A Study in Healthy People to Compare How 2 Different Low Dose Formulations of BI 1015550 Are Taken up in the Body: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study in Healthy People to Compare How 2 Different Low Dose Formulations of BI 1015550 Are Taken up in the Body (NCT06408870), a randomized phase 1 study. The primary outcome is Area Under the Concentration-time Curve of Nerandomilast in Plasma Over the Time Interval From 0 to the Last Quantifiable Data Point (AUC0-tz).

2. Clinical Question

3. Trial Design

This was a Phase 1, randomized, open-label, single-dose, two-way crossover trial sponsored by Boehringer Ingelheim. The registry lists 64 participants and the condition as healthy. The study began on 2024-06-06, reached primary completion on 2024-08-28, and is marked completed.

In a crossover design, participants receive more than one treatment in a planned order. The two sequences in this trial were Formulation C1 followed by Formulation C2, and Formulation C2 followed by Formulation C1. This allows the formulations to be compared within participants, which can reduce the influence of stable differences between people on the treatment comparison. The registry identifies random allocation and no masking. It does not report the sequence allocation method beyond those design descriptions.

Each treatment was described as a single oral dose. The trial's focus was pharmacokinetic exposure rather than a clinical outcome such as symptom change. The primary endpoint was the plasma area under the concentration-time curve from time zero to the last quantifiable data point. Blood sampling was scheduled before dosing and at multiple stated times after intake, extending to 144 hours. The registry also lists an area-under-the-curve outcome extrapolated to infinity as a secondary outcome.

The primary analysis population was the pharmacokinetic parameter analysis set. This included treated participants who contributed at least one specified pharmacokinetic endpoint and were not excluded for a relevant protocol deviation or pharmacokinetic non-evaluability. The reported primary-endpoint group counts were 64 for C1 and 61 for C2. The registry record does not report why these counts differ or provide a participant flow explanation in the registry record for this 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: 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.

Treatment period 1

Washout period

Reasons for not completing this period:

Treatment period 2

Analysis population for the primary outcome, as defined in the registry: Pharmacokinetic parameter analysis set (PKS): included all subjects in the treated set (TS) 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. Thus, a subject was included in the PKS, even if he/she contributed only one PK parameter value for one period to the statistical assessment. Participants in both arms are not mutually exclusive.

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. The treated set was used for safety analyses.

6. Primary Endpoint

The registry lists the primary outcome as Area Under the Concentration-time Curve of Nerandomilast in Plasma Over the Time Interval From 0 to the Last Quantifiable Data Point (AUC0-tz), measured over Within 3 hours (hrs) prior to and at 15minutes (min), 30min, 45min, 1hrs, 1:15hrs, 1:30hrs, 1:45hrs, 2hrs, 2:30hrs, 3hrs, 3:30hrs, 4hrs, 6hrs, 8hrs, 12hrs, 24hrs, 34hrs, 48hrs, 58hrs, 72hrs, 96hrs, 120hrs, and 144hrs after nerandomilast intake..

The primary endpoint is the area under the plasma concentration-time curve for nerandomilast, from time zero until the last quantifiable data point. It is expressed in hours*nanomole/Liter (hrs*nmol/L). The area under a concentration-time curve represents cumulative drug exposure over the measured period: it combines concentration and time rather than describing a concentration at one isolated sampling point.

This measure is useful in a formulation comparison because differences in how a drug is absorbed or otherwise enters the circulation can affect the concentration profile over time. Comparing exposure across the two formulations therefore addresses whether they produce similar overall measured plasma exposure after a single dose. It is a pharmacokinetic measure, not a direct measure of a clinical benefit or harm.

The registry reports geometric least-squares means for each formulation and identifies standard error as the dispersion measure. The primary endpoint was assessed within the stated sampling schedule, which included measurements from before intake through 144 hours after intake. The endpoint ends at the last quantifiable observation; it is distinct from the listed secondary area-under-the-curve measure that extends from zero to infinity.

The posted adjusted geometric-mean ratio is 103.88%, with a 90% confidence interval from 101.08% to 106.75%. A confidence interval describes uncertainty around an estimate under the analysis model. Whether this interval satisfies a formal bioequivalence rule depends on the prespecified acceptance limits, which are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes an analysis of variance (ANOVA) for the primary pharmacokinetic endpoint. In this setting, ANOVA is a model for comparing treatment-related measurements while accounting for features of the crossover design. The stated model included subjects within sequence as a random effect, and sequence, period, and treatment as fixed effects.

