Abstract
The diagnosis of invasive aspergillosis can be challenging in cancer patients. Herein, the analytical and clinical performance of the sōna Aspergillus galactomannan lateral flow assay (GM LFA) was evaluated and its performance compared to that of the Bio-Rad galactomannan enzyme immunoassay (GM EIA). Serum and bronchoalveolar lavage (BAL) fluid samples received for GM EIA testing between March and August 2019 were included. Positive and negative percent agreement (PPA and NPA) were calculated for the GM LFA compared to the GM EIA. Discrepant analysis was performed by review of the patient's medical records assessing for any evidence of a fungal infection. Five hundred thirty-three samples (85 BAL samples and 448 serum samples) from 379 patients were included in the study. The overall PPA and NPA were 100% (95% confidence interval [CI], 72.2 to 100%) and 97.5% (95% CI, 95.5 to 98.4%), respectively. Fourteen of 24 samples were positive by LFA only. The sensitivity of the GM LFA for proven and probable invasive aspergillosis (IA) was 100% (95% CI, 51.0 to 100%) and 87.5% (95% CI, 55.9 to 99.4%), with a specificity of 95.5% (95% CI, 92.3 to 97.2%) and 96.2% (95% CI, 93.4 to 97.7%), respectively. The sensitivity of the GM EIA for proven and probable IA was 25% (95% CI, 1.28 to 69.9%) and 62.5% (95% CI, 30.6 to 86.3%), with a specificity of 98.2% (95% CI, 96.2 to 99.1%) and 99.2% (95% CI, 97.7 to 99.8%), respectively. The Aspergillus GM LFA outperformed the Aspergillus GM EIA for the detection of the galactomannan antigen in our patient population. The simplicity and rapid time to results makes the Aspergillus GM LFA easy to implement in a wide range of laboratory settings.