Rethinking Cytarabine Dosing in Consolidation Therapy for Patients with AML

Over 30 years ago, high-dose cytarabine (HDAC) at 3 g/m² administered over 2 hours every 12 hours on days 1, 3, and 5 was established as a successful postinduction therapy for patients with acute myeloid leukemia (AML) that resulted in significant benefit in overall survival (OS) compared with lower doses (i.e., 100 or 400 mg/ m² over 2 hours every 12 hours on days 1, 3, and 5) of cytarabine.¹ Since then, more nuanced dose–response relationships have been explored, with induction therapy regimens comparing intermediate-dose cytarabine (IDAC) with HDAC establishing IDAC as an equally efficacious therapy with reduced myelosuppression and toxicities.² Until now, no prospective clinical trials have been designed and/or powered to directly compare the efficacy of single-agent IDAC with HDAC in postinduction AML therapy.

In this issue of NEJM Evidence, Hunault and colleagues report the results of a multicenter phase II/III trial (NCT02416388) comparing IDAC (1500 mg/m² administered over 2 hours every 12 hours on days 1, 3, and 5) as postinduction therapy in 1132 randomly assigned patients 18 to 60 years of age with newly diagnosed AML and lacking a core binding factor translocation who had received induction chemotherapy with cytarabine (200 mg/m² by continuous infusion, days 1 to 7) and either daunorubicin (90 mg/m² over 3 hours every 24 hours, days 1 to 3) or idarubicin (9 mg/m² over 15 minutes every 24 hours, days 1 to 5).³ The primary objective was to evaluate the noninferiority of IDAC to HDAC in OS, with key secondary objectives being event-free survival, relapse-free survival, allogeneic transplant rates, cumulative incidences of relapse, death in first complete remission (CR)/complete remission with incomplete recovery (CRi), and salvage efficacy rates. Safety and toxicity, as well as European LeukemiaNet (ELN)–2022 risk groups OS, were also compared between IDAC and HDAC.⁴

For the primary outcome at 5 years, OS was 59.3% in the IDAC group versus 57.5% in the HDAC group, which met the criteria for noninferiority (P=0.0042). Secondary end points, including event-free survival, relapse-free survival, allogeneic transplant rates, cumulative incidences of relapse, and death in first CR/CRi, did not seem to differ between IDAC and HDAC. Interestingly, among the 239 patients in the cohort who were refractory to induction chemotherapy, 38.5% and 51.6% achieved CR/CRi after the first IDAC or HDAC cycle, respectively. Although these data may be used to support a salvage efficacy advantage of HDAC over IDAC, only a modest OS benefit was seen in the HDAC group compared with the IDAC group in this setting, i.e., 38.0% (95% CI, 29.4 to 46.6) versus 31.1% (95% CI, 22.9 to 39.7), a difference in which the 95% confidence intervals substantially overlapped.

ELN-2022 favorable-, intermediate-, and adverse-risk groups showed no apparent difference in OS between IDAC and HDAC groups. Surprisingly, though perhaps due to chance, the only subgroup in which IDAC appeared to be superior compared with HDAC for OS was in the ELN-2022 adverse-risk group (44.9% vs. 40.2%). These findings, in conjunction with the apparent superiority of HDAC in the refractory to induction setting, another high-risk feature, are interesting juxtapositions that do not provide strong support for preferential use of IDAC or HDAC in the adverse-risk group.

Finally, the severity of chemotherapy-induced myelosuppression and the incidences of related adverse events were lower after IDAC, particularly in those 50 years of age or older. IDAC’s improved safety and toxicity profile compared with HDAC, along with its similar OS, event-free survival, and relapse-free survival efficacy in the cohort studied, supports the authors’ claim that lowering cytarabine dosage to intermediate levels preserves antileukemic efficacy while minimizing toxicity. The reduction in myelosuppression with IDAC compared with HDAC may allow for combinatorial consolidative regimens with agents, including venetoclax, which have shown promising results in ongoing phase I/II trials.⁵

This trial provides the foundation for some additional avenues of exploration that could answer the remaining questions regarding IDAC versus HDAC in the consolidative setting. The trial design only randomly assigned patients to IDAC or HDAC if they had received induction therapy with cytarabine and daunorubicin or idarubicin. This allowed for a “clean” comparison of IDAC versus HDAC, but whether these findings apply to alternative induction chemotherapy regimens, including fludarabine, cytarabine, granulocyte-colony stimulating factor, and idarubicin regimens or nonintensive induction regimens, remains unexplored.

This trial only included patients 18 to 60 years of age, but the vast majority of patients diagnosed with AML are 65 years of age or older, with a median age of AML diagnosis of 68 in the United States.⁶ Examining if IDAC is equal to or more efficacious than HDAC in an older patient population, i.e., those 61 to 75 years of age, who can tolerate intensive induction chemotherapy, is an important area for exploration. IDAC would most certainly be safer in this population, as inferred from the trial’s data showing more adverse events with HDAC, particularly in patients over 50 years of age.

Ultimately, when considering cytarabine dosing in the consolidative setting, this trial confirms that “less, (but not too much less,) is more” and provides a sensical dosing strategy for cytarabine.⁷