A Study of the Effect of LY2216684 on Lorazepam: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Effect of LY2216684 on Lorazepam (NCT01275144), a randomized phase 1 study. The primary outcome is Pharmacokinetics of Lorazepam, Maximum Plasma Concentration (Cmax).

2. Clinical Question

The registry describes two treatment sequences: LY2216684 plus lorazepam followed by placebo plus lorazepam, or the reverse. This crossover arrangement means participants were scheduled to receive both conditions, rather than being assigned to remain in only one treatment group. The record does not report how many participants were assigned to each sequence or whether all participants completed both periods.

3. Trial Design

This was a completed, randomized, double-masked, crossover phase 1 study with 28 enrolled participants. The official title describes healthy subjects. Although Major Depressive Disorder appears as a listed condition, the record does not report that the participants had that diagnosis or explain why it is listed.

Each treatment period lasted 6 days for LY2216684 or placebo, with lorazepam given as a single dose on day 3. The LY2216684 dose was 18 mg orally; the lorazepam dose was 1 mg orally. Participants received the alternative study condition in the other period, with a washout of at least 7 days between dosing periods. The two sequences were LY2216684 plus lorazepam followed by placebo plus lorazepam, and placebo plus lorazepam followed by LY2216684 plus lorazepam.

Randomization and double masking are stated in the registry, but the method used to generate the allocation sequence, the process for concealing allocation, and the details of masking are not reported in the registry record. The numbers assigned to each sequence and the handling of period or sequence effects are also not reported.

The study began in 2010-12 and had primary completion in 2011-02. Its status is completed. Enrollment is 28. The primary analysis population is described as all enrolled participants, but that description does not establish how many participants contributed evaluable endpoint data. The number analyzed for the primary endpoint and the completion count for each treatment period 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.

First Intervention (6 Days)

Washout (7 Days)

Second Intervention (6 Days)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All enrolled participants.

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.

6. Primary Endpoint

The registry lists the primary outcome as Pharmacokinetics of Lorazepam, Maximum Plasma Concentration (Cmax), measured over Predose, 1, 2, 3, 4, 6, 8, 12, 24, 48, 72 and 96 hours post lorazepam dose.

The primary endpoint is lorazepam maximum plasma concentration, abbreviated Cmax, measured in nanograms per milliliter. Cmax represents the highest measured concentration of lorazepam in plasma after the dose. The registry lists measurements from predose through 96 hours post dose, at specified sampling times: 1, 2, 3, 4, 6, 8, 12, 24, 48, 72 and 96 hours, as well as predose.

Peak concentration is a pharmacokinetic measure: it describes the level of drug measured in the blood rather than a clinical outcome such as symptom improvement. Comparing Cmax under the two treatment conditions can characterize whether concomitant LY2216684 is associated with a difference in lorazepam exposure at its measured peak. This is relevant to the stated aim of examining how lorazepam behaves when given with LY2216684.

The registry reports geometric least-squares means and 90% confidence intervals. It says the geometric least-squares means were corrected for participant, treatment and random error. The ratio of those means is the primary between-condition comparison. A ratio near 1 indicates similar estimated geometric means; a ratio below 1 indicates a lower estimated mean in the LY2216684 plus lorazepam condition relative to placebo plus lorazepam. This interpretation concerns the endpoint estimate and does not establish a clinical consequence.

The registry does not describe the precise method used to identify or model the maximum from the multiple sampling times, nor does it report the endpoint's missing-data rules.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The central reported comparison is the ratio of geometric least-squares means for lorazepam Cmax. A least-squares mean is a model-based estimate of a group mean adjusted for terms in the model. The registry says the estimates were corrected for participant, treatment and random error. The full model specification, including its link or transformation details, is not reported in the registry record.

Why report a geometric mean?

Pharmacokinetic measurements are often analyzed on a logarithmic scale because concentrations can be skewed and multiplicative differences can be useful. In general, a geometric mean summarizes measurements by averaging on a log scale and transforming the estimate back to the original scale. A ratio of geometric means then expresses a relative contrast. The registry identifies the estimates as geometric least-squares means, but does not report the transformation or modeling steps in detail.

