Bone Mineral Density in Human Immunodeficiency Virus Type 1 (HIV-1)-Infected Adult Subjects Switching From a Tenofovir Regimen to a Dolutegravir Plus Rilpivirine Regimen: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bone Mineral Density in Human Immunodeficiency Virus Type 1 (HIV-1)-Infected Adult Subjects Switching From a Tenofovir Regimen to a Dolutegravir Plus Rilpivirine Regimen (NCT02478632), a non-randomized phase 3 study. The primary outcome is Percent Change From Baseline in Total Hip Bone Mineral Density (BMD) at Week 48.

2. Clinical Question

The endpoint focuses on bone mineral density measured at the total hip by dual energy X-ray absorptiometry. The registry specifies that the total hip measurement included the femoral neck, trochanter and inter-trochanter areas. Baseline was Day 1, and the primary timeframe was baseline through Week 48.

The central statistical question is about the estimated mean percent change and its uncertainty, rather than simply whether participants changed. A comparison between groups could describe the observed difference in those changes, but the substudy’s non-randomized allocation limits how confidently a difference can be attributed to treatment. The registry record does not report the primary endpoint estimate, its confidence interval, or the numerical comparison. It therefore does not establish the direction or size of the observed change.

3. Trial Design

This was a Phase 3, open-label, parallel-group substudy of adults with HIV-1 infection. The registry lists enrollment of 102 and records the study as completed. The recorded study dates are 2015-06-12 for study start and 2016-09-16 for primary completion. Allocation was non-randomized, and masking was none.

The substudy concerned participants receiving a tenofovir disoproxil fumarate-containing antiretroviral regimen at the time of randomization in the parent studies, SWORD-1 and SWORD-2. Those parent studies are described as randomized, open-label, multicentre, parallel-group Phase III studies. The bone-density substudy recruited participants from those parent studies and carried over their treatment groups. The substudy itself did not assign study medication: its arm descriptions state that participants did not receive study medication in study 202094.

The registry labels the carried-over groups DTG + RPV and CAR, including early-switch and late-switch group labels for some outcome periods. The available record does not define the CAR abbreviation or provide enough detail to reconstruct the parent-study treatment sequence for each participant. The fact that the parent studies were randomized does not make this substudy’s allocation randomized; its own allocation is explicitly recorded as non-randomized.

The primary outcome was percent change from baseline in total hip BMD at Week 48. Secondary outcomes addressed lumbar spine and total hip BMD, changes and percent changes, T-scores and Z-scores, later follow-up periods, and analyses by baseline third-agent class. The record lists 13 outcome measures and 24 analyses. Numerical results for the primary outcome and secondary outcomes are not reported in the registry record. Details such as the substudy’s recruitment target, sample-size justification, and any interim analyses 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: Total 146 participants were screened, of which, 44 were screen failures and 102 were registered to study 202094. The study included an early switch phase and a late switch phase.

Early Switch Phase (Up to Week 52)

Reasons for not completing this period:

Late Switch Phase (Week 52 to Week 148)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT-ED Population. Only those participants with data available at specified time point were analyzed.

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 Percent Change From Baseline in Total Hip Bone Mineral Density (BMD) at Week 48, measured over Baseline (Day 1) and Week 48.

The primary endpoint is percent change from baseline in total hip bone mineral density at Week 48. BMD is an areal density measurement expressed in grams per centimeter square, assessed using dual energy X-ray absorptiometry scans. For the total hip, the scan area included the femoral neck, trochanter and inter-trochanter regions. Baseline was defined as Day 1, and the specified follow-up measurement was at Week 48.

In general, a continuous endpoint retains information about the size and direction of measured change rather than reducing participants to categories such as improved or not improved. Percent change expresses the follow-up value relative to baseline, which can make changes easier to interpret across participants whose starting measurements differ. The registry defines the percent-change calculation as the Week 48 value minus baseline, divided by baseline, then multiplied by 100.

The registry describes the primary statistical estimate as a mean and specifies a 95% confidence interval. It also states that an analysis of covariance model was used to compare the difference. These methods are intended to summarize the average change and its uncertainty while making a group comparison. The numerical mean estimates and confidence limits are not reported in the registry record, so the magnitude and precision of the primary endpoint cannot be assessed from the record alone.

BMD is a measured structural outcome; the primary endpoint does not itself report a clinical event or establish the clinical meaning of any observed change. The registry record does not report a threshold for interpreting the size of a BMD change.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: it is a percent change in BMD, and the registry identifies the statistical parameter as a mean. A mean summarizes the central level of a continuous measurement across participants. In general, it is useful when the scientific question concerns average change, although it can be influenced by unusually high or low observations. The registry does not report distributional summaries or diagnostics that would show how individual measurements were distributed.

What does the confidence interval convey?

