Efficacy, Safety, and Pharmacokinetic Profile of Etokimab (ANB020) in Adult Participants With Moderate-to-Severe Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy, Safety, and Pharmacokinetic Profile of Etokimab (ANB020) in Adult Participants With Moderate-to-Severe Atopic Dermatitis (NCT03533751), a randomized phase 2 study. The primary outcome is Percent Change From Baseline to Week 16 in Eczema Area and Severity Index (EASI) Score.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-group, dose-ranging phase 2 study sponsored by AnaptysBio, Inc. The registry lists the study as completed, with an enrollment of 302. The recorded start date is 2018-06-19, and the primary completion date is 2019-12-03. The registry describes masking as quadruple.

Participants were assigned to placebo or one of four etokimab regimens. Three regimens used a loading dose on Day 1 followed by a lower or higher maintenance dose; the fourth used 20 mg every 4 weeks without a listed loading dose. The treatment period was up to 16 weeks. The 150 mg every-8-weeks regimen included placebo administrations at Weeks 4 and 12, matching the schedule of the more frequent injections.

The primary endpoint was percent change from baseline to Week 16 in EASI score. The reported primary population was the Full Analysis Set, and missing data were imputed using multiple imputation. The registry also lists secondary outcomes covering EASI response thresholds, investigator global assessment, itch ratings, and SCORAD. Numerical results are available for the EASI 50, EASI 75, and EASI 90 outcomes described in the posted results; numerical results for the other listed secondary outcomes are not reported in the registry record.

The original protocol’s allocation ratio, planned sample size by arm, sample-size justification, and detailed statistical plan are not reported in the registry record. The posted enrollment does not by itself establish how the original study was planned or powered.

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: Participants were equally randomized on Day 1 to one of five treatment groups.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set. Missing data were imputed using multiple imputation.

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: The Safety Analysis Set included all randomized participants who received at least 1 dose of etokimab or placebo.

6. Primary Endpoint

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

EASI is a clinical score that combines the extent of affected skin with the severity of eczema signs. Assessments cover the head and neck, trunk, upper limbs, and lower limbs. The signs include redness, thickness or swelling, excoriation from scratching, and lichenification. The total score ranges from 0, indicating no disease, to 72, indicating worse disease.

The primary endpoint is the percent change in EASI score from baseline to Week 16. A negative percent change represents a decrease from baseline, so the reported negative group means describe reductions in the score. Comparing changes against a placebo group helps distinguish the observed changes during etokimab treatment from changes observed under the study’s placebo regimen.

The registry reports least-squares means and standard errors for the primary outcome, rather than only raw group averages. Least-squares means are model-based estimates; the registry’s reported between-group analyses used ANCOVA. The Full Analysis Set was used, with missing data imputed using multiple imputation. Details of the imputation model and other model specifications are not reported in the registry record.

EASI is used because it combines the distribution and severity of eczema into a common score that can be compared over time and between randomized groups. It does not, by itself, describe every symptom or aspect of participants’ experience. The listed itch, investigator assessment, and SCORAD outcomes address additional aspects, but numerical results for most of those outcomes are not reported in the registry record.

7. Results Posted in the Registry

• Etokimab 20 mg SC Q4W: mean -41.63 versus -49.38 on control; difference 7.75 (95% CI -11.43 to 26.93), standardized effect d = 0.14; z-test p = 0.428.
• Etokimab 300 mg / 150 mg SC Q8W: mean -55.70 versus -49.38 on control; difference -6.32 (95% CI -24.84 to 12.20), standardized effect d = -0.12; z-test p = 0.504.
• Etokimab 300 mg / 150 mg SC Q4W: mean -47.40 versus -49.38 on control; difference 1.98 (95% CI -16.39 to 20.35), standardized effect d = 0.04; z-test p = 0.833.
• Etokimab 600 mg / 300 mg SC Q4W: mean -44.56 versus -49.38 on control; difference 4.82 (95% CI -15.90 to 25.54), standardized effect d = 0.08; z-test p = 0.648.

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.

