A Study of Lebrikizumab (LY3650150) in Participants With Moderate-to-Severe Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Lebrikizumab (LY3650150) in Participants With Moderate-to-Severe Atopic Dermatitis (NCT03443024), a randomized phase 2 study. The primary outcome is Percent Change From Baseline in Eczema Area and Severity Index (EASI).

2. Clinical Question

3. Trial Design

This was a phase 2, randomized, parallel-group, placebo-controlled, dose-ranging trial of lebrikizumab in participants with moderate-to-severe atopic dermatitis. The registry record describes the study as completed, with enrollment of 280. Its start date is 2018-01-30 and its primary completion date is 2019-02-07. The official title calls the study double-blind, while the registry masking field describes it as triple-masked.

Participants were assigned to one of three lebrikizumab regimens or placebo: 125 mg every 4 weeks, 250 mg every 4 weeks, 250 mg every 2 weeks, or placebo every 2 weeks. The arm descriptions identify subcutaneous administration and describe loading and placebo injections intended to accompany the different schedules. The available description of the 125 mg regimen ends before the full schedule is stated, so additional dosing details for that arm are not reported in the registry record.

The primary outcome was percent change from baseline in EASI at Week 16. The primary analysis population included randomized participants who received at least one dose and had Week 16 EASI data. The registry describes an analysis of covariance with treatment as a factor and baseline EASI as a covariate, and says missing values were imputed using Markov Chain Monte Carlo multiple imputation.

The registry record does not report the trial’s original sample-size justification or the assumptions used to design the trial. It also does not report interim-analysis details or a full statistical analysis plan in the information available for this analysis. The planning illustrations associated with the recalculated comparisons are separate from the trial’s original design.

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 who received at least one dose of study drug and had Week 16 EASI data.

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.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Eczema Area and Severity Index (EASI), measured over Baseline, Week 16.

The primary endpoint is percent change from baseline in the Eczema Area and Severity Index, or EASI, at Week 16. EASI is a clinician-assessed measure that combines the extent of atopic dermatitis over body regions with the severity of four clinical signs: erythema, edema or papulation, excoriation, and lichenification. Each sign is assessed at the head and neck, trunk, upper limbs, and lower limbs. The registry describes a scoring system in which the final score ranges from 0 to 72, with higher scores indicating more severe disease.

Percent change expresses the Week 16 EASI result relative to the participant’s baseline score. The posted arm means are negative, indicating decreases relative to baseline on average. Comparing those means across randomized groups gives a way to examine whether the average change differed among the treatment regimens and placebo during the trial period.

EASI is used because it combines clinical assessments of both disease extent and signs rather than measuring just one feature. It is not a direct measure of every experience important to participants. The trial also listed secondary outcomes relating to investigator assessment, EASI thresholds, sleep loss, and pruritus, but those outcomes do not replace the defined primary endpoint.

The primary outcome is reported as a least-squares mean from an analysis of covariance that included baseline EASI as a covariate. This adjustment accounts statistically for baseline EASI in estimating treatment-group means. It does not make the outcome a measure of change beyond the Week 16 timeframe stated in the registry.

7. Results Posted in the Registry

• 125 mg Lebrikizumab (Q4W): mean -62.34 versus -41.12 on control; difference -21.22 (95% CI -39.51 to -2.93), standardized effect d = -0.44; z-test p = 0.023.
• 250 mg Lebrikizumab (Q4W): mean -69.21 versus -41.12 on control; difference -28.09 (95% CI -46.31 to -9.87), standardized effect d = -0.59; z-test p = 0.003.
• 250 mg Lebrikizumab (Q2W): mean -72.09 versus -41.12 on control; difference -30.97 (95% CI -49.20 to -12.74), standardized effect d = -0.65; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percentage of Participants With a 75% Improvement From Baseline in EASI (EASI75) at Week 16

