Introduction: Invasive fungal infections (IFI) are a leading cause of mortality in patients with acute myeloid leukemia (AML) undergoing induction chemotherapy including "7+3” regimen of cytarabine and daunorubicin.1 Posaconazole improves survival and reduces IFI rates in comparison with fluconazole.2 The aim of our study was to compare outcomes of patients receiving induction chemotherapy for AML at our institution before and after implementation of a priority matrix (PM) including standardized antifungal prophylaxis. Our primary endpoint was incidence of IFI, and our secondary endpoints were mortality at 30 days and 1 year.

Methods: All patients admitted with AML and receiving 7+3 induction chemotherapy at Gundersen Health System between January 2013 and January 2022 were included in our study. Patients were assigned either to Pre-PM or Post-PM group (whether induction was started before or after January 1, 2019, respectively). The priority matrix (Figure) included mandatory treatment with appropriate prophylaxis (posaconazole /voriconazole), patient education, electronic health record (EHR) updates, standardized work up for refractory fevers, and other optimization practices. Antifungal agent was given starting on day 8 after initiation of induction until resolution of neutropenia, defined as an absolute neutrophil count >500/uL. Treatment-related mortality rates (TRMR) and Charlson Comorbidity Index (CCI) were calculated at time of diagnosis. Statistical analyses using chi-square, Fisher exact, Wilcoxon rank sum tests, and Cox proportional hazards were done in SAS version 9.4.

Results: The study population consisted of 37 patients: 23 in the Pre-PM group (prior to Jan 1, 2019) and 14 in the Post-PM group. The population was predominantly white, non-Hispanic with a median age 63 years, and was 60% male. Groups were similar with regard to median age, sex, TRMR, and CCI. Of the 23 Pre-PM patients, 6 received fluconazole. All patients in the Post-PM group received prophylaxis-1 with voriconazole, and 13 with posaconazole. Six Pre-PM patients (26%) developed IFI, compared with 1 Post-PM patient (7%; p=0.22). Five of the 6 Pre-PM patients who received fluconazole developed IFI, as did 1 Post-PM patient who received posaconazole (p<0.001). Four (17%) Pre-PM patients, none of whom received fluconazole died within 30 days, and there were no deaths in the Post-PM group (p=NS). Fifteen (65%) Pre-PM patients died within 1 year (5 had received fluconazole prophylaxis), compared with 3 (21%) Post-PM patients. Mortality rates were not significantly different between groups at 30 days (p=0.22)); however, mortality at 1 year was significantly higher in the Pre-PM group (log rank p=0.004), including in those receiving fluconazole (log rank p=0.0005).

Conclusions: Our study shows significantly improved outcomes with use of a standardized protocol in AML patients receiving 7+3 induction by using posaconazole over fluconazole per our PM intervention. We reduced IFI rates to our goal of <10% in AML patients and improved 30-day and 1 -year survival at our institution, rates comparable to national averages. Because community health care facilities are increasingly treating patients with AML, we share our protocol to help improve care in this setting.


1Hsu A. et al. Supportive Care in Cancer, . 2021;29(2):707-712.

2Cornely OA, et al. N Engl J Med. 2007;356:348-359

No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution