A Study Comparing Risankizumab to Placebo in Participants With Active Psoriatic Arthritis (PsA) Who Have a History of Inadequate Response to or Intolerance to at Least One Disease Modifying Anti-Rheumatic Drug (DMARD) Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Comparing Risankizumab to Placebo in Participants With Active Psoriatic Arthritis (PsA) Who Have a History of Inadequate Response to or Intolerance to at Least One Disease Modifying Anti-Rheumatic Drug (DMARD) Therapy (NCT03675308), a randomized phase 3 study. The primary outcome is Percentage of Participants With an American College of Rheumatology 20% (ACR20) Response at Week 24.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group study comparing risankizumab with placebo in participants with active psoriatic arthritis. The registry reports enrollment of 964 participants. The primary response analysis included 481 participants in the placebo group and 483 in the risankizumab group.

The randomized comparison was blinded during Period 1. The placebo group received placebo at Week 0, Week 4, and Week 16, while the risankizumab group received 150 mg risankizumab at those time points. At Week 24, the groups' assigned treatments changed: participants in the placebo group began risankizumab, while participants in the risankizumab group received blinded placebo. Both descriptions then specify open-label risankizumab beginning at Week 28 and repeated every 12 weeks thereafter. The placebo-controlled primary endpoint is at Week 24, before these later treatment periods can be interpreted as a continued randomized placebo comparison.

The registry records a study start date of 2019-03-25 and a primary completion date of 2020-10-08. Its status is active, not recruiting. The registry does not report the original sample-size justification, the intended allocation ratio, or details of any interim analyses. The planning illustrations associated with the posted effect are not the trial's original design assumptions and should not be read as evidence about how the trial was planned.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Participants were randomized equally (1:1 ratio) to receive double-blind treatment with risankizumab 150 mg or matched placebo for 24 weeks. Randomization was stratified by extent of psoriasis (≥ 3% body surface area \[BSA\] or \< 3% BSA), current conventional synthetic disease-modifying anti-rheumatic drug (csDMARD) use (0 vs ≥ 1), presence of dactylitis (yes vs no), and presence of enthesitis (yes vs no) at Baseline.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set; Non-responder imputation incorporating multiple imputation (MI) to handle missing data due to COVID-19 (NRI-C) was used in the analysis.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: The full analysis set (FAS) included all randomized participants who received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With an American College of Rheumatology 20% (ACR20) Response at Week 24, measured over Baseline and Week 24.

The primary endpoint is the percentage of participants with an American College of Rheumatology 20% response at Week 24. It is a binary outcome: each participant is classified as either meeting or not meeting the response definition. The registry defines response as meeting all three conditions: at least 20% improvement from baseline in the 68-tender joint count, at least 20% improvement in the 66-swollen joint count, and at least 20% improvement in at least three of five additional measures.

Those additional measures are physician global assessment of disease activity, patient global assessment of disease activity, patient assessment of pain, the Health Assessment Questionnaire–Disability Index, and high-sensitivity C-reactive protein. Requiring improvement across joint counts and several other aspects of disease activity makes this a composite response rather than a measure of just one symptom or examination finding.

A response percentage is useful for comparing how often a defined level of improvement was reached in each randomized group. It does not describe the size of improvement among responders, nor does it capture every possible change in disease activity. The registry specifies the full analysis set and a non-responder imputation approach incorporating multiple imputation for missing data due to COVID-19.

7. Results Posted in the Registry

• Risankizumab: 57.3% (277/483) versus 33.5% (161/481) on control; risk difference 23.9 percentage points (95% CI 17.8 to 30.0), risk ratio 1.71 (1.48–1.99); 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 Psoriasis Area Severity Index (PASI) 90 Response at Week 24

• 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: 52.4% (143/273) versus 9.9% (27/272) on control; risk difference 42.5 percentage points (95% CI 35.5 to 49.4), risk ratio 5.28 (3.63–7.68); 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 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With an ACR20 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:
  • Risankizumab: 56.3% (272/483) versus 33.5% (161/481) on control; risk difference 22.8 percentage points (95% CI 16.7 to 29.0), risk ratio 1.68 (1.45–1.95); 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 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Achieving Minimal Disease Activity (MDA) at Week 24

• 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: 25.1% (121/483) versus 10.2% (49/481) on control; risk difference 14.9 percentage points (95% CI 10.1 to 19.6), risk ratio 2.46 (1.81–3.34); 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 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes with numerical results described in the registry show a consistent direction: the risankizumab group had higher reported response percentages than the placebo group for Week 24 PASI90, Week 16 ACR20, and Week 24 minimal disease activity. Each sponsor-reported comparison has p <0.001. These outcomes extend the primary ACR20 result by addressing skin response, an earlier joint-response assessment, and a composite measure of minimal disease activity.

