Dose Ranging Study of the Safety and Efficacy of R115966 in Plaque Psoriasis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Dose Ranging Study of the Safety and Efficacy of R115966 in Plaque Psoriasis (NCT00716144), a randomized phase 2 study. The primary outcome is Psoriasis Area Severity Index (PASI)75 Success at Visit 6.

2. Clinical Question

3. Trial Design

This was a randomized, phase 2, parallel-group, placebo-controlled dose-ranging trial of oral R115866 for plaque psoriasis. The registry lists double masking. The official title describes the study as evaluator-blind, so the record uses both masking descriptions; further details about who was masked are not reported in the registry record. The listed enrollment was 176.

The four randomized groups were placebo and R115866 at 0.5 mg, 1.0 mg, or 2.0 mg. The brief description says eligible participants would be assigned to one of three dose regimens or placebo. Treatment was planned for 12 weeks, with an 8-week post-treatment follow-up. The registry lists the study as completed, with a start date of 2006-06-01 and primary completion date of 2007-05-01.

The primary endpoint was the number of participants with PASI75 success at Week 12, designated Visit 6. The primary analysis population was the Modified Intent-to-Treat population, defined in the record as randomized participants who took study medication and had post-Visit 2 efficacy data. The endpoint arm counts were 41 for placebo and 45 for each dose group. These endpoint-analysis counts should not be treated as a separate report of the number randomized to each group.

The record also lists three secondary outcome descriptions: PASI50 success at post-baseline visits, Investigator's Global Assessment at post-baseline visits, and PASI75 at post-baseline visits except Visit 6. It reports 4 outcome measures and 12 analyses. The protocol's allocation details, the rationale for the planned enrollment, and the trial's original statistical design assumptions are 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: All participants were screened (Visit 1) two weeks prior to randomization (Visit 2) in order to evaluate their eligibility for entry into the trial. At Visit 1, participants signed an informed consent, provided their medical histories, reported their concomitant medication usages, and were evaluated against the inclusion/exclusion criteria.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified Intent-to-Treat (MITT) population. MITT population comprised of all randomized participants who took study medication and provided post-Visit 2 efficacy data (i.e., had one on-therapy follow-up visit).

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 Psoriasis Area Severity Index (PASI)75 Success at Visit 6, measured over Week 12 (Visit 6).

The primary endpoint counted participants who achieved PASI75 at Week 12. Success was defined as at least 75% reduction in the PASI score at Visit 6 compared with Visit 2, identified as baseline. The result is therefore a binary outcome for each participant: the participant either met the threshold or did not. The reported arm values are counts of participants meeting that definition.

PASI combines the extent of affected body surface area with the severity of erythema, induration, and desquamation across the head and neck, upper extremities, trunk, and lower extremities. Each region contributes an assessment of the amount of affected area and the severity of the signs. The score ranges from 0 to 72 in 0.1-unit intervals, with higher scores indicating worse psoriasis.

A threshold endpoint such as PASI75 expresses change in a clinically recognizable way: it records whether a participant crossed a specified improvement threshold from baseline. It does not describe the full distribution of PASI changes, and participants just below and just above the threshold are counted differently even if their scores are close. The endpoint is also tied to the specified assessment visit, so it does not by itself describe the pattern of response at every visit or after treatment ends.

The registry's primary endpoint is a number of participants, analyzed in the Modified Intent-to-Treat population. The result should therefore be read alongside the population definition and the reported arm counts, rather than as a measure of average PASI change or as an outcome for every person enrolled.

7. Results Posted in the Registry

• R115866 0.5 mg: 15.6% (7/45) versus 9.8% (4/41) on control; risk difference 5.8 percentage points (95% CI -8.2 to 19.8), risk ratio 1.59 (0.50–5.05); unadjusted two-proportion z-test p = 0.421.
• R115866 1.0 mg: 8.9% (4/45) versus 9.8% (4/41) on control; risk difference -0.9 percentage points (95% CI -13.2 to 11.4), risk ratio 0.91 (0.24–3.41); unadjusted two-proportion z-test p = 0.890.
• R115866 2.0 mg: 11.1% (5/45) versus 9.8% (4/41) on control; risk difference 1.4 percentage points (95% CI -11.6 to 14.3), risk ratio 1.14 (0.33–3.95); unadjusted two-proportion z-test p = 0.838.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary: each participant either achieved PASI75 at Visit 6 or did not. For a binary outcome, an arm's event count is the number meeting the endpoint, while the denominator is the number included in that arm's analysis. A comparison can express the difference in observed event proportions as a risk difference, or as a relative measure such as a risk ratio or odds ratio. These measures answer related but distinct questions. The registry's sponsor-reported analysis is not one of those pairwise measures: it is a Cochran-Armitage Trend Test across the four ordered groups.

