Long-term Safety and Efficacy Study of Lebrikizumab (LY3650150) in Participants With Moderate-to-Severe Atopic Dermatitis (ADjoin): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Long-term Safety and Efficacy Study of Lebrikizumab (LY3650150) in Participants With Moderate-to-Severe Atopic Dermatitis (ADjoin) (NCT04392154), a non-randomized phase 3 study. The primary outcome is Primary Treatment Period: Percentage of Participants Discontinued From Study Treatment Due to Adverse Events.

2. Clinical Question

In general, a long-term study can describe outcomes accumulated during extended treatment, while interpretation depends on who entered each treatment group, how long participants were observed, and how outcomes were handled. The record identifies the endpoint and its population but does not give the primary endpoint result, a statistical comparison, or details of the analysis method. The clinical question can therefore be described from the registered objectives, but the posted information does not answer it with an outcome estimate.

3. Trial Design

ADjoin is a Phase 3, nonrandomized, double-masked, parallel-group study of lebrikizumab in participants with moderate-to-severe atopic dermatitis. Eli Lilly and Company is listed as the sponsor. The registry records an enrollment of 1153 and a completed status. The recorded study start is 2020-06-15, and the primary completion date is 2024-07-19.

The brief summary describes a study lasting up to 33 months. The primary treatment period is described as lasting up to 100 weeks, and the open-label extension addendum regimens are described as lasting up to 32 weeks. These are maximum durations in the descriptions; actual time on treatment or follow-up for individual participants is not reported in the registry record.

Four treatment-group descriptions appear in the record. Two are labeled as primary-treatment lebrikizumab Q2W and Q4W arms, and two are open-label extension addendum arms labeled Q4W and Q8W. The primary-treatment dosing text does not match the frequency in each arm label: the Q2W-labeled arm is described as 250 mg every 4 weeks, and the Q4W-labeled arm as 250 mg every 2 weeks. That inconsistency should be kept in mind when describing the regimens. The extension descriptions list 250 mg every 4 weeks or every 8 weeks, respectively.

Nonrandomized allocation means the record does not describe treatment assignment by random allocation. The posted information does not state how participants entered each group, whether treatment histories differed, or what criteria governed the extension addendum. Double masking is listed, but details of who was masked and how masking was maintained are not reported in the registry record. The available design information supports a description of the trial structure, not a causal comparison between its treatment groups.

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.

Primary Treatment Period (100 Weeks)

Reasons for not completing this period:

Open-Label Extension Addendum (32 Weeks)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants who 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: All participants who received at least one dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Primary Treatment Period: Percentage of Participants Discontinued From Study Treatment Due to Adverse Events, measured over Baseline to Week 100.

The primary endpoint is the percentage of participants who discontinued study treatment because of adverse events, over the period from baseline to Week 100. The stated analysis population is all participants who received at least one dose of study drug. In practical terms, the endpoint concerns treatment discontinuation attributed to adverse events, not every adverse event and not every reason for stopping treatment.

In general, treatment discontinuation can be a useful safety-related outcome because it captures an event with a direct consequence for continued treatment. It is distinct from counting all adverse events: a participant may experience an adverse event without discontinuing, and discontinuation may depend on clinical decisions and participant circumstances as well as the event itself. The endpoint’s interpretation therefore depends on the registry’s definitions and attribution procedures, which are not reported in the registry record.

The record names a percentage endpoint but does not post the observed percentage, the number of discontinuations, a confidence interval, or a statistical comparison. It also does not describe how participants who left the study before Week 100 were handled in the calculation. The endpoint definition identifies what was intended to be measured, but the available record does not provide an estimate of its frequency or precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies the study structure and outcome definitions, but it does not report the statistical analysis plan or any analyses for the outcomes. The explanations below describe general statistical concepts that help frame these measures; they are not claims about methods used in this trial unless a method is explicitly stated in the record.

What does a percentage endpoint represent?

