A Phase 3, Multi Site, Randomized, Double Blind, Placebo Controlled Study Of The Efficacy And Safety Comparing CP- 690,550 And Etanercept In Subjects With Moderate To Severe Chronic Plaque Psoriasis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 3, Multi Site, Randomized, Double Blind, Placebo Controlled Study Of The Efficacy And Safety Comparing CP- 690,550 And Etanercept In Subjects With Moderate To Severe Chronic Plaque Psoriasis (NCT01241591), a randomized phase 3 study. The primary outcome is Percentage of Participants With a Physician's Global Assessment (PGA) Response of "Clear" or "Almost Clear" at Week 12.

2. Clinical Question

The primary outcome is a binary response: each participant either met the clear-or-almost-clear definition at week 12 or did not. The registry record defines clear as a PGA score of 0 and almost clear as a score of 1. The trial therefore asks about the proportion meeting a specified clinical assessment threshold, not the average amount of change in every participant. The endpoint’s timing and response definition give the comparison a particular scope: the posted primary result concerns the week 12 assessment and this dichotomized PGA outcome.

3. Trial Design

This was a phase 3, randomized, parallel-group study with four treatment groups. The arms were CP-690,550 5 mg twice daily with placebo twice weekly; CP-690,550 10 mg twice daily with placebo twice weekly; oral placebo twice daily with etanercept 50 mg twice weekly; and placebo on both schedules. The registry record lists enrollment of 1101. It describes the study as double blind in the title and records quadruple masking in the design information.

The study was listed as completed. Its recorded start was 2010-11, and its primary completion was 2013-01. The stated purpose was to evaluate the efficacy of CP-690,550 compared with etanercept and the safety of CP-690,550 in moderate to severe chronic plaque psoriasis. The parallel design assigned participants to a treatment group for comparison over the study period; it was not a crossover design.

The primary outcome was the percentage of participants with a PGA response of clear or almost clear at week 12. The primary analysis population included randomized participants who received at least 1 dose of the assigned investigational drug. Missing primary-outcome values were handled using non-responder imputation: participants with missing values were counted as non-responders. The posted arm counts for the primary outcome are 329 for the 5 mg group, 330 for the 10 mg group, 335 for the etanercept group, and 107 for the placebo-only group.

The registry record does not report the original sample-size justification, planned target power, prespecified primary-analysis model, allocation ratio, or planned interim analyses. The study’s masking and randomization are recorded, but further operational details about how those procedures were implemented are not reported in the registry record. These omissions limit what can be inferred about design decisions beyond the listed features and posted analyses.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS): all participants who were randomized to the study and received at least 1 dose of the randomized investigational drug (CP-690,550, etanercept, or placebo); Non-Responder Imputation (NRI) method: participants with missing values were considered to be non-responders.

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: Full Analysis Set (FAS): included all participants who were randomized to the study and received at least 1 dose of the randomized investigational drug (CP-690,550, etanercept, or placebo).

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With a Physician's Global Assessment (PGA) Response of "Clear" or "Almost Clear" at Week 12, measured over Week 12.

The primary endpoint is the percentage of participants whose Physician’s Global Assessment was clear or almost clear at week 12. The PGA is based on a global assessment of erythema, induration, and scaling across psoriatic lesions. These features are scored separately on a 5-point severity scale from 0 to 4, then summed and averaged; the resulting average is rounded to the nearest whole-number category. The categories range from 0, clear, to 4, severe. A response is defined as a score of 0 or 1.

This endpoint converts a graded assessment into a yes-or-no response for each participant, then compares the percentage responding across groups. Such a threshold can make a treatment-group comparison straightforward to communicate, but it does not preserve all detail in the underlying severity scores. Participants just on opposite sides of the clear-or-almost-clear threshold count differently, while participants at different levels within the non-response categories are grouped together.

The endpoint is used to assess the proportion meeting a defined level of overall skin status at a specified time, rather than the mean change in a score or a response sustained over time. The registry record specifies week 12 and the response threshold. It does not report a separate primary endpoint for durability beyond that assessment. Missing values in the primary analysis were treated as non-responses under the stated NRI method, so the reported percentage reflects that analysis rule as well as the observed assessments.

