Impact Of Posaconazole Prophylaxis On Invasive Fungal Infection During Acute Myeloid Leukemia Induction Therapy: Correlation With Posaconazole Serum Levels

The efficacy of posaconazole prophylaxis against invasive fungal infection (IFI) in acute myeloid leukemia (AML) patients during induction therapy has been demonstrated both in randomized and real-life studies. Absorption of posaconazole is enhanced by food intake, which may be impaired by gastrointestinal mucositis. However, correlation between posaconazole serum levels and breakthrough IFI has not been clearly demonstrated and the usefulness of therapeutic drug monitoring (TDM) is still a matter of debate.

In order to clarify the role of posaconazole TDM in preventing opportunistic infections, we correlated posaconazole serum levels with the incidence of breakthrough proven/probable IFI and the need for empiric/pre-emptive/targeted antifungal therapy among AML induction patients. A comparative analysis with an historical cohort receiving mainly itraconazole prophylaxis was also performed.

Since June 2004 a program of active epidemiological surveillance is ongoing at our Institute. Posaconazole prophylaxis was routinely introduced among AML induction patients on July 2011; data concerning incidence of IFI during AML induction were extracted from two consecutive periods (Jul2009-Jun2011 and Jul2011-Jun2013). Patients received posaconazole at 200 mg three times a day; 1^st serum posaconazole level was determined on day 7 and repeated every 7 days. A serum level ≥500 ng/mL was considered adequate. Serum galactomannan (GM) was monitored two times a week; thorax CT scan was performed within 3 days from the onset of fever. Systemic antifungal therapy was started empirically, in the case of persistent fever, with or without radiological signs of fungal infections.

Overall, 120 AML patients were evaluated, 60 in both periods. The two cohorts did not differ in median age, M/F ratio and type of treatment. During the 1^st period 9/60 (15%) breakthrough IFI (2 candidemia, 7 aspergillosis) were observed, as compared to 2/60 (3.3%) in the 2^nd period (1 candidemia and 1 aspergillosis), p=0.027 (Chi-square test). In all but one cases of aspergillosis, GM positivity was the microbiological criterion to meet the diagnosis of probable/proven IFI. The need for systemic antifungal therapy was similar in both periods (40% and 38.3%, respectively). Posaconazole serum levels was available in 50/60 patients (83%) during the 2^nd period. Mean serum concentration did not change significantly over time (mean value±SE: 551±48.7 543±64, 643±107.4, 698±164.7 at 1^st, 2^nd, 3^rd, 4^th week of therapy respectively, p=0.64). Adequate serum levels throughout the induction period were maintained in 19 cases (38%). None of them developed IFI. Posaconazole serum level was below the therapeutic threshold in the patient developing candidemia and not evaluable in the one who developed pulmonary aspergillosis 3 days after starting posaconazole prophylaxis. Systemic antifungal therapy, was given to 4/19 (21.1%) patients with persistently adequate posaconazole serum levels as compared to 13/31 (41.3%) patients with at least 1 posaconazole serum level below the therapeutic threshold (p=0.13).

Our study confirms that posaconazole prophylaxis significantly reduces the incidence of proven/probable breakthrough IFI, particularly aspergillosis, in a “real life” setting. No patient with persistently adequate posaconazole serum levels developed IFI. In this subgroup the need for systemic antifungal therapy was halved compared to patients with below-target posaconazole serum levels. Weather the reduced sensitivity of diagnostic biomarker tests due to azole prophylaxis may contribute to underestimate a diagnosis of IFI remains an open question.

No relevant conflicts of interest to declare.