A Study Investigating the Effect of EDP1815 in the Treatment of Mild, Moderate and Severe Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Investigating the Effect of EDP1815 in the Treatment of Mild, Moderate and Severe Atopic Dermatitis (NCT05121480), a randomized phase 2 study. The primary outcome is Achievement of EASI-50.

2. Clinical Question

3. Trial Design

This was a randomized, quadruple-masked, parallel-group Phase 2 study with multiple cohorts and placebo control. The registry lists atopic dermatitis as the condition and Evelo Biosciences, Inc. as the sponsor. The study started on 2022-01-31, reached primary completion on 2023-03-09, and is recorded as completed.

Enrollment was 421. The four cohort descriptions specify different EDP1815 regimens and matching placebo. Cohorts 1 and 2 studied different once-daily total-cell regimens; Cohort 3 studied twice-daily dosing; and Cohort 4 studied a lower once-daily regimen. Each regimen was planned for 16 weeks. Cohorts 1-3 each described 75 participants assigned to active treatment and 25 to placebo, while Cohort 4 described 70 assigned to active treatment and 35 to placebo.

The primary endpoint results identify a shared Cohorts 1-3 placebo group and a separate Cohort 4 placebo group. The comparisons use the Cohorts 1-3 placebo group as the reference. That arrangement matters when reading the estimates: the posted active-arm comparisons are not direct comparisons between the active and placebo groups within each separately named cohort. The registry record does not report additional details on allocation procedures, stratification, adherence, or the rationale for the cohort sizes.

Masking and randomization are design features intended to reduce bias in treatment assignment and outcome assessment. They do not, by themselves, establish that every result is unbiased or precise. The posted outcome counts and the stated endpoint population still need to be considered alongside the comparison structure, incomplete information about analysis methods, and uncertainty in the estimates.

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.

Only the first 6 of 8 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomized participants who completed the EASI at Week 16 or dropped out before Week 16 for treatment-related reasons.

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.

Only the first 6 of 9 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Achievement of EASI-50, measured over 16 weeks.

The primary endpoint was achievement of EASI-50 at Week 16. It counts participants with at least a 50% reduction from baseline in the Eczema Area Severity Index, a measure combining eczema severity and the body surface area affected across four body regions. The registry describes the EASI score range as 0 - 72, with lower scores indicating a better outcome.

A responder endpoint converts each participant’s change into a binary result: the participant either meets the stated threshold or does not. This can be easier to interpret than a mean score change because it expresses the outcome as the number reaching a prespecified degree of improvement. It also discards information about the size of changes on either side of the threshold. Two participants with different score changes can be classified the same way if both meet the threshold, while a participant just short of the threshold is counted as a non-responder.

The endpoint population included randomized participants who completed the Week 16 EASI assessment or dropped out before Week 16 for treatment-related reasons. This is not necessarily the same as every randomized participant. The registry record does not report how missing Week 16 results for other reasons were handled, or whether another analysis population was used by the sponsor.

7. Results Posted in the Registry

• Cohort 1 - Active: 40.9% (27/66) versus 55.9% (38/68) on control; risk difference -15.0 percentage points (95% CI -31.7 to 1.8), risk ratio 0.73 (0.51–1.05); unadjusted two-proportion z-test p = 0.083.
• Cohort 2 - Active: 37.9% (25/66) versus 55.9% (38/68) on control; risk difference -18.0 percentage points (95% CI -34.6 to -1.4), risk ratio 0.68 (0.47–0.99); unadjusted two-proportion z-test p = 0.037.
• Cohort 3 - Active: 31.5% (23/73) versus 55.9% (38/68) on control; risk difference -24.4 percentage points (95% CI -40.3 to -8.5), risk ratio 0.56 (0.38–0.84); unadjusted two-proportion z-test p = 0.004.
• Cohort 4 - Placebo: 44.7% (17/38) versus 55.9% (38/68) on control; risk difference -11.1 percentage points (95% CI -30.9 to 8.6), risk ratio 0.80 (0.53–1.21); unadjusted two-proportion z-test p = 0.271.

