This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
TB-PRACTECAL was a randomized phase 2/3 parallel-group clinical trial evaluating treatment regimens for multidrug-resistant and extensively drug-resistant tuberculosis. The trial compared drug regimens containing bedaquiline, pretomanid, moxifloxacin, linezolid, and clofazimine with locally accepted standard of care consistent with WHO recommendations.
| Feature | TB-PRACTECAL |
|---|---|
| Condition | Tuberculosis, Multidrug-Resistant; Extensively Drug-Resistant Tuberculosis; Tuberculosis, Pulmonary |
| Phase | Phase 2/3 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None |
| Primary purpose | Treatment |
| Lead sponsor | Medecins Sans Frontieres, Netherlands |
2. Clinical Question
The central clinical question was whether shorter and less toxic multidrug-resistant tuberculosis treatment regimens could be evaluated against locally accepted standard of care in patients with multidrug-resistant tuberculosis and related pulmonary tuberculosis conditions.
Population
Patients with multidrug-resistant tuberculosis, extensively drug-resistant tuberculosis, or pulmonary tuberculosis.
Intervention
Regimens containing bedaquiline, pretomanid, moxifloxacin, linezolid, and clofazimine.
Comparator
Locally accepted standard of care consistent with WHO recommendations for treatment of M/XDR-TB.
Primary question
How do randomized treatment regimens compare when evaluated using culture conversion, treatment discontinuation/death, and unfavorable outcome endpoints?
3. Trial Design
4. Endpoints
| Endpoint | Time frame |
|---|---|
| Stage 1: Percentage of patients with culture conversion in liquid media at 8 weeks post randomisation. | 8 weeks post randomisation |
| Stage 1: Percentage of patients who discontinue treatment for any reason or die | 8 weeks post randomisation |
| Stage 2: Percentage of patients with an unfavourable outcome (failure, death, recurrence, loss to follow-up) | 72 weeks post-randomisation |
5. Planned Analysis
The ClinicalTrials.gov record lists the registered primary endpoints but does not report posted statistical analyses or endpoint results.
The Stage 1 endpoints are binary outcomes: culture conversion and the composite outcome of treatment discontinuation or death. Binary endpoints are typically analyzed using comparisons of proportions between randomized groups, with confidence intervals used to describe uncertainty around differences or ratios.
The Stage 2 endpoint is also a binary clinical outcome: the percentage of patients with an unfavorable outcome defined as failure, death, recurrence, or loss to follow-up at 72 weeks post-randomisation. A comparison of unfavorable outcome rates between treatment groups would typically use methods appropriate for binary clinical trial endpoints.
6. Statistical Methodology
Randomization and causal comparison
Randomization is the central design feature that allows treatment groups to be compared while reducing the influence of measured and unmeasured baseline differences. In a randomized trial, the treatment assignment process creates the framework for estimating comparative treatment effects.
Binary endpoint analysis
Each primary endpoint is expressed as a percentage of patients experiencing a defined outcome. Binary endpoints focus on whether an event occurred rather than the exact timing of the event.
A treatment comparison can be expressed using absolute differences, relative risks, odds ratios, or other prespecified measures depending on the statistical analysis plan.
Composite endpoints
The Stage 1 endpoint combining treatment discontinuation for any reason or death is a composite endpoint. Composite endpoints can increase the number of observed events, but interpretation requires understanding whether individual components are clinically comparable and occur with similar importance.
Missing outcome data
The Stage 2 endpoint includes loss to follow-up as part of the unfavorable outcome definition. The handling of missing data and follow-up status is an important component of interpretation for this type of endpoint.
7. Statistical Methods Explained
Why use culture conversion as an endpoint?
Culture conversion measures microbiological response by assessing whether cultures become negative. In tuberculosis trials, microbiological endpoints can provide information about early response to therapy.
Why analyze an unfavorable outcome composite?
A composite endpoint combines several clinically important outcomes into one measure. This can provide a broader assessment of treatment success or failure, but the contribution of each component must be considered when interpreting results.
What does randomization contribute statistically?
Randomization creates comparable treatment groups on average. This allows differences in outcomes to be more directly attributed to treatment assignment rather than differences in patient characteristics.
Why are confidence intervals important?
Confidence intervals describe uncertainty around an estimated treatment effect. A point estimate alone does not show how precisely the effect has been estimated.
Why is the analysis population important?
The way patients are included in the analysis can affect interpretation. Randomized trials commonly define analysis populations in advance to preserve the advantages of randomization and appropriately evaluate safety or treatment exposure.
8. Important Limitations and Interpretation Issues
- No posted statistical results: The ClinicalTrials.gov record does not provide formal endpoint estimates or confidence intervals.
- Endpoint complexity: Composite outcomes require interpretation of their individual components.
- Loss to follow-up: Because Stage 2 includes loss to follow-up as an unfavorable outcome, follow-up completeness is relevant to interpretation.
- Multiple treatment arms: Four-arm randomized designs require careful interpretation of comparisons between groups.
- Non-masked design: The trial is registered as having no masking, which can influence outcomes involving subjective assessment or treatment decisions.
9. Why This Trial Matters Statistically
TB-PRACTECAL is an important statistical teaching example because it combines a multi-arm randomized design with microbiological and clinical outcome measures in a complex infectious disease setting.
| Concept | How it appears in TB-PRACTECAL |
|---|---|
| Randomization | Randomized allocation among four parallel treatment arms |
| Binary endpoints | Culture conversion and unfavorable outcome percentages |
| Composite endpoint | Failure, death, recurrence, and loss to follow-up combined in Stage 2 |
| Multi-arm design | Four-arm randomized comparison structure |
| Follow-up timing | Endpoints assessed from 8 weeks post randomisation through 72 weeks post-randomisation |
| Clinical interpretation | Requires integration of microbiological and patient-level outcomes |
10. Sources
- ClinicalTrials.gov: NCT02589782.
- PubMed: TB-PRACTECAL publication record.
- PubMed: TB-PRACTECAL publication record.
- PubMed: TB-PRACTECAL publication record.
- PubMed: TB-PRACTECAL publication record.
Continue through the Clinical Biostats knowledge graph
Explore additional clinical trial results, statistical tutorials, and analysis tools.