Persistently High Gam Levels Are Associated with Nonrelapse Mortality in HCT Recipients Irrespective of Invasive Aspergillosis: A Prospective Cohort Study

Detection of galactomannan (GAM) antigen in the serum has been found to strongly correlate with Invasive Aspergillosis (IA) in patients undergoing HCT. We prospectively evaluated the significance of serial GAM measurements in 11 patients undergoing HCT(cohort1) and compared this with 16 patients (cohort2) receiving non-HCT treatment for haematological disorders. GAM values were measured in all patients on admission and weekly thereafter until discharge. A CT scan of the chest was done for all patients with two serial GAM values above 0.5. A throat swab for fungal culture was sent for all patients at diagnosis and weekly thereafter. The median age of the entire group was 38 years (range 6 – 74 years). The cohorts did not differ significantly in age, gender or disease distributions. All patients received antifungal prophylaxis with fluconazole in cohort 2 and mould active azoles in cohort1. The GAM values on admission were not significantly different among two cohorts; however subsequent GAM values (1.32 Vs 0.77, p= 0.03) were significantly higher in cohort 1. The mean, maximum and minimum values of GAM (GAMmean, GAMmax, GAMmin) were significantly higher in cohort1(p <0.05). Two patients in cohort1 and one patient in cohort2 developed proven IA. GAM cutoff of 0.5 didnot correlate with mycological or radiological evidence of IA in either cohort. However a higher GAMmean ( 1.92Vs 0.81, p= 0.03) and GAMmax(3.17 Vs 2.69 p= 0.08) were associated with proven IA. Higher GAM values on admission (1.23 vs 0.54,p = 0.04) and higher GAMmean (1.54 vs0.76, p= 0.009) correlated with higher mortality irrespective of evidence of IA. In multivariate analysis, only GAMmean was associated with higher mortality, particularly in cohort 1. In conclusion, a cut-off value of 0.5 did not correlate with development of IA in our population, however a persistently high GAM value in HCT recipients was an adverse prognostic factor for nonrelapse mortality irrespective of evidence of IA.