Diagnosis and Management of Invasive Aspergillosis in Adult Neutropenic Haemato-Oncology Patients

Abstract

Background Invasive fungal disease (IFD) is difficult to diagnose. For clinical trials the European Organisation for Research in Treatment of Cancer (EORTC) and the Mycology Study Group (MSG) criteria are useful but there are few data on their value in clinical practice. The aims of this study were to: (1) investigate the incidence and risk factors of IFD; (2) assess the utility of galactomannan (GM), β-D glucan (BDG), the UK consensus fungal PCR, and lateral flow device (LFD) assays together with the safety and feasibility of biopsy; (3) assess the role of cytokines in the diagnosis and prognosis of IFD; (4) establish the prevalence of baseline CT abnormalities, and assess diagnostic CT features and spectrum of radiological signs. Methods Patients (N=203) were recruited prospectively and followed for a median (range) of 556 (12-730) days after chemotherapy or haematopoietic stem cell transplantation. Chest CT, Karnofsky score (KS), serum GM, and cytokine profiles were performed at baseline; during admission twice-weekly GM assays were performed on all patients. BDG, serum and whole blood consensus PCR, and LFD assays were performed on a selection of samples from different IFD categories. Neutropenic sepsis refractory to antimicrobials for ≥4 days triggered diagnostic CT and biopsy where feasible. All patients were on antifungal prophylaxis from admission. The revised EORTC/MSG criteria were used to diagnose IFD. Results The cumulative incidence of proven/probable IFD at 6, 12, and 24 months was 16, 19, and 21%, respectively. Using GM or BDG alone (plus host and clinical evidence) the apparent incidence of proven/probable IFD was 11 and 16% respectively. The median time from index treatment to onset of IFD was 142, 17, and 41 days for proven mould, proven candida and probable IFD, respectively (P < 0.001). The 2-year overall survival (95% CI) were 45 (29-61)% for proven/probable IFD vs. 87 (77-97)% for no evidence of IFD (P < 0.001). Baseline CT abnormalities were found in 76 (38%) patients, 19 of which were EORTC signs. Risk factors for IFD on multivariate Cox regression were: EORTC CT signs (Hazard ration [HR] 4.3; 95% CI 1.9-9.8; P<0.001), IL-2R >834 pg/ml (HR 2.3; 95% CI 1.1-5.1; P=0.037), MCP-1 >841 pg/ml (HR 2.7; 95% CI 1.2-6.1; P=0.016), and KS <90 (HR 2.1; 95% CI 1.1-4.2; P=0.034) at baseline as well as monocytopenia >10 days (HR 2.6; 95% CI 1.3-5.4; P=0.009) and bacteraemia (HR 2.5; 95% CI 1.2-5.0; P=0.013). The sensitivity, specificity, PPV, and NPV for GM was 54, 71, 82, and 39%, respectively; coresponding values for BDG were 79, 55, 83, and 49%. The proportion of cases identified by both tests was 38%. The sensitivity of biopsy was 35% and PCR and LFD were less sensitive but more specific than GM or BDG. Conclusions A multi-faceted approach is required to improve the diagnostic accuracy of IFD. No single assay was able to detect all the cases but the combination of BDG and GM seem to offer the best biomarker combination. Baseline chest CT, KS, and cytokine profile as well as monocytopenia and bacteraemia are important risk factors. (ClinicalTrials.gov number NCT00816088).

Date of Award	2014
Original language	English
Awarding Institution	King's College London
Supervisor	Ghulam Mufti (Supervisor) & Melvyn Smith (Supervisor)