LEO 90100 Compared to Vehicle in Subjects With Psoriasis Vulgaris: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for LEO 90100 Compared to Vehicle in Subjects With Psoriasis Vulgaris (NCT01866163), a randomized phase 3 study. The primary outcome is Treatment Success According to IGA.

2. Clinical Question

The record reports treatment success for 53.3% of participants assigned to LEO 90100 and 4.8% assigned to vehicle. The posted comparison gives a risk difference of 0.484, with a confidence interval and an unadjusted p-value. These statistics describe the observed groups and the precision of the comparison; they do not, by themselves, establish what would happen in other populations or settings.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group phase 3 trial. The registry lists an enrollment of 426 and marks the study as completed. The listed study start was 2013-06 and primary completion was 2013-10. LEO Pharma is identified as the sponsor. The trial compared LEO 90100 with vehicle in subjects with psoriasis vulgaris, with treatment intended to continue for up to 4 weeks.

The registry reports 323 participants in the LEO 90100 arm and 103 in the vehicle arm for the primary outcome. The primary endpoint was assessed at Week 4. The outcome population is described as all randomized subjects in the full analysis set, analyzed for efficacy. The record’s masking designation is quadruple, but the specific roles or procedures included in that designation are not reported in the registry record. Details of the randomization sequence, allocation concealment, and the original allocation rationale are also not reported in the registry record.

The registry lists two secondary outcomes: m-PASI at Week 4 and m-PASI at Week 1. It records three outcome measures and three analyses. Further definitions, results, and analysis details for the secondary outcomes are not included in the registry record for this account, so no conclusion about those outcomes can be drawn. The record does not report the original sample-size justification or the assumptions used to design the trial. The planning illustrations associated with the observed primary effect are separate from that original design information.

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: Prior to randomisation, the subject entered a washout phase (if required) where anti-psoriatic treatment and other relevant medication/treatments were discontinued as defined by the exclusion criteria. The wash-out/screening phase could last for up to 4 weeks, depending on which disallowed treatments the subject received.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All randomised subjects were included in the full analysis set and analysed for efficacy.

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 Treatment Success According to IGA, measured over 4 weeks.

The primary endpoint was the percentage of subjects with IGA treatment success at Week 4. The Investigators’ global assessment used a five-point scale: 1 meant clear, 2 almost clear, 3 mild, 4 moderate, and 5 severe. The registry defines success in relation to baseline severity. For participants with at least moderate disease at baseline, success meant clear or almost clear; for participants with mild disease at baseline, it meant clear.

This endpoint turns a clinician’s ordered assessment into a binary outcome: each participant either meets the stated success criterion or does not. Reporting the percentage meeting that criterion makes the result interpretable as a group proportion at a prespecified assessment time. It differs from reporting a mean scale score because it focuses on whether participants cross the defined threshold, rather than the average position on the scale.

Threshold-based endpoints can be useful when the clinical question is whether participants reach a specified state. Their interpretation depends on the scale, the threshold, the timing, and the baseline-dependent definition. The record gives those elements for the primary endpoint, but further detail about assessment procedures or assessor training is not reported in the registry record.

7. Results Posted in the Registry

• LEO 90100: 53.3% (172/323) versus 4.9% (5/103) on control; risk difference 48.4 percentage points (95% CI 41.6 to 55.2), risk ratio 10.97 (4.64–25.95); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary: a participant either met the IGA treatment-success definition at Week 4 or did not. For this kind of outcome, the group percentage estimates the fraction meeting the criterion. The record reports 53.3% for LEO 90100 and 4.8% for vehicle. These are descriptive summaries of the analyzed groups, not guarantees about an individual participant’s outcome.

What does the risk difference express?

The risk difference subtracts the outcome proportion in one group from the outcome proportion in the other. It expresses an absolute contrast in the probability of meeting the endpoint, rather than a relative contrast. The unadjusted recalculation reports a risk difference of 0.484, with a 95% confidence interval from 0.415 to 0.552. A positive estimate means the observed success proportion was higher in the LEO 90100 group than in the vehicle group under the stated ordering. The interval describes uncertainty around that estimated contrast under the calculation’s assumptions.

What does an odds ratio express?

An odds ratio compares the odds of meeting the endpoint between groups. Odds are not the same as probabilities: they express the probability of the event relative to the probability of not meeting the event. An odds ratio of 1 corresponds to equal odds under the comparison’s definition. Values above 1 indicate higher odds in the group named in the comparison when the other group is the reference. Because odds and probabilities are different quantities, an odds ratio should not be read as a probability difference or as a direct multiplier of the success percentage.

