- Design
- Phase 2b/c, open-label, randomised, non-inferiority, proof-of-concept
- Population
- 122 adults in South Africa with drug-susceptible pulmonary TB
- Primary outcome
- Sputum culture conversion at end of treatment
- Effect
- 96.0% pooled new regimen vs 100% standard; difference −4.0% (80% CI −7.4 to 3.4)
This open-label phase 2b/c trial, published in The Lancet Infectious Diseases on 29 May 2026, randomised 122 adults in South Africa with drug-susceptible pulmonary TB to four months of delamanid, bedaquiline and the new agent quabodepistat at one of three doses, or six months of standard rifampicin, isoniazid, ethambutol and pyrazinamide.
At the end of treatment, sputum culture had converted in 96% of the pooled new-regimen groups and 100% of the standard group, meeting the 12% non-inferiority margin. Grade 3 or higher adverse events were more frequent in the new-regimen groups (11% to 20% against 5%), though none led to stopping.
Culture conversion at the end of treatment is a weak endpoint for TB; what matters is relapse after stopping, which this trial was not designed to show. It also uses bedaquiline and delamanid, the backbone of drug-resistant TB treatment, in drug-susceptible disease, which raises stewardship questions in high-burden countries such as India. This is a reason to follow quabodepistat, not to change practice.
- Continue standard six-month or approved four-month regimens for drug-susceptible TB; this is proof-of-concept only.
- Await relapse data before regarding any rifampicin-free four-month regimen as effective.
- Protect bedaquiline and delamanid for drug-resistant TB under national programme guidance.
- Monitor liver function and QT interval in any patient on bedaquiline-containing regimens.
Why it matters
A rifampicin-free short course would matter for drug interactions, but it would spend drugs reserved for resistant TB.
Don't overread it
End-of-treatment culture conversion does not show durable cure; relapse was not assessed.
The statistics, in plain English
Non-inferiority was tested with an 80% confidence interval, which is narrower and less demanding than the usual 95%, as is common in proof-of-concept trials. With only 21 patients on standard therapy, the comparison is imprecise.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for infectious diseases, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free