The reported ratio is 0.897, with a 90% confidence interval of 0.86 to 0.94. In general, a ratio of 1 represents equal estimated values in the two conditions; values below 1 indicate a smaller estimate in the numerator condition. Here the numerator condition is LY2216684 plus lorazepam, and the reference condition is placebo plus lorazepam. The ratio should be interpreted as a relative estimate, not as a percentage difference calculated separately from the registry's result.

What does the confidence interval convey?

In general, a confidence interval describes the precision of an estimate under the statistical model and its assumptions. A narrower interval indicates greater precision than a wider interval for the same type of estimate, while the interval's endpoints mark uncertainty around the estimate. The reported interval is at the 90% confidence level. It does not show the spread of individual participant values, and it does not remove concerns about model misspecification, incomplete data, or study conduct.

The condition-specific intervals are 9.178 to 10.08 for LY2216684 plus lorazepam and 10.23 to 11.24 for placebo plus lorazepam. Those intervals describe uncertainty in the estimated geometric means. The interval for the ratio is the more direct uncertainty statement for the comparison, because it is calculated for the contrast itself. Overlap or non-overlap of separate group intervals is not a substitute for the confidence interval of a treatment contrast.

How does the crossover structure matter?

In a crossover study, each participant is scheduled to receive more than one treatment condition in separate periods. The registry lists both possible orders and a washout of at least 7 days between periods. In general, participant-level comparisons can reduce the influence of stable differences among participants, provided that the statistical analysis uses the within-participant structure appropriately. The record states that participant was included in the correction of the geometric least-squares means. It does not report the model's covariance structure, period terms, sequence terms, or a formal assessment of carryover.

Washout periods are used in crossover designs to allow effects of an earlier treatment to diminish before the next period. The registry reports the washout duration but does not report evidence that carryover was absent, how it was assessed, or whether any participants were excluded because of carryover concerns. Those details should not be inferred from the presence of a washout period alone.

How should the sampling schedule be understood?

Cmax is the maximum plasma concentration assessed after lorazepam dosing. Measurements were scheduled predose and at 1, 2, 3, 4, 6, 8, 12, 24, 48, 72 and 96 hours post dose. In general, sampling times determine which portions of a concentration-time profile can be observed and inform how a maximum is identified or estimated. The registry does not state whether Cmax was taken as the largest observed concentration or derived using another procedure. It also does not report the handling of a missed sample or a participant without an estimable maximum.

What is not established by the reported method?

The registry does not report a p-value for the ratio comparison, a complete statistical analysis plan, the method used to determine the sample size, or the handling of interim analyses. It does not report whether any adjustments were made for multiple endpoints or analyses. The record includes a planning illustration based on the observed effect and specified general assumptions, but no numerical planning results are included in the registry record. Such planning figures, when presented, are not observed power and do not alter the uncertainty attached to the trial's actual estimate.

In general, over-dispersion occurs when variability in the data is greater than a model assumes. If present and unaccounted for, it can make standard errors and confidence intervals misleading. The registry entries do not report an over-dispersion assessment, and there is no basis to claim that it occurred in this study. More broadly, the confidence interval is conditional on the statistical model and available endpoint data; the record's limited model and missing-data details constrain how fully the analysis can be evaluated.

10. Analyses Reported by the Sponsor

The registry reports a sponsor analysis comparing LY2216684 plus lorazepam with placebo plus lorazepam. The reported ratio of geometric least-squares means is 0.897, with a 90% confidence interval of 0.86 to 0.94. The record identifies the treatment estimates as adjusted for participant, treatment and random error, but does not provide a complete model specification or a p-value.

No separate recalculated comparative estimate is reported in the registry record among the registry entries for this analysis. Therefore, a numerical comparison between the sponsor's estimate and an independent recalculation cannot be made from these results. The record does not flag a discrepancy between the sponsor result and a recalculated value. The sponsor's own analysis is the relevant posted inferential result; it should be interpreted with the stated confidence interval and the limits of the documented methods.