The registry’s reported Mantel-Haenszel odds ratio is 30.27, with a 95% confidence interval from 9.72 to 94.30. The posted recalculation is an unadjusted odds ratio of 22.33, with a 95% confidence interval from 8.855 to 56.29. The two estimates are not interchangeable: the sponsor-reported method and the unadjusted recalculation use different approaches. The record does not report enough detail to reconstruct the sponsor’s analysis beyond naming the Mantel-Haenszel method. The discrepancy flag is false, and the sponsor’s reported analysis is the registry’s own analysis.

What does the p-value tell the reader?

The unadjusted p-value is <0.001, and the sponsor-reported p-value is also <0.001. In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions held. It is not the probability that the null model is true. It does not measure the size or importance of an effect, and a small p-value does not remove uncertainty about bias, missing data, measurement, or generalizability.

The registry identifies the sponsor’s test as a Mantel-Haenszel analysis of an odds ratio and labels the analysis as a superiority-or-other legacy type. The detailed test specification, including any stratification variables used in the Mantel-Haenszel calculation, is not reported in the registry record. The unadjusted risk difference, odds ratio, and their intervals are recalculations from posted summary statistics; they are not substitutes for the sponsor’s own analysis.

Why can estimates differ by method?

Statistical methods may target the same general comparison while using different scales or adjustments. A risk difference is an absolute contrast; an odds ratio is a relative comparison of odds. An adjusted or stratified estimate can also differ from a simple unadjusted estimate because it accounts for the analysis structure specified by that method. The registry does not provide the detailed analysis specification needed to determine which particular features account for the difference in odds-ratio estimates. The values should therefore be presented with their methods, rather than treated as conflicting versions of one identical calculation.

What does over-dispersion mean for this outcome?

Over-dispersion is a general issue for some statistical models when variability in the observations is greater than the model expects. It is especially relevant to count outcomes and clustered data. This primary endpoint is reported as a binary percentage, and the registry does not report an over-dispersion model or a related adjustment. There is no basis in the registry record for asserting that over-dispersion affected this analysis.

In general, sound interpretation also depends on how the outcome population and missing measurements are handled. The record says all randomized subjects were included in the full analysis set and analyzed for efficacy, but it does not state the missing-data method or how many Week 4 values, if any, were unavailable. It also does not report interim analyses or a multiple-testing adjustment. Those omissions do not alter the posted numbers, but they limit what can be inferred about the analysis process.

10. Confidence Intervals and Precision

The unadjusted risk difference is 0.484, with a 95% confidence interval from 0.415 to 0.552. This interval gives a range of values reasonably compatible with the observed data under the calculation’s assumptions and method. It is not a range containing a known fixed proportion of individual outcomes, and it does not mean that the true effect has a 95% probability of falling within the interval.

The interval is entirely above zero, the value corresponding to no risk difference. This conveys that the unadjusted comparison is inconsistent with an equal-proportion contrast under the assumptions used for the interval. The unadjusted p-value is <0.001. The interval and p-value are related summaries of evidence for this comparison, but they answer different practical questions: the interval communicates a range of effect sizes, whereas the p-value summarizes compatibility with a specified null model.

The unadjusted odds ratio is 22.33, with a 95% confidence interval from 8.855 to 56.29. The sponsor-reported Mantel-Haenszel odds ratio is 30.27, with a 95% confidence interval from 9.72 to 94.30. Both intervals are wide in absolute terms, so the magnitude of the odds-ratio estimate is not pinned down precisely, even though the intervals lie above the null value for an odds ratio. The intervals should be interpreted in relation to their respective methods, not as if they were estimates on the same scale as the risk difference.

Confidence intervals quantify sampling uncertainty under a statistical model; they do not account automatically for all limitations in design, outcome measurement, missing data, or selective reporting. The registry does not report enough information to assess all of those sources. The percentages themselves are reported without confidence intervals in the record, so the recalculated intervals are the available interval estimates for the primary comparison.

11. 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 illustrations are based on the observed effect, not on the trial’s original design assumptions. They assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario targets a stated fraction of the observed effect. For this binary outcome, the figures are participants per arm, not total event counts.

The 80% power column gives the illustrative number of participants per arm for a design that would have an 80% chance of detecting the target effect under the stated assumptions, if that effect were the true effect and the model assumptions held. The 90% power column gives the corresponding planning figure for a design targeting a 90% chance. These are design calculations, not estimates of the chance that the completed trial’s result is correct.

The listed scenarios are planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The figures are 13 and 17 participants per arm for 100% of the observed effect; 20 and 27 for 75%; and 37 and 49 for 50%, respectively, for the 80% and 90% power targets. As the effect to be detected becomes smaller, more participants are needed under these assumptions. This is not observed power.

These illustrations use the observed effect as an input, so they should not be mistaken for the rationale behind the actual enrollment. The original trial sample-size assumptions and justification are not reported in the registry record. The planning figures also depend on the assumptions stated above and may not represent a design with unequal allocation or a different analysis method.

