A Study to Evaluate the Efficacy and Safety of Bimekizumab Compared to Placebo and an Active Comparator in Adult Subjects With Moderate to Severe Chronic Plaque Psoriasis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Efficacy and Safety of Bimekizumab Compared to Placebo and an Active Comparator in Adult Subjects With Moderate to Severe Chronic Plaque Psoriasis (NCT03370133), a randomized phase 3 study. The primary outcome is Percentage of Participants With a Psoriasis Area and Severity Index 90 (PASI90) Response at Week 16.

2. Clinical Question

The clinical question is framed as a comparison of response proportions at a particular assessment time. It does not ask whether every participant improved, how long any response lasted after treatment, or how participants valued a change in skin symptoms. Those questions require different outcomes or follow-up information.

3. Trial Design

The registry describes a phase 3, multicenter, randomized, double-blind, placebo- and active-comparator-controlled, parallel-group study. The trial was completed. Its recorded start date is 2017-12-06 and its primary completion date is 2019-01-08. The sponsor is UCB Biopharma SRL.

Three groups contribute to the week 16 primary endpoint: placebo, bimekizumab 320 mg Q4W and ustekinumab. The analysis population is called the Randomized Set, and the registry defines that set as all randomized study participants. The reported group sizes for this endpoint are 83 for placebo, 321 for bimekizumab and 163 for ustekinumab. These are the endpoint group sizes, rather than a new estimate of enrollment.

The parallel-group structure means participants were assigned to groups that were followed alongside one another. The record identifies random allocation and quadruple masking. It does not describe the allocation sequence, concealment procedure, or exactly which roles were masked; those details are not reported in the registry record. The record also does not explain how the placebo was matched to each active treatment, beyond noting placebo administration at specified time points in the ustekinumab cohort to maintain blinding.

Treatment differed over time in the placebo cohort: participants received placebo through week 16 and then bimekizumab through week 52. Therefore the week 16 assessment is the period when the placebo group provides a placebo comparison, while later follow-up does not retain an untreated placebo group as described in the arm information. Ustekinumab was given for 52 weeks, with dose determined according to participant weight. Further dosing details are not reported in the registry record.

The registry record does not state the original sample-size justification, the assumptions used in planning the trial, or whether interim analyses were performed. It also does not give the detail needed to reconstruct every feature of the randomization or masking procedures.

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: The study included a 2-5 week Screening Period, a 16-week Initial Period and a 36-week Maintenance Period. After the Maintenance Period participants either enrolled in an open-label study or had a SFU Visit 20 weeks after their final dose (including those withdrawn from IMP). Participant Flow refers to the Randomized Set and Maintenance Set.

Initial Treatment Period (WK 16)

Reasons for not completing this period:

Maintenance Treatment Period (WK 52)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Randomized Set (RS) consisted of all randomized study participants.

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: Baseline characteristics refer to the Randomized Set (RS) consisting of all participants randomized into the study.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With a Psoriasis Area and Severity Index 90 (PASI90) Response at Week 16, measured over Week 16.

The primary endpoint is the percentage of participants with a PASI90 response at week 16. A PASI90 response means at least 90% improvement in the Psoriasis Area and Severity Index (PASI) score from baseline. PASI combines assessments of redness, thickness and scaling across four body areas with an assessment of how much skin is affected. The registry describes component scores and the affected-area score being combined into a final score; a score of 0 indicates no disease and 72 is the maximum possible score.

The endpoint turns change in a multi-part skin assessment into a binary outcome: each participant either meets the specified response threshold or does not. A percentage of participants can then be compared across randomized groups. This makes the result readily interpretable as the frequency of a substantial improvement threshold, but it does not convey the size of improvement among people who fall short of that threshold or distinguish different degrees of improvement among responders.

A threshold such as PASI90 is useful for comparing the proportion reaching a defined level of change at a common time point. It is not a complete description of the clinical experience: it does not, by itself, describe symptoms, quality of life, durability, or safety. The registry lists symptom and other skin response outcomes separately as secondary outcomes. The primary endpoint's particular threshold and week 16 assessment define the scope of the main statistical comparison, not every possible measure of benefit or harm.