The registry specifies a 95% confidence interval for the primary estimate. In general, a confidence interval communicates statistical uncertainty around an estimate under the assumptions of the analysis. It is not a guarantee that the true value lies in the particular interval, nor does it describe the range of individual participant outcomes. Its interpretation depends on the model and data used. Because neither the estimate nor its limits are reported in the registry record, the interval’s location and precision cannot be evaluated.

What is the role of ANCOVA?

The registry states that an analysis of covariance, or ANCOVA, model was used to compare the difference. In general, ANCOVA relates an outcome to a group comparison while accounting for specified covariates. Adjustment can improve precision or address measured baseline differences when the model and covariates are appropriately specified. The registry record does not report which covariates were included, the model specification, or diagnostics. It is therefore not possible to assess how adjustment was implemented or whether it addressed particular differences between groups.

The recorded endpoint is percent change from baseline, and the model’s estimated value is described as a difference. The record does not provide numerical output or additional detail about how the model represented that contrast. No new estimates can be reconstructed from the endpoint definition alone.

Why does allocation matter?

In general, random allocation helps balance both measured and unmeasured characteristics between treatment groups on average. This substudy is recorded as non-randomized, even though its participants came from randomized parent studies. Treatment groups were carried over from those studies; the substudy’s own record does not describe a new random assignment. Thus, a statistical adjustment can account only for variables included and correctly modeled. It cannot ensure that all differences between the carried-over groups have been removed. Any reported contrast should be interpreted as an association in this substudy rather than as a causal effect established by its design.

How should planning calculations be read?

Any sample-size illustrations associated with the results are planning illustrations, not observed power. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. In general, such calculations ask how many participants per arm would be needed to detect a specified effect under those assumptions. They do not show the probability that the completed study detected an effect, and they do not substitute for the observed estimate and its uncertainty. Numerical planning outputs are not reported in the registry record.

More generally, a planning calculation based on an observed effect is sensitive to the estimate used: an uncertain observed estimate can lead to uncertain planning values. The registry’s stated assumptions also do not establish the study’s original design rationale. The protocol’s own sample-size justification and detailed statistical plan are not reported in the registry record.

What can the outcome counts tell readers?

For the serious-event summaries, the registry lists the number at risk and the number affected in each group, alongside deaths. In general, a count of zero affected participants is a descriptive observation within the recorded group and follow-up; it does not establish that an event could never occur or that the risk is zero. The record does not give additional detail about exposure time or a comparative safety model. These counts should not be converted into unreported rates or used as evidence of a causal safety difference.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists several secondary outcomes in addition to the primary endpoint. These include percent and absolute changes at the total hip and lumbar spine, T-scores, Z-scores, later follow-up periods, and analyses by baseline third-agent class. It records 13 outcome measures and 24 analyses. Examining multiple outcomes, timepoints or subgroups can increase the chance of obtaining apparently notable results by chance. In general, multiplicity procedures control or describe that risk, but the registry record does not report whether any such procedure was used. It also does not report a hierarchy distinguishing confirmatory outcomes from exploratory analyses.

The primary analysis population is described as the ITT-ED population, comprising participants in the ITT-E population who received at least one dose of study treatment and registered for study 202094. The registry further states that only participants with data available at the specified timepoint were analyzed. This makes the reported analysis dependent on availability of endpoint measurements at that timepoint.

The number of participants with usable Week 48 measurements, reasons for missing measurements, and any imputation or sensitivity analyses are not reported in the registry record. In general, missing-data methods matter because an analysis limited to observed measurements can be biased if missingness is related to the outcome or to participant characteristics. Without the missingness pattern and analytic details, the extent of this concern cannot be assessed.

The registry record also does not report a detailed analysis plan, the covariates used in ANCOVA, or any interim-analysis plan. Those details should not be inferred from the endpoint description or the number of listed analyses.

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 5% reporting threshold.

The registry lists four group-period summaries. DTG + RPV (Early Switch) had 53 participants at risk in the serious-event and other-event categories; zero were affected in either category, and there were zero deaths. CAR (Early Switch) lists 49 at risk in each category, with zero affected in each and zero deaths. DTG + RPV (Early + Late Switch) lists 53 at risk in each category, with zero affected in each and zero deaths. CAR (Late Switch) lists 44 at risk in each category, with zero affected in each and zero deaths.

These are reported counts for the listed groups, not estimates of comparative risk. A zero count means that no affected participants were recorded in that group and category; it does not establish that the event is impossible or that risk is absent. The record does not report exposure-adjusted rates, a safety comparison model, or further detail about the observation time underlying each summary.

These safety figures also need to be read in light of the substudy design. Participants did not receive study medication in study 202094, and the groups were carried over from parent studies. The counts alone cannot establish that one treatment caused, prevented or changed the occurrence of an event. The registry lists no top serious-event terms.

