Study of Subcutaneous Risankizumab Injection Compared to Oral Apremilast Tablets to Assess Change in Disease Activity And Adverse Events in Adult Participants With Moderate Plaque Psoriasis Who Are Candidates for Systemic Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Subcutaneous Risankizumab Injection Compared to Oral Apremilast Tablets to Assess Change in Disease Activity And Adverse Events in Adult Participants With Moderate Plaque Psoriasis Who Are Candidates for Systemic Therapy (NCT04908475), a randomized phase 4 study. The primary outcome is Percentage of Participants Achieving Psoriasis Area Severity Index (PASI) 90 (Defined as at Least 90% Improvement in PASI From Baseline) in Intent to Treat Population at Week 16 (ITT_A).

2. Clinical Question

3. Trial Design

This was a Phase 4, randomized, parallel-group study with single masking. The study title describes it as open-label and efficacy assessor-blinded: participants and treatment delivery were not described as fully blinded, while the efficacy assessment was blinded. The registry does not report further details about which assessors were masked or how masking was maintained.

The registry lists enrollment of 352 participants. Its study summary says approximately 330 participants would be enrolled across approximately 55 sites globally. The listed start date is 2021-06-09, the primary completion date is 2023-04-20, and the status is completed.

The study had two periods. Period A ran from Baseline to Week 16, with participants randomized to risankizumab or apremilast. The primary endpoint was assessed at Week 16. Period B ran from Week 16 to Week 52. Participants who had received apremilast in Period A were randomized again to receive risankizumab or continue apremilast. The summary also says that apremilast non-responders at Weeks 28 and 40, based on their psoriasis score, would be offered risankizumab.

These later treatment sequences mean that Period B results address a different question from the original Period A comparison. In particular, the Week 52 secondary analyses concern apremilast non-responders and compare specified subsequent treatment sequences. They should not be read as a simple continuation of the original randomized comparison. The registry record does not report further details of the randomization process, allocation concealment, or the rationale for the planned enrollment.

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: Period A: participants were randomized in a 1:2 ratio to risankizumab (RZB) or apremilast (APR). Period B: participants receiving RZB were continued up to Week 52; participants receiving APR were re-randomized in a 1:1 ratio to receive either RZB or APR (with the option to receive RZB as rescue) up to Week 52. Rerandomization was stratified by Psoriasis Area and Severity Index (PASI) 75 response (responder, non-responder) at Week 16.

Period A (Baseline to Week 16)

Reasons for not completing this period:

Period B (Week 16 to Week 52)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT\_A Population: all participants randomized in Period A

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.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Achieving Psoriasis Area Severity Index (PASI) 90 (Defined as at Least 90% Improvement in PASI From Baseline) in Intent to Treat Population at Week 16 (ITT_A), measured over Week 16.

The primary endpoint is the percentage of participants in the ITT A population who achieved PASI 90 at Week 16. PASI 90 means at least 90% improvement in the Psoriasis Area and Severity Index from Baseline. The registry describes PASI as an assessment combining the percentage of skin surface affected with the severity of erythema, induration, and desquamation across the head, upper extremities, trunk, and lower extremities. Scores range from 0 to 72, with higher scores indicating more severe disease.

A threshold endpoint such as PASI 90 turns a change in a broader disease score into a clearly defined response category. It allows the reported result to be expressed as the proportion of participants reaching that threshold, and the groups can be compared on that proportion. The threshold is specific: a participant either meets the stated criterion or does not, even though individual changes in PASI may vary.

The registry identifies the primary population as all participants randomized in Period A. That definition makes the result a comparison based on initial assignment, rather than only on participants who completed treatment or met another post-randomization condition. The registry does not describe additional endpoint handling rules, such as how missing Week 16 assessments were classified. Those details are important for understanding exactly how the reported percentages were produced and are not reported in the registry record.

7. Results Posted in the Registry

• Period A: RZB: 55.9% (66/118) versus 5.1% (12/234) on control; risk difference 50.8 percentage points (95% CI 41.4 to 60.2), risk ratio 10.91 (6.14–19.36); 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 PASI 75 (Defined as at Least 75% Improvement in PASI From Baseline) in Intent to Treat Population at Week 16 (ITT_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:
  • Period A: RZB: 84.7% (100/118) versus 18.8% (44/234) on control; risk difference 65.9 percentage points (95% CI 57.7 to 74.1), risk ratio 4.51 (3.42–5.95); 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 Period A: RZB than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Achieving PASI 75 in Intent to Treat Population for Apremilast Non-Responders at Week 52 (ITT_B_NR)

