Clinical and in vitro analysis of Aspergillus galactomannan antigen false positive results in patients with pulmonary cryptococcosis

T Takazono1,2,3, T Saijo2, K Oshima2, M Tashiro1, K Yamamoto2, Y Imamura2, T Miyazaki1,2, H Mukae2, DC Sheppard3, K Izumikawa1

Author address: 

1Department of Infectious Diseases, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan 2Department of Respiratory Diseases, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki , Japan 3Infectious Diseases and Immunity in Global Health Program, Research Institute of the McGill University Health Centre, Montreal, Canada


Purpose: Platelia Aspergillus assay is a commercially available enzyme-linked immunosorbent assay (ELISA) for the soluble Aspergillus galactomannan (GM) antigen; it indicates high sensitivity and specificity especially in diagnosis of invasive pulmonary aspergillosis and used worldwide. However, it is reported to cause false positive results in several situations; receiving piperacillin-tazobactam or amoxicillin-clavulanate, bifidobacterial lipoglycan, and taking a nutrient containing soybean protein. In 2005, Dalle et al. reported a case of pulmonary cryptococcosis patient with false positive results of GM antigen test, and concluded that Cryptococcus neoformans Galactoxylomannan indicates dose response cross-reactivity with Aspergillus GM in vitro experiment. However, Jesus et al. denied its reproducibility in vitro and in vivo study using mice and rabbit models in 2007. These two reports indicated conflict results despite their in vitro incubation methods were exactly the same; the cause of false positive is still controversial. Therefore, we conducted a retrospective study and in vitro experiments to clarify the cause of GM false positive results in pulmonary cryptococcosis patients.

Methods: A total of 41 pulmonary cryptococcosis patients diagnosed in Second Department of Internal Medicine, Nagasaki University Hospital between March 2008 and December 2014 were enrolled in this retrospective study. Medical information about diagnosis, laboratory tests and treatment were collected from medical records. 

Results: Among 41 pulmonary cryptococcosis patients, 30 patients were performed serum GM assay, and 15 patients of them were also performed GM assays in BALF. The mean of GM antigen titer was 1.37 (min;0.15, max;6.255) and 17 (56.7%) patients indicated positive results when cutoff value were taken as 0.5. No patient showed any evidence of Aspergillus co-infection. There was not a significant positive correlation between GM and Gulucronoxylomannan (GXM) antigens. And their serum GM titers significantly decreased after the treatment of cryptococcosis. Interestingly, their BALF GM titers were significantly lower than their serum ones unlike aspergillosis patients. Therefore, we hypothesized that some compositions in human serum might be contributing the GM false positive in cryptococcosis patients. We checked several serum compositions or incubation conditions to reproduce the GM false positive in in vitro experiment. And we found that C. neoformans showed significantly higher GM titer only when it was co-incubated with gamma globulin. Furthermore, ACAP1 (GXM deficient mutant) showed significantly higher GM than its wild type, this result implied that GalXM, the remaning component of C. neoformans capsule in ACAP1 was showing cross reactivity with GM assay.

Conclusion: More than half of pulmonary cryptococcosis patents showed false positive in GM assays. And GalXM was the most considerable reason of cross reactivity with GM assay.



Full conference title: 

The 8th Advances Against Aspergillus, Lisbon Conference Center, Lisbon, Portugal
    • AAA 8th (2018)