7. Results Posted in the Registry

• CP-690,550 10 mg BID + Placebo BIW: 68.2% (225/330) versus 47.1% (155/329) on control; risk difference 21.1 percentage points (95% CI 13.7 to 28.4), risk ratio 1.45 (1.26–1.66); unadjusted two-proportion z-test p = <0.001.
• Placebo BID + Etanercept 50 mg BIW: 66.3% (222/335) versus 47.1% (155/329) on control; risk difference 19.2 percentage points (95% CI 11.8 to 26.6), risk ratio 1.41 (1.23–1.61); unadjusted two-proportion z-test p = <0.001.
• Placebo BID + Placebo BIW: 15.0% (16/107) versus 47.1% (155/329) on control; risk difference -32.2 percentage points (95% CI -40.8 to -23.5), risk ratio 0.32 (0.20–0.51); 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 PSSM Response of 'Very Satisfied' or 'Somewhat Satisfied' at Week 12

• 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:
  • CP-690,550 10 mg BID + Placebo BIW: 85.9% (262/305) versus 71.0% (218/307) on control; risk difference 14.9 percentage points (95% CI 8.5 to 21.3), risk ratio 1.21 (1.11–1.32); unadjusted two-proportion z-test p = <0.001.
  • Placebo BID + Etanercept 50 mg BIW: 85.9% (263/306) versus 71.0% (218/307) on control; risk difference 14.9 percentage points (95% CI 8.5 to 21.3), risk ratio 1.21 (1.11–1.32); 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 CP-690,550 10 mg BID + Placebo BIW than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Placebo BID + Etanercept 50 mg BIW than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Least Squares (LS) Mean Change From Baseline in EQ-5D Health State Utility Score During the 12-Week Double-Blind Treatment

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • CP-690,550 10 mg BID + Placebo BIW: mean 0.21 versus 0.14 on control; difference 0.07 (95% CI 0.04 to 0.10), standardized effect d = 0.37; z-test p = <0.001.
  • Placebo BID + Etanercept 50 mg BIW: mean 0.19 versus 0.14 on control; difference 0.05 (95% CI 0.02 to 0.08), standardized effect d = 0.26; z-test p = 0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for CP-690,550 10 mg BID + Placebo BIW than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Placebo BID + Etanercept 50 mg BIW than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Mean Change From Baseline in EQ-5D VAS at Week 12

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • CP-690,550 10 mg BID + Placebo BIW: mean 16.60 versus 11.60 on control; difference 5.00 (95% CI 0.98 to 9.02), standardized effect d = 0.20; z-test p = 0.015.
  • Placebo BID + Etanercept 50 mg BIW: mean 15.70 versus 11.60 on control; difference 4.10 (95% CI 0.22 to 7.98), standardized effect d = 0.17; z-test p = 0.038.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for CP-690,550 10 mg BID + Placebo BIW than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Placebo BID + Etanercept 50 mg BIW than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes cover several kinds of measures, including other psoriasis assessments and health-related quality-of-life measures. Among the secondary results with numerical values, the active-treatment groups generally show higher response percentages or larger mean changes than the placebo-only group. The 10 mg CP-690,550 and etanercept groups are often close to one another on these measures, while the 5 mg group is not uniformly at the same level. This broad pattern is directionally consistent with the primary endpoint’s separation of the active-treatment groups from placebo, but secondary outcomes measure different aspects of status and should not be treated as repetitions of the primary PGA result.

Some secondary comparisons have posted unadjusted p-values and recalculated intervals. Those nominal results do not establish that the outcomes were part of a single confirmatory testing sequence. Many outcomes and analyses are registered, but the multiplicity strategy and hierarchy of testing are not reported in the registry record. When multiple outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase unless the analysis accounts for that multiplicity. Consequently, unadjusted secondary findings carry less confirmatory weight than a prespecified primary analysis with a documented testing plan.

The registry lists 46 outcome measures and 8 analyses, and records 44 posted secondary outcomes. The breadth of measurement can be useful for understanding whether a pattern extends beyond one assessment, but it also makes selective emphasis a concern when the analysis plan is unavailable. The secondary results should therefore be read as supporting context rather than independent proof of the primary finding or a definitive ranking of the treatment groups. The record does not report whether all planned secondary outcomes, time points, and comparisons were analyzed under a prespecified multiplicity adjustment.

