This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Study to Test the Efficacy, Safety and Pharmacokinetics of Bimekizumab in Subjects With Moderate to Severe Hidradenitis Suppurativa. (NCT03248531), a randomized phase 2 study. The primary outcome is Percentage of Participants Achieving Clinical Response as Measured by Hidradenitis Suppurativa Clinical Response (HiSCR) at Week 12.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03248531 |
| Phase | Phase 2 |
| Condition | Hidradenitis Suppurativa |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 90 |
| Arms in the primary outcome | 3 |
| Lead sponsor | UCB Biopharma SRL |
| Results posted | Yes |
2. Clinical Question
The clinical question was whether participants with moderate to severe hidradenitis suppurativa achieved a clinical response at Week 12 in the bimekizumab group, and how that response compared with placebo and adalimumab. The endpoint was the percentage achieving Hidradenitis Suppurativa Clinical Response, or HiSCR. Response required at least a 50 % reduction from baseline in total abscess and inflammatory nodule count, with no increase from baseline in abscess or draining fistula count.
The registry record identifies three groups: bimekizumab, adalimumab, and placebo. The primary result is reported for the Per-Protocol Set, with response rates and uncertainty intervals for each group, plus pairwise Bayesian estimates. The bimekizumab response estimate was 57.3, compared with 26.1 for placebo and 59.5 for adalimumab. These are estimated percentages of responders, not counts of participants.
The placebo comparison addresses how the bimekizumab response estimate contrasts with the placebo response estimate in this trial. The reported mean posterior difference for those groups was 31.2, with a 95% credible interval from 11.0 to 50.4. The comparison with adalimumab was reported as a mean posterior difference of -2.2, with a 60% credible interval from -11.2 to 6.6. These estimates describe the trial’s chosen analysis and uncertainty; they do not by themselves establish how results would generalize beyond the study participants.
The registry record describes hidradenitis suppurativa as a painful, long-term skin condition that causes abscesses and scarring. It does not report additional eligibility details beyond the stated condition and moderate to severe disease description.
3. Trial Design
This was a Phase 2, randomized, parallel-group, placebo-controlled study. The official title describes a multicenter study with investigator and subject blinding; the registry also classifies masking as quadruple. Enrollment was 90, and the record lists the study as completed. The recorded start date was 2017-09-22 and the primary completion date was 2018-11-23.
The three groups received bimekizumab, adalimumab, or placebo. The registry describes a bimekizumab loading dose followed by several applications at another dose level. It describes an adalimumab loading dose followed by several applications at two dose levels. The placebo group received several applications to maintain blinding. Dose amounts, administration schedules, and further treatment details are not reported in the registry record.
The primary efficacy result concerns the Per-Protocol Set, a subset of the Full Analysis Set. The registry defines this set as participants without important protocol deviations affecting the primary efficacy variable, as confirmed during a pre-analysis review before unblinding. The displayed primary group sizes in that analysis were 20.0 for placebo, 20.0 for adalimumab, and 44.0 for bimekizumab. These analysis-set counts should not be confused with overall enrollment: they refer to the stated efficacy population and do not describe every safety analysis population.
The registry record does not report the randomization procedure, allocation concealment details, or the number of study centers. It also does not report the planned sample-size justification, the detailed visit schedule, or the full protocol in the information summarized for this analysis. Such details should not be inferred from the trial title or from the observed group sizes.
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 Screening Period (≥ 2 weeks up to a maximum of 4 weeks prior to randomization), a Treatment Period (12 weeks), and a Safety Follow-Up (SFU) Visit (20 weeks after the last dose of investigational medicinal product (IMP)). Participant Flow refers to the Randomized Set.
Overall Study
| Stage | Placebo | Adalimumab | Bimekizumab |
|---|---|---|---|
| STARTED | 22 | 22 | 46 |
| Completed Week 12 | 19 | 18 | 42 |
| COMPLETED | 18 | 17 | 38 |
| NOT COMPLETED | 4 | 5 | 8 |
Reasons for not completing this period:
| Reason | Placebo | Adalimumab | Bimekizumab |
|---|---|---|---|
| Adverse Event | 0 | 0 | 1 |
| Lost to Follow-up | 1 | 0 | 5 |
| Withdrawal by Subject | 3 | 3 | 2 |
| Sponsor Request | 0 | 2 | 0 |
Analysis population for the primary outcome, as defined in the registry: The Per-Protocol Set (PPS) was a subset of the Full Analysis Set (FAS), consisting of those study participants who had no important protocol deviations affecting the primary efficacy variable, as confirmed during a pre-analysis review prior to unblinding of the data (at each of the interim and final analyses).
