BI 655066 (Risankizumab) Compared to Placebo and Active Comparator (Ustekinumab) in Patients With Moderate to Severe Chronic Plaque Psoriasis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for BI 655066 (Risankizumab) Compared to Placebo and Active Comparator (Ustekinumab) in Patients With Moderate to Severe Chronic Plaque Psoriasis (NCT02684370), a randomized phase 3 study. The primary outcome is Percentage of Participants Achieving 90% Improvement in Psoriasis Area and Severity Index (PASI) Score (PASI90) at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A).

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, triple-masked, parallel-group trial. The registry record lists enrollment of 560 and describes the study population as adults with moderate to severe chronic plaque psoriasis. The study status is completed. The start and primary completion entries are 2016-02 and 2016-12, respectively.

Part A included placebo, risankizumab, and ustekinumab groups. Participants assigned to risankizumab received 150 mg by subcutaneous injection at Weeks 0 and 4. The placebo group received matching placebo injections at those weeks. Participants assigned to ustekinumab received 45 mg or 90 mg, based on screening weight, at Weeks 0 and 4. The intervention descriptions also identify separate placebos for risankizumab and ustekinumab, but further details about how those placebos were used are not reported in the registry record.

The primary comparison was risankizumab versus placebo at Week 16. The posted primary endpoint denominators are 304 for risankizumab and 102 for placebo. The safety summary also reports 100 participants at risk in the Part A ustekinumab group. These endpoint and safety denominators should not be treated as a breakdown of the full enrollment unless the registry record explicitly makes that connection.

The randomized, masked, parallel design provides a framework for comparing groups while limiting some sources of bias. The record does not report the details of sequence generation, allocation concealment, masking procedures, or the original sample-size justification. Further follow-up and treatment procedures beyond the descriptions above are not reported in the registry record.

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: A total of 560 subjects were enrolled; 54 subjects failed screening and are excluded from the analyses.

Part A

Reasons for not completing this period:

Part B

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT population

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: Intent-to-treat (ITT) population: All participants who were randomized at Week 0.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Achieving 90% Improvement in Psoriasis Area and Severity Index (PASI) Score (PASI90) at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A), measured over Week 16.

The primary endpoint is the percentage of participants achieving PASI90 at Week 16. PASI is a composite score based on the affected body surface area and the severity and extent of redness, thickness, and scaling of lesions. Each of these signs is scored on a scale from 0, meaning no symptoms, to 4, meaning very marked. The total PASI score ranges from 0 to 72, with 0 indicating no psoriasis and 72 indicating very severe psoriasis.

PASI90 is defined as at least a 90% reduction in PASI score compared with baseline. The endpoint therefore expresses how many participants reached a specified level of improvement by the assessment time; it is not the average change in PASI score. A percentage is useful for a binary response because it makes the proportion meeting the threshold visible in each group and supports a direct comparison between groups.

The endpoint captures a threshold response on the PASI measure, rather than every feature of disease or a participant’s experience. The registry record also lists secondary outcomes involving other PASI thresholds, a physician global assessment, quality of life, and symptoms. Those measures add different perspectives, but they do not change the definition of the primary endpoint. Nonresponder imputation was used for missing primary-endpoint data, as stated in the endpoint description.

7. Results Posted in the Registry

• Risankizumab (Part A): 75.3% (229/304) versus 4.9% (5/102) on control; risk difference 70.4 percentage points (95% CI 64.0 to 76.8), risk ratio 15.37 (6.52–36.21); 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 Achieving sPGA Score of Clear at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A)

• 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:
  • Risankizumab (Part A): 36.8% (112/304) versus 2.0% (2/102) on control; risk difference 34.9 percentage points (95% CI 28.8 to 40.9), risk ratio 18.79 (4.73–74.70); 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 Risankizumab (Part A) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Achieving PASI100 at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A)

• 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:
  • Risankizumab (Part A): 35.9% (109/304) versus 0.0% (0/102) on control; risk difference 35.9 percentage points (95% CI 30.5 to 41.2); 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 Risankizumab (Part A) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Achieving Dermatology Life Quality Index (DLQI) Score of 0 or 1 at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A)

• 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:
  • Risankizumab (Part A): 65.8% (200/304) versus 7.8% (8/102) on control; risk difference 57.9 percentage points (95% CI 50.5 to 65.4), risk ratio 8.39 (4.29–16.39); 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 Risankizumab (Part A) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes extend the primary question across several related dimensions: other PASI response thresholds, physician-rated global status, skin-related quality of life, and psoriasis symptoms. The record lists secondary comparisons against placebo and ustekinumab, and reports 15 secondary outcomes. Taken together, the endpoint set shows that the study assessed more than the primary PASI90 threshold alone.

