Safety and Efficacy of Lebrikizumab (LY3650150) in Combination With Topical Corticosteroid in Moderate-to-Severe Atopic Dermatitis.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Lebrikizumab (LY3650150) in Combination With Topical Corticosteroid in Moderate-to-Severe Atopic Dermatitis. (NCT04250337), a randomized phase 3 study. The primary outcome is Percentage of Participants With an Investigator's Global Assessment (IGA) Score of 0 or 1 and a Reduction ≥2-points From Baseline to Week 16..

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-group phase 3 trial. The registry lists quadruple masking; it does not identify which study roles were masked. The study was designed to evaluate lebrikizumab used in combination with topical corticosteroid treatment, compared with placebo used with topical corticosteroid treatment, in participants with moderate-to-severe atopic dermatitis.

The registry reports enrollment of 228 and a study duration of 16 weeks. The listed start date is 2020-02-03, primary completion date is 2021-08-11, and status is completed. These record details describe the trial’s timing and status, but do not establish how many participants completed follow-up or contributed data to every outcome.

In the lebrikizumab group, the stated regimen was 500 mg as two 250 mg subcutaneous injections at Baseline and Week 2, followed by a single 250 mg injection every two weeks from Week 4 until Week 14. The placebo group received two placebo injections at Baseline and Week 2, followed by one placebo injection every two weeks from Week 4 until Week 14. Topical corticosteroid treatment began at Baseline for all participants. The record says topical corticosteroid use could be tapered or stopped, or restarted as needed, based on treatment response.

The primary analysis population is described as all randomized participants, including those who did not take assigned treatment, did not receive the correct treatment, or otherwise did not follow the protocol. However, one investigational site with seventeen participants was excluded because of Good Clinical Practice issues. The registry does not report further details about the site exclusion, the handling of participants’ data after that exclusion, or the original design assumptions for sample size.

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.

Before assignment: A participant is considered to have completed the study if he/she has completed the last scheduled visit: * For participants continuing into Long-term Extension (LTE), upon completion of week 16 visit and rolling into LTE study * For participants not continuing into LTE, when participant had either week 16 or Early Termination (ET) visit, and safety follow up visit (12 weeks after last study drug administration)

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants, even if the participant did not take the assigned treatment, did not receive the correct treatment, or otherwise did not follow the protocol. One investigational site with seventeen participants was excluded from analysis due to Good Clinical Practice (GCP) issues.

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, even if the participant does not take the assigned treatment, does not receive the correct treatment, or otherwise does not follow the protocol.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With an Investigator's Global Assessment (IGA) Score of 0 or 1 and a Reduction ≥2-points From Baseline to Week 16., measured over Baseline to Week 16.

The primary endpoint is the percentage of participants with an Investigator’s Global Assessment (IGA) score of 0 or 1 and a reduction of at least 2 points from Baseline to Week 16. The IGA is an investigator’s overall assessment of atopic dermatitis severity, using a static numeric scale from 0, meaning clear skin, to 4, meaning severe disease. The assessment considers erythema, papulation or induration, oozing or crusting, and lichenification.

This endpoint combines a final severity threshold with improvement from the participant’s Baseline score. Requiring both conditions distinguishes participants who reach a low severity category and also meet the specified reduction criterion. As a percentage endpoint, it communicates the proportion of participants meeting the defined response rather than the average change in a score.

In general, a binary response endpoint can be straightforward to interpret as the share of participants who meet a stated threshold. It also depends on the chosen threshold and assessment time: participants just short of the criteria are counted differently from those who meet them. The registry identifies Week 16 as the assessment time and reports the endpoint as a percentage with a 95% confidence interval. It does not report additional details about endpoint adjudication or the treatment of missing Week 16 assessments.

7. Results Posted in the Registry

• Lebrikizumab + Topical Corticosteroid: 41.4% (60/145) versus 22.7% (15/66) on control; risk difference 18.7 percentage points (95% CI 5.7 to 31.6), risk ratio 1.82 (1.12–2.96); unadjusted two-proportion z-test p = 0.009.

