A Study in Healthy People to Test Whether BI 730357 Influences the Amount of Caffeine, Warfarin, Omeprazole, and Midazolam in the Blood: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study in Healthy People to Test Whether BI 730357 Influences the Amount of Caffeine, Warfarin, Omeprazole, and Midazolam in the Blood (NCT04679948), a na phase 1 study. The primary outcome is Area Under the Concentration-time Curve of Caffeine in Plasma Over the Time Interval From 0 Extrapolated to Infinity (AUC0-infinity, Caffeine).

2. Clinical Question

3. Trial Design

This was a phase 1, open-label, single-group study in healthy participants. The registry lists an enrollment of 16 and the status as completed. Allocation is recorded as not applicable, and masking is listed as none. The study title describes a two-period fixed-sequence design in which a cocktail of caffeine, warfarin, omeprazole and midazolam was administered before the treatment period combining the cocktail with BI 730357.

The treatment periods were separated by a washout phase of at least 20 days between the two cocktail administrations. The adverse-event record separately lists Cocktail (Period 1), BI 730357 (Period 2), and BI 730357 + Cocktail (Period 2), as well as a BI 730357 total. These are period or treatment group labels within the study, not randomized parallel arms. The sequence therefore does not support a randomized comparison in which treatment assignment balances participant characteristics or time-related influences.

The primary endpoint was caffeine area under the concentration-time curve from time zero extrapolated to infinity. Blood sampling was scheduled before dosing and at multiple time points after caffeine administration in both periods, including measurements through 48 hours. The registry gives specific sampling times for the primary endpoint, but does not report the concentration values or the endpoint estimates. The record also lists eight outcome measures and eight analyses, without enough information to describe the analyses for each outcome.

The trial started on 2020-12-21 and had primary completion on 2021-03-17. Its open-label, fixed-sequence structure is relevant to interpretation: measurements from the later period follow those from the earlier period, so treatment and period are not independently randomized. The registry record does not report a statistical analysis plan, a sample-size justification, or whether interim analyses were planned.

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 2

Analysis population for the primary outcome, as defined in the registry: Pharmacokinetic (PK) parameter analysis set (PKS): This set included all subjects in the TS who provided at least 1 PK endpoint that was defined as primary and was not excluded due to a clinical trial protocol (CTP) deviation relevant to the evaluation of PK or due to PK non-evaluability. Thus, a subject was included in the PKS, even if the subject contributed only 1 PK parameter value for 1 period to the statistical assessment. Only those with non-missing results are included in the analysis.

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 TS included all subjects who were entered and treated with at least 1 dose of trial drug.

6. Primary Endpoint

The registry lists the primary outcome as Area Under the Concentration-time Curve of Caffeine in Plasma Over the Time Interval From 0 Extrapolated to Infinity (AUC0-infinity, Caffeine), measured over Within 2 hours (h) predose for period 1, within 15 minutes predose for period 2 and 0.5 h, 1 h, 1.5 h, 2 h, 3 h, 4 h, 6 h, 8 h, 12 h, 24 h, 48 h after caffeine administration in both periods..

The primary endpoint is the area under the plasma concentration-time curve for caffeine, from time zero extrapolated to infinity. In pharmacokinetics, area under the concentration-time curve (AUC) represents overall exposure across the observed and extrapolated concentration profile. It incorporates both the level of drug in plasma and how that level changes over time. An AUC comparison can therefore address whether exposure differs between treatment conditions, rather than focusing only on a concentration at one sampling time.

The registry identifies the endpoint statistic as a geometric least-squares mean, with standard error as its dispersion measure. A geometric least-squares mean is an estimate on a multiplicative scale, often used when pharmacokinetic measures are analyzed after logarithmic transformation. Standard error describes the statistical precision of an estimated mean; it is not the same as the spread of individual participants’ values. These are general interpretations of the stated statistics, not details of an unreported analysis.

