Background Methods which use Mycobacterium tuberculosis (Mtb)-specific antigens to measure IFN-γ responses (IFN-γ release assays (IGRA)) have been useful in detecting Mtb infection in exposed individuals. We assessed infections in TB cases and their exposed household contacts (HHC) using an in-house optimised IGRA, the QuantiFERON-TB Gold in Tube (QFT-GIT) and QFT-Plus (QFT+).

Methods For the in-house IGRA, we analysed 266 active TB patients and 759 HHC (256 tuberculin skin test-positive and 503 test-negative, TST+ and TST- respectively) at baseline and 6 months. In a separate study we assessed QFT-GIT and QFT-plus responses using samples from 72 TB cases and 69 HHC at baseline. QFT-GIT has 3 Mtb-specific antigens: ESAT6, CFP10 and TB7.7 while QFT-plus has long and short peptides of ESAT-6 and CFP-10, designed to induce CD4+ and CD8+T cell responses respectively.

Results IFN-γ responses were lowest in TST- compared to both TST+ and TB patients at baseline (p<0.0001 for both), with 32% IGRA-positive compared to 76% and 73%, respectively using in-house IGRA. HHC sleeping in the same room with TB patient had a significantly higher IGRA conversion rate by 6 months compared to those sleeping further away (p=0.0004). We also observed a significant decline in IGRA IFN-γ levels by 6 months of TB treatment (p<0.0001). Among QFT-positive TB patients, smear-positive was 57% and culture-positive was 62%. The IFN-γ concentration between TB1 and TB 2 was similar, while, QFT-GIT had a significantly higher response than TB 1 and TB 2 for both TB patients (p=0.003) and HHC (p=0.0005).

Conclusion Our findings show that IGRA conversion is significantly increased in HHC with highest exposure but that IGRA -positive cannot predict risk of progression to active TB. We also found that QFT-GIT was quantifiably better than QTF-plus in our setting, limiting the ‘grey zone’ of indeterminate results.