Azacitidine in Acute Myeloid Leukemia with >30% Bone Marrow Blasts and <15 G/L White Blood Cell Count: Results from the Austrian Azacitidine Registry of the AGMT Study Group Versus Randomized Controlled Phase III Clinical Trial Data

Background

In the Phase III trial (AML-001; NCT01074047) that assessed azacitidine (AZA) vs conventional care regimens (CCR; intensive chemotherapy, low-dose cytarabine or best supportive care as preselected by the treating physician) in older patients (≥65 years) with newly diagnosed acute myeloid leukemia (AML) and >30% bone marrow (BM) blasts, AZA was associated with a clinically meaningful improvement in overall survival (OS) vs CCR.¹ Patient registry data, if performed with adequate quality control, can complement and be of additional value to data generated in clinical trials. The Austrian Azacitidine Registry (AAR; NCT01595295) was initiated to gain a comprehensive view of the use, safety and efficacy of AZA in patients with AML in a ‘real world’ clinical practice setting.^2,3 No formal exclusion criteria existed, as the aim was to include all AML patients treated with AZA, irrespective of age, comorbidities, and/or previous lines of treatment.^2,3

Aims and methods

The aim of this analysis was to assess the efficacy and safety of 1^st-line AZA in patients with AML who were included in the AAR and who fulfilled the BM blast percentage and white blood cell (WBC) count entry criteria of the AML-001 trial (BM blasts >30% and WBC <15G/L). Outcomes of the subgroups of patients with poor-risk cytogenetics and with AML and myelodysplasia-related changes (AML-MRC) who received 1^st-line AZA were also evaluated.

Results

A total of 95 patients were identified that fulfilled the BM blast percentage and WBC count entry criteria of the AML-001 trial (data cut-off June 19 2015). Apart from a higher proportion of AML-MRC and a lower proportion of AML-not otherwise specified (AML-NOS) within the AAR, baseline characteristics were comparable to those observed in AML-001 (i.e. age, BM blast percentage, gender, therapy-related AML, prior myelodysplastic syndrome (MDS), Eastern Cooperative Oncology Group Performance Status (ECOG PS), cytogenetic risk, transfusion dependence, hemoglobin level, WBC count, absolute neutrophil count, and platelet (PLT) count; Figure 1). Patient status at data cut-off, reasons for AZA discontinuation and treatment characteristics were also similar: median number of AZA cycles was 5 (1–51) and 6 (1–28) for the AAR and AML-001 trial, respectively (Figure 1). Patient outcome in terms of overall response according to International Working Group criteria⁴ (31.5 vs 29.0%), red blood cell (42.1 vs 38.5%) and PLT transfusion independence (34.5 vs 40.6%) did not differ significantly between the AAR and the AML-001 trial. Furthermore, median OS was highly concordant between the AAR and AML-001 overall (10.8 vs 10.4 months; Figure 2A) as well as for various patient subgroups: 12.2 vs 12.7 months for patients with AML-MRC (Figure 2B); 14.6 vs 14.1 months for patients with normal cytogenetics; 13.1 vs 13.0 months for patients with intermediate-risk cytogenetics; and 7.2 vs 6.4 months for patients with high-risk cytogenetics (Figure 2C). After 1 year, 47.4% of patients were still alive in the AAR cohort compared with 46.5% in the AML-001 trial (p=0.924). Event-free survival was 5.5 (range: 0–35.3) vs 6.7 (range: 5–8.8) months in the AAR and AML-001 trial, respectively. The 30-day (8.4 vs 6.6%; p=0.642) and 60-day (15.5 vs 16.2%; p=0.903) mortality rates were comparable. The incidence of febrile neutropenia (24.2 vs 28.0%) and Grade 3–4 treatment-emergent (TE) neutropenia were similar between the AAR and AML-001; however, higher rates of TE thrombocytopenia (47.4 vs 15.7%; p=0.023) and anemia (31.6 vs 26.3%; p<0.001) were observed in the AAR.

Conclusion

These data confirm the safety and efficacy of AZA in a patient-population for whom this drug has not yet been approved, i.e. AML with more than 30% BM blasts and without leukocytosis. These data therefore complement prospective clinical trial data and support the role of well-designed and in-depth clinical registries.

References

1.       Dombret H, et al. Blood 2015;126:291–9

2.       Pleyer L, et al. J Hematol Oncol 2013;6:32

3.       Pleyer L, et al. Ann Hematol 2014;93:1825–38

4.       Cheson BD, et al. J Clin Oncol 2003;21:4642–9