Bioequivalence of Two Different Capsule Types of Dabigatran: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bioequivalence of Two Different Capsule Types of Dabigatran (NCT01290757), a randomized phase 1 study. The primary outcome is Area Under the Curve 0 to tz (AUC0-tz) of Total Dabigatran.

2. Clinical Question

This is a formulation comparison, not a comparison of dabigatran with no treatment or with another medicine. The products share the same named drug and dose, while differing in the capsule shell and drug-product batch described in the record. The crossover design means the study was organized around participants receiving treatments across periods, but the registry record does not report the treatment sequence details or the interval between periods. The posted result is therefore best read as evidence about the pharmacokinetic comparison in this study population and under its study conditions. It does not, by itself, answer questions about clinical outcomes or use in populations other than the healthy volunteers studied.

3. Trial Design

This was a Phase I, randomized, open-label, single-dose, replicate-design crossover study. Its official title describes two treatments and four periods. The registry record lists an enrollment of 180 and a completed status. The study began in 2011-01, and primary completion was in 2011-04. No additional calendar dates are reported.

Participants received oral dabigatran etexilate 150 mg capsules from the Capsugel test batch and the Qualicaps reference batch. The intended comparison was between those two capsule types, with the drug dose held at 150 mg. Because this was a crossover study, the treatment comparison was made within a study structured around participants receiving treatments across periods. The allocation is listed as randomized. Masking is listed as none, so the study was open-label.

The primary outcome was total dabigatran AUC0-tz, measured in ng*hr/mL over 60 hours. The registry describes an adjusted analysis that included treatment, period, and sequence as fixed effects and a random subject effect. The record also lists four secondary outcomes: total dabigatran AUC0-∞, free dabigatran AUC0-tz, free dabigatran Cmax, and free dabigatran AUC0-∞. Values for those secondary outcomes are not included in the posted results.

The registry lists 180 enrolled participants, while the primary endpoint summaries show 179 for Capsugel and 180 for Qualicaps. These are the counts attached to the posted endpoint summaries; the record does not explain the difference in those counts. Detailed treatment sequences, period-level completion, the timing between periods, eligibility criteria beyond the healthy-volunteer description, and the reasons for any missing endpoint observations are not reported in the registry record. Those omissions limit how fully the trial conduct can be reconstructed from the public entry.

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: In this crossover study, subjects were randomly assigned to one of the 2 sequences, with 2 treatment arms. In general terms, Ref-Test-Ref-Test and Test-Ref-Test-Ref. The aim was bioequivalence comparing (Ref) Dabigatran 150mg in the currently approved capsule shell (Qualicaps) to the test of Dabigatran 150mg in a new shell (Capsugel).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: This subject set, the pharmacokinetic set (PKS), includes all subjects of the treated set who provide at least one evaluable observation for at least one of the three PK endpoints of total dabigatran, which is obtained in a period without important protocol violations relevant to the evaluation of bioequivalence.

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.

Population: Treated set includes all patients who received at least one study treatment.

6. Primary Endpoint

The registry lists the primary outcome as Area Under the Curve 0 to tz (AUC0-tz) of Total Dabigatran, measured over 60 hours.

The primary endpoint was the area under the plasma concentration-time curve for total dabigatran from time 0 to tz, the time of the last quantifiable data point. AUC summarizes drug concentration over time: it reflects exposure across the observed concentration-time profile rather than a single concentration measurement. The registry reports the endpoint in ng*hr/mL and gives a timeframe of 60 hours.

Pharmacokinetic exposure is used in a bioequivalence study to compare how two product versions deliver the same drug. If the exposure measures are sufficiently similar under the study’s chosen criteria, that supports a conclusion about similarity of drug exposure for the formulations being compared. The registry record states that the study objective was to demonstrate bioequivalence, but the acceptance boundaries for that conclusion are not reported in the registry record. The posted data should therefore be interpreted through the reported sponsor analysis, rather than by applying an unstated threshold.