Number of Participants With a 50% Reduction From Baseline in Eczema Area and Severity Index Score (EASI 50 Response) at Week 16

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Etokimab 20 mg SC Q4W: 31.1% (19/61) versus 35.0% (21/60) on control; risk difference -3.9 percentage points (95% CI -20.6 to 12.9), risk ratio 0.89 (0.54–1.48); unadjusted two-proportion z-test p = 0.652.
  • Etokimab 300 mg / 150 mg SC Q8W: 45.8% (27/59) versus 35.0% (21/60) on control; risk difference 10.8 percentage points (95% CI -6.8 to 28.3), risk ratio 1.31 (0.84–2.04); unadjusted two-proportion z-test p = 0.231.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etokimab 20 mg SC Q4W and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etokimab 300 mg / 150 mg SC Q8W and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With a 75% Reduction From Baseline in Eczema Area and Severity Index Score (EASI 75 Response) at Week 16

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Etokimab 20 mg SC Q4W: 16.4% (10/61) versus 16.7% (10/60) on control; risk difference -0.3 percentage points (95% CI -13.5 to 13.0), risk ratio 0.98 (0.44–2.19); unadjusted two-proportion z-test p = 0.968.
  • Etokimab 300 mg / 150 mg SC Q8W: 23.7% (14/59) versus 16.7% (10/60) on control; risk difference 7.1 percentage points (95% CI -7.3 to 21.4), risk ratio 1.42 (0.69–2.95); unadjusted two-proportion z-test p = 0.337.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etokimab 20 mg SC Q4W and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etokimab 300 mg / 150 mg SC Q8W and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With a 90% Reduction From Baseline in Eczema Area and Severity Index Score (EASI 90 Response) at Week 16

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Etokimab 20 mg SC Q4W: 8.2% (5/61) versus 5.0% (3/60) on control; risk difference 3.2 percentage points (95% CI -5.6 to 12.0), risk ratio 1.64 (0.41–6.56); unadjusted two-proportion z-test p = 0.479.
  • Etokimab 300 mg / 150 mg SC Q8W: 11.9% (7/59) versus 5.0% (3/60) on control; risk difference 6.9 percentage points (95% CI -3.1 to 16.8), risk ratio 2.37 (0.64–8.74); unadjusted two-proportion z-test p = 0.177.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etokimab 20 mg SC Q4W and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etokimab 300 mg / 150 mg SC Q8W and the control. The interval gives the range of effects that remain compatible with the data.

The numerical secondary results reported for EASI response thresholds show no uniform pattern across all four etokimab regimens. At the less stringent threshold, the every-8-weeks regimen had a higher count than placebo, while the other regimens had counts close to or below the placebo count. At the more stringent thresholds, counts were small and the relative ordering varied. The posted response comparisons have substantial uncertainty, and their reported intervals include the null value.

This overall pattern does not supply clear, consistent corroboration of the primary ANCOVA comparisons. It also does not establish that the treatments are equivalent: limited counts and intervals that cover a range of possible differences leave uncertainty. The secondary measures address different definitions of improvement, so they need not produce identical estimates. They should be considered alongside, rather than substituted for, the continuous primary EASI analysis.

The registry lists other secondary outcomes, including investigator global assessment, itch ratings, and SCORAD. Numerical results for those outcomes are not reported in the registry record. Consequently, the secondary evidence available for interpretation is incomplete across the full range of planned or listed measures.

Secondary analyses carry less weight when many endpoints or comparisons are examined and the record does not describe a multiplicity strategy. In general, repeated testing can make at least one result appear unusual by chance, even if no underlying difference exists. The registry lists 11 outcome measures and 40 analyses, but the multiplicity adjustment strategy is not reported in the registry record. Individual secondary p-values should therefore be treated as exploratory context rather than conclusive evidence, especially when they are not supported by a consistent primary result.

