A Study of Tirzepatide (LY3298176) in Healthy Participants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Tirzepatide (LY3298176) in Healthy Participants (NCT03375463), a randomized phase 1 study. The primary outcome is Pharmacokinetics (PK) Part A: Area Under the Concentration Versus Time Curve [AUC (0-∞)] of Tirzepatide.

2. Clinical Question

The study was conducted in healthy participants, and the primary result is pharmacokinetic rather than a comparison of clinical benefit. The registry record describes additional safety and tolerability aims in the other study parts. Those aims do not change the scope of the primary endpoint: it summarizes tirzepatide exposure for the Part A formulation groups. The posted endpoint population is restricted to participants with a dose and evaluable pharmacokinetic data, rather than being defined simply as everyone enrolled in the overall study.

3. Trial Design

The registry record classifies this as a completed phase 1 study with randomized allocation, double masking, and a parallel model. Enrollment was 52. Participants were healthy, and the study was organized into four parts; the record says each participant enrolled in one part. The study began on 2017-12-19, and primary completion was recorded as 2018-12-27.

The parts addressed different formulations, routes, and dosing schedules. Part A compared 5 mg subcutaneous tirzepatide solution with 5 mg subcutaneous lyophilized tirzepatide. Part B evaluated a single 0.5 mg intravenous infusion. Part C evaluated repeated weekly subcutaneous solution doses, with the listed sequence of 5 mg, 5 mg, 7.5 mg, and 10 mg, and included a placebo group. Part D evaluated a single 0.5 mg intravenous bolus of lyophilized formulation. These different objectives and regimens mean that the overall enrollment should not be treated as the sample size for the Part A primary endpoint.

The brief description estimates approximately 70 days for Part A, Part B, or Part D, and 92 days for Part C, excluding screening. It also describes screening within 28 days before study start. These are descriptions of the planned participant timelines, not outcome measurements.

The registry record does not report how participants were randomized within each part, the allocation ratio, the masking procedures, or whether masking was maintained identically across all parts. It also does not report the sample-size justification or enough detail to reconstruct the protocol's statistical plan. The primary posted pharmacokinetic result is specifically for Part A; it should not be read as a pooled result across the four parts.

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.

Period 1

Reasons for not completing this period:

Period 2

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants from Part A group, who received at least one dose of study drug and had evaluable PK data

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: All randomized participants

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Pharmacokinetics (PK) Part A: Area Under the Concentration Versus Time Curve [AUC (0-∞)] of Tirzepatide, measured over Part A: Predose, 8, 12, 24, 48, 72, 96, 120, 144, 168, 336, 480, 816-864 hours postdose.

The primary endpoint is the area under the tirzepatide concentration-time curve from time zero to infinity, recorded in nanograms*hours per milliliter. AUC combines concentration measurements over time into a measure of overall drug exposure. The time points listed for Part A extend from predose through 816-864 hours postdose, supporting the pharmacokinetic description used for this endpoint.

The registry reports the endpoint as a geometric mean, together with a geometric coefficient of variation. A geometric mean is a summary suitable for measurements that are positive and may be asymmetrically distributed; it expresses a typical multiplicative level rather than an arithmetic average. The coefficient of variation describes relative spread around that summary. It is not a confidence interval and does not by itself quantify uncertainty in the difference between formulations.

The posted geometric mean is 113000 for the 5 mg lyophilized formulation, based on 19 participants, with a geometric coefficient of variation of 19. For the 5 mg solution formulation, the posted geometric mean is 109000, based on 20 participants, with a geometric coefficient of variation of 17. No confidence limits are posted for either mean. AUC is used in pharmacokinetic studies because a single summary can characterize exposure across the sampling period, but it does not on its own describe every feature of the concentration-time profile or establish a clinical effect.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General: Pharmacokinetic endpoints describe how drug concentration changes over time. Concentration-time measurements are collected after dosing, and AUC summarizes exposure across a defined interval. For the primary endpoint, the interval is from zero to infinity. The endpoint is reported in nanograms*hours per milliliter, so the unit reflects concentration accumulated over time. This is a pharmacokinetic measure, not a patient-centered clinical outcome.

