5-Azacitidine Induces Hematologic Responses in a High Proportion of Patients with Acute Myeloid Leukaemia Refractory to or Not Eligible for Intensive Chemotherapy.

DNA hypermethylation leads to silencing of gene expression. An association between methylation of the p15 ink4b gene promotor and risk for AML transformation in MDS has been suggested. The DNA hypomethylating pyrimidine analogue 5-azacitidine may reduce hypermethylation and induce re-expression of key tumor suppressor genes in addition to its cytotoxic activity. Azacitidine induces responses in 60% of patients (pts) with MDS. Little is known about the clinical activity of azacitidine in AML patients refractory to or not eligible for chemotherapy. The safety and efficacy of azacitidine in AML pts is assessed at the University of Leipzig.

Patients and methods: Since August, 2005, 17 pts with AML (9 m/8 f), median age 68 (range 58–80) years received 75 mg/m² azacitidine sc for 5 days every 28 days mostly in an outpatient setting. Secondary AML was present in 13 (76%) pts. Abnormal cytogenetics were found in 7 (41%) pts [complex aberrations, n=5, trisomy 8, n=1, deletion of the long arm of chromosome 5, n=1]. Azacitidine was given as first-line therapy in 8 (47%) pts not eligible for chemotherapy or because of relapse or refractory disease after standard chemotherapy, n=9 (53%). Prior to treatment, median WBC was 3.9 (range 0.6–26.4)x10⁹/L. Transfusion-dependent anemia and thrombocytopenia were present in 16 (94%), and 11 (65%) pts respectively. Hematologic Response was assessed according to the International Working Group Criteria for AML. For non-hematologic toxicity, the National Cancer Institute Common Toxicity Criteria and for hematologic toxicity the NCI CTC hematology criteria for leukemia studies or bone marrow infiltrative/myelophthisic processes were used.

Results: A total of 71 treatment cycles were applied. Azacitidine was well tolerated with few non-hematologic side effects (skin irritation at site of injection grade I, n=7, constipation grade I, n=6, transient elevation of SGPT/SGOT grade I, n=5, parasthesia grade I, n=2, creatinin elevation grade I, n=2). One reversible grade IV liver toxicity was observed. Treatment related thrombocytopenia grade 3 and leukopenia grade 3 occurred in 6 (35%) and 3 (18%) pts respectively. ANC and platelet nadirs occurred at days 21 (range 5–23 days) and 14 (range 10–27) days respectively. 6 (35%) pts were refractory to treatment and died within a median of 14 (range 4–29) weeks. After a median follow-up of 33 (range 6–48 weeks), response was observed in 11 (65%) pts after a median of 3 (range 1–8) treatment cycles/patient [CR, n=6 (55%), PR, n=2 (18%), hematologic improvement, n=2 (18%), stable disease, n=1 (9%)]. Two (29%) complete cytogenetic remissions were observed. Responses occurred after a median of 38 (range 17–114) days. Median response duration is 18 weeks (range 15 weeks-not reached).Up till now, 4 (36%) pts lost their responses. Patients refractory to conventional chemotherapy tended to do worse than pts who received azacitidine as first line treatment.

Conclusions: Azacitidine in pts with AML is well tolerated. Five days treatment cycles every 28 days in an outpatient setting are feasible and could induce complete and partial remissions. Longer follow-up is needed to assess the duration of responses in AML pts. Azacitidine may be a promising new treatment option for pts with AML. Larger studies are currently planned to confirm these data, define the number of treatment cycles required and clarify whether a leukaemia-free survival correlates with an improved overall survival.