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

• Percentage of Participants With a PGA Response of "Clear" or "Almost Clear" During the 12-Week Double-Blind Treatment
• Percentage of Participants in Each PGA Category During the 12-Week Double-Blind Treatment
• Percentage of Participants With a PASI75 Response During the 12-Week Double-Blind Treatment
• Mean PASI Score During the 12-Week Double-Blind Treatment
• Mean Change From Baseline in PASI Score During the 12-Week Double-Blind Treatment
• Mean PASI Component Scores by Body Region During the 12-Week Double Blind Treatment
• Mean Change From Baseline in PASI Component Scores by Body Region During the 12-Week Double Blind Treatment
• Mean Percentage of Total Psoriatic BSA During the 12-Week Double-Blind Treatment

9. Statistical Interpretation

10. Statistical Methods Explained

What does the response percentage represent?

For this binary endpoint, the numerator is the number of participants classified as having a PGA response, and the denominator is the analysis group size. The posted primary results give response percentages of 47.11% for CP-690,550 5 mg, 68.18% for CP-690,550 10 mg, 66.27% for etanercept, and 14.95% for placebo only. The corresponding posted response counts are 155, 225, 222, and 16. A percentage is useful for showing the frequency of a defined outcome, while the count and group size indicate how many participant records contribute to that percentage.

What is a risk difference?

A risk difference subtracts one group’s event proportion from another’s. Here the event is a PGA response. A positive difference in a comparison means the first proportion used in that recalculation is higher than the reference proportion; a negative difference means it is lower. The recalculated differences are unadjusted comparisons from the posted summary statistics. They do not account for covariates or other design features through a fitted model. Because the primary outcome is binary, the difference is naturally expressed on the proportion scale. The risk ratio and odds ratio are alternative ways to express a binary comparison, but they answer on different scales and should not be confused with the risk difference.

What does a p-value say?

A p-value describes how incompatible the observed data are with a specified null model, under the assumptions of the test used. A small p-value is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of a difference. The recalculated primary comparisons report p < 0.001. These are labelled unadjusted. The analysis method behind the sponsor’s reported comparisons and their p-values is not reported in the registry record; no sponsor p-value is posted for those comparisons.

What does the confidence interval add?

A confidence interval gives a range of values reasonably compatible with the data under the method used. For example, the 10 mg versus 5 mg recalculated risk difference has a 95% confidence interval from 0.137 to 0.284. Its width indicates that the estimated difference is not known with perfect precision. An interval that excludes zero is consistent with a nonzero difference under the corresponding unadjusted method, but it does not address multiplicity, bias, or whether the endpoint captures every outcome of interest. The sponsor’s comparison of 10 mg with etanercept is reported on a mean-difference scale, and its confidence interval includes zero. The method and interpretation of that sponsor analysis require care because the statistical method and non-inferiority margin are not reported.

How should missing outcomes be understood?

The registry specifies non-responder imputation for the primary endpoint: participants with missing values were treated as non-responders. This assigns a defined outcome to missing assessments rather than excluding those records from the response classification. The resulting estimate depends on that assumption. If missingness and response are related, a different missing-data approach could yield a different estimate. The number and reasons for missing primary assessments, and sensitivity analyses using alternative assumptions, are not reported in the registry record.

Why might an adjusted analysis differ?

An adjusted analysis can account for prespecified participant characteristics or design factors, while the simple recalculations use posted group summaries. Adjustment can change an estimate or its uncertainty, particularly if covariates are related to outcome or treatment allocation. The registry record does not report enough detail about the sponsor’s statistical model to reconstruct it or determine how it differs from the recalculations. The sponsor’s own analysis takes precedence when describing the trial’s reported inferential conclusions; the unadjusted recalculations are complementary descriptive checks, not replacements for the sponsor’s analysis.

Why can binary outcomes be compared in more than one way?

The risk difference reports an absolute contrast in response proportions. The risk ratio expresses one proportion relative to another, and the odds ratio compares response odds. Each measure has its own interpretation, and its numerical scale is not interchangeable with the others. The primary displayed recalculation emphasizes the risk difference, which is readily interpreted as the absolute separation in response frequencies. None of these measures removes uncertainty or resolves questions about missing data, multiple testing, or whether the comparison was prespecified. The record does not report whether multiplicity adjustment was used for the sponsor’s full set of outcome analyses.

