A Study of LY2835219 in Healthy Participants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of LY2835219 in Healthy Participants (NCT02059148), a randomized phase 1 study. The primary outcome is Pharmacokinetics: Maximum Concentration (Cmax) of LY2835219 Fasted vs. High-Fat Arms.

2. Clinical Question

3. Trial Design

This was a Phase 1, randomized, open-label crossover study of LY2835219 in healthy participants. In a crossover design, participants receive more than one study condition at different periods, so comparisons can be made within the same participant. The registry describes three single oral doses, given under fasted, standard-meal, and high-fat-meal conditions. It states that the doses were separated by at least 14 days. The study was not masked.

The planned study duration was approximately 43 days, excluding screening. Screening was required within 28 days before the study began. The registry record lists the study as completed, with a start date of 2014-02 and primary completion date of 2014-08. It lists enrollment as 24. Those details describe the overall study record; they do not mean every participant contributed to every posted outcome.

The primary outcome is maximum concentration of LY2835219 for the fasted-versus-high-fat comparison. The arm summaries list 23 participants for the fasted condition and 24 for the high-fat condition. These are condition-specific counts in the posted outcome and should not be read as separate parallel groups: the design is crossover, and participants could contribute observations under multiple conditions. The registry record does not report the order in which participants received conditions, the allocation sequence, or details of any washout rationale.

The record also lists secondary pharmacokinetic outcomes, including Cmax, AUC(0-tlast), and Tmax for the standard-meal condition, as well as an AUC measure for the metabolite LSN3106726 (M20). Four secondary outcomes are listed. Values for some comparisons are missing or not estimable in the posted results. The record does not provide enough information to reconstruct all planned comparisons or the complete analysis workflow.

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.

Treatment 1

Reasons for not completing this period:

Treatment 2

Treatment 3

Analysis population for the primary outcome, as defined in the registry: All randomized participants receiving at least one dose of the investigational product.

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: Single Oral Dose of 200 mg LY2835219 on 3 Occasions: Standard Meal, Fasted State, High-Fat Meal.

6. Primary Endpoint

The registry lists the primary outcome as Pharmacokinetics: Maximum Concentration (Cmax) of LY2835219 Fasted vs. High-Fat Arms, measured over Predose, 0.5,1,2,3,4,6,8,10,12,24,48,72,96,120,144,168,192 hours post-LY2835219 dose in each period.

The primary endpoint is the maximum concentration, or Cmax, of LY2835219 in nanograms per milliliter. It represents the highest measured concentration in the blood during the stated sampling schedule, which runs from predose through 192 hours after dosing in each period. In pharmacokinetic studies, Cmax is used to describe the peak concentration reached after a dose. It is one aspect of drug exposure; it does not by itself describe total exposure over time or establish a clinical effect.

The registry reports Cmax using a geometric mean and a geometric coefficient of variation. A geometric mean is commonly used for positive concentration measurements that are right-skewed: it is calculated on a logarithmic scale and expressed back on the original scale. The geometric coefficient of variation describes relative dispersion on a multiplicative scale. It is not the same as a confidence interval and does not directly give the precision of the estimated mean.

Comparing Cmax under fasted and high-fat conditions addresses whether the observed peak concentration differed with those meal conditions in this study. This can characterize a food-related pharmacokinetic contrast, but it does not alone establish what happens to other pharmacokinetic features, such as the timing of the peak or total exposure. The registry lists AUC and Tmax among secondary outcomes, but their values are not included in the primary result summarized in this comparison.

The endpoint’s clinical interpretation depends on the study context and other evidence. The posted Cmax values describe concentration measurements in healthy participants following single doses. They do not establish how the drug performs in a patient population or whether any observed concentration contrast changes a clinical outcome.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a geometric mean describe?

Concentration data are positive and can have a right-skewed distribution, with some observations farther above the center than others are below it. A geometric mean summarizes such data using logarithms: values are transformed, averaged on that scale, and then transformed back. This makes the summary suitable for multiplicative variation and for contrasts naturally expressed as ratios. It differs from an arithmetic mean, which averages the raw values directly.

