A Comparative Study of LY900014 With Insulin Lispro (Humalog) in Participants With Type 2 Diabetes Mellitus (T2DM): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Comparative Study of LY900014 With Insulin Lispro (Humalog) in Participants With Type 2 Diabetes Mellitus (T2DM) (NCT03341312), a randomized phase 1 study. The primary outcome is Pharmacokinetics (PK): Insulin Lispro Area Under the Concentration Curve Zero to Seven Hours (AUC0-7hr).

2. Clinical Question

3. Trial Design

This was a randomized, double-masked, phase 1 crossover study. The registry lists an enrollment of 36 and a completed status. It records a start date of 2017-11-13 and a primary completion date of 2018-03-31. The study summary says each participant was expected to take part in 4 periods and spend approximately 12 weeks from screening to final study follow-up.

In the four treatment-period descriptions, participants received LY900014 immediately before a meal, LY900014 20 minutes after the start of a meal, insulin lispro immediately before a meal, or insulin lispro 20 minutes after the start of a meal. Each dose was individualized and administered subcutaneously. The primary pharmacokinetic description pools the two LY900014 timing conditions and separately pools the two insulin lispro timing conditions.

General: a crossover design gives participants more than one treatment condition over time, allowing treatment comparisons to be made within the same participant when the analysis uses the corresponding paired data. The registry record identifies the crossover structure but does not report the treatment sequence, the period-specific analysis method, or how possible period or sequence effects were handled. The posted primary results are pooled by treatment and do not include a formal comparison estimate.

The registry describes the purpose as evaluating pharmacokinetics and the effect on blood sugar following a standard meal. The primary endpoint concerns pharmacokinetics; the posted secondary endpoint concerns glucose response. Details of the randomization procedure, masking implementation, and sample-size justification are not reported in the registry record.

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

Period 3

Period 4

Analysis population for the primary outcome, as defined in the registry: All randomized participants who received at least 1 dose of study drug and evaluable PK data. LY900014 before and after meal and insulin lispro (Humalog) before and after meal were pooled by treatment to evaluate the overall PK of LY900014 and insulin lispro (Humalog) per protocol .

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 who received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Pharmacokinetics (PK): Insulin Lispro Area Under the Concentration Curve Zero to Seven Hours (AUC0-7hr), measured over Predose, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 70, 90, 120, 150, 180, 240, 300, 360, and 420 minutes (7 hours) postdose.

The primary endpoint is insulin lispro area under the concentration curve from zero to seven hours after dosing, abbreviated in the registry as AUC0-7hr. It is measured in picomol times hour per Liter. The registry describes blood sampling from predose through 420 minutes postdose, which is seven hours, and reports a geometric mean and geometric coefficient of variation for each pooled treatment group.

General: an area under a concentration-time curve summarizes exposure across a stated observation period. It uses information from the concentration profile over time rather than describing only one sampling point. In pharmacokinetic studies, this kind of measure is used to characterize the amount of drug exposure during the measured interval. The endpoint therefore addresses drug concentration exposure, not a clinical outcome or a direct measure of long-term glucose control.

The registry pools LY900014 before-meal and after-meal data to describe the overall pharmacokinetics of LY900014, and pools the corresponding insulin lispro periods. This pooling means the primary result does not separately report the pharmacokinetic endpoint for each meal timing. The registry lists 35 participants in each pooled treatment group. It provides no confidence interval or formal between-treatment test for this primary endpoint, so the posted geometric means and variability summaries should be read as descriptive results rather than as a complete inferential comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry classifies the study as randomized, double-masked, and crossover. It reports the primary endpoint as a geometric mean with geometric coefficient of variation. For the secondary endpoint, it reports means and standard deviations and gives separate reported analyses as ratios of least-squares means with 95% confidence intervals. These are distinct summaries and should not be treated as if they were estimates from one common model.

What does a geometric mean summarize?

General: the geometric mean is a measure of central tendency suited to values that are naturally interpreted on a multiplicative scale. It is commonly used for concentration and exposure data. The geometric coefficient of variation describes relative variability on that scale. These summaries differ from an arithmetic mean and standard deviation; they should be read using the statistic named in the registry rather than translated into another scale without the individual-level data or a stated analysis.

For the primary endpoint, the record posts the geometric mean and geometric coefficient of variation for each pooled treatment. It does not provide a primary treatment contrast, its standard error, confidence interval, or p-value. A difference between the two descriptive geometric means is not itself a formal estimate of the treatment effect, particularly for a crossover study where a paired analysis may use within-participant information.