The secondary results shown in the tables have a generally consistent pattern for the reported placebo comparisons: response percentages were higher in the risankizumab group on the listed outcomes, and the sponsor analyses report small p-values. This pattern is directionally compatible with the primary PASI90 result. The endpoints nevertheless measure different constructs or use different response thresholds, so they should not be treated as repeated measurements of precisely the same outcome. Nor should the primary result be used to infer the result of every secondary comparison, including those with ustekinumab.

Secondary outcomes usually carry less confirmatory weight when many outcomes are examined and no multiplicity adjustment is reported. Each additional test creates another opportunity for a result to appear statistically notable by chance. The registry record does not report a multiplicity procedure, such as a prespecified hierarchy or adjustment across the family of secondary tests. The p-values should therefore be interpreted as posted results for individual outcomes rather than as a single confirmatory family of tests.

The record reports 17 outcome measures and 17 analyses, with 15 secondary outcomes posted. The number of analyses alone does not reveal how the outcomes were prioritized, whether all were prespecified in the same way, or whether there were unposted analyses. Those details are not reported in the registry record. The tables are the place to examine each endpoint’s figures; the overall pattern is informative context, but it does not replace the primary analysis or resolve multiplicity concerns.

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

• Percentage of Participants Achieving sPGA Score of Clear at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A)
• Percentage of Participants Achieving PASI100 at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A)
• Percentage of Participants Achieving Dermatology Life Quality Index (DLQI) Score of 0 or 1 at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A)
• Percentage of Participants Achieving Psoriasis Symptoms Scale (PSS) Total Score of 0 at Week 16 in Participants Who Received Risankizumab Compared With Placebo (Part A)
• Percentage of Participants Achieving PASI90 at Week 16 in Participants Who Received Risankizumab Compared With Ustekinumab (Part A)
• Percentage of Participants Achieving sPGA Score of Clear or Almost Clear at Week 16 in Participants Who Received Risankizumab Compared With Ustekinumab (Part A)
• Percentage of Participants Achieving PASI100 at Week 16 in Participants Who Received Risankizumab Compared With Ustekinumab (Part A)
• Percentage of Participants Achieving sPGA Score of Clear at Week 16 in Participants Who Received Risankizumab Compared With Ustekinumab (Part A)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is binary for analysis: each participant either achieved PASI90 by Week 16 or did not. The registry reports the percentage in each group and a sponsor analysis of the difference in percentages. For a binary endpoint, comparing percentages gives a direct summary of the contrast in response frequency. It does not show the distribution of PASI changes among participants who did not meet the threshold or among those who did.

What the reported analysis does

The sponsor used a Cochran-Mantel-Haenszel analysis and reported an adjusted difference in percentage. In general, the Cochran-Mantel-Haenszel method combines information across strata to estimate an association or group contrast while accounting for stratification. The registry record does not identify the strata or explain the adjustment variables, so the particular adjustment cannot be reconstructed from this information. The reported adjusted estimate is therefore the appropriate description of the sponsor’s own analysis; the simple calculation shown separately is unadjusted.

For an unadjusted risk difference, the response proportion in one group is compared with the response proportion in the other. Here the simple re-calculation uses the posted event counts and denominators: 229 of 304 for risankizumab and 5 of 102 for placebo. The estimate and interval are shown in the statistical results. This is a descriptive comparison based on the posted summaries and does not reproduce the sponsor’s adjustment. The registry record identifies no discrepancy between the recalculation and sponsor analysis, but the methods answer slightly different statistical questions because one is unadjusted and the other is adjusted.

Confidence intervals and hypothesis tests

In general, a confidence interval gives a range of estimates compatible with the data and the method used, subject to the method’s assumptions. It conveys precision as well as direction: a narrower interval indicates less statistical uncertainty in the estimate than a wider interval on the same scale. It is not a range containing a fixed proportion of individual participant outcomes, and it does not state the probability that the true effect lies in the observed interval.

The sponsor reported a 95% confidence interval for the adjusted difference. The posted p-value is <0.001. In general, a p-value is calculated relative to a null hypothesis and a statistical model; it is not the probability that the null hypothesis is true, nor does it quantify clinical importance. The p-value should be read alongside the effect estimate, its interval, the endpoint definition, and the study design.