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

• Number of Participants With a 50% Reduction From Baseline in Eczema Area and Severity Index Score (EASI 50 Response) at Week 16
• Number of Participants With a 75% Reduction From Baseline in Eczema Area and Severity Index Score (EASI 75 Response) at Week 16
• Number of Participants With a 90% Reduction From Baseline in Eczema Area and Severity Index Score (EASI 90 Response) at Week 16
• Number of Participants Who Achieved a Reduction of ≥ 2 Points From Baseline in the Validated Investigator's Global Assessment for Atopic Dermatitis (vIGA-AD) at Week 16
• Number of Participants Who Achieved a vIGA-AD Response of 0 (Clear) or 1 (Almost Clear) at Week 16
• Number of Participants Who Achieved a Reduction of ≥ 4 Points From Baseline in Weekly Averaged Peak Numerical Rating Scale (NRS) for Pruritus Score at Week 16
• Percent Change From Baseline in Peak Weekly Averaged Numerical Rating Scale (NRS) for Pruritus Score at Week 16
• Percent Change From Baseline in Scoring Atopic Dermatitis (SCORAD) Score at Week 16

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a percent change in an EASI score. For each randomized group, the registry reports a least-squares mean and its standard error. A least-squares mean is an estimate from a statistical model, rather than necessarily the simple arithmetic average of the observed participants’ changes. The registry identifies ANCOVA as the method used for the reported primary comparisons.

What does the group estimate represent?

The group estimates describe the modeled average percent change from baseline to Week 16 within each arm. Negative values correspond to decreases in EASI score. A group estimate does not on its own answer whether the group changed differently from placebo. That question is addressed by the between-group estimate, which expresses the contrast between an etokimab arm and placebo.

For the registry’s ANCOVA results, the contrast is a least-squares mean difference. Its sign indicates the direction of the contrast as recorded, but its interpretation should be paired with the group estimates and confidence interval. A confidence interval gives a range of values compatible with the estimate and the analysis assumptions. When an interval includes zero, it includes no between-group difference as one plausible value. That does not prove the groups are identical; it indicates that the estimate is not precise enough to exclude no difference at the interval’s stated confidence level.

Why use ANCOVA?

ANCOVA is a regression-based method for comparing a continuous outcome between groups while accounting for model terms. In randomized trials it is commonly used to estimate adjusted group contrasts. The registry record names ANCOVA and reports least-squares mean differences, but the covariates and full model specification are not reported in the registry record. Therefore, the precise adjustment used in the sponsor’s model cannot be reconstructed from the brief method label alone.

The recalculated comparisons use the posted arm-level summary estimates and standard errors to form simple, unadjusted contrasts. The record indicates that the pooled standard deviation used in these calculations is approximate. These recalculations provide a useful check on the scale and direction of the posted differences, but they are not a replication of the sponsor’s ANCOVA. The registry’s own reported analysis is the more relevant result for describing the trial’s prespecified or sponsor-reported comparison; whether it was prespecified cannot be confirmed from the available record.

How should binary response outcomes be read?

EASI 50, EASI 75, and EASI 90 classify participants according to whether they reached a specified reduction from baseline by Week 16. The counts are therefore binary outcomes at the participant level: each participant either met the threshold or did not. The registry’s reported analyses for selected comparisons use logistic regression and report odds ratios. An odds ratio compares the odds of meeting the response definition between groups; it is not the same quantity as a difference in response probabilities.

The recalculated binary comparisons use risk differences, which express the difference between the groups’ observed response proportions. Their confidence intervals express uncertainty in that difference. The odds ratios in the sponsor analyses and the risk differences in the simple recalculations use different scales and may not have identical numerical values. Neither scale removes uncertainty, especially where the response counts are limited. The registry does not report sponsor analyses for every possible arm contrast on these secondary outcomes.

Why does over-dispersion matter in general?

In general, over-dispersion occurs when observed variability in count or binary data exceeds what a simple model assumes. If a model underestimates variability, standard errors may be too small and confidence intervals too narrow. This is a general statistical consideration, not a finding about this trial. Whether an over-dispersion assessment or adjustment was used is not reported in the registry record.

For the primary outcome, missing data were imputed using multiple imputation in the Full Analysis Set. In general, multiple imputation creates several plausible completed datasets under a specified model, analyzes each, and combines the results to reflect uncertainty from missing observations. The imputation variables, assumptions about why data were missing, number of imputations, and sensitivity analyses are not reported in the registry record. Those details matter because an imputation-based estimate depends on its modeling assumptions.

