Skip to main content

Droplet digital PCR for diagnosing invasive aspergillosis in hematological malignancies: a prospective cohort study

In brief

Aspergillus ddPCR detects infection well in lavage fluid, poorly in breath samples

In 203 patients with blood cancers, droplet digital PCR on bronchoalveolar lavage fluid had good diagnostic performance, with 88% overall accuracy by area under the curve, 65% sensitivity and 96% specificity. Testing exhaled breath condensate performed poorly; results were stronger in patients not already taking mold-active antifungals, but need further evaluation before routine use.

Journal
Journal of clinical microbiology (Q1)
Published
27 October 2026
Study design
Prospective / inception cohort
Evidence level
Level 2, Moderate (CEBM 2b)
Authors
Roni Bitterman, Chun Fai Ng, Simran Arora, Milena Casciato, Ursula Magyar, Sophia Salazar, et al.
PMID
42937364
DOI
10.1128/jcm.01140-26

Why clinicians should know about it

  • Picked for Microbiology (medical) (paper of the day, 29 October 2026): Prospective diagnostic accuracy of Aspergillus ddPCR versus qPCR

Abstract

Our objective was to evaluate the diagnostic accuracy of Aspergillus droplet digital PCR (ddPCR) in exhaled breath condensate (EBC) and bronchoalveolar lavage fluid (BALF) for diagnosing invasive aspergillosis (IA), and to compare its performance with quantitative real-time PCR (qPCR) in patients with hematologic malignancies. We conducted a prospective single-center study, including adults admitted to leukemia and allogeneic hematopoietic stem cell transplant units who underwent bronchoscopy for pulmonary infiltrates between October 2021 and December 2024. EBC and BALF were collected concurrently. ddPCR was performed on EBC and BALF (supernatant and cell pellet); qPCR was performed on BALF samples. IA was classified according to EORTC/MSGERC criteria without incorporating PCR into the reference standard. Diagnostic performance characteristics, including area under the receiver operating characteristic curve (AUC-ROC), were calculated. Among 203 participants, 26 (12.8%) were diagnosed with probable or proven IA. EBC ddPCR showed poor performance (AUC-ROC 0.55, 95% CI 0.43-0.67). BALF supernatant ddPCR demonstrated the best performance (AUC-ROC 0.88, 95% CI 0.80-0.96), with sensitivity of 0.65 (95% CI 0.43-0.84) and specificity of 0.96 (95% CI 0.92-0.99). BALF supernatant outperformed BALF cell pellet testing for both ddPCR and qPCR. Compared with qPCR, ddPCR showed higher sensitivity and similarly high specificity. Among participants without prior mold-active antifungal therapy, BALF supernatant ddPCR showed improved performance (AUC-ROC 0.95, 95% CI 0.89-1.00). In this large prospective study, EBC ddPCR demonstrated limited utility for the diagnosis of IA. BALF supernatant was the optimal specimen type, and ddPCR outperformed qPCR. Diagnostic performance improved substantially without prior mold-active antifungal exposure.IMPORTANCEInvasive aspergillosis is difficult to diagnose in patients with hematologic malignancies, and non-invasive diagnostic approaches are needed. In this prospective study of 203 patients, Aspergillus droplet digital PCR performed poorly in exhaled breath condensate (area under the receiver operating characteristic curve 0.55; 95% CI 0.43-0.67) but showed good diagnostic performance in bronchoalveolar lavage supernatant fluid (area under the receiver operating characteristic curve 0.88; 95% CI 0.80-0.96). Droplet digital PCR demonstrated higher sensitivity than quantitative PCR with similarly high specificity, and bronchoalveolar lavage supernatant fluid consistently outperformed cell pellet testing. Among patients without prior mold-active antifungal therapy, the area under the receiver operating characteristic curve was 0.95 (95% CI 0.89-1.00). These findings support bronchoalveolar lavage supernatant fluid as the preferred specimen for Aspergillus droplet digital PCR and warrant further evaluation of this approach for diagnosing invasive aspergillosis.

Abstract as published, via PubMed.

View on PubMedFull text at the publisherOpen in the app

For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.