Potential of host markers produced by infection phase-dependent antigen-stimulated cells for the diagnosis of tuberculosis in a highly endemic area

Abstract

Background: Recent interferon gamma (IFN-γ)-based studies have identified novel Mycobacterium tuberculosis (M.tb) infection phase-dependent antigens as diagnostic candidates. In this study, the levels of 11 host markers other than IFN-γ, were evaluated in whole blood culture supernatants after stimulation with M.tb infection phase-dependent antigens, for the diagnosis of TB disease.

Methodology and principal findings: Five M.tb infection phase-dependent antigens, comprising of three DosR-regulon-encoded proteins (Rv2032, Rv0081, Rv1737c), and two resucitation promoting factors (Rv0867c and Rv2389c), were evaluated in a case-control study with 15 pulmonary TB patients and 15 household contacts that were recruited from a high TB incidence setting in Cape Town, South Africa. After a 7-day whole blood culture, supernatants were harvested and the levels of the host markers evaluated using the Luminex platform. Multiple antigen-specific host markers were identified with promising diagnostic potential. Rv0081-specific levels of IL-12(p40), IP-10, IL-10 and TNF-α were the most promising diagnostic candidates, each ascertaining TB disease with an accuracy of 100%, 95% confidence interval for the area under the receiver operating characteristics plots, (1.0 to 1.0).

Conclusions: Multiple cytokines other than IFN-γ in whole blood culture supernatants after stimulation with M.tb infection phase-dependent antigens show promise as diagnostic markers for active TB. These preliminary findings should be verified in well-designed diagnostic studies employing short-term culture assays.

Figure 1. Host markers detected in ESAT-6/CFP-10-stimulated whole blood cell cultures in TB and no-TB cases and receiver operating characteristics (ROC) plots showing the accuracies of these markers in discriminating between TB disease and no-TB.

Only analytes that discriminated between TB and no-TB disease with AUC ≥0.70 are shown. Error bars in the scatter dot plots represent the median analyte levels. AUC  =  Area under the ROC curve.

Figure 2. Host markers detected in Rv2389c stimulated whole blood cell cultures of TB and no-TB cases and receiver operating characteristics (ROC) plots showing the accuracies of these markers in discriminating between TB disease and no-TB.

Representative plots for analytes that discriminated between TB and no-TB disease with AUC ≥0.70, are shown. Error bars in the scatter plots represent the median analyte levels. AUC  =  Area under the ROC curve.

Figure 3. Host markers detected in Rv0081-stimulated whole blood cell cultures of TB and no-TB cases and receiver operating characteristics (ROC) plots showing the accuracies of these markers in discriminating between TB disease and no-TB.

Representative plots for the most potentially useful analytes are shown. The unstimulated control levels of the markers with negative responses were higher than the analyte levels detected in Rv0081 stimulated culture supernatants. Error bars in the scatter plots represent the median analyte levels. AUC  =  Area under the ROC curve.

Figure 4. Host markers detected in Rv2032-stimulated whole blood cell cultures of TB and no-TB cases and receiver operating characteristics (ROC) plots showing the accuracies of these markers in discriminating between TB disease and no-TB.

Representative plots for analytes that discriminated between TB and no-TB disease with AUC ≥0.70 are shown. The unstimulated control levels of the markers with negative responses were higher than the analyte levels detected in Rv2032 stimulated culture supernatants. Error bars in the scatter plots represent the median analyte levels. AUC  =  Area under the ROC curve.

Figure 5. Host markers detected in Rv1737c-stimulated whole blood cell cultures and receiver operating characteristics (ROC) plots showing the accuracies of these markers in discriminating between TB disease and no-TB.

Only the top 4 Rv1737c-specific analytes are shown. Error bars in the scatter dot plots represent the median analyte levels. AUC  =  Area under the ROC curve.

Figure 6. Number of times each analyte was included in the best 20 general discriminant analysis models that best classified the TB cases and HHCs into their respective groups.

A, Analytes detected in the unstimulated and ESAT-6/CFP-10 stimulated supernatants. B, Analytes detected in Rv0867c-, Rv2389c-, Rv2032-, Rv0081- and Rv1737c-stimulated supernatants.