In general, a percentage is a proportion expressed on a percentage scale: the number of participants meeting a specified outcome is considered in relation to a defined population. For this primary endpoint, the outcome is discontinuation from study treatment due to adverse events, and the stated population is all participants who received at least one dose. The record does not report the observed numerator or percentage. It also does not explain how the calculation treats incomplete follow-up, participants who discontinue for another reason, or participants whose adverse events are not judged to have caused treatment discontinuation.

A percentage is easier to interpret when the numerator, denominator, outcome definition, and observation period are clear. The record gives the endpoint wording, treated population, and timeframe, but not all operational details. In particular, the stated period extends to Week 100, while individual follow-up may vary; the method for accounting for that variation is not reported in the registry record.

Why does a confidence interval matter?

In general, a confidence interval describes the uncertainty around an estimated quantity under a specified statistical method. A narrow interval indicates greater precision than a wide interval, while neither width nor location alone establishes clinical importance. For a percentage endpoint, an interval would help readers judge how precisely the observed frequency was estimated. No confidence interval for the primary endpoint is reported in the registry record, and no interval method is stated.

A point estimate without an interval can give an incomplete picture because the same estimate may be compatible with different degrees of uncertainty. Conversely, a confidence interval is not a guarantee that the true value lies within its limits. Its interpretation depends on the method and assumptions used. Because neither an endpoint estimate nor an interval is posted for the primary outcome, precision cannot be assessed from the available primary-endpoint information.

What can group counts say?

Counts of participants affected by an event describe how many people in each listed group were recorded with that event category. The safety entries give counts for serious adverse events, other adverse events, and deaths, along with the number at risk for each group. These are descriptive figures. They do not supply a randomized contrast, and a difference in counts between groups cannot be interpreted as a treatment effect without considering group sizes, follow-up, participant characteristics, and the analysis method.

In general, event-count summaries may also depend on whether the unit is a participant or an event. The posted safety entries are phrased as participants affected, but further definitions of event collection, coding, and attribution are not reported in the registry record. The record also does not clarify whether the same participant could appear in more than one adverse-event category. The listed death counts are descriptive and do not establish cause.

What does nonrandomized allocation imply?

In general, random allocation is used to reduce systematic differences between treatment groups at the point of assignment. In a nonrandomized study, group membership may be associated with participant characteristics, prior treatment, timing, or other factors. These possibilities can complicate a direct comparison because outcomes may reflect both treatment and differences in who received it. The registry record labels allocation as nonrandomized and explicitly indicates that the arms are not a like-for-like control comparison. It does not report adjustment methods or other procedures intended to address group differences.

The double-masked designation is part of the registered design, but masking does not make allocation random and does not remove all sources of confounding. The record does not describe masking roles, its implementation, or whether it continued across the open-label extension. No assumptions about these details should be used to fill in the missing description.

What do the planning assumptions mean?

The registry’s sample-size assumptions describe a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is defined as a stated fraction of the observed effect. In general, such calculations estimate how many participants would be needed under those assumptions to detect a chosen effect with a chosen power. They are planning illustrations, not observed trial results and not evidence that the study had a particular probability of finding an effect.

The record does not give numerical planning estimates or the effect fractions in the information available for this analysis. It also does not report the trial’s own sample-size justification. The illustration should not be confused with the enrolled sample or an observed-power calculation. In general, observed power is not a substitute for reporting an effect estimate and its uncertainty.

The registry lists zero analyses. That means no analysis results are listed in the record; it does not establish that investigators conducted no analyses outside the registry. The statistical methods used for the primary and secondary outcomes, including any model choice, covariate adjustment, or handling of repeated observations, are not reported in the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Primary Treatment Period: Lebrikizumab Q4W", "Primary Treatment Period: Lebrikizumab Q2W")
n      <- c(141, 1012)
events <- c(3, 46)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Primary Treatment Period: Lebrikizumab Q4W  141  3  2.1  0.4  6.1  0.7  6.1  0.8  6.0
ARM  Primary Treatment Period: Lebrikizumab Q2W  1012  46  4.5  3.3  6.0  3.4  6.0  3.4  6.0

11. Multiplicity, Missing Data and Analysis Populations

The registry record names one primary endpoint and two secondary efficacy outcomes. The primary endpoint is discontinuation from study treatment due to adverse events through Week 100. The secondary outcomes are the percentage with IGA 0 or 1 at Week 100 and the percentage achieving EASI-75 at Week 100. No analysis results are listed for these outcomes, and the record does not describe a plan for controlling multiplicity across endpoints or treatment groups.