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) which consisted of all study participants randomized into the study.
| Characteristic | Unit | Placebo | Adalimumab | Bimekizumab | Total Title |
|---|---|---|---|---|---|
| Participants analysed | 22 | 22 | 46 | 90 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 2 | 2 |
| Age, Categorical: Between 18 and 65 years | Participants | 21 | 22 | 44 | 87 |
| Age, Categorical: >=65 years | Participants | 1 | 0 | 0 | 1 |
| Age, Continuous | years | 40.7 (SD 12.5) | 31.0 (SD 9.2) | 37.4 (SD 11.9) | 36.6 (SD 11.9) |
| Sex: Female, Male: Female | Participants | 15 | 18 | 30 | 63 |
| Sex: Female, Male: Male | Participants | 7 | 4 | 16 | 27 |
| Race/Ethnicity, Customized: American Indian or Alaskan Native | Participants | 1 | 0 | 0 | 1 |
| Race/Ethnicity, Customized: Asian | Participants | 1 | 3 | 0 | 4 |
| Race/Ethnicity, Customized: Black or African American | Participants | 6 | 5 | 10 | 21 |
| Race/Ethnicity, Customized: White | Participants | 12 | 14 | 35 | 61 |
| Race/Ethnicity, Customized: Other or Mixed | Participants | 2 | 0 | 1 | 3 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Participants Achieving Clinical Response as Measured by Hidradenitis Suppurativa Clinical Response (HiSCR) at Week 12, measured over Week 12.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Percentage of responders |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was the percentage of participants achieving HiSCR at Week 12. HiSCR combines change in abscess and inflammatory nodule counts with a condition on abscess and draining fistula counts. A participant had to reach at least a 50 % reduction from baseline in the combined abscess and inflammatory nodule count and have no increase from baseline in abscess or draining fistula count.
This kind of response endpoint turns several clinical findings into a binary outcome: each participant is classified as a responder or nonresponder under the stated definition. The group result is then expressed as a response percentage. In general, a composite response definition can make a trial endpoint easier to interpret than separate counts alone, while its meaning depends on the exact components and threshold chosen. Here, the registry gives those components and the Week 12 assessment point.
The endpoint was used to compare response rates across the randomized groups. Participants with missing Week 12 data were considered nonresponders in the primary analysis. That rule affects how missing outcomes contribute to the reported rates and is important when interpreting them. The registry reports posterior mean response rates and credible intervals for each group, rather than presenting only a single percentage.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
In the Per-Protocol Set, the posterior mean HiSCR response estimate at Week 12 was 57.3 for bimekizumab, with a 95% credible interval from 42.4 to 71.4. The corresponding estimate was 26.1 for placebo, with a 95% credible interval from 13.8 to 40.5, and 59.5 for adalimumab, with a 95% credible interval from 44.2 to 73.9. The intervals describe uncertainty around the group estimates under the stated Bayesian analysis. They are not ranges containing a fixed proportion of individual participant outcomes.
For bimekizumab compared with placebo, the registry reports a mean posterior difference of 31.2 and a 95% credible interval from 11.0 to 50.4. The reported posterior probability that the difference was greater than 0% was 99.8%. Taken together, these posted estimates place the posterior contrast in favor of a higher response percentage for bimekizumab than placebo within this analysis. They do not remove uncertainty about the size of that contrast, as indicated by the interval.
For the bimekizumab and adalimumab comparison, the reported mean posterior difference was -2.2, with a 60% credible interval from -11.2 to 6.6. The reported posterior probability that the difference was greater than 0% was 42.1%. This interval is explicitly a 60% interval, not a 95% interval. Its credibility level should be kept in view when comparing it with the 95% intervals reported for group response estimates or the placebo contrast. The available results do not establish equivalence or noninferiority between the treatments; the registry does not report such a conclusion or a corresponding decision criterion.
These findings are based on a Per-Protocol Set rather than the full randomized study population, and missing Week 12 data counted as nonresponse. The record does not provide a parallel primary result for another analysis population in the details. The estimates therefore answer the stated question for the defined analysis set and missing-data rule. They should not be interpreted as a complete account of all outcomes: the record lists many secondary outcomes, including plasma concentrations and adverse events, but those do not replace the primary efficacy result.
