New England Journal of Medicine SEPTEMBER 07, 2014

Four-Month Moxifloxacin-Based Regimens for Drug-Sensitive Tuberculosis

Stephen H. Gillespie, et al.

Bottom Line

The REMoxTB trial demonstrated that substituting either isoniazid or ethambutol with moxifloxacin in a four-month treatment regimen for drug-sensitive pulmonary tuberculosis was not non-inferior to the standard six-month regimen due to higher rates of relapse.

Key Findings

1. The primary outcome of unfavorable status (failure or relapse) at 18 months was observed in approximately 15% of patients in the isoniazid-substituted arm and 20% in the ethambutol-substituted arm, compared with 9% in the standard control arm.
2. The study failed to demonstrate the non-inferiority of the four-month moxifloxacin-containing regimens to the standard six-month regimen, as the lower bounds of the confidence intervals for the treatment differences exceeded the pre-specified non-inferiority margin of 6%.
3. Moxifloxacin-containing regimens demonstrated enhanced early bactericidal activity, with patients achieving culture-negative status significantly faster than those in the control group (P<0.01).
4. Safety profiles were favorable, with no significant differences in the incidence of grade 3 or 4 adverse events between the experimental and control arms.

Study Design

Design
Phase 3 RCT
Double-Blind
Sample
1,931
Patients
Duration
18 mo
Median
Setting
Multicenter, 9 countries
Population Adult patients with newly diagnosed, previously untreated, sputum smear-positive, pulmonary tuberculosis with rifampin-susceptible strains.
Intervention Two four-month regimens: one substituting moxifloxacin for isoniazid and the other substituting moxifloxacin for ethambutol.
Comparator Standard six-month regimen (2 months of isoniazid, rifampin, pyrazinamide, and ethambutol, followed by 4 months of isoniazid and rifampin).
Outcome Proportion of patients with bacteriologically or clinically defined failure or relapse during the 18 months following randomization.

Study Limitations

The study was specifically designed to demonstrate non-inferiority for a treatment duration reduction, and the higher relapse rate in the experimental arms precluded shortening the standard of care.
The trial was conducted in multiple global settings with varying baseline characteristics and health system resources, which may introduce heterogeneity, though no evidence of significant regional response differences was found.
The study design required adherence to a strict protocol, and while adherence was high, real-world application of such regimens might face different challenges.

Clinical Significance

While the trial did not achieve its primary goal of reducing treatment duration to four months, it established moxifloxacin as a safe and potent agent for TB therapy. The results underscored the difficulty of shortening TB treatment using a single drug substitution and emphasized the need for novel regimens combining multiple new agents to achieve shorter, effective therapy.

Historical Context

At the time of the study, the standard treatment for drug-sensitive TB—a six-month regimen of isoniazid, rifampin, pyrazinamide, and ethambutol—had remained largely unchanged for decades. The REMoxTB trial represented a landmark, large-scale Phase 3 global effort to address the urgent clinical and public health need for shorter, more manageable treatment regimens to improve patient compliance and reduce the burden on healthcare systems.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Moxifloxacin is a potent bactericidal agent against Mycobacterium tuberculosis. Based on its mechanism of action, why was it hypothesized that replacing ethambutol with a fluoroquinolone could potentially shorten the standard 6-month treatment duration?

Key Response

Moxifloxacin inhibits DNA gyrase, which is essential for DNA replication. Unlike ethambutol, which is primarily bacteriostatic and inhibits cell wall synthesis, fluoroquinolones like moxifloxacin have high bactericidal activity against both actively dividing and semi-dormant bacilli. The hypothesis was that this superior killing power would sterilize lesions faster, allowing for a 4-month cure.

Resident
Resident

In the REMoxTB trial, the 4-month moxifloxacin-based regimens failed to meet non-inferiority criteria compared to the standard 6-month regimen. If you have a patient with drug-sensitive TB who is struggling with the 6-month duration, can you substitute moxifloxacin for isoniazid or ethambutol to safely stop treatment at 4 months?

Key Response

No. The trial demonstrated that substituting moxifloxacin into the standard backbone (using rifampin) resulted in significantly higher rates of relapse compared to the 6-month control. Clinicians should not shorten drug-sensitive TB treatment to 4 months based on moxifloxacin substitution alone; a 6-month duration remains the standard unless using the specific rifapentine-moxifloxacin 4-month regimen validated in later trials (e.g., Study 31).

Fellow
Fellow

The REMoxTB trial showed that while culture conversion happened faster in the moxifloxacin arms, the relapse rates were higher. What does this discrepancy reveal about the limitations of using 'time to culture conversion' as a surrogate endpoint for 'stable cure' in tuberculosis drug development?

