BACKGROUND: Invasive pulmonary aspergillosis (IPA) is a severe complication of hematological malignancies. The European Organization for Research and Treatment of Cancer and the Mycoses Study Group Education and Research Consortium (EORTC/MSGERC) includes Aspergillus polymerase chain reaction (PCR), as a mycological criterion for IPA. METHODS: We retrospectively studied patients who underwent at least two Aspergillus fumigatus (Af) real-time PCR (qPCR) tests. Patients were classified according to the EORTC/MSGERC definitions, with or without considering qPCR results. We assessed the impact of qPCR on antifungal management and evaluated analytical specificity using non-Af molds DNA. RESULTS: We analyzed 296 serum and 136 respiratory specimens from 153 adults with hematological malignancies.Overall, using one positive qPCR as an IPA diagnostic criterion yielded a sensitivity and specificity of 0.73 and 0.71, respectively, whereas requiring two positive qPCR results reduced sensitivity to 0.31 and increased specificity to 0.93. Diagnostic performance differed according to specimen type. In blood specimens, sensitivity ranged from 0.33 with one positive qPCR result to 0.13 with two positive qPCR results, while specificity increased from 0.73 to 0.95. In respiratory specimens, sensitivity ranged from 0.70 to 0.60, with specificity from 0.60 to 0.87. Four patients with possible infection were reclassified as probable IPA and were treated with antifungals during routine care. Analytical testing revealed cross-reactivity with non-fumigatus Aspergillus species and non-Aspergillus molds. CONCLUSIONS: The performance of Af-qPCR in real-life settings depends on specimen type and positivity criteria. Cross-reactivities with other molds exist. Af-qPCR results should be interpreted cautiously, considering mycological and clinical data.
Gibert et al. (Thu,) studied this question.