7. Results Posted in the Registry

• Bimekizumab 320 mg Q4W (RS): 85.0% (273/321) versus 4.8% (4/83) on control; risk difference 80.2 percentage points (95% CI 74.2 to 86.3), risk ratio 17.65 (6.78–45.96); unadjusted two-proportion z-test p = <0.001.
• Ustekinumab (RS): 49.7% (81/163) versus 4.8% (4/83) on control; risk difference 44.9 percentage points (95% CI 35.9 to 53.8), risk ratio 10.31 (3.91–27.16); unadjusted two-proportion z-test p = <0.001.

8. Secondary Efficacy Endpoints

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

Percentage of Participants With a PASI100 Response at Week 16

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Bimekizumab 320 mg Q4W (RS): 58.6% (188/321) versus 0.0% (0/83) on control; risk difference 58.6 percentage points (95% CI 53.2 to 64.0); unadjusted two-proportion z-test p = <0.001.
  • Ustekinumab (RS): 20.9% (34/163) versus 0.0% (0/83) on control; risk difference 20.9 percentage points (95% CI 14.6 to 27.1); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Bimekizumab 320 mg Q4W (RS) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab (RS) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With an IGA 0 Response at Week 16

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Bimekizumab 320 mg Q4W (RS): 58.6% (188/321) versus 0.0% (0/83) on control; risk difference 58.6 percentage points (95% CI 53.2 to 64.0); unadjusted two-proportion z-test p = <0.001.
  • Ustekinumab (RS): 22.1% (36/163) versus 0.0% (0/83) on control; risk difference 22.1 percentage points (95% CI 15.7 to 28.5); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Bimekizumab 320 mg Q4W (RS) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab (RS) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With a PASI75 Response at Week 4

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Bimekizumab 320 mg Q4W (RS): 76.9% (247/321) versus 2.4% (2/83) on control; risk difference 74.5 percentage points (95% CI 68.9 to 80.2), risk ratio 31.93 (8.11–125.72); unadjusted two-proportion z-test p = <0.001.
  • Ustekinumab (RS): 15.3% (25/163) versus 2.4% (2/83) on control; risk difference 12.9 percentage points (95% CI 6.5 to 19.4), risk ratio 6.37 (1.54–26.22); unadjusted two-proportion z-test p = 0.002.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Bimekizumab 320 mg Q4W (RS) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab (RS) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes with numerical results in the registry include PASI100 response and IGA 0 response at week 16, as well as PASI75 response at week 4. Across these outcomes, the reported response percentages follow the same broad ordering as the primary result: the bimekizumab group is higher than placebo, and the reported bimekizumab values are higher than those for ustekinumab. The earlier week 4 PASI75 assessment adds a different time point and response threshold; it is not a substitute for the week 16 PASI90 primary endpoint.

The registry also lists secondary outcomes relating to pain, itch, scaling, scalp assessment and PASI90 at week 12. Numerical results for those listed outcomes are not reported in the registry record available for this analysis. The record indicates 17 posted secondary outcomes, but that count does not mean that all outcome estimates are available in the results. No conclusions about the unreported numerical results should be inferred.

Secondary outcomes can help show whether the primary pattern is accompanied by responses on other measures, but they have a different inferential status. When many outcomes, time points or group comparisons are tested, the chance of at least one nominally small p-value can increase. If no multiplicity adjustment is specified, individual p-values should be treated as exploratory rather than as a set of confirmatory findings. The registry record does not report a multiplicity procedure for these secondary analyses.

The secondary results are best read as supporting context for the primary endpoint, not as independent proof of every possible treatment effect. Their different thresholds, assessment times and constructs also mean that apparent consistency does not make the outcomes interchangeable. The primary comparison remains the pre-eminent statistical question in the registry record; interpretation of secondary outcomes should account for multiplicity and the incomplete reporting of numerical results for several listed measures.