What does a confidence interval say?

General: a confidence interval describes uncertainty around an estimate under the model and assumptions used to construct it. It is not a guarantee that the true value lies within the interval, and it does not directly express the probability that a treatment is better. For a ratio, the value 1 represents no ratio difference. For a difference, the value 0 represents no difference. Interpretation also depends on the question and any clinical or scientific threshold defined in advance.

The secondary table includes reported ratios of least-squares means and 95% confidence intervals for selected timing-specific comparisons. The registry does not identify the statistical method used to obtain those reported analyses, and it supplies no p-value for them. The accompanying tables also show simple unadjusted recalculations from posted summary statistics for some secondary comparisons. Those recalculations should be regarded as approximations based on the summary data, not as a replacement for the study's own analysis. In particular, an unadjusted calculation may not reflect the paired structure or other features of a crossover analysis.

Why can dispersion matter?

General: substantial variability makes estimates less precise, all else being equal. When data are highly variable or have a distribution for which a simple normal model is unsuitable, analysis choices can affect standard errors and intervals. The secondary results show standard deviations alongside means, and the primary results use geometric coefficients of variation. These measures describe spread differently and should not be compared as if they were interchangeable. The registry does not report diagnostic checks or a rationale for the selected analysis models.

How should the planning illustrations be read?

The sample-size section contains planning illustrations based on the observed secondary effects. The stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect set to a stated fraction of the observed effect. These are planning scenarios, not the trial's own design justification and not observed power. Their values are sensitive to the effect assumed: smaller target effects require larger planned groups under the illustration's assumptions. The registry's original sample-size rationale is not reported in the registry record.

The crossover design and pooled primary results are important when interpreting all these methods. A treatment comparison in a crossover study can depend on within-participant contrasts, but the registry does not provide the participant-level observations or enough analysis detail to reproduce such a model. It also does not report how period effects, sequence effects, missing observations, or multiplicity were handled. The displayed results support descriptive reading and limited interpretation of the reported secondary analyses, not reconstruction of an unreported full statistical plan.

10. Multiplicity, Missing Data and Analysis Populations

The primary population is defined as randomized participants who received at least 1 dose of study drug and had evaluable pharmacokinetic data. The registry gives an enrollment of 36, while the posted pooled primary results list 35 participants for each treatment. The record does not explain the difference between enrollment and the endpoint analysis counts, so the reason for any participant not appearing in a posted endpoint group is not reported in the registry record.

The registry does not report the amount or pattern of missing data, the method used to handle missing endpoint observations, or whether any imputation was performed. It also does not report a separate analysis population for the secondary endpoint. The description of the primary analysis population should not be assumed to apply to every secondary analysis.

Multiplicity is also not described. The secondary endpoint has several treatment and meal-timing conditions, and the registry posts more than one secondary comparison. No adjustment for multiple comparisons, prespecified hierarchy, or multiplicity strategy is reported in the registry record. The reported secondary confidence intervals and the unadjusted recalculations therefore should not be interpreted as if they were adjusted for the full set of analyses.

The record does not report interim analyses, stopping rules, or a rationale for the planned sample size. The sample-size section on this page is a planning illustration based on observed effects and stated assumptions; it is not evidence of the study's original design rationale or observed power. These omissions limit assessment of how closely the posted analyses follow a prespecified statistical plan.

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 table reports zero serious events in each of the four treatment-period groups and zero deaths in each group. The numbers at risk vary by group, with 35 participants listed for three groups and 36 for the LY900014 after-meal group. The record lists no serious adverse event terms.

These figures are counts within the listed groups, not estimates of comparative safety. In a crossover study, participants can contribute to more than one treatment condition, so the group entries should not automatically be treated as independent sets of people. The registry does not report a formal statistical comparison of serious adverse events or a detailed exposure-adjusted analysis.

The table also includes counts for other adverse events, but those figures are distinct from serious events. A zero count of serious events in these groups does not demonstrate that serious events cannot occur, and the registry summary cannot establish a safety profile beyond the observations recorded for this study. The duration, size, and phase of the study are important context when reading sparse event counts.

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

The posted primary data describe insulin lispro exposure over a seven-hour period for pooled LY900014 and insulin lispro treatment conditions. They provide geometric means and measures of relative variability, but not a formal primary comparison or a confidence interval around one. The page therefore shows what the registry reports for each pooled group without establishing the precision of a treatment contrast.

