Utility of interferon-gamma ELISPOT assay responses in highly tuberculosis-exposed patients with advanced HIV infection in South Africa

Abstract

Background: Interferon-gamma (IFN-gamma) ELISPOT assays incorporating Mycobacterium tuberculosis-specific antigens are useful in the diagnosis of tuberculosis (TB) or latent infection. However, their utility in patients with advanced HIV is unknown. We studied determinants of ELISPOT responses among patients with advanced HIV infection (but without active TB) living in a South African community with very high TB notification rates.

Methods: IFN-gamma responses to ESAT-6 and CFP-10 in overnight ELISPOT assays and in 7-day whole blood assays (WBA) were compared in HIV-infected patients (HIV+, n = 40) and healthy HIV-negative controls (HIV-, n = 30) without active TB. Tuberculin skin tests (TSTs) were also done.

Results: ELISPOTs, WBAs and TSTs were each positive in >70% of HIV- controls, reflecting very high community exposure to M. tuberculosis. Among HIV+ patients, quantitative WBA responses and TSTs (but not the proportion of positive ELISPOT responses) were significantly impaired in those with CD4 cell counts <100 cells/mul compared to those with higher counts. In contrast, ELISPOT responses (but not WBA or TST) were strongly related to history of TB treatment; a much lower proportion of HIV+ patients who had recently completed treatment for TB (n = 19) had positive responses compared to those who had not been treated (11% versus 62%, respectively; P < 0.001). Multivariate analysis confirmed that ELISPOT responses had a strong inverse association with a history of recent TB treatment (adjusted OR = 0.06, 95%CI = 0.10-0.40, P < 0.01) and that they were independent of CD4 cell count and viral load. Among HIV+ individuals who had not received TB treatment both the magnitude and proportion of positive ELISPOT responses (but not TST or WBA) were similar to those of HIV-negative controls.

Conclusion: The proportion of positive ELISPOT responses in patients with advanced HIV infection was independent of CD4 cell count but had a strong inverse association with history of TB treatment. This concurs with the previously documented low TB risk among patients in this cohort with a history of recent treatment for TB. These data suggest ELISPOT assays may be useful for patient assessment and as an immuno-epidemiological research tool among patients with advanced HIV and warrant larger scale prospective evaluation.

Figure 1

Proportion of positive IFN-γ ELISPOT responses and 7-day IFN-γ whole blood assay (WBA) responses to ESAT-6 and/or CFP-10 (RD1) and purified protein derivative (PPD) among HIV- controls (n = 30) and HIV+ patients (n = 40) subdivided into those with (n = 19) or without (n = 21) a history of previous tuberculosis treatment (TB Rx). ELISPOT assay responses in HIV+ patients were strongly associated with history of TB treatment. *In HIV+ patients who had previously received TB treatment, the proportions of positive responses to RD1 and PPD antigens were significantly lower using the ELISPOT assay than the WBA (P < 0.05 for each comparison). ns = not significant.

Figure 2

Proportion of positive ELISPOT IFN-γ responses to M. tuberculosis-specific antigens (ESAT-6 and/or CFP-10) among HIV- controls (n = 30) compared to HIV+ patients (n = 40) stratified by CD4 cell count >100 cells/μl (n = 23) or <100 cells/μl (n = 17) and according to history of recent TB treatment (n = 19) or no history of such treatment (n = 21). ns = not significant