Interpretation of Mycobacterium tuberculosis antigen-specific IFN-γ release assays (T-SPOT.TB) and factors that may modulate test results

• Published in: The Journal of infection Vol. 55; no. 2; pp. 169 - 173
• Main Authors: Dheda, K, Pooran, A, Pai, M, Miller, R.F, Lesley, K, Booth, H.L, Scott, G.M, Akbar, A.N, Zumla, A, Rook, G.A
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01-08-2007; Elsevier
• Subjects: Adult; Antigens, Bacterial - immunology; Antigens, Bacterial - isolation & purification; Bacterial Proteins - immunology; Biological and medical sciences; CFP-10; ESAT-6; Female; Human; Humans; IFN-γ; Infectious Disease; Infectious diseases; Interferon-gamma - biosynthesis; Male; Medical sciences; Mycobacterium tuberculosis; South Africa; Tuberculosis; Tuberculosis, Pulmonary - drug therapy; Tuberculosis, Pulmonary - immunology; Tuberculosis, Pulmonary - metabolism; South Africa; Human; IFN-γ; Tuberculosis; CFP-10; ESAT-6; Antigen; Infection; Mycobacterium tuberculosis; Mycobacteriales; Mycobacteriaceae; Bacteria; Actinomycetes; Release
• ISSN: 0163-4453; 1532-2742
• DOI: 10.1016/j.jinf.2007.02.005

Summary Background Data about T cell antigen-specific (ESAT-6 and CFP-10) IFN-γ release assays (IGRAs) during and after completion of anti-tuberculous (TB) treatment are limited and highly discordant. Thus, the utility of IGRAs as a surrogate marker of mycobacterial burden remain unclear. Methods To investigate factors that modulate IGRA responses during anti-TB treatment we used a standardised assay (T-SPOT.TB) in 33 patients with culture positive tuberculosis. Results Significantly more patients in the early (≤4 months of anti-TB treatment) rather than the late phase (>4 months or completed anti-TB treatment) had positive IGRA responses [10/12 (83%) vs 4/21 (19%); p ≤ 0.01]. Thus, 17/21 (81%) in the late phase or who had completed treatment (mean duration of treatment = 8.7 months) were IGRA negative, despite having robust antigen-specific recall proliferative responses. In these 17 patients prolonged incubation (5 days vs overnight), use of different antigen preparations (protein vs peptide) and addition of endotoxin, failed to elicit positive responses. Conclusions In treated TB patients the discordant IGRA data remain unexplained by variation in laboratory protocols and are more likely due to host or environmental factors. In a low burden setting IGRAs may be a promising surrogate marker of mycobacterial disease burden.