The registry lists 11 outcome measures and 40 analyses. The number of outcomes and comparisons is relevant when interpreting isolated p-values: repeated testing can increase the chance of a nominally unusual result somewhere among many tests. The multiplicity adjustment strategy, if any, is not reported in the registry record. The reported secondary p-values should therefore be read as unadjusted evidence unless the record specifies otherwise; no such adjustment is described in the available information.

11. Confidence Intervals and Precision

Precision is reflected by the width of a confidence interval. A narrower interval indicates that the estimate is more tightly localized under the analysis assumptions; a wider interval leaves more uncertainty about the size and direction of the contrast. The primary ANCOVA intervals are broad enough to include zero for each etokimab-versus-placebo comparison. They also span values on both sides of zero, so the posted analyses do not precisely determine whether any regimen’s mean change was greater or smaller than placebo’s.

The simple recalculated intervals show the same broad conclusion. For etokimab 20 mg every 4 weeks, the estimated difference is 7.75, with a confidence interval from -11.43 to 26.93. For the 150 mg every-8-weeks regimen, the estimate is -6.32, with an interval from -24.84 to 12.2. For the 150 mg every-4-weeks regimen, the estimate is 1.98, with an interval from -16.39 to 20.35. For the 300 mg every-4-weeks regimen, the estimate is 4.82, with an interval from -15.9 to 25.54. Each interval includes zero.

The participant counts in the arms range from 59 to 61 for the primary endpoint. With groups of this size, estimates can remain uncertain even when the group means differ numerically. The standard errors reported for the group estimates indicate sampling uncertainty around each modeled mean; they are not standard deviations of individual participant outcomes. The registry does not report confidence limits for the individual least-squares means in the arm summaries, but does report intervals for the between-group analyses.

For the binary secondary outcomes, response counts are modest, particularly at the more stringent response threshold. Their comparison intervals are correspondingly broad and include no difference. The uncertainty is important: a point estimate above or below the placebo result should not be treated as definitive evidence about the underlying difference. The sponsor’s logistic-regression intervals likewise extend across the null value for the listed EASI response comparisons.

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 illustrations translate an observed contrast into hypothetical future study sizes. For these continuous and binary comparisons, the figures are participants per arm, not total participants and not total events. The planning assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect.

The 80% power column gives the estimated number of participants per arm for a hypothetical study to have an 80% chance of detecting an effect of the specified size under those assumptions. The 90% power column gives the corresponding number for a higher chance of detection. These are planning illustrations based on the observed effect, not the original design calculations for this trial. The original trial’s design assumptions and sample-size justification are not reported in the registry record.

The rows labeled as planning for 100%, 75%, or 50% of the observed effect illustrate how the assumed target effect changes the required enrollment. A smaller target effect generally requires more participants because a weaker signal is harder to distinguish from random variation. When the observed contrast is very small, using it as the effect to detect can lead to very large hypothetical sample-size estimates. Such values should not be read as a recommendation for a future trial without considering its scientific question and design.

These figures are not observed power. They do not measure the probability that the completed trial was capable of finding its result, and they do not change the uncertainty in the posted estimates. They are conditional planning calculations based on the listed assumptions and observed contrasts.

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 = 7.750; pooled standard deviation σ = 53.789.
2. n = 2 × (1.96 + 0.842)² × 53.789² / 7.750² = 756.2, rounded up to 757 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 <- 7.75                      # observed difference in means
sigma <- 53.789252494021                     # 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(3533751)
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 = 757 (80% power), 1,013 (90% power)
 75% of observed effect: n per arm = 1,345 (80% power), 1,800 (90% power)
 50% of observed effect: n per arm = 3,025 (80% power), 4,050 (90% power)
power.t.test: 757.1 
Simulated power at n = 757 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 sponsor analyses using ANCOVA for the primary continuous endpoint. Those analyses give least-squares mean differences of 7.75, -6.32, 1.97, and 4.82 for the four etokimab regimens compared with placebo. The corresponding reported confidence intervals all include zero, and the reported p-values are all greater than 0.05.

