Azacitidine (AZA) has emerged as the standard of care for treatment of higher risk myelodysplastic syndromes (MDS) based upon results of several studies. However, the rate and duration of response are insufficient and the focus of research will shift towards further optimization of clinical outcome. Studies for a dosing schedule and a combination therapy with other drugs are important. In general, there are some difficulties to fix the dosing schedule of AZA due to its side effects (infection etc.) and complications. We had carried out the strict managements for protection of infection and bleeding to keep a fixed dosing schedule of 75mg/m2/day for 5 days, followed by 21 days drug-off.


Data from consecutive 62 patients with higher risk MDS treated with AZA at our institute between May 2011 and April 2013 were reviewed. AZA was strictly administered at a dosage of 75mg/m2/day intravenously for 5 days, followed by 21 days drug-off, every 26 days. Patients, whose neutropenia (≤500/mm3) were sustained, were hospitalized in clean rooms. Prophylaxis of antibiotics and the administration of G-CSF were carried out if necessary. When patients in poor risk cytogenetics achieved complete remission (CR) by AZA alone, they were subsequently received a sequential combination therapy with 5-day AZA, followed by 21 days drug-off and chemotherapy (including idarubicine or aclarubicine), followed by 35 days drug-off to maintain CR.


A total of 62 patients were classified as RA 2, RAEB-1 14, RAEB-2 32, and RAEB-T 14. There were 40 male patients and 22 female. Median age was 73 years (range, 40-90). At the time of AZA treatment, 40 patients were stratified as Int-2 and 22 as High by IPSS. Sixteen patients had good risk cytogenetics, 17 intermediate, and 29 poor. Fourteen patients had 20-29% of bone marrow blasts, 32 10-19% , and 16 2-9%. Forty-seven patients were de novo MDS and 15 therapy-ralated MDS (t-MDS). Forty-five patients had the time from MDS diagnosis to AZA treatment within two month, 4 2-12 months, and 13 13-102 months.

The median time of follow up was 11 months (3-27) and the median cycle of a 5-day AZA treatment was 8 (2-29). Totally, 537 cycles had been done. Days of drug-off within 21 days/cycle were 435 cycles(81%), 22-28 days/cycle 82 cycles (15%), and over 29 days/cycle 20 cycles (4%).

The IWG2006 responses to AZA included 29 CR (47%), 10 marrow CR (16%), 9 stable disease (SD) with hematological improvement (HI) (15%), 6 SD without HI (10%), 3 progressive disease (PD) (5%), and 1 failure (2%). The overall response rate (ORR) was 78%. HI includes 35 HI-E (57%), 37 HI-P (59%), and 26 HI-N (42%).

The CR/mCR rate in Int-2 was 23/40 (58%), High 16/22 (73%), good risk cytogenetics 11/16 (69%), intermediate 9/17 (53%), poor 19/29 (66%), 20-29% bone marrow blasts at initial diagnosis 9/14 (64%), 10-19% 21/32 (66%), and ≤9% 9/16 (56%). About the time to AZA from MDS diagnosis, the CR/mCR rates were 29/45 (64%) in 0-2 months and 10/17 (59%) in 3-102 months. Among t- MDS, 8/15 (53%) achieved to CR/mCR and among de novo MDS, 31/47 (66%) achieved to CR/mCR. Cytogenetic responses, excluding 9 normal karyotype patients, were 15/28 (54%) CR and 12/28 (42%) PR.

Among 62 patients, 10 patients died and 12 received bone marrow transplantation (BTM) including 7 CR, 2 SD, and 3 PD. The median time to death from the initiation of AZA treatment was 11 months (3-23). The causes of death were 3 BMT-related, 3 PD, and 4 infection. Eight patients, including 4 CR and 4 SD, progressed to AML. They were immediately treated with chemotherapy, and 7 are still alive. Ten patients in poor risk cytogenetics, who achieved CR by AZA alone, subsequently switched to a sequential combination therapy with AZA and chemotherapy from March 2013. No relapse have been seen at present and they have been followed up.


The AZA regimen fixed dosing schedule of 75mg/m2 for 5 days, followed by 21 days drug-off showed promising efficacy for higher risk MDS (ORR; 78%). Based upon our results, the response rates of AZA against Int-2 vs High and de novo MDS vs t-MDS were almost the same. When patients with poor risk cytogenetics achieved CR by AZA treatment alone, the following sequential combination therapy with AZA and chemotherapy was effective to maintain CR.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract