Fungal infections caused by Aspergillus fumigatus are now identified in up to 45% of patients dying from haematological malignancy. There has been a significant increase in deaths from IA over the last 20 years. Our current diagnostic approach is neither sensitive nor specific. The purpose of this study is to prospectively assess the value of current diagnostic tools, as well as test other new diagnostic methods for the diagnosis of IA among haemato-oncology patients undergoing chemotherapy or stem cell transplantation.

The basis of the diagnosis of invasive aspergillosis is usually based on radiological appearances, which are neither sensitive nor specific. The burden of this problem is also not properly documented and there is paucity of prospective data in the literature. Therefore, we want to prospectively collect complete epidemiological data on all our patients. In addition we want to collaborate with radiological, respiratory, and microbiological colleagues to develop a unified approach to diagnosis. In the initial 12-18 months of the study all adult haemato-oncology patients likely to be rendered neutropenic during their treatment will be enrolled. All clinical data will be collected including dose, duration, and side-effects of anti-fungals administered. We will evaluate accepted diagnostic modalities for IA including the determination of optimum cut-offs for galactomannan and B-D-glucan in serum and urine in this cohort. In addition we would investigate the utility of other approaches for the diagnosis of IA such as markers for tissue injury and cytokine profiling.

We would test the urine in parallel with blood for galactomannan and B-D glucan to assess its usefulness with respect to blood.

CT scanning forms an important cornerstone of our diagnostic workup currently. However, there is paucity of data on the natural history and spectrum of CT changes in neutropenic patients with IA. Thus, our aim is to carefully document such changes in our cohort. We aim to rationalise CT imaging in the following way:

1. Baseline CT:

   We aim to perform an initial non-contrast enhanced thin-section continuous volume acquisition thoracic CT study on all the study patients. This will allow us to establish a "baseline" of normality in addition to potentially identifying those patients with pre-existing but indeterminate pulmonary lesions prior to chemotherapy or stem-cell transplantation.

2. Diagnostic CT:

   Neutropaenic patients with febrile episodes that are unresponsive to standard second-line broad-spectrum antibiotics combination (currently meropenem and vancomycin) will be referred for a contrast-enhanced thin section continuous volume CT scan. The purpose of this CT study is primarily to support the clinical suspicion of a diagnosis of IA and to determine its morphological extent. The purpose of the contrast injection is to test the hypothesis that in patients with IA, regions of necrotic lung (in contrast to other "inflammatory" or infective lesions) should not demonstrate any contrast enhancement.

3. Follow-up CTs (x2):

In patients with CT features of IA on the Diagnostic CT (see No. 2 above) and who have been commenced on antifungal chemotherapy, two follow-up, low-dose CTs (without iv contrast) will be performed at 10 days and 4 weeks after the diagnostic CT. These CT studies will not only allow us to evaluate the serial changes on CT but also determine the potential relationships between the initial CT features, haematological factors and outcome.

To increase the diagnostic yield, patients who are referred for lung biopsy, will undergo the technique of preoperative "labeling": small indeterminate lung nodules are frequently invisible and impalpable. There is an encouraging literature which indicates that preoperative labeling of lung lesions with a small (0.3-0.5ml) volume of methylene blue which acts a guidance track for the surgeon, may significantly improve the diagnostic yield from surgical (open or video-assisted thoracoscopic) biopsy.

Transplant patients typically would have 2-4 cycles of chemotherapy prior to admission for transplant. As such they have more chance of developing neutropenic infection and IA. Therefore we would perform a baseline bronchoscopy and washing (BAL) to assess the cytokine profile at admission and ensure that no infection is apparent before the initiation of transplant conditioning. A small amount of the BAL sample would be frozen and stored for future studies. Additional bronchoscopy may be done later during admission for both transplant and non-transplant patients if the clinical situation warrants it according to our current clinical practice.

Management strategies would also be assessed prospectively to evaluate the role of both prophylaxis and treatment. We are currently using itraconazole as our prophylactic agent of choice. Serum itraconazole levels will be measured on a weekly basis in all patients to ensure therapeutic levels are achieved.

We will be conducting costing analysis.

Inclusion Criteria:

• All adult haemato-oncology patients admitted for transplant or high dose chemotherapy and able to consent.

Exclusion Criteria:

• children (< 18 years old) or inability or refusal to consent.

Haemato-oncology patients undergoing high dose chemotherapy or stem cell transplantation likely to render them neutropenic.