Invasive fungal infections continue to form a major threat to hemato-oncologic patients, especially in those undergoing allogeneic stem cell transplantation.1 Whereas fluconazole offers effective prophylaxis for candidal infections, up to now no prophylactic measure has proven to be satisfactory against invasive mold infections. By far the most important of the invasive mold infections is aspergillosis, both in terms of frequency of occurrence and prognosis. Prophylaxis trials performed so far have been hampered by the relatively low risk of aspergillosis in the population studied, thereby masking a potential beneficial effect in high-risk subpopulations.2 In this issue, Marr and colleagues (page 1527) describe a carefully conducted prophylaxis trial in patients at high risk of fungal infection. They show that itraconazole is more effective than fluconazole in preventing invasive aspergillosis, as long as patients are able to take the drug. Does this finding mean that we should give itraconazole prophylaxis to these patients or even to all patients at risk for Aspergillus infection?
Because of difficulties in diagnosing and treating invasive fungal infection in severely immunocompromised patients, treatment strategies have concentrated on prophylaxis (preventing infections in all patients at risk), empirical treatment (early treatment in symptomatic patients who may have fungal infection, notably patients with “neutropenic fever”), or preemptive treatment (early treatment in those patients who are likely to have fungal infection). Whereas prophylaxis with itraconazole may now seem the best option at first glance, several questions remain unanswered. First, the time that the patient is at risk is variable and certainly not restricted to the neutropenic episode. In this respect it would be interesting to know how many patients in the study of Marr et al developed aspergillosis during neutropenia as opposed to during treatment with glucocorticosteroids for graft-versus-host disease (GVHD). Second, there is a safety issue. Itraconazole offers its broader antifungal spectrum as an advantage over fluconazole but at the same time is known to have more serious side effects and problematic interactions with many other drugs. In fact, the protocol of the study by Marr et al had to be adjusted because of a possible interaction of itraconazole with cyclophosphamide, and other interactions may still be undetected. This problem of side effects and drug interactions may also be a limiting factor with voriconazole, a new highly potent azole against Aspergillus species.3 Third, continuous administration of antimicrobial drugs will inevitably lead to development of resistance, as has already been described in Aspergillus after long-term treatment with itraconazole through emergence of multidrug resistance against all antifungal azoles.4
Therefore, what options do we have? Recently, progress has been made in the diagnosis of early, invasive aspergillosis, using both computed tomography (CT) scanning and serologic techniques. In a recent study, Becker et al5 claim a positive predictive value of 100% and a negative predictive value of 88% using a combination of CT scanning and Aspergillus galactomannan detection in serum or bronchoalveolar lavage (BAL) fluid. Moreover, new treatment modalities, although still far from perfect, offer a much better treatment outcome of established aspergillosis than before, voriconazole being far superior over conventional amphotericin B.3
The present study of Marr et al certainly provides a rationale for using itraconazole prophylaxis in high-risk patients, such as those receiving allogeneic stem cell transplantation, provided special attention is given to drug interactions and serum levels of itraconazole are measured for step-up (or even step-down) dosing strategies. An alternative approach would be to improve the identification of high-risk patients for primary itraconazole prophylaxis (eg, patients given glucocorticosteroids for GVHD) and in other patients at risk choose for a preemptive strategy with voriconazole, based on new diagnostic techniques, for instance sequential galactomannan assay–guided CT scanning. Which of the 2 strategies is best can only be determined by another carefully developed clinical trial.