11. Confidence Intervals and Precision

The primary recalculated confidence intervals show that the estimated response differences are not single, exact quantities. For the 10 mg dose compared with the 5 mg reference group, the risk difference is 0.211, with a 95% confidence interval from 0.137 to 0.284. For etanercept compared with that reference, the estimate is 0.192, with a 95% confidence interval from 0.118 to 0.266. Both intervals lie above zero under the unadjusted recalculation. The placebo-only comparison is -0.322, with a 95% confidence interval from -0.408 to -0.235.

These ranges describe uncertainty in the estimated contrasts under the recalculation method; they are not ranges of individual participant outcomes and do not mean that every value inside is equally likely. The confidence level refers to the long-run behavior of the interval procedure under its assumptions, rather than a probability statement about the fixed true effect after observing this particular interval.

Precision depends in part on the number of participants and the frequency of the outcome. The primary analysis group sizes differ, with 329 in the 5 mg group, 330 in the 10 mg group, 335 in the etanercept group, and 107 in the placebo-only group. These group sizes and observed response frequencies contribute to the estimated uncertainty. The intervals do not incorporate uncertainty about unreported design choices, missing-data assumptions, possible bias, or selection among multiple outcomes and comparisons.

The sponsor’s 10 mg versus etanercept estimate is 1.91, with a 95% confidence interval from -5.22 to 9.05. Its range includes zero, unlike the recalculated comparison of each of those groups with the 5 mg reference. These comparisons ask different questions and are reported on the sponsor’s mean-difference scale. The record does not report the sponsor’s analysis method or non-inferiority margin, so the interval alone cannot establish the full basis for an equivalence or non-inferiority conclusion.

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 figures are planning illustrations based on the observed effect, not the trial’s original design calculation. For this binary primary outcome, the figures are participants per arm, not total events. 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 to the stated fraction of the observed effect. The registry record does not report the assumptions used for the trial’s own original sample-size justification.

The 80% power column gives the estimated number of participants per arm for a hypothetical two-group study to have an 80% chance of detecting the specified effect under those planning assumptions, if that effect and the assumptions were correct. The 90% power column uses a higher target chance of detection, so it generally requires more participants per arm. For example, planning for 75% of the observed 10 mg versus 5 mg effect gives 155 participants per arm for 80% power and 207 participants per arm for 90% power.

The rows that use smaller fractions of the observed effect require larger sample sizes. This is expected: smaller differences are harder to distinguish from sampling variation, so more information is needed to detect them at the chosen significance and power levels. The curve is read as the same relationship across effect sizes: as the target effect becomes smaller, the planning sample size rises. It is an illustration of assumptions, not a report of enrollment or a guarantee about a future study.

These values are not observed power. They do not calculate the probability that the completed trial detected its true effect after results were known. They are prospective-style planning illustrations using the observed estimate as an input, and they can be sensitive to that estimate and to the specified assumptions.

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 = 0.471; treatment proportion p₁ = 0.682; average p̄ = 0.576.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6988 = 1.3696.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6827 = 0.5746.
4. n = (1.3696 + 0.5746)² / (0.2107)² = 85.1, rounded up to 86 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 <- 0.47112462006079                          # control proportion
p1 <- 0.681818181818182                          # 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(1241591)
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 = 86 (80% power), 114 (90% power)
 75% of observed effect: n per arm = 155 (80% power), 207 (90% power)
 50% of observed effect: n per arm = 352 (80% power), 471 (90% power)
power.prop.test: 85.1 
Simulated power at n = 86 per arm: 0.813 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The registry reports sponsor analyses for four comparisons involving the two CP-690,550 doses, etanercept, and placebo. For the comparison of CP-690,550 10 mg with etanercept, the sponsor reports a mean difference of 1.91 and a 95% confidence interval from -5.22 to 9.05. The registry labels this analysis non-inferiority or equivalence, but does not report the analysis method, margin, or p-value. The 5 mg versus etanercept comparison is reported as -19.16, with a 95% confidence interval from -26.55 to -11.76, also under that registry classification. Comparisons with placebo are reported as 53.23, with an interval from 44.81 to 61.65 for 10 mg, and 32.16, with an interval from 23.51 to 40.81 for 5 mg.

