A Study of Mirikizumab (LY3074828) in Participants With Crohn's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Mirikizumab (LY3074828) in Participants With Crohn's Disease (NCT03926130), a randomized phase 3 study. The primary outcome is Percentage of Adult Participants Achieving Clinical Response at Week 12 and Endoscopic Response at Week 52 (Placebo and Mirikizumab).

2. Clinical Question

This is a question about a defined combination of patient-reported symptoms and endoscopic change, rather than a single symptom or a general measure of disease status. Requiring both response milestones makes the endpoint reflect outcomes at distinct times: clinical response at Week 12 and endoscopic response at Week 52. The registry reports the primary analysis for participants meeting its adult endoscopic eligibility criteria and receiving at least one dose. It does not establish that every randomized participant was included in that analysis, nor does the registry record describe all details of how missing outcome information was handled. The posted result should therefore be read in the context of this stated analysis population and endpoint definition, not as a result for every possible population or outcome of interest.

3. Trial Design

This was a phase 3, multicenter, randomized, double-blind, placebo- and active-controlled, treat-through study in participants with moderately to severely active Crohn's disease. The registry classifies the allocation as randomized, masking as triple, and the study model as parallel. Total enrollment was 1158. The study is recorded as completed. The registry lists Eli Lilly and Company as sponsor.

The treatment arms included placebo, mirikizumab, and ustekinumab, along with a mirikizumab arm for adolescents. The adult mirikizumab regimen was 900 mg intravenously every 4 weeks for 3 doses and then 300 mg subcutaneously every 4 weeks. The ustekinumab regimen was 6 mg/kg intravenously for one dose and then 90 mg subcutaneously every 8 weeks. The placebo arm used intravenous or subcutaneous placebo every 4 weeks. Placebo-arm non-responders at Week 12 switched to mirikizumab for the rest of the study, as described in the registry.

Although the trial included an active comparator, the primary reported comparison was placebo versus mirikizumab. Its analyzed groups contained 199 placebo participants and 579 mirikizumab participants. Those are the denominators reported for the primary outcome and should not be confused with total trial enrollment or assumed to represent all randomized participants. The registry record does not report the rationale for the allocation pattern or a sample-size justification.

The primary outcome spans Week 12 through Week 52. Secondary outcomes include symptom, clinical, endoscopic, and laboratory measures at different follow-up times. For some outcomes, placebo non-responders had crossed over to active treatment by later follow-up. Consequently, long-term comparisons involving the original placebo arm require attention to the treatment changes and the analysis used. Details sufficient to reconstruct all aspects of the trial's analysis plan 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants in Placebo and Mirikizumab arms who have baseline SES-CD ≥7 (or ≥4 for isolated ileal disease) and received at least one dose of study drug.

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: Adults: All randomized participants; Adolescents: All participants who received at least one dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Adult Participants Achieving Clinical Response at Week 12 and Endoscopic Response at Week 52 (Placebo and Mirikizumab), measured over Week 12 to Week 52.

The primary endpoint is the percentage of adult participants achieving clinical response at Week 12 and endoscopic response at Week 52. Clinical response by patient-reported outcome required at least a 30% decrease in stool frequency and/or abdominal pain, with neither score worse than baseline. Stool frequency refers to liquid or very soft stools, defined using Bristol Stool Scale categories 6 or 7. Abdominal pain was assessed on a 4-point scale from none to severe.

Endoscopic response required at least a 50% reduction from baseline in the total SES-CD score. The SES-CD assesses ulcers, ulcerated surface, affected surface, and narrowing across 5 bowel segments. Each feature contributes to a total score ranging from 0 to 56, with higher scores indicating more severe disease. The combined endpoint therefore includes a symptom-based component and an endoscopic component, assessed at different follow-up times.

This type of endpoint is used to evaluate whether a response is evident in both the participant-reported clinical domain and an endoscopic measure. It is not equivalent to complete symptom resolution, endoscopic remission, or improvement on every individual measure. Participants who meet the endpoint have to satisfy the specified response criteria at both milestones; the registry's percentage is the proportion meeting that composite definition. General statistical interpretation: combining distinct outcomes can make a result more clinically multidimensional, but the meaning depends on the exact criteria and timing, and the combined percentage does not reveal how each component behaved on its own.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a percentage, so the arm-level estimates describe the proportion of analyzed participants who met the endpoint criteria. The registry reports 9.0% for placebo and 38.0% for mirikizumab, with separate 95% confidence intervals for those percentages. It also reports a between-arm risk difference of 28.7 percentage points and a 95% confidence interval from 23.0 to 34.4. The risk difference expresses the contrast in absolute percentage points, rather than a relative change. General explanation: an absolute difference is often easier to interpret as a change in the proportion meeting an outcome, while a relative measure answers a different question and can appear larger or smaller depending on the underlying event frequency.