Why account for subjects and sequence?

In a crossover study, each participant receives both formulations, so observations from the same person are related rather than independent. A subject-within-sequence random effect represents participant-level variation within the allocated order. This helps distinguish differences between participants from differences associated with the formulations. Sequence is included as a fixed effect to account for the assigned order in which the treatments were received.

Period represents the position of a treatment in the crossover sequence. A period effect can capture systematic differences between the first and later treatment periods that are not specifically due to which formulation was used. Treatment represents the contrast of interest: the comparison of Formulation C1 and Formulation C2. The registry identifies these model components but does not report further details such as the estimation procedure, transformations, handling of carryover, or diagnostics.

Why use geometric means?

Pharmacokinetic exposure measures are often summarized on a multiplicative scale. A geometric mean is formed from logarithmically transformed values and then expressed on the original scale. This approach is useful when proportional rather than absolute changes are the focus. In a formulation comparison, the ratio of adjusted geometric means expresses the relative exposure associated with the two formulations. A ratio of 100% corresponds to equal adjusted geometric means; values on either side indicate a relative difference in the estimated exposure.

The registry reports a ratio of adjusted geometric means of 103.88%, with a 90% confidence interval from 101.08% to 106.75%. The confidence interval represents uncertainty in the ratio estimate under the fitted model. Its endpoints are not individual participant values and do not describe the range of responses in the study population. A confidence interval also does not specify the formal decision rule. For a bioequivalence conclusion, the interval must be interpreted against prespecified limits, which are not reported in the registry record.

What does the endpoint capture?

The primary endpoint is area under the concentration-time curve from zero to the last quantifiable data point. It summarizes measured plasma exposure over time rather than using a single concentration measurement. The registry reports geometric least-squares means and standard error as the dispersion measure. The listed arm values are 1271.0 for C1 and 1320.0 for C2; the corresponding analysis counts are 64 and 61. These values should be understood in the context of the model-based comparison, not as a substitute for its adjusted ratio and confidence interval.

The outcome's sampling schedule spans multiple post-dose times through 144 hours, as well as a pre-dose measurement. The endpoint is truncated at the last quantifiable point for each participant, whereas the secondary endpoint listed in the registry extends the area under the curve from zero to infinity. The record does not provide the individual concentration profiles or participant-level endpoint values in the registry record for this analysis.

What can and cannot be inferred?

General statistical interpretation: an estimate is a summary of the observed data under a model, while its confidence interval conveys uncertainty in that estimate. The interval does not account for every possible limitation, such as selection into the analysis set or departures from model assumptions. The registry describes the analysis set and model components, but does not report diagnostic results or a detailed account of missing pharmacokinetic measurements.

General statistical interpretation: over-dispersion is especially relevant when a model assumes less variability than the data contain, often in count outcomes. The primary endpoint is a continuous exposure measure, and the registry describes an ANOVA model rather than a count model. The registry entries do not report a separate over-dispersion assessment, and it should not be inferred that one was conducted.

General statistical interpretation: in a crossover trial, valid treatment comparisons rely on the planned sequence and period structure and on usable measurements for the relevant endpoint. The registry's definition of the pharmacokinetic analysis set excludes certain protocol deviations and non-evaluable data, but does not explain the reasons for individual exclusions or missing values. No further assumptions about how those cases were handled should be made.

The registry reports no p-value for the primary ratio. A test of whether a ratio differs from 100% would answer a different question from whether an interval lies within prespecified bioequivalence limits. The limits and detailed decision procedure are not reported in the registry record. The posted result therefore supports interpretation of the estimated exposure ratio and its precision, but does not independently establish the trial's formal bioequivalence determination.

10. Analyses Reported by the Sponsor

The registry reports an adjusted geometric-mean ratio of 103.88%, with a 90% confidence interval from 101.08% to 106.75%. It also describes an ANOVA model with subjects within sequence as a random effect and sequence, period, and treatment as fixed effects. These are the posted sponsor analysis results and model details for the primary endpoint.

