Although the clinical outcome in patients with Invasive Aspergillosis (IA) is largely dependent on early initiation of effective treatment with antifungal drugs, diagnosing IA is still a critical problem. Symptoms are non-specific and available diagnostic tools are either invasive or have low sensitivity and specificity. This often results in a diagnostic delay, with patients developing more extensive disease. Furthermore, as long as IA is present, oncological follow-up treatment is not feasible. Inaccuracy in diagnosing IA can cause serious treatment delay and increased mortality. However, an empirical strategy with prophylactic anti-mould therapy is not feasible considering both possible side effects and costs. In order to safely continue the use of a pre-empirical strategy, improved (non-invasive) diagnostic tools are desirable.

In a pilot study de Heer et al. showed that it is possible to discriminate between patients with IA and their neutropenic controls by exhaled breath analysis using an electronic nose (eNose). In this study the investigators aim to test whether an eNose could be useful as a diagnostic tool in a prospective setting.

The gold standard in exhaled breath analysis is Gas Chromatography - Mass Spectrometry (GC-MS). This technique enables identification of volatile organic compounds (VOCs) in breath of patients. It is possible that there are Aspergillus specific VOCs in the breath of patients with IA.

The composition of the lung microbiome seems to be an important factor in both health and disease. It is likely that the microbiome of the lung changes in prolonged neutropenia, therefore possibly creating a niche for molds and yeasts. Comparing the microbiome of patients with prolonged neutropenia who develop IA with those who do not, can learn us more about the pathogenesis of this disease. This knowledge could be used to investigate new treatment options for Invasive Aspergillosis.

Hypothesis The investigators hypothesize that airway microbial (viral, bacterial) presence and exhaled molecular profiles as obtained from patients with prolonged neutropenia due to treatment of hematological malignancies, are different between patients who develop IA and patients who do not.

Aims

1. To compare the exhaled molecular profiles (GC-MS and eNose) between neutropenic patients who develop IA and neutropenic controls.
2. To investigate whether exhaled molecular profiles can serve as surrogate to predict IA at an early stage.
3. To compare the alterations in the viral/bacterial microbial profiles during the neutropenic episode between patients who develop IA and controls.
4. To examine the relationship between microbial and molecular profiles in order to generate mechanistic hypotheses.

Inclusion Criteria:

Patients are:

• aged 18 years or older;
• diagnosed with a hematological malignancy;
• treatment is expected to result in prolonged (>7 days) neutropenia (<0.5 x 10^9/L)
• able to give written and dated informed consent prior to any study specific procedures.

Exclusion Criteria:

• Patients are unable to perform the breathing manoeuvre needed for eNose-analysis of exhaled air

All patients aged 18 or older, admitted at the hematology department of the AMC or UMCU, that will undergo treatment for a hematological malignancy expected to result in prolonged neutropenia (neutrophil counts <0.5 x 10 ^9/L for more than seven days).