Key findings

Six randomised controlled trials (RCTs) were identified, of which two were of good methodological quality. Two systematic reviews, incorporating four of these RCTs alongside comparative and non-comparative observational studies, were also identified.

Pre-emptive strategies are not standardised and the contribution of test findings to prescribing decisions varies. Protocols vary widely across the published literature, as does the use of prophylactic antifungal agents. Trials employ a wide range of patient inclusion criteria meaning that invasive fungal infection risk varies across studies. There is also heterogeneity across empirical strategies, measures of antifungal use and means of classifying invasive fungal infections.

Whilst it is not possible to reach robust conclusions from this heterogeneous evidence base, trials suggest that the use of pre-emptive strategies incorporating test findings may result in improved rates of identification of invasive fungal infection and reduced use of empirical antifungal medications.

There is no strong evidence that these potential benefits can be attained without adverse effects on mortality, meaning that there is substantial uncertainty surrounding the safety of pre-emptive strategies.

Cost-effectiveness evidence is subject to similar issues of heterogeneity, and the applicability of some economic analyses to the UK context may be limited.

One Australian economic study reported a non-significant difference in costs between a pre-emptive diagnostic-driven strategy and an empirical approach. Another study found a small difference in costs between the strategies, but was undermined by a lack of clarity around reporting and transcription. There was heterogeneity in the strategies analysed and the applicability of these findings may be limited by the local approach to empirical/pre-emptive antifungal treatment and cost setting. A UK economic study reported a large cost saving associated with the diagnostic-driven strategy, but methodological weaknesses relating to model design and data inputs were identified, and the reporting was unclear.

Uncertainty may be resolved on publication of a large (n=556) multicentre European trial comparing an empirical therapy strategy with a strictly defined diagnostic driven strategy. The primary outcome is mortality at 42 days and the investigators anticipate reporting their findings in 2019. NCT01288378

What were we asked to look at?

We were asked to look at the evidence on pre-emptive strategies incorporating galactomannan (GM) testing and/or polymerase chain reaction (PCR) assays compared with empirical antifungal treatment strategies for invasive Aspergillus infection in patients with haematological malignancies.

Why is this important?

Patients receiving treatments for haematological malignancies often experience severe and prolonged neutropenia which puts them at risk of life-threatening invasive fungal infections including Aspergillus infection. Since current diagnostic methods have poor sensitivity and can be slow, overuse of empirical antifungal therapies in this clinical setting is a significant problem which exposes patients to potential harms of treatment from which they derive no benefit. Resistance to antifungal medications is emerging as a serious threat. Our review assesses whether pre-emptive antifungal strategies incorporating novel biomarker tests could potentially address these issues.

Referred by

The Scottish Antimicrobial Prescribing Group (SAPG)