The registry reports geometric means of 111.0 ng/mL for the fasted condition and 137.0 ng/mL for the high-fat condition. It also reports geometric coefficients of variation of 61.0 and 52.0, respectively. A geometric coefficient of variation summarizes relative spread on a multiplicative scale. It is a description of variability among measurements, not a confidence interval for a group mean, and not an estimate of the uncertainty in the ratio between conditions.

Why use a ratio for the comparison?

When an outcome is summarized geometrically, a ratio is a natural way to express a contrast. A ratio of 1 corresponds to equal values under the two conditions; values on either side of 1 represent a multiplicative contrast. The registry reports a ratio of geometric least-squares means of 1.24 for the fasted-versus-high-fat comparison. It does not clearly identify the numerator and denominator convention in the posted analysis, so the ratio is best quoted as reported rather than given an inferred direction.

“Least-squares means” are model-based averages, often estimated after transforming the outcome or accounting for features in an analysis model. They may not be identical to a simple ratio calculated directly from the two arm-level geometric means. The registry does not specify the model used for this estimate. It does not report the adjustment terms, model diagnostics, or the exact handling of within-participant correlation. Those omissions prevent an independent reconstruction of the sponsor’s analysis from the summary statistics alone.

How should the confidence interval be read?

The reported 90% confidence interval extends from 1.11 to 1.38. A confidence interval gives a range of estimates compatible with the data and the assumptions of the statistical procedure. Its width conveys precision: a narrower interval indicates more precise estimation than a wider one, all else equal. It is not a statement that individual measurements fall within the interval, nor does it give the probability that a fixed, unknown value lies in this particular realized interval.

For a ratio, 1 is the reference value for no difference in the geometric means. The reported interval lies on one side of that reference. This is a statistical feature of the posted estimate, not a decision about whether the contrast is clinically meaningful. The registry does not report a p-value for this analysis or a clinical threshold for interpreting the estimate.

Why does crossover design matter?

In a crossover study, participants receive multiple conditions in separate periods. Measurements from conditions within the same participant are correlated; they are not independent observations in the way measurements from separate parallel groups might be. A suitable analysis can account for that pairing and may use within-participant information to estimate the contrast. General statistical principle: ignoring relevant within-participant correlation can give an inaccurate estimate of uncertainty.

The registry identifies the study as randomized and crossover, and describes dosing under fasted, standard-meal, and high-fat-meal conditions. It does not report the sequence allocation, the analysis model, or whether period and sequence effects were included. It also does not describe how incomplete crossover records were handled. As a result, readers should not attempt to recreate the reported interval by treating the posted condition summaries as independent groups.

What does variability mean in this setting?

The geometric coefficients of variation describe relative dispersion among the observed concentrations. Variability can be important in pharmacokinetic studies because participants may have differing concentration profiles, and an average does not capture every individual pattern. General statistical principle: when data are more variable, an estimate may be less precise unless other aspects of the design or analysis compensate. The coefficients alone cannot establish the precision of the treatment-condition contrast, which is represented more directly by its confidence interval.

Over-dispersion, in general, means that observed variability is greater than a model assumes. If a model underestimates variability, standard errors can be too small and intervals too narrow. The registry record does not report whether over-dispersion was assessed or whether any correction was used. The available summaries are insufficient to judge that issue.

What can planning calculations tell readers?

Planning calculations based on an observed effect are illustrations of how a future study might be sized under stated assumptions. In this case, the stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is described as a fraction of the observed effect. For non-time-to-event outcomes, the stated convention is participants per arm for 80% power or another listed power target. Such calculations are not observed power and do not establish the design assumptions used for the original trial.

The record does not provide the original study’s sample-size justification or statistical analysis plan. Planning illustrations based on the observed contrast are conditional on the chosen effect, allocation, significance level, and approximation. They should not be interpreted as proof that the completed study was adequately or inadequately powered. The registry does not report a protocol-based justification that would allow that question to be assessed directly.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s comparison as a ratio of geometric least-squares means of 1.24, with a 90% confidence interval from 1.11 to 1.38. It also gives condition-level geometric mean Cmax values and geometric coefficients of variation. The sponsor’s model-based ratio is the reported inferential result; the condition summaries are descriptive and do not by themselves reproduce the model-based estimate or its interval.