In general, testing several outcomes can increase the chance of obtaining at least one apparently notable result by chance if each is assessed separately. A prespecified multiplicity strategy can define how to interpret a family of tests. The registry record does not report such a strategy, so it is not possible to determine whether any adjustment or testing hierarchy was planned. This is a reporting limitation, not evidence that no strategy was used outside the registry record.

The primary endpoint population is stated as all participants who received at least one dose. The record does not explain how missing endpoint information, early withdrawal, or incomplete observation through Week 100 was handled. It does not report imputation methods, censoring rules, sensitivity analyses, or assumptions about missing outcomes. Those details matter because a percentage can depend on how participants without a complete observation are treated.

Other analysis populations, such as a per-protocol or intention-to-treat population, are not reported in the registry record. The nonrandomized allocation also means the record would need to describe the analysis population and any adjustment approach to support a comparative interpretation. No such analysis details or results are listed. These gaps prevent assessment of multiplicity control, missing-data sensitivity, and the relationship between the stated endpoint population and any analysis actually performed.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: food allergy (0.7%), cellulitis (0.7%), accident (0.7%), hip fracture (0.7%), multiple injuries (0.7%).

The serious adverse-event entries are participant counts in four treatment groups, paired with the number at risk. In the primary-treatment Q4W-labeled group, 5 of 141 participants at risk were affected by serious adverse events; in the Q2W-labeled group, 45 of 1012 were affected. In the open-label extension addendum Q4W group, 0 of 52 were affected, and in the Q8W group, 0 of 51 were affected. These are the posted counts, not rates calculated for comparison.

The record also lists participants affected by other adverse events: 89 of 141 in the primary-treatment Q4W-labeled group, 638 of 1012 in the Q2W-labeled group, 17 of 52 in the extension Q4W group, and 13 of 51 in the extension Q8W group. Death counts are 0, 2, 0, and 0 in those groups, respectively. The record does not establish that deaths were caused by treatment.

The listed serious-event terms are food allergy, cellulitis, accident, hip fracture, multiple injuries, and depression, each reported at 0.7%. The denominator and group attribution for these terms are not reported. Because the study is nonrandomized and the groups differ in size and treatment context, these counts should be read descriptively rather than as direct evidence of comparative risk.

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

The registry record establishes that ADjoin was a completed Phase 3, nonrandomized, double-masked, parallel-group study of lebrikizumab in moderate-to-severe atopic dermatitis. It records an enrollment of 1153 and describes primary-treatment regimens and open-label extension regimens. It also identifies a primary safety-related endpoint and two secondary efficacy outcomes.

What the posted record does not establish is the observed result for the primary endpoint. It does not provide the percentage of participants who discontinued treatment because of adverse events, an uncertainty interval, or a statistical comparison. It also does not report results for the Week 100 IGA 0 or 1 and EASI-75 outcomes. Therefore, the record cannot be used by itself to quantify the primary endpoint or evaluate those secondary outcomes.

The adverse-event counts offer limited descriptive information. They show the numbers of participants affected in the listed groups, along with deaths and a list of serious-event terms. Those counts do not show whether an event was caused by treatment, and the nonrandomized design does not support treating the groups as equivalent controls. The uncertainty and comparability of these safety summaries cannot be fully assessed from the posted information.