• 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:
  • Period B: APR/RZB, NR: 83.1% (69/83) versus 11.5% (9/78) on control; risk difference 71.6 percentage points (95% CI 60.9 to 82.3), risk ratio 7.20 (3.87–13.42); 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 Period B: APR/RZB, NR than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Achieving Static Physician Global Assessment (sPGA) 0 or 1 With at Least 2-grade Improvement From Baseline in Intent to Treat Population for Apremilast Non-Responders at Week 52 (ITT_B_NR)

• 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:
  • Period B: APR/RZB, NR: 77.1% (64/83) versus 7.7% (6/78) on control; risk difference 69.4 percentage points (95% CI 58.6 to 80.2), risk ratio 10.02 (4.61–21.82); 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 Period B: APR/RZB, NR than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes show a consistent broad pattern across the posted comparisons: the risankizumab-containing group or sequence has a higher reported response percentage than the apremilast comparison group or sequence. This pattern appears both in the Week 16 outcome assessed in the full Period A intent-to-treat population and in the Week 52 analyses among apremilast non-responders. The Week 52 endpoints use different response definitions and a selected population, so they add context rather than serving as direct replications of the primary analysis.

The Week 16 PASI 75 result is related to the primary PASI 90 endpoint because both use improvement in PASI, but they use different thresholds. The Week 52 PASI 75 and sPGA outcomes address later responses among apremilast non-responders after further treatment assignment. Their population and treatment history differ from the primary Week 16 comparison. These design distinctions matter when considering whether the secondary results reinforce, extend, or qualify the primary finding.

The posted secondary comparisons are reported with p-values below 0.001, but the registry record does not report a multiplicity adjustment or a complete prespecified testing hierarchy. When several outcomes are tested, the chance of at least one apparently notable result can increase unless the analysis plan accounts for repeated testing. Therefore, unadjusted secondary results generally carry less confirmatory weight than a clearly designated primary endpoint. The pattern can be described as supportive and coherent with the primary result, while recognizing that the public record does not establish how the secondary tests were controlled as a family.

These secondary findings also should not be combined into a single overall effect. The endpoints measure different response criteria and use different analysis populations and time points. In particular, a result among non-responders is conditional on not having responded to apremilast, and does not represent the whole original randomized population. The tables provide the specific percentages and intervals; this narrative emphasizes their shared direction and the limits on how strongly they can be interpreted.

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

• Percentage of Participants Achieving PASI 75 (Defined as at Least 75% Improvement in PASI From Baseline) in Intent to Treat Population at Week 16 (ITT_A)
• Percentage of Participants Achieving PASI 75 in Intent to Treat Population for Apremilast Non-Responders at Week 52 (ITT_B_NR)
• Percentage of Participants Achieving Static Physician Global Assessment (sPGA) 0 or 1 With at Least 2-grade Improvement From Baseline in Intent to Treat Population for Apremilast Non-Responders at Week 52 (ITT_B_NR)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant either achieved PASI 90 at Week 16 or did not. The registry summarizes this outcome as a percentage in each randomized group. In general, a percentage describes the frequency of an outcome within a group, while a between-group difference describes how far apart the groups are on that measure. These summaries are useful together: arm-level percentages show the observed results, and a contrast focuses on the comparison.

What does the adjusted analysis mean?

The registry reports a Cochran-Mantel-Haenszel analysis with an adjusted difference. In general, this method combines information across strata to estimate a group contrast while accounting for stratification. The specific stratification factors and the detailed model specification are not reported in the registry record. Therefore, the reported adjusted result should be presented as the sponsor’s registry-reported analysis, rather than treated as a calculation that can be reproduced from the arm percentages alone.

The reported adjusted difference is 50.7, with a 95% confidence interval from 41.3 to 60.1; the reported p-value is p < 0.001. The record labels the analysis as superiority. These are the sponsor’s posted analysis results. They should not be confused with the simple unadjusted re-calculation based on the posted group summaries, which uses a different level of information and does not reproduce the adjustment. The registry record does not give enough detail to independently reproduce the adjusted analysis.

How should the p-value be read?

In general, a p-value evaluates how compatible the observed data are with a specified null hypothesis, assuming the statistical model and analysis conditions hold. A p-value below the chosen threshold is often described as statistically significant, but it does not show the probability that the null hypothesis is true. It also does not measure the probability that the result will be repeated, or the importance of the difference for patients. Here, the reported p-value is p < 0.001. The registry does not report the full prespecified statistical analysis plan, so the threshold and any rules for selecting among analyses are not documented in the posted record.

What does a confidence interval contribute?