The reported primary analysis used the Cochran-Mantel-Haenszel method and is labeled as a risk-difference analysis. In general, this method can estimate an association between treatment group and a categorical outcome while accounting for strata. The registry record does not identify the stratification variables or give enough detail to reconstruct the full procedure. The primary result should therefore be interpreted using the sponsor's posted estimate and confidence interval, rather than assuming a particular set of adjustments.

A confidence interval expresses uncertainty in an estimate under the model and analysis procedure used. It does not mean that the true value has a particular probability of lying in the calculated interval after the data have been observed, and it does not capture uncertainty from every possible design or reporting issue. Here, the primary interval is reported at the 95% level. Its endpoints indicate that the estimated contrast is more precisely constrained than a point estimate alone would suggest, but readers should not treat its width as a complete summary of uncertainty about all outcomes or populations.

Some secondary outcomes are continuous measures reported as least-squares means with standard errors. A least-squares mean is a model-based group estimate, often adjusted for terms in the analysis model; the registry identifies ANCOVA as the sponsor's method for these reported comparisons. A standard error describes the estimated sampling variability of a mean estimate. General explanation: standard errors are not the same as standard deviations, which describe variation among individual observations. The ANCOVA comparison is reported as a net least-squares mean difference with a confidence interval. The registry record does not provide the complete model specification, covariates, or assumptions needed to reproduce that adjusted analysis in full.

Separate unadjusted re-calculations are displayed for selected continuous outcomes using posted summary statistics. These are not the sponsor's ANCOVA results. They use simple unadjusted methods and, where stated, an approximate standard deviation derived from the posted standard errors. This distinction matters: a calculation based on summary statistics may not match an adjusted model that accounts for baseline or other covariates. For those secondary outcomes, the sponsor's reported ANCOVA analysis is the relevant posted analysis; the re-calculation offers a complementary check and should not replace it.

General explanation: over-dispersion occurs when outcome variability is greater than a model's assumed variability. If it is ignored, standard errors and confidence intervals can be too small, making the evidence appear more precise than it is. The registry does not report whether over-dispersion was assessed or relevant to these analyses. No specific claim about it can be made from the posted results.

Sample-size values accompanying the secondary calculations are planning illustrations based on the observed effect, not the trial's design calculations and not observed power. They assume a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the target effect is a stated fraction of the observed effect. General explanation: planning calculations illustrate the sample size that would be needed under those assumptions for a future comparison. They do not demonstrate that the completed trial had a particular probability of finding a result, and they do not substitute for the registry's unreported original sample-size justification.

The registry record does not report the full statistical analysis plan, the handling of missing outcomes, or whether interim analyses were conducted. Those details can affect how estimates are interpreted, especially for outcomes assessed over extended follow-up and in the presence of treatment crossover. They should not be inferred from the fact that results are posted or from the method labels alone.

10. Analyses Reported by the Sponsor

The registry reports the primary comparison as a Cochran-Mantel-Haenszel risk difference of 28.7 percentage points, with a 95% confidence interval from 23.0 to 34.4. It also reports arm-level percentages of 9.0% for placebo and 38.0% for mirikizumab. The primary result is therefore interpreted from the sponsor's own posted analysis; a separate recalculation is not reported for this binary endpoint.

For selected continuous secondary outcomes, the posted sponsor analyses use ANCOVA and report least-squares mean differences. Simple recalculations from the posted summary statistics are also available for those outcomes. Those recalculations are unadjusted and use approximate variability derived from the arm-level standard errors, so they are not replacements for the sponsor's model-based estimates or confidence intervals. The reported ANCOVA estimates and the recalculated estimates are consistent in their displayed direction and rounded mean-difference values, while their inferential procedures differ.

There is no flagged discrepancy between sponsor estimates and the recalculations. Where a comparison is reported by the sponsor, its stated analysis remains the primary basis for interpretation. The recalculations are best read as checks based on limited summary information, not as a competing analysis of participant-level data. The registry record does not supply the full ANCOVA specification needed for an independent reproduction of the sponsor's adjusted results.

