A noninvasive urine assay demonstrated 92% sensitivity and 88% specificity for detecting invasive aspergillosis in a case-control study involving 50 patients with proven or probable infection and 240 controls.
The assay also yielded positive results in 79% of patients whose infections were diagnosed through bronchoalveolar lavage or biopsy despite negative serum galactomannan results, according to findings published in Clinical Infectious Diseases.
The study, led by Irina Baburina and colleagues, evaluated the MycoMEIA Aspergillus assay (Pearl Diagnostics), an FDA-cleared test designed to detect Aspergillus glycans in urine. The investigators conducted a case-control study and a prospective cohort study to assess its performance as an aid in diagnosing invasive aspergillosis, a fungal infection that can be difficult to identify in immunocompromised patients.
The assay uses two monoclonal antibodies to detect galactofuranose-containing glycans, including low-molecular-weight antigens associated with fungal extracellular vesicles. Unlike diagnostic approaches requiring bronchoscopy or tissue sampling, MycoMEIA uses urine specimens. The assay employs a microplate-based enzyme immunoassay, with results available in approximately 3.5 hours.
Study design and patient characteristics
The case-control analysis included 469 urine samples from 290 patients: 220 samples from 50 patients with proven or probable invasive aspergillosis and 249 samples from 240 controls without infection.
Patients were recruited through specimen repositories and clinical studies conducted at institutions in the United States and Belgium. Clinical diagnoses were independently adjudicated using the 2020 European Organisation for Research and Treatment of Cancer/Mycoses Study Group Education and Research Consortium consensus criteria.
Among the 50 patients with invasive aspergillosis, 47 (94%) had probable disease and three (6%) had proven disease. The median age was 53.4 years among patients with invasive aspergillosis and 52.6 years among controls.
Underlying risk factors included hematologic malignancy without hematopoietic cell transplantation in 27 patients (54%), allogeneic hematopoietic cell transplantation in 19 (38%) and solid organ transplantation in two (4%). Twenty-four patients (48%) had neutropenia, while 11 (22%) had both neutropenia and other immunosuppression.
The investigators assessed assay sensitivity, specificity, predictive values, agreement with serum galactomannan testing, potential cross-reactivity and timing of positive results relative to clinical diagnosis. A separate prospective cohort evaluated specificity using freshly collected urine specimens, while reproducibility was assessed across three laboratories.
Diagnostic performance in the case-control study
At the prespecified optical density index (ODI) cutoff of 0.6, MycoMEIA detected invasive aspergillosis in 46 of 50 patients, yielding 92% sensitivity. Specificity was 88%, with 31 of 249 control samples producing false-positive results.
Increasing the positivity cutoff to 0.7 reduced sensitivity to 88% but improved specificity to 93%.
The investigators also modeled positive and negative predictive values using estimated pretest probabilities of 10% for screening patients at high risk and 30% for evaluating patients with suspected invasive aspergillosis.
At a 10% pretest probability, the assay had a positive predictive value of 45% and a negative predictive value of 99%. At a 30% pretest probability, the corresponding values were 76% and 96%.
Four patients with probable invasive aspergillosis had false-negative urine results.
Comparison with serum galactomannan testing
Among 49 patients with available serum galactomannan enzyme immunoassay results, 25 (51%) had at least one positive test. Urine MycoMEIA testing was positive in 46 of 50 patients (92%) with invasive aspergillosis.
All patients with positive serum galactomannan results also had positive MycoMEIA results.
Among 24 patients whose invasive aspergillosis diagnoses were established through bronchoalveolar lavage fluid or biopsy despite negative serum galactomannan results, 19 (79%) had at least one positive urine assay.
Urine assay positivity was similar across neutropenia subgroups. MycoMEIA was positive in 31 of 33 patients with neutropenia (94%) and 15 of 16 patients without neutropenia (94%). Corresponding serum galactomannan positivity rates were 48% and 56%, respectively.
Among 21 patients with urine samples collected on or before the date of clinical diagnosis, 13 (62%) had positive MycoMEIA results before invasive aspergillosis was confirmed through standard clinical evaluation. The median lead time was five days.
The investigators noted that because serum galactomannan results contributed to the consensus definition of probable invasive aspergillosis, the study design did not permit a direct comparison of sensitivity and specificity between the two assays.
