Efficacy and Safety Study of Apremilast to Treat Active Psoriatic Arthritis (PsA): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety Study of Apremilast to Treat Active Psoriatic Arthritis (PsA) (NCT01307423), a randomized phase 3 study. The primary outcome is Percentage of Participants With an American College of Rheumatology 20% (ACR20) Response at Week 16.

2. Clinical Question

3. Trial Design

The registry describes a phase 3, multicenter, randomized, double-blind, placebo-controlled, parallel-group study. Its masking field specifies quadruple masking. Participants were assigned to placebo, apremilast 20 mg, or apremilast 30 mg. The active treatments were administered orally twice daily. The trial title and summary identify the study population as people with active psoriatic arthritis who had not previously received DMARD treatment.

The registry records enrollment of 529 and lists the study as completed. The recorded study start date is 2010-12-09, and the primary completion date is 2013-02-21. Those dates describe the registry record; they do not, by themselves, specify how long each participant was followed. The primary outcome was assessed at Week 16. ACR20 response at Week 24 and several other outcomes at Week 16 or Week 24 are also listed.

The arm descriptions include placebo-controlled treatment and placebo tablets administered alongside apremilast tablets. The registry record does not report the detailed allocation ratio, the sequence-generation procedure, the exact masking roles, or the original sample-size justification. It also does not report enough detail to reconstruct the complete visit schedule or the full treatment-switching arrangements. Those design details should not be inferred from the arm labels alone.

For the primary outcome, the registry defines the analysis population as randomized participants specified by the protocol, except for one participant randomized in error who received no investigational product. It notes that participants withdrawing early or lacking sufficient data for a definitive response were part of the endpoint description, but the exact procedure used to assign or handle their outcomes is not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: This study consisted of a 24-week randomized, double-blind, placebo-controlled phase, a 28-week randomized, double-blind active treatment phase and a 4-year open-label safety phase, for an overall study duration of 5 years.

Only the first 6 of 9 groups are shown.

Placebo-controlled Phase (Week 0 - 24)

Reasons for not completing this period:

Active Treatment Phase (Week 25-52)

Reasons for not completing this period:

Long-term Safety Phase (Year 2)

Reasons for not completing this period:

Long-term Safety Phase (Year 3)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set consisting of all participants randomized as specified in the protocol; one participant randomized in error and not receiving any dose of investigational product was excluded. Participants who withdrew early or did not have sufficient data for a definitive determination of response status at Week 16 were counted as 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 was defined as participants who were randomized as specified in the protocol; one participant who was randomized in error and did not receive any dose of investigational product was excluded.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With an American College of Rheumatology 20% (ACR20) Response at Week 16, measured over Baseline and Week 16.

The primary endpoint was the percentage of participants with an ACR20 response at Week 16, assessed from baseline to Week 16. A participant counted as a responder only when all of the stated response conditions were met: at least 20% improvement in the 78-joint tender count, at least 20% improvement in the 76-joint swollen count, and at least 20% improvement in three or more of five additional measures. Those measures covered pain, patient and physician global assessments of disease activity, physical function assessed with the HAQ-DI, and C-reactive protein.

ACR20 is a binary endpoint: each participant is classified as either meeting or not meeting the response definition. The reported percentage therefore describes how often the full threshold was reached in each group. It is a composite definition, not a measure of the average change in each component. A participant who improves in some components but does not meet every required threshold is not counted as an ACR20 responder.

Using a prespecified response definition allows the comparison to focus on a consistent combination of joint counts and other disease assessments. At the same time, the result does not show which component accounts for a group difference, nor does it capture outcomes outside the definition. The registry record does not report the detailed rationale for selecting this endpoint or the original statistical planning for it.

7. Results Posted in the Registry

• Apremilast 20mg: 28.0% (49/175) versus 15.9% (28/176) on control; risk difference 12.1 percentage points (95% CI 3.5 to 20.7), risk ratio 1.76 (1.16–2.66); unadjusted two-proportion z-test p = 0.006.
• Apremilast 30mg: 30.7% (54/176) versus 15.9% (28/176) on control; risk difference 14.8 percentage points (95% CI 6.1 to 23.5), risk ratio 1.93 (1.29–2.89); 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.