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when statistical tests are conducted across several outcomes, time points, treatment comparisons, or subgroups. General explanation: without a prespecified strategy for managing multiple tests, the chance of at least one apparently strong result arising by chance can increase. Possible strategies include defining a testing hierarchy or adjusting the threshold for a family of comparisons, but the registry record does not report whether either approach was used for this trial. The primary result is identified, but the record does not establish the full inferential status of the secondary outcomes.

The registry lists multiple secondary endpoints and analyses at different times. It does not report which secondary outcomes, if any, were included in a formal confirmatory sequence, whether results were adjusted for multiplicity, or whether the analysis plan changed. Those details are not reported in the registry record. As a result, secondary p-values and confidence intervals should be read as the posted outcome-specific results, not automatically as multiplicity-controlled evidence.

Missing data are another important issue for outcomes assessed over time. The registry record does not describe the imputation method, the assumptions used for missing values, the handling of discontinuation, or sensitivity analyses for missingness. It also does not report interim analyses. These omissions do not show that such procedures were absent; they mean the registry record does not provide enough detail to evaluate them.

The primary population is described as all randomized participants in the placebo and mirikizumab arms who met the baseline SES-CD criterion and received at least one dose. This is the population definition stated for the primary outcome. The record does not provide further detail about exclusions after randomization, the number with missing primary outcomes, or whether alternative analysis populations were used. The reported arm denominators should therefore be read as the analysis counts attached to the posted result, rather than presumed to represent every randomized participant.

Finally, treatment crossover for placebo non-responders at Week 12 affects interpretation of later follow-up. The record identifies the switch but does not provide enough detail to evaluate how crossover was handled in every secondary analysis. The statistical analysis plan and full missing-data procedures are not reported in the registry record, so conclusions about their influence cannot be drawn from the posted summary alone.

12. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: crohn's disease (6.6%), anal fistula (1.9%), ileus (1.2%), intestinal obstruction (1.2%), intestinal perforation (1.2%).

The registry lists serious adverse events and deaths by the stated groups, with the number at risk shown for each group. These are counts of participants affected, not event rates calculated. The listed groups include placebo, placebo non-responders, mirikizumab, ustekinumab, and mirikizumab in adolescents. The denominators differ across groups, and placebo non-responders are presented separately, so raw counts should not be compared as though each group had the same size or treatment history.

For example, the serious adverse event counts are 37 among 211 at risk for placebo, 7 among 85 for placebo non-responders, 65 among 630 for mirikizumab, 35 among 309 for ustekinumab, and 0 among 6 for the adolescent mirikizumab group. The registry also lists deaths of 1, 1, 0, 1, and 0 in those respective groups. These figures describe the posted safety record; they do not establish the cause of an event or show whether groups differed after accounting for exposure, follow-up, or participant characteristics.

The registry lists Crohn's disease as the most frequent serious-event term at 6.6%, followed by anal fistula at 1.9%; ileus, intestinal obstruction, intestinal perforation, and procedural intestinal perforation are each listed at 1.2%. The record excerpt does not provide enough context to assign these term percentages to a particular comparison or infer a causal relationship. Safety interpretation requires the complete event definitions, exposure and follow-up context, and clinical assessment; the posted counts alone are descriptive.

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

The primary result means that, in the registered analysis population, the percentage meeting the combined clinical and endoscopic response definition was higher in the mirikizumab arm than in the placebo arm. The estimated absolute difference was 28.7 percentage points, and the 95% confidence interval ranged from 23.0 to 34.4. That is the statistical contrast reported for this endpoint and this analysis; it is not a measure of every dimension of Crohn's disease or a prediction for an individual participant.

The finding does not mean that all participants assigned to mirikizumab achieved the endpoint, or that participants who did not meet it experienced no change. It also does not mean that the endpoint captures complete remission: the definition requires a specified clinical response at Week 12 and endoscopic response at Week 52, not the separate endpoint of endoscopic remission. A percentage meeting a composite criterion cannot reveal the full distribution of symptom and endoscopic changes without component-level results.

The comparison also does not answer every question about longer-term treatment because placebo non-responders switched to mirikizumab at Week 12. Nor does the primary placebo comparison establish the result of a direct comparison with ustekinumab. The study included an ustekinumab arm, but the reported primary estimate is limited to placebo and mirikizumab.

General statistical explanation: randomization is intended to make treatment groups comparable on average, but interpretation still depends on how the analysis population was defined, how missing data were handled, and whether the reported analysis followed the planned procedures. Details of missing-data methods and the full statistical plan are not reported in the registry record. The confidence interval reflects uncertainty under the reported analysis, not every uncertainty about the study or how broadly the finding applies.

