Abstract
Objectives
To evaluate a non-invasive panel integrating urinary triacetylfusarinine C normalised to creatinine (uTafC/Cr), serum galactomannan (sGM) and serum pentraxin-3 (sPtx3) for detection of invasive aspergillosis (IA) in patients with haematological malignancies.Methods
Prospectively collected urine and serum from 148 clinical episodes in 138 high-risk patients with haematological malignancies (Leuven, 2018-2021) were analysed in a diagnostic accuracy study (37 proven/probable IA episodes from 37 patients; 111 control episodes from 104 patients). Median sampling days relative to the diagnostic reference day (day 0) were +1 for urine uTafC/Cr, 0 for sGM and +2 for sPtx3. uTafC/Cr was quantified by liquid chromatography-mass spectrometry (LC-MS); sGM and sPtx3 by enzyme immunoassays. Prespecified cut-offs were used for sGM and uTafC/Cr; sPtx3 positivity was receiver operating characteristic-derived. Analytical TafC stability and the relative analytical sensitivity of LC-MS and MALDI-MS were assessed in validation experiments.Results
sGM and uTafC/Cr showed no false positives in available control episodes (sGM, 0/98; uTafC/Cr, 0/111) and each detected 51.4% of IA episodes (19/37). sPtx3 alone showed sensitivity of 70.3% (26/37) and specificity of 58.3% (21/36). Adding sPtx3 to the uTafC/Cr or sGM pair increased case detection to 83.8% (31/37), with lower specificity in the complete-case control episode subset (57.6%; 19/33 control episodes). In a ±72 h paired-window analysis, bronchoalveolar lavage fluid (BALF) mycology, defined as BALF GM, Aspergillus PCR and/or fungal culture, was positive in 19/23 tested IA episodes (82.6%; 95% CI, 62.9-93.0%). TafC was stable at 4 °C, and LC-MS was more sensitive than MALDI-MS.Conclusions
uTafC/Cr or sGM provides an Aspergillus-focused non-invasive adjunct to BALF-based testing; sPtx3 increased case detection but had limited specificity and requires prospective validation.