Assessing PA-824 for Tuberculosis (the APT Trial): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Assessing PA-824 for Tuberculosis (the APT Trial) (NCT02256696), a randomized phase 2 study. The primary outcome is Time to Sputum Culture Conversion on Liquid Medium.

2. Clinical Question

In practical statistical terms, the trial asked whether the time until sputum culture changed from positive to negative differed among the PA-824-containing regimens and the standard-treatment regimen. The endpoint concerns the timing of a laboratory milestone, not a direct measure of long-term disease status. The registry record lists several additional outcomes, including culture conversion by 8 weeks, time to culture conversion on solid medium, permanent discontinuation of the assigned regimen, pharmacokinetic measures, and the relationship between PA-824 exposure and change in time to positivity.

The registry record does not report numerical results for the primary endpoint or these secondary outcomes. It therefore does not establish from the posted results whether conversion occurred earlier or later in any arm, or how precise any arm comparison was. The clinical question can be described from the trial design, but its answer requires results that are not reported in the registry record.

3. Trial Design

This was a Phase 2 randomized, parallel-group, open-label trial sponsored by Johns Hopkins University. It enrolled 157 participants and is listed as completed. The registry dates are 2015-04-29 for study start and 2022-05-01 for primary completion. These dates describe the study record; they do not supply the timing or duration of individual participants’ follow-up.

Three treatment arms are described. Arm 1 received PA-824 200 mg once daily with rifampin 600 mg once daily, isoniazid 5 mg/kg once daily, and pyrazinamide 25 mg/kg once daily for 8 weeks. For the following 4 weeks, the regimen continued PA-824, rifampin, and isoniazid. Arm 2 used the same PA-824, isoniazid, and pyrazinamide schedule, with rifabutin 300 mg once daily instead of rifampin; rifabutin and the other continued medicines were used in the following 4 weeks. Arm 3 received rifampin, ethambutol, isoniazid, and pyrazinamide for 8 weeks, followed by rifampin and isoniazid for 4 weeks.

Randomization is reported, but the allocation ratio, sequence-generation method, and concealment procedures are not reported in the registry record. The trial was open-label, meaning masking was not used. The record does not report whether laboratory personnel assessing cultures were masked to assignment. It also does not report the planned sample size, a sample-size justification, or whether enrollment differed from the original target.

The primary endpoint was time to sputum culture conversion on liquid medium over 12 weeks. The primary description defines conversion as the time in days for sputum to change from positive to negative. The reported summary measure is the median, with inter-quartile range as the dispersion measure. The record lists eight outcome measures and eight analyses, but does not provide enough detail to identify every analysis population or the prespecified model for each comparison.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Adult (\> 18 years) drug sensitive pulmonary TB patients with weight 40 - 80 kg, HIV Negative or Positive with CD4 \> 350, not on antiretroviral therapy (ART)

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Time to Sputum Culture Conversion on Liquid Medium, measured over 12 weeks.

Time to sputum culture conversion on liquid medium measures how long it takes for a participant’s sputum culture to change from positive to negative. In this trial the timeframe was 12 weeks, and the registry specified days as the unit, with the median and inter-quartile range as the summary statistics. The endpoint captures the timing of a microbiological change during treatment rather than only whether conversion has occurred at a single visit.

As a general statistical concept, a time-to-event endpoint retains information about when an event occurs. A participant who converts earlier and one who converts later are not treated as equivalent merely because both eventually experience conversion. Such outcomes also require care when participants have not experienced the event by the end of observation or when their event time is otherwise unavailable: their information may be censored rather than treated as a completed event time. The registry record does not report how censoring was handled for this endpoint.

Culture conversion is used as a microbiological measure because it expresses a change in culture status over a defined treatment period. It is not, by itself, a complete account of every clinical outcome. The registry record does not report long-term outcomes or provide the primary endpoint’s arm-specific medians and inter-quartile ranges. Without those results, readers cannot determine the observed conversion times, their spread, or the precision of a comparison between regimens.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical interpretation is constrained by what the registry record reports. It describes a randomized, parallel-group study and names a median time-to-event endpoint, but it does not report the primary endpoint estimates or the sponsor’s statistical analysis plan. General concepts below explain how such outcomes are often interpreted; they should not be mistaken for methods confirmed as used in this trial.

