Statistical evaluation of a commercial Neospora caninum competitive ELISA in the absence of a gold standard: application to wild elk (Cervus elaphus) in Alberta

Abstract

Neospora caninum is an important abortive agent of domestic ruminants, but few diagnostic tools are available to reliably assess the exposure of wild cervid species such as elk (Cervus elaphus) to this pathogen, which limits our ability to understand their role in the life cycle of this parasite. In the absence of a gold standard test or panels of samples from individuals of known infection status, classical laboratory-based validation methods are not applicable. However, there are a number of statistical methods that can help in selecting an appropriate cut-off value and estimating the resulting diagnostic test performances. In this paper, the performance of a commercial competitive enzyme-linked immunosorbent assay (cELISA) on elk serum samples was evaluated with two statistical approaches: a mixture distribution model fitted to the cELISA results, and a Bayesian latent class analysis combining results from the cELISA and an indirect immuno-fluorescence antibody test. Both methods indicated that the commercial kit could be used on elk serum with the specifications recommended by the manufacturer. In particular, the optimal combination of sensitivity and specificity were obtained for a percentage of inhibition cutoff of 30%. The 95% probability interval of the proportion of elk exposed to N. caninum, adjusting for the sensitivity and specificity of this test in elk, was estimated between 1.3 and 7.4%. There was no association between the serological status of female elk and their pregnancy status. These results point out to the involvement of elk in a sylvatic cycle of N. caninum in this area.