Plasma caffeine was measured before dosing and at scheduled times after administration in both periods, with sampling extending through 48 hours. The AUC is extrapolated to infinity, so its estimate depends partly on the measured concentration profile and partly on the terminal portion projected beyond the last observed time. The registry record lists the endpoint definition and sampling schedule but does not report the estimated AUC values, standard errors, treatment contrast, confidence interval or significance test. The size and precision of the observed caffeine effect therefore cannot be described from the record details available.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical concepts below are general explanations of the measures and design identified in the registry record. The record names the primary statistic and dispersion measure but does not report the full statistical analysis plan or numerical results. The methods actually used for transformations, model specification, confidence intervals and hypothesis testing are therefore not reported in the registry record.

What does AUC measure?

Area under the concentration-time curve combines information across a series of plasma concentrations taken over time. It is used in pharmacokinetics as a measure of overall exposure. AUC from time zero extrapolated to infinity includes the observed profile and an estimated contribution after the last sample. The extrapolated portion depends on the estimated decline in concentration near the end of sampling, so the quality of the profile and the terminal estimate can affect precision. The registry gives the sampling schedule for caffeine through 48 hours but does not give the concentration data or the proportion of AUC that was extrapolated.

AUC is not itself a direct measure of clinical benefit or harm. It answers a pharmacokinetic question about exposure. Interpretation of an exposure contrast requires the estimated values and their uncertainty, neither of which is reported for the primary endpoint in the record details available. Results for the other cocktail components are also not reported in the registry record.

What does a geometric least-squares mean represent?

A geometric mean is appropriate for data that are naturally interpreted multiplicatively or have a right-skewed distribution. A common general approach is to analyze logarithmically transformed values, estimate means on that scale, and transform the result back. The resulting geometric mean is a central estimate on the original scale, but the registry record does not specify whether this exact procedure was used or describe a model. “Least-squares” indicates a model-based estimate adjusted according to the analysis specification; which terms or adjustments were included is not reported in the registry record.

With pharmacokinetic outcomes, treatment effects are often expressed as ratios because a multiplicative comparison can be more meaningful than an additive difference. That is a general statistical convention, not a reported result or confirmed analysis method for this trial. No ratio, contrast, confidence interval, or p-value for caffeine AUC is reported in the registry entries available, so no effect estimate can be interpreted.

How should standard error and confidence intervals be understood?

Standard error describes the precision of an estimated quantity across repeated samples under a statistical model. It is not a description of the range of individual participant values. A small standard error relative to an estimate generally indicates greater precision, while a larger one indicates more uncertainty. Here, the record names standard error as the dispersion measure for the primary statistic but does not report its value.

A confidence interval is a range constructed by a statistical procedure to express uncertainty around an estimate. Its interpretation depends on the model and assumptions used. A narrow interval indicates greater precision than a wide interval, while an interval spanning values on both sides of a null contrast may indicate that the data are compatible with a range of effects. These are general explanations only: the registry record does not report a confidence interval for the primary endpoint.

Why does the design matter?

The study is described as open-label, single-group and fixed sequence. Participants received the cocktail in Period 1 before the cocktail-plus-BI 730357 treatment in Period 2. A within-participant comparison can account for stable differences between people, because each person contributes measurements under the sequence. However, fixed order means treatment condition and period are linked. Changes over time, learning or procedural effects, and any residual influence from earlier dosing may complicate interpretation. The washout was at least 20 days between the two cocktail administrations, but a washout interval does not by itself demonstrate the absence of carryover.

Open-label conduct means masking was not used. This may be more consequential for subjective outcomes than for laboratory concentration measurements, but it can still affect conduct or decisions around data collection. That is a general concern, not evidence that measurements in this trial were biased. The registry does not report procedures for blinded laboratory analysis or other safeguards.

Over-dispersion is a general issue when count data vary more than a chosen model expects; a model that ignores it can produce standard errors that are too small. That concept is not directly the primary endpoint method here, which is a continuous pharmacokinetic measure, and the registry does not report a count model for the primary analysis. Likewise, no method for handling repeated measures, missing values, or outliers is reported. These should not be inferred from the endpoint label alone.

What can be said about planning calculations?