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

• Percentage of Participants With a PASI100 Response at Week 16
• Percentage of Participants With an IGA 0 Response at Week 16
• Percentage of Participants With a PASI75 Response at Week 4
• Percentage of Participants With a Patient Symptom Diary Response for Pain at Week 16
• Percentage of Participants With a Patient Symptom Diary Response for Itch at Week 16
• Percentage of Participants With a Patient Symptom Diary Response for Scaling at Week 16
• Percentage of Participants With a Scalp IGA Response (Clear or Almost Clear) at Week 16 for Participants With Scalp Psoriasis (PSO) >=2 at Baseline
• Percentage of Participants With a PASI90 Response at Week 12

9. Statistical Interpretation

10. Statistical Methods Explained

What does the response percentage represent?

For this endpoint, a response is a PASI90 result at week 16. The group percentage is the share of participants in that group who met the threshold. Because the outcome is binary, it can be described as a proportion or as a percentage. This presentation is direct, but it reduces a graded change in PASI to a yes-or-no classification. It does not show the distribution of individual PASI changes.

What is a risk difference?

General statistical explanation: a risk difference subtracts the response proportion in one group from the response proportion in another. It describes an absolute separation in response proportions, rather than a relative comparison. The re-calculated comparisons use this scale and are based on the posted response counts and denominators without adjustment. The registry's arm percentages are rounded, so the re-calculation uses the underlying counts rather than attempting to reconstruct counts from rounded percentages.

An absolute difference can be useful because it keeps the comparison tied to the percentage-point scale of the endpoint. Its size depends on the response frequency in both groups. It is distinct from a risk ratio or an odds ratio, which compare groups on relative scales. Different scales answer different questions and can make the same data look numerically very different.

What is an odds ratio?

General statistical explanation: odds are the probability of an event divided by the probability of no event. An odds ratio compares the odds between groups. It is not the same as a ratio of response percentages, and it is not an absolute difference. When responses are common, an odds ratio can appear farther from 1 than a risk ratio for the same data. The registry reports odds ratios for its formal primary analyses, while the simple re-calculations use risk differences. Interpret each measure on its own scale.

What does the reported analysis method do?

The registry identifies the Cochran-Mantel-Haenszel method for the reported primary comparisons. General statistical explanation: this method can estimate a common association across strata, accounting for specified stratification factors rather than treating all observations as one undifferentiated comparison. The factors and detailed implementation are not reported in the registry record. Accordingly, the reported odds ratios should be read as the registry's formal analyses, not as unadjusted estimates reconstructed from the overall counts.

The reported p-values are less than 0.001 for both primary comparisons. General statistical explanation: a p-value evaluates how compatible the observed data are with a stated null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor is it a measure of the probability that the result will replicate. A small p-value alone does not establish the magnitude, practical importance, or completeness of an effect.

How should confidence intervals be read?

General statistical explanation: a confidence interval gives a range of values compatible with the data and method at a stated confidence level. The registry reports 95% confidence intervals for the primary odds ratios. The range expresses statistical uncertainty in the estimated comparison; it is not a range containing 95% of participants' outcomes. Its width reflects factors such as event counts and sample sizes, as well as the method used. The interval does not include uncertainty from every possible source, such as departures from the analysis assumptions or differences in populations.

Why do outcome counts matter?

General statistical explanation: a binary endpoint is informed by the number of participants who experience the event and the number who do not. Small event counts can make estimates less stable and intervals wider, particularly for relative measures. In the placebo group, 4 participants had a PASI90 response among 83 in the primary analysis. This count is relevant to interpreting precision and helps explain why the comparison should be read with its interval, rather than from the point estimate alone.

Allocation was randomized, and the registry describes the study as double-blind, with quadruple masking recorded. Random allocation is intended to create comparable groups on average, while masking can reduce the influence of expectations on care or assessment. General explanation does not establish that bias was absent in practice. Details about implementation, adherence, deviations, and outcome assessment procedures are not reported in the registry record.

The registry calls the primary population the Randomized Set and defines it as all randomized participants. The record labels the safety groups separately, including Initial Period and Any Bimekizumab or Any Ustekinumab, but does not provide detailed definitions of all safety populations or analysis rules. Those labels should not be treated as interchangeable with the primary efficacy population.

The statistical plan, including any planned covariates, stratification factors, missing-data method, multiplicity procedure, and interim analyses, is not reported in the registry record. The reported method and comparison results can be described, but additional choices should not be inferred.