In general, a confidence interval conveys uncertainty around an estimate. A narrower interval indicates greater precision under the method used; a wider interval indicates more uncertainty about the value of the underlying contrast. The 95% confidence interval around the adjusted difference is reported as 41.3 to 60.1. Its interpretation depends on the reported analysis and its assumptions. A confidence interval is not a prediction interval for individual outcomes, and it does not account for sources of uncertainty that were not included in the model.

Why might over-dispersion matter?

In general, over-dispersion occurs when outcome variability is greater than a model expects. For binary outcomes, the practical concern is that a model may understate uncertainty if its assumptions do not adequately represent variation across participants or groups. The registry does not report an over-dispersion assessment or a model-based diagnostic for this endpoint. This is a general statistical consideration, not a reported finding about this trial.

What can the posted counts support?

The posted group sizes and response percentages allow a simple, unadjusted comparison of proportions. Such a calculation is a descriptive re-analysis of the summary results, not a substitute for the sponsor’s reported adjusted analysis. It does not incorporate unreported stratification, covariates, missing-data rules, or analysis details. The discrepancy flag is false, but that does not make the two approaches identical: adjusted and unadjusted estimates may differ because they answer somewhat different questions and use different information. The primary interpretation should retain the distinction between the registry’s reported analysis and a simple calculation from the posted summaries.

For the secondary binary outcomes, the registry also gives arm percentages and reported comparisons. The endpoint definitions and populations differ, particularly for the Week 52 analyses restricted to apremilast non-responders. Each such comparison is conditional on the treatment history and the eligibility definition used for that later analysis. The statistical analysis plan, missing-data procedures, and any multiplicity adjustment are not reported in the registry record. Those omissions limit how comprehensively the significance tests can be assessed across the full collection of outcomes.

11. Confidence Intervals and Precision

The primary arm-level percentages have 95% confidence intervals of 2.3 to 8.0 for Period A apremilast and 47.0 to 64.9 for Period A risankizumab. These intervals show the uncertainty around each group’s estimated response percentage under the method used for those summaries. The sponsor’s adjusted between-group estimate is 50.7, with a 95% confidence interval from 41.3 to 60.1.

In general, a confidence interval is useful because it presents a range of estimates compatible with the observed data and analysis assumptions, rather than only a single point estimate. The width of an interval is one indication of statistical precision. It should not be interpreted as covering a fixed proportion of future participants or as accounting for every possible source of uncertainty. In particular, unreported analytic choices and missing-data handling are not resolved simply by seeing a confidence interval.

The adjusted interval excludes no difference, and its limits indicate that the reported contrast is not estimated with unlimited precision. The interval still allows a range of plausible magnitudes. The arm-level intervals and the adjusted interval refer to different quantities, so they should not be compared as if they were interchangeable measures of uncertainty. The registry does not provide the detail needed to evaluate how much the adjustment, or any unreported analysis decisions, affected 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 table and curve are planning illustrations based on the observed effect, not a record of the trial’s original design calculation. The figures are participants per arm for 80% power and participants per arm for 90% power; this is not a time-to-event outcome, so the entries are participants per arm rather than total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect.

In plain language, the 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting the specified effect if that effect were the true one, under the planning assumptions. The 90% power column gives the corresponding illustration for a 90% chance. The table shows scenarios for planning for 100%, 75%, and 50% of the observed effect. For the primary endpoint, The table shows 12 participants per arm for 80% power and 16 for 90% power when planning for 100% of the observed effect; the entries are larger for the smaller-effect scenarios.

The curve is a visual presentation of the planning relationship shown in the table; it should be read as an illustration, not as an observed trial result. Smaller effects require more participants to detect with the same power under otherwise unchanged assumptions. These calculations are not observed power. The registry record does not report the original sample-size justification or the assumptions used by the trial designers, so this illustration should not be used to infer those choices or to assess whether the trial met its original enrollment target.

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 = 0.051; treatment proportion p₁ = 0.559; average p̄ = 0.305.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6513 = 1.2765.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5433 = 0.4572.
4. n = (1.2765 + 0.4572)² / (0.5080)² = 11.6, rounded up to 12 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 <- 0.0512820512820513                          # control proportion
p1 <- 0.559322033898305                          # 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(4908475)
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 = 12 (80% power), 16 (90% power)
 75% of observed effect: n per arm = 19 (80% power), 25 (90% power)
 50% of observed effect: n per arm = 35 (80% power), 46 (90% power)
power.prop.test: 11.6 
Simulated power at n = 12 per arm: 0.846 (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 own primary analysis is reported as a Cochran-Mantel-Haenszel adjusted difference of 50.7, with a 95% confidence interval from 41.3 to 60.1 and p < 0.001. The posted group percentages are 5.1% for Period A apremilast and 55.9% for Period A risankizumab. The adjusted result is the registry’s stated analysis and is the appropriate result to quote as the sponsor-reported primary comparison.

