Risk for Invasive Fungal Infections during Acute Myeloid Leukemia Induction Therapy: a True Association with Echinocandins?

LETTER

We read the report by Gomes and colleagues characterizing the incidence of invasive fungal infections (IFIs) in patients with acute myeloid leukemia (AML) during remission induction chemotherapy (1). Given the high density of IFIs in the first 42 days after induction chemotherapy, as well as a randomized prospective trial showing the benefit of posaconazole prophylaxis in patients with AML, we agree that mold-active prophylaxis is most useful during this period (1, 2). However, this conclusion is not supported by the Gomes et al. article, as the rate of IFIs was no different in the fluconazole group. It is interesting that receipt of echinocandins (ECs) was associated with an increase in IFIs; however, there are many other potential explanations for this association that must be given due attention.

In the aforementioned study, rates of IFIs were reported as the incidence per 1,000 prophylaxis days (1). From Fig. 1 in reference 1, it is clear that anti-Aspergillus azole antifungals were evenly distributed throughout the 120 days postchemotherapy. In contrast, EC use was concentrated in the first 30 days postchemotherapy (1). It is well accepted that prolonged and profound neutropenia, typically seen in the first 30 days after induction chemotherapy, is associated with IFIs (3). The use of anti-Aspergillus azole antifungals during the nonneutropenic, low-risk period may have diluted the rate of IFIs associated with these agents and skewed the results against EC-based prophylaxis.

The authors acknowledge that the choice of primary antifungal prophylaxis (PAP) was left to the discretion of the treating hematologist (1). This inherent selection bias must be emphasized. EC-based prophylaxis is often utilized in patients who cannot tolerate oral therapy, typically due to severe mucositis related to intensive induction chemotherapy. Disruption of the intestinal tract, whether due to intensive chemotherapy or other comorbidities, is a major risk factor for the development of IFIs due to Candida spp. (3–6). ECs also often replace azoles in patients with liver dysfunction or in those receiving chemotherapy that may also cause hepatic injury. Patients with liver disease are at high risk for IFIs, including Candida spp., Cryptococcus spp., Aspergillus spp., and Coccidioides spp. (7, 8). In addition, compared to ECs, more patients discontinued anti-Aspergillus azoles due to adverse events (1). It is likely that these patients were then started on ECs, potentially enriching the EC prophylaxis group with patients with baseline toxicities and greater comorbidities. Inability to tolerate oral medications, switching to EC prophylaxis due to toxicity, and liver disease are also likely surrogate markers for severity of illness, another predictor of IFIs (9, 10). The lowest rate of IFI in this study was actually seen in the fluconazole group (Fig. 3), further reinforcing the importance of selection bias.

Although Gomes and colleagues found that 6/7 culture-positive IFIs were caused by fungi innately resistant or sporadically nonsusceptible to ECs, no MICs were reported, and overall, only 2/14 IFIs represent documented infections by organisms innately resistant in vitro to ECs (Geotrichum capitatum and Paecilomyces spp.) (1). Additionally, ECs have been shown to be not inferior to itraconazole, an anti-Aspergillus azole, for prophylaxis in patients with AML (11).

Baseline characteristics, specific chemotherapy regimens received, treatment response, and the aforementioned covariates should be reported and accounted for in the future multivariate analysis; however, the inherent limitations of a retrospective analysis in this population may still limit the applicability of these results to clinical practice. Without randomized controlled trials demonstrating a clear difference in rates of IFIs with ECs and anti-Aspergillus azole antifungals, it is difficult to support the conclusion that ECs may be inferior to anti-Aspergillus azole antifungals as PAP in patients receiving induction chemotherapy for AML.