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 Achieving EASI-90 (≥90% Reduction From Baseline in EASI Score) 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:
  • Lebrikizumab + Topical Corticosteroid: 41.4% (60/145) versus 21.2% (14/66) on control; risk difference 20.2 percentage points (95% CI 7.5 to 32.9), risk ratio 1.95 (1.18–3.23); unadjusted two-proportion z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Lebrikizumab + Topical Corticosteroid than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change in Pruritus Numerical Rating Scale (NRS) Score From Baseline to Week 16

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Lebrikizumab + Topical Corticosteroid: mean -50.68 versus -35.47 on control; difference -15.21 (95% CI -30.53 to 0.11), standardized effect d = -0.29; z-test p = 0.052.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Lebrikizumab + Topical Corticosteroid and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants With a Pruritus NRS of ≥4-Points at Baseline Who Achieve a ≥4-Point Reduction From Baseline to 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:
  • Lebrikizumab + Topical Corticosteroid: 50.8% (66/130) versus 31.6% (18/57) on control; risk difference 19.2 percentage points (95% CI 4.4 to 34.0), risk ratio 1.61 (1.06–2.44); unadjusted two-proportion z-test p = 0.015.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Lebrikizumab + Topical Corticosteroid than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary results cover several related but distinct measures: a stringent EASI response threshold, change in pruritus score, and a pruritus improvement criterion among participants who met a specified Baseline score threshold. Across the reported results, the broad pattern is generally aligned with the primary endpoint: the lebrikizumab plus topical corticosteroid group has more participants meeting the posted binary response criteria, and the sponsor’s reported continuous analysis also favors a larger reduction in pruritus score.

These outcomes complement the primary IGA result rather than duplicate it. IGA is an investigator’s global severity assessment; EASI and pruritus measures capture other aspects of disease signs and symptoms. Their convergence can provide context across different measures, but each outcome has its own definition, analysis population, and uncertainty. The secondary findings do not change what the primary endpoint measures.

The registry reports 29 secondary outcomes. The posted information includes outcome-specific estimates and analyses, but the record does not report whether a multiplicity procedure was planned or applied. When many outcomes are examined, the chance of at least one apparently notable result can increase. Unless multiplicity is controlled through a prespecified strategy, unadjusted secondary p-values should carry less inferential weight than a designated primary analysis.

Secondary outcomes may be informative and hypothesis-generating, but they should not be treated as if each were an independently confirmed primary finding. Their interpretation also depends on missing-data handling, model specification, and the number of analyses considered; those details are not fully reported in the registry record. The overall direction of the posted secondary findings is useful context, but it does not establish that all outcomes show the same effect or have equal evidential strength.

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

• Percentage of Participants Achieving EASI-90 (≥90% Reduction From Baseline in EASI Score) at Week 16
• Percent Change in Pruritus Numerical Rating Scale (NRS) Score From Baseline to Week 16
• Percentage of Participants With a Pruritus NRS of ≥4-Points at Baseline Who Achieve a ≥4-Point Reduction From Baseline to Week 16
• Percentage of Participants With a Pruritus NRS of ≥5-points at Baseline Who Achieve a ≥4-point Reduction From Baseline to Week 16
• Percent Change in EASI Score From Baseline at Week 16
• Change From Baseline to Week 16 in Percent Body Surface Area (BSA)
• Percentage of Participants Achieving EASI-90 at Week 4
• Percent Change in Sleep-loss Score From Baseline to Week 16

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports a Cochran-Mantel-Haenszel analysis for the primary comparison, with the result expressed as a risk difference. It also reports group response percentages and 95% confidence intervals. It does not report enough detail to reconstruct every aspect of the sponsor’s statistical plan, including any stratification variables, covariate handling, or missing-data strategy.

What does a risk difference describe?

