Patterns of Infection in Patients with Myeloid Malignancies Receiving 5-Azacytidine: identification of Candidates for Antifungal Prophylaxis

Infectious complications are among the most recurrent causes of mortality in patients (pts) with acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) undergoing intensive chemotherapy (IC). These pts routinely receive anti-infective prophylaxis (AIP) with flourquinolones and antifungals. 5-azacytidine has recently been incorporated to treatment options for AML and MDS. However, the evidence of the effectiveness of AIP in patients treated with 5 azacytidine (AZA) is limited [1–3].

To analyze the incidence of episodes of infectious fever (IF), type of microbiological isolation and clinical relevance of infectious complications in AML and MDS pts treated with AZA who did not received prophylaxis. Identification a subgroup of pts who may benefit from AIP in this setting.

We retrospectively analyzed 48 pts with AML and MDS who received AZA from 2008, with a total of 365 cycles administered. Median age was 68 years (29–83y). Distribution: LMA (n=17) and MDS (n=31). One third of these pts had an absolute neutrophil count (ANC)<0. 5×10e⁹/L at time of starting AZA. Another 33% of pts had received prior IC, being all refractory to previous treatment. Baseline characteristics in table 1.

Forty-eight febrile episodes were recorded (13% of IF/cycles of AZA). There was no difference in IF in pts with ANC<0. 5×10⁹/L vs ANC>0. 5×10e⁹/L (p=0. 53). A total of 17 pts suffered at least one episode of IF (35% of the pts). Hospital admission was required in 14 of these 17 pts with a median time of hospitalization of 14 days (4–80). Mortality attributed to infectious complications ocurred only in 3/48 pts (6%). Twelve microbiological isolations were documented, the most common being: Gram negative bacilli (E Coli=4) and aspergillus reported as probable (n=4) and shown in table 1. Upon comparing pts who received prior IC (n=16; 33%) vs AZA as first line treatment, a higher risk for IF per cycle was observed in first group (18% vs 11. 5%; p=0. 06). Double of these pts developed fever (56% vs 25%; p=0. 03), required more hospital admissions (44% vs 22%; p=0. 21) and had longer duration of hospital stay (22 vs 14 days; p=0. 71). Finally, the group of patients that underwent previous IC, had higher rate of fungal infection by aspergillus and candida (5/9 isolations; 55% vs 0/5; 0%. P <0. 001), although no difference was observed in terms of mortality attributed to infection (6% each group) because of the reduced number of pts who died of this complication overall (3/48).

To our knowledge, this is the first study to evaluate the frequency and impact of IF in pts treated with AZA not receiving routinely AIP. Overall, the incidence of IF is lower than the reported in similar series. These results allow to identify pts that previously were treated with IC as those at highest risk of fungal infection. Thus, prophylaxis should be considered in this group. Prospective studies are needed to assess the requierement of prophylaxis during treatment with 5 azacytidine.

Jain N et al. Benefit of Anti-infectious Prophylaxis in Patients with Acute Myeloid Leukemia or High-Risk Myelodysplastic Syndrome receiving Frontline “Targeted Therapy”. Blood (ASH) 2007, 110:Abstract 2858

Je-Hwan Lee et al. Decreased incidence of febrile episodes with antibiotic prophylaxis in the treatment of decitabine for myelodysplastic syndrome. Leuk Res 35 (2011):499–503

Merkel D et al. Predictive Parameters for Infections During Azacitidine Therapy in High Risk MDS Patients. Blood (ASH) 2011, 118:Abstract 3811

No relevant conflicts of interest to declare.