A rifampicin-free four-month TB regimen matched standard care in a small trial

A four-month, all-oral tuberculosis regimen that contains no rifampicin cleared the bacterium from patients' sputum about as reliably as the standard six-month treatment in a small South African trial [s1]. In the 122-patient proof-of-concept study, sputum culture conversion by the end of treatment reached 96.0% across the shortened regimen's dose groups against 100.0% for standard care, a gap of 4.0 percentage points that fell within the margin the trialists had set for calling the short course no worse [s1].

The regimen, abbreviated DBQ, combines two newer anti-tuberculosis drugs — delamanid and bedaquiline — with an investigational one, quabodepistat [s1]. Its interest is not simply that it is short. Drug-susceptible tuberculosis has been treatable in four months since a landmark 2021 trial showed a regimen built around high-dose rifapentine with moxifloxacin was non-inferior to the six-month standard [s2]. What DBQ offers, if it holds up, is a four-month course that leaves out the rifamycin class entirely — the family that includes rifampicin and rifapentine and that interacts heavily with other medicines, most importantly the antiretrovirals millions of people with HIV and tuberculosis take together [s1].

What the trial did

This was an open-label, randomised, proof-of-concept, non-inferiority, phase 2b/c trial at six clinical sites in South Africa, sponsored by Otsuka, which is developing quabodepistat [s1]. Adults aged 18 to 65 with newly diagnosed, drug-susceptible pulmonary tuberculosis were randomly assigned in a 1:2:2:1 ratio to one of four groups: delamanid (300 mg once daily) and bedaquiline (400 mg daily for two weeks, then 200 mg three times a week) plus quabodepistat at 10 mg, 30 mg or 90 mg once daily for four months; or the six-month standard regimen of rifampicin, isoniazid, ethambutol and pyrazinamide, known as RHEZ [s1]. Every drug was taken by mouth [s1].

The main question was whether the short regimen cleared the infection as well as the standard one. The primary efficacy endpoint was the proportion of patients whose sputum cultures turned negative by the end of treatment, with non-inferiority declared if the pooled DBQ groups fell no more than 12 percentage points short of RHEZ, judged against a two-sided 80% confidence interval [s1]. Between April 2022 and May 2024, 306 people were screened and 122 enrolled: 20 to the 10 mg group, 42 to 30 mg, 39 to 90 mg and 21 to RHEZ; 121 who received at least one dose formed the main analysis set [s1].

What it found

Culture conversion was high across the board. It reached 100.0% in the 10 mg group (all 20 patients; 95% CI 83.2 to 100.0), 92.9% at 30 mg (39 of 42; 80.5 to 98.5) and 97.4% at 90 mg (37 of 38; 86.2 to 99.9), for a pooled DBQ figure of 96.0% (96 of 100; 90.1 to 98.9), against 100.0% for RHEZ (all 21; 83.9 to 100.0) [s1]. The pooled comparison met the non-inferiority bar, with a difference of −4.0 percentage points (80% CI −7.4 to 3.4) [s1].

Safety was mostly unremarkable, with no serious adverse events or discontinuations attributed to the study drugs, though the newer combination was not entirely benign [s1]. Grade 3 or higher adverse events were more common in the DBQ groups — 20% (four of 20) at 10 mg, 14% (six of 42) at 30 mg and 11% (four of 38) at 90 mg — than in the 5% (one of 21) seen with standard care, and 105 of 121 patients (87%) had at least one treatment-emergent adverse event [s1]. One patient, in the 90 mg group, died after worsening tuberculosis with pneumonia, which the investigators judged unrelated to treatment [s1].

How to read it

The result is promising and clearly preliminary — and the trial's own design says so. "Proof of concept" and "phase 2b/c" mean this was an early test of whether the combination is worth taking forward, not a definitive one [s1]. Three limits matter most. It was small: 122 patients, with as few as 20 in some groups, so the confidence intervals around each conversion rate are wide, and the headline non-inferiority rested on an 80% confidence interval, a looser standard than the 95% used for confirmatory trials [s1]. It was open-label, so everyone knew which regimen a patient was taking, though the laboratory staff reading the cultures did not [s1]. And, most important, the endpoint was a surrogate: clearing bacteria from sputum by the end of treatment predicts cure but is not the same as it, because tuberculosis can relapse months after treatment stops [s1]. The measure that decides whether a short regimen truly works — sustained cure without relapse over a year or more — is what a larger phase 3 trial would have to show.

Charted against standard care, the pooled short regimen sits just below it, with overlapping intervals; the trial's claim is that the two are close enough, not that the short course is better [s1]. Related coverage has examined an experimental tuberculosis vaccine tested head-to-head against BCG in newborns, how genomic sequencing traced tuberculosis transmission inside Brazilian prisons, and a post-war prevalence survey that measured the true tuberculosis burden in Tigray.

What to watch

The case for a rifamycin-free short regimen rests on the people for whom rifampicin is most awkward — those on antiretrovirals, and others taking medicines that the rifamycins disrupt [s1]. Whether DBQ delivers on that promise now depends on a larger, longer trial with relapse-free cure as the endpoint, and on which of the three quabodepistat doses is carried forward [s1]. Until then, the six-month standard, and the rifapentine-based four-month regimen already endorsed by guidelines, remain the proven options [s1][s2].

This article describes research and is not medical advice. Tuberculosis treatment is a decision for patients and their clinicians, following national guidelines.

Sources

• Efficacy and safety of a 4-month quabodepistat, delamanid, and bedaquiline regimen for drug-susceptible pulmonary tuberculosis: a multicentre, open-label, randomised, proof-of-concept, non-inferiority, phase 2b/c trial — The Lancet Infectious Diseases, 29 May 2026
• Four-Month Rifapentine Regimens with or without Moxifloxacin for Tuberculosis — New England Journal of Medicine, 6 May 2021