Evaluation of the solid-phase competition ELISA for detecting SAT foot-and-mouth disease virus vaccination and infection in goats

Abstract

Foot-and-mouth disease (FMD) is a highly contagious and widely distributed disease affecting cloven-hoofed animals. The exact role played by goats in the epidemiology of the Southern African Territories (SAT) serotypes is an area currently under investigation. In this study, diagnostic properties of the solid phase competition ELISA (SPCE) were estimated using serum samples collected from goats that had participated in a FMD vaccination and challenge study. The goats were vaccinated with a pentavalent vaccine containing two SAT1 viruses, two SAT2 viruses and one SAT3 virus. After vaccination, the goats were challenged with a pool of three field SAT1 viruses. The SPCE were performed initially using a single-spot version (SS-SPCE) that was followed by a half-titration SPCE (T-SPCE) performed specifically for this study. The two independent runs (SS-SPCE and T-SPCE) were conducted in duplicate on two separate occasions. The repeatability of the assay was estimated from the duplicate wells (intra-aassay) and also from mean percentage inhibition from the separate tests (inter-assay). The coefficients of variations calculated from the duplicate percentage inhibitions from each of the two test runs were used to calculate within run repeatability while the mean percentage inhibitions for each run were used to measure inter-assay repeatability. The mean percentage inhibitions from the independent SPCE runs were used to estimate the level of correlation, agreement and relationship between them. The mean percentage inhibitions from each of the two SPCE runs were also compared to titer and the correlation, agreement and relationships were estimated. The ROC curve and area under the curve were used to estimate the accuracy of the SPCE and the optimum threshold cut off of the SPCE method across all three SAT serotypes was determined based on the Youden index (Y = maxc (Se (c) + Sp (c) – 1)). Approximately 80% of all intra-assay and 60% of all inter-assay SPCE results across all SAT serotypes had a good repeatability (<20% coefficient of variation). There was a very strong correlation between the two SPCE test runs and the titer results (Spearman’s rho=0.8 to 0.97). The agreement between the both SPCE runs was substantial across all serotypes, with kappa values ranging from 0.74 to 0.82. However, when the individual SPCE runs were compared with titer, the agreement was not as strong and differed according to serotype. SAT 1 agreement was moderate to substantial (k=0.592 and 0.612), SAT2 was moderate (k=0.423) and SAT3 was fair (k=0.24 and 0.309). The area under the T-SPCE ROC curve for SAT1, SAT2 and SAT3 was 0.98, 0.979 and 0.953 respectively, indicating a high diagnostic accuracy. The optimum SPCE percentage inhibition cut off based on the Youden index was found to be 75.63% PI for SAT1, 76.6% PI for SAT2 and 71.6% PI for SAT3. The cut off determined by this study were significantly higher when compared to the one normally used for this assay.