No separate recalculated estimate for this primary comparison is reported in the registry record. Consequently, there is no recalculated value to compare with the sponsor's estimate. The registry entries do not indicate a discrepancy between the posted analysis and a recalculation.

The result should therefore be read as the sponsor-reported, model-adjusted comparison, rather than as an independent reconstruction from summary values. The listed arm summaries—1271.0 for C1 and 1320.0 for C2—do not provide enough detail to reproduce the adjusted comparison. Individual participant data, the full model output, and the prespecified bioequivalence limits are not reported in the registry record. The posted ratio and interval describe the statistical estimate and its uncertainty; they do not on their own document whether a formal acceptance criterion was satisfied.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary pharmacokinetic endpoint: area under the concentration-time curve from zero to the last quantifiable data point. It also lists a secondary area-under-the-curve endpoint extending from zero to infinity. The record identifies three outcome measures and three analyses overall, but the registry entries do not provide a complete account of the hypotheses, testing order, or any adjustment for multiple comparisons. Multiplicity procedures are therefore not reported in the registry record.

Multiplicity matters when several outcomes or comparisons are tested because the chance of at least one misleading result can increase across a collection of tests. The number of listed outcomes alone does not establish how many formal hypotheses were tested, nor whether a multiplicity adjustment was planned. No claim about such an adjustment can be made from the listed information.

The primary analysis set is defined as treated participants who contributed at least one primary or secondary pharmacokinetic endpoint and were not excluded because of a relevant protocol deviation or pharmacokinetic non-evaluability. This identifies broad eligibility criteria for the analysis but does not give a participant-level account of exclusions or missing observations. The reasons for the difference between the reported endpoint counts of 64 for C1 and 61 for C2 are not reported in the registry record.

The registry record does not report a detailed missing-data method, imputation procedure, sensitivity analysis, or an analysis of the effect of missing pharmacokinetic measurements. It also does not report whether any missingness was related to treatment, sequence, or endpoint values. The stated analysis-set definition should not be treated as evidence that missingness had no effect.

These omissions limit assessment of how robust the result is to alternative analysis choices. They do not change the numerical estimate that is posted, but they constrain what can be concluded about the broader analysis plan and the impact of unavailable or excluded data.

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

The registry reports zero serious adverse events in both treatment groups: 0 among 64 participants in the C1 group and 0 among 61 participants in the C2 group. It also reports zero deaths in each group. These are counts observed in the groups at risk, not estimates that serious events are impossible or that the treatments have identical safety profiles.

Other adverse events were reported for 15 participants in the C1 group and 10 participants in the C2 group. These counts are separate from the serious adverse event figures. The registry entries do not provide the event terms, severity breakdown, relatedness assessments, or timing for these other events. The registry lists no top serious adverse event terms.

Safety counts in a trial of this size and design describe events recorded during the study; they do not, by themselves, establish comparative safety or capture every possible adverse effect. The trial involved healthy participants receiving single doses, and the registry record does not report longer-term safety follow-up in the registry record for this analysis. No inference about unobserved events should be drawn from the absence of reported serious events or deaths.

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

The main statistical finding is a model-adjusted geometric-mean ratio of 103.88% for the primary exposure endpoint, with a 90% confidence interval from 101.08% to 106.75%. This indicates that the estimated adjusted geometric means were close to the equal-exposure reference point of 100%. The interval conveys uncertainty around that estimate; it is not a range of individual participant outcomes.

The trial was designed to compare the pharmacokinetic exposure after single doses of two BI 1015550 formulations in healthy participants. Its randomized two-way crossover structure allows the formulations to be compared across treatment periods within participants. The reported analysis accounts for participant sequence, period, and treatment according to the ANOVA description.

The result does not establish that the formulations meet a formal bioequivalence criterion, because the acceptance limits and decision rule are not reported in the registry record. Nor does the result show whether either formulation produces a clinical benefit or harm. The primary endpoint measures drug exposure, not a patient-centered clinical outcome.

The findings are specific to the analyzed healthy-participant population, the single-dose setting, and the reported pharmacokinetic endpoint. They should not be treated as evidence about repeated dosing or other populations. The endpoint counts differ between groups, and the reasons for that difference are not reported in the registry record. Detailed missing-data handling, multiplicity procedures, and a complete participant flow explanation are also not reported.