The simple recalculations use the posted response percentages and produce unadjusted risk differences for comparisons with the 5 mg group. They are not attempts to reproduce the sponsor’s model. The sponsor’s analysis takes precedence when describing the trial’s own inferential analysis; the recalculated estimates provide a separate, transparent contrast from summary statistics. The different comparison groups and statistical scales mean the estimates should not be treated as interchangeable. The registry does not report enough detail to reconcile all methodological differences or assess the non-inferiority framework fully. No discrepancy between the sponsor results and the recalculations is flagged in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates multiple outcomes, treatment contrasts, time points, or analysis choices. Even if each individual test uses a conventional significance threshold, a larger collection of tests raises the possibility of finding one or more small p-values by chance. The registry records numerous outcome measures and analyses, including a primary endpoint and a range of secondary psoriasis and quality-of-life outcomes. The multiplicity strategy, confirmatory testing hierarchy, and any adjustment for multiple comparisons are not reported in the registry record. The recalculated p-values are explicitly unadjusted and should be interpreted accordingly.

The primary endpoint’s analysis population is described as the Full Analysis Set: participants randomized and given at least 1 dose of randomized investigational drug. Missing primary outcomes were classified as non-responses under non-responder imputation. This rule makes the analysis definition explicit, but it does not show how many assessments were missing or whether missingness differed between groups. Reasons for missing data and sensitivity analyses under alternative assumptions are not reported in the registry record.

The registry record also does not report the detailed primary statistical model, covariates, any stratification factors, or the original analysis plan. It does not report a planned interim analysis or stopping rule. These details matter because they clarify which comparisons were planned, how uncertainty was calculated, and whether the reported tests followed a prespecified procedure. Their absence does not demonstrate that such procedures were not used; it means they cannot be assessed from the registry record.

The sponsor’s reported comparisons include labels for non-inferiority or equivalence, but the margin and analysis method are not reported. A non-inferiority interpretation depends on the prespecified margin and the method used, not only on whether an interval crosses zero. The sponsor’s own result should be distinguished from the unadjusted recalculations, and neither should be extended beyond the stated outcome and comparison. Together, the unreported multiplicity and analysis details limit the strength of claims that can be made from the registry record alone.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: psoriasis (1.9%), oedema peripheral (0.9%), myocardial infarction (0.3%), diverticulitis (0.3%), extradural abscess (0.3%).

The registry posts serious adverse event counts by treatment group. There were 7 serious adverse event-affected participants among 329 at risk in the CP-690,550 5 mg group, 5 among 330 in the 10 mg group, 7 among 335 in the etanercept group, and 2 among 107 in the placebo group. These are counts of participants affected, not counts of every event, and they describe the recorded groups rather than proving that a particular treatment caused an event.

The registry also lists serious adverse event terms, including psoriasis at 1.9%, peripheral oedema at 0.9%, and myocardial infarction, diverticulitis, extradural abscess, and foot fracture at 0.3% each. These posted terms should be read as registry-reported event categories. The record does not report a comparative statistical analysis for serious adverse events, detailed exposure-adjusted rates, or a causal assessment for each event. Death counts are not reported in the posted safety information.

The non-serious adverse event-affected counts are also recorded by group, but a simple comparison of counts would not account for exposure duration, event severity, or other differences in the safety data. The safety figures provide descriptive context; they are not sufficient to determine comparative safety on their own.

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

The primary results describe the proportion of participants in each randomized group who met the registry-defined clear-or-almost-clear PGA response at week 12, using the stated Full Analysis Set and non-responder imputation. They show different response percentages across the four groups and unadjusted recalculated differences with confidence intervals. This is evidence about a specific endpoint at a specific assessment time, under a specified missing-data rule.

The results do not by themselves establish that every participant’s psoriasis improved, that changes persisted beyond week 12, or that one treatment is preferable for an individual. A response proportion does not describe the full distribution of severity scores, the size of change among non-responders, or other outcomes that may matter to participants. Secondary outcomes add context but have their own definitions and analysis limitations.

The sponsor’s comparison of the 10 mg dose with etanercept is reported with an interval that includes zero, and the registry classifies it as a non-inferiority or equivalence analysis. Because the non-inferiority margin and analysis method are not reported, the sponsor result cannot be independently evaluated from the registry record alone. The unadjusted recalculations answer comparisons against the 5 mg reference group, not the same sponsor question, and should not be substituted for the sponsor’s analysis.