Missing outcomes and the analysis population

The primary endpoint description specifies the intention-to-treat population and nonresponder imputation for missing data. Intention-to-treat generally refers to analyzing participants according to randomized group, regardless of subsequent treatment received, though the registry description does not provide further operational details. Nonresponder imputation assigns missing endpoint values to the nonresponse category. This provides a defined rule for the primary analysis but relies on that chosen rule for participants whose status is missing.

The record does not report a sensitivity analysis for alternative missing-data assumptions in the information. It also does not identify additional analysis populations for the primary endpoint. Missing-data methods for secondary endpoints are not reported unless explicitly stated in their endpoint descriptions.

Why variability matters differently for this outcome

This endpoint is a percentage of participants with a response, rather than a continuous measurement such as an average score. The uncertainty in a percentage depends on the number of participants and the observed response pattern. In general, estimates based on fewer observations tend to be less precise, and binary outcomes can have uncertainty that changes with the response frequency. The posted confidence interval captures uncertainty according to its analysis method; it does not remove uncertainty about generalizing beyond the study.

Over-dispersion is a general concern when observed binary outcomes vary more than a model assumes. If present, it can make standard errors too small and confidence intervals too narrow under an overly simple model. The registry record does not report an over-dispersion assessment for this endpoint, so it is not possible to say whether it was evaluated or whether a correction was used. This is a general methodological consideration, not a reported finding about this trial.

The methods information should not be expanded into assumptions about interim analyses, covariate adjustment, multiplicity control, or other statistical procedures. Those details are not reported in the registry record. The named analysis method, the stated analysis population, the imputation rule, and the posted estimates are the limits of what can be described confidently.

11. Confidence Intervals and Precision

The sponsor’s adjusted estimate for the primary comparison is a 70.3 difference in percentage, with a 95% confidence interval from 64.0 to 76.7. This interval is the main reported measure of precision for the sponsor analysis. Its endpoints are both above zero, which is consistent with a positive adjusted difference under the reported method. The interval also makes clear that the point estimate should not be treated as exact: a range of values remains compatible with the analysis.

The simple unadjusted re-calculation gives a risk difference of 0.704, with a confidence interval from 0.640 to 0.768. This estimate is on the risk-difference scale used by the recalculation, whereas the sponsor reports an adjusted difference in percentage. The numerical scales should therefore be read in the context of the analysis that produced them, not treated as a single interchangeable output. The sponsor analysis is adjusted; the recalculation is not.

In general, a 95% confidence interval describes the long-run coverage properties of the method when its assumptions hold. It does not mean there is a 95% probability that the true difference is inside this particular interval. Nor does it account for all sources of uncertainty, such as uncertainty about the suitability of the endpoint, missing-data assumptions, or generalizability to people outside the trial population.

The p-value of <0.001 is not a substitute for the interval. It conveys evidence against a null hypothesis under the analysis model, while the interval shows a range of effect estimates. Both are most informative when considered with the response percentages and study design. The registry record does not report alternative interval methods or sensitivity intervals here, so the posted interval is the available description of statistical precision.

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 display is a planning illustration derived from the observed effect, not a record of the trial’s original design calculation. Because the primary outcome is a binary percentage rather than a time-to-event outcome, its figures are participants per arm, not total events. The curve should be read as a visual presentation of the same planning relationship, not as an additional observed study result.

The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at a stated fraction of the observed effect. The table shows planning for the full observed effect and for smaller fractions of that effect. Smaller effects need more participants to detect under the same assumptions, because the groups’ expected response rates are closer together.

The 80% power column describes the number of participants per arm indicated by the calculation for an 80% chance of detecting the specified effect under the planning assumptions, if that effect were the true effect and the assumptions held. The 90% power column gives the corresponding planning quantity for a 90% chance. These values are prospective illustrations, not outcomes measured in the trial.

This is not observed power. It does not report the probability that the completed study found a result, and it should not be used to reinterpret the primary p-value. The original trial’s sample-size justification and its design assumptions are not reported in the registry record. The displayed planning figures instead use the posted observed effect and the assumptions stated for the illustration.

