A Study of LY3009104 in Healthy Participants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of LY3009104 in Healthy Participants (NCT01536951), a randomized phase 1 study. The primary outcome is Change From Baseline Through 24 Hours Postdose in Population-Corrected QT (QTcP) Interval.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, Phase 1 study with a crossover model. It had two parts and enrolled 62 healthy participants. The registry record lists the study as completed, with a start date of 2012-02 and a primary completion date of 2013-05. Eli Lilly and Company is identified as the sponsor.

Part A was a placebo-controlled, single-dose dose-escalation study. The stated purpose was to identify a safe and tolerable single oral dose that would yield drug exposures slightly exceeding those anticipated from repeated administration of an efficacious dose. The listed LY3009104 doses were 20 mg, 30 mg, and 40 mg. Placebo was assigned in one of three study periods, and LY3009104 in two of three periods; the LY3009104 doses were separated by at least a 3 day wash-out period.

Part B assessed the ECG effect of a single LY3009104 dose selected in Part A. It included placebo and moxifloxacin, which served as a positive control. Each treatment was assigned to a study period. The arm description specifies at least a 3 day wash-out period between moxifloxacin periods. The complete wash-out wording for the LY3009104 Part B description is not reported in the registry record. The treatment sequence and allocation details are also not reported in the registry record.

The study is described as participant- and investigator-blind, consistent with the registry classification of double masking. The record does not give numerical primary-endpoint findings or explain the statistical model used to compare treatments. Thus, the design and endpoint structure are available, but the treatment-effect estimates needed to interpret the ECG comparison are not.

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: The study had 2 parts. Part A: single-dose, dose-escalating study of LY3009104 \[up to 40 milligrams (mg)\] or placebo administered in each period. Part B: assessed the electrophysiological effects of a single supratherapeutic LY3009104 dose compared to a positive control (moxifloxacin) and placebo. Participants enrolled in either Part A or Part B.

Only the first 6 of 9 groups are shown.

First Intervention and Washout Period 1

Reasons for not completing this period:

Second Intervention and Washout Period 2

Reasons for not completing this period:

Third Intervention

Analysis population for the primary outcome, as defined in the registry: Participants enrolled in Part B of the study who had at least 1 dose of study drug (LY3009104, moxifloxacin, or placebo).

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: Participants who received at least 1 dose of LY3009104, moxifloxacin, or placebo and had at least 1 postdose safety assessment.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline Through 24 Hours Postdose in Population-Corrected QT (QTcP) Interval, measured over Part B, Periods 1 through 3: Baseline, 1 h, 1.5 h, 2 h, 3 h, 4 h, 6 h, 12 h, and 24 h postdose.

The primary endpoint was change from baseline through 24 hours postdose in the population-corrected QT interval, or QTcP, measured in milliseconds. The QT interval is the time between the start of the Q wave and the end of the T wave on an electrocardiogram. QTcP adjusts that interval for heart rate and the RR interval, the time between two R waves.

Heart rate correction is used because the QT interval can vary with heart rate. The record defines QTcP as QT/RR^beta. The population correction factor, beta, was estimated from a log-linear model fitted to predose QT and RR measurements from all periods and participants. Baseline was the average of measurements collected during the two hours before dosing on Day 1 of each period.

The endpoint was assessed at baseline and at 1 h, 1.5 h, 2 h, 3 h, 4 h, 6 h, 12 h, and 24 h after dosing in Part B. QTcP was not assessed in Part A. The moxifloxacin-to-placebo comparison at 1 h, 2 h, and 4 h was intended to establish assay sensitivity: whether the study measurements could detect an ECG response to the positive control. These endpoint details explain what was planned to be measured, but the numerical QTcP results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does QTcP measure?

The QT interval is taken from the ECG and represents the time from the start of the Q wave to the end of the T wave. Because QT duration varies with heart rate, the endpoint uses a corrected value, QTcP. The registry record defines the correction as QT/RR^beta, where RR is the interval between two R waves and beta is a population correction factor. Beta was estimated from a log-linear model of QT and RR using predose measurements across all periods and participants. This approach aims to reduce the influence of heart-rate variation when comparing postdose QT measurements.

General statistical explanation: A correction formula does not make the measurement error-free. The correction factor and model used to estimate it matter, and residual association with heart rate can remain. Interpretation therefore depends on the correction method, the data used to estimate its parameters, and the resulting treatment estimates. Here the registry record describes the correction method, but does not provide the numerical beta estimate or the primary outcome results.

Why use a baseline change?

The endpoint is expressed as change from baseline. Baseline was defined as the average of ECG data collected during the two hours before dosing on Day 1 of each period. Comparing postdose measurements with a period-specific baseline helps describe the shift after that period's treatment while accounting for the participant's starting measurement in that period.

General statistical explanation: A change score is one way to account for baseline. Statistical analyses may instead adjust for baseline in a model, or use another specified approach. Those choices can affect estimates and their uncertainty. The registry record defines baseline and the change endpoint but does not report the model used to analyze it, whether baseline was also included as a model term, or how missing measurements were handled.

