TB Research

Predictive Markers of Incident Tuberculosis in Close Contacts in Brazil and India.

Betânia M F Nogueira, Francys Rangel, Alice M S Andrade, Evangeline Ann Daniel, Marina C Figueiredo, Cody Staats, Valeria C Rolla, Afrânio L Kritski, et al. (14 authors)

The Journal of infectious diseases · 2025-09

Abstract

There are insufficient predictors of progression to tuberculosis among contacts. A case-control study within RePORT-Brazil matched 20 QuantiFERON-positive progressors and 40 nonprogressors by sex, age, and exposure duration. Twentynine cytokines were measured using a Luminex assay with QuantiFERON-TB Gold Plus supernatants collected at baseline and evaluated using machine learning for tuberculosis prediction. The same markers were evaluated in 8 QuantiFERON-positive progressors and 12 nonprogressors from India. Interleukin 8, interleukin 10, and CCL3 levels predicted incident tuberculosis (area under the receiver operating characteristic curve, 0.75) in 2 years with sensitivity and specificity >80%, in both cohorts. This signature predicted tuberculosis progression in close contacts meeting World Health Organization goals.

MeSH terms

  • Humans
  • India
  • Brazil
  • Male
  • Female
  • Biomarkers
  • Adult
  • Tuberculosis
  • Case-Control Studies
  • Young Adult
  • Middle Aged
  • Cytokines
  • Incidence
  • Adolescent
  • Interleukin-8
  • Disease Progression
  • Machine Learning
  • ROC Curve