In general, a risk difference is the difference between the proportions experiencing an event in two groups. For this endpoint, the event is meeting the IGA response definition. A risk difference is expressed in percentage points when the group proportions are expressed as percentages. It focuses on the absolute separation between the groups, rather than how many times as common the response is in one group relative to the other.

The primary registry analysis reports a risk difference of 18.3, with a 95% confidence interval from 5.1 to 31.5 and p = 0.011. The confidence interval describes a range of values compatible with the analysis and its assumptions; it is not a range containing a fixed proportion of individual participants. The p-value assesses how compatible the observed data are with a null comparison under the stated analysis. Neither quantity measures the probability that a treatment works.

Why use the Cochran-Mantel-Haenszel method?

In general, the Cochran-Mantel-Haenszel method can estimate an association or treatment contrast across strata, allowing a combined comparison that accounts for the strata used in the analysis. The registry names this method for the primary risk-difference analysis, but does not report the strata or enough further detail to describe how they were formed. It would therefore be inappropriate to infer which participant characteristics, if any, were incorporated.

The posted group percentages and the method-based comparison are related but are not interchangeable descriptions. The group percentages describe each arm. The risk difference and its confidence interval summarize the sponsor’s between-group analysis. A simple unadjusted recalculation from summary statistics can be useful as a check, but it does not reproduce an analysis that may account for features not represented in those summaries.

How should the other statistical approaches be understood?

The registry reports an ANCOVA analysis for the continuous secondary outcome measuring percent change in pruritus Numerical Rating Scale score. In general, analysis of covariance compares groups while adjusting for specified covariates. The registry reports a least-squares mean difference, confidence interval, and p-value for that outcome. The covariates and other model details are not reported in the registry record.

For the binary secondary outcomes, the registry reports Cochran-Mantel-Haenszel risk differences. Those outcomes include reaching the stated EASI-90 threshold and meeting a specified pruritus improvement criterion among participants meeting the Baseline eligibility threshold. They are related to the primary result in that they also describe response proportions or improvement, but they measure different features and use different definitions.

What does an unadjusted recalculation add?

In general, recalculating from posted group summaries can provide a transparent, simple comparison using the displayed counts or means. It is not a substitute for a sponsor analysis that uses a specified model. For the primary endpoint, the unadjusted p-value is 0.009, while the reported primary p-value is 0.011. The record marks no discrepancy, and the sponsor-reported method remains the principal reported analysis.

For continuous outcomes, a summary-based calculation may use an approximate standard deviation and may not reproduce the sponsor’s covariate-adjusted estimate. For binary outcomes, a simple comparison may likewise differ from a stratified analysis. These calculations are best treated as checks on the broad pattern, not as a replacement for the registry’s reported inferential method.

What is not established by the analysis labels?

Randomization and masking describe important features of trial conduct, but they do not fill in unreported analytical details. The record does not report the original sample-size justification, planned interim analyses, detailed statistical analysis plan, multiplicity adjustment, or missing-data procedures. Those omissions limit how fully the inferential framework can be assessed from the registry record alone.

In general, statistical significance and clinical importance are distinct questions. A confidence interval and p-value characterize uncertainty under an analysis; they do not define what magnitude is important to participants or clinicians. No threshold for a clinically important difference is reported in the registry record.

11. Confidence Intervals and Precision

The primary arm estimates are 22.1% for placebo plus topical corticosteroid and 41.2% for lebrikizumab plus topical corticosteroid. Their reported 95% confidence intervals are 11.6 to 32.7 and 33.0 to 49.4, respectively. The intervals show that the point estimates are not exact values; plausible values for each group’s response percentage remain uncertain.

For the between-group comparison, the sponsor reports a risk difference of 18.3 percentage points, with a 95% confidence interval from 5.1 to 31.5. This interval is compatible with a positive difference under the reported analysis, while allowing a range of possible magnitudes. Its width is relevant to precision: the result does not pinpoint a single exact difference. The interval should be interpreted as uncertainty about the group-level contrast, not as a prediction interval for individual participants.