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • 125 mg Lebrikizumab (Q4W): 43.8% (32/73) versus 25.0% (13/52) on control; risk difference 18.8 percentage points (95% CI 2.5 to 35.2), risk ratio 1.75 (1.02–3.00); unadjusted two-proportion z-test p = 0.031.
  • 250 mg Lebrikizumab (Q4W): 56.2% (45/80) versus 25.0% (13/52) on control; risk difference 31.2 percentage points (95% CI 15.2 to 47.3), risk ratio 2.25 (1.35–3.74); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for 125 mg Lebrikizumab (Q4W) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for 250 mg Lebrikizumab (Q4W) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With an Investigator Global Assessment (IGA) Score of 0 (Clear) or 1 (Almost Clear) and a Reduction ≥2 Points From Baseline to Week 16 (5-point Scale)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • 125 mg Lebrikizumab (Q4W): 26.0% (19/73) versus 15.4% (8/52) on control; risk difference 10.6 percentage points (95% CI -3.4 to 24.7), risk ratio 1.69 (0.80–3.57); unadjusted two-proportion z-test p = 0.154.
  • 250 mg Lebrikizumab (Q4W): 33.8% (27/80) versus 15.4% (8/52) on control; risk difference 18.4 percentage points (95% CI 4.1 to 32.6), risk ratio 2.19 (1.08–4.45); unadjusted two-proportion z-test p = 0.020.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 125 mg Lebrikizumab (Q4W) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for 250 mg Lebrikizumab (Q4W) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With EASI <7 at Week 16

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • 125 mg Lebrikizumab (Q4W): 42.5% (31/73) versus 28.8% (15/52) on control; risk difference 13.6 percentage points (95% CI -3.1 to 30.4), risk ratio 1.47 (0.89–2.44); unadjusted two-proportion z-test p = 0.120.
  • 250 mg Lebrikizumab (Q4W): 61.3% (49/80) versus 28.8% (15/52) on control; risk difference 32.4 percentage points (95% CI 16.1 to 48.7), risk ratio 2.12 (1.34–3.37); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 125 mg Lebrikizumab (Q4W) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for 250 mg Lebrikizumab (Q4W) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Across the posted secondary outcomes, the arm-level percentages generally follow a pattern in which the lebrikizumab groups have higher values than placebo for the listed Week 16 responder or threshold outcomes. The available secondary results also include sleep-loss and pruritus measures, but a complete set of numerical results for those outcomes is not included in the posted secondary summaries described in the record. The overall pattern is broadly consistent with the direction of the primary EASI comparison: the posted group summaries tend to favor a larger change or a greater proportion meeting the specified endpoint in lebrikizumab groups.

That broad pattern should not be treated as a single, uniform confirmatory finding. Secondary endpoints measure different features, use different scales, and may have different analysis populations or methods. A threshold outcome such as EASI75 does not convey the same information as a continuous mean change in EASI. Nor should percentages across different endpoints be directly compared as though they measured the same quantity.

The registry lists 11 posted secondary outcomes, and the record includes analyses across multiple treatment comparisons. When many outcomes and comparisons are examined, the chance of at least one apparently notable result can increase unless the analysis accounts for multiplicity. The multiplicity approach is not reported in the registry record. The recalculated secondary p-values are unadjusted, and the sponsor’s reported tests are available only for some of the listed outcomes. Accordingly, these findings carry less inferential weight than the prespecified primary result unless the analysis plan establishes how they were intended to support inference.

Secondary outcomes can still add context by showing whether the primary result is accompanied by changes in other assessed dimensions. They should be read as complementary evidence, with attention to endpoint definitions, missing results, and the distinction between sponsor analyses and unadjusted recalculations. The registry record does not justify treating every secondary comparison as independently confirmatory.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Percentage of Participants With a 75% Improvement From Baseline in EASI (EASI75) at Week 16
• Percentage of Participants With an Investigator Global Assessment (IGA) Score of 0 (Clear) or 1 (Almost Clear) and a Reduction ≥2 Points From Baseline to Week 16 (5-point Scale)
• Percentage of Participants With EASI <7 at Week 16
• Percentage of Participants Achieving EASI50 at Week 16
• Percentage of Participants Achieving EASI90 at Week 16
• Percent Change From Baseline in the Sleep Loss Scale Score
• Percent Change From Baseline in Pruritus Numeric Rating Score (NRS)
• Percentage of Participants With Pruritus NRS Change of ≥3 at Week 16

9. Statistical Interpretation

10. Statistical Methods Explained

The primary analysis is described as an analysis of covariance, often abbreviated ANCOVA. In general, ANCOVA estimates group outcomes while adjusting for a covariate; in this trial, the registry says treatment was the factor and baseline EASI was the covariate. The reported least-squares means are model-adjusted group means, rather than simple averages alone. Adjustment can account for baseline score differences in the model, but the registry information does not provide enough detail to reconstruct every aspect of the sponsor’s fitted model.