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 vIGA of 0

• 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:
  • Cohort 1 - Active: 1.5% (1/65) versus 2.9% (2/68) on control; risk difference -1.4 percentage points (95% CI -6.4 to 3.6), risk ratio 0.52 (0.05–5.63); unadjusted two-proportion z-test p = 0.586.
  • Cohort 2 - Active: 3.0% (2/66) versus 2.9% (2/68) on control; risk difference 0.1 percentage points (95% CI -5.7 to 5.9), risk ratio 1.03 (0.15–7.10); unadjusted two-proportion z-test p = 0.976.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cohort 1 - Active 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 Cohort 2 - Active and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants Achieving a Reduction of ≥4 in the DLQI, of Those With a Score of ≥4 at Baseline

• 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:
  • Cohort 1 - Active: 40.7% (24/59) versus 65.6% (40/61) on control; risk difference -24.9 percentage points (95% CI -42.2 to -7.6), risk ratio 0.62 (0.43–0.89); unadjusted two-proportion z-test p = 0.006.
  • Cohort 2 - Active: 37.7% (23/61) versus 65.6% (40/61) on control; risk difference -27.9 percentage points (95% CI -44.9 to -10.8), risk ratio 0.57 (0.40–0.83); unadjusted two-proportion z-test p = 0.002.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Cohort 1 - Active than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Cohort 2 - Active than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Achieving a Reduction of ≥4 in the Worst PR-NRS, of Those With a Score of ≥4 at Baseline

• 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:
  • Cohort 1 - Active: 17.0% (9/53) versus 26.0% (13/50) on control; risk difference -9.0 percentage points (95% CI -24.8 to 6.8), risk ratio 0.65 (0.31–1.39); unadjusted two-proportion z-test p = 0.264.
  • Cohort 2 - Active: 10.0% (5/50) versus 26.0% (13/50) on control; risk difference -16.0 percentage points (95% CI -30.7 to -1.3), risk ratio 0.38 (0.15–1.00); unadjusted two-proportion z-test p = 0.037.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Cohort 1 - Active and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Cohort 2 - Active than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists secondary outcomes spanning EASI response thresholds and score changes, global assessment categories, and participant-reported measures. These outcomes cover related but not identical aspects of atopic dermatitis and its effects. Their results can help describe whether the primary endpoint pattern is accompanied by similar patterns in other measures, but a secondary measure is not automatically confirmatory evidence for the primary question.

The selected secondary outcomes with posted comparisons do not form a uniformly precise or consistent pattern. Some unadjusted comparisons show differences in the direction of fewer participants reaching the stated threshold in active groups; others are highly uncertain, including intervals that include no difference. The vIGA 0 outcome is particularly sparse in the posted counts, with very few participants reaching that category. Sparse binary outcomes can yield unstable estimates and broad intervals, so those results merit caution.

The registry record lists 33 secondary outcomes, but detailed results in the available data cover only selected outcomes. It does not report sponsor analyses for the selected secondary outcome entries. The endpoint results should therefore be treated as exploratory descriptions rather than as a complete confirmatory analysis of the secondary outcome family.

Multiplicity is important because examining numerous outcomes and treatment comparisons creates more opportunities for chance findings. The registry record does not report a hierarchy, adjustment method, or other plan for controlling this issue. Unadjusted secondary results carry less weight than a clearly prespecified primary analysis with an appropriate testing strategy. They can suggest questions for further study, but they should not be read as independent proof or combined informally into a single conclusion.

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

• Percentage of Participants Achieving EASI-50
• Percentage of Participants Achieving EASI-75
• Percentage of Participants Achieving EASI-90
• Mean Absolute Change in EASI
• Mean Percentage Change in EASI
• Percentage of Participants Achieving Investigator's Global Assessment (vIGA) of 0 or 1 With a ≥2 Point Improvement
• Percentage of Participants Achieving vIGA of 0 or 1
• Percentage of Participants Achieving vIGA of 0

9. Statistical Interpretation

10. Statistical Methods Explained

For a binary endpoint such as EASI-50, the basic outcome is whether each participant met the responder definition by the specified visit. The registry reports counts of participants achieving the endpoint and the endpoint population size for each arm. A risk, in general statistical terms, is the proportion of participants in a group who experience the defined event. The risk difference compares those proportions by subtraction. A negative risk difference for this endpoint indicates a lower observed responder proportion in the active group than in its stated reference group.

What does the risk difference say?

The risk difference is an absolute contrast between two observed proportions. It is expressed on the proportion scale, rather than as a relative comparison. Its direction depends on how the event is defined. Here, the event is achievement of EASI-50, so a negative estimate indicates fewer observed responders in the active group relative to the reference. The estimate describes the data and does not by itself establish a causal explanation.

The registry also posts relative-risk and odds-ratio intervals for the primary comparisons. In general, a relative risk compares event probabilities, while an odds ratio compares event odds. These measures can answer related but different questions; odds and probabilities are not interchangeable. When outcomes are not rare, an odds ratio may appear farther from the null than a relative risk. The registry record does not report the sponsor’s preferred effect measure or a rationale for selecting among these measures.

