Targeted Pre-Emptive Anti-Aspergillus Therapy in Neutropenic Patients Not Receiving Empirical Antifungal Therapy.

Background. For several decades, empirical antifungal therapy has been the standard of care for the early treatment of invasive aspergillosis (IA) in neutropenic cancer patients. However, this approach results in the undue exposure to potentially toxic drugs, may select for resistant strains, and jeopardizes the hospital budget. We conducted a prospective study to evaluate the feasibility and safety of a targeted pre-emptive approach (while avoiding empirical therapy), based on the incorporation of sensitive, non-invasive diagnostic tests.

Methods. A total of 136 treatment episodes at high risk for IA (41 % ≥40 years of age; 26 % relapsed leukemia; 30 % high dose cytarabine; mean duration of neutropenia: 21.5 days; mean duration of broad-spectrum antibiotics: 15.9 days) were screened daily for circulating galactomannan by the Platelia (Bio-Rad) enzyme immunoassay (EIA) while receiving fluconazole prophylaxis. An index of ≥0.5 was considered positive if confirmed by the subsequent sample. In addition, well-defined clinical, radiological, and microbiological criteria triggered a diagnostics work-up for invasive fungal infection, including thoracic computed tomography scanning and bronchoscopy with bronchoalveolar lavage. Antifungal therapy (AmBisome 5 mg/kg/day) was withheld unless the patient became EIA-positive or had a positive culture or microscopy for molds with supportive radiological findings.

Results A total of 4170 serum samples (mean 30.6/episode) was analyzed. The combined EIA-CT-scan approach identified 19 episodes (14 percent) of IA. Less than one quarter of the episode that qualified for empirical therapy (unexplained persisting or relapsing fever) received antifungal therapy (6.6 percent versus 30 percent). Conversely, early antifungal therapy was started in 10 episodes (7.3 percent) not otherwise suspected for IA due to treatment with corticosteroids or due to the presence of confounding co-factors/co-infections. Undetected cases of IA were not found at autopsy. Unexpectedly, autopsy identified one patient with disseminated zymogycosis. At 12 weeks, a 77 percent survival rate was noted in case patients with controlled underlying hematological disease (versus 30 percent for patients with refractory underlying disease)

Conclusions. Replacing empirical therapy by a targeted, pre-emptive therapy offered effective antifungal control and spared patients from exposure to potentially toxic and expensive drugs without negative impact on mortality. This strategy should further be explored in a multicenter, randomized setting.