Risk of Invasive Fungal Infections in Patients with High-Risk MDS and AML Receiving Hypomethylating Agents

Gee Youn (Geeny) Kim 1, Jonathan Burns 1, Craig W. Freyer 1, Keith Hamilton 2, Mindy Schuster 2, Noelle V. Frey 3, Elizabeth O. Hexner 4, James K. Mangan 4, Mary Ellen Martin 3, Shannon R. McCurdy 3, Alexander E. Perl 3, David L. Porter 5, Selina Luger 4 and Alison W. Loren 6

Author address: 

1Department of Pharmacy, Hospital of the University of Pennsylvania, Philadelphia, PA 2Division of Infectious Diseases, Hospital of the University of Pennsylvania, Philadelphia, PA 3Division of Hematology-Oncology/Department of Medicine, University of Pennsylvania, Philadelphia, PA 4Division of Hematology and Oncology, Abramson Cancer Center, Hospital of the University of Pennsylvania, Philadelphia, PA 5Division of Hematology and Oncology, Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA 6Division of Hematology-Oncology/Department of Medicine, Perelman Center For Advanced Medicine, Philadelphia, PA



Acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) patients are at risk of invasive fungal infection (IFI) due to prolonged neutropenia related to the underlying diseases and the induction chemotherapy required to treat them. The risk of IFI specifically in AML or MDS patients treated with hypomethylating agents (HMA) has not been fully elucidated, with previous publications reporting various prophylaxis practices and IFI incidences from 3.3% to 14.4%. We sought to characterize the incidence of IFI in patients with MDS and AML who received azacitidine (AZA) or decitabine (DAC) at our institution.


A retrospective cohort study was performed to evaluate the incidence of IFI in adults receiving between 10/1/10 and 10/1/17. Patients were included if they received at least two cycles of either drug and were diagnosed with AML, MDS with a revised international prognostic scoring system (R-IPSS) score of at least 3.5, or chronic myelomonocytic leukemia (CMML) categorized as intermediate 2 or higher risk using CMML-specific prognostic scoring system (CPSS). Patients were excluded if they received mold-active antifungal prophylaxis at any time during HMA therapy prior to the development of an IFI, or if they received mold-active antifungal treatment when starting a HMA. The primary endpoint was the incidence of IFI, defined as proven, probable, or possible using the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group criteria (EORTC).


Six hundred and fifty-one consecutive patients were screened, and 203 patients met inclusion criteria. The three most common reasons for exclusion were incomplete data (n=158), receipt of fewer than two cycles of a HMA (n=138), and receipt of mold-active antifungal treatment at the time of HMA initiation or mold-active antifungal prophylaxis at any point during HMA therapy (n=86). Six patients had two distinct episodes of HMA therapy separated by either an allogeneic stem cell transplant or a treatment other than HMA; thus 209 episodes of treatment occurred amongst 203 patients. Overall, the study population was elderly with the median age of 69, and 59% of the patients were male. Fifty-one percent of patients had AML, while 49% had MDS or CMML; 72% received AZA. The median number of cycles of HMA utilized was five. Most patients were treatment-naïve (59% AML and 87% MDS had no prior therapy). A significant portion of the group (25.8%) was neutropenic on cycle 1 day 1 (C1D1) of HMA. Other pertinent baseline characteristics of the study population are displayed in Table 1. We identified 20 cases of IFI, with an overall incidence of 9.6%. Most IFI were possible cases according to the EORTC criteria (13/20, 65%), but there were 3 culture-proven cases. The possible cases all met the host criterion of neutropenia and 12 of 13 met the clinical criterion of lower respiratory tract infection based on imaging. Organisms isolated on culture included Aspergillus spp., Fusarium spp., and Scedosporium spp. A diagnosis of IFI was made at a median of 3 cycles of HMA, with the majority (85%) diagnosed within the first four cycles. Among the 20 cases of IFI, 12 were diagnosed while receiving AZA and 8 were receiving DAC. More than half of the patients who developed IFI were neutropenic on C1D1. These patients had a significantly higher risk of IFI compared to non-neutropenic patients (20.4% vs. 5.8%, p=0.0051). When other baseline characteristics were analyzed, patients with therapy-related MDS (t-MDS) compared to de novo MDS were more likely to have an IFI (20.8% vs. 5.1%, p=0.0314). Patients with de novo AML also had a higher incidence of IFI compared to AML from an antecedent MDS/MPN (15.4% vs. 2.4%), however this did not reach statistical significance (p=0.07). Additional comparisons are described in Table 2.


IFI occurred in 9.6% of patients treated with HMA, but was observed in 20.4% of those who were neutropenic on C1D1 of HMA. Another potential risk factor identified was t-MDS versus de novo MDS. Prospective evaluation of mold-active antifungal prophylaxis in patients with MDS or AML who are neutropenic at the start of HMA therapy is warranted.



abstract No: 


Full conference title: 

60th American Society of Hematology Annual Meeting 2018
    • ASH 60th (2018)