ABSTRACT
This study aimed to determine the sensitivity (Se) and specificity (Sp) of a circulating pathogen-specific biomarker (polyketide synthetase 5, Pks5)-based enzyme-linked immunosorbent assay (ELISA) independently or in conjunction with a caudal fold tuberculin (CFT) test for bovine tuberculosis (bTB) screening in dairy cattle. We enrolled 987 dairy cows from 34 herds in Chiang Mai province, Thailand. A conditionally independent Bayesian model with a single population was inferred from the test results. The percentage of positive results for the Pks5-ELISA using 0.4 OD cutoff test and CFT test were 9.0% (89/987) and 10.5% (104/987), respectively. The median of posterior estimates of Se for the Pks5-ELISA test was 90.2% (95% posterior probability interval [PPI] = 76.6-97.4%), while the estimated Sp was slightly higher (median = 92.9, 95% PPI = 91.0-94.5%). The median estimated Se of the CFT test was 85.9% (95% PPI = 72.4-94.6%), while the estimated Sp was higher, with a median of 90.7% (95% PPI = 88.7-92.5%). The posterior estimate for true disease prevalence was 2.4% (95% PPI = 1.2-3.9%). The Pks5-ELISA test yielded characteristics at or above the acceptable standards for bTB detection. Therefore, the pathogen-specific biomarker, Pks5, is a potential detection system for bTB screening and may be applied as an ancillary test together with the currently applied standard method (CFT test) to reinforce the bTB control and eradication programs.