The response estimates are not a measure of how much each individual improved, nor do they describe duration of response beyond the stated assessment. The primary endpoint is specifically a Week 12 classification based on the HiSCR definition. Longer-term efficacy conclusions are not reported in this primary result.
9. Statistical Methods Explained
The primary analysis used Bayesian logistic regression. In general, logistic regression models the probability of a binary outcome, such as responder versus nonresponder, in relation to the groups being compared. A binomial distribution is appropriate for representing the number of responders among participants assessed in a group, because each participant contributes a binary response classification under the endpoint definition. The registry says that the responder counts were assumed to follow a binomial distribution.
What does a posterior mean response rate represent?
In a Bayesian analysis, the observed data update a probability distribution for quantities of interest, such as a group response rate or a contrast between groups. The posterior mean is the average of that updated distribution. It is a single summary of the estimated response rate, not a raw count and not necessarily the response percentage obtained by dividing a count by the displayed group size. The registry reports posterior mean rates for the three groups.
Bayesian results depend on the model and prior distributions as well as the observed data. The registry identifies the logistic regression model and binomial response assumption, but prior distributions and further model specifications are not reported in the registry record. Without those details, the reported estimates can be described, but the model cannot be fully reconstructed from the information recorded.
How should the credible intervals be read?
In general, a Bayesian credible interval is an interval containing a stated amount of posterior probability for the parameter, conditional on the model and data. The group response rates are accompanied by 95% credible intervals. For example, the bimekizumab group estimate is 57.3, with an interval from 42.4 to 71.4. This communicates both the central estimate and uncertainty under the Bayesian analysis. The interval is not a guarantee that the true value lies inside it, and it is not an interval for individual participant responses.
The pairwise results include a 95% credible interval for the bimekizumab-placebo mean posterior difference and a 60% credible interval for the bimekizumab-adalimumab difference. A credible interval’s stated probability level matters: intervals at different levels are not directly comparable as if they had the same coverage. The record does not explain why different levels were reported for these two contrasts. A 60% interval is less demanding as a probability coverage statement than a 95% interval, so its endpoints should not be read as if they were a 95% interval.
What do the posterior probabilities add?
The registry also reports the posterior probability that each contrast was greater than 0%. In general, this is the share of the posterior distribution for the contrast that lies on the positive side of zero. The reported values were 99.8% for the bimekizumab-placebo contrast and 42.1% for the bimekizumab-adalimumab contrast. These probabilities are model-based summaries, not frequentist p-values. The registry reports no p-value for these comparisons. A posterior probability does not state the probability that a treatment works, and it does not by itself determine clinical importance.
The pairwise contrast values must be read alongside the response rates and interval levels. The reported bimekizumab-placebo difference was 31.2, with a 95% credible interval from 11.0 to 50.4. The bimekizumab-adalimumab difference was -2.2, with a 60% credible interval from -11.2 to 6.6. The direction of a contrast is meaningful only in relation to the group ordering used by the analysis; the group estimates provide context for that ordering.
How were missing outcomes handled?
The primary analysis counted participants with missing Week 12 data as nonresponders. In general, this is a conservative-looking rule when missing participants might otherwise have responded, but it does not guarantee conservative estimates in every setting. Its impact depends on why data are missing and on the outcomes those participants would have had. The registry does not report the missing-data counts, reasons, or sensitivity analyses in the information, so the robustness of the result to alternative assumptions cannot be assessed from these details.
What do the planning illustrations mean?
General sample-size planning uses assumptions about the effect to detect, the significance level, allocation, and statistical approximation. The planning illustration associated with these results is based on the observed effect, uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It considers effects stated as fractions of the observed effect. Such planning values illustrate design implications under assumptions; they are not observed power and do not show that the trial was planned using those assumptions. The trial’s own sample-size justification is not reported in the registry record.
Because the primary outcome is binary, planning also depends on the baseline response rate and the chosen comparison. The listed planning assumptions alone do not establish which response rates or target effect were used in the original trial design. Those design details, including any formal power calculation, are not reported in the registry record. A planning illustration based on the observed effect is retrospective and should not be mistaken for an independent confirmation of the observed result.