The result is evidence about the registered outcome in the trial population, not medical advice and not a determination of which treatment is suitable for any person. It should be considered alongside the other trial outcomes, safety information, trial conduct, and limitations in the available record.

14. Why This Trial Matters Statistically

This trial is statistically notable for pairing a symptom-based clinical response with an endoscopic response at a later time. That construction asks whether participants met a defined clinical criterion and a defined endoscopic criterion, rather than relying on either domain alone. The result is consequently tied to a specific composite definition and follow-up schedule; it should not be generalized to a broader notion of improvement without additional evidence.

The analysis also illustrates why the absolute effect estimate and its confidence interval matter together. The risk difference gives the contrast in percentage points, while the interval communicates the uncertainty around that estimate under the reported method. Readers can see both the magnitude and the precision of the posted comparison without reducing the conclusion to a significance label.

The trial's treatment structure matters for interpretation. It includes placebo, an active comparator, and an adolescent mirikizumab arm, while the primary result addresses the adult placebo-versus-mirikizumab comparison. In addition, placebo non-responders switched treatment at Week 12. A statistical reading that ignores the analysis population, treatment history, or outcome timing would miss important parts of what the estimate represents.

The record also demonstrates the distinction between model-based sponsor analyses and simple summary-statistic recalculations. The former use ANCOVA for selected continuous outcomes, while the recalculations are unadjusted illustrations based on posted summaries. Keeping those methods distinct avoids treating a rough reconstruction as if it were the original adjusted analysis. Finally, multiple secondary outcomes and incomplete reporting of multiplicity and missing-data procedures limit how strongly those secondary findings can be interpreted beyond their descriptive pattern.

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 a full protocol or statistical analysis plan, including the complete specifications for the primary Cochran-Mantel-Haenszel analysis and the secondary ANCOVA models. The record also does not describe the rationale for the planned enrollment or provide a sample-size justification. These omissions restrict independent evaluation of the analysis choices and their prespecification.

Important details about missing data are not reported in the registry record. The posted summary does not specify how unavailable outcomes, treatment discontinuation, or other post-randomization events were handled, nor does it provide sensitivity analyses for those issues. The record also does not report interim analyses. These are limits of what can be assessed from the registry entry, not proof that the study lacked procedures for these matters.

The primary analysis applies to adults meeting the stated baseline endoscopic criterion who received at least one dose. The result should not automatically be generalized to all enrolled participants, to participants outside those criteria, or to the adolescent arm. The primary reported comparison is placebo versus mirikizumab, despite the presence of an ustekinumab arm; the primary estimate does not provide a direct comparison with that active comparator.

Placebo non-responders switched to mirikizumab at Week 12, complicating interpretation of later comparisons that retain the placebo label. The registry record does not provide enough detail to assess crossover handling across every outcome. The primary composite also combines outcomes at different times, so it does not by itself disclose the separate component results or establish remission.

Finally, multiple secondary endpoints are posted, but the record does not report a multiplicity strategy. The selected recalculations for continuous outcomes rely on summary statistics and approximate variability; they cannot reproduce the sponsor's adjusted ANCOVA analyses. These limitations mean that secondary findings and recalculated results should be treated as contextual rather than as substitutes for a complete participant-level analysis.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Mirikizumab (LY3074828) in Participants With Crohn's Disease — NCT03926130.
• Registry record last updated 2024-12-27; status last verified 2024-12.
• Results first posted 2024-12-27.
• 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

The registry describes a completed, randomized, triple-masked, parallel phase 3 study of mirikizumab in participants with moderately to severely active Crohn's disease. Total enrollment was 1158. The primary reported comparison was mirikizumab versus placebo in adults meeting the stated baseline endoscopic eligibility criteria and receiving at least one dose.

The primary endpoint combined clinical response at Week 12 with endoscopic response at Week 52. The reported response percentages were 38.0% for mirikizumab and 9.0% for placebo; the risk difference was 28.7 percentage points, with a 95% confidence interval from 23.0 to 34.4. The trial also included ustekinumab and an adolescent mirikizumab group, and placebo non-responders switched to mirikizumab at Week 12.

Secondary outcomes address additional clinical, symptom, endoscopic, and laboratory measures. The registry record does not report complete details of the statistical plan, multiplicity handling, or missing-data methods. The posted estimates should therefore be interpreted in light of their stated populations, outcome definitions, treatment changes, and reporting limitations.