Change From Baseline in Health Assessment Questionnaire- Disability Index [HAQ-DI]) at Week 16

• 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:
  • Apremilast 20 mg: mean -0.16 versus 0.01 on control; difference -0.17 (95% CI -0.26 to -0.07), standardized effect d = -0.36; z-test p = <0.001.
  • Apremilast 30mg: mean -0.20 versus 0.01 on control; difference -0.22 (95% CI -0.31 to -0.12), standardized effect d = -0.47; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Apremilast 20 mg 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 lower for Apremilast 30mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With an ACR 20 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:
  • Apremilast 20 mg: 29.1% (51/175) versus 13.1% (23/176) on control; risk difference 16.1 percentage points (95% CI 7.7 to 24.4), risk ratio 2.23 (1.43–3.48); unadjusted two-proportion z-test p = <0.001.
  • Apremilast 30 mg: 24.4% (43/176) versus 13.1% (23/176) on control; risk difference 11.4 percentage points (95% CI 3.3 to 19.4), risk ratio 1.87 (1.18–2.97); unadjusted two-proportion z-test p = 0.006.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Apremilast 20 mg 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 Apremilast 30 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Health Assessment Questionnaire- Disability Index [HAQ-DI]) at Week 24

• 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:
  • Apremilast 20 mg: mean -0.16 versus 0.01 on control; difference -0.17 (95% CI -0.27 to -0.07), standardized effect d = -0.34; z-test p = 0.001.
  • Apremilast 30 mg: mean -0.21 versus 0.01 on control; difference -0.22 (95% CI -0.32 to -0.12), standardized effect d = -0.45; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Apremilast 20 mg 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 lower for Apremilast 30 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary results described in the registry include HAQ-DI change at Week 16 and Week 24, and ACR20 response at Week 24. Across those reported comparisons, the active groups' estimates are directionally consistent with the primary Week 16 result: the recorded ACR20 proportions are higher than placebo, and the reported HAQ-DI changes are lower than placebo. This broad alignment can provide context across different outcome types and assessment times, but it does not constitute an independent confirmation of the primary finding.

These outcomes measure different things. ACR20 at Week 24 uses the same responder concept at a later assessment, while HAQ-DI is reported as a continuous change from baseline. A binary response proportion and a mean change answer different questions and should not be treated as interchangeable measures. The registry lists many additional secondary outcomes, but their individual results are not all described in the available outcome summaries.

Secondary results generally carry less confirmatory weight when many outcomes are tested and the analysis plan for controlling multiplicity is not reported. Without knowing which outcomes were prespecified as key, whether adjustment was used, or how the testing sequence was organized, isolated secondary p-values can overstate the strength of evidence. The registry record does not report a multiplicity-adjustment procedure for these results.

The secondary pattern is therefore best read as supportive descriptive context, not as proof that every measured outcome showed the same degree of difference or that the effects were sustained beyond the recorded assessment times. The registry record does not report a single combined analysis that integrates these secondary outcomes with the primary endpoint.

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 16
• Percentage of Participants With an ACR 20 Response at Week 24
• Change From Baseline in Health Assessment Questionnaire- Disability Index [HAQ-DI]) at Week 24
• Change From Baseline in 36-item Short Form Health Survey (SF-36) Physical Functioning Domain at Week 16
• Percentage of Participants With a Modified Psoriatic Arthritis Response Criteria (PsARC) Response at Week 16
• Change From Baseline in Patient's Assessment of Pain at Week 16
• Change From Baseline in Maastricht Ankylosing Spondylitis Entheses Score (MASES) at Week 16
• Change From Baseline in Dactylitis Severity Score at Week 16

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is binary because each participant either met the ACR20 definition by Week 16 or did not. A natural group summary is the proportion of participants classified as responders. Comparing two such proportions can be expressed as a risk difference: the response proportion in one group minus the response proportion in the other. A positive risk difference in these comparisons indicates a higher observed response proportion in the apremilast group than in placebo. The sponsor reports the primary contrasts as risk differences in percentage points.

The registry identifies the sponsor's primary comparisons as chi-squared analyses. In general, a chi-squared test assesses whether observed counts across categories are compatible with the no-association or no-difference model. For a binary endpoint, the categories are response and non-response. The reported p-value is conditional on the analysis model and its assumptions; it is not the probability that the null hypothesis is true, and it does not express the magnitude of the effect.

A confidence interval for a risk difference describes a range of values reasonably compatible with the observed data under the analysis approach. Its width reflects statistical uncertainty, which depends on such features as the observed proportions and the amount of information in the groups. An interval that excludes zero indicates that a zero risk difference is not among the values covered by that interval at its stated confidence level. It does not rule out bias, limitations in the endpoint, or uncertainty arising from unreported analysis choices.

