Evaluation of INSTAND e.V.'s external quality assessment for Cryptococcus antigen testing autumn 2016-spring 2021

Background: Cryptococcus antigen testing is a highly sensitive and specific diagnostic test, as documented predominantly in studies performed in areas with high prevalence of AIDS associated cryptococcosis. Here, we analyse its performance based on the results of an external quality assessment (EQA) scheme in Germany. Methods: We analysed method specific results on Cryptococcus antigen testing collected by the Society for Promotion of Quality Assurance in Medical Laboratories (INSTAND e.V.) from eight EQA's performed between 2016 and 2021. In every EQA the qualitative and semi-quantitative results with regard to a negative and a positive sample were evaluated. The positive samples consisted of patient sera, while the negative samples were taken from blood donors. The range of acceptable semi-quantitative results for the positive samples was defined as two dilution titres below and above the median of method-specific collectives. Results: Between 75 and 96 laboratories participated, two-thirds from Germany. The most common methods were a lateral flow assay (LFA: 40.5%) and latex agglutination tests, either by monoclonal antibodies (mAb LAT: 23.1%) or by polyclonal antibodies (pAb LAT: 34%). Overall, most of the participants were able to distinguish the positive from the negative samples, but occasionally false-positive titres (mAb LAT: 3%; pAb LAT: 2.8%) were reported. For the semi-quantitative results of the positive samples, the median titre was highest in the LFA (1:160-1:2560) and lowest in the mAb LAT (1:32-1:512). After exclusion of missing answers and potentially swapped samples, the titres were in the acceptable range in 92.8% (LFA), 94.9% (mAb LAT) and 84.7% (pAb LAT). Conclusion: While qualitative detection of Cryptococcus antigen was accurate among participants, semi-quantitative tests' results, i.e. titres, vary considerably depending on methods and between participating laboratories.