In general, a confidence interval is calculated by a statistical procedure that would cover the target parameter at its stated rate over repeated use under its assumptions. It does not mean there is a 95% probability that the fixed parameter lies inside this particular interval. Interpretation also depends on whether the analysis model and assumptions are suitable.

The arm summaries list 66 participants in the placebo group and 145 in the lebrikizumab group for the primary outcome. The registry reports enrollment of 228 and says one site with seventeen participants was excluded from analysis. The available record does not explain the full path from enrollment to each analysis denominator. The difference in group sizes and the site exclusion are relevant context when considering precision and representativeness, but the registry does not provide enough detail to quantify their effect on the estimate.

Confidence intervals for individual arms and for the difference answer different questions. The primary between-group interval is the direct uncertainty summary for the treatment contrast. Overlap or non-overlap of arm-specific intervals should not be used in place of the reported comparison analysis.

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 planning figures are expressed as participants per arm, not as total events. The planning illustration uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers an effect equal to 100%, 75%, or 50% of the observed effect. For the primary endpoint, the figures are 98 participants per arm for 80% power and 130 participants per arm for 90% power when planning for 100% of the observed effect; 167 and 223 when planning for 75%; and 358 and 479 when planning for 50%.

In plain terms, the 80% power column gives the estimated number of participants per arm for a design with an 80% chance of detecting the specified effect under the planning assumptions, if that effect and those assumptions held. The 90% power column gives the corresponding planning target for a 90% chance. These are prospective planning illustrations, not observed power for this completed trial.

The figures are based on the observed effect, with smaller fractions representing smaller effects to detect. Smaller effects generally require more participants to distinguish them from chance variation under the same planning framework. The registry’s original design assumptions and sample-size justification are not reported in the registry record. These illustrations should not be read as evidence that the trial was originally planned using these exact targets, or as a reconstruction of its design process.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.227; treatment proportion p₁ = 0.414; average p̄ = 0.321.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6600 = 1.2936.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6467 = 0.5443.
4. n = (1.2936 + 0.5443)² / (0.1865)² = 97.1, rounded up to 98 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
p0 <- 0.227272727272727                          # control proportion
p1 <- 0.413793103448276                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4250337)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 98 (80% power), 130 (90% power)
 75% of observed effect: n per arm = 167 (80% power), 223 (90% power)
 50% of observed effect: n per arm = 358 (80% power), 479 (90% power)
power.prop.test: 97.1 
Simulated power at n = 98 per arm: 0.805 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 sponsor’s reported primary analysis is a Cochran-Mantel-Haenszel comparison. It gives a risk difference of 18.3 percentage points, a 95% confidence interval from 5.1 to 31.5, and p = 0.011. This is the primary inferential result to use when describing the sponsor’s analysis.

A simple unadjusted recalculation from the posted primary summaries gives p = 0.009. The recalculation is based on the posted group summaries and is not the same analysis method as the sponsor’s reported comparison. The registry does not mark a discrepancy. The small difference between the p-values should not be interpreted as evidence of a contradiction; the two analyses need not be identical in method or assumptions.

For the reported continuous secondary endpoint, the sponsor’s ANCOVA analysis gives a least-squares mean difference of -15.21, with a 95% confidence interval from -27.7 to -2.7 and p = 0.017. The summary-based recalculation gives a different interval and p-value. ANCOVA can adjust for covariates, while the recalculation is based on posted summary statistics, so they do not necessarily estimate uncertainty in the same way. The registry does not give enough model detail to reproduce the sponsor’s analysis.

In general, when the registry reports a sponsor analysis, a simple recalculation is a secondary check rather than a replacement. It can help readers understand what the posted summaries imply under a simpler approach, but it cannot establish that the sponsor’s model was correctly implemented or explain all differences between methods.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary endpoint and lists 29 secondary outcomes. It reports 31 outcome measures and 31 analyses. However, a multiplicity plan—such as a hierarchy among outcomes or an adjustment to control the overall chance of false-positive findings—is not reported in the registry record. Without that information, the secondary p-values should be interpreted cautiously, particularly when considering the collection of outcomes rather than any single result.