How does a crossover design affect comparisons?

In a crossover study, participants receive treatments in different periods, allowing comparisons that may use within-participant information. Part A assigned placebo in one of three periods and LY3009104 in two of three periods. Part B included LY3009104, placebo, and moxifloxacin across periods. The specific randomized sequences are not reported in the registry record.

General statistical explanation: Within-participant comparisons can reduce the influence of stable participant characteristics because the same person contributes measurements under different treatments. However, crossover analysis must account appropriately for the repeated observations and the order and timing of treatment periods. Carryover, period effects, and sequence effects can matter if they are plausible and not addressed by the design or analysis. The registry record does not report the analysis model or its treatment of these factors. The Part A description specifies a wash-out of at least 3 days between LY3009104 doses. Part B's moxifloxacin description also specifies at least a 3 day wash-out between periods; the full LY3009104 Part B wash-out wording is not reported in the registry record.

Why are several postdose times important?

QTcP was assessed repeatedly from 1 h through 24 h postdose, with additional scheduled times at 1.5 h, 2 h, 3 h, 4 h, 6 h, and 12 h. A treatment effect may vary over time, so estimates at separate times can answer different questions. The record specifically identifies moxifloxacin-to-placebo comparisons at 1 h, 2 h, and 4 h for assay sensitivity.

General statistical explanation: Repeated measurements from one participant are correlated rather than independent. A model for repeated data can use that correlation to estimate treatment effects and uncertainty. The choice of model, covariance structure, and contrast determines how measurements across time contribute to the result. Those choices are not reported in the registry record, and the numerical estimates at the scheduled times are also not reported.

What do uncertainty and significance mean?

General statistical explanation: A treatment estimate describes the observed difference or change under the chosen analysis. A confidence interval describes the range of values compatible with the data and model at the interval's stated confidence level; it helps show precision as well as the direction and size of an estimate. A p-value measures how incompatible the observed data are with a specified null hypothesis under the analysis assumptions. It does not measure the probability that a treatment effect is true, nor does it express clinical importance.

The registry record does not report confidence intervals or p-values for the primary endpoint. It also does not report the statistical analysis plan, the primary comparison's exact contrast, or a designated summary across times. Consequently, no inferential conclusion can be drawn from the endpoint definition alone.

How should planning calculations be read?

General statistical explanation: A sample-size illustration based on an observed effect is a planning exercise, not an observed feature of the completed trial and not a measure of the trial's achieved power. A planning calculation depends on assumptions about the effect to detect, allocation, variability, significance level, and analysis method. The record describes planning assumptions based on the observed effect, with equal allocation and a normal approximation, but no usable primary comparison or numerical planning illustration is reported for this endpoint. Such planning assumptions should not be mistaken for the trial's actual design justification; that justification is not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint was measured at multiple postdose times, and the study also included distinct parts and treatment comparisons. This creates more than one possible statistical question: treatment differences at particular times, an overall time-profile question, and the moxifloxacin-to-placebo comparisons used to assess assay sensitivity. The record lists these measurements and the selected positive-control comparisons, but does not report a multiplicity procedure or identify how the primary inference was protected against the chance of false-positive findings across multiple comparisons.

General statistical explanation: When several hypotheses are tested, the chance of at least one false-positive result can increase if each is assessed without adjustment. A prespecified testing sequence, an adjustment procedure, or a clearly designated primary contrast can help define how the evidence should be interpreted. The statistical analysis plan and any such approach are not reported in the registry record. The missing numerical estimates also prevent assessment of any individual comparison.

The primary analysis population is described as Part B participants who had at least one dose of LY3009104, moxifloxacin, or placebo. The registry record does not report the number analyzed for each endpoint time, the number with missing ECG measurements, reasons for missing measurements, or the method used to handle missing data. It does not describe whether all eligible participants contributed to every treatment-period comparison.

The record also does not report whether the primary analysis was intention-to-treat, per-protocol, or another defined analysis set. In a crossover study, the population included in each period-specific comparison and the treatment of incomplete sequences can affect interpretation. These details, along with any sensitivity analyses, are not reported in the registry record. Thus, the endpoint population is defined broadly, but the actual analysis set and missing-data assumptions cannot be assessed.

11. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent non-serious adverse event terms

The registry record lists serious adverse events affected as 0 in every displayed study group. In Part A, the listed groups are placebo, 20 mg LY3009104, 30 mg LY3009104, and 40 mg LY3009104; their respective numbers at risk are 8, 6, 6, and 5. In Part B, the listed groups are placebo, 40 mg LY3009104, and moxifloxacin; their respective numbers at risk are 52, 53, and 53.