11. Confidence Intervals and Precision

The registry reports a 95% confidence interval of 34.020 to 293.175 for the bimekizumab-versus-placebo odds ratio of 99.869. For bimekizumab versus ustekinumab, the reported odds ratio is 6.056 and its 95% confidence interval is 3.874 to 9.466. Both intervals are on the odds-ratio scale. They do not express uncertainty in percentage points and should not be read as intervals for the arm-specific response percentages.

Both intervals lie above 1, the value corresponding to equal odds under the odds-ratio measure. The interval for the placebo comparison is broad in absolute terms. That breadth means the exact size of the odds ratio is less precisely estimated than the direction of the comparison; the small placebo response count, 4 among 83 participants, is part of the context for that uncertainty. The interval for the comparison with ustekinumab is narrower in absolute terms, though it still allows a range of possible odds ratios.

General statistical explanation: an interval does not say that every value inside it is equally plausible, and a 95% confidence procedure does not mean that there is a 95% probability that this particular fixed interval contains the true value. It is a method-based description of uncertainty. The interval also does not account for unreported design or analysis decisions, missing-data assumptions, or all forms of bias.

The arm percentages are descriptive estimates, while the formal confidence intervals accompany the registry's reported odds ratios. The posted record does not give confidence intervals for each arm percentage. Precision should therefore be judged using the intervals attached to the reported comparisons, alongside the group sizes and event counts, rather than treating rounded percentages as exact population values.

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 are planning calculations based on the observed effect, not the trial's original design justification. For the non-time-to-event outcomes shown, the figures mean participants per arm. They are not total participants across the trial. The registry's stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The original design assumptions are not reported in the registry record.

The 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting the selected effect under the stated assumptions, if that effect and assumptions were true. The 90% power column gives the corresponding per-arm planning figure for a design with a 90% chance under those same assumptions. These are prospective design concepts, not probabilities that this completed trial produced a correct conclusion.

The rows using 100%, 75% and 50% of the observed effect illustrate progressively smaller target effects. Smaller effects are harder to distinguish from random variation, so the planning illustrations require more participants per arm. The row based on the entire observed effect can give especially small figures when the observed contrast is large; it should not be treated as a recommended sample size without considering whether that effect is realistic for future planning.

These figures are not observed power. They are illustrations generated from the observed effect and the listed assumptions, not a retrospective measure of the chance that the trial succeeded. They also do not reveal how the sponsor designed the original trial; that justification is not reported in the registry record. The sample-size curve should be read in the same planning context, with its axes and scale interpreted according to the outcome and comparison displayed.

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 = 0.048; treatment proportion p₁ = 0.850; average p̄ = 0.449.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7035 = 1.3788.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4160 = 0.3501.
4. n = (1.3788 + 0.3501)² / (0.8023)² = 4.6, rounded up to 5 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 <- 0.0481927710843374                          # control proportion
p1 <- 0.850467289719626                          # 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(3370133)
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 = 5 (80% power), 6 (90% power)
 75% of observed effect: n per arm = 9 (80% power), 11 (90% power)
 50% of observed effect: n per arm = 18 (80% power), 23 (90% power)
power.prop.test: 4.6 
Simulated power at n = 5 per arm: 0.860 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The registry reports the formal primary comparisons as Cochran-Mantel-Haenszel odds ratios. For bimekizumab versus placebo, the reported odds ratio is 99.869, with a 95% confidence interval of 34.020 to 293.175 and p < 0.001. For bimekizumab versus ustekinumab, it is 6.056, with a 95% confidence interval of 3.874 to 9.466 and p < 0.001. Both are identified as superiority analyses.

The re-calculated estimates use a different measure: simple, unadjusted risk differences based on posted summary counts. A risk difference expresses an absolute difference in response proportions, while an odds ratio compares odds and the reported Cochran-Mantel-Haenszel analysis uses a stratified method. The values are therefore not expected to match numerically. Their intervals also arise from different estimators and should not be compared as if they described an identical quantity.