The registry describes the primary population as all randomized participants, including those who did not take assigned treatment, did not receive the correct treatment, or otherwise did not follow the protocol. It also states that one investigational site with seventeen participants was excluded from analysis because of Good Clinical Practice issues. The record does not report the rationale in further detail, how the exclusion was applied across outcomes, or whether any sensitivity analyses assessed its impact.

Missing-data methods for the primary and secondary outcomes are not reported in the registry record. The record does not specify how participants without a Week 16 assessment were handled, whether values were imputed, or whether the analysis used another missing-data approach. It also does not provide a complete participant-flow account linking enrollment, randomization, site exclusion, and each outcome-specific denominator.

In general, missing-data methods matter because results can depend on which participants contribute observed outcomes and on assumptions about unobserved values. An analysis described as including all randomized participants does not, by itself, explain how missing outcome measurements were addressed. The available population description and arm counts should therefore be reported as stated, without assuming a particular imputation method or a complete intention-to-treat analysis for every outcome.

The registry does not report a detailed statistical analysis plan, planned interim analyses, or the original sample-size justification. These omissions limit assessment of whether all analyses followed prespecified decisions and how the study’s error rates were managed.

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: dehydration (1.3%), acute kidney injury (1.3%), sinus node dysfunction (0.7%), fall (0.7%).

The registry lists serious adverse events for 1 of 75 participants at risk in the placebo plus topical corticosteroid group and 2 of 153 participants at risk in the lebrikizumab plus topical corticosteroid group. It reports zero deaths in either group. These counts describe the serious adverse event information posted for the trial; they do not by themselves establish whether an event was caused by study treatment.

The registry also lists 26 participants affected in the placebo group and 66 in the lebrikizumab group under the other adverse-event count, with 75 and 153 participants at risk, respectively. The record lists dehydration and acute kidney injury at 1.3% each, and sinus node dysfunction and fall at 0.7% each among the top serious terms. It does not identify the treatment-group attribution for each listed term in the information.

In general, adverse-event counts should be read in relation to the number at risk, follow-up, and event definitions. The registry record does not provide enough detail to assess exposure-adjusted rates, duration, severity beyond the serious-event classification, or causal attribution. The posted counts are a descriptive safety summary, not a comparative safety conclusion.

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

The primary result describes the proportion of analyzed participants in each group who met a specified IGA response definition at Week 16. The sponsor’s reported comparison estimates a positive risk difference between the lebrikizumab plus topical corticosteroid group and the placebo plus topical corticosteroid group. Its confidence interval and p-value characterize uncertainty under the reported analysis.

The result does not establish that every participant experienced improvement, nor does it describe the size or timing of response for each individual. A binary endpoint classifies participants according to whether they cross the stated threshold. It does not show the full distribution of IGA changes or explain what happened to participants who did not meet the response definition.

The comparison concerns lebrikizumab added to topical corticosteroid treatment versus placebo added to topical corticosteroid treatment. Because topical corticosteroid treatment was part of both groups, the contrast is not a comparison of lebrikizumab alone with no treatment. The registry also states that topical corticosteroid use could be adjusted based on response. The result should be understood within that treatment context.

The primary assessment was at Week 16. The posted result does not establish persistence beyond that time, effects on outcomes not reported, or results in populations not represented by this trial. The registry record does not report a clinically important difference threshold, so statistical evidence should not be equated with a judgment about clinical importance.

Finally, the record reports an enrollment of 228 but outcome-specific analysis counts that vary, and it describes exclusion of one site with seventeen participants from the analysis. Missing-data handling and the full participant flow are not reported. These details constrain how fully the result can be evaluated from the registry entry alone. The findings are best described as the posted Week 16 comparison under the stated trial design and analysis, without extending them beyond that evidence.