What does a mean difference represent?

A mean difference compares the average outcome in one group with the average outcome in another. For the recalculated primary comparisons, the reference group is placebo. Because the outcome is percent change from baseline, a negative difference means the lebrikizumab group’s average percent change was more negative than the placebo group’s average. The value describes the estimated contrast on the outcome scale; it is not a percentage-point response rate and should not be interpreted as a ratio.

The posted comparison calculations are labelled as recalculated and are based on the arm means and standard deviations. They are unadjusted summary-statistic comparisons, whereas the registry’s own primary analyses use ANCOVA. The pooled standard deviations used in the recalculations are marked as approximate. Therefore, the recalculated intervals and p-values may not match the sponsor’s results. Since a discrepancy is flagged, the sponsor’s own analysis takes precedence for interpretation of the trial’s planned analysis.

How should the p-values be read?

A p-value describes how compatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The recalculated p-values are explicitly unadjusted. The registry lists sponsor p-values for the primary comparisons, but the full statistical plan, including any multiplicity strategy, is not reported in the registry record. That missing information limits how confidently the p-values can be interpreted across several treatment comparisons and outcomes.

What does the interval add?

A confidence interval gives a range of values compatible with the data and the method used to calculate it, at the stated confidence level when one is reported. For the recalculated primary comparisons, the posted interval endpoints are available, but the recalculated comparison record does not state a confidence level. The sponsor’s posted analysis identifies a 95% interval for two comparisons but does not include the endpoints; it does not identify an interval level for the remaining comparison. Keeping these analyses separate avoids treating approximate re-calculations as though they were the sponsor’s adjusted estimates.

For secondary binary outcomes, the recalculated comparisons use risk differences. In general, a risk difference compares the proportions meeting an outcome in two groups and expresses the contrast on the proportion scale. The posted record also includes relative risks and odds ratios for some recalculations. These are different ways of expressing binary contrasts: a relative risk compares proportions as a ratio, while an odds ratio compares odds. They are not interchangeable with a risk difference, and their interpretation depends on the outcome and baseline frequency. The sponsor’s reported secondary tests use the Cochran-Mantel-Haenszel method for the outcomes with reported analyses.

Why does over-dispersion matter?

In general, over-dispersion means that outcome variability is greater than a statistical model expects. When variability is underestimated, standard errors can be too small and intervals too narrow, making results appear more precise than they are. The primary endpoint is continuous, and the posted arm summaries include standard deviations. The registry’s stated primary model is ANCOVA; it does not report an over-dispersion adjustment. No claim about over-dispersion in this trial can be made from the registry information alone.

The primary population definition also matters. The outcome analysis includes randomized participants who received at least one dose and had Week 16 EASI data, with missing values imputed using MCMC multiple imputation. The registry does not give further details about the imputation model or the assumptions used for missing outcomes. These details affect how one evaluates the analysis, particularly when some outcome data are unavailable. The population definition and stated missing-data method should therefore be considered part of the interpretation, not incidental technical choices.

11. Confidence Intervals and Precision

The recalculated primary mean differences are accompanied by intervals that show the range of estimates compatible with the summary-statistic calculation. For 125 mg lebrikizumab every 4 weeks versus placebo, the estimate is -21.22 and the interval is -39.51 to -2.929. For 250 mg every 4 weeks, the estimate is -28.09 and the interval is -46.31 to -9.873. For 250 mg every 2 weeks, the estimate is -30.97 and the interval is -49.2 to -12.74. Each interval spans values on the same side of zero, but these are recalculated, unadjusted intervals, not the sponsor’s model-based intervals.

