Factors influencing the performance level of Candida mannan antigen testing in allogeneic stem cell transplant recipients not receiving fluconazole prophylaxis

Abstract

In this study, we evaluated the value of the Platelia(®) Candida mannan antigen (Ag) sandwich enzyme-linked immunosorbent assay test in the diagnosis of invasive candidiasis (IC) and the degree of oral colonization by Candida species in 102 allogeneic stem cell transplantation recipients who were not receiving fluconazole prophylaxis. Of the 2071 serum samples, 98 (4.7%) yielded positive and 78 (3.8%) borderline results with a cut-off value of 0.5 ng/mL. One patient had IC. In this patient, 6 out of 9 serum samples were positive, the first one 49 days before Candida albicans candidemia. False-positive results occurred in 92 (4.4%) samples and in 54 (52.9%) patients. Use of valacyclovir and acyclovir was associated with false-positive or borderline results. The median Ag concentration of the true-positive results was significantly higher than the concentration of the false-positive results (1.60 versus 0.62 ng/mL, P<0.001). With higher cut-off values of 0.75 and 1.0 ng/mL, false-positive Ag test results were seen in 17 and 7 patients, respectively. Of the 657 oral samples, a total of 92 (14%) samples in 39 (38.2%) patients turned out to be positive. C. albicans grew in 82 samples (89.1%), other Candida species in 9 (9.8%), and Aspergillus fumigatus in 1 sample (1.1%). In conclusion, despite the lack of fluconazole prophylaxis, the incidence of IC was low (1%). False-positive Ag test results were common with a test cut-off value of 0.5 ng/mL, and a single positive result does not seem to predict IC. Multiple positive results might predict IC, as 6 out of 9 samples were positive in the only patient with IC, the first one 7 weeks before positive blood cultures.