What does the geometric mean summarize?

The registry reports geometric means rather than arithmetic means. In general, a geometric mean is useful for positive measurements with a multiplicative pattern of variation. Conceptually, observations are considered on a logarithmic scale and then expressed back on the original scale. This summary can be more representative of a typical value when measurements have a long upper tail. The registry record does not describe the exact calculation procedure used for the posted result, so the general explanation should not be mistaken for a statement about undocumented analysis steps.

For Part A, the geometric mean is 113000 for the lyophilized formulation and 109000 for the solution formulation. The registry gives the geometric coefficient of variation as the dispersion measure, with values of 19 and 17 respectively. In general, a coefficient of variation expresses variability relative to the size of a measurement, which can make it informative for pharmacokinetic data whose spread increases with the measurement level. It describes variation among measurements; it is not an interval estimate for a group mean or for a treatment contrast.

What would an inferential comparison add?

A descriptive comparison places the group summaries side by side. An inferential comparison would additionally estimate the contrast between formulations and its uncertainty. A confidence interval gives a range of values compatible with the data and the specified analysis procedure; a narrow interval indicates greater precision, while a wide interval indicates less precision. A p-value addresses compatibility with a specified null hypothesis under an analysis model, but does not measure the size or clinical importance of an effect. Neither a confidence interval nor a p-value for the primary contrast is reported in the registry record.

For positive pharmacokinetic measurements, comparisons are often conducted on a logarithmic scale and expressed as a ratio on the original scale. That is a general approach, not a claim about the analysis used in this study. The registry does not report a ratio, a model, adjustment factors, a test, or confidence limits for the Part A contrast. The posted geometric means alone do not establish what inferential method, if any, was used by the investigators.

Why can variability affect precision?

General: Greater variability makes it harder to estimate a group summary or a contrast precisely when other design features are unchanged. The geometric coefficients of variation describe the observed spread in the Part A data, but they cannot determine the precision of the between-group contrast without the analysis details and an uncertainty estimate. Small group counts can also limit precision, but no claim about achieved power follows from the counts alone. Observed power is not a substitute for an interval estimate and is not reported in the registry record.

Over-dispersion is a general issue when observed variation exceeds what an analysis model assumes. If a model understates variation, its uncertainty estimates may be too small and its inferential conclusions too optimistic. This is a general statistical concern, not a finding about this study: the registry record does not report a model fit assessment or an over-dispersion evaluation. The posted geometric coefficients of variation by themselves do not show whether any model assumption was violated.

What do the planning illustrations mean?

General: A sample-size planning calculation asks how many participants a future study might need under specified assumptions about the effect, variability, allocation, and testing procedure. The study record's planning description says the illustrations use 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. These are illustrative assumptions, not reported features of the original design. They do not establish that the original trial was planned with those assumptions, and they are not observed power. The registry record does not report the original sample-size justification.

The distinction between planning and inference matters. Planning uses assumptions to explore a possible future design; an analysis uses observed data to estimate an effect and its uncertainty. A planning illustration based on the observed effect is especially dependent on the observed result and should not be taken as independent confirmation. No planning figure can fill the registry's missing confidence interval or formal comparison.

The registry classifies the study as randomized, double masked, and parallel. In general, randomization can reduce systematic allocation differences, masking can reduce certain expectation-related biases, and a parallel design assigns participants to a study group rather than having each participant receive every study condition. The record does not describe implementation details sufficient to assess how these features operated in each part. These general design concepts do not resolve the absence of a posted inferential estimate for the primary pharmacokinetic contrast.