In general, an interval’s width reflects both the observed variability and the amount of information available for the comparison. Wider intervals indicate less precision than narrower intervals, all else equal. The posted standard deviations are substantial relative to the group means, and the arm sizes differ. These features contribute to uncertainty in group contrasts. The recalculated calculations use approximate pooled standard deviations, which is an additional reason not to treat their endpoints as exact substitutes for the sponsor’s ANCOVA results.

The registry reports sponsor p-values for the primary comparisons, but it gives interval endpoints for none of those sponsor analyses. It identifies 95% confidence intervals for two of the sponsor comparisons, without reporting their limits, and does not identify a confidence interval level for the other. The sponsor’s adjusted estimate precision therefore cannot be read from the posted summary fields. Statistical significance and precision are related but not identical: a p-value does not show the range of plausible effect sizes.

Confidence intervals do not account for every source of uncertainty. They depend on the chosen model and its assumptions, and they do not address unreported multiplicity procedures, missing-data assumptions, or whether the outcome difference is clinically important. The registry record does not report a clinical-importance threshold for the primary endpoint. Consequently, the interval should be read as uncertainty around the specified statistical contrast, not as a complete measure of certainty about all aspects of the result.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size display gives planning illustrations in participants per arm for the non-time-to-event comparisons in this trial. It does not report total event counts. The calculations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. For each comparison, the assumed effect is a stated fraction of the observed effect. The trial’s original sample-size assumptions are not reported in the registry record; these illustrations are calculations based on the posted observed effects, not a reconstruction of the original design.

The 80% power column gives the number of participants per arm that the planning calculation indicates would be needed to have an 80% chance of detecting the chosen effect under its assumptions. The 90% power column gives the corresponding planning number for a 90% chance. These are planning quantities, not evidence that the completed trial had that probability of detecting an effect. This is not observed power.

Rows that plan for a smaller fraction of the observed effect generally require more participants per arm than rows that plan for the full observed effect. The reason, in general, is that a smaller difference is harder to distinguish from random variation while maintaining the same target power and significance level. The figures depend on the assumptions and on the observed effect used as the starting point. They should not be read as a recommendation for a future trial without a separate design assessment.

The planning display is comparison-specific: the observed effect and variability differ among comparisons, so the resulting participant counts differ as well. It does not incorporate any trial-specific design assumptions that are absent from the registry record, and it does not establish what sample size would be appropriate for a different population, endpoint, analysis, or target effect.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -21.220; pooled standard deviation σ = 47.756.
2. n = 2 × (1.96 + 0.842)² × 47.756² / -21.220² = 79.5, rounded up to 80 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -21.22                      # observed difference in means
sigma <- 47.7560550154025                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3443024)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 80 (80% power), 107 (90% power)
 75% of observed effect: n per arm = 142 (80% power), 190 (90% power)
 50% of observed effect: n per arm = 319 (80% power), 426 (90% power)
power.t.test: 80.5 
Simulated power at n = 80 per arm: 0.798 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparisons as ANCOVA analyses. Their p-values are 0.017 for 125 mg lebrikizumab every 4 weeks versus placebo, 0.002 for 250 mg every 4 weeks versus placebo, and 0.001 for 250 mg every 2 weeks versus placebo. The registry identifies the last two sponsor intervals as 95% confidence intervals, but does not report their endpoints. The interval level for the first sponsor comparison is not reported in the registry record.

The posted recalculations instead compare arm summary statistics without the baseline adjustment described for the sponsor’s ANCOVA. They use approximate pooled standard deviations and produce different unadjusted p-values: 0.023, 0.003, and <0.001 for the corresponding comparisons. The differences between the reported and recalculated results are expected to be possible because the analyses use different information and methods; the recalculation is not a reproduction of the sponsor’s model.

Because a discrepancy is flagged, the sponsor’s own analysis takes precedence. The recalculated values can help readers understand the direction and scale of the posted group summaries, but they should not replace the registry’s reported ANCOVA results. The registry record does not provide all model details or sponsor interval endpoints needed to independently verify the sponsor’s estimates.

15. Multiplicity, Missing Data and Analysis Populations

The study compares three lebrikizumab regimens with placebo on the primary endpoint and includes multiple secondary outcomes. This creates a multiplicity issue: the more statistical comparisons made, the greater the opportunity for one or more results to appear notable by chance. A prespecified strategy may control this problem by defining which tests are confirmatory and how error rates are handled. That strategy is not reported in the registry record.