These are group-specific descriptive counts. They indicate that the registry record reports no serious adverse events among the participants at risk in each listed group. They do not show that the true risk is zero, and the record does not provide uncertainty intervals for these counts. Because the study used a crossover design, the groups are treatment-period groups rather than necessarily independent sets of participants; the figures should not be added or interpreted as independent group totals.

Death values are listed as null, so deaths are not reported in the registry record. The serious-event counts are separate from the other adverse-event counts and do not provide a complete account of tolerability. They also do not answer the primary ECG question: the numerical QTcP treatment estimates are not reported in the registry record.

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

This record establishes that the study was designed to assess QTcP after a single LY3009104 dose in healthy participants, using placebo as the comparator and moxifloxacin as a positive control in Part B. It defines how QTcP was corrected, how baseline was formed, when ECG measurements were scheduled, and which participants were eligible for the primary endpoint analysis. Those details clarify the question the trial set out to address.

The registry record does not report the numerical primary endpoint results. It therefore does not establish the size or direction of the QTcP change under LY3009104, the difference relative to placebo, or the precision of that difference. Without treatment estimates and confidence intervals, readers cannot judge how much uncertainty surrounds a potential effect. Without the moxifloxacin comparison results, they also cannot determine whether the planned assay-sensitivity assessment was met.

The listed serious adverse-event counts are zero in each displayed group, but this is a descriptive observation within the recorded study groups. It does not establish that serious events cannot occur, and death values are not reported in the registry record. The safety counts should not be used to fill the gap in the primary ECG results.

General interpretation: A trial's design and endpoint definition explain what was intended to be learned; only the outcome estimates and their uncertainty show what the observed data support. Even a precise estimate in a study of healthy participants would apply to the study's population, treatments, dose, timing, and analysis. The registry record does not provide the evidence needed for a quantitative conclusion about this trial's primary ECG question.

13. Why This Trial Matters Statistically

This study is statistically notable for its use of a crossover design and repeated ECG measurements. The same participant can contribute measurements under different treatments across periods, which can make within-participant comparisons informative. The design also requires appropriate analysis of repeated observations and period structure; those analysis details are not reported in the registry record.

The endpoint uses population-based heart-rate correction rather than relying only on an uncorrected QT interval. The record describes estimating the correction factor from predose QT and RR measurements across periods and participants. This makes clear that the endpoint depends on both ECG observations and a model-based correction procedure. Understanding the correction is important when interpreting any estimated treatment effect.

The positive control gives the study a second statistical purpose: the moxifloxacin-to-placebo comparison at selected times was intended to establish assay sensitivity. A positive-control result can help show that the measurement and study conditions were capable of detecting a response. However, the registry record does not report those comparison results, so that purpose cannot be evaluated from the recorded results.

Finally, this record illustrates why endpoint definitions and outcome results must be distinguished. It contains a detailed description of the QTcP measure and schedule, along with group-specific serious-event counts, but it does not report the numerical primary outcome estimates or their uncertainty. The design can be described; the primary statistical finding cannot be reconstructed from the registry record.

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 main limitation for interpreting the primary outcome is that numerical QTcP results are not reported in the registry record. There are no treatment estimates, confidence intervals, or p-values for the LY3009104-to-placebo comparison, and no reported results for the moxifloxacin assay-sensitivity comparisons. The outcome definition alone cannot establish the magnitude, direction, or precision of an effect.

The registry record describes a crossover design but does not report randomized treatment sequences, the full Part B wash-out description for LY3009104, or the statistical model used for the repeated measurements. It also does not report how period or sequence effects were handled. These omissions limit assessment of whether the analysis reflected the design features.

The record does not report the analysis plan, multiplicity handling, missing ECG measurement counts, reasons for missingness, or methods used to address missing data. Although the primary endpoint population is defined as Part B participants who received at least one study dose, the exact number contributing to each time-specific analysis is not reported. Demographic details beyond healthy males and females are not reported in the registry record.

The safety listing is also limited: serious adverse-event counts are reported by group, while death values are not reported in the registry record. These descriptive data do not replace the missing primary endpoint estimates or provide a full account of safety. Taken together, the record supports a description of the study's design and planned measurements, but not an independent quantitative assessment of its central ECG comparison.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of LY3009104 in Healthy Participants — NCT01536951.
• Registry record last updated 2017-06-06; status last verified 2017-05.
• Results first posted 2017-04-21.
• 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

NCT01536951 is a completed randomized, double-blind, crossover Phase 1 study sponsored by Eli Lilly and Company. It enrolled 62 healthy participants and had two parts: a placebo-controlled dose-escalation study and a placebo- and positive-control study of ECG effects. Part B assessed change from baseline in population-corrected QT interval through 24 hours after a single dose.

The registry record defines the QTcP calculation, baseline period, assessment times, and primary endpoint population. It also lists zero serious adverse events affected in each displayed group. Numerical primary-endpoint results, assay-sensitivity results, confidence intervals, p-values, and deaths are not reported in the registry record. The record therefore describes the trial question and design but does not permit a quantitative interpretation of the primary ECG findings.