The posted results also permit simple, unadjusted recalculations from summary counts. Those calculations can be useful as a check on the direction and approximate size of a contrast, but they are not a replacement for the sponsor's analysis. The registry reports sponsor analyses using a chi-squared method, and the recalculations do not incorporate any unreported adjustment or other details of the trial analysis. Because the registry reports no discrepancy between the recalculations and sponsor results, the sponsor's analysis is the primary reference for the trial findings.

For continuous outcomes such as HAQ-DI change, the registry reports least-squares means and standard errors, with sponsor comparisons using ANCOVA. In general, ANCOVA compares groups while adjusting for specified covariates; which covariates and model details were used are not reported in the registry record. A least-squares mean is a model-based group estimate, and its standard error describes uncertainty in that estimate. A mean difference gives the estimated contrast between groups. The posted recalculations for those outcomes are simpler unadjusted comparisons and should not be mistaken for the sponsor's adjusted analysis.

Risk differences, risk ratios, and odds ratios are different ways of expressing a binary-outcome contrast. A risk difference expresses an absolute difference in proportions. A risk ratio compares the proportions as a ratio, while an odds ratio compares the odds of response. These measures answer related but distinct questions and can have different numerical values. The primary registry analysis is presented as a risk difference, so that is the clearest measure for interpreting the primary contrast.

Statistical significance and clinical importance are not interchangeable. A p-value summarizes evidence against a specified null model under assumptions; it does not say whether a difference is large enough to matter to participants. Clinical interpretation also depends on what the endpoint measures, how the response threshold was defined, the uncertainty interval, and whether results are robust to analysis choices. The registry record does not report a threshold for a clinically important between-group difference.

Many outcomes and analyses are recorded for this study. When multiple hypotheses are examined, the chance of at least one apparently unusual result can rise unless the analysis accounts for the multiplicity or distinguishes confirmatory from exploratory outcomes. The registry record does not report whether the secondary analyses were adjusted for multiplicity or how their inferential status was specified. Their p-values should therefore be read as outcome-specific results, not as a complete account of the chance of false-positive findings across all analyses.

11. Confidence Intervals and Precision

The sponsor-reported 95% confidence interval for the primary comparison of apremilast 20 mg with placebo runs from 3.5 to 20.7 percentage points. For apremilast 30 mg compared with placebo, it runs from 6.1 to 23.5 percentage points. Both intervals are on the positive side of zero, while their widths show that the data remain compatible with a range of possible differences rather than one exact value.

In general, a confidence interval is a method for conveying uncertainty in an estimate. Under repeated use of the same statistical procedure and its assumptions, intervals constructed at a stated confidence level would cover the underlying quantity at that frequency. It is not correct to interpret a particular interval as assigning a probability to the fixed true difference. The interval also reflects only the uncertainty represented by the analysis; it does not account for unmeasured bias, errors in measurement, or every possible limitation in the trial conduct.

The registry reports confidence intervals for the between-group primary differences, not confidence intervals for each arm's response percentage. The endpoints of the sponsor-reported intervals are rounded as recorded. Their interpretation should be tied to the risk-difference scale: they describe uncertainty about the absolute contrast in response proportions, not about a relative effect.

The intervals overlap substantially, but overlap between separate intervals is not a formal test of the difference between the two active groups. The registry record does not report a direct dose-to-dose comparison for the primary endpoint. Therefore, the placebo comparisons do not establish a statistically reliable difference between the apremilast doses.

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 illustrations based on the observed effects, not a reconstruction of the trial's original design. For these binary outcomes, the figures are participants per arm. The stated 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 the stated fraction of the observed effect. The original trial's sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm under those assumptions for an 80% probability of detecting the specified effect if that effect were the true one. The 90% power column uses a higher target probability of detection and therefore generally requires more participants per arm. These are design-planning quantities, not evidence about what was achieved in the completed trial.

Rows that use a smaller fraction of the observed effect represent a more modest target difference. Smaller effects are harder to distinguish from random variation and need more participants to reach the same target power under the stated assumptions. The illustration is conditional on its assumptions; changes in allocation, outcome frequency, analysis method, or other design features could change the planning requirement.

These estimates are not observed power. They are hypothetical planning calculations based on the effect estimates in the results, not a measure of the probability that the trial succeeded after its results were known. The planning curve, where read alongside the figures, expresses the same general relationship: the required number of participants per arm rises as the target effect becomes smaller or the desired power increases.

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 = 0.159; treatment proportion p₁ = 0.280; average p̄ = 0.220.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5854 = 1.1474.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5791 = 0.4874.
4. n = (1.1474 + 0.4874)² / (0.1209)² = 182.8, rounded up to 183 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 <- 0.159090909090909                          # control proportion
p1 <- 0.28                          # 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(1307423)
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 = 183 (80% power), 245 (90% power)
 75% of observed effect: n per arm = 310 (80% power), 414 (90% power)
 50% of observed effect: n per arm = 659 (80% power), 881 (90% power)
power.prop.test: 182.8 
Simulated power at n = 183 per arm: 0.808 (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's sponsor-reported primary analyses compare each apremilast group with placebo using chi-squared tests and report risk differences with 95% confidence intervals. For the 20 mg group, the reported difference is 12.1 percentage points, with a 95% confidence interval from 3.5 to 20.7 and p = 0.006. For the 30 mg group, it is 14.8 percentage points, with a 95% confidence interval from 6.1 to 23.5 and p = 0.001.

Simple unadjusted recalculations from the posted summary counts give estimates close to the sponsor-reported contrasts. The registry's discrepancy indicator is false. Under the stated interpretation of these results, the sponsor's own analysis takes precedence over the recalculations, because it is the analysis reported for the trial rather than a reconstruction from summary statistics.

The recalculations are useful as a transparent check, but they cannot reproduce unreported details of the sponsor's analysis. They should not be presented as an alternative primary analysis or used to override the sponsor's results. The registry record does not provide enough detail to independently reconstruct every aspect of the sponsor's analysis procedure.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and numerous secondary outcomes and analyses. Testing several outcomes creates a multiplicity issue: even when there is no true effect on any outcome, a large set of tests can produce one or more small p-values by chance. Statistical plans can address this through adjustment procedures, ordered testing strategies, or by clearly distinguishing confirmatory outcomes from exploratory ones.

The registry record does not report whether the primary comparisons were adjusted for multiplicity, whether the two dose comparisons were treated as a family, or whether a gatekeeping or hierarchical testing plan was used. It also does not report whether secondary outcomes were adjusted for multiple testing. The sponsor-reported p-values should be understood as reported for their individual comparisons; the record does not establish how they relate to a study-wide error rate.

The endpoint population is described as the full analysis set consisting of randomized participants specified by the protocol, with one participant randomized in error and receiving no investigational product excluded. The description also refers to participants who withdrew early or lacked sufficient data for a definitive response. The precise missing-outcome rule, including how those cases were classified or imputed, is not reported in the registry record. As a result, the handling of missing Week 16 response data cannot be independently assessed from this information.

The registry record also does not report the detailed statistical analysis plan, the model covariates for ANCOVA outcomes, interim-analysis procedures, or a rationale for the enrollment target. These omissions limit assessment of whether all analysis decisions followed a prespecified plan. They do not change the numerical results recorded, but they constrain how fully the inferential framework can be evaluated.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: inguinal hernia (1.2%), clostridium difficile colitis (0.8%), meningitis bacterial (0.8%), pneumococcal bacteraemia (0.8%), pneumonia (0.8%).

The registry reports serious adverse events for the placebo-controlled period labeled Weeks 0-24 and for longer apremilast exposure groups labeled up to 5 years. During Weeks 0-24, the recorded serious-event counts are 5 among 176 at risk for placebo, 3 among 175 for apremilast 20 mg, and 1 among 175 for apremilast 30 mg. These are counts alongside the registry's at-risk denominators; they are not adjusted comparisons or evidence by themselves that a treatment caused or prevented an event.

The longer-exposure groups are separately described as apremilast 20 mg, apremilast 20 mg/30 mg, and apremilast 30 mg, with serious-event counts and at-risk numbers recorded for each. The registry does not provide a randomized placebo comparison for these longer-exposure summaries in the details. Group composition and exposure time may differ, so the figures should not be interpreted as directly comparable rates.

The registry lists serious-event terms including inguinal hernia, Clostridium difficile colitis, bacterial meningitis, pneumococcal bacteraemia, pneumonia, and incisional hernia, with the recorded percentages ranging from 0.8% to 1.2%. Death counts are not reported in the registry record. These safety summaries are descriptive; they do not establish causality or provide a complete benefit-risk assessment.

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

The primary result describes a difference in the recorded proportion of participants meeting a defined ACR20 threshold at Week 16. In this trial, the sponsor-reported risk differences compared each apremilast dose with placebo and their 95% confidence intervals were above zero. This is evidence about those randomized groups, the endpoint definition, and the assessment time in this study.