Finally, model-based inference does not eliminate design and population limitations. Randomization supports comparison across assigned groups, but the reported primary estimates are from the Per-Protocol Set, which excludes participants with important protocol deviations affecting the primary variable. The registry’s definition and analysis rule should therefore accompany any interpretation of the Bayesian estimates.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Regression, Logistic | Mean posterior difference 31.2 | 11.0 to 50.4 | not reported |
| Regression, Logistic | Mean posterior difference -2.2 | -11.2 to 6.6 | not reported |
| Regression, Logistic | Pr [Diff>0%] (%) 99.8 | not reported | not reported |
| Regression, Logistic | Pr[Diff > 0%](%) 42.1 | not reported | not reported |
The registry’s primary analysis is a Bayesian logistic regression analysis, with posterior mean response rates and credible intervals. Its reported pairwise results are also Bayesian: the bimekizumab-placebo comparison is a mean posterior difference of 31.2 with a 95% credible interval, and the bimekizumab-adalimumab comparison is a mean posterior difference of -2.2 with a 60% credible interval. The record additionally reports posterior probabilities for each difference being greater than 0%.
No separate recalculated primary comparison is reported in the registry record, so there is no second estimate to set against the sponsor’s analysis. The posted sponsor analysis should therefore be read on its own terms, including its model, analysis population, missing-data rule, and differing interval levels. The registry does not report a disagreement between the sponsor’s result and an independent re-analysis. Nor does it provide enough model detail, such as prior distributions, to reproduce the Bayesian analysis in full from the record alone.
The pairwise estimate and its interval are more informative together than either alone: the mean difference gives a central summary, while the interval describes uncertainty under the model. The posterior probability is a complementary summary of the distribution’s position relative to zero, not a replacement for the interval or a p-value.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists 42 outcome measures and four analyses. The primary endpoint is one part of a broader set of efficacy, pharmacokinetic, and safety outcomes. When many outcomes, groups, time points, or contrasts are examined, the chance of at least one apparently notable finding can increase. In general, multiplicity procedures address that issue by controlling error rates or by defining a hierarchy of hypotheses. The registry record does not report a multiplicity adjustment, a testing hierarchy, or a plan for interpreting the primary result alongside all secondary outcomes.
The absence of a reported multiplicity plan does not show that no plan existed; it means the relevant detail is not reported in the registry record. The pairwise posterior probabilities should be understood as the reported model-based quantities, not as multiplicity-adjusted probabilities unless the record says so. No such adjustment is described in the information recorded for these results.
Missing Week 12 observations were classified as nonresponders for the primary analysis. This rule is explicit and applies to the primary endpoint. However, the number of missing outcomes and reasons for missingness are not reported in the registry record. The record also does not report a sensitivity analysis using alternative missing-data assumptions. As a result, readers cannot assess from these details whether different handling of missing outcomes would materially alter the group estimates or contrasts.
The primary result uses the Per-Protocol Set, defined as a subset of the Full Analysis Set that excluded participants with important protocol deviations affecting the primary efficacy variable after a pre-analysis review. The record does not report the corresponding primary estimates for the Full Analysis Set in the results. A per-protocol analysis may be informative about outcomes among participants meeting the stated protocol criteria, but it is not identical to an analysis retaining all participants in the Full Analysis Set. The details available do not permit a direct comparison of those populations.