The registry lists 13 secondary outcomes, including measures of function, skin involvement, joint response, nail psoriasis, enthesitis, and dactylitis. Numerical results for the other secondary outcomes are not reported in the registry record. The available secondary findings therefore support a coherent pattern across the listed measures with results, but do not establish that every measured secondary outcome had the same result.

Secondary outcomes are generally interpreted with more caution when many outcomes are tested. If multiple hypotheses are examined without an adjustment or a prespecified testing hierarchy, the chance of obtaining at least one small p-value by chance can increase. The registry record does not report whether multiplicity was controlled or how the secondary tests were ordered. As a result, the secondary p-values should be read as reported comparisons rather than as a complete, multiplicity-adjusted set of confirmatory conclusions.

The secondary endpoints complement the primary endpoint; they do not replace it. Differences in outcome definition, timing, and analysis population can also affect how findings compare. The registry does not provide enough detail to infer unreported analysis procedures for the secondary outcomes, so their estimates and sponsor-reported tests should be considered within those limits.

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

• Change From Baseline In Health Assessment Questionnaire-Disability Index (HAQ-DI) at Week 24
• Percentage of Participants Achieving Psoriasis Area Severity Index (PASI) 90 Response at Week 24
• Percentage of Participants With an ACR20 Response at Week 16
• Percentage of Participants Achieving Minimal Disease Activity (MDA) at Week 24
• Change From Baseline in Modified Nail Psoriasis Severity Index (mNAPSI) Score at Week 24
• Change From Baseline in Fingernail-Physician Global Assessment (PGA-F)
• Percentage of Participants With Resolution of Enthesitis at Week 24
• Percentage of Participants With Resolution of Dactylitis at Week 24

9. Statistical Interpretation

10. Statistical Methods Explained

What does a response percentage represent?

For a binary endpoint, each participant contributes to one of two outcome categories: meeting the response definition or not meeting it. The response percentage is the proportion classified as responders in that group. It summarizes how frequently the defined outcome occurred, but it does not show the range of individual changes or the severity of disease among participants who did not meet the threshold.

For the primary endpoint, response requires improvement in both joint-count measures and in at least three of five additional measures. Because all listed requirements must be met, a participant who improves in some areas but not enough areas overall is not classified as an ACR20 responder. That threshold-based definition makes the outcome straightforward to compare between groups, while leaving other kinds of improvement outside the response count.

What is the response-rate difference?

A response-rate difference compares the proportions in the two groups by subtraction. It is expressed in percentage points when reported as a difference between percentages. The primary sponsor analysis reports a difference of 24.0 percentage points, with a 95% confidence interval from 18.0 to 30.0. The unadjusted recalculation uses the posted summary counts and reports the estimate on the proportion scale, 0.239, with an interval from 0.178 to 0.300.

The sponsor's method is Cochran-Mantel-Haenszel. In general, this method can estimate a comparison across strata while accounting for stratification factors. The registry record does not identify the strata or provide enough detail to determine which factors were used. The sponsor's reported result should therefore be described as a Cochran-Mantel-Haenszel analysis without inferring the specific adjustment.

How should the p-value be understood?

The sponsor reports p <0.001 for the primary comparison. In general, a p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions are correct. A small p-value is not a measure of effect size, does not show the probability that a treatment effect is real, and does not settle whether an effect is important to patients. The response percentages and the confidence interval provide complementary information about magnitude and precision.

The recalculated p-value is an unadjusted comparison based on posted summary counts. Such a calculation does not necessarily reproduce the sponsor's stratified analysis or its handling of missing outcomes. Here the sponsor-reported estimate and the recalculation are consistent, but the trial's reported analysis remains the primary inferential result.

What does the confidence interval add?

A confidence interval describes statistical uncertainty around an estimate under the method used to calculate it. In general, a 95% confidence interval is generated by a procedure that would cover the underlying parameter in 95% of repeated applications under its assumptions. It is not a probability statement that the particular interval contains the parameter, and it does not account for every source of uncertainty, such as unreported analysis choices or limitations in the study design.

