Open-label Study of Dolutegravir (DTG) or Efavirenz (EFV) for Human Immunodeficiency Virus (HIV) - Tuberculosis (TB) Co-infection: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Open-label Study of Dolutegravir (DTG) or Efavirenz (EFV) for Human Immunodeficiency Virus (HIV) - Tuberculosis (TB) Co-infection (NCT02178592), a randomized phase 3 study. The primary outcome is Percentage of Participants With Plasma Human Immunodeficiency Virus-1 Ribonucleic Acid (HIV-1 RNA) < 50 Copies/Milliliter at Week 48 in DTG Arm Using the Modified United States (US) Food and Drug Administration (FDA) Snapshot Algorithm.

2. Clinical Question

The study was designed to evaluate safety and efficacy in the setting of HIV and tuberculosis treatment together. Its open-label design means participants and study staff were not masked to assigned treatment. The primary question is specifically an outcome within the DTG arm, rather than a direct contrast between DTG and EFV. Although the study randomized participants to both regimens, the primary endpoint as defined in the registry does not itself establish a comparative treatment effect. The registry record does not report the primary endpoint response percentage, interval estimate, or a primary between-arm test.

3. Trial Design

This was a Phase IIIb, randomized, open-label, parallel-group study sponsored by ViiV Healthcare. The registry record gives enrollment as 113 and study status as completed. The listed start date is 2015-01-23, and the primary completion date is 2017-11-02. No masking was used. The registry record does not report a more detailed method for generating or concealing the random allocation.

Participants were antiretroviral therapy-naïve adults with HIV-1 infection who were starting treatment for rifampicin-sensitive tuberculosis. Both regimens included two NRTIs and were administered with rifampicin-containing tuberculosis treatment. In the DTG arm, participants received DTG 50 mg twice daily during tuberculosis therapy and for 2 weeks after its discontinuation, followed by once-daily DTG 50 mg. The EFV arm received EFV 600 mg once daily through Week 52 along with tuberculosis therapy. The registry describes tuberculosis treatment as including isoniazid, rifampicin, pyrazinamide, and ethambutol at standard doses under national tuberculosis program conditions or acceptable alternatives.

The primary endpoint was assessed at Week 48 in the DTG arm. Secondary outcomes include viral suppression in the EFV arm at Week 48, viral suppression in both arms at Week 24, status without confirmed virologic withdrawal or treatment-related discontinuation, CD4+ count changes, adverse events, and chemistry toxicities. The registry record lists 18 outcome measures. It does not report the full protocol, the planned statistical analysis document, or the rationale used to determine enrollment. A sample-size justification and any interim-analysis plan are not reported in the registry record.

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.

Before assignment: A total of 263 participants were screened of which 150 were screen failures. A total of 113 participants were enrolled in this study.

Randomized Phase (Up to Week 52)

Reasons for not completing this period:

OLE Phase (Week 52 to Week 252)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-treat exposed (ITT-E) Population comprised of all randomly assigned participants who received at least one dose of IP.

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 Percentage of Participants With Plasma Human Immunodeficiency Virus-1 Ribonucleic Acid (HIV-1 RNA) < 50 Copies/Milliliter at Week 48 in DTG Arm Using the Modified United States (US) Food and Drug Administration (FDA) Snapshot Algorithm, measured over Week 48.

The primary endpoint is the percentage of participants in the DTG arm with plasma HIV-1 RNA <50 copies/milliliter at Week 48. HIV-1 RNA is a measure of viral load. The threshold classifies participants according to whether the measured viral load is below the stated level at the specified visit. The registry describes quantitative analysis of plasma samples and identifies the response measure as a percentage.

The endpoint uses a modified US FDA Snapshot algorithm. Under the registry definition, participants with HIV-1 RNA at or above 50 copies/milliliter are non-responders. Participants without Week 48 HIV-1 RNA data, including because of missing data or discontinuation of investigational product before the visit window, are also counted as non-responders. Participants with antiretroviral substitutions that were not permitted are likewise counted as non-responders. A single protocol-allowed background therapy substitution does not count against response, even if it occurs after the first trial visit; DTG or EFV itself could not be substituted.