A simple re-calculation from posted summary statistics is unadjusted. It can help readers understand the scale and direction of the contrast, but it is not a reproduction of the Cochran-Mantel-Haenszel method. The adjustment details are not reported in the registry record, so the adjusted estimate cannot be recreated from the summaries alone. The record marks the discrepancy flag as false; nevertheless, a descriptive re-calculation and a reported adjusted analysis are conceptually distinct and should be identified separately.

The reported analysis takes account of details that are unavailable in the public summary. The re-calculation should therefore be treated as an explanatory check, not as a replacement, correction, or independent confirmation of the sponsor’s result. For the primary endpoint, the tables and plot display the arm-level summaries alongside the registry-reported comparison so that readers can see both the observed percentages and the adjusted estimate.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies PASI 90 at Week 16 as the primary endpoint and lists three secondary outcomes. It reports analyses for the primary and secondary outcomes, but the registry record does not report a full statistical analysis plan, a testing hierarchy, or whether p-values were adjusted for multiplicity. Consequently, the public record does not show whether the reported secondary p-values were intended as confirmatory tests or as supportive analyses.

In general, multiplicity arises when researchers test several outcomes, time points, populations, or contrasts. Without an appropriate plan, interpreting each test as if it were the only test can overstate the strength of evidence across the collection of analyses. This concern does not erase the reported results; it limits how confidently a reader can treat multiple nominally significant findings as independent confirmations. The primary endpoint is the clearest prespecified focus in the registry, while the secondary outcomes require more caution in the absence of multiplicity details.

The registry defines the primary population as all participants randomized in Period A. It also names intent-to-treat populations for the secondary outcomes, including a population of apremilast non-responders at Week 52. The precise analysis-population definitions beyond those descriptions, and the handling of participants with missing endpoint data, are not reported in the registry record. In general, missing data can affect an estimated response percentage if the probability of having an observed outcome is related to treatment or response. The record does not specify imputation methods, non-responder rules for missing assessments, or sensitivity analyses, so their influence cannot be assessed from the posted summary.

The analysis method is given for the primary outcome and for the posted secondary comparisons, but further details are not. The factors used in the primary adjusted analysis, the planned sample-size justification, interim analyses, and the complete statistical plan are not reported in the registry record. These gaps make it difficult to evaluate all analytic choices or determine whether additional planned analyses were performed. They should be understood as limits of the available record, not as evidence that a particular procedure was or was not used.

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: pulmonary embolism (1.7%), coronary artery disease (1.0%), prostate cancer (1.0%), gastrointestinal haemorrhage (1.0%), hypersensitivity (1.0%).

The registry summarizes serious adverse events by treatment group and period. In Period A, serious adverse events affected 4 participants among 234 at risk in the apremilast group and 1 participant among 118 at risk in the risankizumab group. In Period B, the posted counts vary across the treatment sequences: 8 among 116 at risk for RZB/RZB, 2 among 97 for APR/APR, 3 among 102 for APR/RZB, and 0 among 48 for APR/APR/RZB. The registry reports 0 deaths in each listed group.

These are counts of participants affected, alongside the group sizes at risk; they are not a complete comparison of safety over equivalent exposure. Period B groups reflect different treatment sequences and participant histories, and the registry summary does not provide all information needed to assess duration of exposure or event timing. The record also lists serious adverse event terms with percentages, including pulmonary embolism at 1.7% and several other terms at 1.0%; those listed terms should not be interpreted as group-specific without additional information.

Serious adverse event counts are descriptive safety information. They do not establish that an event was caused by a study treatment, and they do not alone provide a complete assessment of all adverse events. The registry also reports counts for other adverse events in the tables; interpretation should account for the study periods and treatment sequences rather than treating every group as directly comparable.

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

The primary result describes the percentage of participants who met a defined PASI response threshold at Week 16 in a randomized comparison. The posted data show a higher observed PASI 90 percentage in the risankizumab group than in the apremilast group, and the sponsor’s adjusted analysis reports a difference with a 95% confidence interval that excludes no difference. These statements summarize the endpoint and analysis in this trial; they are not medical advice.

The result does not mean that every participant in one group had the same outcome, or that an individual’s response can be predicted from the group percentage. The endpoint is a threshold applied to a clinical score, so it does not describe every dimension of disease experience. It also does not establish the effects of treatment beyond the reported time point or for populations that were not represented in the study.