For the primary result, the reported interval runs from 18.0 to 30.0 percentage points. Its position entirely above no difference is consistent with the sponsor's reported p-value. The interval still allows a range of possible underlying differences; the point estimate alone should not be treated as exact.

How does missing-data handling matter?

The primary analysis used non-responder imputation incorporating multiple imputation for missing data due to COVID-19, described as NRI-C in the registry. Non-responder imputation generally classifies participants without an observed response as non-responders. Multiple imputation is a general statistical approach that represents uncertainty about missing values by creating and analyzing multiple plausible completed datasets. The registry's combined label indicates the stated approach, but the precise implementation, assumptions, and sensitivity analyses are not reported in the registry record.

Missingness can affect a response percentage if the probability of missing data is related to the outcome or differs between groups. The analysis method makes an assumption about how missing outcomes should be treated; it cannot recover unobserved outcomes without assumptions. The registry does not report the number of missing primary outcomes or the reasons for missingness beyond identifying COVID-19 as a reason addressed by the method.

What can be learned from the planning illustrations?

In general, a sample-size calculation links an assumed effect, a significance threshold, and a desired probability of detecting that effect if it exists. The displayed planning illustrations use the observed effect or stated fractions of it, two-sided significance level 0.05, equal allocation, and a normal approximation. They are scenario calculations rather than evidence about the original design process. The trial's own sample-size justification and original assumptions are not reported in the registry record.

Power is a property of a specified study design and assumed effect, not a probability that the observed result is correct. In general, estimates from a completed study can be used to illustrate how sample size might vary under selected assumptions, but those illustrations do not establish what the trial was powered to detect at the planning stage. The sample-size reading explains the reported per-arm figures and their assumptions in more detail.

11. Confidence Intervals and Precision

The primary response percentages are accompanied by 95% confidence intervals: 29.3 to 37.8 for placebo and 52.9 to 61.8 for risankizumab. These intervals describe uncertainty in each group's estimated response percentage. The sponsor's comparison has a 95% confidence interval from 18.0 to 30.0 percentage points for the response-rate difference. The recalculated unadjusted difference is reported on the proportion scale as 0.239, with a 95% confidence interval from 0.178 to 0.300.

Precision is not the same as certainty about every aspect of the result. The interval reflects sampling uncertainty under the analysis method; it does not resolve questions about missing-data assumptions, outcome definitions, generalizability, or unreported analysis choices. Nor does it indicate the probability that a particular participant will respond.

The two groups had analysis sizes of 481 and 483. The interval around the group comparison remains more informative than the point estimate alone because it shows a range of effects compatible with the sponsor's method. The interval excludes no difference, consistent with the reported p <0.001, but its width still represents uncertainty about the magnitude. Confidence intervals for the secondary outcomes also describe uncertainty and should be interpreted alongside the multiplicity limitations.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are illustrations based on the observed effect, with scenarios using 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The registry does not report the trial's original sample-size justification or its original design assumptions.

The 80% power column gives the participants per arm estimated to be needed for an 80% chance of detecting the specified effect under those planning assumptions, if that effect exists and the assumptions hold. The 90% power column gives the corresponding per-arm estimate for a 90% chance. These are design probabilities calculated for hypothetical future study conditions; they are not probabilities that the observed finding is true.

Smaller effects require more participants to distinguish them from no difference with the same planned power and significance threshold. The figures therefore rise as the assumed effect moves from the full observed effect to smaller fractions. They are useful for illustrating how sensitive planning can be to the effect size chosen, but they should not be mistaken for the original trial's enrollment target or justification. This is not observed power.

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 = 0.335; treatment proportion p₁ = 0.573; average p̄ = 0.454.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7041 = 1.3801.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6836 = 0.5753.
4. n = (1.3801 + 0.5753)² / (0.2388)² = 67.1, rounded up to 68 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 <- 0.334719334719335                          # control proportion
p1 <- 0.573498964803313                          # 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(3675308)
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 = 68 (80% power), 90 (90% power)
 75% of observed effect: n per arm = 119 (80% power), 158 (90% power)
 50% of observed effect: n per arm = 262 (80% power), 351 (90% power)
power.prop.test: 67.1 
Simulated power at n = 68 per arm: 0.806 (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 reports a Cochran-Mantel-Haenszel response-rate difference of 24.0 percentage points for the primary endpoint, with a 95% confidence interval from 18.0 to 30.0 and p <0.001. The unadjusted recalculation from the posted summary counts gives a risk difference of 0.239 on the proportion scale, with a 95% confidence interval from 0.178 to 0.300, and an unadjusted p-value below 0.001. The recalculation is broadly consistent with the sponsor's reported comparison.