The registry record does not mark a discrepancy between its reported analysis and the re-calculation. The reported analysis is the trial's formal registered result, while the re-calculation is a transparent summary-statistic comparison. The registry does not report the detailed strata or full statistical plan needed to reproduce the formal analysis independently.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and numerous secondary outcomes, including outcomes at different weeks and with different response definitions. Each additional endpoint or comparison creates another opportunity for a chance finding. General statistical explanation: a multiplicity procedure can control the overall error rate across a defined family of tests, but it may make individual tests more stringent or otherwise alter how results are interpreted. The registry record does not report whether a multiplicity adjustment was planned or used, how the family of tests was defined, or whether any secondary analyses were designated as confirmatory.

The reported primary p-values are less than 0.001 for the two comparisons. These are the registry's reported values, but the record does not state whether or how the two comparisons were handled together for multiplicity. A small p-value does not resolve that reporting gap. The secondary outcome results should be interpreted cautiously because the multiplicity plan is not reported, and numerical results are absent for several outcomes listed in the registry.

The primary population is the Randomized Set, defined as all randomized participants. This definition identifies who was included, but it does not by itself specify the outcome handling for participants without a week 16 measurement, participants who discontinued, or participants with protocol deviations. The missing-data approach, imputation method, treatment of intercurrent events and any sensitivity analyses are not reported in the registry record.

The record describes quadruple masking and random allocation, but detailed allocation and concealment procedures are not reported. The registry identifies a Cochran-Mantel-Haenszel method for the formal primary comparisons, yet the stratification factors and complete analysis specification are not reported. Interim analyses and any related stopping rules are also not reported in the registry record. These absences limit independent assessment of how the analysis was prespecified and how robust the results are to alternative assumptions.

The registry's safety reporting uses several labeled populations, including initial-period groups and groups with any bimekizumab or any ustekinumab exposure. The precise membership and exposure windows for those analysis populations are not fully described in the record. This makes it important not to assume that the safety denominators correspond to the primary Randomized Set in the same way or cover equivalent observation time.

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: tendon injury (1.2%), toxicity to various agents (1.2%), oesophageal adenocarcinoma (1.2%), cardiac arrest (0.6%), macular hole (0.6%).

The serious adverse event figures are counts of participants affected, reported for distinct registry safety groups. In the initial-period groups, the registry records 2 affected participants among 83 in the placebo group, 5 among 321 in the bimekizumab group and 5 among 163 in the ustekinumab group. In the groups labeled Any Bimekizumab and Any Ustekinumab, the counts are 24 among 395 and 13 among 163, respectively. These are different group definitions; the figures should not be treated as direct comparisons over equal exposure periods without further information.

The record also lists deaths: 1 in each of the three initial-period groups, 2 in the Any Bimekizumab group and 1 in the Any Ustekinumab group. The registry record does not establish from these counts alone whether an event was caused by treatment. It does not provide enough detail to compare duration of exposure or follow-up across the safety groups.

Other adverse events are reported separately from serious adverse events. The registry counts affected participants, not necessarily the number of separate event episodes. Serious-event terms listed include tendon injury, toxicity to various agents, oesophageal adenocarcinoma, cardiac arrest, macular hole and haemorrhoids. These counts describe safety observations in the record and should not be interpreted as causal estimates.

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

The posted primary results show different week 16 PASI90 response proportions across the three randomized groups: 85.0% for bimekizumab, 49.7% for ustekinumab and 4.8% for placebo. The reported formal comparisons have 95% confidence intervals above 1 on the odds-ratio scale and p-values less than 0.001. The evidence in this record therefore supports a difference in the observed response patterns for the specified week 16 endpoint and comparisons.

That conclusion is limited to the endpoint, population, assessment time and analysis reported. PASI90 captures whether a participant crossed a particular improvement threshold; it does not describe all degrees of improvement or the full experience of people with psoriasis. The reported percentages do not show how long responses persisted, and the registry record does not provide numerical results for several listed secondary outcomes. The placebo group began bimekizumab after week 16, so the arm description does not support a continuing placebo-only comparison afterward.

Statistical significance is not the same as proof that every participant will respond, that the observed percentages will recur unchanged in another setting, or that one option is preferable for a particular person. Nor do the primary response results establish comparative safety. The adverse event counts are descriptive, arise from safety groups with different labels and denominators, and do not by themselves establish causation.

