-free, but not overall survival, advantage was observed in the intermediate-dose and high-dose ara-C cohorts compared with lower doses.⁹  The current JALSG study suggests that these patients do just as well with standard-dose regimens. None of the studies have demonstrated an advantage for ara-C dose escalation in poor-risk patients. Thus it can be argued that the routine use of high doses of ara-C may not be justified in patients other than those with a CBF mutation. It is unfortunate that data on NPM1 and FLT-3 mutations are not available from the JALSG study. Retrospective trials have demonstrated that patients with NPM1 mutations, without FLT-3 mutations, have similar outcomes to those of patients with CBF mutations, and it would be helpful to know if high-dose ara-C is an important factor in this outcome.¹⁰