18. Why This Trial Matters Statistically

This trial illustrates how a clinical endpoint becomes a statistical comparison. The primary outcome is not simply an average score; it is the percentage meeting both a final IGA threshold and a specified reduction from Baseline. That definition determines who counts as a responder and frames the meaning of the estimated group difference.

The trial also demonstrates why the analysis method matters. The sponsor reports a Cochran-Mantel-Haenszel risk difference, while a simple recalculation from summary statistics gives a separate unadjusted result. These analyses can be informative in different ways, but the recalculation does not recreate a method that may account for design or analysis features not represented in the summaries. Clear separation of the sponsor analysis and an unadjusted check helps avoid treating them as interchangeable.

Precision matters alongside the point estimate. The reported confidence interval for the primary risk difference spans a range of possible values, so the estimate should not be presented as an exact effect. The interval and p-value provide complementary information about uncertainty and compatibility with a null difference under the analysis assumptions.

The secondary outcomes broaden the statistical picture to other measures of disease signs and symptoms, but their interpretation depends on multiplicity, missing-data handling, and analysis details. Several of those details are not reported in the registry record. This makes the distinction between a designated primary result and a broader set of secondary findings especially important.

Finally, enrollment, analysis denominators, and the site exclusion are separate pieces of information. A reader should not assume that all enrolled participants contributed to every endpoint. Attending to those distinctions is central to evaluating how strongly the posted results support conclusions about the trial population and its defined outcomes.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record leaves several trial-specific uncertainties. It reports a 16-week study and a Week 16 primary endpoint, but does not report follow-up beyond that assessment. The primary result therefore cannot describe durability after the stated endpoint time.

One investigational site with seventeen participants was excluded from analysis because of Good Clinical Practice issues. The record does not explain the nature of those issues, the process for deciding on exclusion, or whether the exclusion was applied identically across all outcomes. This matters because the primary analysis description otherwise refers to all randomized participants.

The registry reports enrollment of 228, while the primary arm summaries include 66 and 145 participants. Outcome-specific denominators also differ for secondary measures. The record does not give a complete participant-flow account or explain all changes in denominators. Missing-data handling is not reported, so readers cannot assess from this record how incomplete Week 16 measurements affected estimates.

The sponsor’s statistical methods are only partly described. The primary analysis is named as Cochran-Mantel-Haenszel, but the strata and further model details are not reported. The continuous secondary analysis is described as ANCOVA, but the covariates and implementation details are not reported. Multiplicity procedures, interim analyses, and the original sample-size justification are also not reported in the registry record.

Topical corticosteroid treatment was given in both groups and could be adjusted according to response. The observed contrast is therefore specific to the combination-treatment comparison and cannot isolate the contribution of topical corticosteroid treatment. The registry’s serious adverse event summaries are descriptive, and the available record does not provide enough detail for causal attribution or exposure-adjusted safety comparisons.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Safety and Efficacy of Lebrikizumab (LY3650150) in Combination With Topical Corticosteroid in Moderate-to-Severe Atopic Dermatitis. — NCT04250337.
• Registry record last updated 2022-05-09; status last verified 2022-05.
• Results first posted 2022-05-02.
• 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

NCT04250337 is a completed, randomized, double-blind, placebo-controlled, parallel-group phase 3 trial of lebrikizumab with topical corticosteroid treatment versus placebo with topical corticosteroid treatment for moderate-to-severe atopic dermatitis. The registry reports enrollment of 228 and a study duration of 16 weeks.

The primary endpoint was the percentage of participants with an IGA score of 0 or 1 and a reduction of at least 2 points from Baseline to Week 16. The sponsor-reported Cochran-Mantel-Haenszel risk difference was 18.3 percentage points, with a 95% confidence interval from 5.1 to 31.5 and p = 0.011. One investigational site with seventeen participants was excluded from analysis because of Good Clinical Practice issues.

The record lists 29 secondary outcomes and reports serious adverse event counts and no deaths in either group. Details about missing-data handling, multiplicity, the original sample-size justification, and several aspects of the statistical plan are not reported in the registry record.