In short, the posted estimate is informative about the relative exposure measured in this trial, and its confidence interval describes the precision of that estimate under the stated model. Broader conclusions require information not reported in the registry record, including the prespecified bioequivalence limits and details of the analysis plan.

14. Why This Trial Matters Statistically

This trial illustrates how a crossover design can be used to compare formulations through pharmacokinetic measurements. Because participants received both treatments in different orders, the analysis could account for participant-level variation as well as sequence and period. The registry's model description makes those design features explicit and identifies treatment as the contrast of interest.

It also illustrates why a point estimate must be interpreted with its uncertainty. The adjusted geometric-mean ratio is 103.88%, while the 90% confidence interval extends from 101.08% to 106.75%. The estimate gives a central summary; the interval shows that the comparison is not known without uncertainty. Neither the point estimate nor the interval alone specifies the formal criterion for bioequivalence.

A further statistical distinction is between testing for a difference and assessing equivalence within defined limits. A result that appears close to 100% is not, by itself, proof of equivalence. A formal assessment needs prespecified limits and a decision rule. Since these are not reported in the registry record, the reported pharmacokinetic result can be described, but the formal conclusion cannot be independently assessed from these registry entries.

The listed analysis population and unequal endpoint counts also show why the denominator matters. The record defines who could be included in the pharmacokinetic analysis but does not explain the count difference or describe missing-data handling in detail. For readers evaluating evidence, the estimate, precision, analysis population, and reporting gaps all contribute to a careful 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 trial's primary outcome is a pharmacokinetic exposure measure after a single dose in healthy participants. The registry record therefore does not support conclusions about clinical outcomes, repeated use, or populations other than the healthy participants enrolled. The endpoint reflects measured exposure over the stated interval, rather than a direct measure of efficacy or longer-term safety.

The registry reports a 90% confidence interval for the adjusted geometric-mean ratio, but does not report the prespecified bioequivalence limits or decision rule. This is a central limitation for interpreting the stated objective of establishing bioequivalence: the estimate and interval are available, but the formal benchmark for judging them is not.

The primary analysis counts were 64 for C1 and 61 for C2, while overall enrollment was 64. The reason for the differing endpoint counts is not reported in the registry record. The registry entries also do not provide a detailed participant flow, a description of missing-value handling, or sensitivity analyses addressing exclusions or non-evaluable pharmacokinetic data.

The ANOVA model components are described, but additional technical details are not reported in the registry record. These include model diagnostics, estimation details, and a full account of the prespecified analysis plan. Multiplicity procedures across the listed outcomes are also not reported. The absence of these details limits independent assessment of the full analytical strategy, although it does not alter the numerical values posted for the primary comparison.

Finally, the available safety information gives group counts for serious and other adverse events and deaths, but not event terms or detailed assessments for other adverse events. The scope of follow-up beyond the trial periods is not reported in the registry record. These limits should be kept distinct from the primary pharmacokinetic estimate, which addresses a narrower question about measured exposure.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study in Healthy People to Compare How 2 Different Low Dose Formulations of BI 1015550 Are Taken up in the Body — NCT06408870.
• Registry record last updated 2025-12-01; status last verified 2025-07.
• Results first posted 2025-12-01.
• 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

NCT06408870 is a completed Phase 1 randomized, open-label, single-dose, two-way crossover trial sponsored by Boehringer Ingelheim. It enrolled 64 healthy participants and compared BI 1015550 Formulation C2, the test formulation, with Formulation C1, the reference formulation. The study began on 2024-06-06 and reached primary completion on 2024-08-28.

The primary endpoint was plasma area under the concentration-time curve from time zero to the last quantifiable data point. The registry reports an adjusted geometric-mean ratio of 103.88%, with a 90% confidence interval from 101.08% to 106.75%. The described analysis used ANOVA with subjects within sequence as a random effect and sequence, period, and treatment as fixed effects.

The registry lists zero serious adverse events and zero deaths in both groups. It reports other adverse events for 15 participants in the C1 group and 10 in the C2 group. The formal bioequivalence limits, detailed missing-data handling, and reasons for the differing primary endpoint counts are not reported in the registry record.