The trial’s later period involved re-randomization and treatment sequences, including analyses among apremilast non-responders. Those findings answer different questions from the original Period A comparison. They should not be collapsed into one continuous head-to-head comparison, nor should results from a selected non-responder group be assumed to apply to all participants.

The statistical evidence is bounded by what the registry reports. Adjustment details, missing-data methods, multiplicity procedures, and the full analysis plan are not reported in the registry record. The confidence interval describes uncertainty under the reported method, but it cannot address unreported sources of uncertainty or limitations in generalizability. Safety summaries also require separate consideration: event counts alone do not establish causation or fully characterize the balance of potential benefits and harms. The result is therefore best read as a precise statement about the posted Week 16 comparison, alongside its population, endpoint, analysis, and reporting limitations.

18. Why This Trial Matters Statistically

This trial is statistically informative because it has a clearly defined primary endpoint, a randomized Period A comparison, and a reported between-group analysis with a confidence interval. Reporting arm-level response percentages alongside an adjusted contrast helps distinguish the observed results in each group from the estimated comparative effect. The interval adds information about precision that a p-value alone would not convey.

The study also illustrates why design and analysis populations matter. The primary question concerns participants randomized in Period A at Week 16, whereas later outcomes concern re-randomized treatment sequences and, for some analyses, apremilast non-responders. Those results may look related because they concern the same condition and treatments, but their populations and estimands differ. Statistical interpretation should retain those distinctions.

The registry’s incomplete reporting of multiplicity, missing-data handling, adjustment factors, and original sample-size planning is also important. Such details influence how fully readers can evaluate reported estimates and tests. The sample-size illustration offers a way to understand planning under stated assumptions, but it does not reveal the original design rationale or the trial’s observed power. This distinction is especially useful when a posted result looks statistically clear: statistical significance does not eliminate uncertainty about the analysis choices, broader applicability, or other outcomes.

Overall, the record offers a substantial amount of outcome information while leaving important methodological details unavailable. Reading the tables together with the design and reporting context makes the numerical results more interpretable and reduces the risk of treating distinct comparisons as interchangeable.

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 full statistical analysis plan, the factors used in the adjusted primary analysis, or the details of allocation concealment. It also does not report how missing endpoint data were handled, whether sensitivity analyses were conducted, or whether any multiplicity adjustment was applied. These omissions limit independent assessment of the analysis and the strength of conclusions from the secondary tests.

The study was open-label with efficacy assessor masking, so participant and treatment awareness may have influenced aspects of care or reporting even though efficacy assessment was blinded. The record does not provide further details about which assessments were blinded or how masking was implemented. The primary endpoint is a threshold response based on PASI; the registry describes its components, but the result does not capture every possible outcome relevant to participants.

Interpretation of the later period is constrained by re-randomization, treatment switching, and the focus on apremilast non-responders in some analyses. Those groups have treatment histories and eligibility conditions that differ from the original Period A population. The registry also presents safety summaries across groups with different sequences and periods, without a complete exposure-based comparison in the posted results.

The posted enrollment is 352, while the study summary describes approximately 330 participants. The record does not explain that difference. The sample-size figures displayed with the results are planning illustrations based on the observed effect, not the trial’s own design justification. The original sample-size assumptions are not reported in the registry record. Finally, a completed, multicenter study provides evidence about its defined population and treatment periods, but the registry summary alone cannot establish how well the findings generalize to other settings or longer-term treatment questions.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Subcutaneous Risankizumab Injection Compared to Oral Apremilast Tablets to Assess Change in Disease Activity And Adverse Events in Adult Participants With Moderate Plaque Psoriasis Who Are Candidates for Systemic Therapy — NCT04908475.
• Registry record last updated 2024-04-30; status last verified 2024-04.
• Results first posted 2024-04-30.
• 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

NCT04908475 is a completed Phase 4, randomized, parallel-group study comparing risankizumab with apremilast in adults with moderate plaque psoriasis who were candidates for systemic therapy. It is described as open-label and efficacy assessor-blinded, with a primary Week 16 assessment in the ITT A population.

The primary endpoint was the percentage achieving PASI 90. The posted arm percentages were 5.1% for Period A apremilast and 55.9% for Period A risankizumab. The sponsor reported an adjusted difference of 50.7, with a 95% confidence interval from 41.3 to 60.1 and p < 0.001.

The study also included a Week 16 PASI 75 outcome and Week 52 outcomes among apremilast non-responders. Safety summaries are posted for the Period A and Period B treatment groups. The registry record does not report several details needed for fuller evaluation, including missing-data procedures, multiplicity handling, adjustment factors, and the original sample-size justification.