The incidence of acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) increases with age. Intensive chemotherapy is unsuitable for older patients because of its intolerable toxicity. CAG (cytarabine, aclarubicin, and G-CSF) regimen is widely used for such elderly patients because of its relatively low toxicity mainly in Eastern Asia such as China and Japan. Several reports showed the efficacy of CAG, but their sample sizes are relatively small. Thus, we here report a retrospective, single-center analysis of 86 cases with AML and MDS treated with CAG regimen

Patients and methods

Patients of AML and MDS who were treated with CAG regimen at Sapporo Hokuyu hospital from 2008 to 2012 were analyzed retrospectively. To examine the efficacy of CAG, we examined response rate, overall survival (OS), event-free survival (EFS), and rate and duration of being free from transfusion. We also performed subgroup analysis on response rate, OS, and EFS by underlying diseases, chromosomal karyotypes, WBC count and LDH level. Chi-squared test was used to compare the response rate of chemotherapy between subgroups. Overall survival was defined from the time of starting chemotherapy to the time of death from any cause or to the time of last follow-up. Kaplan- Meier method was used to compare survival curves. Cox regression analysis was used to analyze differences between subgroups.


Among 86 patients enrolled, 60 patients were with AML, and 26 were with MDS. Median age of the patients was 71 (range 17-87) years old and 57 (66.3%) patients were male. Forty-six patients had previous history of chemotherapy other than CAG, and seven had previous history of hematopoietic transplantation. Median time from diagnosis to first cycle of CAG was 169 (range 1-6104) days. Eighty-six patients received total of 136 series of CAG regimen. Each series consisted of several courses of CAG, and median number of cycles was two (range 1-6). Median follow-up duration was 276 (range 18-1718) days and 28 patients (32.6%) were alive at the last follow up. After the first series of chemotherapy, 39 patients achieved complete response (CR), 11 patients remained stable disease, and 36 patients experienced progression of disease. Overall response rate was 58.1%. CR was achieved in 30 out of 60 patients with AML (50%) and 8 out of 23 with MDS (34.8%), respectively. No significant difference of response rate was seen between the two groups (p=0.70). While 6 cases of AML and 4 cases of MDS showed partial response, 24 cases of AML and 11 cases of MDS were resistant to CAG, respectively. Half of the patients of AML were diagnosed as AML with MDS-related changes (AML with MRC). No differences of response rate were seen between AML and AML with MRC.

Significant differences of CR rate were seen between three groups according to classification scheme of karyotype reported by SWOG/ECOG. In detail, CR rates were 62.9% in favorable karyotype group, 40.0% in intermediate group and 22.7% in unfavorable group, respectively (p=0.026). Favorable group was also superior on EFS compared to the other two groups (p=0.010). Median duration of OS and EFS were 349 and 457 days in favorable group, 495 and 228 days in intermediate group, 217 and 174 days in unfavorable group, respectively. Patients with normal LDH level survive significantly longer compared to patients with high LDH level (717 days vs. 296 days, p=0.004), although LDH level was not a predictive factor for CR rate (55.8% vs. 34.9%, p=0.084). Low WBC count (<3000/μl) at the time of chemotherapy also predicted longer survival time after treatment (p<0.001). Twenty-five out of 58 patients who were dependent on transfusion at the time of chemotherapy became independent from transfusion. Median duration of free from transfusion was 156 (range 35-483) days. Forty-five patients received two or more series of CAG regimen. CR rate on second or more series of the chemotherapy was 45.7%, and PR or SD was 12.5%, which were comparable to those of prior CAG treatment.


This study showed the efficacy of CAG regimen for patients with AML and MDS who were not candidate for intensive chemotherapy. CAG was first reported more than 10 years ago. But there are no prospective studies comparing this strategy with other low-dose chemotherapy, hypomethylating agents, and supportive care. To determine the place of this useful approach within the many therapeutic options for AML and MDS, prospective clinical trial is needed.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.