The simple recalculations use posted group estimates and standard errors. They yield differences of 7.75, -6.32, 1.98, and 4.82. Their intervals also include zero. The small variation between the recalculated estimate of 1.98 and the sponsor’s ANCOVA estimate of 1.97 reflects that these are not identical analyses: the sponsor reports model-based least-squares contrasts, while the recalculation is a simple unadjusted contrast from summaries.

The registry does not flag a discrepancy between the sponsor analysis and the recalculations. The sponsor’s ANCOVA remains the principal reported analysis for the primary endpoint; the recalculations are a secondary check, not a replacement for the model-based analysis. The registry record does not provide enough detail to reconstruct the complete ANCOVA model or determine the extent to which the difference between methods reflects specific adjustments.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several hypotheses are tested in one study. Each additional test creates another opportunity for a result to appear statistically notable by chance. This trial has multiple arms, multiple outcomes, and multiple analyses. The registry lists 11 outcome measures and 40 analyses. The multiplicity adjustment strategy is not reported in the registry record, so the posted p-values cannot be assumed to control the overall false-positive risk across all comparisons.

The primary endpoint is the central basis for the main interpretation. The reported ANCOVA comparisons do not establish a difference from placebo at the conventional 0.05 level. Secondary outcomes can add context, but the registry’s numerical secondary results cover only selected EASI thresholds, and those comparisons are also uncertain. The other listed secondary outcomes do not have numerical results reported in the registry record. This limits the ability to assess consistency across the full set of outcome domains.

The record states that the primary endpoint analysis used the Full Analysis Set and that missing data were imputed using multiple imputation. That identifies the broad approach but does not describe the imputation model, the amount or pattern of missing data, the assumptions about missingness, or sensitivity analyses. These details are important because an imputed estimate relies on assumptions about the unobserved outcomes.

The original sample-size justification, planned statistical analysis plan, interim analyses, and treatment of multiplicity are not reported in the registry record. Their absence means the posted material does not establish whether the analyses followed a prespecified hierarchy or whether any additional planned analyses were performed. This is a limitation of what can be inferred from the registry entry, not evidence that a particular procedure was or was not used.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: dermatitis atopic (5.0%), joint instability (1.7%), hysterectomy (1.7%), dermatitis exfoliative generalized (1.7%), eczema herpeticum (1.7%).

The safety summaries report participants affected by serious adverse events and by other, non-serious adverse events in each randomized group. Serious adverse events were reported for 1 participant in placebo, 2 in etokimab 20 mg every 4 weeks, and 3 in each of the three loading-dose regimens. The counts at risk in those groups ranged from 59 to 61. No deaths were reported in any group.

Other adverse events were reported for 24 participants in placebo, 22 in the 20 mg regimen, 23 in the every-8-weeks regimen, 18 in the 150 mg every-4-weeks regimen, and 26 in the 300 mg every-4-weeks regimen. These are counts, not adjusted comparisons. They do not establish that a regimen caused or prevented an event, and differences between counts should be interpreted cautiously in groups of this size.

The registry lists serious-event terms including atopic dermatitis, joint instability, hysterectomy, generalized exfoliative dermatitis, eczema herpeticum, and increased troponin I. The record does not provide enough detail in these summaries to assess timing, duration, causal attribution, or the clinical course of each event. The safety figures describe events recorded in the study; they are not a substitute for a full safety analysis.

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

The primary results show substantial decreases in mean EASI score from baseline in every study group, including placebo. The key comparison is not whether scores decreased within a group, but whether the change differed between an etokimab regimen and placebo. The registry’s ANCOVA estimates and the simple recalculated estimates have confidence intervals that include zero. The posted results therefore leave uncertainty about the direction and size of the between-group differences.

This interpretation does not mean that the groups were proven identical. A confidence interval that includes no difference can also include some degree of benefit or worsening relative to the comparator. The estimates are not precise enough to exclude those possibilities. Similarly, a numerical difference between arm estimates is not, on its own, proof of a treatment effect.

The EASI response counts provide additional but limited information. The pattern varies across doses and response thresholds, and the available comparisons do not establish clear separation from placebo. Numerical results for several other listed secondary outcomes are not reported in the registry record, which prevents a complete assessment of whether the pattern is consistent across symptoms and assessment methods.

