endTB-Q: A phase III randomised clinical trial on pre-XDR tuberculosis
Context
Pre-extensively drug-resistant tuberculosis (pre-XDR-TB), that is tuberculosis resistant to rifampicin and fluoroquinolones, is difficult to treat. endTB-Q evaluated the efficacy and safety of a regimen comprising bedaquiline, delamanid, linezolid, and clofazimine (BDLC) compared to WHO-recommended longer individualized regimen (control) in people with pre-XDR-TB.
ENDTB-Q is a randomized, controlled, open-label, multi-country Phase III trial (Randomization (2:1 in favor of the experimental arm) is stratified by country and extent-of-TB-disease phenotype).
The objective is to Identify an effective, shorter, less toxic and injection-free regimen for FQ-resistant MDR-TB as compared to the current standard of care. The primary objective is to assess whether the efficacy of experimental 6 to 9 months regimen is non-inferior to that of the control at 73 weeks.
Our role
Statistics & Data Management
Partners
- Médecins Sans Frontières
- Partners In Health
- IRD
- Unitaid
- Ministries of Health of the countries involved in the study
Methods
endTB-Q was a randomized, controlled, open-label, non-inferiority, stratified Phase 3 trial conducted in six countries. Participants aged 15 years or older with pulmonary pre-XDR-TB were randomized 2:1, stratified by country and disease extent, to BDLC or the control (individualized, longer regimens). BDLC duration was 6 months, extended to 9 months for extensive disease or in case of positive culture at 2 months or later. The primary outcome was bacteriological favorable outcome or favorable bacteriological, radiological, and clinical evolution at 73 weeks post-randomization. We report the risk difference adjusted for stratification variables, with a non-inferiority margin of -12%, in the modified intention-to-treat (mITT) and per-protocol (PP) populations. We report serious adverse events and deaths in the safety population at 73 weeks.
Results
Between 2020 and 2023, 324 participants were randomized: 219 (68%) to BDLC and 105 (32%) to control. In mITT, extensive disease was present in 105/163 (64%) and 52/84 (62%) of BDLC and control groups, respectively. At 73 weeks, favorable outcomes were observed in 141/163 (87%) BDLC and 75/84 (89%) control participants in the mITT population (adjusted risk difference [aRD] mITT: 0∙2% [95%CI: -9∙1%;9∙5%]), and in 138/157 (88%) BDCL and 71/76 (93%) control participants in the PP population (aRD PP: -3∙5% [95% CI: -12∙8%,5∙9%]). In a prespecified subgroup analysis, the RD in mITT was -7∙5% [95% CI: -18∙3%,3∙2%] for extensive and 5∙6% [95% CI: -7∙6%,18∙8%] for limited disease. At 73 weeks in the safety population, at least one serious adverse event occurred in 42/213 (20%) BDLC and 23/105 (22%) control participants, including 8 (4%) and 2 (2%) all-cause deaths, respectively.
Conclusion
Although non-inferiority was not established, endTB-Q demonstrated excellent outcomes with a shortened BDLC regimen for pre-XDR-TB with limited disease. Accumulating evidence suggests that individuals with extensive pre-XDR-TB may require longer, reinforced regimens.