The registry record does not identify the analysis model or state the numerator and denominator convention for the ratio. Therefore, the comparison can be described as the reported fasted-versus-high-fat analysis, while the ratio should not be assigned a direction from the label alone. A separate recalculated ratio and interval are not reported in the registry record. The record does not flag a discrepancy between the sponsor result and a recalculation.

The p-value is not reported in the registry record. The interval is the available measure of precision for the sponsor’s estimate. No independent conclusion about the adequacy of the model can be drawn from the posted arm summaries alone, particularly because this is a crossover study and the within-participant analysis details are not reported.

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary Cmax comparison and four secondary outcomes. These include pharmacokinetic measures for the standard-meal condition and an AUC measure for the metabolite LSN3106726 (M20). The primary result concerns fasted versus high-fat conditions; the listed secondary measures are not interchangeable with that primary contrast.

When several endpoints or comparisons are tested, the chance of at least one apparently unusual result can increase. General statistical principle: multiplicity procedures can be used to control that issue, for example by adjusting significance thresholds or interpreting a prespecified primary endpoint separately. The registry record does not report whether multiplicity was addressed, which comparisons were prespecified, or how the secondary outcomes were intended to support interpretation. It also does not state whether there were interim analyses.

Values for some comparisons are described as missing or not estimable for a two-arm comparison for another endpoint. The registry does not explain why those values are unavailable, whether the outcome was not collected, or whether estimation failed. Missing outcome values can affect interpretation, especially if their absence is related to participant characteristics or measured outcomes. The record does not describe missing-data methods or sensitivity analyses.

The primary outcome population is defined as all randomized participants receiving at least one dose of the investigational product. The record does not provide a fuller analysis-population specification, such as how participants with incomplete dosing periods were handled. Although condition-specific counts are reported for the Cmax summaries, the registry does not explain the participant flow into each analysis or the reasons for any difference in those counts.

The analysis plan, sample-size justification, and details of outcome-data handling are not reported in the registry record. Without those details, it is not possible to assess how closely the posted analyses followed a prespecified plan or whether missing outcomes and multiple comparisons altered the interpretation.

12. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for the standard-meal, fasted, and high-fat conditions. In each condition, the reported number of participants at risk is 23, 23, and 24, respectively, and the reported number affected by a serious adverse event is 0 in each condition. These are condition-specific safety entries in a crossover study, not counts of separate parallel treatment groups.

The record lists other adverse events affecting 4 participants in the standard-meal condition, 6 in the fasted condition, and 4 in the high-fat condition. These figures should not be combined across conditions: participants could contribute to more than one condition, and the registry does not report how events were counted across periods. No serious adverse-event terms are listed.

The death counts are not reported in the registry record. The absence of reported serious adverse events in these entries is a description of the posted safety data, not proof that no serious event could occur or that the conditions have equivalent safety. The study was conducted in healthy participants and the registry record does not provide detailed event descriptions, severity, causality assessments, or a broader account of follow-up beyond the listed results.

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

The result answers a narrow pharmacokinetic question: the registry reports Cmax summaries and a sponsor analysis comparing fasted and high-fat-meal conditions after single oral doses of LY2835219 in healthy participants. The geometric mean Cmax values were 111.0 ng/mL under fasted conditions and 137.0 ng/mL under high-fat conditions. The reported ratio of geometric least-squares means was 1.24, with a 90% confidence interval from 1.11 to 1.38.

The interval conveys statistical precision under the analysis used, but the registry does not report a clinical interpretation threshold. The result therefore does not establish whether the size of the contrast is important for a clinical decision. It also does not determine whether the meal conditions affect total exposure, peak timing, or other outcomes. AUC and Tmax appear among the listed secondary outcomes, but their results are not part of the primary comparison.

The study’s crossover structure means that the conditions were assessed across periods in the same participants. That can provide a useful within-participant comparison, but the record does not give the analysis details needed to assess period, sequence, or incomplete-observation effects. The posted condition summaries alone are not a substitute for the sponsor’s crossover analysis.

The study was conducted in healthy participants and involved single doses. It does not establish what the same comparison would show in a patient population, with repeated dosing, or for clinical outcomes. The registry does not report a complete statistical plan, multiplicity approach, or detailed missing-data handling. These limitations leave uncertainty about the broader evidence base and about how the posted analysis fits into all planned comparisons.