What does a median time-to-event measure?

As a general statistical concept, a median event time is the point by which half of the relevant event-time distribution has experienced the event, under the analysis definition. Here the event is sputum culture conversion on liquid medium, and the unit is days. The inter-quartile range describes the middle half of the distribution by marking its lower and upper quartiles. It conveys spread, but it is not a confidence interval for the median and does not, on its own, quantify uncertainty in a comparison between arms.

Time-to-event data can include participants whose event was not observed during the period of follow-up. In general, such observations may be censored: they contribute information up to the last time their event status is known, without being assigned an event time that was not observed. The handling of censoring, the time origin, rules for confirming conversion, and the analysis population are not reported in the registry record. These details affect interpretation and cannot be inferred from the endpoint title.

How are time-to-event comparisons commonly expressed?

In general, a hazard ratio compares the instantaneous event rates between groups over follow-up, conditional on remaining event-free up to each time. A hazard ratio is not the same as a ratio of medians, and its interpretation depends on the model and assumptions used. The record’s statistical-methods description says that only sponsor-reported hazard ratios are used for time-to-event outcomes. No numerical sponsor-reported hazard ratio for the primary endpoint is included in the registry record, so no such comparison can be interpreted.

A confidence interval is a range of values compatible with the data and analysis under the method’s assumptions. It indicates the precision of an estimate: a narrower interval represents more precision than a wider one. A p-value addresses how compatible the observed data are with a specified null hypothesis under a statistical model; it does not express the probability that the null hypothesis is true or measure clinical importance. The primary endpoint’s confidence interval and p-value are not reported in the registry record.

Why can assumptions matter?

General time-to-event models can rely on assumptions about how the relative event rate behaves over time, and on appropriate handling of censoring. If the relative rate changes over follow-up, a single summary measure may not describe the entire pattern well. The record does not report diagnostics or model details that would show whether such assumptions were assessed. No particular model should therefore be attributed to the trial based on the endpoint label alone.

As a general concept, over-dispersion occurs when variability in count data exceeds what a simpler model assumes. It can make standard errors too small if the extra variability is ignored. This trial’s primary endpoint is described as time to event, not as a count outcome, and the record does not report an over-dispersion analysis. The concept is relevant only if a count-based secondary analysis were used; the posted information does not establish that one was used.

What do planning illustrations mean?

Any sample-size illustrations associated with these analyses are planning illustrations based on an observed effect, not the trial’s own design justification and not observed power. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. For time-to-event outcomes, the illustration gives a total number of events using the Schoenfeld method. These assumptions describe the illustrations, not necessarily the trial’s original planning or actual analysis.

Planning illustrations are conditional calculations. They do not change the evidence observed in the trial, and they cannot replace estimates with uncertainty intervals. A sample-size target is not proof that a study was adequately powered, while a retrospective power calculation is not a substitute for interpreting the effect estimate and its precision. The registry record does not report the trial’s own sample-size justification, and no new planning calculation is inferred in this explanation.

Why is the sponsor’s analysis important?

Recalculated results, when available, are simple unadjusted calculations from posted summary statistics. For time-to-event outcomes, the stated approach uses the sponsor’s reported hazard ratios rather than creating an alternative comparison from summary values. If a discrepancy between a recalculation and the sponsor’s analysis is identified, the sponsor’s own analysis takes precedence. The registry record does not include the primary endpoint results needed to assess such a comparison.