Interim analyses, stopping boundaries, and any adjustments related to interim looks are not reported in the registry record. Their absence from the record should not be filled in by assumption.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Placebo (SS) | 2 | 21 | 9.5% |
| Adalimumab (SS) | 1 | 21 | 4.8% |
| Bimekizumab (SS) | 2 | 46 | 4.3% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Placebo (SS) | 0 of 21 |
| Adalimumab (SS) | 0 of 21 |
| Bimekizumab (SS) | 0 of 46 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Placebo (SS) | 5 of 21 |
| Adalimumab (SS) | 12 of 21 |
| Bimekizumab (SS) | 21 of 46 |
Most frequent serious adverse event terms
| Event term | Organ system | Placebo (SS) | Adalimumab (SS) | Bimekizumab (SS) |
|---|---|---|---|---|
| Anaemia | Blood and lymphatic system disorders | 0 of 21 | 0 of 21 | 1 of 46 |
| Myocardial infarction | Cardiac disorders | 1 of 21 | 0 of 21 | 0 of 46 |
| Empyema | Infections and infestations | 0 of 21 | 0 of 21 | 1 of 46 |
| Headache | Nervous system disorders | 1 of 21 | 0 of 21 | 0 of 46 |
| Dizziness | Nervous system disorders | 1 of 21 | 0 of 21 | 0 of 46 |
| Hypoaesthesia | Nervous system disorders | 1 of 21 | 0 of 21 | 0 of 46 |
| Hidradenitis | Skin and subcutaneous tissue disorders | 0 of 21 | 1 of 21 | 0 of 46 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo (SS) | Adalimumab (SS) | Bimekizumab (SS) |
|---|---|---|---|---|
| Hidradenitis | Skin and subcutaneous tissue disorders | 3 of 21 | 6 of 21 | 8 of 46 |
| Fatigue | General disorders | 0 of 21 | 2 of 21 | 4 of 46 |
| Headache | Nervous system disorders | 3 of 21 | 0 of 21 | 3 of 46 |
| Diarrhoea | Gastrointestinal disorders | 0 of 21 | 1 of 21 | 3 of 46 |
| Nausea | Gastrointestinal disorders | 0 of 21 | 1 of 21 | 3 of 46 |
| Injection site pruritus | General disorders | 0 of 21 | 2 of 21 | 2 of 46 |
| Oral candidiasis | Infections and infestations | 0 of 21 | 1 of 21 | 3 of 46 |
| Nasopharyngitis | Infections and infestations | 0 of 21 | 1 of 21 | 3 of 46 |
| Upper respiratory tract infection | Infections and infestations | 0 of 21 | 2 of 21 | 2 of 46 |
| Arthralgia | Musculoskeletal and connective tissue disorders | 1 of 21 | 2 of 21 | 1 of 46 |
Serious events with the highest arm-level rate: myocardial infarction (4.8%), headache (4.8%), dizziness (4.8%), hypoaesthesia (4.8%), hidradenitis (4.8%).
The safety results use the Safety Set, which is distinct from the Per-Protocol Set used for the primary efficacy result. Serious adverse events were reported for 2 participants among 21 in the placebo group, 1 among 21 in the adalimumab group, and 2 among 46 in the bimekizumab group. The record reports 0 deaths in each group. These are counts with the stated at-risk group sizes; no additional calculation is needed to understand them.
For other adverse events, the registry reports 5 participants among 21 in the placebo group, 12 among 21 in the adalimumab group, and 21 among 46 in the bimekizumab group. These totals should not be interpreted as counts of distinct events, because a participant may experience more than one event and the record’s summary does not describe event-level overlap. The listed adverse-event categories and event terms also should not be treated as interchangeable with the serious adverse event counts.
The record lists myocardial infarction, headache, dizziness, hypoaesthesia, hidradenitis, and anaemia among the leading serious-event terms, with percentages of 4.8%, 4.8%, 4.8%, 4.8%, 4.8%, and 2.2%, respectively. It does not identify those percentages by treatment group in the available summary, and the listed terms do not establish a causal relationship to study treatment. The safety counts are descriptive and should be interpreted in light of the study’s size and reporting definitions. Further detail about event timing, duration, relatedness, and participant-level overlap is not reported in the registry record.
13. What the Result Does, and Does Not, Mean
The primary result describes Week 12 HiSCR response estimates in the Per-Protocol Set under a specific endpoint definition and missing-data rule. The bimekizumab estimate was 57.3, while the placebo estimate was 26.1 and the adalimumab estimate was 59.5. The Bayesian contrast with placebo was reported as 31.2, with a 95% credible interval from 11.0 to 50.4. The contrast with adalimumab was -2.2, with a 60% credible interval from -11.2 to 6.6. These are the results of this trial’s stated analysis, not guarantees about outcomes for every participant or every setting.
The response endpoint does not describe all aspects of hidradenitis suppurativa, and a Week 12 responder classification does not indicate how long a response lasted. It also does not quantify each participant’s symptom experience, quality of life, or longer-term course. Those outcomes cannot be inferred from the primary response percentage. Although the registry lists secondary outcomes concerning drug concentrations and adverse events, the primary result does not answer every safety, pharmacokinetic, or clinical question.
The estimates do not establish that one treatment is equivalent to another, or that one is superior for all relevant outcomes. The registry record does not report an equivalence or noninferiority margin, and the bimekizumab-adalimumab interval is reported at a 60% credibility level. The posterior probability for a difference above zero is a model-based summary and should not be translated into a blanket treatment claim.
Missing Week 12 data were treated as nonresponse, and the primary analysis population excluded participants with important protocol deviations affecting the endpoint. Those choices define what the reported estimates mean. The number and reasons for missing outcomes, and results under alternative assumptions, are not reported in the registry record. These limitations make it important to keep the result tied to the recorded population, time point, and analysis rather than generalizing beyond them.