General statistical explanation: a fixed viral-load threshold turns a continuous laboratory measurement into a responder/non-responder outcome. This can make the result easier to interpret as the proportion meeting a prespecified criterion, but it does not retain all information about the actual viral-load values. The endpoint also incorporates treatment discontinuation and missing measurements into the response definition. That approach gives a defined rule for classifying participants, but the resulting percentage reflects both the measured viral load and the algorithm’s handling of missingness and treatment changes. The registry record does not report the observed percentage or its confidence interval.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry defines the primary outcome as a percentage of participants with plasma HIV-1 RNA <50 copies/milliliter at Week 48 in the DTG arm. It identifies a 95% confidence interval as the measure of dispersion. However, the response percentage and interval endpoints are not reported in the registry record. No primary statistical test or comparative estimate is reported. The following explains the stated methods and their general statistical meaning.

What does the Snapshot approach classify?

The modified US FDA Snapshot algorithm applies a prespecified response rule at a defined visit. For this endpoint, a participant is a responder only if the Week 48 viral-load result meets the threshold and the participant’s treatment and data status do not trigger a non-response classification. A value at or above 50 copies/milliliter is classified as non-response. Missing Week 48 data and discontinuation of investigational product before the visit window are also classified as non-response. The registry further identifies unpermitted antiretroviral substitutions as non-response.

The modification described in the record allows a single protocol-permitted background therapy substitution without treating it as failure, even if it takes place after the first trial visit. Substitution of the assigned study drug, DTG or EFV, was not allowed. This distinction matters because the endpoint counts a treatment-policy outcome under explicit rules, rather than simply summarizing viral-load measurements available at the visit. The registry does not give the number of participants assigned to each response category or explain how many classifications arose from missing measurements, discontinuations, or treatment changes.

What does the confidence interval contribute?

General statistical explanation: a confidence interval conveys uncertainty around an estimate. For a response percentage, the interval gives a range reflecting the data and the method used to calculate it. It does not remove uncertainty, establish that an effect is clinically important, or by itself test whether two treatments differ. The registry specifies a 95% confidence interval for the primary endpoint, but does not report its limits. The response estimate’s precision therefore cannot be evaluated from this record.

The record does not identify the interval calculation method, any adjustment for covariates, or a primary between-arm hypothesis test. Those details are not reported in the registry record and should not be inferred from the endpoint description. A percentage and its interval would describe the endpoint in the DTG arm; a comparative conclusion would require an appropriate between-arm estimate and its uncertainty.

How should the planning assumptions be read?

The study record’s planning illustration assumptions describe a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The illustration is based on the observed effect and considers an effect to detect expressed as a fraction of that observed effect. It is not the trial’s own design justification. The record does not report the resulting planning values in the registry record for this analysis, so no participant requirement can be stated. Such illustrations are prospective planning scenarios, not observed power and not evidence that the study had a particular probability of detecting an effect.

General statistical explanation: power is a design concept tied to an assumed effect, variability, significance threshold, and sample size. It is not a property calculated from a study’s observed result that can establish whether a result is reliable. A planning exercise using an observed effect as its starting point is especially dependent on that effect estimate and should be treated as an illustration rather than as the trial’s original rationale. The registry record does not report a formal sample-size justification.

What is not available for re-analysis?

The primary result, its interval limits, and arm-level endpoint counts are not reported in the registry record. There is therefore no basis in the posted information for an unadjusted recalculation of the primary endpoint or a treatment contrast. Nor are individual participant data, outcome denominators for the primary endpoint, or the full analysis plan reported. The registry lists 18 outcome measures and 0 analyses in the record details used for this description. The absence of listed analyses means there is no posted set of analysis results from which to assess consistency across the listed outcomes.

The endpoint definition gives useful information about the classification rule, and the reported safety counts allow a limited descriptive reading by treatment group and study phase. These details do not replace the missing efficacy estimate, interval, or full statistical analysis. General statistical explanation: interpretation depends on both the chosen estimator and the population included in the analysis. Results using different populations or different rules for missing data may answer different questions. The primary endpoint population and its missing-data classifications are specified, but the actual endpoint result is not.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("DTG 50 mg")
n      <- c(69)
events <- c(52)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  DTG 50 mg  69  52  75.4  63.5  84.9  64.0  84.0  64.0  84.0

11. Multiplicity, Missing Data and Analysis Populations

The registry lists several outcomes in addition to the primary Week 48 DTG-arm endpoint. These include viral suppression in the EFV arm at Week 48, viral suppression in both arms at Week 24, treatment-related discontinuation and virologic withdrawal outcomes, CD4+ count changes, adverse events, and chemistry toxicities. The record lists 18 outcome measures. It does not report a multiplicity adjustment or explain how confirmatory and exploratory analyses were distinguished. Any family-wise error control or other approach to multiple testing is therefore not reported in the registry record.