What does the trend test assess?

In general, a Cochran-Armitage Trend Test assesses whether a binary outcome changes in an ordered way across groups. In this trial the groups are placebo and three ascending R115866 dose levels. The test is suited to asking whether the response pattern follows a trend with dose, rather than testing every active dose separately against placebo. The registry reports p = 0.884 for this test. The record does not describe the scoring or other implementation details used for the test, nor does it report a confidence interval for the trend statistic.

A trend test and pairwise comparisons need not give the same kind of conclusion. A trend test asks about an ordered pattern across all groups. A pairwise comparison asks about the contrast between one dose and placebo. A dose can differ from placebo without producing a consistent trend across all doses; conversely, a trend test can be informative about ordered change without establishing that a particular dose differs from placebo. The observed event counts do not display a steadily increasing pattern across the active doses.

How should the recalculated comparisons be read?

The recalculated risk differences use the posted arm counts and denominators to compare each active dose with placebo. They are described as unadjusted because they do not incorporate covariates or a more complex model. Their confidence intervals quantify uncertainty around the estimated contrast under the calculation method; their p-values assess compatibility with a null difference under the corresponding unadjusted test. The recalculated p-values are 0.421 for R115866 0.5 mg, 0.89 for R115866 1.0 mg, and 0.838 for R115866 2.0 mg.

These comparisons are not the sponsor's reported primary analysis. They are useful for interpreting the arm-level differences but should not replace the sponsor's trend test when describing the registry's reported analysis. The record does not report a sponsor-reported pairwise test for each dose against placebo, or whether the sponsor performed additional comparisons not posted in the registry record.

What does a confidence interval communicate?

In general, a confidence interval gives a range of values compatible with the observed data and the method used to calculate the interval. It conveys precision as well as direction. An interval that includes the null value does not rule out no difference, while an interval that extends in both directions indicates that effects of different signs remain compatible with the data. A confidence interval is not a guarantee that the true effect lies within its bounds, and it does not by itself establish clinical importance.

For all three recalculated risk differences, the intervals include no difference. This does not establish that the treatments are equivalent. An equivalence conclusion requires a prespecified margin and an analysis designed for that purpose; such a margin and analysis are not reported in the registry record. The record describes the trial as dose-ranging and reports a superiority-or-other analysis type, not an equivalence analysis.

What is known about populations and missing data?

The primary outcome population was the Modified Intent-to-Treat population. It includes randomized participants who took study medication and provided post-Visit 2 efficacy data, defined as having at least one on-therapy follow-up visit. This is not a full intention-to-treat definition that necessarily includes every randomized participant, because medication use and post-baseline data are required. The registry does not report how participants without qualifying follow-up data were handled, whether any endpoint values were imputed, or how the analysis addressed missingness.

In general, excluding participants without post-baseline efficacy data can affect an estimate if their outcomes differ systematically from those included. The registry does not provide enough information to assess whether that occurred. The reported analysis denominators should therefore be interpreted in light of the stated Modified Intent-to-Treat definition, not as proof that outcomes were available for every randomized participant.

Why does multiplicity matter?

Multiplicity arises when analyses include several outcomes, visits, doses, or comparisons. Testing many questions can increase the chance of finding an apparently unusual result by chance if no adjustment or prespecified hierarchy is used. The record lists 4 outcome measures, 12 analyses, and three secondary outcome descriptions. It does not report a multiplicity adjustment, a testing hierarchy, or which analyses were designated confirmatory. The trend test is the posted sponsor analysis for this primary endpoint; the recalculated dose-versus-placebo comparisons are unadjusted and should not be treated as multiplicity-controlled evidence.

General statistical interpretation also depends on the analysis plan, endpoint definitions, and handling of missing observations. Those details are not fully available in the registry record. The appropriate reading is therefore bounded: the record provides the endpoint definition, analysis population, counts, and a sponsor trend-test p-value, while important aspects of implementation and multiplicity handling are not reported.

10. Confidence Intervals and Precision

The recalculated risk-difference confidence intervals for each dose-versus-placebo contrast include no difference. This means the data do not pin down a single direction of effect for any of these comparisons. For each contrast, values on both sides of the null remain compatible with the interval and the calculation method.