In general, statistical uncertainty describes uncertainty about an estimated group quantity, while clinical importance depends on the relevance of the endpoint and the size of a difference in context. The registry provides the response estimates and their uncertainty summaries, but it does not state a threshold for a clinically important difference. The posted statistics alone therefore do not settle that judgment.
14. Why This Trial Matters Statistically
This trial is statistically instructive because it reports a binary clinical response endpoint through a Bayesian model rather than only as a raw proportion. The HiSCR definition combines a specified reduction in abscess and inflammatory nodule count with a no-increase condition for abscess or draining fistula count. That makes the construction of the endpoint central to interpreting the response percentages: a responder is someone meeting the entire stated rule.
The report also puts several kinds of Bayesian summaries side by side. Group posterior mean response rates have 95% credible intervals, while the reported pairwise comparison with adalimumab has a 60% credible interval. The posterior probability of a difference greater than zero offers another perspective on the contrast. Readers need to distinguish these quantities rather than treating all intervals or probabilities as the same kind of evidence.
The study highlights how analysis population and missing-data rules shape a result. The primary efficacy result is for the Per-Protocol Set, and missing Week 12 observations count as nonresponse. Those are not minor labels: they specify which participants contribute to the analysis and how incomplete outcomes are classified. The Safety Set has different group sizes, illustrating why efficacy and safety denominators must not be mixed.
Finally, the results show the value of reading the reported comparison alongside both group estimates and uncertainty. A central contrast alone can conceal how uncertain it is, while a posterior probability alone does not communicate the range of plausible contrasts. Interpreting the complete set of reported summaries requires attention to the endpoint, model, population, missing-data rule, and interval level.
15. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The registry record does not report the prior distributions or enough additional model specifications to reproduce the Bayesian logistic regression. It also does not explain why the bimekizumab-adalimumab comparison is accompanied by a 60% credible interval while group estimates and the placebo comparison use 95% intervals. That difference in interval level limits direct visual comparison of interval widths or endpoints.
The primary result is presented for the Per-Protocol Set. The registry defines the set in relation to important protocol deviations affecting the primary efficacy variable, but the corresponding Full Analysis Set result is not reported in the registry record. The record also does not report the count or reasons for missing Week 12 values, or sensitivity analyses that would show how the nonresponder rule performs under other assumptions.
The registry reports a broad collection of outcome measures, but a multiplicity plan is not reported in the registry record. A sample-size justification for the trial is also not reported. The planning illustration based on the observed effect is not a substitute for the original design calculation and should not be interpreted as observed power.
For safety, the summary gives counts of participants with serious and other adverse events and lists several serious-event terms. Details about event attribution, timing, duration, severity beyond the serious classification, and overlap among event categories are not reported in the registry record. The group summaries therefore support a descriptive reading, not a detailed comparative safety analysis.
The registry record does not report the randomization procedure, allocation concealment, detailed dose amounts, or additional eligibility criteria in the information used for these results. These omissions restrict assessment of design and generalizability. They are limitations of the available record, not evidence that those details were absent from the study documentation.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: A Study to Test the Efficacy, Safety and Pharmacokinetics of Bimekizumab in Subjects With Moderate to Severe Hidradenitis Suppurativa. — NCT03248531.
- Registry record last updated 2022-04-11; status last verified 2022-01.
- Results first posted 2022-02-09.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT03248531 is a completed Phase 2 randomized, parallel-group, placebo-controlled study of bimekizumab in participants with moderate to severe hidradenitis suppurativa. The registry identifies bimekizumab, adalimumab, and placebo groups and describes investigator and subject blinding, with masking classified as quadruple.
The primary endpoint was the percentage achieving HiSCR at Week 12. The analysis used a Bayesian logistic regression model in the Per-Protocol Set; missing Week 12 outcomes were counted as nonresponse. The posterior mean response estimates were 57.3 for bimekizumab, 59.5 for adalimumab, and 26.1 for placebo. Pairwise estimates and credible intervals are reported for bimekizumab versus placebo and bimekizumab versus adalimumab.
The registry also reports safety summaries, including serious adverse event counts and deaths, and lists pharmacokinetic and other secondary outcomes. Details about the multiplicity plan, trial sample-size justification, and several model specifications are not reported in the registry record.