Treatment with Azacitidine: A Retrospective Study Evaluating the Real Life Clinical Course and Impact on Infectious Complications

Introduction: Demethylating agents such as azacitidine (AZA) and decitabine have become available therapies for higher risk MDS and low blast count (20-30%) AML. Compared to intensive chemotherapy and best supportive care both agents have been found to improve overall survival rates (OS). Yet, infectious complications can impede AZA or decitabine therapy. Since most of the patients are managed on an outpatient basis with often low level of clinical and microbiological documentation, the impact of the hematological response on demethylating agents and the consequences of following infectious complications often are not well documented. Therefore, the aim of this study was to evaluate the real life incidence, character, time of occurrence and clinical role of infectious complications of MDS patients treated with AZA at our institute and to correlate findings with hematological response. We especially focused on the amount of resulting days in hospital, mortality rates and the extent of antimicrobial therapies.

Methods: We retrospectively evaluated the outcome of 77 patients with MDS treated with AZA from 08/2004 to 01/2015. All patients received AZA cycles at standard dosage (75mg/m² for 7 days every 28 days). Clinical and microbiological work-up included parameters like severity and kind of infectious complication, days as an inpatient and days with antimicrobial therapy, response to therapy, and overall survival (OS).

Results: Between August 2004 and January 2015, 77 MDS patients (M/F 57%/43%, median age 69 years, range 41-81 years) treated with azacitidine were included in the study. IPSS-R was high or very high in 57 % of the cases. The median number of administered AZA cycles was 6 (range 1-43 cycles). The response rates were 18% complete remission (CR), 26% partial remission (PR), 6% hematological improvement (HI), 18% stable disease (SD) and 34% progressive disease (PD). Median overall survival was 12 months (range 1-103). In Kaplan-Meyer-curve analysis AZA showed a trisection of all patients in three groups with significantly different survival, as well as incidence and duration of infectious complications. In the first group were all patients who reached a CR with a median OS of 37 months (range 7-72), a median of 0 cycles as an inpatient (range 0-4), 6% complication rate per cycle with a median of 6 (range 2-9) days of i.v. antibiotics and no days of i.v. antimycotics per complication and a median of 3.5 (range 0-84) days in hospital during all cycles. In the second group were all patients who reached PR or HI with a median OS of 23 months (range 1-103), a median of 0 cycles as an inpatient (range 0-7), 10% complication rate per cycle with a median of 8 (1-19) days of i.v. antibiotics and 10.5 (10-11) days of i.v. antimycotics and a median of 9 (range 0-115) days in hospital during all cycles. In the third group were all patients who remained stable or progressed with a median OS of 11 months (range 0-67), a median of 0 cycles as an inpatient (range 0-6), 32 % complication rate per cycle with a median of 9 (3-47) days of i.v. antibiotics and 10 (3-28) days of i.v. antimycotics and a median of 16 days in hospital during all cycles (range 0-78). In total, 614 treatment cycles were administered, and 81 infectious complications occurred. 18% of all infectious complications were lung infections documented by chest CT, 15% skin-/soft tissue infections, 10 % gastrointestinal infections, 3 % urinary tract infections and 58 % infections of unknown origin. Looking at all documented AEs (n=100), 36 were documented in the first 2 cycles of treatment, 30 during the cycles 3-5 and 22 after the fifth cycle.

Conclusion: The clinical course, incidence of infectious complications and the outcome of patients treated with 5-Azazytidine can be trisected in 3 groups only according to their hematological response. Therefore, any response, even only HI, diminishes the incidence rate of infectious complications and improves the outcome of patients. Of note, patients with stable disease had the same outcome and infection rates as patients with progressive disease. The best clinical course and outcome was seen in patients with complete remission. Most of the infectious complications occurred during the first treatment cycles. An early treatment response seems to reduce the incidence of infections and days in hospital and improves the overall outcome.