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 = 0.049; treatment proportion p₁ = 0.753; average p̄ = 0.401.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6932 = 1.3586.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4821 = 0.4058.
4. n = (1.3586 + 0.4058)² / (0.7043)² = 6.3, rounded up to 7 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 <- 0.0490196078431373                          # control proportion
p1 <- 0.753289473684211                          # 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(2684370)
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 = 7 (80% power), 8 (90% power)
 75% of observed effect: n per arm = 11 (80% power), 15 (90% power)
 50% of observed effect: n per arm = 21 (80% power), 28 (90% power)
power.prop.test: 6.3 
Simulated power at n = 7 per arm: 0.870 (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 sponsor’s reported primary analysis used the Cochran-Mantel-Haenszel method and estimated an adjusted difference in percentage of 70.3, with a 95% confidence interval from 64.0 to 76.7 and p < 0.001. This is the trial’s reported analysis. It should be distinguished from the simple unadjusted re-calculation based on the posted response counts.

The recalculation uses 229 responses among 304 participants in the risankizumab group and 5 among 102 in the placebo group. It gives a risk difference of 0.704, with an interval from 0.640 to 0.768. Since this calculation does not apply the sponsor’s Cochran-Mantel-Haenszel adjustment, it is a check based on the marginal summary data, not a reproduction of the sponsor method. The registry record does not report the strata or other details needed to reproduce that adjustment.

The two analyses show a similar direction and broadly similar magnitude when their scales are understood, but they are not identical estimands or methods. The record does not flag a discrepancy. In general, when a discrepancy is flagged between a sponsor analysis and a recalculation, the sponsor’s own analysis takes precedence; that situation is not indicated. The sponsor result remains the primary interpretation because it is the posted trial analysis, while the recalculation is useful as a transparent unadjusted comparison.

15. Multiplicity, Missing Data and Analysis Populations

The registry record identifies the primary outcome and reports a collection of secondary outcomes. It lists 17 outcome measures and 17 analyses, and 15 secondary outcomes are posted. Examining multiple outcomes can increase the chance of obtaining at least one apparently notable result even when no true effect exists for any of them. Statistical approaches such as family-wise error control, a testing hierarchy, or false-discovery control can address particular multiplicity goals, but no such approach is reported in the registry information.

It is therefore not possible to determine whether the secondary analyses were adjusted for multiplicity, arranged in a prespecified testing sequence, or treated as exploratory. Their individual reported p-values should not automatically be interpreted as if each were the sole primary test. The primary result is clearly identified, but the full statistical plan and any multiplicity strategy are not reported in the registry record.

For the primary endpoint, the registry specifies the intention-to-treat population and nonresponder imputation for missing data. This means missing primary outcomes were handled as nonresponses under the stated rule. The record does not provide, in the information, the number of missing primary outcomes, the timing or reasons for missingness, or sensitivity analyses using alternative assumptions. The reported percentages should consequently be read in light of the specified imputation approach.

Missing-data methods for the secondary endpoints are not reported unless specified in an individual endpoint description. The record also does not report details of the statistical analysis plan, including any interim analyses, model covariates, or prespecified subgroup analyses. These omissions do not invalidate the posted result, but they limit independent assessment of how all analytic decisions were made and how robust the findings are to alternative choices.

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: cholelithiasis (1.0%), cellulitis (1.0%), osteomyelitis (1.0%), wound complication (1.0%), basal cell carcinoma (1.0%).

The safety summary reports serious adverse events for the Part A groups as counts among participants at risk: 3 of 102 in the placebo group, 8 of 100 in the ustekinumab group, and 7 of 304 in the risankizumab group. No deaths are listed for these groups. The record also includes Part B safety groups, but the primary efficacy comparison is the Part A risankizumab-versus-placebo comparison.

These counts describe reported events in the registry summary; they do not by themselves establish that a treatment caused an event or that groups had equivalent safety. The group sizes differ, and the registry information does not provide the event timing, exposure-adjusted rates, or details needed to assess individual cases. The record lists serious-event terms including cholelithiasis, cellulitis, osteomyelitis, wound complication, basal cell carcinoma, and squamous cell carcinoma of skin, each shown as 1.0%. The summary does not establish a group-specific attribution for those terms.

Other adverse events are listed separately in the safety data. The serious-event figures should be read alongside the full registry safety record and the differing denominators, not as a complete account of all possible harms. The registry record does not report further clinical interpretation.

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

The posted primary result shows that a larger percentage of participants assigned to risankizumab met the PASI90 threshold at Week 16 than participants assigned to placebo. The sponsor’s adjusted difference estimate and its confidence interval quantify that comparison under the reported method. The simple unadjusted recalculation points in the same direction. Together, the posted results describe a clear between-group difference for this endpoint and assessment time.

This result is specific to PASI90, Week 16, the intention-to-treat population, and the primary comparison with placebo. It does not say that every participant responded, or that the same result would occur for every individual. It also does not describe average PASI change, long-term outcomes, or the full range of participant experiences. Other endpoints in the registry address some additional dimensions, but each needs its own interpretation.

