A Study in Healthy People to Compare 2 Different Formulations of BI 1015550 Taken With or Without Food: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study in Healthy People to Compare 2 Different Formulations of BI 1015550 Taken With or Without Food (NCT05428436), a randomized phase 1 study. The primary outcome is Area Under the Concentration-time Curve of BI 1015550 in Plasma Over the Time Interval From 0 to the Last Quantifiable Data Point (AUC0-tz).

2. Clinical Question

The registry describes the trial’s main objective as investigating the relative bioavailability of the intended commercial formulation compared with Trial Formulation 2, and the effect of food on the pharmacokinetics of the intended commercial formulation. These are related but distinct comparisons: the first contrasts formulations under fasting conditions, while the second contrasts food conditions for the same formulation.

The posted primary endpoint is an exposure measure based on plasma concentrations collected over time. It is not a clinical health outcome. The trial enrolled healthy people and does not report a disease-treatment comparison or a clinical benefit endpoint. The record identifies the study as Phase 1, randomized, open-label, and crossover. Its results therefore describe pharmacokinetic comparisons in the studied setting, rather than establishing how a treatment performs in a broader clinical population.

3. Trial Design

This was a randomized, open-label, single-dose, three-way crossover trial in healthy people. The registry lists the study as Phase 1, with 24 participants enrolled, and reports that it is completed. The recorded study start was 2022-08-15 and primary completion was 2022-09-30. Boehringer Ingelheim is listed as sponsor.

In a crossover design, participants receive more than one treatment condition in separate periods, so comparisons can be made within individuals. That is the general rationale for this design: each participant can serve as their own reference, which can reduce the influence of stable differences between people. A crossover comparison still depends on the conduct and analysis of periods and sequences. The registry record does not report the statistical model, any carryover assessment, or period-specific analysis details, so those aspects cannot be characterized further.

The three conditions were a fasted reference treatment using Trial Formulation 2, a fasted intended commercial formulation, and a fed intended commercial formulation. The reference consisted of three 6 milligram film-coated tablets taken after an overnight fast of at least 10 hours. The fasted intended commercial formulation was one 18 milligram film-coated tablet taken after an overnight fast of at least 10 hours. The fed treatment was one 18 milligram film-coated tablet taken following a high-calorie breakfast. Each participant’s treatment order followed one of the listed three-period sequences.

Randomization concerns the assigned treatment sequence; masking was listed as none, so the study was open-label. The record does not describe allocation concealment or other implementation details. The arm-level pharmacokinetic results include 24 observations for the fasted reference and 23 for each intended commercial formulation condition. The reason for the difference in these analysis counts is 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: All subjects were screened for eligibility prior to participation in the trial. Subjects attended a specialist site which ensured that they (the subjects) strictly met all inclusion and none of the exclusion criteria. Subjects were not to be allocated to a treatment group if any of the entry criteria were violated.

Period 1

Period 1 - 2 washout

Reasons for not completing this period:

Period 2

Period 2 - 3 washout

Analysis population for the primary outcome, as defined in the registry: Pharmacokinetic parameter analysis set (PKS): This set included all subjects who were treated with at least one dose of trial drug and who provided at least one Pharmacokinetic (PK) endpoint that was defined as primary or secondary and was not excluded due to a protocol deviation relevant to the evaluation of PK or due to PK non-evaluability.

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 (TS): The treated set included all subjects who were treated with at least one dose of trial drug.

6. Primary Endpoint

The registry lists the primary outcome as Area Under the Concentration-time Curve of BI 1015550 in Plasma Over the Time Interval From 0 to the Last Quantifiable Data Point (AUC0-tz), measured over Within 3 hours before and 0.5, 0.75, 1, 1.25, 1.5, 1.75, 2, 2.5, 3, 4, 6, 8, 12, 24, 48, 72, 96, 120, 144 hours following drug administration..

The primary endpoint was the area under the plasma concentration-time curve for BI 1015550 from time zero to the last quantifiable data point, abbreviated AUC0-tz in the registry record. The endpoint describes accumulated plasma exposure across the observed concentration-time profile. It is reported in hours*nanomol/Liter, with a geometric mean as the summary measure and geometric coefficient of variation as the measure of dispersion.

In general, an area under a concentration-time curve is used in pharmacokinetic studies because it combines concentration information across multiple sampling times into an exposure measure. Comparing this measure between formulations can address relative bioavailability, while comparing the same formulation under fed and fasted conditions can address whether food condition is associated with a different exposure profile. These are pharmacokinetic questions, not direct measures of a clinical outcome.

