Abstract 4148


5-azacytidine (AZA) is an hypomethylating drug. The international multicenter AZA-001 trial established that AZA significantly improves overall survival (OS) in patients with high risk myelodysplastic syndromes (MDS) compared with conventional care regimens (Fenaux, Lancet Oncol 2009). Some recent reports have raised the question of a possible efficacy of AZA in selected patients with acute myeloid leukaemia (AML). In this study we retrospectively analysed the safety and efficacy of a 7 days-schedule of AZA alone or in combination with an HDAC inhibitor, Valproic acid (VA) and with All-trans retinoic acid (ATRA) in patients with newly-diagnosed and refractory/relapsed AML not eligible for intensive chemotherapy.

Patients and Methods

A monocentric retrospective study from October, 2006 until March, 2009 analysed 29 patients with AML. Among these patients. There were 11 males and 18 females, median age 70,8 years (range 51,2-84,1), AML de novo in 15 patients (3 relapse) and secondary in 14 patients (2 post MPD and 12 post MDS). Median WBC count was 2,5 (range 0,7-140).109/L, 4 patients had WBC more than 10.109/L. The median rate of bone marrow blasts is 30%. 12/27 (44%) patients and 15/26 (56%) have respectively an intermediate and poor risk caryotype. Fifteen (54%) were newly-diagnosed patients, 14 (46%) were refractory/relapsed patients. Median co morbidity index (Sorror, J Clin Oncol 2007) of patients is 2 (0-7).

Patients received daily AZA 75mg/m2 J1-J7, ± VA 35 to 50 mg/kg J1-J7 and ATRA 45mg/m2 J8-J28 every 4 weeks.


5 azacytidine was used alone for 6/29 (21%) patients and in combination with VA and ATRA for 23/29 (79%) patients. Compliance to the planned therapy was good. Average number of AZA administration was 6 days. To date a total 150 treatment-cycles with a median of 5 cycles/patient were applied (1-14). Treatment was well tolerated. Neutropenia grade3III and thrombopenia grade3III occurred respectively in 26/150 cycles (17%) and in 31/150 (20%). Infections grade3III were observed in 14/150 cycles (9,3 %).

Overall response was 62% (17/29): 9 complete response (CR=31%), 3 partial response (PR=10%), 5 haematological improvement (HI=21%), There were 2 stable diseases (SD=7%). 28% of responses were obtained after 1 cycle, 56% after 3 and 89% after 4. Median overall survival (OS) was 13,2 months (0.3-26). We did not observe any significant difference on OS regarding: age, cytogenetics, de novo vs secondary AML, newly diagnosed vs refractory/relapsed patients. OS for patients with SD was similar to patients with CR, PR or HI. WBC >10.109/L before treatment was not correlated with a shorter survival (7.73 months vs 13.2 months p=0,6). Correlation was found between OS and clearance of the creatinine (p=0.005).

In conclusion, AZA based regimens seems well tolerated and an effective treatment in AML, with an overall response of 62% and an OS of 13,2 months. A minimum of 4 cycles of treatment is necessary to evaluate the efficacy. OS of patients achieving CR, PR or HI is not significantly different of those with SD. Treatment should be continued until progression of the disease.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.