• Combined failure of bacteriological cure and relapse within one year of completion of therapy as defined by culture using solid media. [ Time Frame: 18 months (within one year of completion of therapy) ] [ Designated as safety issue: No ]
  
• Proportion of patients with grade 3 or 4 adverse events (using a modified DAIDS scale of adverse event reporting) [ Time Frame: 18 months (within one year of completion of therapy) ] [ Designated as safety issue: Yes ]
  

• Combined failure of bacteriological cure and relapse as defined by culture using liquid media [ Time Frame: 18 months (within one year of completion of therapy) ] [ Designated as safety issue: No ]
  
• Proportion of patients who are culture negative (solid and liquid media) [ Time Frame: 8 weeks ] [ Designated as safety issue: No ]
  
• Time to first culture negative sputum sample (solid and liquid media) [ Time Frame: Estimated 3 to 6 months ] [ Designated as safety issue: No ]
  
• Speed of decline of sputum viable count by culture using solid or liquid media [ Time Frame: 6 months ] [ Designated as safety issue: No ]
  
• Sensitivity analyses assuming all losses to follow-up and non-tuberculous deaths have an unfavourable outcome using both solid and liquid media
  
• Sensitivity analyses assuming all losses to follow-up and non-tuberculous deaths have a favourable outcome using both solid and liquid media.
  

The current recommended treatments for tuberculosis (TB) require a patient to take multiple drugs for six to eight months. Because the course of therapy is long, many patients do not adhere to treatment and as a consequence they have a poor outcome. In these cases either the sputum is not cleared of the bacteria causing tuberculosis, or the disease returns again (called relapse). Response to medication can be monitored during treatment by collecting regular sputum samples and examining these samples by culture, for the organisms that cause tuberculosis.

The commonly used drugs to treat tuberculosis are rifampicin, isoniazid, ethambutol and pyrazinamide. Previous studies in animals and in humans suggest that a new drug called moxifloxacin may also be an effective treatment in tuberculosis. Moreover, promising laboratory studies on mice suggest that moxifloxacin may enable the total duration of chemotherapy to be reduced to four months, which would be a significant improvement for patients taking medication for tuberculosis.

This study will involve comparisons that are designed to assess whether substituting moxifloxacin for individual drugs in existing treatment combinations will enable tuberculosis treatment to be shortened. Patients selected for the study will be allocated to one of three treatment groups. The first group will be given six months standard treatment. A second group will receive moxifloxacin substituted for ethambutol, as part of a four month regimen, to see whether the shorter treatment is not inferior to the standard six month treatment. The third group will receive moxifloxacin substituted for isoniazid, as part of a four month regimen, to see whether the shorter treatment is not inferior to the standard six month treatment.

Hypotheses:

1. In treatment-naïve adults with active pulmonary TB treated with eight weeks of moxifloxacin (M), isoniazid (H), rifampicin (R) and pyrazinamide (Z) (i.e. a standard regimen where moxifloxacin is substituted for ethambutol (E)), followed by nine weeks of moxifloxacin, isoniazid and rifampicin, followed by nine weeks of placebo, the proportion of patients who experience treatment failure or disease relapse in the twelve months following treatment completion will not be inferior to that observed in patients who are treated with a standard regimen (eight weeks of ethambutol, isoniazid, rifampicin and pyrazinamide followed by eighteen weeks of isoniazid plus rifampicin) (Comparison 1). See Fig. 1 REMox Trial Design.
2. In treatment-naïve adults with active pulmonary TB treated with eight weeks of ethambutol, moxifloxacin, rifampicin and pyrazinamide (i.e. a standard regimen where moxifloxacin is substituted for isoniazid), followed by nine weeks of moxifloxacin and rifampicin followed by nine weeks of placebo, the proportion of patients who experience treatment failure or disease relapse in the twelve months following treatment completion will not be inferior to that observed in patients who are treated with a standard regimen (eight weeks of ethambutol, isoniazid, rifampicin and pyrazinamide followed by eighteen weeks of isoniazid plus rifampicin) (Comparison 2). See Fig. 1 REMox Trial Design.