The two analyses need not be numerically identical because the sponsor reports a Cochran-Mantel-Haenszel analysis, while the recalculation is an unadjusted calculation from summary counts. The registry does not report the stratification details needed to reproduce the sponsor's analysis. The sponsor's own analysis is the trial-reported inferential result; the recalculation serves as a straightforward comparison of the posted group summaries, not a replacement for that analysis.

The registry record does not flag a discrepancy between the posted values and the recalculation. Both indicate a response-rate difference favoring the risankizumab group in the analyzed data. Neither the recalculation nor the sponsor's p-value alone addresses all questions about clinical importance, missing data, or generalizability.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports the primary endpoint analysis and several secondary analyses, but a multiplicity strategy is not reported in the registry record. It is therefore unclear whether the secondary comparisons were adjusted for testing multiple outcomes, whether a hierarchical testing sequence was used, or whether the reported secondary p-values are intended as confirmatory. When many hypotheses are tested, the chance of one or more apparently strong results arising by chance can increase unless the testing plan accounts for multiplicity. This general concern makes the primary endpoint especially important to distinguish from exploratory or multiplicity-unadjusted secondary findings.

The primary analysis used the full analysis set. The registry states that non-responder imputation incorporating multiple imputation was used to handle missing data due to COVID-19. The number of missing primary outcomes, the imputation model, the assumptions about missingness, and sensitivity analyses are not reported in the registry record. Those details would help readers judge how dependent the result is on the chosen missing-data approach.

The analysis method reported for the primary comparison is Cochran-Mantel-Haenszel. The registry does not identify the stratification variables or provide a full statistical analysis plan. The sample-size justification, interim-analysis details, and original design assumptions are also not reported in the registry record. These absences do not negate the posted results, but they limit independent evaluation of the complete analysis strategy.

The study is randomized and double-blind during the primary comparison, which supports interpreting between-group differences within the trial. Still, an analysis population, missing-data method, and multiplicity plan each shape the interpretation. The registry's stated information supports a careful description of the reported result, while leaving these methodological details unresolved.

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

Serious events with the highest arm-level rate: pneumonia (0.4%), anaemia (0.2%), iron deficiency anaemia (0.2%), angina unstable (0.2%), vertigo (0.2%).

The registry reports serious adverse events among 18 of 481 participants in the placebo group and 12 of 483 participants in the risankizumab group. It also reports 0 deaths in the placebo group and 1 death in the risankizumab group. These are counts in the registry record; they do not by themselves establish that an event was caused by the assigned treatment or explain its clinical circumstances.

The listed serious-event terms include pneumonia (0.4%), anaemia (0.2%), iron deficiency anaemia (0.2%), unstable angina (0.2%), vertigo (0.2%), and duodenal ulcer (0.2%). The registry's listed term percentages should be read as reported, without assuming from this summary which treatment group each term occurred in. The record reports 0 in the other adverse-event count field for each group, but does not provide further detail in the registry record.

Serious adverse event counts are descriptive safety information. Differences in counts alone do not establish comparative risk, particularly without event definitions, follow-up context, attribution, and detailed event-level information. The registry summary supports reporting the counts and terms, not a causal conclusion.

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

The results show that, under the registry's primary analysis, a higher percentage of participants assigned to risankizumab met the defined ACR20 response criteria at Week 24 than participants assigned to placebo. The sponsor-reported difference is 24.0 percentage points, with a 95% confidence interval from 18.0 to 30.0 and p <0.001. This is a between-group result in the analyzed trial population, not a guarantee of an individual outcome.

The endpoint is a threshold-based composite. It records whether a participant met the specified joint-count and additional-measure criteria; it does not describe all dimensions of disease, the amount of change for every participant, or outcomes beyond the measured time point. The placebo group began risankizumab at Week 24, so the randomized placebo comparison should not be treated as continuing after that scheduled change.

