Abstract
Tuberculosis (TB) treatment is highly effective, but response to therapy varies by geography and population subgroups. We assessed differences in TB treatment response in a representative and heterogeneous Brazilian population. We estimated genetic ancestry according to major genetic ancestry groups (African, European, and Amerindian) in the Regional Prospective Observational Research in Tuberculosis (RePORT)-Brazil cohort using ADMIXTURE software. RePORT-Brazil is an observational prospective cohort study of individuals with newly-diagnosed, culture-confirmed, pulmonary TB. Outcomes attributed to TB treatment included Grade 2 or higher adverse drug reaction (ADR), Grade 3 or higher ADR, hepatic ADR, and failure/recurrence. Genetic ancestry proportions were evaluated as predictors in univariate and multivariable logistic regression models for each outcome, and in stratified models for each genetic ancestry group. There were 930 pulmonary TB patients included in this study. In multivariable models we observed a decreased risk of Grade 2 + ADR when African ancestry proportion increased by 10% (Odds Ratio [OR] 0.41, 95% Confidence Interval [CI] 0.20-0.85) and an increased risk for Grade 2 + ADR with increasing European genetic ancestry (OR 2.33, 95% CI 1.14-4.76). In secondary analyses evaluating the interaction of HIV and genetic ancestry, we observed a statistically significant interaction between HIV and African genetic ancestry, but significantly decreased risk for Grade 2 + ADR with increasing African ancestry proportion. There were no associations with Amerindian genetic ancestry or for other treatment outcomes. In this Brazilian TB cohort, increased toxicity risk was associated with decreased African and increased European ancestry proportion.