The endpoint is described as adjusted for treatment, period, and sequence, with those terms treated as fixed effects and subject included as a random effect. This reflects the crossover structure: exposure can vary between people and may also be affected by the period or sequence in which a treatment is received. The primary endpoint concerns total dabigatran, while the listed secondary outcomes concern additional total- or free-dabigatran exposure measures and peak concentration. The registry does not report the results of those secondary measures in this summary, so they cannot be used to broaden the primary endpoint interpretation.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The principal inferential result is a mixed-model analysis. This choice is consistent with the crossover structure described in the registry: participants received treatments across periods, and the analysis included terms for treatment, period, sequence, and subject. The registry says treatment, period, and sequence were fixed effects and subject was a random effect. The public record does not provide the full model equation, covariance structure, estimation method, or diagnostics, so those technical details are not reported in the registry record.

Why use a mixed model?

In a crossover study, measurements from the same participant are related rather than independent. A random subject effect is a way to represent participant-level differences and account for that repeated structure. Treatment terms estimate the contrast of interest, while period and sequence terms allow the analysis to account for features of the study schedule. This is a general explanation of the model concepts; the registry does not report enough detail to determine how every modelling choice was implemented.

The reported primary endpoint is a geometric mean measure, and the adjusted treatment comparison is reported as a geometric mean ratio. Geometric summaries are commonly used for positive pharmacokinetic measurements that may be unevenly distributed. A ratio expresses the estimated exposure for one treatment relative to the other on a multiplicative scale. The posted ratio is 126.33%, with a 90% confidence interval from 119.45% to 133.60%. The interpretation of this interval for bioequivalence depends on acceptance boundaries specified by the study’s statistical plan. Those boundaries are not reported in the registry record.

What does the confidence interval show?

A confidence interval communicates uncertainty in an estimated comparison. A narrower interval indicates greater precision under the model and data used, while a wider interval indicates more uncertainty. It does not describe the spread of individual participant values. Here, the sponsor reports a 90% interval for the adjusted geometric mean ratio and a separate upper 95% confidence limit for a linearized criterion. These are different reported quantities and should not be treated as interchangeable. The registry does not state the acceptance boundary for the linearized criterion or explain its scale in enough detail for an independent decision-rule assessment.

The arm summaries also include geometric coefficients of variation: 32.32 for Capsugel and 49.83 for Qualicaps. A coefficient of variation describes relative dispersion around a summary, rather than uncertainty in the estimated treatment contrast. In general, dispersion matters because substantial variability among measurements can make a treatment comparison less precise. These arm-level dispersion values do not replace the model-based confidence interval for the adjusted ratio.

How should the analysis population be read?

The registry describes the pharmacokinetic set as treated participants with at least one evaluable observation for at least one of the total-dabigatran endpoints, obtained in a period without important protocol violations relevant to evaluating that endpoint. The description ends before all criteria are stated, so the complete analysis-population definition is not reported in the registry record. The primary endpoint summaries show 179 for Capsugel and 180 for Qualicaps. The record does not explain why the counts differ, nor does it provide a participant flow accounting for all enrolled participants.

In a crossover study, incomplete data for one period can affect which observations contribute to a treatment comparison. The registry’s abbreviated description gives a general eligibility rule for the pharmacokinetic set, but it does not explain the handling of incomplete sequences, protocol deviations, or observations excluded from analysis. It also does not report a sensitivity analysis. Readers should therefore distinguish the posted endpoint counts from the total enrollment and avoid assuming that every enrolled participant contributed complete data to every endpoint.

What is not specified?

The statistical plan, including the prespecified bioequivalence margins, is not reported in the registry record. The record also does not report whether interim analyses were conducted, how missing observations were handled beyond the partial pharmacokinetic-set description, or what multiplicity procedure, if any, was used for the outcome measures and analyses. No sample-size justification is reported in the registry record. The sample-size illustrations elsewhere on the page are planning exercises based on their stated assumptions, not observed power and not evidence of the design’s original rationale.

These omissions matter for interpretation, not because they invalidate the posted result, but because they constrain independent verification. A reader can describe the adjusted estimate, confidence interval, and model terms that are reported. A reader cannot reconstruct the full statistical plan or confirm all decision rules from the registry entry alone. General methodological explanations clarify what the reported methods mean; they do not fill in unreported protocol details.

10. Analyses Reported by the Sponsor

The registry reports a sponsor mixed-model analysis, including an adjusted geometric mean ratio of 126.33% and a 90% confidence interval from 119.45% to 133.60%. It also reports an upper 95% confidence limit of -0.082 for the linearized criterion. These are the posted inferential results. The arm-level summaries of 1148.0 and 908.5 are descriptive geometric means, not a substitute for the adjusted analysis.

