Intensive chemotherapy (IC) is associated with significant mortality and morbidity in elderly with acute myeloid leukemia (AML); particularly in patients aged ≥ 70 years, it is currently debated whether or not to deliver IC. Recent findings suggest that hypomethylating agents have encouraging clinical and biological activity in AML, with limited toxicity. Based on this background we wanted to evaluate retrospectively: 1) the efficacy of azacitidine (AZA) in elderly patients with untreated AML, diagnosed according to WHO criteria; 2) the impact of AZA on overall survival (OS) of this category of patients. For comparative purposes a historical group of matched patients who received IC was also analyzed. The AZA group included 57 patients, median age was 76 yrs (range 63–88) with 91% (52) of the patients being 70 yrs old or more, 32 were males and 25 females. Forty-two (78%) had a white blood cell count (WBCc) < 10×109/L, 17 (30%) a secondary AML supervening after an antecedent hematological disorder and 28 (49%) a bone marrow (BM) blast count < 30%. Karyotype was evaluable in 38 (67%) patients and, according to the refined Medical Research Council criteria, 27 (47%) had intermediate-risk abnormalities (25 normal karyotype) and 11 (20%) an unfavorable risk karyotype. A diagnosis of AML according to WHO classification, absence of uncontrolled infections, adequate renal and hepatic function, life expectancy longer than 4 months were the criteria to receive AZA. IC historical control group consisted in 83 patients enrolled sequentially in the AML13 and AML17 EORTC/GIMEMA protocols between 1995 and 2008. In these prospective trials patients aged 61–80 years received an induction course consisting in mitoxantrone, cytarabine (ARA-C), and etoposide (ICE) followed by 2 consolidation cycles including idarubicin, ARA-C, and etoposide (mini-ICE). In AML17 trial, the patients were also randomized to receive or not gemtuzumab ozogamicin, during induction and consolidation courses. Median age was 67 yrs (range 61–78) with 34% (28) of the patients being 70 years old or more, 47 were males and 36 females; 43 (52%) had a WBCc < 10×109/L and 79 (95%) a BM blast count ≥ 30%. Karyotype was evaluable in 65 (78%) patients: 52 (63%) belonged to the intermediate category (43 normal karyotype), 2 (2%) and 11 (13%) carried a favorable and unfavorable risk karyotype, respectively. Thirty-nine patients (68%) received AZA subcutaneously at the conventional dose of 75 mg/m2, the remaining 18 at a flat dose of 100 mg daily; all patients were given a schedule of 7 consecutive days per month. The median number of cycles delivered was 6 (range 1–39) and 77% of the patients received at least 4 cycles of therapy. Median follow up was 244 days (range 30–1281). Overall response rate, estimated according to the revised recommendation of the International Working Group in AML, was 17% and consisted in 11 (8%) complete remission (CR), 4 (3%) CR with incomplete blood count recovery (CRi) and 9 (6%) partial remission (PR). The chance of obtaining a response to AZA was significantly associated with a WBCc < 10×109/L (p=0.002). The median duration of response was 183 days (range 60–1067). Projected 2-years OS for patients on AZA treatment vs IC was 14% and 38%, respectively (p=0.008). Since the median age of the AZA group was 76 years, we broke down the analysis focusing on patients aged 70 years or more. By doing so, we observed that the projected 2-years OS for patients receiving AZA vs those receiving IC was 12% and 28%, respectively (p=0.11). Finally, among the category of patients aged ≥ 70 yrs, we extrapolated those who achieved a response with AZA (22 patients) or IC (22 patients); for these individuals, 2-years OS rate was 25% and 36%, respectively (p=0.46). In conclusion, our retrospective analysis suggests that in AML patients aged ≥ 70 yrs and with a WBCc < 10×109/L, AZA is as effective as IC and could be considered a valid therapeutic option. In this context, the manageable and limited toxicity of AZA will also allow for the quality of life objective to be accomplished. Patients aged < 70 yrs or those who, whatever the age, have a WBCc ≥ 10×109/L are likely to benefit more from IC than AZA. Controlled, randomized, clinical trials are warranted to further explore this matter and confirm our conclusions.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.