Randomization can support a fair comparison by balancing measured and unmeasured characteristics on average, but it does not guarantee that every group is identical in a particular trial. Interpretation still depends on outcome ascertainment, missing data, adherence, and the analysis population. The registry record does not provide enough information to evaluate these issues for the primary endpoint. The appropriate conclusion is therefore limited: the design is described, but the reported numerical evidence for the primary comparison is unavailable in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The record lists multiple outcomes, including the primary time to culture conversion on liquid medium and secondary measures of conversion on solid medium, conversion by 8 weeks, permanent discontinuation, pharmacokinetics, and change in time to positivity. With multiple outcomes or comparisons, the chance of obtaining at least one apparently notable result can increase if many tests are conducted. As a general statistical approach, investigators may prespecify a testing hierarchy, adjust for multiplicity, or treat secondary analyses as exploratory. The multiplicity strategy is not reported in the registry record.

The record indicates eight outcome measures and eight analyses, but does not report a complete analysis plan or enough detail to connect each listed outcome to a specific model and comparison. It does not report whether all listed analyses were prespecified, whether any interim analyses were conducted, or whether adjustments for multiple testing were planned. These matters are not reported in the registry record and should not be inferred from the count of outcomes or analyses.

Missing-data handling is also not described for the primary endpoint. The record does not report how missing cultures, withdrawals, treatment discontinuations, or participants who did not convert during the 12-week timeframe were handled in the efficacy analysis. For a time-to-event endpoint, the distinction between censoring and other missingness can affect both the estimate and its interpretation. The record does not report sensitivity analyses examining these possibilities.

The analysis populations are not reported in enough detail to establish whether the primary comparison used all randomized participants, participants who received treatment, or another defined group. The registry record also does not report adherence or protocol-deviation summaries for interpreting those populations. Without these details and numerical outcome estimates, readers cannot judge the effects of missingness, analysis choices, or multiplicity on the posted primary result.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: increased liver enzymes (5.7%), hemoptysis (1.9%), pneumothorax (1.8%), pleuritic chest pain (1.8%), deep vein thrombosis (1.8%).

The safety record reports serious adverse events among 56 participants at risk in Arm 1, 53 in Arm 2, and 48 in Arm 3. The number of participants affected by a serious adverse event was 3, 4, and 0, respectively. The record separately reports other adverse events among 56, 53, and 48 participants at risk, with 3, 3, and 2 participants affected, respectively. No deaths are recorded in any arm.

These are counts of participants affected, not necessarily counts of every event. A participant may experience more than one event, and the registry record does not clarify event-level totals or provide definitions beyond the categories shown. The leading serious-event terms listed include increased liver enzymes at 5.7%, hemoptysis at 1.9%, pneumothorax at 1.8%, pleuritic chest pain at 1.8%, and deep vein thrombosis at 1.8%. The record does not provide arm-specific rates for each listed term in the information described.

Safety counts should be read descriptively. They do not, by themselves, establish that a regimen caused or prevented an event, and the record does not report comparative confidence intervals or statistical tests for these figures. Different numbers at risk across arms also mean that counts alone are not a complete basis for comparing event frequency. The registry record does not report further detail on event timing, severity grading, relatedness, or follow-up completeness.

12. What the Result Does, and Does Not, Mean

The record establishes that the APT trial compared two PA-824-containing regimens with a standard-treatment regimen in a randomized, open-label Phase 2 design. It also specifies the primary endpoint as time to sputum culture conversion on liquid medium over 12 weeks and gives the planned treatment schedules. These details define the question the trial was designed to address.

The record does not report the primary endpoint results. It therefore does not show the median conversion time in any arm, the spread of those times, or the size and precision of a treatment comparison. It would be inappropriate to infer that PA-824 shortened or lengthened conversion time, or that the regimens were equivalent, from the absence of numerical results. Absence of a reported estimate is not an estimate of no effect.

The safety information describes reported counts of participants with serious and other adverse events, along with a list of leading serious-event terms. These counts are not a full comparative safety analysis. The registry record does not include uncertainty intervals or tests for the arm comparisons, nor does it provide enough detail to determine how every event was attributed or counted.

More generally, a Phase 2 trial’s microbiological endpoint is not interchangeable with every longer-term clinical outcome. The record lists secondary measures and pharmacokinetic outcomes, but their numerical results are not reported in the registry record. No conclusion about those outcomes, or about a broader treatment effect, can be drawn from their inclusion in the outcome list alone.

