Impact of treatment intensity on infectious complications in patients with acute myeloid leukemia

Abstract

Background: Infectious complications reflect a major challenge in the treatment of patients with acute myeloid leukemia (AML). Both induction chemotherapy and epigenetic treatment with hypomethylating agents (HMA) are associated with severe infections, while neutropenia represents a common risk factor. Here, 220 consecutive and newly diagnosed AML patients were analyzed with respect to infectious complications dependent on treatment intensity and antifungal prophylaxis applied to these patients.

Patients and methods: We retrospectively analyzed 220 patients with newly diagnosed AML at a tertiary care hospital between August 2016 and December 2020. The median age of AML patients undergoing induction chemotherapy (n = 102) was 61 years (25-76 years). Patients receiving palliative AML treatment (n = 118) had a median age of 75 years (53-91 years). We assessed the occurrence of infectious complication including the classification of pulmonary invasive fungal disease (IFD) according to the EORTC/MSG criteria at diagnosis and until day 100 after initiation of AML treatment. Furthermore, admission to intensive care unit (ICU) and subsequent outcome was analyzed for both groups of AML patients, respectively.

Results: AML patients subsequently allocated to palliative AML treatment have a significantly higher risk of pneumonia at diagnosis compared to patients undergoing induction chemotherapy (37.3% vs. 13.7%, P < 0.001) including a higher probability of atypical pneumonia (22.0% vs. 10.8%, P = 0.026). Furthermore, urinary tract infections are more frequent in the palliative subgroup at the time of AML diagnosis (5.1% vs. 0%, P = 0.021). Surprisingly, the incidence of pulmonary IFD is significantly lower after initiation of palliative AML treatment compared to the occurrence after induction chemotherapy (8.4% vs. 33.3%, P < 0.001) despite only few patients of the palliative treatment group received Aspergillus spp.-directed antifungal prophylaxis. The overall risk for infectious complications at AML diagnosis is significantly higher for palliative AML patients at diagnosis while patients undergoing induction chemotherapy have a significantly higher risk of infections after initiation of AML treatment. In addition, there is a strong correlation between the occurrence of pneumonia including atypical pneumonia and pulmonary IFD and the ECOG performance status at diagnosis in the palliative AML patient group. Analysis of intensive care unit (ICU) treatment (e.g. in case of sepsis or pneumonia) for both subgroups reveals a positive outcome in 10 of 15 patients (66.7%) with palliative AML treatment and in 15 of 18 patients (83.3%) receiving induction chemotherapy. Importantly, the presence of infections and the ECOG performance status at diagnosis significantly correlate with the overall survival (OS) of palliative AML patients (315 days w/o infection vs. 69 days with infection, P 0.0049 and 353 days for ECOG < 1 vs. 50 days for ECOG > 2, P < 0.001, respectively) in this intent-to-treat analysis.

Conclusion: The risk and the pattern of infectious complications at diagnosis and after initiation of AML therapy depends on age, ECOG performance status and subsequent treatment intensity. A comprehensive diagnostic work-up for identification of pulmonary IFD is indispensable for effective treatment of pneumonia in AML patients. The presence of infectious complications at diagnosis contributes to an inferior outcome in elderly AML patients.

Keywords: AML; Epigenetic therapy; IFD; Induction chemotherapy; Infections; Pneumonia.

Fig. 1

Kaplan–Meier estimates for overall survival (OS) dependent on the ECOG performance score A or on the modified Charlson comorbidity index B for AML patients undergoing palliative treatment

Fig. 2

Comparison of infectious complications after initiation of palliative AML treatment (until day 100) or following AML induction chemotherapy. The percentage of distinct subsets of infections is indicated: * significant difference (P < 0.001), ** not significant, *** not applicable. A more detailed overview is given in Table S2

Fig. 3

CONSORT diagram of time-dependent pneumonia subtypes for both patient groups. Distribution of absolute numbers of patients presenting with pneumonia either at the time of AML diagnosis (upper part) or after initiation of AML therapy (lower part) for patients allocated to palliative AML treatment (left) or induction chemotherapy (right)