Risk stratification of latent tuberculosis defined by combined interferon gamma release assays

Abstract

Background: Most individuals infected with Mycobacterium tuberculosis develop latent tuberculosis infection (LTBI). Some may progress to active disease and would benefit from preventive treatment yet no means currently exists to predict who will reactivate. Here, we provide an approach to stratify LTBI based on IFN-γ responses to two antigens, the recombinant Early-Secreted Antigen Target-6 (rESAT-6) and the latency antigen Heparin-Binding Haemagglutinin (HBHA).

Methods: We retrospectively analyzed results from in-house IFN-γ-release assays with HBHA (HBHA-IGRA) and rESAT-6 (rESAT-6-IGRA) performed during a 12-year period on serial blood samples (3 to 9) collected from 23 LTBI subjects in a low-TB incidence country. Both the kinetics of the absolute IFN-γ concentrations secreted in response to each antigen and the dynamics of HBHA/rESAT-6-induced IFN-γ concentrations ratios were examined.

Results: This analysis allowed the identification among the LTBI subjects of three major groups. Group A featured stable HBHA and rESAT-6-IGRA profiles with an HBHA/rESAT-6 ratio persistently higher than 1, and with high HBHA- and usually negative rESAT-6-IGRA responses throughout the study. Group B had changing HBHA/rESAT-6 ratios fluctuating from 0.0001 to 10,000, with both HBHA and rESAT-6 responses varying over time at least once during the follow-up. Group C was characterized by a progressive disappearance of all responses.

Conclusions: By combining the measures of IFN-γ concentrations secreted in response to an early and a latency antigens, LTBI subjects can be stratified into different risk groups. We propose that disappearing responses indicate cure, that persistent responses to HBHA with HBHA/rESAT-6 ratios ≥ 1 represent stable LTBI subjects, whereas subjects with ratios varying from >1 to <1 should be closely monitored as they may represent the highest-risk group, as illustrated by a case report, and should therefore be prioritized for preventive treatment.

Figure 1. Definition of the heat map color and of the score values for HBHA- and rESAT-6-IGRA.

The first column shows the intensity of the colors used in the heat map to represent the intensity of the IFN-γ responses, from dark blue for insignificant IFN-γ responses to dark red for very high responses. The last column defines the IFN-γ scores used to calculate the HBHA/rESAT-6 ratios. These scores were attributed to the different ranges of IFN-γ concentrations illustrated in the heat map. Values corresponding to the grey zone just below or above the cut-off values (dotted line), were attributed the same score of 1.

Figure 2. LTBI subjects displaying stable HBHA- and rESAT-6-IGRA profiles to mycobacterial antigens over time (group A).

HBHA- and rESAT-6-IGRAs were performed on 12 LTBI subjects, numbered from 1 to 12, at different time points between the years 2000 and 2011, as indicated. When indicated, the numbers 1 and 2 correspond to tests performed twice, i. e. in the first and second part of the year, respectively. A heat-map is represented on the left panel, with positive HBHA- and rESAT-6-IGRA results shown in red and negative results shown in blue as defined in Figure 1. The right panel represents ratios between HBHA- and rESAT-6-induced IFN-γ concentrations. White boxes correspond to time points for which no data was available for the particular antigen.

Figure 3. LTBI subjects displaying unstable HBHA- and rESAT-6-IGRA profiles to mycobacterial antigens over time (group B).

HBHA- and rESAT-6-IGRAs were performed on 7 LTBI subjects, numbered from 13 to 19, at different time points between the years 2000 and 2011, as indicated. Heat-map illustrating the intensity of IFN-γ responses over time, and ratios between HBHA- and rESAT-6-induced IFN-γ concentrations are represented in panel A and B, respectively. When indicated, the numbers 1 and 2 correspond to tests performed twice, i. e. in the first and second part of the year, respectively. White boxes correspond to time points for which no data was available for the particular antigen.

Figure 4. LTBI subjects displaying disappearing HBHA- and rESAT-6-IGRA responses to mycobacterial antigens over time (group C).

HBHA- and rESAT-6-IGRAs were performed on 4 LTBI subjects, numbered from 20 to 23, at different time points between the years 2001 and 2011, as indicated. Heat-map illustrating the intensity of IFN-γ responses over time, and ratios between HBHA- and rESAT-6-induced IFN-γ concentrations are represented in panel A and B, respectively. When indicated, the numbers 1 and 2 correspond to tests performed twice, i. e. in the first and second part of the year, respectively. White boxes correspond to time points for which no data was available for the particular antigen.

Figure 5. HBHA- and rESAT-6-IGRA kinetics of LTBI subjects representative of group A, B and C.

IFN-γ secretion was measured by ELISA after four days of PBMC stimulation with PPD (black circle), HBHA (violet circle) or rESAT-6 (green circle) at the indicated time points. The limits of positivity for PPD, HBHA and rESAT-6 are represented as dotted lines with the corresponding color. When available, the negative or positive results of the QFT-G-IT are shown on the bottom of the figure as “−” or “+”, respectively. Subject shown on panel A belongs to group A (n°1) characterized by stable HBHA- and rESAT-6-IGRA responses. Subject shown on panel B belongs to group B (n°14) characterized by unstable HBHA- and rESAT-6-IGRA responses. Subject shown on panel C belongs to group C (n°23) presenting disappearing HBHA- and rESAT-6-IGRA responses.

Figure 6. Spectrum of immune responses during LTBI.

We propose to relate the spectrum of LTBI to the IFN-γ responses to two different mycobacterial antigens, HBHA represented in the left arrow, and rESAT-6 represented in the right arrow. The intensity of the colors depicts the intensity of IFN-γ responses with a scale described in Figure 1 (Adapted from Young et al. [10]).