General statistical explanation: when multiple outcomes or comparisons are tested, the chance of at least one apparently notable finding can increase if each is evaluated without adjustment. Whether an adjustment is appropriate depends on the study’s aims and the prespecified analysis plan. Because the registry record does not report that plan or an adjustment strategy, the statistical status of results across the outcomes cannot be assessed from the record alone.

For the primary endpoint, missing Week 48 viral-load data and discontinuation of investigational product before the visit window are counted as non-response. HIV-1 RNA at or above 50 copies/milliliter and unpermitted antiretroviral substitutions are also classified as non-response. A single permitted background therapy substitution is allowed under the modified algorithm. These are explicit endpoint rules, not evidence that missing values were imputed as measured viral loads.

The endpoint population is defined as randomly assigned participants who received at least one dose of investigational product. This is the intent-to-treat exposed population described in the record. The record does not report the number included in that endpoint analysis or provide a detailed accounting of exclusions and missing endpoint data. The reasons for missing values beyond those described in the algorithm, sensitivity analyses, and alternative missing-data methods are not reported in the registry record.

12. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: psoas abscess (5.3%), headache (5.3%), pneumonia (2.3%), patella fracture (2.3%), immune reconstitution inflammatory syndrome (2.3%).

The registry reports serious adverse event counts separately for the randomized phase and the OLE phase. In the randomized phase, 5 participants in the DTG group were affected among 69 at risk, and 5 in the EFV group were affected among 44 at risk. In the OLE phase, the corresponding counts are 5 among 47 in the DTG group and 2 among 19 in the EFV group. The record lists 0 deaths in each randomized arm, 1 death in the DTG OLE group, and 0 in the EFV OLE group.

These are participant counts, not a randomized comparison of adverse-event rates. The OLE groups are reported separately from the randomized phase and should not be interpreted as if they preserve the original randomized comparison. The registry does not provide an adjusted safety comparison, detailed exposure-adjusted rates, or causal attribution for the listed events. A serious adverse event count also does not establish that the study drug caused the event.

The record lists serious-event terms including psoas abscess and headache at 5.3%, and pneumonia, patella fracture, immune reconstitution inflammatory syndrome, and spontaneous abortion at 2.3%. The listed percentages are not accompanied by enough context in the record to assign them to a particular arm or phase. They should not be used to infer comparative risks.

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

The record establishes the study’s design and the way its primary response outcome was defined. It does not report the primary Week 48 response percentage for the DTG arm, the associated 95% confidence interval, or a direct comparison with EFV. As a result, the posted information does not show the size or precision of the primary endpoint result. It also does not support a conclusion that one regimen had better viral suppression than the other.

The endpoint has a specific interpretation: it is the proportion in the DTG arm classified as having HIV-1 RNA <50 copies/milliliter at Week 48 under a modified Snapshot rule. Missing data, early discontinuation of investigational product, a viral-load result at or above the threshold, and unpermitted antiretroviral substitutions count as non-response. The resulting percentage, if reported, would reflect both measurements and these classification rules. It would not be a simple summary of only the observed viral-load values among participants with a measurement.

The safety information can be described as counts of participants affected during two study phases. Those counts do not establish that an event was caused by the assigned medicine, and the OLE counts do not form a randomized comparison. Event counts also do not describe the nature, timing, duration, or clinical course of every event. The registry record does not provide a complete comparative safety analysis.

General statistical explanation: the absence of a reported estimate is different from a finding of no effect. It means the result cannot be evaluated from the registry information, rather than showing that the outcome was the same across treatments. Likewise, a confidence interval is needed to understand uncertainty around an estimate; merely knowing that one was planned does not convey its precision. The design and endpoint definitions provide context, but they cannot substitute for numerical outcome results.

