- 2026-03
- Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 32(3)
- Feng Sun
- Yang Li
- Yilin Zhang
- Cui Cai
- Yuanyuan Chen
- Hengzhong Yi
- Qianhong Wu
- Yuan Qian
- Hongying Yu
- Yuanbo Lan
- Jichan Shi
- Ya Huang
- Yungui Zhang
- Shun Feng
- Mingying Xiao
- Jing Wang
- Yiming Li
- Zebao He
- Haiqing Liu
- Yena Zhang
- Yu Chen
- Yong Zhou
- Jinlan Li
- Bin Chen
- Lin Xu
- Jianfeng Luo
- Yajiao Xing
- Xiaolin Wei
- Yanlin Zhao
- Wenhong Zhang
Study Design
- Type
- Randomized Controlled Trial (RCT)
- Sample size
- n = 660
- Population
- participants with rifampicin-resistant pulmonary tuberculosis
- Methods
- open-label, randomized noninferiority trial; 1:1 assignment to a 6- to 9-month all-oral regimen (levofloxacin, linezolid, cycloserine, clofazimine and/or pyrazinamide) and a 9-month injectable-containing control regimen
- Blinding
- Open-label
- Duration
- 6 to 9 months (all-oral), 9 months (control)
- Funding
- Unclear
Objectives
Despite advancements in shorter, oral regimens for drug-resistant tuberculosis, more options are needed to extent patient benefits. We conducted a trial to evaluate the noninferiority of an all-oral regimen compared with the injectable-containing regimen.Methods
In an open-label, randomized noninferiority trial in China, participants with rifampicin-resistant pulmonary tuberculosis were assigned 1:1 to a 6- to 9-month all-oral regimen (levofloxacin, linezolid, cycloserine, clofazimine and/or pyrazinamide) and a 9-month injectable-containing control regimen. The primary outcome was a favourable outcome at 84 weeks after treatment initiation, defined by two consecutive, negative cultures with no previous unfavourable outcome, using a noninferiority margin of 10%.Results
Between 2 June 2020, and 1 December 2021, 660 participants were enrolled and 354 underwent randomization. A total of 312 and 260 participants were included in the modified intention-to-treat and the per-protocol analysis, respectively. In the modified intention-to-treat analysis, 76.1% of participants in the oral group and 63.7% of those in the control group had a favourable outcome at 84 weeks (difference, 12.4%; 95% CI, 2.4-22.5%; noninferiority p < 0.0001). In the per-protocol analysis, 84.4% of participants in the oral group and 73.5% of participants in the control group had a favourable outcome (difference, 10.9%; 95% CI, 1.1-20.7%; noninferiority p < 0.0001). Grade 3 to 5 adverse events occurred in 69.1% of participants in the control group and 59.5% in the oral group. The most common grade 3 to 5 adverse events were QTcF (corrected QT interval calculated with Fridericia's formula) prolongation, affecting 44.6% of participants in the control group and 29.5% of participants in the oral group. Hepatobiliary disorder occurred more frequently in the control group (21.7%) compared with the oral group (7.5%).Conclusions
The all-oral regimen was noninferior to the 9-month injectable-containing regimen, offering an alternative for patients lacking access to bedaquiline, delamanid or pretomanid. However, its efficacy against the latest WHO-recommended bedaquiline-containing regimens requires further validation.