For this endpoint, blood sampling was scheduled within 3 hours before administration and at multiple times after dosing, from 0.5 through 144 hours. The registry describes the primary analysis population as a pharmacokinetic parameter analysis set, including treated participants with at least one primary or secondary pharmacokinetic endpoint that was not excluded for a relevant protocol deviation. Further details of the endpoint calculation and exclusions are not reported in the registry record.

The record also lists two secondary outcomes concerning AUC extrapolated to infinity for BI 1015550 and R-BI 1015550. The primary endpoint remains the finite-interval AUC to the last quantifiable point; it should not be conflated with those secondary measures.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is summarized using geometric means and geometric coefficients of variation. In general, geometric summaries are commonly used for positive pharmacokinetic measurements that can be skewed, because analysis on a logarithmic scale can make relative comparisons more appropriate than comparisons of raw arithmetic averages. The registry does not report the data distribution or the reason for selecting this summary.

What does a geometric mean ratio express?

In general, a geometric mean ratio compares the geometric mean exposure under one condition with the geometric mean under another. It expresses a relative contrast rather than an absolute difference in exposure. The registry reports a ratio of 121.12 for fasted intended commercial formulation versus fasted Trial Formulation 2, and 108.13 for fed versus fasted intended commercial formulation. The record does not state whether these values are displayed on a percentage scale or describe the calculation convention in detail, so the reported values are best quoted as recorded rather than rescaled.

The arm-level geometric means are 2290.0, 2740.0, and 3000.0 hours*nanomol/Liter for the fasted reference, fasted intended commercial formulation, and fed intended commercial formulation, respectively. Their geometric coefficients of variation are 194.0, 34.7, and 31.2. In general, a coefficient of variation describes dispersion relative to the scale of the measurement. It is not a confidence interval, and it does not by itself show how certain a between-condition estimate is.

How should the confidence intervals be read?

A confidence interval expresses statistical uncertainty around an estimated quantity. In general, a 90% confidence interval is generated by a procedure intended to cover the underlying parameter at the stated rate across repeated uses of that procedure, under its assumptions. It is not a guarantee that the true value lies in the interval, and it is not a range containing a fixed proportion of individual participants’ measurements. The registry reports 90% intervals for both comparisons: 86.14 to 170.30 for the formulation comparison and 76.68 to 152.48 for the food-condition comparison.

The intervals are wide relative to their point estimates, which signals imprecision in the reported relative comparisons. Their limits communicate a range of estimates compatible with the analysis procedure; they should not be replaced by a binary interpretation based only on whether an interval includes a particular value. Although equivalence assessments often use confidence intervals and prespecified bounds, no such bounds are reported in the registry record. The record also does not report a formal equivalence conclusion or p-value.

Why does the crossover design matter?

In general, a crossover design compares conditions received by the same participants in different periods. This can control for stable between-person differences, since each participant contributes information under more than one condition. Randomization of treatment order can help distribute sequence effects. But interpretation also depends on aspects such as period effects, possible carryover, washout, and the model used to account for the within-person structure. Specific washout details, model terms, and diagnostics are not reported in the registry record.

For these reasons, the two reported ratios should not be treated as if they came from unrelated parallel groups. The registry gives arm-level counts of 24 for the reference and 23 for each intended commercial formulation condition, while not reporting the paired participant-level data or exact analysis set by comparison. The reason for the unequal counts is not reported in the registry record.

What analysis details remain unclear?

The registry identifies the endpoint, summary type, dispersion measure, sampling timeframe, and reported comparison estimates. It does not report the statistical model, transformation details, confidence-interval method, assumptions checking, treatment of values below quantification limits, or the handling of missing pharmacokinetic observations. These omissions matter because they affect how fully an independent reader can reproduce or assess the comparisons.

General statistical practice distinguishes the uncertainty of an estimated effect from the variability of participant measurements. The geometric coefficient of variation addresses the latter, while the confidence interval addresses uncertainty in a treatment contrast. A variable endpoint can therefore have a different level of uncertainty than its arm-level dispersion alone suggests, particularly in a crossover analysis where within-person correlation matters. That correlation is not reported.