No separate recalculated treatment comparison is listed in the posted results. The surrounding tables and plots present the posted values rather than a newly calculated inferential estimate. The sponsor analysis should therefore be read as the registry’s reported model-based comparison, with its own stated measures and confidence levels. The record does not give the acceptance limits needed to verify a bioequivalence decision from those estimates alone, and it does not list p-values for the reported analyses.

The registry does not flag a discrepancy between the posted analysis and a recalculation. That does not supply missing protocol details or establish whether a decision threshold was met. It means the appropriate account of the result is the sponsor’s stated analysis and the limits of interpretation visible in the record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and eight analyses. The primary outcome is total dabigatran AUC0-tz. Four secondary outcomes are named: total dabigatran AUC0-∞, free dabigatran AUC0-tz, free dabigatran Cmax, and free dabigatran AUC0-∞. The names of the remaining measures are not reported in the record. Results for the secondary outcomes are also not included in this account.

When researchers examine several outcomes or analyses, the chance of finding an apparently notable result can change across the collection of tests. A multiplicity plan may specify which results are confirmatory, how comparisons are prioritized, or whether significance thresholds are adjusted. This is a general statistical consideration; the registry record does not report whether a multiplicity procedure was planned or used. It also does not identify which of the eight analyses were confirmatory versus exploratory. Readers should not infer that all analyses share the same inferential status.

The registry gives a partial description of the pharmacokinetic analysis population, but the full eligibility and exclusion criteria are not reported in the registry record. It states that included observations came from periods without important protocol violations relevant to evaluating the endpoint. The handling of missing observations, incomplete treatment sequences, or participants excluded from particular analyses is not described in enough detail to reconstruct the analysis dataset. The primary endpoint summaries list 179 for Capsugel and 180 for Qualicaps, while enrollment was 180; the registry does not explain the count difference.

No interim-analysis plan, missing-data sensitivity analysis, or complete statistical analysis plan is reported in the registry record. The reason for any absent endpoint observations is not reported either. These gaps leave uncertainty about how the analysis set was finalized and how robust the result was to alternative data-handling choices. They do not change the values reported, but they limit independent assessment of the analysis process and the scope of conclusions that can be drawn from the posted entry.

12. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for 180 participants at risk in each treatment group and zero participants affected in each group. It reports 37 participants affected by other adverse events in the Qualicaps group and 34 in the Capsugel group, with 180 at risk in each group. The record does not list serious adverse event terms. Death counts are not reported in the registry record.

These figures are counts of participants recorded as affected, not a detailed account of event timing, duration, severity, or cause. The entry does not provide a breakdown of the other adverse events in this summary, so the kinds of events cannot be characterized. Nor does it establish whether an event was considered related to treatment. The absence of reported serious adverse events in these groups is a statement about the registry’s posted count, not proof that no serious event could occur.

Safety figures in a short, single-dose study in healthy volunteers have a limited scope. The record does not provide enough detail to compare event patterns by period or sequence, or to assess follow-up beyond the information displayed. The numbers should therefore be read as the reported participant counts in this study, not as a broad estimate of safety in other populations or settings.

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

The posted result describes a pharmacokinetic comparison between two 150 mg dabigatran etexilate capsule types in healthy volunteers. The sponsor’s mixed model reports an adjusted geometric mean ratio of 126.33%, with a 90% confidence interval from 119.45% to 133.60%. It also reports an upper 95% confidence limit of -0.082 for the linearized criterion. These quantities convey an estimated comparison and its uncertainty under the analysis described in the registry.

The result is not, on its own, a complete record of the bioequivalence decision process. The registry states that demonstrating bioequivalence was the objective, but does not report the acceptance boundaries needed to judge the estimate against the study’s decision rule. Without those boundaries and the full statistical plan, readers cannot independently verify a criterion-based conclusion from the public entry. The reported confidence interval is informative, but it cannot stand in for an unreported margin.

The endpoint measures total dabigatran exposure over the stated 60-hour period. It is not a measure of clinical benefit, long-term outcomes, or the experience of patients with a medical condition. This study enrolled healthy male and female volunteers and used single-dose administration. Its direct subject is the comparison of the Capsugel and Qualicaps capsule types under the trial conditions, not every product batch, dosing situation, or patient population.