The secondary results describe glucose response under different treatment and meal-timing conditions. They include descriptive summaries and selected reported comparisons. Those results are not a substitute for the missing primary inferential comparison, and unadjusted secondary analyses carry additional uncertainty when several comparisons are examined.

General: absence of a statistically significant result, where one is reported, is not by itself evidence that two treatments are equivalent. An equivalence or noninferiority conclusion requires a prespecified margin and an analysis designed for that question. Such a margin is not reported in the registry record. Nor does a pharmacokinetic or meal-related glucose endpoint alone establish an effect on longer-term clinical outcomes.

This was a phase 1 study with a crossover design and a stated focus on pharmacokinetics and glucodynamics. Its posted results can inform description of measured exposure and glucose response under the study conditions. They do not establish what would happen in other populations, over longer periods, or outside the study setting. The registry does not provide enough detail to reconstruct the full statistical analysis or to resolve the uncertainty around the primary comparison.

13. Why This Trial Matters Statistically

This trial is statistically interesting because its crossover structure and its posted summaries do not line up as a simple independent two-group comparison. In general, crossover data can use within-participant comparisons, but the primary results are pooled by treatment and presented as geometric means with geometric coefficients of variation. The registry does not post a formal contrast or enough model detail to determine how the crossover structure entered the primary analysis.

The study also distinguishes pharmacokinetics from glucodynamics. The primary endpoint summarizes drug exposure over time, while the secondary endpoint summarizes glucose response around a meal. These address related but different questions. Keeping the endpoint types separate prevents an exposure summary from being mistaken for a direct clinical outcome or for a complete account of glucose response.

The record illustrates why numerical summaries should be read alongside their uncertainty and analysis context. The primary geometric means are informative descriptions, but no primary interval or test is posted. The secondary comparisons include both reported analyses and unadjusted recalculations, and the registry does not report a multiplicity plan. Each fact limits how strongly the observed patterns can be interpreted.

Finally, the sample-size illustrations are planning scenarios based on observed effects, not proof that the study had a particular probability of detecting an effect. The study's own sample-size rationale is not reported in the registry record. This distinction matters whenever a small exploratory study's estimates are used to consider the scale of a future study: planning assumptions should be explicit, and uncertainty in the observed effect should not be overlooked.

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 lacks a posted formal treatment comparison, confidence interval, and p-value. The descriptive values and variability summaries alone do not quantify uncertainty around a difference between treatments. The record also does not report the analysis model or explain how within-participant information from the crossover design was used.

The primary results pool before-meal and after-meal periods within each treatment. This is appropriate to the stated pooled pharmacokinetic description, but it means the primary table does not show separate pharmacokinetic estimates for each timing condition. The registry does not report the treatment sequence or how period and sequence effects were addressed.

The secondary endpoint covers multiple treatment and timing conditions, with selected reported analyses and unadjusted recalculations. A multiplicity strategy is not reported. The reported analyses do not include p-values, and their statistical method is not identified in the registry record. The recalculated comparisons use summary statistics and are not a replacement for an analysis that uses the crossover data directly.

The registry does not report missing-data handling, the amount or pattern of missing observations, or a sample-size justification. The endpoint counts are not the same as the overall enrollment count, but the record does not explain why. The study is also described as phase 1 and focuses on pharmacokinetics and meal-related glucose response; it does not report longer-term clinical outcomes. These limitations constrain both the strength and the scope of conclusions that can be drawn from the posted results.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Comparative Study of LY900014 With Insulin Lispro (Humalog) in Participants With Type 2 Diabetes Mellitus (T2DM) — NCT03341312.
• Registry record last updated 2020-06-09; status last verified 2018-04.
• Results first posted 2020-06-09.
• 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

The registry record describes a completed, randomized, double-masked, phase 1 crossover study of LY900014 and insulin lispro (Humalog) in participants with type 2 diabetes. Enrollment was 36. Participants received treatment in four periods, including before-meal and after-meal conditions for each insulin treatment.

The primary endpoint was insulin lispro AUC0-7hr, reported as a geometric mean and geometric coefficient of variation for pooled treatment groups. The registry also reports a secondary glucose-response endpoint across treatment and timing conditions. A formal primary comparison, detailed statistical methods, missing-data handling, and multiplicity procedures are not reported in the registry record.