Prospective cohort findings
A prospective study conducted between January 2023 and February 2024 evaluated MycoMEIA performance using fresh urine specimens tested in a hospital microbiology laboratory.
The study initially enrolled 210 patients who provided 254 urine samples. After exclusions, 171 suspected infection episodes involving 168 patients were eligible for analysis.
The median patient age was 64 years, and 67% were male. Hematologic malignancies were present in 79% of patients, while 17% had undergone solid organ transplantation. Neutropenia occurred in approximately 32% of eligible episodes.
Seven episodes were classified as probable invasive aspergillosis, 39 as having no infection, 102 as possible invasive aspergillosis and 23 as mixed or other infections.
Among the seven episodes classified as probable invasive aspergillosis, six (86%) had positive MycoMEIA results. Among 39 episodes classified as having no infection, four had false-positive urine results, corresponding to a specificity of 90%.
Of the 102 episodes classified as possible invasive aspergillosis, 28 (27%) had positive urine MycoMEIA results. These patients had clinical suspicion of infection but lacked sufficient microbiologic evidence to meet criteria for proven or probable invasive aspergillosis.
The investigators emphasized that positive findings in this group could not establish whether patients had invasive aspergillosis because of nonspecific radiographic findings and the absence of confirmatory microbiologic evidence.
Cross-reactivity and reproducibility
Cross-reactivity was observed in patients with other fungal infections caused by Ascomycetes, including Blastomyces, Histoplasma and Fusarium species.
Positive results occurred in two of nine episodes of histoplasmosis, one of two episodes involving Fusarium species and the single episode of blastomycosis evaluated. One of six episodes of candidemia also produced a positive result.
Two patients with severe gastrointestinal mucositis had positive MycoMEIA results despite having no radiographic evidence of invasive aspergillosis.
In contrast, the assay was negative in the evaluated episodes of bacterial urinary tract infection, bacteremia and bacterial pneumonia with bacteremia.
Assay reproducibility was evaluated at three laboratories using blinded urine specimens. Results showed intraclass correlation coefficients of 0.99 for both optical density measurements and normalized ODI values, indicating high agreement across sites.
Study limitations
The investigators identified several limitations. Most urine specimens from patients with invasive aspergillosis had been frozen before testing. Although the prospective study demonstrated similar specificity using fresh urine specimens, the small number of patients meeting diagnostic criteria for probable invasive aspergillosis prevented a reliable estimate of sensitivity in the prospective setting.
The use of consensus diagnostic criteria also complicated comparisons with serum galactomannan testing because galactomannan positivity contributed to the classification of probable invasive aspergillosis. The absence of a definitive diagnostic reference standard further limited comparative interpretation.
Cross-reactivity with other fungal pathogens and false-positive results near the assay's positivity threshold were additional concerns. The investigators suggested that repeat testing may be reasonable for samples with an ODI between 0.6 and 0.7, provided patients receive close clinical follow-up.
The study population consisted predominantly of patients with hematologic malignancies or other immunosuppressive conditions. Additional research is needed to establish assay performance in broader populations, including lung transplant recipients and patients with severe pulmonary disease.
The investigators also noted that additional prospective studies using synchronized test timing and latent class modeling are needed to compare the performance of diagnostic assays.
Conclusions
The investigators concluded that the urinary MycoMEIA Aspergillus assay may serve as a noninvasive diagnostic aid for invasive aspergillosis. Combining the assay with other diagnostic methods may provide complementary clinical value.
“Deployment as an aid to diagnose IA may support clinical decision-making, with sensitivity sufficient to warrant further exploration as a screening tool in people at high risk,” the authors said.
Disclosures
The study was funded by Pearl Diagnostics and a National Institute of Allergy and Infectious Diseases grant.
Irina Baburina, Kieren A. Marr and Patricia X. Marques reported holding shares in Pearl Diagnostics. Baburina and Marques also reported employment with the company, while Marr reported consulting income from Pearl Diagnostics and employment income from Elion Therapeutics.
Additional authors reported financial relationships involving consulting fees, research funding, grants, travel support or other payments from pharmaceutical and diagnostic companies, including Pearl Diagnostics, Pfizer, Gilead, Merck, Scynexis and others. All authors submitted International Committee of Medical Journal Editors conflict-of-interest disclosure forms.