Advances in the Diagnosis of Human Opisthorchiasis: Development of Opisthorchis viverrini Antigen Detection in Urine

Abstract

Background: Many strategies to control opisthorchiasis have been employed in Thailand, but not in the other neighbouring countries. Specific control methods include mass drug administration (MDA) and health education to reduce raw fish consumption. These control efforts have greatly shifted the epidemiology of Opisthorchis viverrini (OV) infection over the last decade from presenting as densely concentrated "heavy" infections in single villages to widespread "light" OV infections distributed over wide geographical areas. Currently, the "gold standard" detection method for OV infection is formalin ethyl-acetate concentration technique (FECT), which has limited diagnostic sensitivity and diagnostic specificity for light OV infections, with OV eggs often confused with eggs of minute intestinal flukes (MIFs) in feces. In this study, we developed and evaluated the diagnostic performance of a monoclonal antibody-based enzyme-linked immunosorbent assay for the measurement of OV excretory-secretory (ES) antigens in urine (urine OV-ES assay) for the diagnosis of opisthorchiasis compared to the gold standard detection FECT method.

Methodology: We tested several methods for pre-treating urine samples prior to testing the diagnostic performance of the urine OV-ES assay. Using trichloroacetic acid (TCA) pre-treated urine, we compared detection and quantification of OV infection using the urine OV-ES assay versus FECT in OV-endemic areas in Northeastern Thailand. Receiver operating characteristic (ROC) curves were used to determine the diagnostic sensitivity and specificity of the urine OV-ES assay using TCA pre-treated urine, and to establish diagnostic positivity thresholds. The Positive Predictive Value as well as the likelihood of obtaining a positive test result (LR+) or a negative test result (LR-) were calculated for the established diagnostic positivity threshold. Diagnostic risks (Odds Ratios) were estimated using logistic regression.

Results: When urine samples were pre-treated with TCA prior to use in the urine OV-ES assay, the analytical sensitivity was significantly improved. Using TCA pre-treatment of urine, the urine OV-ES assay had a limit of detection (LoD) of 39 ng/ml compared to the LoD of 52 ng/mL reported for coprological antigen detection methods. Similarly, the urine OV-ES assay correlated significantly with intensity of OV infection as measured by FECT. The urine OV-ES assay was also able to detect 28 individuals as positive from the 63 (44.4%) individuals previously determined to be negative using FECT. The likelihood of a positive diagnosis of OV infection by urine OV-ES assay increased significantly with the intensity of OV infection as determined by FECT. With reference to FECT, the sensitivity and specificity of the urine OV-ES assay was 81% and 70%, respectively.

Conclusion: The detection of OV-infection by the urine OV-ES assay showed much greater diagnostic sensitivity and diagnostic specificity than the current "gold standard" FECT method for the detection and quantification of OV infection. Due to its ease-of-use, and noninvasive sample collection (urine), the urine OV-ES assay offers the potential to revolutionize the diagnosis of liver fluke infection and provide an effective tool for control and elimination of these tumorigenic parasites.

Fig 1. Receiver operating characteristic (ROC) curve analyses comparing trichloroacetic acid (TCA), frozen, heating and alkaline methods in detecting O. viverinni infection in Sample Set 1 (n = 50).

ROC curve, using individuals from Sample Set 1 were used to compare the diagnostic performance of antigen detection TCA, frozen, heating and alkaline treatment methods in the O. viverrini Excretory Secretory assay. Sensitivity against 1-specificity for the antigen levels detected by each ELISA method from confirmed positives and negatives by the gold standard method formalin ethyl-acetate concentration technique (FECT).

Fig 2. Receiver operating characteristic (ROC) curves comparing urine O. viverrini excretory secretory (OV-ES) assay method to the gold standard formalin ethyl-acetate concentration technique (FECT) (n = 235).

The ROC curve illustrates the diagnostic performance of antigen detection using the urine OV-ES assay compared to the gold standard FECT method. The assay had AUC of 0.8460. The ROC curve diagnostic sensitivity was modeled as included negative controls (OV negative and other infections) and diagnostic sensitivity were modeled as individuals who were OV+ with and without co-infections.

Fig 3. Cross reactivity with other helminthes of using the urine O. viverrini Excretory Secretory (OV-ES) assay with trichloroacetic acid (TCA) pre-treated urine.

Most helminth infections resulted in negative tests except for 2 of 56 S. stercoralis infections (3.57%) and 1 of 10 hookworm infections (10%). Data points shown are antigen levels (OD 492) of individuals infected with OV, O. viverrini; Ss, S. stercoralis; MIF, minute intestinal flukes; Hw, hookworms; T, Taenia; Ech, Echinostomes and Tt, Trichuris trichiura.