The individual geometric least-squares means are also reported: 9.62 ng/mL for LY2216684 plus lorazepam and 10.72 ng/mL for placebo plus lorazepam. The ratio is the direct contrast reported by the registry. It is preferable to treat that ratio and its interval as the comparative analysis rather than to infer a different test from the separate treatment estimates.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 24 outcome measures, 21 posted secondary outcomes and 3 analyses. The secondary outcomes include cognitive-function measures such as simple and choice reaction time, digit vigilance, false alarms, accuracy, and numeric working memory. These counts indicate that the record covers more than the primary pharmacokinetic endpoint, but do not specify a multiplicity strategy.

Multiplicity arises when several outcomes, contrasts, or statistical tests are considered. In general, testing many hypotheses can increase the chance of at least one apparently notable result by chance if no adjustment or prespecified hierarchy is used. The registry record does not report whether the primary endpoint was designated within a hierarchy, whether secondary outcomes were confirmatory or exploratory, or whether any adjustment for multiple testing was applied. No conclusions about the secondary outcomes can be drawn from their names alone.

The primary population is described as all enrolled participants, and enrollment is 28. That population label does not say how many contributed endpoint measurements or how missing observations were handled. The number analyzed for the primary endpoint, reasons for missing data, imputation procedures, and sensitivity analyses are not reported in the registry record. It is therefore not possible to determine whether the analysis included every enrolled participant for this endpoint or how incomplete sampling affected the estimate.

The registry lists randomized allocation and a crossover structure but does not report participant flow by sequence, treatment-period completion, protocol deviations, or exclusions from analysis. It also does not report whether an intention-to-treat, per-protocol, or pharmacokinetic evaluable analysis set was used beyond the description of the primary population as all enrolled participants. Those distinctions matter because an analysis population can affect which participants contribute to an estimate.

Interim analyses and stopping rules are not reported in the registry record. The record also does not report a detailed statistical analysis plan or a sample-size justification. The absence of these details means the planned treatment of multiplicity, missing data and analysis populations cannot be reconstructed from the posted record.

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 3% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports zero participants with serious adverse events in each of the four listed groups: LY2216684, placebo, LY2216684 plus lorazepam, and placebo plus lorazepam. Each group lists 28 participants at risk. The record does not list serious adverse-event terms.

The registry separately reports participants affected by other adverse events: 13 in the LY2216684 group, 3 in the placebo group, 6 in the LY2216684 plus lorazepam group, and 4 in the placebo plus lorazepam group. These are group-specific counts, not a basis for assuming that participants across periods are distinct. Because this was a crossover study and the record does not explain how adverse-event counts were attributed or deduplicated, comparisons between these counts should be made cautiously.

The deaths field is not populated in the record; deaths are not reported in the registry record, which is different from a reported count of zero. The adverse-event information should be read as descriptive registry reporting, not as a complete assessment of risk. Definitions, severity, timing, causal attribution, duration, and the method of collecting adverse events are not reported in the registry record. The absence of listed serious events does not establish that no safety concerns occurred outside the reporting presented in the record.

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

The result describes a pharmacokinetic comparison in a randomized, double-masked crossover study of 28 enrolled participants. The posted estimate for lorazepam Cmax is a ratio of 0.897 for LY2216684 plus lorazepam relative to placebo plus lorazepam, with a 90% confidence interval of 0.86 to 0.94. The condition-specific geometric least-squares means are 9.62 ng/mL and 10.72 ng/mL, respectively. The estimate and interval indicate how the two study conditions compared on this endpoint under the reported analysis.

This result does not establish whether the observed concentration difference has a clinically important consequence. The primary endpoint is a blood concentration measure, not a direct measure of patient benefit or harm. Nor does the result establish a conclusion about every aspect of lorazepam pharmacokinetics, such as total exposure or elimination, because the registry entries for this analysis report the Cmax comparison rather than a result for each pharmacokinetic parameter.

The study was conducted in healthy subjects according to its official title, while Major Depressive Disorder is listed as a condition. The record does not clarify that listing or report participant characteristics beyond the stated primary population. The findings should therefore not be assumed to apply to a particular patient group, clinical setting, dose, or regimen not described in the study.