How should a confidence interval be read?

A confidence interval represents uncertainty around an estimated effect under the statistical method used to construct it. In repeated sampling under the method’s assumptions, intervals constructed in this way would cover the underlying parameter at the stated confidence level at the corresponding long-run rate. It is not a probability statement that the particular fixed parameter lies within the particular calculated interval. A narrower interval generally indicates greater precision, while a wider one leaves more plausible values compatible with the data.

For a risk difference, an interval that includes zero is compatible with no difference under the interval method. If the interval lies entirely below zero, the data are less compatible with zero at that interval’s confidence level, but the interval does not account for every source of uncertainty. In particular, it does not automatically adjust for testing several cohorts or outcomes. The registry record does not state the confidence level or the method used for the posted intervals, so those details are not reported in the registry record.

What does an unadjusted p-value convey?

A p-value measures how unusual the data, or data more extreme, would be under a specified null hypothesis and statistical model. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The posted p-values are described as unadjusted. When many comparisons are examined, the chance of at least one apparently noteworthy result can increase. Without a reported multiplicity plan, an unadjusted p-value should not be treated as confirmatory evidence on its own.

The registry reports no sponsor analyses in the reported-analyses fields for the secondary outcomes described in the results record. It does post outcome counts and selected unadjusted comparisons. This distinction matters: a count-based comparison can be recalculated from arm summaries, but it does not reveal whether the sponsor used a different prespecified model, population, or missing-data method. The registry record does not report a detailed statistical analysis plan, adjustment covariates, or the methods used for the sponsor’s primary analysis.

Why can cohort structure affect interpretation?

The primary result lists a common placebo reference for Cohorts 1-3 and a separate Cohort 4 placebo group. A shared control can be used in designs with several active regimens, but its interpretation depends on the analysis plan and on how comparisons are made. The posted results name the common reference and give separate estimates, but the registry record does not explain whether the comparisons were intended to support a single overall hypothesis, separate dose-specific questions, or another objective.

Randomization supports comparisons between assigned groups when the analysis respects the randomized design. It does not make comparisons between separately randomized cohorts automatically equivalent to a direct within-cohort contrast. In particular, the posted Cohort 4 active comparison uses the Cohorts 1-3 placebo reference, while a distinct Cohort 4 placebo arm is also reported. No direct active-versus-Cohort-4-placebo comparison is reported in the registry record.

What is not known about the analysis?

The registry record does not report the original sample-size justification, the planned statistical model, a hierarchy of endpoints, interim analyses, or the sponsor’s missing-data rules. It also does not report whether the primary analysis adjusted for baseline EASI, cohort, or other variables. These omissions prevent a full assessment of whether the posted unadjusted comparisons correspond to the trial’s prespecified confirmatory analysis.

In general, a well-specified trial analysis identifies the primary population, endpoint, treatment contrast, handling of missing measurements, and approach to multiple testing before results are examined. Those choices affect how estimates and uncertainty should be interpreted. Because the registry record does not report those details, the posted results should be read as the available outcome summaries and not as a complete account of the trial’s statistical analysis.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data locate an effect estimate. The primary confidence intervals are wide enough to leave meaningful uncertainty about the size and direction of some comparisons. For Cohort 1, the interval spans no difference, so the posted data do not distinguish that estimate from no difference using the interval criterion. The Cohort 2 and Cohort 3 intervals lie below no difference, but their width still allows a range of negative effects.

These intervals reflect sampling uncertainty under their construction method. They do not, on their own, account for multiplicity across cohorts and outcomes, possible differences between the posted and prespecified analyses, or uncertainty related to missing outcomes. The confidence-interval method and confidence level are not reported in the registry record.

Arm sizes in the primary endpoint results range from 38 to 73 participants. In general, estimates based on limited group sizes can be imprecise, especially for binary outcomes where the number of responders also affects precision. The registry’s posted counts help readers see the amount of observed information, but they do not resolve uncertainty about the analysis population or missing-data handling.

A confidence interval is not a range containing all plausible results in a clinical sense, and it does not indicate whether an effect is important to participants. Clinical importance requires a relevant benchmark, which is not reported in the registry record. The interval should therefore be read together with the endpoint definition, effect direction, cohort comparison, and limits on inference.

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 illustrations based on the observed effect, not the trial’s original design assumptions. For these binary outcomes, the figures are participants per arm, not total events. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. The original trial’s sample-size justification is not reported in the registry record.