The primary result is not a direct comparison of risankizumab with ustekinumab. Although an ustekinumab group was included, conclusions about that active-comparator contrast require its own reported results. Nor does a difference on the primary efficacy endpoint alone settle questions about safety. The registry safety summary reports adverse-event counts, but those counts are not a substitute for a full assessment of benefits and harms.

Statistical evidence and clinical importance are related but distinct. A small p-value does not quantify how important an outcome is to participants, and the endpoint’s threshold does not capture all aspects of disease. The registry record does not report a threshold for judging clinical importance or enough information to evaluate how participants valued the observed difference.

Finally, the record does not support claims about outcomes beyond those reported, treatment decisions for a particular person, or results in populations not represented in the study description. The result should be understood as a group-level finding from the randomized comparison at the specified time point, with uncertainty and analytic limitations.

18. Why This Trial Matters Statistically

This trial illustrates why the definition of an endpoint matters. PASI90 is a threshold response: it classifies participants by whether they achieved at least a specified reduction from baseline. A response percentage is readily interpretable, but it compresses a range of underlying score changes into two categories. Readers should keep the threshold and timing visible when interpreting the group contrast.

It also provides a practical example of distinguishing an adjusted sponsor analysis from an unadjusted calculation based on summary counts. The reported Cochran-Mantel-Haenszel result is the sponsor’s analysis, while the recalculation offers a transparent comparison using the posted margins. They address related but not identical analyses. Recognizing that distinction helps prevent a numerical check from being mistaken for an exact replication of the trial’s statistical method.

The result also shows how effect estimates, confidence intervals, and p-values provide complementary information. The estimate describes the size and direction of the difference, the interval conveys statistical precision, and the p-value summarizes evidence against a null hypothesis under the model. None of these alone answers every question about clinical relevance, safety, missing outcomes, or generalizability.

Finally, the set of secondary outcomes raises familiar issues in clinical-trial interpretation. Multiple measures can broaden understanding of a study, but their results need context about prespecification and multiplicity. Where that context is not reported, a coherent pattern across outcomes can be noted without treating every secondary p-value as independent confirmation. This is a useful statistical lesson from the structure of the posted record, rather than a claim beyond its reported results.

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 registry information leaves important design and analysis details unreported. The original sample-size justification, allocation procedures, masking procedures, and full statistical analysis plan are not described. The listed planning illustration is based on the observed effect and should not be confused with the trial’s original design calculation.

The primary endpoint is assessed at Week 16. The record does not establish how results changed at other times or whether the Week 16 comparison represents longer-term outcomes. It also does not provide enough information to describe treatment exposure after the stated Part A injections or the details of any later treatment transitions.

The primary missing-data rule is specified as nonresponder imputation, but the number and reasons for missing outcomes and sensitivity analyses are not reported. Multiplicity procedures for the collection of secondary outcomes are also not reported. These gaps limit assessment of how robust secondary findings are and how the primary analysis might change under alternative missing-data assumptions.

The sponsor’s adjusted result cannot be independently reconstructed from the summarized details because the strata or adjustment details are not reported. The available recalculation is unadjusted and therefore is not a substitute for the sponsor’s method. The safety summary provides event counts, but the information does not include event timing, exposure-adjusted comparisons, or a full account of harms. These limitations define what can be concluded from the registry results.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: BI 655066 (Risankizumab) Compared to Placebo and Active Comparator (Ustekinumab) in Patients With Moderate to Severe Chronic Plaque Psoriasis — NCT02684370.
• Registry record last updated 2021-07-30; status last verified 2021-07.
• Results first posted 2019-06-18.
• 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

The registry record describes a completed Phase 3, randomized, triple-masked, parallel-group study of risankizumab, placebo, and ustekinumab in adults with moderate to severe chronic plaque psoriasis. The primary question was whether the percentage achieving PASI90 at Week 16 differed between risankizumab and placebo.

The posted response percentages were 75.3% for risankizumab and 4.9% for placebo. The sponsor’s adjusted analysis reported a difference in percentage of 70.3, with a 95% confidence interval from 64.0 to 76.7 and p < 0.001. A separate unadjusted recalculation is presented in the results, and should be distinguished from the sponsor analysis.

The record specifies the intention-to-treat population and nonresponder imputation for the primary endpoint. It also lists secondary efficacy outcomes and safety counts. Details about multiplicity control, the original sample-size justification, and several aspects of the statistical plan are not reported in the registry record.