Identification of soluble immune markers for the diagnosis of aspergillosis and pneumocystosis: an exploratory study

Invasive fungal diseases (IFDs) are increasingly recognized in immunocompromised patients, but diagnosis remains challenging because current pathogen-targeted perform variably across heterogeneous immune backgrounds. We investigated whether bronchoalveolar lavage fluid (BALF) immune profiling could identify disease-specific signatures for aspergillosis and Pneumocystis pneumonia. BALF samples from 27 patients were analyzed by shotgun proteomics using a Q Exactive Plus mass spectrometer coupled to an EASY-nLC1200 system. Clinical groups included invasive pulmonary aspergillosis (IPA, n = 7), chronic pulmonary aspergillosis (CPA, n = 4), Pneumocystis pneumonia (PCP, n = 4), Pneumocystis colonization (PCC, n = 4), bacterial infection (BI, n = 4), and controls (n = 4). Overall, 2,278 proteins were identified, with a median of 1,000 proteins per sample (IQR 902-1,384). Within-group pairwise Jaccard similarity was high for controls (median, 0.89), PCP (0.91), PCC (0.87) and BI (0.79), but lower for IPA (0.73) and CPA (0.62), indicating greater heterogeneity in aspergillosis. Hierarchical clustering largely grouped controls, BI, PCC, and PCP according to clinical diagnosis, whereas aspergillosis profiles were more variable. Two IPA clusters emerged: kidney transplant recipients, whose proteomic profiles resembled controls, and non-neutropenic patients with hematological malignancies. Analysis of 77 soluble immune mediators produced broadly similar clustering, suggesting that immune proteins may distinguish clinically relevant disease patterns. In this exploratory analysis, BALF immune proteomic profiles differed according to clinical diagnosis, with the most pronounced heterogeneity observed within aspergillosis. Complement proteins and antimicrobial peptides emerged as candidate biomarkers that warrant validation in larger, independent cohorts before their diagnostic utility can be established.