Lesion penetration predicts tuberculosis regimen success: a validated translational tool for de-risking late-stage drug development

Anti-tuberculosis (TB) drugs must reach the disease site to be effective, but the complex TB pathology limits drug distribution and efficacy, as bacteria in the necrotic center of lung cavities are difficult to access and kill. Our study generated the most comprehensive database of site-of-disease distribution for 24 drugs in a rabbit model of cavitary TB. We validated a lung lesion penetration model with clinical data and developed an integrated approach that combines microenvironment-specific pharmacokinetics and pharmacodynamics. Next, we simulated site-of-disease coverage of shortened regimens tested in major Phase 3 trials over the past decade, allowing direct comparisons with trial outcomes. We propose that regimens including at least two drugs achieving bactericidal concentrations in pulmonary cavities are more likely to meet primary clinical endpoints in patients with cavitary disease. Our translational platform constitutes a validated tool to support drug evaluation, smart regimen design, and de-risking of late-stage trials. One-sentence Summary: Translational site-of-disease pharmacokinetic-pharmacodynamic modeling quantitatively predicts lesion concentrations of anti-tuberculosis drugs in diverse lung pathology of patients and discerns Phase 3 efficacy outcomes of successful regimens from failed regimens.