Interferon-gamma inducible protein 10 as a biomarker for active tuberculosis and latent tuberculosis infection in children: a case-control study.

BACKGROUND

Interferon-gamma (IFN-γ) release assays (IGRAs) are suboptimally sensitive to diagnose tuberculosis (TB) and latent TB infection (LTBI) in young children. In this study we compared Mycobacterium tuberculosis antigen-stimulated IFN-γ inducible protein 10 (IP-10) responses in children with active TB and LTBI to responses from children with non-tuberculous mycobacterial (NTM) lymphadenopathy and respiratory tract infection (RTI). We also assessed test agreement between IP-10 and the QuantiFERON(®)-TB Gold In-Tube (QFT-IT) test results, and investigated whether IP-10 release upon mitogen stimulation is associated with age.

METHODS

We recruited 48 children (median age 54 months) diagnosed in Germany with either active TB (n = 11), LTBI (n = 14), NTM lymphadenopathy (n = 8), or common RTI (n = 15). IFN-γ levels were measured using the QFT-IT. These plasma supernatants were used to determine IP-10 concentrations using an in-house enzyme-linked immunosorbent assay (ELISA).

RESULTS

The median antigen-stimulated IP-10 levels in children with active TB, LTBI, NTM lymphadenopathy, and RTI were 12,702 pg/ml, 9109 pg/ml, 97 pg/ml, and 84 pg/ml, respectively. We observed a strong correlation between IP-10 and IFN-γ plasma concentration in children with active TB and LTBI (r(2) = 0.69). Overall agreement between IP-10 and QFT-IT assays was high (kappa = 0.95). IP-10 levels after mitogen stimulation showed no association with age.

CONCLUSIONS

IP-10 and IFN-γ were both induced with antigen stimulation in blood from children in the TB and LTBI groups, in contrast to the NTM and RTI groups. Compared to IFN-γ the IP-10 levels were higher and IP-10 was released independently of age. IP-10 therefore may represent an additional biomarker in the paediatric population.