Bioavailability Study of 10 Milligram (mg) and 5 mg Tablets Versus Conventional Tablets of Dolutegravir: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bioavailability Study of 10 Milligram (mg) and 5 mg Tablets Versus Conventional Tablets of Dolutegravir (NCT03095638), a randomized phase 1 study. The primary outcome is Area Under the Plasma Concentration-time Curve From Time of Dose Extrapolated to Infinite Time (AUC[0-infinity]) of DTG for Part 1.

2. Clinical Question

The study’s focus was pharmacokinetic comparison after single doses, rather than treatment of HIV infection or assessment of clinical benefit. The use of a crossover design means participants were assigned to treatment sequences and received more than one formulation across study periods. In Part 1, the two formulations were administered in different orders. In Part 2, the three conditions were the dispersible tablet taken as a dispersion, the dispersible tablet taken directly by mouth, and the conventional 25 mg tablet. The record describes washout periods between dosing periods, but the primary result concerns Part 1 only.

3. Trial Design

This was a Phase 1, randomized, open-label, single-dose crossover study, recorded as completed. The registry record describes two parts and an enrollment of 38. It gives an approximate participation target of 14 healthy subjects for Part 1 and approximately 24 for Part 2. The start date was 2017-05-03 and primary completion was 2017-06-23.

Part 1 used a two-period crossover. Participants were randomized to receive the conventional 10 mg formulation, administered as five tablets, and the conventional 50 mg tablet in one of two sequences. Each dose was given in the fasting state. The planned washout between dosing periods was at least 7 days. This design allows each participant with evaluable observations in both periods to contribute information about both formulations, reducing the influence of stable differences between participants on the comparison. That statistical advantage depends on usable within-participant data and does not remove possible period, sequence, or measurement effects.

Part 2 used a three-period crossover and randomized participants among sequences involving the 5 mg dispersible tablets, administered either dispersed with water or directly to the mouth, and the conventional 25 mg tablet. The described washout period between doses was at least 7 days. A follow-up visit was planned 7-10 days after the last dose. The primary endpoint reported for this analysis is not a Part 2 outcome; it is the Part 1 AUC measure.

Open-label means that masking was not used, according to the registry record. Because the primary endpoint is a laboratory-based pharmacokinetic measure, the open-label status does not by itself determine whether the measured concentrations are biased. It remains relevant to the overall design, especially for any outcomes susceptible to participant or investigator expectations. The record does not report a detailed statistical analysis plan, a sample-size justification, or how period and sequence effects were handled in the primary analysis.

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 106 participants were screened however 68 were screen failures the reasons for which were did not meet inclusion/exclusion, adverse event, lost to follow up, investigator discretion, withdrew consent and screened but enrollment target reached prior to enrollment. All of the 38 enrolled participants completed the study.

Only the first 6 of 8 groups are shown.

P1: Period 1 (4 Days)

P1: Period 1, Washout (7 [-4 Hours])

P1: Period 2 (4 Days)

P1: Period 2, Washout (7[-4 Hours])

Analysis population for the primary outcome, as defined in the registry: PK Population

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 Area Under the Plasma Concentration-time Curve From Time of Dose Extrapolated to Infinite Time (AUC[0-infinity]) of DTG for Part 1, measured over Pre-dose, 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 12, 16, 24, 48, and 72 hours post-dose in Part 1.

The primary endpoint is the area under the plasma concentration-time curve from the time of dose extrapolated to infinite time, written as AUC[0-infinity]. It measures overall dolutegravir exposure after dosing, including the observed concentration-time profile and an extrapolated contribution beyond the last measured time. The registry record gives the unit as hours*nanograms/mL (h*ng/mL) and identifies the reported summary as a geometric mean with geometric coefficient of variation.

Blood samples were collected before dosing and at the stated post-dose time points through 72 hours in Part 1. The record says pharmacokinetic parameters were calculated using standard non-compartmental analysis, based on actual sampling times, with Phoenix WinNonlin Version 6.4 or higher. AUC is used in relative bioavailability comparisons because it summarizes systemic exposure over time; comparing formulations can indicate whether their measured exposure profiles are similar under the study conditions. It does not, by itself, describe clinical efficacy or safety.