The registry record therefore permits a cautious account of who was studied, what regimens were compared, how the primary outcome was classified, and what safety counts were listed. It does not permit a conclusion about comparative efficacy, the precision of the primary result, or a complete benefit-risk assessment.

14. Why This Trial Matters Statistically

This study addresses a statistical question in a setting where participants were receiving treatment for HIV-1 and rifampicin-sensitive tuberculosis at the same time. The randomized parallel-group design places DTG- and EFV-based regimens within the same study framework. That design is relevant to comparative inference, while the open-label status means that treatment assignment was not masked.

The primary endpoint also illustrates why outcome definitions matter. A viral-load threshold gives a clear response criterion, while the modified Snapshot algorithm specifies how to classify missing measurements, discontinuation, and permitted or unpermitted treatment changes. These rules can change the interpretation of a response percentage: it reflects the outcome definition as well as the laboratory measurements. Being explicit about the algorithm helps readers understand what a reported percentage would represent.

At the same time, the registry’s primary outcome is defined for the DTG arm rather than as a direct contrast between arms. The record lists secondary viral-suppression outcomes in EFV and across both arms, but does not report their results. This illustrates an important distinction between a randomized study design and a reported comparative estimand: randomization alone does not supply a treatment-effect estimate.

The record also makes clear why precision and analysis details matter. It specifies a confidence interval for the primary percentage, but the estimate and interval are not reported. Without them, neither the magnitude nor uncertainty can be judged. Likewise, the safety counts can be read descriptively, but the randomized and OLE phases should remain distinct in interpretation. The study is statistically informative as an example of endpoint construction and phase-specific reporting, while the missing outcome results limit conclusions about treatment differences.

15. 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 most important limitation for interpreting efficacy is that the registry record does not report the primary endpoint response percentage or its 95% confidence interval. It also does not report the primary analysis denominator, the number classified as responders, or a direct between-arm effect estimate. The results of the listed secondary outcomes are not included in the record used for this description. Accordingly, the magnitude, precision, and comparative interpretation of efficacy cannot be assessed.

The primary outcome applies to the DTG arm only. This means that the primary endpoint definition is not itself a randomized DTG-versus-EFV contrast. The record lists additional outcomes that could support other comparisons, but their estimates are not reported. The registry record also does not provide a detailed statistical analysis plan, the method used to construct the confidence interval, a multiplicity strategy, or any interim-analysis plan.

The open-label design is another consideration when interpreting outcomes. The registry states that masking was not used; it does not describe additional procedures to reduce possible effects of treatment awareness. The primary outcome is based on laboratory measurement, but other outcomes, including adverse events and treatment discontinuations, may involve assessment or reporting. No further assessment procedures are reported in the registry record.

Safety reporting is limited to participant counts in the randomized and OLE phases, with no adjusted comparative analysis or detailed exposure information. OLE counts should not be treated as randomized comparisons. Several serious-event term percentages are listed without clear arm- or phase-specific context in the record. Finally, the registry does not report a sample-size justification, so the basis for the planned enrollment and the precision anticipated for the primary endpoint cannot be evaluated.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Open-label Study of Dolutegravir (DTG) or Efavirenz (EFV) for Human Immunodeficiency Virus (HIV) - Tuberculosis (TB) Co-infection — NCT02178592.
• Registry record last updated 2021-01-12; status last verified 2020-12.
• Results first posted 2019-02-21.
• 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.

19. Record Summary

This completed Phase IIIb study randomized 113 adults with HIV-1 infection who were starting treatment for rifampicin-sensitive tuberculosis to open-label regimens based on DTG or EFV, each with two NRTIs. The primary endpoint was the percentage in the DTG arm with HIV-1 RNA <50 copies/milliliter at Week 48, classified using a modified US FDA Snapshot algorithm.

The registry record does not report the primary endpoint estimate, its 95% confidence interval, or a direct comparative efficacy result. It lists serious adverse event counts separately for the randomized and OLE phases and gives death counts by group. Those safety figures are descriptive and do not establish causation or a randomized OLE comparison. The record does not report a sample-size justification, a detailed statistical analysis plan, or a multiplicity strategy.