Statistical significance is not the same as clinical importance. The registry record does not specify a minimally important difference for the primary endpoint or provide enough information to connect the risk differences to such a threshold. Confidence intervals convey statistical precision under their method, not certainty about bias, generalizability, or benefit and harm together. The safety counts are descriptive and do not establish causal relationships. Overall, the findings should be read as estimates from this trial and outcome framework, with the unreported analysis details and multiplicity plan kept in view.

18. Why This Trial Matters Statistically

This trial is statistically informative because it compares two doses of an investigational oral treatment with both an active comparator and placebo in a randomized, parallel-group design. That structure supports more than one type of contrast: comparisons with placebo address separation from the placebo-only group, while comparison with etanercept addresses a different question. Keeping those questions distinct is important when interpreting the reported mean differences and recalculated risk differences.

The primary outcome is a binary response derived from a graded global assessment. This makes the result accessible as a percentage of participants meeting a clear threshold, but it also illustrates how outcome definition shapes interpretation. Dichotomization can discard information about the degree of severity within response and non-response categories. Reading the response percentage alongside the endpoint definition, group sizes, and uncertainty makes the result more meaningful than considering a p-value alone.

The record also illustrates why sponsor analyses and simple recalculations should not be conflated. The recalculations make unadjusted group contrasts from posted summaries, while the sponsor reports comparisons on a different scale and labels some as non-inferiority or equivalence. The latter interpretation depends on methodological details that are not reported. Recognizing this distinction prevents an easily computed result from being mistaken for the trial’s full confirmatory analysis.

Finally, the study’s multiple secondary measures, missing-data rule, and unavailable multiplicity details show how statistical conclusions depend on more than the headline endpoint. Confidence intervals, analysis populations, assumptions, and the role of each outcome all matter. The registry record supports a careful account of the posted estimates, while also marking limits on how far they can be generalized or interpreted.

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.

Several trial-specific limitations follow from the registry information. The primary endpoint is a thresholded PGA response at week 12, so it does not retain the full range of severity scores or describe response at later times. The posted primary analysis uses non-responder imputation, but the number and reasons for missing assessments and sensitivity analyses under different missing-data assumptions are not reported in the registry record.

The original sample-size justification and planned power are not reported. Although planning illustrations can be generated from observed effects, they do not explain the trial’s design rationale. The statistical plan, primary model, covariate adjustment, and planned interim analyses are also not reported. This makes it difficult to determine from the registry record whether all posted analyses correspond to prespecified hypotheses or how the sponsor’s analysis was implemented.

The sponsor labels some comparisons as non-inferiority or equivalence, yet the margin and analysis method are not reported. Those details are central to evaluating such claims. In addition, multiplicity handling across the many outcomes and comparisons is not reported, so unadjusted secondary findings should be considered exploratory or supportive rather than treated as a set of independent confirmatory tests.

The posted safety information is descriptive. It does not include a comparative analysis of serious adverse events, exposure-adjusted rates, or a causal assessment. The registry record also does not provide detailed information in the registry record about participant characteristics, adherence, or outcomes beyond the listed measures and timeframes. These omissions limit assessment of generalizability and of the balance of outcomes across groups.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Phase 3, Multi Site, Randomized, Double Blind, Placebo Controlled Study Of The Efficacy And Safety Comparing CP- 690,550 And Etanercept In Subjects With Moderate To Severe Chronic Plaque Psoriasis — NCT01241591.
• Registry record last updated 2018-12-26; status last verified 2018-12.
• Results first posted 2014-05-07.
• 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, parallel-group trial in moderate to severe chronic plaque psoriasis, with quadruple masking recorded and 1101 participants enrolled. The four groups received CP-690,550 5 mg, CP-690,550 10 mg, etanercept 50 mg, or placebo, using the oral and twice-weekly placebo schedules specified in the arm descriptions.

The primary endpoint was the percentage with a PGA response of clear or almost clear at week 12. The Full Analysis Set included randomized participants who received at least 1 dose, and missing primary assessments were treated as non-responses. Posted response percentages were 47.11% for 5 mg CP-690,550, 68.18% for 10 mg, 66.27% for etanercept, and 14.95% for placebo only.

The registry also reports sponsor comparisons, secondary outcomes, and descriptive safety counts. Details of the original sample-size justification, analysis plan, multiplicity approach, and non-inferiority margin are not reported in the registry record. The unadjusted recalculations are complementary comparisons from the posted summary statistics, not a substitute for the sponsor’s analysis.