The record describes a sample-size planning illustration based on the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect set as a stated fraction of the observed effect. It also specifies that planning values are participants per arm except for time-to-event calculations. Numerical planning outputs are not reported in the registry record. Such planning illustrations are not observed power and cannot substitute for the uncertainty intervals from the actual trial.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s comparison estimates as geometric mean ratios with 90% confidence intervals: 121.12 (86.14 to 170.30) for the fasted formulation comparison, and 108.13 (76.68 to 152.48) for the fed-versus-fasted comparison. The analysis method is not specified in the registry record, and neither comparison has a posted p-value.

The registry record does not indicate a discrepancy between sponsor-reported results and a simple recalculation from the posted summary statistics. However, it does not report a separate set of recalculated estimates or the underlying participant-level data, so no independent reconstruction of the sponsor’s analysis can be assessed from these results alone. The ratios should therefore be read as the registry’s reported analyses. Their confidence intervals convey substantial uncertainty, and the record does not state a prespecified decision threshold or formal equivalence conclusion.

In general, a recalculation from summary statistics would be an unadjusted calculation and might not reproduce a crossover model that uses within-person data. The sponsor’s analysis is the relevant reported analysis for the trial; any lack of modeling details limits independent evaluation rather than changing the estimates reported in the registry.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and 12 analyses, including two posted secondary outcomes. The two named secondary outcomes concern AUC extrapolated to infinity for BI 1015550 and R-BI 1015550. The primary endpoint is AUC from time zero to the last quantifiable data point. The record does not explain how the analyses map across all outcome measures or identify a single family of hypotheses for multiplicity control.

Multiplicity arises in general when multiple outcomes or comparisons are tested: the chance of at least one misleadingly extreme result can increase if each is interpreted independently. Methods such as adjustment procedures, hierarchical testing, or prespecified prioritization can address that issue, but the statistical plan and any multiplicity approach are not reported in the registry record. No adjustment should be assumed from the posted results.

Missing-data handling is also not described. The record defines the pharmacokinetic parameter analysis set as treated participants with at least one primary or secondary pharmacokinetic endpoint not excluded because of a relevant protocol deviation. The full exclusion criteria and the counts excluded for each reason are not reported in the registry record. The primary endpoint counts are 24 for the fasted reference and 23 for each test condition; the record does not explain the difference.

The analysis population is therefore identifiable in broad terms, but the record does not provide a participant flow or a complete account of observations contributing to each comparison. The handling of missing samples, incomplete concentration profiles, and values that could not be quantified is not reported in the registry record. These details are important for judging whether the analyzed data could differ systematically from the planned data collection.

Without a stated statistical analysis plan, it is not possible to determine from the registry record whether the primary endpoint and comparisons were designated before data review, how secondary analyses were prioritized, or how missing data and multiple comparisons were intended to be handled. Those uncertainties limit assessment of the inferential framework, even though the endpoint and two comparison estimates are posted.

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 non-serious adverse event terms

For serious adverse events, the registry reports zero affected participants in each treatment condition: the fasted reference had 0 of 24, the fasted intended commercial formulation had 0 of 23, and the fed intended commercial formulation had 0 of 23. Deaths are reported as zero in each condition. These figures describe the posted counts during the trial; they do not establish that serious events cannot occur.

Other, non-serious adverse events were reported for 5 of 24 participants in the fasted reference condition, 3 of 23 in the fasted intended commercial formulation condition, and 2 of 23 in the fed intended commercial formulation condition. The registry record does not provide the event terms in the listed safety summary or explain the method used to attribute events to treatment conditions. The number of participants affected is not necessarily the same as the total number of events.

In general, safety counts from a small, short, single-dose study are descriptive and have limited ability to characterize uncommon events or longer-term experience. This trial involved healthy people and three treatment periods. The posted counts should be understood in that study context, not as a comprehensive safety profile or as evidence about risks outside the trial.

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

The results quantify two pharmacokinetic contrasts: the intended commercial formulation versus Trial Formulation 2 under fasting conditions, and the intended commercial formulation under fed versus fasted conditions. The reported geometric mean ratios are 121.12 and 108.13, respectively, with 90% confidence intervals of 86.14 to 170.30 and 76.68 to 152.48. The breadth of those intervals indicates that the relative estimates are not precise.

The study measures plasma exposure, not how participants feel, function, or experience a clinical outcome. It does not compare treatment effects in people with a disease, and it does not establish outcomes after repeated dosing. The trial’s healthy population, open-label conduct, single-dose design, and 24 enrolled participants define the setting to which the findings directly pertain.