The 80% power column gives the planned number of participants per arm that the illustration estimates would be needed for an 80% chance of rejecting the null hypothesis if the assumed effect were the true effect and the assumptions held. The 90% power column gives the corresponding number for a 90% chance. These are design-planning quantities, not probabilities that the observed result is correct.

Rows for smaller fractions of the observed effect require more participants per arm. This is expected in general: smaller effects are harder to distinguish from random variation, so more information is needed to detect them at a given significance level and power target. The observed effects may themselves be unstable, particularly when group sizes are limited. Planning calculations based on them should not be treated as reliable estimates of the number needed for a future study.

This is not observed power. The calculations do not describe the probability that the completed trial detected an effect after seeing its results. They are hypothetical planning illustrations under the listed assumptions and do not replace the trial’s original design rationale, which is not reported in the registry record.

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 = 0.559; treatment proportion p₁ = 0.409; average p̄ = 0.484.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7067 = 1.3852.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6988 = 0.5881.
4. n = (1.3852 + 0.5881)² / (-0.1497)² = 173.7, rounded up to 174 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 <- 0.558823529411765                          # control proportion
p1 <- 0.409090909090909                          # 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(5121480)
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 = 174 (80% power), 233 (90% power)
 75% of observed effect: n per arm = 310 (80% power), 415 (90% power)
 50% of observed effect: n per arm = 698 (80% power), 934 (90% power)
power.prop.test: 173.7 
Simulated power at n = 174 per arm: 0.788 (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. Multiplicity, Missing Data and Analysis Populations

The study includes multiple cohorts, several dose regimens, a primary endpoint, and many secondary outcomes. Testing multiple questions creates a multiplicity problem: even when there is no true effect for any question, some results may appear notable by chance. The registry record does not report whether the trial used a testing hierarchy, adjusted significance thresholds, or another method to control multiplicity. The posted p-values are unadjusted, so they should be interpreted in that context.

Missing outcome data can also affect a responder analysis. The primary population includes participants who completed the EASI at Week 16 or dropped out before Week 16 for treatment-related reasons. The registry record does not report how participants missing for other reasons were treated, whether imputation was used, or whether sensitivity analyses assessed alternative assumptions. It also does not report how many randomized participants were excluded from the primary endpoint population or the reasons for missing assessments.

These details matter because conclusions can depend on who is included and on what is assumed about missing outcomes. For example, counting missing participants as non-responders, imputing outcomes, or analyzing only observed assessments can produce different summaries. This is a general statistical concern; the registry record does not say which of these approaches, if any, was used by the sponsor.

The record does not report the full analysis population definitions, a prespecified statistical analysis plan, interim analyses, or the original sample-size justification. It also does not report whether baseline covariates were used in the sponsor’s analysis. Without those details, the posted summaries do not establish whether the unadjusted comparisons were the planned primary analysis or how the sponsor accounted for multiplicity and missingness.

15. 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: pneumonia (1.4%), dermatitis atopic (1.4%), ischaemic stroke (1.4%), hepatitis toxic (1.4%), cerebrovascular accident (1.2%).

The registry reports serious adverse events and other adverse events by group. For serious adverse events, the listed affected counts are 1 among 113 in the placebo group, 2 among 74 in Cohort 1 active, 2 among 74 in Cohort 2 active, 1 among 85 in Cohort 3 active, and 0 among 74 in Cohort 4 active. These are counts of participants with at least one serious adverse event in the reported groups; they do not convey event timing, severity detail, or causal attribution.

Other adverse events were reported in 36 among 113 in the placebo group, 21 among 74 in Cohort 1 active, 25 among 74 in Cohort 2 active, 31 among 85 in Cohort 3 active, and 18 among 74 in Cohort 4 active. The registry reports zero deaths in each listed group. Serious-event terms include Pneumonia, Dermatitis Atopic, Ischaemic Stroke, Hepatitis Toxic, Cerebrovascular Accident, and Papillary Thyroid Cancer, with posted percentages ranging from 0.9% to 1.4%.

These summaries are descriptive. They do not establish whether an event was caused by treatment, and small counts limit comparisons. The registry record does not report detailed event narratives or an analysis of between-group safety differences.

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

The posted primary results describe EASI-50 achievement at Week 16 for the reported endpoint population. In the active arms of Cohorts 1-3, the observed responder counts were lower than in the shared placebo group. The posted comparisons vary in precision, with some intervals spanning no difference and others lying below it. The registry does not report a full sponsor analysis plan or multiplicity adjustment, which limits how strongly unadjusted findings can be interpreted.

These results do not establish that EDP1815 is beneficial or harmful, and they should not be used as medical advice. They are a record of observed endpoint counts and associated unadjusted comparisons in a particular Phase 2 study. They do not show what would happen for every participant, in other populations, over a different timeframe, or with a different treatment regimen.

