Galactomannan serves as a surrogate endpoint for outcome of pulmonary invasive aspergillosis in neutropenic hematology patients

Abstract

A noninvasive, objective, reproducible, and quantitative Aspergillus-specific surrogate marker is needed for a more accurate assessment of the outcome of invasive aspergillosis (IA) in patients with a hematologic disorder. The quantitative serum galactomannan index (GMI) assay seems to fulfill the requirements of surrogacy for outcome evaluation. Kappa statistics were used to determine the strength of correlation between GMI outcome and clinical outcome (survival or death), autopsy data, and response outcome of IA in 70 adults with prolonged neutropenia. All patients underwent serial GMI monitoring until discharge or death. The overall correlation between GMI and clinical outcome was good at 6 weeks (kappa=0.5882; 95% confidence interval [95% CI], 0.4023-0.7741) and was excellent at 12 weeks (kappa=0.8857; 95% CI, 0.7766-0.9948). Concordance with autopsy findings was perfect (kappa=1). At 6 weeks, the correlation between GMI and response outcome (favorable or unfavorable) was excellent (kappa=0.7523; 95% CI, 0.5803-0.9243). Survival was significantly better in patients who became GMI-negative (P<.0001). In neutropenic patients with seropositive IA, serum galactomannan index outcome strongly correlates with survival, autopsy findings, and response outcome. This finding may have implications for patient management and for clinical trial design.