The sponsor’s primary analyses are reported as ANCOVA comparisons, but the registry information does not describe a complete testing hierarchy, adjustment procedure, or other multiplicity plan. The sponsor-reported p-values therefore should be interpreted with the scope of the available record in mind. The recalculated p-values are explicitly unadjusted and should not be treated as multiplicity-controlled. The posted secondary analyses also include multiple outcomes and comparisons, while the registry reports sponsor tests for only some of them.

The primary analysis population is defined as randomized participants who received at least one dose and had Week 16 EASI data. The registry states that missing primary endpoint values were imputed using Markov Chain Monte Carlo multiple imputation. In general, multiple imputation uses repeated draws to represent uncertainty about missing values, then combines analyses across imputed datasets. The registry does not report the imputation model, variables included, number of imputations, or assumptions about the missingness process. These details are not reported in the registry record.

It is also not reported how missing data were handled for each secondary outcome, or whether sensitivity analyses assessed alternative missing-data assumptions. The registry record does not report interim-analysis details, the original sample-size justification, or a full statistical analysis plan. Those omissions prevent a complete assessment of how the endpoint hierarchy, missing outcomes, and planned testing strategy were handled. They do not invalidate the posted results, but they limit how fully readers can assess the confirmatory strength and robustness of the findings.

16. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: oedema peripheral (1.9%), chronic obstructive pulmonary disease (1.9%), pulmonary embolism (1.9%), hernial eventration (1.4%), periprosthetic fracture (1.4%).

The registry reports serious adverse events by treatment group. Serious adverse events were recorded for 2 participants in the 125 mg lebrikizumab every-4-weeks group, 0 in the 250 mg every-4-weeks group, 2 in the 250 mg every-2-weeks group, and 2 in the placebo group. The at-risk group sizes listed for these summaries are 73, 80, 75, and 52, respectively. The registry reports 0 deaths in each group.

These are descriptive counts over the reported study record. They do not by themselves establish whether a treatment caused an event or whether the groups differ in safety. Serious-event counts can be influenced by the kinds of events observed, follow-up, and the number of participants at risk; the registry record does not give enough detail to assess those factors comprehensively. The posted serious-term list includes oedema peripheral, chronic obstructive pulmonary disease, pulmonary embolism, hernial eventration, periprosthetic fracture, and chest pain, each with the percentage stated in the record. Those terms are not assigned to specific treatment groups in the posted list.

The trial’s serious adverse event figures should be considered separately from the counts for other adverse events and from efficacy comparisons. The registry information is not a complete basis for drawing broad conclusions about safety beyond the recorded events and stated follow-up.

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

The central statistical finding is a difference in Week 16 EASI percent change between the lebrikizumab groups and placebo in the posted results. The group means are more negative for each lebrikizumab regimen than for placebo. The registry’s sponsor analyses report ANCOVA comparisons, while the associated recalculations are unadjusted and approximate. Those analyses answer related but not identical statistical questions, and the sponsor’s own analysis takes precedence where the results differ.

The finding applies to the trial’s studied population, intervention schedules, comparator, outcome definition, and Week 16 assessment. The primary analysis population was restricted to randomized participants who received at least one dose and had Week 16 EASI data, with missing values handled by the stated multiple-imputation method. It does not directly describe outcomes beyond the reported timepoint or in groups not represented by that analysis population.

A statistically notable comparison is not automatically a measure of clinical importance. The registry does not report a threshold for a clinically meaningful difference in the primary endpoint, and the statistical estimates do not independently establish how participants experienced the changes. EASI assesses physician-rated disease extent and clinical signs; it is not a complete account of symptoms, daily functioning, or every other outcome relevant to participants.

The results also do not settle questions that the registry record does not address, including the trial’s original sample-size justification, detailed multiplicity strategy, full missing-data assumptions, or the sponsor’s confidence interval endpoints for the primary comparisons. The safety counts are descriptive and do not establish causality. Overall, the posted data support interpretation of a Week 16 group comparison in this phase 2 dose-ranging trial, not a broader conclusion beyond its design and recorded analyses.

