RESUMO
The efficacy of azacitidine in the treatment of high-risk myelodysplastic syndromes (MDS), chronic myelomonocytic leukaemia (CMML) and acute myeloid leukaemia (AML) (20-30% blasts) has been demonstrated. To investigate the efficacy of azacitidine in daily clinical practice and to identify predictors for response, we analysed a cohort of 90 MDS, CMML and AML patients who have been treated in a Dutch compassionate named patient programme. Patients received azacitidine for a median of five cycles (range 1-19). The overall response rate (complete/partial/haematological improvement) was 57% in low risk MDS, 53% in high risk MDS, 50% in CMML, and 39% in AML patients. Median overall survival (OS) was 13·0 (9·8-16·2) months. Multivariate analysis confirmed circulating blasts [Hazard Ratio (HR) 0·48, 95% confidence interval (CI) 0·24-0·99; P = 0·05] and poor risk cytogenetics (HR 0·45, 95% CI 0·22-0·91; P = 0·03) as independent predictors for OS. Interestingly, this analysis also identified platelet doubling after the first cycle of azacitidine as a simple and independent positive predictor for OS (HR 5·4, 95% CI 0·73-39·9; P = 0·10). In conclusion, routine administration of azacitidine to patients with variable risk groups of MDS, CMML and AML is feasible, and subgroups with distinct efficacy of azacitidine treatment can be identified.