Immune activation from M. tuberculosis screening tests predicts mortality.

Abstract

Impaired immune responses are a key feature of aging; however, there are few laboratory tests that link these responses to clinical outcomes. Interferon-gamma release assays (IGRAs) for tuberculosis screening quantify release of interferon-gamma by T-cells, and the difference between unstimulated and mitogen-stimulated T-cells is assessed for test validity. We assess this measure's relationship with all-cause mortality. We obtained the most recent negative and indeterminate outpatient IGRAs from a large health system along with demographics, frailty, lymphocyte count, and inflammatory markers. We removed individuals on hemodialysis or immunosuppressive medications. We assessed the association of mitogen-nil with mortality at 6 months, 1 year, and 5 years by Kaplan-Meier analysis and Cox regression. Among 16,104 individuals, reported mitogen-nil ranged from < 0.01 to ≥ 10 IU/mL, and median (IQR) age was 64 (57, 72). Cumulative mortality (95% CI) at 5 years was estimated at 28% (23%-33%) for values 0-1 versus 19% (18%-19%) ≥ 10. In Cox regression, relative to values ≥ 10, values from 0-1 had hazard ratios for mortality at 6 months, 1 year, and 5 years of 2.77 (1.47-5.22), 2.22 (1.41-3.50), and 1.76 (1.31-2.36). Among those with data, adding lymphocyte count did not alter associations. Lower T-cell response to mitogen stimulation in IGRAs is associated with greater mortality. This common test may provide additional information to risk-stratify patients and as a phenotype of impaired immune response.