In general, a study objective is not the same as an outcome result, and a treatment-group description is not the same as evidence of a comparative effect. Interpretation requires outcome estimates, definitions, follow-up information, and statistical methods. Several of those elements are not reported in the registry record. The responsible conclusion is limited to describing the registered design and the posted counts, without asserting that lebrikizumab does or does not change safety or efficacy outcomes.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of the distinction between a registered question and a reportable result. The primary endpoint is clearly framed as a percentage of treated participants discontinuing treatment due to adverse events by Week 100, but the registry record does not include an estimate or precision measure. Without those, readers cannot judge the magnitude or uncertainty of the endpoint from the registry entry.

The design also matters. The study is nonrandomized, so its treatment groups are not like-for-like controls. The registry’s own description cautions against treating the arms as a controlled comparison. This is important because a numerical difference between groups, even if reported, would need interpretation in light of how participants entered each group and whether their characteristics and observation periods differed. The record does not report those details or adjustment methods.

The listed safety counts still have descriptive value: they show the number of participants affected within the groups recorded. But descriptive counts answer a narrower question than a comparative analysis. Group sizes differ, the extension groups belong to a distinct treatment context, and the record does not give sufficient detail to establish comparable follow-up or event attribution.

The entry also illustrates why statistical interpretation depends on complete reporting. Endpoint estimates, confidence intervals, missing-data methods, multiplicity procedures, and analysis populations help distinguish an observed pattern from an uncertain or potentially biased one. Their absence from the registry record limits what can be concluded, even though the study is marked completed and the objectives are specified. The useful statistical reading is therefore careful and bounded: describe what is registered and posted, and separate it from what remains unreported.

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.

Several limitations arise from the information recorded for this specific study. Most importantly, the registry does not report the primary endpoint result, despite identifying treatment discontinuation due to adverse events as the primary outcome. It also does not report results for either Week 100 efficacy outcome. This prevents assessment of the registered questions from outcome estimates.

The primary-treatment arm descriptions contain an internal frequency inconsistency. The Q2W-labeled arm is described as dosing every 4 weeks, and the Q4W-labeled arm as dosing every 2 weeks. The extension descriptions list Q4W and Q8W schedules. The discrepancy makes it important to preserve the labels and descriptions as registered rather than silently infer which schedule was intended.

The nonrandomized design limits direct comparison between groups, and the registry does not describe how participants were assigned or entered each regimen. The record does not report baseline characteristics, group comparability, follow-up duration by group, or adjustment for potential differences. It also does not provide a statistical analysis plan, confidence intervals, missing-data procedures, multiplicity methods, or outcome-specific analysis populations beyond the stated treated population for the primary endpoint.

The serious-event term percentages are listed without a denominator or group attribution. The safety counts are therefore not sufficient to calculate or compare event rates from the information in the record, and no additional rates should be inferred. The listed deaths are not accompanied by causal attribution. Finally, a completed status and a stated enrollment do not substitute for posted endpoint results or a detailed account of the analysis.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Long-term Safety and Efficacy Study of Lebrikizumab (LY3650150) in Participants With Moderate-to-Severe Atopic Dermatitis (ADjoin) — NCT04392154.
• Registry record last updated 2025-11-05; status last verified 2025-10.
• Results first posted 2025-08-03.
• 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

ADjoin is a completed Phase 3, nonrandomized, double-masked, parallel-group study of lebrikizumab for moderate-to-severe atopic dermatitis, sponsored by Eli Lilly and Company. The registry records enrollment of 1153, a study start of 2020-06-15, and a primary completion date of 2024-07-19. The brief summary describes a duration of up to 33 months.

The primary endpoint is the percentage of treated participants discontinuing study treatment due to adverse events from baseline to Week 100. Week 100 IGA 0 or 1 and EASI-75 responses are listed as secondary outcomes. Their results, the primary endpoint result, and statistical analyses are not reported in the registry record. Safety entries list serious and other adverse-event counts, deaths, and serious-event terms, but do not establish comparative or causal effects. The primary-treatment arm labels and dosing descriptions are inconsistent about dosing frequency.