Azacitidine in Acute Myeloid Leukemia: Comparison of Patients with AML-MRF Vs AML-NOS Enrolled in the Austrian Azacitidine Registry

Background

Few data exist reporting the incidence of World Health Organization (WHO)-defined acute myeloid leukemia (AML) subgroups in adults in Europe,¹ but the incidence of AML with myelodysplastic-related features (MRF) may be higher than initially reported.^2–4 Whether the presence of multilineage dysplasia has an independent prognostic impact in AML is still controversial.^5,6 However, AML patients with preceding myelodysplastic syndrome (MDS)/myeloproliferative neoplasm and/or MDS-related cytogenetics (Medical Research Council [MRC]) have been shown to have poorer survival than AML-not otherwise specified (NOS).⁷ Previous studies of AML patients treated with azacitidine (AZA) included AML-MRF patients, but did not report on outcomes separately.^{3,4, 8–12}

Methods

Due to the lack of data assessing the prognostic impact of AML-MRF in elderly AML patients treated with AZA, we analyzed patients with AML-MRF from the Austrian AZA Registry, which was initiated to gain a comprehensive view of the safety and efficacy of AZA in ‘real-life’ patients. Similar to the approach taken by others who assessed the effect of AML-MRF irrespective of treatment modality,⁷ patients with AML and recurrent cytogenetic abnormalities (RCA), and treatment-related AML (tAML) based on the WHO 2008 classification¹³ were excluded, as these subsets have particularly good (AML-RCA) or dismal prognosis (tAML), respectively.

Results

The AML-MRF group comprised 217 patients (AZA 1^st line, n=121; AZA ≥2^nd line, n=96) and the AML-NOS group comprised 90 patients (AZA 1^st line, n=33; AZA ≥2^nd line, n=57). Baseline characteristics were comparable except AML-MRF patients had worse Hematopoietic Cell Transplant Comorbidity Index (HCT-CI) scores and a higher proportion of high-risk cytogenetics (Figure 1). The latter is expected as most high-risk cytogenetic abnormalities are defined as MDS-related.¹³ Median time from diagnosis to AZA start was <1 month for AZA 1^st line and >6 months for AZA ≥2^nd line.

The median number of AZA cycles was 4 (range 1–35) for AML-MRF patients and 4.5 (range 1–46) for AML-NOS patients.

The overall response rate (ORR)¹⁴ was similar for AML-MRF vs AML-NOS patients (complete response [CR] + CR with incomplete blood count recovery [CRi] + partial response [PR]: 29.0 vs 33.3%, respectively; p=0.586). Rates of hematologic improvement¹⁵ were similar for AML-MRF vs AML-NOS patients (57.9 vs 57.4%; p=0.964). Median duration of response was 7.0 vs 5.5 mo, respectively. ORR was similar for AML-MRF and AML-NOS patients irrespective of AZA treatment line (Figure 1).

Median overall survival (OS) for AML-MRF vs AML-NOS patients was 9.4 vs 9.7 mo for the total cohort (p=0.490; Figure 2), and 13.1 vs 10.8 mo for patients treated with AZA 1^st line. For responding patients, median OS, response duration and relapse-free survival were numerically longer for AML-MRF than AML-NOS patients within the total cohort (17.1 vs 12.6; 7.0 vs 5.5; and 9.8 vs 8.5 mo), and even more so in the AZA 1^st line cohort (19.7 vs 12.6; 7.4 vs 3.9; and 10.5 vs 7.9 mo). In univariate analyses, baseline factors that significantly negatively impacted OS in AML-MRF patients treated with AZA were peripheral blood blasts >0% (p=0.025), Eastern Cooperative Oncology Group Performance Status (ECOG PS) ≥2 (p=0.004), >3 comorbidities (p=0.004), and poor cytogenetics (p=0.001). Notably, none of these factors showed statistical significance for AML-NOS patients.

Conclusions

This represents the first and largest report comparing outcomes of AML-MRF vs AML-NOS patients treated with AZA. Outcomes of AML-MRF patients were similar to patients with other AML-NOS subgroups. AZA seems a feasible treatment option for these patients despite the reported poor prognostic impact of MRF.

1. Sant M, et al. Blood 2010;116:3724–34

2. Quintas-Cardama A, et al. Blood 2012;120:4840–5

3. Pleyer L, et al. Ann Hematol 2014 [Epub ahead of print]

4. Thepot S, et al. Am J Hematol 2014;89:410–6

5. Wandt H, et al. Blood 2008;111:1855–61

6. Gahn B, et al. Leukemia 1996;10:946–51

7. Miesner M, et al. Blood 2010;116:2742–51

8. van der Helm L, et al. Leuk Res 2013;37:877–82

9. Gavillet M, et al. Haematologica 2012;97:1929–31

10. Maurillo L, et al. Cancer 2012;118:1014–22

11. Ivanoff S, et al. Am J Hematol 2013;88:601–5

12. Al-Ali HK, et al. Leuk Lymphoma 2011;53:110–7

13. Swerdlow SH, et al. IARC press 2008

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

15. Cheson BD, et al. Blood 2006;108:419–25

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal

Figure 2

View largeDownload slide

Figure 2

View largeDownload slide

Close modal