The result should be read as a recorded estimate of a concentration contrast, with the stated interval indicating its precision. It neither proves a clinical benefit nor establishes a clinical harm. It also does not support advice about how any individual should take a medicine. Questions about clinical relevance require evidence beyond this pharmacokinetic comparison.

14. Why This Trial Matters Statistically

This study is statistically useful as an example of a crossover pharmacokinetic comparison. Participants received LY2835219 under fasted, standard-meal, and high-fat-meal conditions, allowing the study to examine concentration outcomes under different food conditions within a crossover framework. That design differs from a parallel-group trial: condition measurements may be linked within participants, so the analysis should reflect that dependence.

The primary outcome also illustrates why the reported summary and its uncertainty should be read together. Geometric means describe concentration levels on a multiplicative scale, while the geometric coefficient of variation describes relative spread. The sponsor’s ratio of geometric least-squares means and its confidence interval present the comparison and its precision. These quantities answer different statistical questions and should not be substituted for one another.

The record’s level of reporting also demonstrates an important interpretive limit. It identifies randomization and crossover design and gives an inferential result, but does not report the analysis model, ratio orientation, or details about period and sequence effects. Those omissions matter because readers cannot fully reconstruct a crossover analysis from condition-level summaries alone.

Finally, a pharmacokinetic contrast is not itself a clinical outcome. Cmax describes a peak concentration, whereas other listed measures such as AUC and Tmax address other aspects of drug exposure. The distinction helps keep conclusions aligned with the endpoint actually reported. The study can inform how the posted Cmax summaries compare across food conditions, while broader clinical interpretation remains outside what these registry results establish.

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 registry record does not report the statistical analysis model, the numerator and denominator convention for the reported ratio, or whether the model accounted for period, sequence, or within-participant correlation. These details are particularly relevant for interpreting a crossover comparison. The arm-level geometric means and coefficients of variation cannot independently reproduce the model-based ratio and confidence interval.

The record does not report a p-value for the primary comparison, a prespecified threshold for clinical importance, or a sample-size justification. It also does not describe interim analyses or any multiplicity approach for the primary and secondary outcomes. As a result, the relationship between the posted analysis and the complete planned statistical strategy cannot be assessed from the registry record.

Although the study describes three dosing conditions, the posted primary result concerns only fasted and high-fat conditions. The record lists secondary outcomes, but does not provide their complete results in the registry record for this comparison. Some values are missing or not estimable for a two-arm comparison for another endpoint, and the reason is not reported in the registry record.

The record gives condition-specific numbers for the Cmax summaries but does not explain the participant flow into each analysis or the difference between the listed condition counts. It does not describe how incomplete periods or missing measurements were handled. This makes it difficult to judge whether all eligible observations contributed consistently to the comparison.

Generalizability is limited by the recorded population and dosing context: healthy participants received single oral doses. The registry does not establish that results apply to patients, repeated use, or clinical outcomes. The record also does not provide detailed demographic characteristics or a fuller account of adverse-event descriptions and follow-up. These constraints should accompany interpretation of both the pharmacokinetic estimates and the safety entries.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of LY2835219 in Healthy Participants — NCT02059148.
• Registry record last updated 2019-01-07; status last verified 2018-12.
• Results first posted 2018-08-16.
• 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

NCT02059148 is a completed Phase 1 randomized, open-label crossover study of LY2835219 in healthy participants. The registry lists enrollment of 24 and describes single oral doses under fasted, standard-meal, and high-fat-meal conditions. The primary outcome is Cmax for the fasted-versus-high-fat comparison, measured through the stated post-dose sampling schedule.

The posted geometric mean Cmax values are 111.0 ng/mL for the fasted condition and 137.0 ng/mL for the high-fat condition. The sponsor-reported ratio of geometric least-squares means is 1.24, with a 90% confidence interval from 1.11 to 1.38. The registry does not report a p-value or the model details needed to reconstruct the crossover analysis.

The record lists secondary pharmacokinetic outcomes and condition-specific adverse-event summaries. It does not report the full analysis plan, multiplicity handling, or missing-data methods. The primary result is a pharmacokinetic comparison in healthy participants, not a clinical outcome.