The endpoint is one thresholded measure of eczema severity. It does not capture every possible change in disease, patient experience, or safety. The record lists additional outcomes, but selected secondary comparisons are also unadjusted and are subject to multiplicity and missing-data uncertainties. Their patterns should not be treated as independent confirmation of the primary result.

Interpretation also depends on the comparison structure. The active arms are compared with the shared Cohorts 1-3 placebo reference, while the separate Cohort 4 placebo arm is not used as the stated reference for the Cohort 4 active comparison. The posted results therefore do not provide a direct active-versus-placebo comparison within Cohort 4.

A registry result is not a complete clinical evidence assessment. The record does not report the sponsor’s detailed analysis methods, missing-data rules, or original sample-size rationale. The posted results are best understood as bounded evidence from this study, with uncertainty about the size of the effects and about how the available comparisons relate to any prespecified analysis.

17. Why This Trial Matters Statistically

This trial is statistically interesting because it examined several EDP1815 regimens across multiple cohorts while using placebo control and a binary responder endpoint. Such designs can address several dose and schedule questions, but they also require a clear plan for sharing or separating controls, defining primary comparisons, and accounting for multiple tests.

The common placebo group for Cohorts 1-3 makes the comparison structure especially important. Readers need to distinguish each active-versus-common-control contrast from a within-cohort comparison. Cohort 4 has its own placebo arm, but the posted active comparison uses the shared Cohorts 1-3 placebo reference. That distinction illustrates why arm labels alone are not enough to identify the exact randomized contrast.

The trial also illustrates how an endpoint can be clinically interpretable yet statistically limited. EASI-50 marks a defined response threshold, but the binary classification loses information about the degree of change. The observed counts and intervals show why effect direction, precision, and uncertainty matter alongside p-values.

Finally, the record highlights the importance of reporting the analysis plan, missing-data rules, and multiplicity approach. Without these details, readers can describe the posted outcomes but cannot fully assess whether they reflect the planned confirmatory analyses. This is a general lesson in registry interpretation: transparent design and analysis details are essential for evaluating what a result can support.

18. 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 are specific to the available record. The primary endpoint population is defined by Week 16 completion or treatment-related dropout, but the registry does not explain how other missing assessments were handled. Participant flow into the primary analysis population and reasons for missing outcomes are not reported in the registry entries.

The comparison structure also limits direct interpretation of Cohort 4. A separate Cohort 4 placebo arm is reported, but the posted Cohort 4 active comparison uses the Cohorts 1-3 placebo group. A direct Cohort 4 active-versus-placebo contrast is not reported. Readers should not infer one from the separate arm summaries.

The registry record does not report the original sample-size rationale, statistical analysis plan, adjustment for multiple comparisons, interim analyses, or detailed model specifications. The posted p-values are unadjusted. This limits the ability to distinguish exploratory findings from results intended to support a prespecified confirmatory claim.

The primary endpoint is binary and focuses on crossing a responder threshold. It does not describe the full distribution of EASI changes. The record lists other outcomes, but the secondary results available in detail cover only selected endpoints and do not include reported sponsor analyses. Safety summaries provide group counts but limited context on event timing, attribution, or clinical course.

These limitations do not invalidate the posted data, but they constrain the conclusions that can be drawn from them. The registry record is not a substitute for a complete protocol and statistical analysis report.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study Investigating the Effect of EDP1815 in the Treatment of Mild, Moderate and Severe Atopic Dermatitis — NCT05121480.
• Registry record last updated 2023-08-16; status last verified 2023-08.
• Results first posted 2023-08-16.
• 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.

22. Record Summary

NCT05121480 is a completed randomized, quadruple-masked, parallel-group Phase 2 study of EDP1815 versus matching placebo in participants with mild, moderate or severe atopic dermatitis. The registry reports enrollment of 421 and four dose-regimen cohorts, each planned for 16 weeks.

The primary endpoint was achievement of EASI-50 at Week 16. In the shared Cohorts 1-3 placebo group, 38 of 68 participants achieved the endpoint; active-arm counts were 27 of 66, 25 of 66, and 23 of 73. The registry posts unadjusted comparisons and selected secondary outcome results. It does not report the sponsor’s detailed analysis plan, multiplicity approach, or missing-data rules.

Interpretation should account for the shared placebo comparison structure, the separate Cohort 4 placebo group, and uncertainty in the estimates. The posted results are descriptive registry evidence and do not provide a complete account of the trial’s prespecified statistical analyses.