The most defensible interpretation is that the registry record documents the design and selected safety data but leaves the primary efficacy comparison unresolved. A complete assessment would require the sponsor’s reported endpoint estimates, uncertainty measures, analysis methods, and details about missing observations and analysis populations. Those details are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial raises several statistical issues that are important in clinical research. First, its primary outcome is a time-to-event measure rather than a single status assessed at one visit. The timing of conversion carries information, but interpretation depends on how the event is defined, how follow-up is handled, and what happens when conversion is not observed within the measurement period. The registry record identifies the endpoint but does not report all of these analysis details.

Second, the study has three treatment arms and multiple listed outcomes. Comparisons across arms and outcomes require a clear plan for which questions are primary and how other findings will be interpreted. Without a reported multiplicity strategy, readers cannot determine whether statistical testing accounted for the range of analyses or whether secondary results were intended to be exploratory.

Third, the open-label design and laboratory-based endpoint make outcome-assessment procedures relevant. The record states that masking was not used, but does not report whether culture assessment was masked or describe procedures for resolving uncertain results. These details matter to understanding how outcome measurement may have been protected from knowledge of treatment assignment.

Finally, the registry record illustrates the difference between a well-defined research question and a reportable result. It provides eligibility criteria, regimen schedules, the endpoint definition, and selected safety counts, but not the primary estimates and their precision. For statistical interpretation, design information is necessary but cannot replace the results. The record’s limitations leave the central comparative question unanswered.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main trial-specific limitation for interpreting efficacy is that arm-specific primary endpoint results are not reported in the registry record. The median time to culture conversion on liquid medium and its inter-quartile range are not available, and no sponsor-reported primary comparison or uncertainty interval is given. This prevents assessment of the observed magnitude, direction, and precision of the primary outcome.

The record describes the trial as open-label, with no masking. It does not report whether laboratory staff assessing sputum cultures were masked, or how conversion was confirmed. It also does not report the timing rules used to define the event, the approach to censored observations, or how missing culture measurements were handled. These limitations are specific to interpreting a culture-conversion time endpoint.

Although the total enrollment was 157, the record does not report the original target enrollment or its justification. The allocation ratio and randomization procedures are also not reported in the registry record. Arm-specific enrollment or efficacy analysis counts are not provided in the information described, limiting assessment of whether the enrolled participants contributed to the primary analysis as planned.

The record lists several secondary outcomes and analyses but does not report a multiplicity plan, a full statistical analysis plan, interim-analysis information, or defined analysis populations. Numerical secondary outcome results are not reported in the registry record. The safety counts are descriptive and are not accompanied by comparative uncertainty measures or further detail on event-level counting and attribution. These omissions constrain conclusions about both efficacy and comparative safety.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Assessing PA-824 for Tuberculosis (the APT Trial) — NCT02256696.
• Registry record last updated 2023-07-18; status last verified 2023-05.
• Results first posted 2023-07-18.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT02256696, titled “Assessing PA-824 for Tuberculosis (the APT Trial),” is listed as a completed Phase 2 randomized, open-label, parallel-group trial sponsored by Johns Hopkins University. It enrolled 157 participants with pulmonary tuberculosis. The study compared two PA-824-containing regimens, one with rifampin and one with rifabutin, against a regimen using rifampin, ethambutol, isoniazid, and pyrazinamide.

The primary endpoint was time to sputum culture conversion on liquid medium over 12 weeks, summarized as a median and inter-quartile range. The registry record does not report numerical primary endpoint results or the sponsor’s primary comparison. It lists secondary microbiological, discontinuation, and pharmacokinetic outcomes, but their numerical results are not reported in the registry record.

Safety counts are reported for serious and other adverse events, and no deaths are recorded in the listed arms. The posted record therefore supports a description of the trial’s design and selected safety information, but does not provide the primary efficacy estimates needed to answer the central comparative question.