Invasive Aspergillosis and AML

By Andrew S. Artz, Division of Hematology/Oncology, University of Chicago. Dr. Artz reports no financial relationship relevant to this field of study.

Abstract & Commentary

Synopsis: Invasive aspergillosis remains a problematic complication of therapy. The authors prospectively evaluated IA infections among AML patients at 21 centers. Among 140 cases of probable or definite IA, attributable mortality was 27%. Relapsed/refractory AML and lack of recovery from neutropenia were the strongest adverse prognostic factors. Most patients underwent empirical or pre-emptive therapy. Response rate varied by agent from amphotericin, caspofungin, and voriconazole but was not statistically different. Invasive aspergillosis-related mortality for AML patients continues to improve although mortality remains substantial.

Source: Pagano L, M. et al. Invasive Aspergillosis in patients with acute myeloid leukemia: A SEIFEM-2008 registry study. Haematologica. 2010;95:644-650.

Invasive fungal infections have been a long-feared complication in treating patients with hematologic malignancies. Aspergillosis is the most common invasive fungal infection, as many have adopted routine anti-yeast prophylaxis for Candida albicans. A recent series confirmed for AML patients, aspergillosis persisted as the most common invasive fungal infection.1 Neutropenia and immune suppression are the principal risk factors for invasive aspergillosis (IA).2 The epidemiology of IA may be changing with the availability of better diagnostic tools and additional aspergillosis agents. Some data do suggest improving outcomes over time related to IA infections.1

The authors embarked on a prospective registry study to characterize the epidemiology and outcomes related to IA among AML patients; twenty-one Italian centers participated. Prior hematopoietic cell transplantation was excluded, or patients who had already received two or more chemotherapy regimens. Between 2004 and 2007, 140 cases of proven or probable IA were identified. The median age was 57 years, and 85 (60%) occurred during aplasia from initial induction therapy. The lung represented the most common site of infection (90%), and almost all cases occurred in the setting of severe neutropenia (93%). On average, 12 days elapsed between symptom onset and diagnosis of IA. Aspergillosis fumigatus was the most common subspecies. The majority of patients (86%) generally received antifungal prophylaxis using itraconazole or fluconazole. Therapy was empiric in 62% and pre-emptive in 29%. G-CSF was initiated in two-thirds of patients as part of the treatment plan.

Mortality by day 120 was 33%. Death was related to IA in 38/140 (27%). As expected, patients having relapsed/refractory AML harboring an IA infection fared much worse than those in remission (43% compared to 19%, p = 0.002). Prolonged neutropenia of 10 days or more was associated with increased IA-attributable mortality. G-CSF administration was associated with reduced neutropenia, but did not alter IA-related mortality. No anti-mold drug was linked to better outcomes, with the three most common drugs being amphotericin preparations, caspofungin, and voriconazole. A good response to therapy for IA occurred in 71%.


Invasive fungal infections remain one of the most feared complications from immunosuppressive and highly myelosuppressive therapy. Of these, IA is the most common.

The findings of these studies mirror other published studies. Most infections were identified in the lung during neutropenia. Prolonged secondary G-CSF therapy (i.e., started for signs or symptoms of infection) was associated with faster neutrophil recovery than non-G-CSF therapy, but did not improve outcomes related to IA. Granulocytic transfusions were only given to two patients, so could not be assessed. The response rates to first-line therapy in 101 evaluable patients were 68%, 61%, and 84%, respectively, for liposomal amphotericin, caspofungin, and voriconazole. This was not statistically different, and baseline characteristics were not necessarily similar. Around 16% of patients received two aspergillosis-active drugs, usually added sequentially.

Overall, aspergillosis-attributable mortality was 27%, indicating continued improvements in outcomes over time. As expected, patients having relapsed/refractory AML harboring an IA infection fared much worse than those in remission (43% compared to 19%, p = 0.002). Prolonged neutropenia of 10 days or more was associated with increased IA-attributable mortality.

To put the results in context, patients only received prophylaxis with fluconazole or itraconazole. Fluconazole is inactive against aspergillosis, and itraconazole has limited activity. Whether aspergillosis-active drugs such as voriconazole and posaconazole should be used for prophylaxis remains an area of controversy. Routine prophylaxis would add significant cost and possibly toxicity. The strategy of empiric therapy (started with signs or symptoms of infection) or pre-emptive therapy (signs and symptoms and suspicion for aspergillosis) appears to have had a high success rate. Specifically, response rate overall to single therapy was 71%, and as high as 84% for voriconazole. Interestingly, most patients received empirical or pre-emptive therapy, suggesting better diagnosis, awareness, and/or low threshold to employ IA treatment. IA-related mortality was 27%, indicating marked improvement from historic results.

Voriconazole has improved activity compared to conventional (non-lipid) amphotericin.3 The response rate in a large randomized trial for voriconazole was 53%, but many patients had undergone hematopoietic stem-cell transplantation, a higher-risk group of patients.

The improvement in historically dismal outcomes certainly stems, in part, from more tolerable, if not more effective, drugs, such as echinocandins and extended spectrum azoles (e.g., voriconazole). Still, other advances, such as improved detection through high-resolution CT scanning, rapid bronchoscopy, and/or galatomman assay, may also play important roles.

In summary, aspergillosis remains a problematic infection for AML patients. Although mortality is significant, response rates and attributable mortality continues to improve, probably related to more active anti-aspergillosis therapy. For physicians treating AML, one must have a high index of suspicion for IA and consider early pre-emptive or empirical therapy.


1. Pagano L, et al. The epidemiology of fungal infections in patients with hematologic malignancies: the SEIFEM-2004 study. Haematologica. 2006;91:1068-1075.

2. Denning DW, Stevens DA: Antifungal and surgical treatment of invasive aspergillosis: review of 2,121 published cases. Rev Infect Dis. 1990;12:1147-1201.

3. Herbrecht R, et al. Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. N Engl J Med. 2002;347:408-415.