12. 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 = 0.049; treatment proportion p₁ = 0.533; average p̄ = 0.291.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6421 = 1.2584.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5433 = 0.4572.
4. n = (1.2584 + 0.4572)² / (0.4840)² = 12.6, rounded up to 13 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 <- 0.0485436893203883                          # control proportion
p1 <- 0.532507739938081                          # 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(1866163)
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 = 13 (80% power), 17 (90% power)
 75% of observed effect: n per arm = 20 (80% power), 27 (90% power)
 50% of observed effect: n per arm = 37 (80% power), 49 (90% power)
power.prop.test: 12.6 
Simulated power at n = 13 per arm: 0.853 (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.

13. Analyses Reported by the Sponsor

The registry’s own analysis reports a Mantel-Haenszel odds ratio of 30.27, with a 95% confidence interval from 9.72 to 94.30 and p < 0.001. The recalculation from posted summary statistics gives an unadjusted odds ratio of 22.33, with a 95% confidence interval from 8.855 to 56.29. It also gives an unadjusted risk difference of 0.484, with a 95% confidence interval from 0.415 to 0.552 and p < 0.001.

The sponsor’s odds ratio and the recalculated odds ratio use different methods. The sponsor’s estimate is Mantel-Haenszel; the recalculation is explicitly unadjusted. The registry record does not report the detailed specifications needed to reproduce the Mantel-Haenszel analysis. The discrepancy flag is false, but the method distinction remains important when interpreting the estimates.

The registry’s reported analysis takes precedence as the sponsor’s own analysis. The recalculated estimates are useful as simple, unadjusted comparisons of the posted group summaries, not as a replacement or correction to the sponsor’s analysis. Both approaches report p < 0.001, but p-values do not convey effect magnitude on their own; interpretation should also consider the reported effect estimates and confidence intervals.

14. Multiplicity, Missing Data and Analysis Populations

The primary endpoint is identified as IGA treatment success at Week 4. The registry lists two secondary outcomes, m-PASI at Week 4 and m-PASI at Week 1, and records three outcome measures and three analyses. The registry entries available for this account contain no result details for those secondary outcomes. They also do not state whether the trial used a prespecified hierarchy, adjusted significance thresholds, or another strategy to account for testing multiple outcomes. Multiplicity handling is therefore not reported in the registry record.

In general, testing multiple outcomes or multiple hypotheses can increase the chance of at least one apparently unusual result when all null hypotheses are true. A prespecified testing strategy can help control that issue. Without the analysis plan or a stated adjustment procedure, the primary p-value should be read as the reported result for that endpoint, not as evidence that all possible testing decisions have been accounted for.

The primary analysis population is described as all randomized subjects in the full analysis set, analyzed for efficacy. The registry gives the arm counts for the primary outcome, but it does not report how missing Week 4 assessments were handled, whether any values were imputed, or how participants without endpoint data were classified. It also does not give a separate count of participants contributing evaluable endpoint data. The reported population definition alone is not enough to determine the exact missing-data procedure.

The record does not report interim analyses, a statistical analysis plan, or the original sample-size justification. These details cannot be inferred from the posted outcome summary. Their absence limits assessment of prespecification, multiplicity, and the relationship between the analysis population and any missing observations; it does not change the numerical endpoint results quoted above.

15. 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 0.6% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: substance induced psychotic disorder (0.3%), bipolar disorder (0.3%).

The adverse-event summary reports serious adverse events for 2 participants in the LEO 90100 group and 0 in the vehicle group. The corresponding at-risk counts are 323 and 103. The record lists the serious-event terms Substance induced psychotic disorder and Bipolar disorder, each reported as 0.3%. It does not provide additional detail about timing, duration, severity beyond the serious-event classification, or the relationship of each event to treatment in the registry record used.

For other adverse events, the registry reports 18 in the LEO 90100 group and 15 in the vehicle group, with at-risk counts of 323 and 103. These are event-affecting counts as presented in the record; they should not be interpreted as rates without a specified calculation, and no rates are calculated. The death fields have no reported values in the registry record.

Adverse-event counts are descriptive safety information. They do not establish that a treatment caused an event or that an event was unrelated to treatment. The summary does not report event-level narratives, exposure-adjusted rates, or statistical comparisons for these outcomes. Differences in group counts should therefore be read alongside the group sizes and the limitations of the reported safety detail, without inferring causality.

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

The posted primary results show that IGA treatment success at Week 4 was reported for 53.3% of participants assigned to LEO 90100 and 4.8% assigned to vehicle. The unadjusted risk difference is 0.484, with a 95% confidence interval from 0.415 to 0.552. The sponsor-reported Mantel-Haenszel odds ratio is 30.27, with a 95% confidence interval from 9.72 to 94.30 and p < 0.001. These figures describe a difference between the trial groups for the registry-defined endpoint and time point.

The result does not establish that every participant receiving LEO 90100 met the success criterion, or that no participant receiving vehicle did so. It does not describe the course of disease for each individual. Nor does a result at Week 4 establish how long a response lasted, what happened after the assessed period, or how outcomes differed on the listed secondary m-PASI measures; those secondary results are not reported in the registry record.

The result is specific to the population, endpoint definition, treatment comparison, and follow-up described in the record. The registry identifies randomized allocation and quadruple masking, but details of the procedures are not reported in the registry record. Information about missing outcome handling, multiplicity, and the original sample-size justification is also not reported in the registry record. These omissions constrain assessment of the full analysis context.

Statistical evidence is not the same as a complete judgment about clinical importance, safety, or applicability to other settings. The confidence intervals express uncertainty under their statistical methods, while the p-values express compatibility with a null model under test assumptions. Neither resolves questions about measurement quality, missing data, or broader relevance. This statistical description does not constitute medical advice or a recommendation about treatment.

17. Why This Trial Matters Statistically

This trial illustrates how the choice of effect measure shapes interpretation of a binary endpoint. The primary outcome is the percentage meeting a defined IGA success threshold at a fixed assessment time. The risk difference communicates an absolute contrast, while the odds ratio communicates a relative contrast in odds. They answer related but distinct questions, and the odds ratio should not be read as though it were a percentage-point difference.

The comparison also shows why the analysis method should accompany an estimate. The registry reports a Mantel-Haenszel odds ratio, while the simple recalculation gives an unadjusted odds ratio. Their numerical values differ, and the registry does not include enough method detail to reproduce the sponsor’s estimate. Reporting the method, estimate, and interval together gives readers a clearer basis for understanding what each number represents.

Precision matters as well as the direction of an estimate. The risk-difference interval is separated from zero, while the odds-ratio intervals span a broad range of values. A small p-value alone cannot communicate that range or indicate whether an effect is important in practice. Confidence intervals help show the scale of uncertainty, although they do not account for every source of uncertainty in a trial.

Finally, a strong primary comparison does not answer every question about a treatment or study. The record lists secondary endpoints and safety outcomes, but the available details are limited. Missing-data handling, multiplicity procedures, and original design assumptions are not reported in the registry record. Recognizing these boundaries is part of interpreting the primary result responsibly.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry summary provides a clear primary endpoint definition and group percentages, but it does not report the original sample-size justification or the assumptions used in planning the completed trial. The planning illustrations based on the observed effect cannot fill that gap. They answer a separate hypothetical design question and do not reveal what assumptions guided the trial.

The analysis description is limited. The sponsor’s primary comparison is identified as a Mantel-Haenszel odds ratio, but the registry record does not state the stratification variables or other details needed to reproduce the analysis. The record also does not describe randomization sequence generation, allocation concealment, or the specific procedures covered by quadruple masking.

The primary population is identified as the full analysis set of all randomized subjects, but missing Week 4 data handling is not reported in the registry record. The record does not state whether data were imputed, how incomplete assessments were treated, or how many participants lacked endpoint measurements. The registry entries used also do not include results for the listed m-PASI secondary outcomes, despite the registry listing those measures.

Multiplicity procedures and interim analyses are not reported in the registry record. Further safety detail is limited: the summary provides serious-event and other-event counts and two serious-event terms, but not event narratives or treatment-relatedness assessments. Death figures are not reported in the registry record. These limitations restrict independent assessment of the full analysis and safety context and the extent to which the primary estimate applies beyond the recorded comparison.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: LEO 90100 Compared to Vehicle in Subjects With Psoriasis Vulgaris — NCT01866163.
• Registry record last updated 2025-03-10; status last verified 2019-09.
• Results first posted 2016-05-09.
• 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.

22. Record Summary

The registry describes a completed, randomized, quadruple-masked, parallel-group phase 3 trial sponsored by LEO Pharma. It compared LEO 90100 aerosol foam, containing calcipotriol 50 mcg/g plus betamethasone 0.5 mg/g as dipropionate, with aerosol foam vehicle in subjects with psoriasis vulgaris. The listed enrollment is 426, and treatment was intended for up to 4 weeks.

The primary endpoint was IGA treatment success at Week 4, defined according to baseline disease severity. The reported success percentages were 53.3% for LEO 90100 and 4.8% for vehicle. The sponsor-reported Mantel-Haenszel odds ratio was 30.27, with a 95% confidence interval from 9.72 to 94.30 and p < 0.001. The record also lists m-PASI at Week 4 and Week 1 as secondary outcomes.

The registry says all randomized subjects were included in the full analysis set and analyzed for efficacy. Details of missing-data handling, multiplicity, and the original design assumptions are not reported in the registry record.