The record describes planning illustrations based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the target effect is a stated fraction of the observed effect. These assumptions describe an illustrative calculation, not the actual fixed-sequence design. Such planning values are not observed power and do not show that a study had a particular chance of detecting an effect after the trial was completed. The record details available do not include participant-per-arm values for 80% power or other planning outputs, nor do they report the trial’s own sample-size justification. Any calculation depending on an observed effect cannot be evaluated when the endpoint estimate is not reported.

Overall, the record supports a description of the endpoint, design and named summary statistic, but not a reconstruction of the primary analysis. The precise model, handling of the paired or sequential data, uncertainty interval, hypothesis test, multiplicity control and missing-data procedures are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures and eight analyses. The record details available identify caffeine AUC as the primary endpoint, but do not report the complete outcome definitions or the analysis results for the remaining measures. It is therefore not possible to tell how many statistical comparisons were treated as confirmatory, which outcome tests were prioritized, or whether multiplicity was controlled.

Multiplicity arises when several outcomes, time points or contrasts are tested. If each is evaluated without adjustment, the chance of at least one apparently notable result can increase. General approaches include prespecifying a primary comparison, adjusting significance thresholds, or presenting estimates with emphasis on uncertainty rather than isolated p-values. The registry record does not report which, if any, approach was used. No p-values or confidence intervals for the primary endpoint are reported in the details available, so the impact of multiplicity cannot be assessed.

The primary population is described as a pharmacokinetic analysis set comprising subjects who provided at least one primary pharmacokinetic endpoint and were not excluded for relevant protocol deviations or pharmacokinetic non-evaluability. The description is incomplete in the registry record, and the number included in that analysis is not reported. The overall enrollment was 16, but enrollment does not establish how many participants contributed evaluable data to each endpoint.

Missing-data handling, exclusions, imputation, and sensitivity analyses are not reported in the registry record. The record also does not describe interim analyses or the prespecified statistical analysis plan. These omissions limit assessment of whether the reported analysis population and methods were chosen in advance or how incomplete pharmacokinetic profiles were handled.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent non-serious adverse event terms

The registry reports zero serious adverse events among 16 participants in each listed group: Cocktail (Period 1), BI 730357 (Period 2), BI 730357 + Cocktail (Period 2), and BI 730357 Total. It also reports zero deaths in each group. These are observed counts in a small phase 1 study and should be read as counts recorded during the study, not as proof that serious events or deaths cannot occur.

For other adverse events, the registry lists 3 affected participants in Cocktail (Period 1), 2 in BI 730357 (Period 2), and none in BI 730357 + Cocktail (Period 2). BI 730357 Total lists 2 affected participants. Participants could appear in more than one period or group summary, so the group counts should not be added to infer a unique participant total. The record does not list the non-serious event terms, severity details or attribution assessments.

The group summaries are not randomized, like-for-like safety comparisons. They reflect a fixed sequence and distinct period labels, so differences in these counts cannot by themselves establish a treatment-caused difference in risk. Serious-event counts and non-serious-event counts also describe different categories and should not be conflated.

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

The registry record describes a phase 1 pharmacokinetic study in healthy participants, with caffeine AUC as the primary endpoint and a fixed treatment sequence involving BI 730357. It also records adverse-event counts. These details explain what the study set out to assess and how its treatment sequence was organized; they do not establish the size or direction of the pharmacokinetic effect.

The primary caffeine result is not reported in the registry record. Consequently, the posted information cannot show whether caffeine AUC was higher, lower or similar after BI 730357, nor how uncertain any contrast might be. The same limitation applies to numerical pharmacokinetic results for warfarin, omeprazole and midazolam. Without estimates and uncertainty measures, there is no basis for judging the magnitude or precision of an observed pharmacokinetic difference.

The design also limits causal interpretation. Participants followed a fixed sequence rather than being randomized to concurrent comparison groups. The washout between cocktail administrations was at least 20 days, but sequence-related influences cannot be excluded solely from that interval. Open-label conduct and the absence of a randomized comparator should be kept in mind when reading any contrast that might be reported elsewhere.

The zero serious adverse events and zero deaths are the registry counts for the listed groups. They do not establish that no risk exists, and the other adverse-event counts do not by themselves attribute events to a particular treatment. The record does not provide event terms for the non-serious events.