The result does not by itself establish how the findings apply to people outside the trial population, how long differences persist, or how results compare with treatments not studied in this trial. The registry record does not supply the detailed information needed to answer those questions. It also does not resolve all uncertainty related to missing data, multiple secondary comparisons, or the clinical importance of the response-rate difference.

The safety counts are descriptive and should not be interpreted as proving that an event was caused by treatment or that one group had a definitive safety advantage. Decisions about care require clinical judgment and information beyond these trial summaries; these statistics are not medical advice.

18. Why This Trial Matters Statistically

This trial illustrates how a binary clinical response can be analyzed and communicated. The ACR20 endpoint combines improvement in joint counts with improvement in other disease-related measures, and the result is reported as a percentage in each randomized group. The response-rate difference and its confidence interval then express the comparison in a form that is more directly interpretable than a p-value alone.

It also shows why the analysis method matters. The sponsor used a Cochran-Mantel-Haenszel analysis, while a simple recalculation from posted summary counts is unadjusted. The two give consistent descriptions here, but they are not interchangeable methods. Distinguishing the trial-reported analysis from a summary-statistic recalculation helps readers understand what can be independently checked and what depends on details of the sponsor's analysis.

The primary endpoint's missing-data approach is explicitly identified as non-responder imputation incorporating multiple imputation for missing data due to COVID-19. At the same time, essential details such as the number of missing outcomes and sensitivity analyses are not reported in the registry record. This contrast highlights the value of transparent reporting: naming a method helps, while describing its implementation and assumptions is necessary for fuller assessment.

Finally, the secondary outcomes raise the general issue of multiplicity. Several measures can add breadth to a clinical picture, but their inferential weight depends on a planned testing strategy. Reporting outcome estimates alongside uncertainty, analysis methods, and missing-data limitations helps separate the primary evidence from supportive findings.

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 record does not report the original sample-size justification, full statistical analysis plan, or details of interim analyses. It also does not identify the stratification factors used in the reported Cochran-Mantel-Haenszel analysis. These omissions limit independent assessment of the design assumptions and reproduction of the sponsor's primary analysis.

For the primary endpoint, the registry identifies the full analysis set and the use of non-responder imputation incorporating multiple imputation for missing data due to COVID-19. It does not report the number of missing outcomes, the imputation model, the assumptions used, or sensitivity analyses. Those details matter because missing-data assumptions can affect estimated response rates.

The registry does not report a multiplicity adjustment or a testing hierarchy for secondary endpoints. Although the reported numerical secondary results have small p-values, their inferential status is difficult to assess without knowing the prespecified multiplicity strategy. Numerical results for the other listed secondary outcomes are not reported in the registry record.

The randomized placebo-controlled comparison ends at the Week 24 treatment change described in the registry. Later treatment periods do not provide the same continuing placebo comparison, so the Week 24 result should not be extended to later periods. The registry summary also provides limited detail about serious adverse events, including event-level circumstances, attribution, and follow-up context. These limitations constrain interpretation of both efficacy and safety summaries without erasing the reported findings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study Comparing Risankizumab to Placebo in Participants With Active Psoriatic Arthritis (PsA) Who Have a History of Inadequate Response to or Intolerance to at Least One Disease Modifying Anti-Rheumatic Drug (DMARD) Therapy — NCT03675308.
• Registry record last updated 2025-08-05; status last verified 2025-08.
• Results first posted 2022-02-24.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT03675308 is a Phase 3, randomized, double-blind, parallel-group study of risankizumab versus placebo in participants with active psoriatic arthritis and a history of inadequate response to or intolerance of at least one DMARD therapy. The registry reports enrollment of 964, a study start date of 2019-03-25, a primary completion date of 2020-10-08, and status active, not recruiting.

The primary endpoint was ACR20 response at Week 24. The reported response percentages were 33.5% for placebo and 57.3% for risankizumab. The sponsor-reported response-rate difference was 24.0 percentage points, with a 95% confidence interval from 18.0 to 30.0 and p <0.001. The primary analysis used the full analysis set and addressed missing data due to COVID-19 using non-responder imputation incorporating multiple imputation.

Reported secondary outcomes show a consistent direction among the numerical results available in the registry. Serious adverse event counts were 18 in the placebo group and 12 in the risankizumab group; deaths were 0 and 1, respectively. Important details about multiplicity, missing-data implementation, and the original design assumptions are not reported in the registry record.