The posted adverse-event counts give limited safety information for the study groups. They do not describe event types in enough detail here, establish causation, or support conclusions beyond the study and reporting shown. Likewise, the secondary pharmacokinetic outcomes are listed but their results are not reported in this summary. The most defensible reading is therefore specific: the registry reports the sponsor’s adjusted primary exposure comparison, but important details for independently judging the protocol decision rule and extending the result are unavailable.

14. Why This Trial Matters Statistically

This trial is statistically instructive because it compares formulations using a randomized crossover design rather than comparing unrelated groups in parallel. In a crossover, the same participant can contribute observations under both treatments, allowing the analysis to account for individual-level differences. The registry’s mixed model explicitly includes a random subject effect, as well as fixed effects for treatment, period, and sequence. Those features help explain why the adjusted result should be distinguished from the separate arm-level geometric means.

It also illustrates why a point estimate is not enough to interpret a bioequivalence study. The reported adjusted geometric mean ratio is accompanied by a 90% confidence interval, and a separate linearized criterion result is also posted. The confidence interval describes estimation uncertainty; the formal decision depends on a prespecified acceptance rule. Since that rule is not reported in the registry record, the statistical interpretation must remain bounded by the available information rather than assuming a conventional margin.

The entry also shows the value of distinguishing enrollment from endpoint-specific counts. It lists 180 participants enrolled, while the primary summaries list 179 and 180. Without a participant flow or explanation, the count difference cannot be resolved from the registry. Similarly, the primary result is available while the secondary outcomes are named but not presented. This separation matters when assessing completeness and the scope of claims.

More broadly, the record demonstrates how a registry can communicate a model-based estimate while leaving important analysis details unavailable. The result can be described precisely, yet full independent evaluation requires the decision boundaries, complete analysis-population rules, and missing-data and multiplicity procedures. Statistical importance lies in reading the reported estimate accurately and recognizing what the posted record cannot answer.

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 registry record gives only a partial view of the protocol and analysis plan. Although the official title describes a randomized, open-label, single-dose, replicate-design crossover study with two treatments and four periods, the treatment sequences, period schedule, and interval between periods are not reported in the registry record. Those details would help readers assess how the crossover was implemented and what assumptions underlie the analysis.

The acceptance boundaries for bioequivalence are not reported. The record supplies an adjusted ratio, its 90% confidence interval, and an upper 95% confidence limit for the linearized criterion, but not the decision limits against which those values should be interpreted. It also does not report the full statistical analysis plan, multiplicity approach, interim-analysis information, missing-data procedures, or sample-size justification. The abbreviated pharmacokinetic-set definition and absence of participant flow leave the reasons for the endpoint counts unclear.

The primary endpoint concerns total dabigatran AUC0-tz. Although four secondary outcomes are listed and the record indicates six outcome measures and eight analyses, their results are not presented in this summary. The entry also provides limited adverse-event detail: serious event counts are posted, but serious event terms are absent, and deaths are not reported in the registry record. Other adverse events are summarized only as affected-participant counts.

The trial population and intervention are narrow: healthy volunteers, single-dose administration, and the two named capsule types. The registry entry does not establish how the findings apply to other populations, dosing conditions, or product batches. These limitations define the reach of the posted information; they do not supply evidence for or against results that are not reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Bioequivalence of Two Different Capsule Types of Dabigatran — NCT01290757.
• Registry record last updated 2014-05-19; status last verified 2013-12.
• Results first posted 2012-06-29.
• 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 I trial compared oral, single-dose dabigatran etexilate 150 mg capsules using the Capsugel and Qualicaps shells. It was randomized, open-label, and described as a two-treatment, four-period crossover study. The registry lists 180 enrolled healthy male and female volunteers.

The primary endpoint was total dabigatran AUC0-tz over 60 hours. The sponsor’s mixed-model analysis reported an adjusted geometric mean ratio of 126.33%, with a 90% confidence interval from 119.45% to 133.60%, and an upper 95% confidence limit of -0.082 for the linearized criterion. The registry does not report the bioequivalence acceptance boundaries, so the decision rule cannot be independently checked from this entry.

Four secondary outcomes are listed, but their results are not reported. Serious adverse events affected zero participants in each group; other adverse events affected 37 in Qualicaps and 34 in Capsugel. The registry does not report serious event terms or death counts. Important protocol and analysis details, including the full analysis-population definition and missing-data procedures, are not reported in the registry record.