Novel N Chegou; Kim GP Hoek; Magdalena Kriel; Robin M Warren; Thomas C Victor; Gerhard Walzl Disclosures Expert Rev Anti Infect Ther. 2011;9(4):457-469.
In This ArticleSerological diagnosis of TB disease has been practiced since 1898 (cited in [135] ). While the initial agglutination tests evaluated in the 1890s made use of crude cell preparations of M. tuberculosis or M. bovis BCG, the serological tests that are currently being developed and evaluated often employ highly purified native or recombinant antigens. Antigens that are commonly used in these assays include the 38 kDa antigen, antigen 60, lipoarabinomannan, members of the antigen 85 complex, MPT32 and MPT51, amongst others, and have previously been extensively reviewed. [136–138]
Serological tests, especially those in immunochromatographic format, are known to be rapid, easy-to-use and therefore highly suitable as point-of-care tests in resource-limited settings. However, results of meta-analysis of these tests have consistently revealed very poor accuracies. [65,139,140] In spite of the superior technical advantages that these tests offer over other diagnostic techniques, the poor accuracies of the currently available serological tests suggest that these assays do not make any meaningful contribution to the diagnosis of TB disease.
It has been hypothesized that the accuracies of serological tests could be enhanced by the use of combinations of antigens in addition to the detection of multiple classes of antibodies (IgG plus IgA and IgM). The unavailability of useful commercial tests using such approaches suggest that this concept may not be practical. There is evidence of strong serological recognition of diverse M. tuberculosis infection phase-specific antigens that are traditionally regarded as T-cell-based antigens. [141–143] A number of these antigens, including heat shock protein 65, have been shown to have some diagnostic potential in serodiagnostic assays, [142–144] but these results are yet to be confirmed by other investigators.