Key Response

This discrepancy highlights that early bactericidal activity (killing actively replicating bacteria) does not necessarily correlate with the sterilization of 'persister' populations. Persisters are metabolically distinct and require prolonged exposure to drugs like rifampin. REMoxTB proved that faster sputum conversion does not guarantee the elimination of these dormant niches, necessitating long-term follow-up (18 months or more) to confirm non-inferiority in relapse-free survival.

Attending
Attending

Reflecting on the failure of REMoxTB to shorten treatment using rifampin and moxifloxacin, contrasted with the success of the subsequent Study 31 (using rifapentine and moxifloxacin), what does this suggest about the pharmacokinetic requirements for shortening TB therapy?

Key Response

It suggests that shortening therapy requires more than just adding a potent fluoroquinolone; it likely requires a 'stronger' rifamycin backbone. The success of Study 31 relied on the long half-life and high-dose exposure of rifapentine combined with moxifloxacin. REMoxTB showed that standard-dose rifampin (10mg/kg) combined with moxifloxacin is insufficient to prevent relapse if the duration is reduced to 17 weeks.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The REMoxTB trial utilized a non-inferiority margin of 6 percentage points. Critically evaluate how the selection of this margin and the distribution of 'unfavorable' outcomes (including deaths and dropouts) influence the statistical robustness of the study's conclusions regarding the experimental arms.

Key Response

A 6% margin is relatively wide in TB research, where the standard control relapse rate is often <5%. Because the experimental arms in REMoxTB showed an absolute difference in unfavorable outcomes that exceeded this margin (up to 10% vs 15-17%), the trial clearly demonstrated inferiority. However, the inclusion of non-adherent patients and those who died from non-TB causes as 'unfavorable' in ITT analyses can sometimes mask drug efficacy; notably, the higher relapse rate in the per-protocol population confirmed the biological failure of the 4-month regimen.

Journal Editor
Journal Editor

As a peer reviewer for a high-impact journal, if you were presented with the REMoxTB data today, how would you address the ethical and editorial significance of publishing a 'negative' trial that failed its primary non-inferiority endpoint?

Key Response

The editorial significance is immense because it prevents the global adoption of an ineffective shortened regimen that would have led to an epidemic of relapses and potentially increased fluoroquinolone resistance. A tough reviewer would flag the high rates of missing data and 'unclassifiable' outcomes (patients lost to follow-up), demanding rigorous sensitivity analyses to ensure that the finding of inferiority wasn't merely a result of trial conduct issues in high-burden settings.

Guideline Committee
Guideline Committee

Current WHO and CDC guidelines now include a 4-month regimen for drug-sensitive TB, but they specify the use of high-dose rifapentine and moxifloxacin. How did the REMoxTB trial findings influence these recommendations, and why did it not lead to a recommendation for moxifloxacin substitution in standard rifampin-based therapy?

Key Response

REMoxTB serves as the definitive evidence against shortening standard rifampin-based (HRZE) therapy by simply substituting moxifloxacin. Guideline committees (like those for the 2022 WHO update) used REMoxTB to establish that 6 months of HRZE remains the minimum for rifampin-based therapy. The committee only recommends 4-month regimens when the specific pharmacodynamic profile of rifapentine is utilized, as seen in Study 31/ACTG A5349, rather than the moxifloxacin-rifampin combinations tested in REMoxTB.

Clinical Landscape

Noteworthy Related Trials

2014

REMoxTB Trial

n = 1,931 · NEJM

Tested

Moxifloxacin-substituted 4-month regimens

Population

Patients with smear-positive pulmonary tuberculosis

Comparator

Standard 6-month regimen

Endpoint

Status at 18 months post-randomization (culture-negative)

Key result: The moxifloxacin regimens were not non-inferior to the standard 6-month regimen for the treatment of drug-sensitive tuberculosis.
2014

RIFAQUIN Trial

n = 827 · NEJM

Tested

4-month regimen with moxifloxacin and rifapentine

Population

Patients with newly diagnosed pulmonary tuberculosis

Comparator

Standard 6-month regimen

Endpoint

Unfavorable outcome (relapse or failure) at 18 months

Key result: The 4-month moxifloxacin-rifapentine regimen was non-inferior to the 6-month standard regimen in patients with non-cavitary disease but failed in the overall intention-to-treat analysis.
2021

TBTC Study 31/A5349 Trial

n = 2,516 · NEJM

Tested

4-month rifapentine-moxifloxacin regimen

Population

Patients with newly diagnosed pulmonary tuberculosis

Comparator

Standard 6-month rifampin-based regimen

Endpoint

Survival free of tuberculosis at 12 months

Key result: A 4-month regimen containing high-dose rifapentine and moxifloxacin was non-inferior to the standard 6-month treatment.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis