Tapinarof for the Treatment of Plaque Psoriasis in Adults (3002): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tapinarof for the Treatment of Plaque Psoriasis in Adults (3002) (NCT03983980), a randomized phase 3 study. The primary outcome is Percent of Subjects Who Achieve a Physician Global Assessment (PGA) Score of Clear (0) or Almost Clear (1) With a Minimum 2-grade Improvement From Baseline at Week 12. Analyses Were Done Using Multiple Imputation.

2. Clinical Question

The primary measure combines a final disease-severity category with a change requirement. A participant must meet both parts of the endpoint: a Week 12 score of 0 or 1 and improvement of at least 2 grades from baseline. This makes the outcome more restrictive than simply reaching a score of 0 or 1, because the required improvement depends on the starting score as well as the final assessment.

The registry record also posts a sponsor analysis comparing the groups with a Cochran-Mantel-Haenszel method and reports a risk ratio. A separate unadjusted re-calculation expresses the posted group summaries as a mean difference in percentages. These are different statistical summaries of the comparison; they should not be read as interchangeable estimates.

3. Trial Design

The study is registered as a Phase 3, randomized, parallel-group trial with triple masking. Its brief summary describes it as randomized, double-blind, and vehicle-controlled. The two descriptions of masking are both part of the registry record; an explanation of the difference is not reported in the registry record.

The trial compared once-daily tapinarof cream, 1%, with matching vehicle cream, also applied once daily. The brief summary describes a planned 2:1 randomization and treatment for 12 weeks. The posted group sizes are 343 for tapinarof and 172 for vehicle. The registry record lists enrollment of 515. The study is marked completed, with a start date of 2019-06-06 and a primary completion date of 2020-05-13.

The control group helps distinguish the group contrast from changes that may occur during the study without the active cream. Random allocation is intended, in general, to make assigned groups comparable on average at the start of a trial. Masking is intended, in general, to reduce the possibility that knowledge of assigned treatment influences assessments or other trial processes. The registry record does not report further details about how masking was implemented or maintained.

The primary endpoint and the posted secondary outcomes are evaluated at Week 12 or describe change from baseline to Week 12. The primary analysis is stated to use multiple imputation. The registry record does not report the original sample-size justification, the protocol's planned statistical analysis in full, or the trial's detailed analysis-population definitions. Those omissions limit how completely the reported estimates can be compared with a prespecified analysis plan.

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

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 Percent of Subjects Who Achieve a Physician Global Assessment (PGA) Score of Clear (0) or Almost Clear (1) With a Minimum 2-grade Improvement From Baseline at Week 12. Analyses Were Done Using Multiple Imputation, measured over Baseline to Week 12.

The primary endpoint is the percentage of participants who have a Physician Global Assessment (PGA) score of Clear (0) or Almost Clear (1) at Week 12 and have improved by a minimum of 2 grades from baseline. The PGA is a clinical assessment of the current state or severity of psoriasis. The registry describes a static 5-point scale based on erythema, scaling, and plaque thickness or elevation: Clear (0), Almost clear (1), Mild (2), Moderate (3), and Severe (4). Higher scores represent more severe disease.

The endpoint is used to summarize the proportion meeting a defined clinical response threshold, rather than the average score on the PGA scale. Requiring both a low Week 12 score and a minimum improvement means the endpoint reflects the final assessment and the change from baseline. Participants who reach a score of 0 or 1 without the required improvement do not satisfy the full endpoint definition.

In general, a threshold-based percentage can be easy to interpret as the proportion meeting a response criterion, but it does not show how far each participant's score changed or describe the full distribution of scores. The registry reports the endpoint as a percentage of subjects and states that analyses were done using multiple imputation. The registry record does not provide details of the imputation procedure or define the analysis population beyond the outcome description.

7. Results Posted in the Registry

• Tapinarof (DMVT-505) Cream Group: mean 40.20 versus 6.30 on control; difference 33.90 (95% CI 27.14 to 40.66), standardized effect d = 0.75; 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.

Percent of Subjects With ≥ 75% Improvement in Psoriasis Area and Severity Index (PASI) From Baseline at Week 12. Analyses Were Done Using Multiple Imputation.

• 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:
  • Tapinarof (DMVT-505) Cream Group: mean 47.60 versus 6.90 on control; difference 40.70 (95% CI 33.93 to 47.47), standardized effect d = 0.90; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Tapinarof (DMVT-505) Cream Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent of Subjects With a PGA Score of 0 or 1 at Week 12. Analyses Were Done Using Multiple Imputation.

• 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:
  • Tapinarof (DMVT-505) Cream Group: mean 43.60 versus 8.10 on control; difference 35.50 (95% CI 28.43 to 42.57), standardized effect d = 0.77; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Tapinarof (DMVT-505) Cream Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Mean Change in Percent of Total Body Surface Area (%BSA) Affected From Baseline to 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:
  • Tapinarof (DMVT-505) Cream Group: mean -4.22 versus 0.10 on control; difference -4.32 (95% CI -5.46 to -3.18), standardized effect d = -0.66; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Tapinarof (DMVT-505) Cream Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary results follow a broadly consistent direction across the reported types of outcome: the posted summaries favor the tapinarof group in the response percentages and show a larger reduction in affected body-surface area. The outcomes include PASI response thresholds, a PGA threshold, and change in percent of total body surface area affected. Together, they address related but nonidentical aspects of psoriasis status at Week 12.

This overall pattern is compatible with the primary endpoint result, because the primary and secondary measures all compare outcomes between the same randomized treatment groups over the study period. They should not be treated as independent confirmations, however. The endpoints are related, and the registry record does not report a multiplicity adjustment for the secondary analyses. The registry lists four secondary outcomes, while the posted summary describes three secondary results with group comparisons.

Unadjusted secondary p-values do not account for the number of outcomes or comparisons examined. In general, testing multiple hypotheses increases the chance of at least one apparently positive result when there is no real effect across the family of tests. Without a stated multiplicity strategy, secondary findings are more appropriately read as supportive or exploratory evidence than as a collection of separately definitive claims. The record does not report whether a hierarchical testing sequence, family-wise error control, or another multiplicity approach was prespecified.

The endpoints also differ in scale and interpretation. Threshold-based response percentages describe the proportion meeting a criterion, whereas change in body-surface-area involvement is a mean change measure. That distinction matters when comparing results: a consistent direction across outcomes may strengthen the overall descriptive picture, but it does not make their effect estimates interchangeable. The secondary results do not replace the primary endpoint or resolve missing details about the analysis plan.

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

• Percent of Subjects With ≥ 75% Improvement in Psoriasis Area and Severity Index (PASI) From Baseline at Week 12. Analyses Were Done Using Multiple Imputation.
• Percent of Subjects With a PGA Score of 0 or 1 at Week 12. Analyses Were Done Using Multiple Imputation.
• Mean Change in Percent of Total Body Surface Area (%BSA) Affected From Baseline to Week 12
• Percent of Subjects With ≥90% Improvement in PASI Score From Baseline to Week 12. Analyses Were Done Using Multiple Imputation.

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports a Cochran-Mantel-Haenszel analysis for the primary endpoint, with a risk ratio and a superiority classification. It also reports a p-value of < 0.001. Separately, the posted group summaries have been used for a simple unadjusted re-calculation of the mean difference, its confidence interval, and an unadjusted p-value. The calculation is not the sponsor's analysis and should be interpreted as a descriptive statistical cross-check of the posted summaries.

What does the reported risk ratio express?

In general, a risk ratio compares the response proportion in one group with the response proportion in another. A value above 1 indicates that the response proportion is higher in the numerator group than in the reference group, given the stated ordering. The registry reports a risk ratio of 6.10 for the primary comparison. This ratio is not a percentage-point difference, and it should not be interpreted as a count of additional participants or as an individual-level prediction.

The Cochran-Mantel-Haenszel method is, in general, a method for comparing categorical outcomes across strata. The registry names this method but does not report the stratification details for the primary analysis. The registry record also does not report the precise risk-ratio confidence interval, so the reported risk ratio cannot be accompanied by an interval-based precision description from the posted primary analysis.

What does the re-calculated mean difference express?

The re-calculated contrast is 33.9 percentage points, formed as the difference between the posted percentages of 40.2% and 6.3%. The associated 95% confidence interval is 27.14 to 40.66, and the unadjusted p-value is <0.001. In general, a mean difference on a percentage scale describes the absolute gap between group percentages. It is distinct from the relative comparison expressed by a risk ratio.

The group summaries list standard errors of 2.82 and 1.99. A standard error describes the estimated sampling variability of a statistic; it is not the spread of individual participant measurements. In general, smaller standard errors indicate a more precise estimate under the assumptions used. The registry presents the outcome as a mean percentage, but the endpoint itself is a response percentage, not a continuous measurement for each participant.

Why can the reported and re-calculated analyses differ in form?

The sponsor's analysis is a Cochran-Mantel-Haenszel risk-ratio analysis. The re-calculation is an unadjusted mean-difference analysis based on the posted group summaries. Different measures of effect and different methods need not yield the same numerical estimate or p-value. The posted re-calculation does not reproduce the sponsor's analysis, and no adjustment variables or stratification details are available to align the two methods fully.

General statistical explanation: a p-value measures how incompatible the observed data are with a specified null hypothesis under a statistical model. It is not the probability that the null hypothesis is true, nor does it measure the clinical importance of an effect. A small p-value is not a measure of precision by itself; confidence intervals provide information about the range of effect values compatible with the data and method.

How does multiple imputation fit the endpoint?

The primary endpoint description states that analyses were done using multiple imputation. In general, multiple imputation addresses missing outcome information by creating and analyzing multiple completed datasets, then combining results to account for uncertainty due to missingness. Its validity depends on the imputation model and assumptions about the missing data. The registry record does not report the number of imputations, model variables, missingness assumptions, or sensitivity analyses. Therefore, those details cannot be evaluated from the posted record.

The registry's analysis method and its imputation statement are not enough to establish whether every decision followed a prespecified protocol. The protocol's full statistical analysis plan, interim-analysis information, and analysis-population definitions are not reported in the registry record. General statistical interpretation should therefore distinguish the reported numerical result from unreported design and analysis details.

11. Confidence Intervals and Precision

The unadjusted re-calculated primary mean difference is 33.9 percentage points, with a 95% confidence interval from 27.14 to 40.66. This interval describes uncertainty around the estimated difference under the re-calculation's method and assumptions. It does not mean that 95% of individual participants had outcomes inside that range. In general, a confidence interval is a range produced by a statistical procedure that, under repeated use of the same procedure and its assumptions, would cover the target parameter at the stated rate.

The interval does not include zero, so the re-calculated difference is distinguishable from no group difference at the corresponding conventional two-sided significance level. This is consistent with the unadjusted p-value of <0.001. The interval applies to the mean difference, not the sponsor's risk ratio of 6.10. A confidence interval for the risk ratio is not reported in the registry record.

The width of a confidence interval reflects both the amount of information in the data and the variability estimated by the method. In general, greater precision corresponds to a narrower interval, while less precision corresponds to a wider one. Precision does not establish that an endpoint captures every aspect of the condition, and it does not eliminate uncertainty about missing-data assumptions, endpoint definitions, or analysis choices.

The posted group standard errors are 2.82 for tapinarof and 1.99 for vehicle. These are measures of uncertainty in the group percentage estimates, not standard deviations describing variation among individual participants. Because the primary analysis used multiple imputation and its details are not reported, the interval from the simple unadjusted re-calculation should not be mistaken for a complete account of uncertainty under every possible analysis of missing outcomes.

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 displayed planning values are participants per arm, not total event counts. They are illustrations derived from the observed effect, rather than a record of the trial's original design calculation. The registry's original sample-size assumptions and justification are not reported in the registry record.

The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each illustration sets the effect to detect as a stated fraction of the observed effect. The rows describe planning for 100%, 75%, or 50% of the observed effect. The columns describe participants per arm for 80% power and participants per arm for 90% power.

In general, power is the probability that a planned statistical test will detect an effect of a specified size under its assumptions. The 80% and 90% columns correspond to different target probabilities of detecting the assumed effect, if that effect and the planning assumptions were correct. They are not guarantees about the result of a future study. Planning for 90% power generally requires more participants than planning for 80% power under otherwise identical assumptions.

Smaller effects require more participants to detect with the same target power and significance level, which is why the planning figures increase as the assumed fraction of the observed effect becomes smaller. These figures are not observed power: they are prospective planning illustrations based on the observed result, not an assessment of the probability that this completed trial detected its true effect. A curve based on the same planning assumptions should be read as illustrating how the required per-arm sample size changes with the effect size, not as evidence about the trial's original design.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 33.900; pooled standard deviation σ = 45.227.
2. n = 2 × (1.96 + 0.842)² × 45.227² / 33.900² = 27.9, rounded up to 28 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
delta <- 33.9                      # observed difference in means
sigma <- 45.2271437671376                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3983980)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 28 (80% power), 38 (90% power)
 75% of observed effect: n per arm = 50 (80% power), 67 (90% power)
 50% of observed effect: n per arm = 112 (80% power), 150 (90% power)
power.t.test: 28.9 
Simulated power at n = 28 per arm: 0.787 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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 the sponsor's primary comparison as a Cochran-Mantel-Haenszel risk ratio of 6.10, with p < 0.001 and a superiority classification. The re-calculation from the posted group summaries gives an unadjusted mean difference of 33.9 percentage points, with a 95% confidence interval from 27.14 to 40.66 and p <0.001.

These results point in the same direction, but they are not identical estimates. The sponsor reports a relative measure, the risk ratio, using a named categorical-outcome method. The re-calculation gives an absolute percentage-point difference using an unadjusted comparison of the posted summaries. A difference in measure and method means the re-calculation is not a reproduction or replacement of the sponsor analysis.

The registry does not report a confidence interval for the sponsor's primary risk ratio. It also does not report the stratification details needed to reconstruct the Cochran-Mantel-Haenszel analysis. The unadjusted confidence interval therefore describes only the re-calculated difference. The results should be read as complementary descriptions of the posted comparison, with the sponsor's reported analysis representing the trial's stated analysis and the re-calculation offering a simple summary contrast.

15. Multiplicity, Missing Data and Analysis Populations

The primary and secondary outcome descriptions state that analyses were done using multiple imputation for the primary endpoint and several threshold-based secondary endpoints. Multiple imputation is, in general, a method for addressing missing outcome data by representing uncertainty across multiple plausible completions of the data. Its interpretation depends on the modeling assumptions, variables included, and relationship between missingness and unobserved outcomes.

The registry record does not report the imputation model, number of imputed datasets, missing-data assumptions, or sensitivity analyses. It also does not give the extent or pattern of missing outcome data. As a result, the robustness of the reported estimates to alternative missing-data assumptions cannot be assessed from the registry record. The statement that multiple imputation was used identifies an approach, but does not describe enough of its implementation to evaluate it fully.

Multiplicity concerns arise because several endpoints are reported. The registry record does not report whether adjustment for multiple comparisons was planned or performed, or whether the secondary outcomes were arranged in a testing hierarchy. The primary and secondary p-values should therefore be distinguished: the primary analysis is the designated primary result, while unadjusted secondary results carry less inferential weight when the approach to multiple testing is unreported.

The analysis populations are not reported in the registry record. In particular, it is not stated whether the reported analyses used all randomized participants, a modified subset, or another defined population. The registry also does not report interim analyses or the complete prespecified statistical analysis plan. Without those details, it is not possible to determine from this record how closely the posted analyses follow the original analysis plan, or how all analysis decisions were made.

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 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: coronary artery stenosis (0.3%), pancreatitis (0.3%), cholelithiasis (0.3%), appendicitis (0.3%), headache (0.3%).

The posted safety figures list serious adverse events for 7 participants in the tapinarof group and 0 in the vehicle group. The corresponding groups at risk are 343 and 172. The record lists 123 participants with other adverse events in the tapinarof group and 28 in the vehicle group. Deaths are listed as 0 in both groups.

These are counts, not estimates of comparative risk adjusted for exposure or follow-up. They should be read alongside the group sizes and the trial's 12-week treatment period, but no rate calculation is made. In general, serious adverse events are a subset defined by seriousness criteria, while the broader other-adverse-event count covers events not categorized as serious in this summary. The registry record does not provide event onset, duration, severity grading, relatedness assessments, or exposure-adjusted rates.

The record lists six serious-event terms, each with a frequency of 0.3%: coronary artery stenosis, pancreatitis, cholelithiasis, appendicitis, headache, and nerve compression. The listed terms are not assigned to a particular group in the available summary. The counts and terms describe events recorded during the trial; by themselves they do not establish causation or provide a comprehensive assessment of safety.

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

The posted primary results show a higher Week 12 percentage meeting the defined PGA response criterion in the tapinarof group than in the vehicle group. The sponsor's reported comparison is statistically classified as superiority, and the simple unadjusted re-calculation also gives a positive mean difference with a confidence interval that excludes zero. These statements describe the observed trial results and the statistical analyses recorded.

They do not establish that every participant experienced improvement, predict an individual participant's response, or describe results beyond Week 12. The endpoint records whether a participant met a particular threshold; it does not represent the full range of symptom experience, the duration of response, or every aspect of disease severity. The registry record does not report outcomes beyond the specified trial period or comparisons with other active treatments.

A statistical difference is not the same as a complete account of clinical importance. The record provides a response percentage, a risk ratio, and an unadjusted percentage-point difference, but it does not report a prespecified threshold for what constitutes a clinically important between-group difference. Nor do the group-level results establish why the groups differed.

Interpretation is also limited by details not reported in the registry record, including the full analysis plan, missing-data assumptions, analysis populations, and multiplicity handling. Multiple imputation is stated for the primary outcome, but its implementation is not described. Safety counts are available, but the summary does not provide enough information to determine causality or compare event rates across differing follow-up exposure.

Accordingly, the findings should be understood as evidence about the recorded Week 12 group comparison in this trial, under the sponsor's reported analysis and the assumptions of the unadjusted re-calculation. They do not answer questions about longer-term outcomes, other populations, other treatments, or individual treatment decisions.

18. Why This Trial Matters Statistically

This trial illustrates how the same clinical comparison can be expressed through different statistical measures. The sponsor's primary result is a risk ratio from a Cochran-Mantel-Haenszel analysis, while the unadjusted re-calculation is a percentage-point difference with a confidence interval. The first describes a relative contrast and the second an absolute contrast. Reporting both makes clear why a single p-value cannot convey the size, scale, and precision of a result.

The primary endpoint also illustrates the interpretive trade-off in using a threshold-based response. A clear or almost clear PGA score paired with a minimum improvement requirement gives a defined response criterion. At the same time, dichotomizing a graded assessment discards information about differences among participants who fall on the same side of the threshold and differences among those who do not meet it. Understanding the endpoint definition is therefore central to understanding the reported percentage.

The secondary outcomes add context by covering related response thresholds and change in affected body-surface area. Their broadly consistent direction is descriptively informative, but the unreported multiplicity strategy means they should not be treated as a set of independent confirmatory tests. This is a practical example of why the prespecified role of each endpoint matters when interpreting several outcomes.

Finally, the record shows how an apparently clear numerical result can coexist with important reporting gaps. Multiple imputation is named, but the details needed to assess its assumptions are absent. The analysis population, original sample-size justification, and full statistical analysis plan are also not reported. Statistical interpretation therefore requires both attention to the effect estimates and care about what the registry record leaves unknown.

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.

The registry record does not report the full protocol or statistical analysis plan, original sample-size justification, analysis-population definitions, interim-analysis information, or detailed handling of missing data. Multiple imputation is named for the primary endpoint, but the imputation model, assumptions, and sensitivity analyses are not reported. These omissions prevent a full assessment of how the posted analyses relate to prespecified methods and how sensitive the conclusions are to analytic choices.

The sponsor's primary risk ratio has no confidence interval reported in the available results. The confidence interval discussed for the primary comparison comes from an unadjusted re-calculation of the posted group summaries and should not be attributed to the sponsor's analysis. The registry does not report stratification details for the Cochran-Mantel-Haenszel method, limiting independent reconstruction of that analysis.

The secondary outcomes cover related measures and multiple comparisons, but the registry record does not report a multiplicity-control strategy. The secondary findings should therefore be interpreted with caution rather than as individually definitive claims. The registry also lists a masking classification of triple while the brief summary calls the study double-blind; the record does not explain this difference.

The safety summary provides counts and selected serious-event terms, but does not give event-level details, exposure-adjusted rates, or relatedness assessments. The trial's reported treatment period is 12 weeks, so the record does not establish longer-term safety or outcomes beyond that period. These limitations concern what can be concluded from the registry record and do not by themselves establish that the underlying trial lacked additional information.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Tapinarof for the Treatment of Plaque Psoriasis in Adults (3002) — NCT03983980.
• Registry record last updated 2025-06-12; status last verified 2022-09.
• Results first posted 2022-10-13.
• 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 randomized, parallel-group Phase 3 study of once-daily tapinarof cream, 1%, versus vehicle cream in adults with plaque psoriasis. Enrollment is listed as 515, and the posted groups are 343 for tapinarof and 172 for vehicle. The study summary describes treatment for 12 weeks.

The primary endpoint is the percentage achieving a PGA score of 0 or 1 with a minimum 2-grade improvement from baseline at Week 12. The sponsor reports a Cochran-Mantel-Haenszel risk ratio of 6.10 and p < 0.001. The unadjusted re-calculation gives a mean difference of 33.9 percentage points, with a 95% confidence interval from 27.14 to 40.66 and p <0.001.

Multiple imputation is stated for the primary outcome, but its implementation and the analysis populations are not reported in the registry record. The original sample-size justification, multiplicity approach, and full statistical analysis plan are also not reported.