This study addresses pharmacokinetics under the specified study conditions in healthy people. It does not, on the information reported, answer questions about clinical outcomes or establish how results would apply to other populations or settings. The record lists eight outcome measures and eight analyses, but does not provide the numerical results or enough detail to evaluate their multiplicity, missing data or full analysis methods.

13. Why This Trial Matters Statistically

This trial illustrates how a pharmacokinetic study can be organized around exposure changes rather than clinical outcomes. Its primary endpoint, caffeine AUC extrapolated to infinity, uses concentration measurements over time to represent overall exposure. The listed sampling schedule makes clear that the endpoint depends on a series of measurements, not a single concentration reading.

The study also highlights the importance of distinguishing a fixed-sequence design from a randomized comparison. Participants received the cocktail before the later cocktail-plus-BI 730357 condition, with at least 20 days between cocktail administrations. This can support within-participant pharmacokinetic assessment, but period and treatment are linked, so the design does not offer the protection against order effects that random assignment can provide. The record’s allocation and masking entries reinforce that this was not a randomized, masked parallel-group trial.

Another statistical lesson is that endpoint definitions and labels are not a substitute for results. The registry identifies a geometric least-squares mean and standard error, but without the actual estimate and its uncertainty, readers cannot assess effect size or precision. Naming a model-based summary does not reveal the model specification, handling of incomplete profiles, or the analysis population size.

Finally, the safety counts show why small-study event summaries require restraint. Zero serious adverse events and deaths are useful descriptions of what was recorded, but they do not establish the absence of risk. With no event terms or attribution details for other adverse events, the counts cannot support detailed comparisons. The record is informative about the study structure and measurement intent, while the absence of reported endpoint estimates sharply limits statistical interpretation.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

A central limitation is that the registry record does not report numerical results for the primary caffeine AUC endpoint. The geometric least-squares mean and standard error are identified, but their values, any treatment contrast, confidence interval and p-value are not reported. The corresponding pharmacokinetic results for warfarin, omeprazole and midazolam are also not reported in the registry record details available.

The study enrolled 16 participants and used an open-label fixed sequence. It did not use randomized allocation or masking, and the registry explicitly cautions that the arms are not a like-for-like control comparison. Treatment and period are linked, and the extent to which period effects or carryover could matter cannot be assessed from the reported details. The stated washout phase was at least 20 days between the two cocktail administrations, but no evidence about residual carryover is reported.

The record’s description of the pharmacokinetic analysis population is incomplete, and the number of participants included in the primary analysis is not reported. Exclusions, missing samples, non-evaluable profiles, imputation and sensitivity analyses are not described. The analysis model and full statistical analysis plan are not reported, nor are interim analyses or the sample-size justification.

Although the record lists eight outcome measures and eight analyses, it does not provide enough information to assess multiplicity or compare the primary outcome with secondary findings. The other adverse-event counts lack event terms and attribution information. These limitations mean that the record supports a clear description of study intent and structure, but not a detailed assessment of efficacy, precision, safety patterns or robustness.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study in Healthy People to Test Whether BI 730357 Influences the Amount of Caffeine, Warfarin, Omeprazole, and Midazolam in the Blood — NCT04679948.
• Registry record last updated 2023-08-14; status last verified 2023-07.
• Results first posted 2023-07-17.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT04679948 is a completed phase 1, open-label, single-group study sponsored by Boehringer Ingelheim. It enrolled 16 healthy participants and evaluated caffeine, warfarin, omeprazole and midazolam pharmacokinetics in a fixed sequence involving BI 730357. The cocktail treatment preceded the cocktail-plus-BI 730357 treatment, with a washout phase of at least 20 days between cocktail administrations.

The primary endpoint was caffeine plasma AUC from time zero extrapolated to infinity, summarized as a geometric least-squares mean with standard error. Numerical endpoint results are not reported in the registry record details available. The safety record lists zero serious adverse events and zero deaths in the groups, with other adverse events affecting 3 participants in Cocktail (Period 1), 2 in BI 730357 (Period 2), and none in BI 730357 + Cocktail (Period 2). The full statistical analysis plan, missing-data methods and multiplicity approach are not reported in the registry record.