It does not mean that every participant in an active-treatment group met the response definition, or that a particular participant's outcome can be predicted from the group average. It does not identify which endpoint component drove the observed response proportions. Nor does it show that all symptoms, physical function, or other aspects of psoriatic arthritis changed in the same way.

The comparison does not establish which apremilast dose is preferable, because the primary results described are comparisons with placebo rather than a direct dose-to-dose test. It also does not establish how results would apply to people outside the studied population, including people with prior DMARD treatment, or whether outcomes persist beyond the recorded assessment times.

Statistical evidence is not the same as a judgment about clinical importance. The registry record does not report a threshold for a clinically important difference, and the p-values alone cannot supply one. The intervals communicate uncertainty about the estimated contrasts, but they do not cover every source of uncertainty, such as missing-data handling or unreported analysis decisions.

These findings are a description of trial results, not individual medical advice. Decisions about care require clinical context and information beyond the trial summaries, including the full safety profile and the needs and preferences of the person concerned.

18. Why This Trial Matters Statistically

This trial provides a clear setting for understanding the distinction between an outcome proportion and a treatment contrast. The arm-specific ACR20 percentages describe the observed response level in each group; the risk difference puts the placebo comparison on an absolute scale. Reading both helps prevent a contrast from being mistaken for the response rate itself.

The endpoint also illustrates how a composite responder definition works. ACR20 combines joint-count requirements with thresholds across additional measures. That approach creates a single interpretable response classification, but it compresses several domains into one binary outcome. The result cannot reveal whether the group difference was concentrated in one component or shared across them.

The registry's sponsor estimates and their confidence intervals make uncertainty visible rather than reducing the finding to a p-value. The secondary outcomes offer additional perspectives, while the number of recorded outcomes highlights why multiplicity and prespecified analysis plans matter. Together, these features make the study useful for learning how endpoint choice, effect scale, precision, and analysis transparency shape interpretation.

It is also a reminder that planning calculations derived from an observed effect are not the same as the assumptions used to design a trial. A planning illustration can show how required group size changes with target effect and desired power, but it cannot recover the original design rationale. Keeping that distinction clear avoids retrospective calculations being mistaken for evidence about trial planning or observed power.

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 primary endpoint is a threshold-based composite. It summarizes whether participants met all required ACR20 criteria, but does not describe the size of change in each component or outcomes among participants who narrowly missed the threshold. The registry record does not report a clinically important difference threshold for interpreting the between-group estimates.

The study population was described as having active psoriatic arthritis and no prior DMARD treatment. Findings may not directly generalize to people with different treatment histories or other populations not described in the registry record. The registry does not provide demographic or clinical baseline details in the information, so the representativeness of the enrolled population cannot be evaluated from these results alone.

The exact procedure for handling early withdrawal and insufficient data for a definitive Week 16 response is not reported. The record also does not report the detailed statistical analysis plan, multiplicity strategy, original sample-size justification, or the covariates used in the ANCOVA analyses. Those omissions limit independent assessment of missing-data assumptions and the full inferential plan.

The primary reported comparisons are each against placebo; a direct comparison between the active doses is not reported. Secondary results cover other outcomes and time points, but the registry record does not report whether multiplicity was controlled. Safety summaries include a placebo-controlled period and distinct longer-exposure groups, which should not be treated as equivalent randomized comparisons. These limitations do not negate the reported estimates, but they define how narrowly those estimates should be interpreted.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety Study of Apremilast to Treat Active Psoriatic Arthritis (PsA) — NCT01307423.
• Registry record last updated 2022-06-14; status last verified 2020-04.
• Results first posted 2014-08-15.
• 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 describes a completed, randomized, phase 3, multicenter, quadruple-masked, parallel-group, placebo-controlled study of apremilast for active psoriatic arthritis in participants not previously treated with DMARDs. Enrollment was 529. The primary endpoint was ACR20 response at Week 16.

The sponsor reported higher ACR20 response proportions in both apremilast groups than in placebo, with positive risk differences and 95% confidence intervals above zero. Secondary results include ACR20 at Week 24 and HAQ-DI changes at Week 16 and Week 24. Serious adverse events are recorded for a placebo-controlled period and for longer apremilast exposure groups.

The registry record does not report full details of missing-data handling, multiplicity adjustment, or the original sample-size rationale. These limits, along with the endpoint definition and population, are important when interpreting how far the results apply.