18. Why This Trial Matters Statistically

This trial is statistically informative because it compares multiple dose schedules with a concurrent placebo group under randomized, masked, parallel-group conditions. Randomization is intended to support fair comparisons by reducing systematic differences between assigned groups, while masking can limit some forms of bias in treatment delivery and outcome assessment. The registry record describes the study as triple-masked and the official title as double-blind; the distinction in wording should be noted rather than resolved by assumption.

The primary endpoint is continuous and reported as percent change from baseline, with ANCOVA adjustment for baseline EASI. This makes the analysis more specific than a simple comparison of raw group changes. At the same time, the posted summary-statistic recalculations show why statistical method matters: unadjusted comparisons based on group means and variability need not reproduce a model-adjusted sponsor analysis.

The trial also illustrates the interpretive challenge of dose-ranging studies. Multiple regimens and several secondary outcomes create a broader set of comparisons than a single treatment-versus-control test. Without a reported multiplicity plan, nominal p-values require caution, especially for secondary analyses. The registry’s incomplete reporting of sponsor interval endpoints also constrains assessment of the adjusted estimates’ precision.

Finally, the distinction between observed results and planning illustrations is important. Recalculated sample-size figures use observed effects and specified hypothetical assumptions; they are not the trial’s original design rationale and are not observed power. Recognizing these differences helps separate what the data show from what additional analysis-plan details would be needed for a fuller statistical appraisal.

19. 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.

Several limitations arise from the information reported for this trial. The registry gives the primary endpoint definition, analysis population, ANCOVA description, and missing-data method, but does not provide full details of the fitted model or imputation model. It also does not report the original sample-size justification, interim-analysis details, or a full statistical analysis plan. These omissions limit independent assessment of design assumptions and analysis choices.

The sponsor’s primary analysis record lacks interval endpoints. It identifies a 95% confidence interval for two comparisons but does not give the limits, and does not state the interval level for the remaining comparison. The recalculated intervals are based on summary statistics and approximate pooled standard deviations, rather than the sponsor’s adjusted model. They offer a secondary descriptive check, not an exact substitute.

Multiplicity is a particular concern because the trial includes three treatment comparisons and numerous secondary outcomes. The registry record does not state how multiplicity was handled. Secondary results are therefore difficult to classify as confirmatory from the available information, and not every secondary endpoint has a corresponding sponsor analysis in the posted record.

The primary analysis population requires receipt of at least one dose and Week 16 EASI data, and missing primary outcomes were imputed. The registry does not describe the assumptions behind that imputation or report sensitivity analyses. It also does not provide enough detail in this record to judge whether missingness or departures from the analysis population could affect the findings.

Finally, the study is a phase 2 trial with a Week 16 primary outcome. The reported results are tied to this design and timeframe. The registry record does not report a clinical-importance threshold for the primary measure, and the safety summary is descriptive. These constraints should temper interpretation without implying that unreported information had any particular value.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of Lebrikizumab (LY3650150) in Participants With Moderate-to-Severe Atopic Dermatitis — NCT03443024.
• Registry record last updated 2021-05-04; status last verified 2020-07.
• Results first posted 2021-05-04.
• 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.

23. Record Summary

NCT03443024 is a completed, randomized, parallel-group phase 2 trial of lebrikizumab in participants with moderate-to-severe atopic dermatitis. The study compared three subcutaneous regimens—125 mg every 4 weeks, 250 mg every 4 weeks, and 250 mg every 2 weeks—with placebo. The registry reports enrollment of 280 and identifies the primary outcome as percent change from baseline in EASI at Week 16.

The sponsor’s primary analyses used ANCOVA with treatment and baseline EASI as specified in the registry description. The posted group means show more negative average EASI percent changes in each lebrikizumab group than in placebo. Recalculated unadjusted comparisons are approximate and do not replace the sponsor’s analysis. The registry states that primary missing values were imputed using MCMC multiple imputation, but does not report the full model details or multiplicity plan.

Serious adverse event counts and deaths are reported by group. They are descriptive and do not establish causality. The original sample-size justification and several details needed for fuller assessment are not reported in the registry record.