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

The record identifies a focused measurement question: average percent change in total hip BMD from baseline to Week 48 among participants in a non-randomized substudy of adults with HIV-1 infection. It specifies the measurement method, analysis population, use of ANCOVA, and a 95% confidence interval. These details clarify what the primary endpoint was intended to represent.

However, the numerical primary outcome is not reported in the registry record. Without the mean changes, the between-group estimate and its confidence limits, the observed direction, magnitude and precision cannot be described. No conclusion about whether the groups had similar or different BMD changes follows from the outcome definition or analysis method alone. The record also does not provide numerical secondary outcome results.

Even if a numerical difference were available, the non-randomized allocation would limit causal interpretation. Participants carried over groups from parent studies, but study 202094 itself did not randomize participants or administer study medication. A measured contrast could reflect treatment history, group differences or other factors, and the registry record does not provide enough model detail to evaluate adjustment for those factors.

The endpoint is a measured BMD outcome. In general, a change in a measurement does not by itself establish a change in a clinical event or an individual participant’s experience. The record does not report an interpretation threshold or link between a particular BMD change and a clinical outcome. Nor do zero adverse-event counts establish that no harms occurred outside the categories, groups or observation represented by the reported summaries.

The defensible reading is therefore limited: the registry describes the planned Week 48 comparison and its analysis, while the numerical efficacy results and important details about missing data and model specification are not reported in the registry record. This information is insufficient to quantify the observed effect or determine its clinical significance.

13. Why This Trial Matters Statistically

This substudy raises a useful statistical distinction between the design of a parent trial and the design of an analysis embedded within it. The parent studies are described as randomized, but the bone-density substudy is explicitly recorded as non-randomized. Carrying over treatment groups does not mean that the substudy’s own comparison inherits random allocation for every question it addresses. That distinction is important when interpreting group differences in a secondary measurement.

The endpoint also illustrates why an estimate and its uncertainty are both needed. A mean percent change describes an average, while a confidence interval communicates uncertainty around that estimate. Neither can be interpreted from an endpoint title alone. Because the registry does not report the numerical estimate or interval, the statistical question is clear but the answer cannot be quantified from this record.

The population definition and available-case statement matter as well. The analysis included participants meeting the ITT-ED definition, and only those with data available at the specified timepoint were analyzed. Without the number measured and information about missingness, readers cannot evaluate how representative the analyzed group was of those registered.

Finally, the list of outcomes spans measurement sites, scales, timepoints and subgroup analyses. That breadth makes prespecification and multiplicity handling relevant statistical issues, yet those procedures are not reported in the registry record. The study is therefore informative as an example of how design, endpoint definition, analysis population and reporting completeness shape interpretation, even when numerical conclusions cannot be drawn.

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.

A central limitation for interpreting the registry record is that the primary numerical result is not reported. The estimate, 95% confidence limits, p-value and number analyzed at Week 48 are not reported in the registry record. Numerical results for the listed secondary outcomes are also not reported. This prevents assessment of effect direction, magnitude and precision.

The substudy is non-randomized, and its groups were carried over from parent studies. Although the parent studies are described as randomized, the substudy’s own allocation is not. The registry record does not report baseline characteristics for the analyzed groups or enough detail to evaluate whether important differences were adjusted for.

The primary outcome description identifies ANCOVA but does not state which covariates were used, how the model was specified, or whether diagnostic checks or sensitivity analyses were performed. The record also says that only participants with data available at the specified timepoint were analyzed, but it does not report missing-data counts, reasons for missingness, or methods to address missing values.

Several secondary outcomes and analyses are listed, but multiplicity handling and the distinction between confirmatory and exploratory analyses are not reported in the registry record. The record does not report the substudy’s sample-size justification or interim-analysis plan. Finally, participants did not receive study medication in study 202094, which limits attribution of safety summaries to treatment administered within the substudy itself.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Bone Mineral Density in Human Immunodeficiency Virus Type 1 (HIV-1)-Infected Adult Subjects Switching From a Tenofovir Regimen to a Dolutegravir Plus Rilpivirine Regimen — NCT02478632.
• Registry record last updated 2020-10-19; status last verified 2020-09.
• Results first posted 2017-10-20.
• 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

NCT02478632 is a completed Phase 3, open-label, parallel-group, non-randomized substudy of bone mineral density among adults with HIV-1 infection. Enrollment was 102. The study began on 2015-06-12, and primary completion was recorded on 2016-09-16.

Participants carried over treatment groups from SWORD-1 or SWORD-2; the substudy itself did not administer study medication. The primary endpoint was percent change from baseline in total hip BMD at Week 48, assessed by dual energy X-ray absorptiometry. The registry describes an ANCOVA comparison in the ITT-ED population and specifies a 95% confidence interval.

The numerical primary and secondary outcome results are not reported in the registry record. Serious-event and other-event summaries list zero affected participants and zero deaths in each displayed group. Detailed missing-data handling, multiplicity procedures, and the substudy’s sample-size justification are not reported in the registry record.