Evaluation of the CCA Immuno-Chromatographic Test to Diagnose Schistosoma mansoni in Minas Gerais State, Brazil

BACKGROUND: The Kato-Katz (KK) stool smear is the standard test for the diagnosis of Schistosoma mansoni infection, but suffers from low sensitivity when infections intensities are moderate to low. Thus, misdiagnosed individuals remain untreated and contribute to the disease transmission, thereby forestalling public health efforts to move from a modality of disease control to one of elimination. As an alternative, the urine-based diagnosis of schistosomiasis mansoni via the circulating cathodic antigen immuno-chromatographic test (CCA-ICT) has been extensively evaluated in Africa with the conclusion that it may replace the KK test in areas where prevalences are moderate or high. METHODS AND FINDINGS: The objective was to measure the performance of the CCA-ICT in a sample study population composed of residents from non-endemic and endemic areas for schistosomiasis mansoni in two municipalities of Minas Gerais state, Brazil. Volunteers (130) were classified into three infection status groups based on duplicate Kato-Katz thick smears from one stool sample (2KK test): 41 negative individuals from non-endemic areas, 41 negative individuals from endemic areas and 48 infected individuals from endemic areas. Infection status was also determined by the CCA-ICT and infection exposure by antibody ELISA (enzyme-linked immunosorbent assay) to S. mansoni soluble egg antigen (SEA) and soluble (adult) worm antigen preparation (SWAP). Sensitivity and specificity were influenced by whether the trace score visually adjudicated in the CCA-ICT was characterized as positive or negative for S. mansoni infection. An analysis of a two-graph receiver operating characteristic was performed to change the cutoff point. When the trace score was interpreted as a positive rather than as a negative result, the specificity decreased from 97.6% to 78.0% whereas sensitivity increased from 68.7% to 85.4%. A significantly positive correlation between the CCA-ICT scores and egg counts was identified (r = 0.6252, p = 0.0001). However, the CCA-ICT misdiagnosed as negative 14.6% of 2KK positive individuals, predominantly those with light infections (fewer than 100 eggs/g feces). Considering 2KK as reference test, the discriminating power of the CCA-ICT (the area under the curve [AUC] = 0.817) was greater than the SEA-ELISA (AUC = 0.744) and SWAP-ELISA (AUC = 0.704). CONCLUSION: Our data for the performance of the CCA-ICT in the Brazilian communities endemic for schistosomiasis mansoni support those from Africa, i.e., in areas with greater infection prevalence and intensities, the CCA-ICT may be useful as a tool to indicate community-based preventative chemotherapy without individual diagnosis. However, because of the Brazilian Ministry of Health’s recommendation for individual diagnosis in areas where prevalence is less than 15%, i.e., those areas in which infection intensities are likely to be lowest, the CCA-ICT lacks the sensitivity to be used as standalone diagnostic tool.