In general, a pharmacokinetic comparison can help describe how formulations or food conditions relate to drug exposure. It cannot by itself establish clinical benefit, effectiveness, or a complete safety profile. Nor does a point estimate alone settle whether two conditions should be treated as equivalent: that interpretation ordinarily requires prespecified bounds and an analysis plan, neither of which is reported in the registry record.

The registry lists no serious adverse events and no deaths in the treatment conditions, and it reports other adverse events in each condition. These are observed counts in this study, not proof that an event is impossible or that risks are absent. No medical advice or treatment decision follows from the posted data alone.

Important interpretive details—including the full model, missing-data procedures, multiplicity handling, and reasons for the differing endpoint counts—are not reported in the registry record. The results can be read as the sponsor’s posted estimates with their stated uncertainty, but should not be extended beyond the measurements and population described.

14. Why This Trial Matters Statistically

This trial illustrates why pharmacokinetic interpretation involves more than comparing arm-level averages. The primary endpoint compresses a concentration-time profile into an exposure measure, while the geometric mean ratio expresses a relative comparison between conditions. Both pieces are useful, but neither should be read without the uncertainty interval and design context.

The crossover design is also statistically important. Because participants received treatment conditions across periods, within-person comparisons are central to the design’s rationale. The posted arm summaries do not reveal the within-person correlation or the specific model used to account for the crossover. This shows why aggregate means and variability may not be enough to independently reproduce a treatment contrast.

The two comparisons answer separate questions: formulation under fasting conditions, and food condition for the intended commercial formulation. Their ratios and confidence intervals should not be merged into one general statement. The wide intervals underscore the importance of reporting precision alongside point estimates, especially when the registry does not include a p-value or a formal decision criterion.

The record also illustrates transparency issues relevant to trial interpretation. It identifies the primary endpoint and analysis population broadly, but does not report several details needed to assess the analysis fully, including the model, missing-data methods, and multiplicity plan. The posted results are informative about the recorded pharmacokinetic comparisons, while important analytic uncertainties remain.

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.

Interpretation is limited by the size and setting of the study: the registry reports 24 healthy participants enrolled in a Phase 1, single-dose trial. The results concern plasma pharmacokinetics under the specified formulation and food conditions. Clinical outcomes, repeated-dose experience, and results in other populations are not reported in the registry record.

The comparison estimates have broad 90% confidence intervals, indicating uncertainty around both relative effects. The statistical model and confidence-interval calculation are not reported, limiting independent assessment of how period, sequence, and within-participant structure were addressed. The registry also does not report the within-person correlation or paired participant-level results.

There were 24 primary endpoint observations for the fasted reference and 23 for each intended commercial formulation condition. Why these analysis counts differ is not reported in the registry record. Detailed participant flow, missing sample handling, and reasons for any exclusions are also not reported.

The record lists six outcome measures and 12 analyses, but does not describe a multiplicity strategy or a complete statistical analysis plan. Numerical results for sample-size planning illustrations are not reported in the registry record. It is therefore not possible to evaluate the trial’s planned precision from those illustrations.

Finally, safety results are counts from a small, single-dose study. The record reports zero serious adverse events and zero deaths in each condition, but these observations cannot characterize uncommon or longer-term events. The listed summary also does not detail event terms or attribution procedures.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study in Healthy People to Compare 2 Different Formulations of BI 1015550 Taken With or Without Food — NCT05428436.
• Registry record last updated 2025-12-19; status last verified 2025-12.
• Results first posted 2025-12-19.
• 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

NCT05428436 is a completed, randomized, open-label, three-period crossover Phase 1 trial sponsored by Boehringer Ingelheim. It enrolled 24 healthy people and compared BI 1015550 as Trial Formulation 2 with the intended commercial formulation under fasting conditions, and compared fed with fasted administration of the intended commercial formulation.

The primary endpoint was plasma AUC from time zero to the last quantifiable data point. The reported geometric mean ratios were 121.12 for the fasted formulation comparison and 108.13 for the food-condition comparison, with 90% confidence intervals of 86.14 to 170.30 and 76.68 to 152.48. The intervals indicate uncertainty around both estimates.

The record reports zero serious adverse events and deaths in each condition. The analysis model, missing-data approach, multiplicity plan, and reasons for differing endpoint counts are not reported in the registry record. The findings describe pharmacokinetic results in the study population and conditions; they are not clinical outcome comparisons.