The primary outcome population was the pharmacokinetic population: participants in the All Subjects Population for whom a pharmacokinetic sample was obtained and evaluable assay results were available. The registry-reported comparison gives a ratio of 1.008 and a 90% confidence interval from 0.863 to 1.179. The record does not report a prespecified acceptable range for the ratio, so the result should be read as an estimate with uncertainty rather than as a formal equivalence finding.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary comparison concerns a continuous pharmacokinetic measure summarized with geometric means. A geometric mean is commonly used for positive measurements that are naturally analyzed on a multiplicative scale, such as exposure measures. On that scale, the comparison is expressed as a ratio rather than as an arithmetic difference. A ratio of 1 represents equal values on the ratio scale. The posted ratio is 1.008, and the posted 90% confidence interval is 0.863 to 1.179.

What does the confidence interval communicate?

A confidence interval describes uncertainty around an estimate under the statistical procedure used to construct it. The 90% interval in the registry record spans values below and above 1. That means the result does not isolate a single relative exposure value. A confidence interval is not a probability statement about the particular true value after the data have been observed, and its interpretation depends on the method and assumptions used. The registry record does not report the primary comparison method, so the interval’s exact construction cannot be evaluated from the posted description alone.

For equivalence or bioequivalence, a ratio and interval are generally judged against prespecified limits. Such limits define how far apart the formulations may be while still satisfying the study’s chosen criterion. No such limits are reported in the registry record. It is therefore possible to describe the estimate and interval, but not to claim that a formal equivalence criterion was satisfied. A point estimate near 1 is not sufficient by itself to establish equivalence.

Why is a crossover design relevant?

In a crossover study, participants receive multiple treatments in separate periods, and randomization assigns the order. A within-participant comparison can control for characteristics that remain stable across periods. This can improve the precision of a comparison relative to a design that compares entirely separate groups, provided the data are complete and the analysis accounts appropriately for the design. Crossover analyses also need to consider period and sequence effects and whether the washout period was adequate for the question. The record describes at least 7 days of washout but does not report how period or sequence effects were modeled, or whether carryover was assessed.

The registry describes Part 1 as a two-period crossover and Part 2 as a three-period crossover. The primary result is for Part 1, comparing five 10 mg conventional tablets with one 50 mg conventional tablet. Part 2’s formulation and administration comparisons should not be treated as part of this primary estimate. The record’s primary endpoint population is limited to participants with a pharmacokinetic sample and evaluable assay results, rather than being defined simply as everyone randomized.

How should variability summaries be read?

The arm summaries report geometric coefficients of variation. This statistic describes relative spread in the pharmacokinetic measurements around their geometric mean. It is not an uncertainty interval for the treatment ratio and is not a measure of the precision of the comparison by itself. In general, greater variability can make an estimated treatment contrast less precise, especially when the number of evaluable paired observations is limited. The posted arm-level summaries do not show the within-participant covariance needed to reconstruct a crossover comparison from marginal summaries alone.

The registry says pharmacokinetic parameters were calculated by standard non-compartmental analysis using actual sampling times and Phoenix WinNonlin Version 6.4 or higher. Non-compartmental analysis estimates exposure and other pharmacokinetic parameters directly from the concentration-time observations without requiring a specified compartment model. The record describes blood collection from pre-dose through 72 hours after dosing for the primary Part 1 outcome. The analysis population required an obtained pharmacokinetic sample and evaluable assay results. Further details about handling missing samples, values below assay limits, or incomplete concentration profiles are not reported in the registry record.

What do the planning illustrations mean?

The posted sample-size illustrations are described as planning calculations based on the observed effect, not as the trial’s original design justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. Such calculations show how planning requirements would vary under those assumptions. They are not observed power, do not show the probability that the completed study detected an effect, and do not establish that the original study was designed around the observed result. The registry record does not report the original sample-size justification or a prospective power target.

For binary secondary outcomes, the posted recalculations are unadjusted comparisons based on the arm summaries. In general, an unadjusted analysis does not account for multiple testing or other covariates unless such adjustment is specified. The registry does not report a multiplicity procedure for the secondary outcomes. These calculations should therefore be treated as descriptive or exploratory rather than as confirmatory evidence. The original sponsor analysis takes precedence if a discrepancy is identified; the registry record does not flag a discrepancy for the primary result.

10. Analyses Reported by the Sponsor

The registry record reports a primary ratio of 1.008, with a 90% confidence interval from 0.863 to 1.179. It does not report a primary p-value or the statistical method used to obtain this comparison. The result is presented as a reported analysis, rather than as a newly reconstructed comparison from the two arm-level geometric means.