The crossover design and the stated adjustment for participant are relevant to interpreting the comparison, but important analysis details remain unavailable. The record does not report endpoint missingness, the number analyzed, a complete model specification, or treatment of period and sequence effects. The absence of a reported p-value also means the registry result should be communicated using its estimate and confidence interval rather than an unreported significance test.

Finally, the safety counts are descriptive. The registry reports no serious adverse events in the listed groups and reports other adverse events separately, but it does not provide the details needed to characterize event severity, timing or relatedness. The statistical result alone cannot resolve those limitations.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of a pharmacokinetic comparison using a randomized crossover design. Instead of assigning participants to only one treatment condition, the design scheduled each participant to receive both LY2216684 plus lorazepam and placebo plus lorazepam in separate periods. The registry reports a washout of at least 7 days and two possible treatment sequences. That structure makes participant-level modeling particularly relevant to the comparison.

The reported ratio places the contrast directly on a relative scale. Alongside the geometric least-squares means, the 90% confidence interval communicates the precision of the estimate. This illustrates why a treatment estimate should be presented with an uncertainty interval: the point estimate alone does not convey how precisely the comparison was measured.

The record also illustrates the difference between a well-defined endpoint and a fully documented analysis. It specifies Cmax, measurement units, the sampling schedule, and the stated population. At the same time, it does not report a complete model, missing-data procedures, analysis counts for the endpoint, or multiplicity handling. A reader can interpret the posted estimate while still recognizing that the available detail is insufficient to reproduce every analytic choice.

Finally, pharmacokinetic estimates and clinical outcomes answer different questions. The reported Cmax ratio describes a comparative concentration measure under the study regimens. It is not, on its own, a measure of effectiveness or a complete safety evaluation. Keeping the endpoint's scope clear helps avoid extending a numerical contrast beyond the question the trial assessed.

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 leaves several trial-specific details unresolved. It describes healthy subjects in the official title but also lists Major Depressive Disorder as a condition; it does not explain this combination or report participant characteristics that would resolve it. The primary population is called all enrolled participants, but the number contributing evaluable Cmax data is not reported.

The crossover design is described through its two sequences and a washout of at least 7 days. However, the record does not report sequence allocation counts, period-specific completion, exclusions, or how period and sequence effects were handled. It also does not state whether carryover was assessed or how the maximum concentration was derived from the scheduled samples.

Other important analytic details are not reported in the registry record: the full statistical model, endpoint missing-data rules, sensitivity analyses, a p-value for the primary comparison, multiplicity procedures, and a sample-size justification. Randomization and double masking are stated, but their operational methods are not described. These omissions limit assessment of reproducibility and of the assumptions underlying the reported interval.

The adverse-event information does not include event definitions, severity, timing, relatedness, or an explanation of how counts were handled across crossover periods. The deaths field is unreported rather than explicitly recorded as zero. The registry's safety figures should therefore be treated as a limited description of the posted record, not as a comprehensive accounting of safety.

The trial is phase 1 and enrolled 28 participants. Its design and population constrain how far the endpoint estimate can be generalized. The record does not establish results in other populations or regimens, nor does the primary Cmax result answer questions about all pharmacokinetic parameters or clinical outcomes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of the Effect of LY2216684 on Lorazepam — NCT01275144.
• Registry record last updated 2019-03-11; status last verified 2018-11.
• Results first posted 2019-03-11.
• 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

This completed phase 1 study was randomized, double-masked and crossover, with 28 participants enrolled. The official title identifies healthy subjects. Participants received LY2216684 plus lorazepam in one treatment period and placebo plus lorazepam in the other, with a washout of at least 7 days.

The primary endpoint was lorazepam Cmax. The reported ratio of geometric least-squares means for LY2216684 plus lorazepam versus placebo plus lorazepam was 0.897, with a 90% confidence interval of 0.86 to 0.94. The registry reports zero serious adverse events in each listed group and separate counts of other adverse events.

Key details—including the number analyzed, missing-data handling, full statistical model, and multiplicity procedures—are not reported in the registry record.