The confidence intervals describe uncertainty in the reported odds-ratio estimates under the specified method. They do not capture every limitation in design, reporting, missing data or generalizability. The registry record does not report the full statistical plan or sample-size justification, and it does not provide detail on missing outcome handling or multiplicity. These gaps should remain part of the interpretation.

In short, the results address a narrowly defined randomized comparison of week 16 PASI90 response. They do not answer every question about longer-term response, other symptoms, harms, or applicability beyond the trial population.

18. Why This Trial Matters Statistically

This trial is statistically informative because it includes both placebo and an active comparator, and reports the same primary binary endpoint across all three randomized groups. Readers can distinguish the observed response proportions from the formal odds-ratio analyses and see why those measures answer related but different questions. The concurrent comparisons also place the week 16 bimekizumab result in more than one comparative context.

The group sizes are not equal: the primary analysis includes 83 participants in placebo, 321 in bimekizumab and 163 in ustekinumab. Unequal group sizes affect the precision available for each comparison, and the relatively small placebo response count is relevant to the width of its odds-ratio interval. Presenting the interval alongside the large point estimate prevents the estimate from being read as exact.

The registry also illustrates the distinction between an absolute effect measure and a relative effect measure. The re-calculated risk difference summarizes a change in response proportion, while the reported formal analyses use odds ratios. A careful reading avoids treating numbers on these scales as interchangeable, or assuming that a small p-value conveys the clinical magnitude by itself.

Finally, the record makes clear why statistical interpretation extends beyond a significant primary result. There are multiple secondary outcomes, differing assessment times, a placebo-to-bimekizumab transition after week 16, and safety summaries using distinct analysis populations. The incompleteness of the reported multiplicity, missing-data and sample-size planning details places boundaries on what can be independently evaluated. These are not reasons to discard the results; they are reasons to state precisely which question the estimates answer.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The placebo group changes treatment after week 16, receiving bimekizumab through week 52. This limits placebo-based interpretation after the primary assessment. The main endpoint is a binary threshold, so it does not preserve the full range of PASI change or show how far participants moved beyond the response cutoff.

Group sizes differ, and the placebo group has 4 PASI90 responses among 83 participants. Estimates involving a low event count can be less precise, as reflected in the broad reported interval for the bimekizumab-versus-placebo odds ratio. The registry reports arm percentages rounded to one decimal place, while the formal comparisons use the Cochran-Mantel-Haenszel method; the detailed stratification information needed to reproduce that method is not reported.

The registry does not report the original sample-size justification, the full statistical analysis plan, a multiplicity procedure, missing-data methods, or interim analyses. Numerical results are not reported for several listed secondary outcomes. The safety results use distinct initial-period and any-exposure populations, and the registry does not fully describe their exposure windows or provide equivalent follow-up detail for a direct comparison.

The trial record describes a specific population of adults with moderate to severe chronic plaque psoriasis. The record does not provide enough information to assess how closely the enrolled participants represent other populations or care settings. The trial's recorded design and results should not be generalized beyond the population, endpoint and follow-up comparisons reported without additional evidence.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate the Efficacy and Safety of Bimekizumab Compared to Placebo and an Active Comparator in Adult Subjects With Moderate to Severe Chronic Plaque Psoriasis — NCT03370133.
• Registry record last updated 2026-04-15; status last verified 2026-04.
• Results first posted 2022-02-03.
• 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.

23. Record Summary

NCT03370133 is a completed phase 3, multicenter, randomized, double-blind study comparing bimekizumab with placebo and ustekinumab in adults with moderate to severe chronic plaque psoriasis. The trial enrolled 567 participants and used a parallel-group design. Its primary endpoint was PASI90 response at week 16 in the Randomized Set.

The registry reports response percentages of 85.0% for bimekizumab, 49.7% for ustekinumab and 4.8% for placebo. Formal Cochran-Mantel-Haenszel odds-ratio analyses report p < 0.001 for both primary comparisons involving bimekizumab. The registry also lists secondary outcomes and safety counts, with limitations in the available detail on multiplicity, missing-data handling, sample-size justification and safety-population definitions.