Treatment failure and relapse may affect many tuberculosis (TB) patients who undergo standard anti-TB therapy. Several independent studies suggested unsuccessful sputum culture conversion at month 2 of treatment (slow response) as risk factor for treatment failure and relapse. However, earlier than month 2 identification of patients with a high risk for poor treatment outcome would offer significant clinical trial and individual patient care benefits.

The sensitivity and specificity of serological IgG and IgA responses against four recombinant mycobacterial antigens (ABC transporter PstS3, secreted l-alanine dehydrogenase, culture filtrate protein Tpx and 6 kDa early secretory antigenic target esxa (ESAT-6)) were evaluated separately in a blinded fashion in 21 smear-positive pulmonary TB patient sera taken at diagnosis before commencement of directly observed anti-TB treatment short course comprising 13 slow responder and eight fast responder subjects.

We observed a general pattern of higher antibody levels in sera of slow responders. Most pronounced were high levels of anti-alanine dehydrogenase IgG, anti-Tpx IgG, anti-ESAT-6 IgG and anti-ESAT-6 IgA antibodies at diagnosis being associated with slow response with 100% specificity each and 46.2, 53.8, 53.8 or 53.8% sensitivity, respectively, when compared to fast response (P = 0.020, 0.021, 0.040 and 0.011, respectively). Discriminant analysis showed that the combined use of anti-Tpx IgG and anti-ESAT-6 IgA antibody titers before treatment predicted slow responders with 90.5% accuracy.

These preliminary results suggest that combinations of serodiagnostic markers measured prior to initiation of treatment may be suitable for the prediction of early treatment response. This approach holds promise and requires further evaluation for its utility in the prediction of treatment failure and relapse, the evaluation of new TB therapeutics, as well as in the care of individual patients.