These findings apply to the adults enrolled in this phase 2 study, the regimens evaluated, and the Week 16 endpoint. They do not establish what would happen with other doses, schedules, durations, or populations. The registry summaries also do not report the detailed model specifications, missing-data diagnostics, or multiplicity plan needed to fully evaluate every analytic assumption. A result from one study should therefore be interpreted as evidence with stated uncertainty, not as a universal conclusion.

18. Why This Trial Matters Statistically

This study is statistically informative because it compares multiple dose regimens against a concurrent placebo group using a common clinical score. The parallel-group randomized design gives a structured basis for comparing changes over the same period, while the double-blind and placebo-controlled features are intended to reduce influences on outcome assessment and participant expectations. The registry describes quadruple masking, though it does not detail which roles were masked.

The study also illustrates why within-group change and between-group treatment contrasts answer different questions. All arm estimates show reductions in EASI from baseline, but the primary question concerns whether those reductions differ from placebo. The confidence intervals for the reported contrasts include zero, underscoring that numerical movement in group means is not enough to establish a between-group difference.

Comparing several regimens can describe a dose-ranging pattern, but it also creates multiple comparisons. The record lists many outcomes and analyses, while its multiplicity strategy is not reported. This makes the hierarchy of evidence important: the primary outcome provides the central comparison, and secondary results require cautious interpretation, especially when only some numerical outcomes are available.

Finally, the registry’s reporting makes the distinction between sponsor analyses and summary-statistic recalculations visible. The recalculations can help readers understand scale and uncertainty, while the sponsor’s ANCOVA is the reported model-based analysis. Neither should be mistaken for greater certainty than the intervals support. The study therefore offers a useful example of interpreting estimates, uncertainty, and incomplete analysis detail together.

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.

The registry record does not report the original sample-size justification, planned enrollment by arm, or assumptions used in designing the trial. The retrospective planning illustrations based on observed effects cannot fill that gap. The full statistical analysis plan, including any hierarchy for the multiple doses and endpoints, is also not reported in the registry record.

Although the primary population and missing-data approach are identified, the registry does not give the amount of missing data, the imputation model details, or sensitivity analyses. Without these details, readers cannot assess how dependent the primary estimates may be on imputation assumptions. The complete ANCOVA specification and covariates are also not reported, limiting independent reconstruction of the sponsor’s analysis.

The groups were modest in size, with 59 to 61 participants in the primary endpoint summaries. The primary confidence intervals are broad and include no difference. The secondary binary response counts are also limited, particularly at the more stringent thresholds, and their estimates are uncertain. These features constrain precision rather than proving that any specific effect is absent.

The registry lists a broader set of secondary measures than the numerical results available for review. Results for several listed outcomes are not reported in the registry record, so consistency across itch, investigator ratings, and SCORAD cannot be evaluated from these data. Safety summaries provide affected-participant counts but do not supply detailed event narratives, timing, or causal assessments. These reporting limits narrow the conclusions that can be drawn from the registry entry.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy, Safety, and Pharmacokinetic Profile of Etokimab (ANB020) in Adult Participants With Moderate-to-Severe Atopic Dermatitis — NCT03533751.
• Registry record last updated 2023-05-24; status last verified 2023-04.
• Results first posted 2023-05-24.
• 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

This completed phase 2 study randomized adults with moderate-to-severe atopic dermatitis to placebo or one of four etokimab regimens. The primary endpoint was percent change from baseline to Week 16 in EASI score, analyzed in the Full Analysis Set with multiple imputation for missing data.

All groups had negative mean percent changes in EASI. The sponsor-reported ANCOVA contrasts against placebo had confidence intervals that included zero, as did the simple recalculations from arm summaries. The posted primary analyses therefore do not establish a between-group difference, while leaving uncertainty about the possible size and direction of effects.

Available secondary EASI response results are mixed and imprecise; numerical results for several other listed secondary outcomes are not reported in the registry record. The registry also does not report the original sample-size justification, complete statistical plan, or multiplicity strategy. Safety summaries report serious adverse events in each group and no deaths. These results should be interpreted in light of their uncertainty and the registry’s reporting limits.