10. Multiplicity, Missing Data and Analysis Populations

The study has several parts, multiple interventions, and six outcome measures recorded, including the primary endpoint and secondary pharmacokinetic outcomes. When several outcomes or comparisons are tested, multiplicity can increase the chance of a false-positive finding if no adjustment or prespecified testing strategy is used. The registry record does not report a multiplicity strategy, a hierarchy of hypotheses, or whether any adjustment was applied. No formal primary comparison is posted, so the effects of multiplicity on a reported primary test cannot be assessed from this record.

The primary endpoint population is defined as randomized Part A participants who received at least one dose and had evaluable pharmacokinetic data. The endpoint summaries include 19 participants for the lyophilized formulation and 20 for the solution formulation. The record does not report the number randomized to each Part A group before applying the dose and evaluability criteria, nor does it describe missing-data handling, reasons for missing pharmacokinetic data, or sensitivity analyses.

In general, missing outcome data can affect the interpretation of a comparison when the availability of measurements is related to participant characteristics or outcomes. An analysis restricted to evaluable measurements may not represent every randomized participant if data are missing for systematic reasons. That is a general concern, not a claim that missingness biased this result. The registry record does not report enough information to evaluate the missingness mechanism or the robustness of the endpoint summaries.

The record also does not report a statistical analysis plan, analysis model, or detailed analysis-population rules beyond the stated endpoint definition. As a result, it is not possible to determine how protocol deviations, measurements below quantification limits, or incomplete concentration profiles were handled. Those details are not reported in the registry record.

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 serious adverse event counts are zero in every listed group, and deaths are zero in every listed group. For Part A, the registry lists 20 participants at risk in each formulation group, with zero serious adverse events in each. It lists 9 participants with other adverse events in the lyophilized group and 5 in the solution group. These are counts, not rates, and they should be read with their corresponding group sizes.

Other listed groups are the 0.5 mg intravenous Part B group, with 8 at risk and 1 with other adverse events; the Part C placebo group, with 4 at risk and 1 with other adverse events; and the Part C tirzepatide groups, with 12 at risk and 4, 12 at risk and 2, 12 at risk and 6, and 11 at risk and 4, respectively, across the listed dosing stages. The Part D 0.5 mg intravenous bolus group had 8 at risk and 3 with other adverse events. Serious adverse event counts and deaths were zero for each of these groups.

These summaries do not establish that no adverse events occurred: the record separately lists other adverse events. Nor do zero serious adverse events establish the absence of risk. The record does not list serious event terms, and it does not provide enough detail to assess event duration, severity grading, or relatedness. Part C group labels refer to stages of a dosing schedule; their counts should not be combined as if they were independent, concurrent treatment arms.

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

The posted primary result gives descriptive pharmacokinetic summaries for two Part A formulations in healthy participants. The recorded geometric means are 113000 for the 5 mg subcutaneous lyophilized formulation and 109000 for the 5 mg subcutaneous solution formulation. The associated geometric coefficients of variation are 19 and 17. These values describe the endpoint summaries and spread in the analyzed groups.

The result does not, by itself, establish that the formulations have equal or different exposure. The registry record does not report a contrast estimate, confidence interval, or formal test for the primary comparison. Without such information, the precision of any apparent difference cannot be judged. A conclusion about equivalence would also require a prespecified acceptable range and an analysis against that range, neither of which is reported.

The endpoint measures drug exposure over time. It does not directly measure clinical benefit, and this study involved healthy participants rather than a stated patient population. The registry also describes safety and tolerability objectives in other parts, but their separate counts do not resolve the Part A pharmacokinetic comparison.

The safety record lists zero serious adverse events and zero deaths across the listed groups, alongside participants with other adverse events. These are observations during the recorded study period, not proof that serious events cannot occur or that the regimens are safe in other settings. The study's phase, population, duration, sample sizes, and limited posted inferential detail constrain how broadly the results can be interpreted.