The record does not flag a discrepancy for the primary result. For the posted secondary binary outcomes, the additional comparisons are unadjusted recalculations from the reported arm summaries. They should not be treated as replacing a sponsor analysis if one is reported or as adjusting for the crossover structure. No sponsor comparison is listed for those two secondary outcomes. The primary result remains the registry-reported analysis, and the secondary recalculations have a more limited, descriptive role.

Interpretation should remain tied to the information actually recorded: a ratio and interval for the primary pharmacokinetic outcome, without a reported equivalence margin or a detailed analysis method. The result can be described as an estimate of relative exposure with uncertainty. It cannot be converted into a formal equivalence conclusion solely by treating the estimate as close to 1.

11. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a procedure for controlling multiplicity across the secondary outcomes. Multiplicity arises when several outcomes or comparisons are examined: without a prespecified strategy, the chance of at least one apparently unusual result can increase. This matters particularly for secondary findings, which are often less strongly protected against false-positive interpretations than a primary outcome. The secondary results should be read as exploratory unless a confirmatory role and adjustment plan are documented.

Missing-data handling is not reported in the registry record. The primary pharmacokinetic population is defined as participants in the All Subjects Population with a pharmacokinetic sample and evaluable assay results. This definition indicates that availability and evaluability of pharmacokinetic data affected inclusion in that analysis, but the record does not state how many randomized participants lacked evaluable results, why any data were unavailable, or how incomplete profiles were treated. It also does not report whether imputation was used.

The crossover design makes completeness of paired measurements relevant: a within-participant comparison depends on usable data for the required treatment periods. The record does not give a detailed account of how participants with incomplete periods were handled, whether any participants were excluded from particular endpoints, or whether sensitivity analyses assessed the effect of missing observations. These omissions limit assessment of possible selection bias in the primary analysis.

The record describes a pharmacokinetic analysis based on actual sampling times and evaluable assay results, but it does not give the full analysis plan, model specification, or rules for handling assay values that could not be quantified. It also does not report interim analyses or a sample-size justification. The absence of these details should not be taken to mean that no such procedures existed; they are simply not reported in the registry record.

The primary endpoint was defined for the pharmacokinetic population, while the original randomized population and the analysis population may not be identical. The registry record does not provide enough participant-flow detail to assess that distinction fully. These limitations affect transparency and reproducibility, not the numerical value of the posted estimate itself.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse event figures report zero affected participants in each posted group: DTG 10 mg, DTG 50 mg, DTG 5 mg (Test 1), DTG 5 mg (Test 2), and DTG 25 mg (Reference). Deaths are also recorded as zero in each group. The registry lists no serious adverse event terms.

These figures describe serious events recorded during the study, not the probability of serious events under longer-term use or in a broader population. The study involved single-dose pharmacokinetic comparisons in healthy adults, and the registry record does not provide a basis for extending the observed safety counts beyond those conditions. Zero reported events does not prove that the risk is zero; uncommon events may not appear in a study of this size and duration.

Groups in a crossover study may include participants who received more than one treatment, so group counts should not be added to infer a total number of distinct people. The serious-event figures are best read separately by the posted treatment group. They provide a limited description of observed serious events and do not substitute for a comprehensive assessment of safety.

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

The primary result is an estimate of relative dolutegravir exposure after two conventional tablet regimens in fasted healthy adults. The reported ratio is 1.008, and the 90% confidence interval ranges from 0.863 to 1.179. The estimate is close to 1, but the interval indicates uncertainty around that point estimate. The registry record does not report an equivalence margin, so the result cannot on its own support a formal statement that the formulations are equivalent.

The endpoint measures plasma exposure, not whether a treatment prevents or controls HIV infection, improves symptoms, or changes long-term outcomes. The trial was a single-dose Phase 1 pharmacokinetic study. Its findings do not establish what would occur with repeated dosing, in people with HIV infection, under fed conditions, or in populations not described in the record.

The comparison is also specific to the formulations and administration conditions studied. The 10 mg conventional tablets were given as five tablets and compared with one 50 mg conventional tablet in Part 1. Part 2 examined the 5 mg dispersible tablets by two administration approaches against a conventional 25 mg tablet, but those comparisons are separate from the primary Part 1 result. No conclusions about those Part 2 comparisons should be inferred from the Part 1 ratio.

The confidence interval includes values below and above 1, and no decision margin is recorded. This means the displayed estimate should not be restated as proof that exposure is identical or that a meaningful difference has been ruled out. A near-1 point estimate does not remove uncertainty. Likewise, lack of a reported p-value does not invalidate the estimate; the ratio and interval are the available primary comparison information.