Overall, the record supports a narrow description of the measured exposure summaries and listed safety counts. It does not provide the uncertainty estimates, formal comparison, or broader clinical evidence needed to make a more definitive claim about the relative formulations or outcomes in other populations.

13. Why This Trial Matters Statistically

This study illustrates how an early-phase clinical trial can address several related but distinct questions within one registry record. Part A focused on pharmacokinetic exposure for two subcutaneous formulations, while the other parts addressed intravenous administration, repeated weekly subcutaneous dosing, placebo, or another intravenous formulation. Distinguishing those objectives prevents a reader from treating the overall enrollment as though it were the sample size for a single comparison.

The primary endpoint also illustrates why an endpoint's summary measure matters. AUC condenses concentration measurements over time into an exposure measure, while the geometric mean and geometric coefficient of variation describe the center and relative spread of those data. These summaries are useful for characterizing pharmacokinetics, but they are not substitutes for an estimated contrast and its uncertainty when the question is whether formulations differ.

The record is a useful reminder that a randomized, double-masked design does not guarantee that every inferential detail is visible in a registry result. The record classifies the design, reports group summaries, and defines the endpoint population, yet does not report a formal Part A comparison or confidence interval. Readers need to distinguish what is posted from what might have been specified or analyzed elsewhere; undocumented analysis choices should not be inferred.

Finally, the safety summaries demonstrate why counts need context. Serious adverse events and deaths are listed as zero, while other adverse events are recorded in several groups. Separating these categories and keeping each group attached to its own denominator avoids conflating the absence of recorded serious events with an absence of all adverse events. The study is statistically informative as a description of early pharmacokinetic and safety data, while its limited posted inferential detail calls for restraint in interpreting comparative claims.

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.

The primary endpoint result is limited to Part A and to the endpoint population of randomized participants who received at least one dose and had evaluable pharmacokinetic data. The registry record does not report the number initially randomized to each Part A formulation, a participant flow, or reasons why any measurements may not have been evaluable. This limits assessment of how the analyzed groups relate to all randomized participants.

The record reports geometric means and geometric coefficients of variation but no confidence limits, treatment contrast, p-value, or analysis model. Thus, the precision of the primary comparison and the method used to make it cannot be assessed. It is also not reported whether the protocol specified an equivalence or non-inferiority objective, or what criterion would have been used to interpret a formulation contrast.

Details of randomization, masking, allocation concealment, protocol deviations, and handling of incomplete or unusual pharmacokinetic profiles are not reported in the registry record. The record does not describe multiplicity control across the study's parts and outcome measures or provide sensitivity analyses for missing data.

The study was in healthy participants and included distinct regimens across four parts. Its results therefore describe the populations and regimens recorded, rather than establishing findings for other populations or treatment contexts. The safety summaries are counts; the registry does not list serious event terms or provide enough detail to assess duration, grading, or relatedness. The overall enrollment of 52 does not remove these limitations because participants were enrolled into different study parts and the primary endpoint uses a narrower population.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Tirzepatide (LY3298176) in Healthy Participants — NCT03375463.
• Registry record last updated 2024-02-05; status last verified 2024-02.
• Results first posted 2024-02-05.
• 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

NCT03375463 is a completed phase 1 randomized, double-masked, parallel study in healthy participants, with enrollment of 52. It began on 2017-12-19, and primary completion was recorded as 2018-12-27. The study comprised four parts with different pharmacokinetic, safety, and tolerability objectives.

The primary posted endpoint is Part A tirzepatide AUC from zero to infinity. The geometric mean was 113000 among 19 participants receiving the 5 mg subcutaneous lyophilized formulation and 109000 among 20 receiving the 5 mg subcutaneous solution formulation. The geometric coefficients of variation were 19 and 17. No formal comparison or confidence interval is reported in the registry record.

Serious adverse event counts and deaths were zero in all listed groups; other adverse events were recorded. The record does not report the original sample-size rationale, a detailed statistical analysis plan, or missing-data and multiplicity procedures.