The result is most appropriately understood as a registry-reported pharmacokinetic comparison under defined study conditions, with limits in precision and analysis detail. It is not medical advice, a treatment recommendation, or a substitute for a full assessment of clinical evidence.

14. Why This Trial Matters Statistically

This study illustrates why pharmacokinetic endpoints and clinical outcomes answer different questions. AUC[0-infinity] summarizes systemic exposure based on measured concentrations and an extrapolated tail. It is a relevant quantity for relative bioavailability, but it does not directly measure a patient-centered outcome. Keeping the endpoint’s meaning distinct from clinical efficacy prevents a technical exposure comparison from being overstated.

The crossover design is statistically notable because it compares formulations within participants across periods. That approach can reduce noise from stable differences between individuals, while making the analysis dependent on the quality and completeness of paired measurements. The registry describes randomized treatment sequences and washout periods, but does not report enough analysis detail to evaluate period, sequence, or missing-data handling.

The reported ratio is close to 1, yet its confidence interval spans values on either side of 1. This makes the distinction between a point estimate and its precision central to interpretation. It also shows why a formal equivalence claim needs a decision margin set in advance: closeness of the estimate alone is not a statistical criterion.

The record also offers a useful example of the limitations of secondary analyses. Several outcomes can characterize pharmacokinetics or laboratory findings, but unadjusted comparisons across multiple endpoints require caution. Small event counts and a lack of reported multiplicity procedures reduce the weight that should be placed on isolated secondary patterns.

Finally, the study’s narrow, single-dose setting helps define rather than broaden its conclusions. The analysis is informative about the posted exposure comparison in fasted healthy adults; it does not settle questions about repeated dosing, clinical effectiveness, or broader safety. That distinction is a core part of sound biostatistical interpretation.

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 primary analysis is reported with a ratio and confidence interval, but the registry record does not identify the statistical method used to produce them. It also does not state the numerator and denominator explicitly in the comparison description. This restricts interpretation of direction and makes independent reproduction of the primary analysis difficult.

No equivalence margin is reported. Without the prespecified acceptance limits and the study’s decision rule, the ratio and interval cannot be used to conclude that the formulations met an equivalence criterion. The record also does not report the original sample-size justification, so the rationale for the planned enrollment cannot be assessed from the registry entry.

The primary endpoint applies to the defined pharmacokinetic population, which required an obtained sample and evaluable assay results. Participant flow into that population and the handling of incomplete concentration profiles are not reported. The crossover structure makes paired data and period-specific completeness relevant, but the registry does not describe the model’s treatment of period or sequence effects, possible carryover, or missing observations.

The study was open-label, used single doses, and was conducted in fasted healthy adults. These design features define the scope of the result and limit generalization to other populations and dosing conditions. The Part 1 primary result does not answer the separate Part 2 questions about dispersible tablet administration.

The registry does not report a multiplicity strategy for secondary endpoints, nor a detailed analysis plan for those outcomes. The secondary binary comparisons are unadjusted and include sparse counts, so they should not be treated as confirmatory evidence. Serious adverse event counts are also limited to the study observation and cannot characterize uncommon or longer-term risks. These limitations concern the amount and scope of available evidence, rather than changing the posted primary estimate.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Bioavailability Study of 10 Milligram (mg) and 5 mg Tablets Versus Conventional Tablets of Dolutegravir — NCT03095638.
• Registry record last updated 2019-07-23; status last verified 2019-07.
• Results first posted 2018-12-21.
• 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

The registry record describes a completed Phase 1 randomized, open-label crossover study of dolutegravir tablet formulations in fasted healthy adults. Part 1 compared five conventional 10 mg tablets with one conventional 50 mg tablet as single doses. Part 2 assessed 5 mg dispersible tablets administered in two ways against a conventional 25 mg tablet.

The primary endpoint was Part 1 AUC[0-infinity] in the pharmacokinetic population. The reported ratio was 1.008, with a 90% confidence interval from 0.863 to 1.179. The registry record does not report an equivalence margin or the detailed primary analysis method, so the estimate should be interpreted as a relative exposure result with uncertainty, not as a formal equivalence finding.

Secondary pharmacokinetic and toxicity outcomes offer additional context but lack a reported multiplicity plan. Serious adverse events and deaths were recorded as zero in each posted group. The conclusions are limited to the study’s single-dose, fasting, healthy-adult setting.