In general, a confidence interval is a way to express the uncertainty around an estimate, not a range of outcomes that individual participants experienced. A wider interval indicates less precision, while a narrower interval indicates more precision. Here, the intervals span effects in both directions, so the recalculated comparisons do not precisely distinguish an advantage from a disadvantage relative to placebo. The corresponding unadjusted p-values are 0.421 for R115866 0.5 mg, 0.89 for R115866 1.0 mg, and 0.838 for R115866 2.0 mg.

The event counts are 4 in the placebo group and 7, 4, and 5 in the dose groups. With these relatively small observed counts, the estimated comparisons are sensitive to a small number of outcome classifications. That is a general property of binary-outcome estimates when event counts are limited; it does not mean that any particular classification was wrong.

The confidence intervals describe the recalculated unadjusted risk differences, not the sponsor's trend test. The registry does not report a confidence interval for the sponsor's test result. The intervals also do not resolve whether an observed difference would be clinically meaningful: the record does not provide a threshold for a clinically important difference. Their main statistical message is uncertainty, not confirmation of no effect, equivalence, or a precise dose response.

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 sample-size figures are planning illustrations based on the observed effect, not the trial's original design calculation. The registry does not report the trial's own sample-size assumptions or justification. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row uses the stated fraction of the observed effect as the effect to detect.

These figures are participants per arm for 80% power and participants per arm for 90% power. The 80% column describes a planning target that would give the test an 80% chance of detecting the specified effect under the stated assumptions, if that effect were real and the assumptions held. The 90% column applies the corresponding higher power target. These are hypothetical planning quantities, not counts of people who took part in this trial.

The table gives separate illustrations for planning at 100%, 75%, and 50% of the observed effect. In general, a smaller effect is harder to distinguish from random variation, so detecting a smaller effect requires more participants. The planning quantities can become especially large when the observed contrast is close to zero, as in the recalculated comparison for R115866 1.0 mg. Such values reflect the arithmetic of planning around the observed estimate; they do not show that the trial had a particular probability of success.

The illustration compares two arms at a time and specifies equal allocation between those arms. It should not be read as a total sample size for the four-group trial. These figures are not observed power.

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 = 0.098; treatment proportion p₁ = 0.156; average p̄ = 0.127.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4702 = 0.9216.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4684 = 0.3942.
4. n = (0.9216 + 0.3942)² / (0.0580)² = 514.7, rounded up to 515 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 <- 0.0975609756097561                          # control proportion
p1 <- 0.155555555555556                          # 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(716144)
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 = 515 (80% power), 689 (90% power)
 75% of observed effect: n per arm = 871 (80% power), 1,166 (90% power)
 50% of observed effect: n per arm = 1,857 (80% power), 2,485 (90% power)
power.prop.test: 514.7 
Simulated power at n = 515 per arm: 0.802 (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 reports the sponsor's analysis of PASI75 as a Cochran-Armitage Trend Test across placebo and the dose groups, with p = 0.884. This is the sponsor-reported result for the primary endpoint. It addresses an ordered trend across groups rather than supplying a separate sponsor-reported risk difference for each dose against placebo.

The recalculated estimates are unadjusted pairwise contrasts built from the posted event counts and analysis denominators. Their p-values are 0.421 for R115866 0.5 mg, 0.89 for R115866 1.0 mg, and 0.838 for R115866 2.0 mg. These calculations answer a different question from the trend test and do not replace it. The registry does not report whether the sponsor used additional adjusted pairwise analyses or other methods beyond the posted trend test.

No discrepancy is flagged between the posted outcome data and the recalculated summaries. Even so, the sponsor's analysis remains the appropriate description of the analysis reported by the registry. The recalculations are supplementary context, not a correction to or substitute for the sponsor result. Neither the sponsor trend test nor the unadjusted comparisons establish that the dose groups are equivalent.

14. Multiplicity, Missing Data and Analysis Populations

The record lists 4 outcome measures and 12 analyses, including the primary PASI75 endpoint and secondary measures concerning PASI50, Investigator's Global Assessment, and PASI75 at other post-baseline visits. The record does not report a multiplicity adjustment, a testing hierarchy, or a clear account of which analyses were intended to be confirmatory. Without that information, the analyses should not be treated as if each were an isolated test with no risk of chance findings across multiple outcomes or visits.

The sponsor-reported primary analysis is a single Cochran-Armitage Trend Test across the four groups, with p = 0.884. The recalculated pairwise p-values are unadjusted. They should be read as descriptive contrasts rather than as multiplicity-controlled tests. The record does not say whether the sponsor adjusted other analyses for multiple comparisons.

The endpoint population was the Modified Intent-to-Treat population: randomized participants who took study medication and had post-Visit 2 efficacy data, meaning at least one on-therapy follow-up visit. The record reports arm analysis counts for the endpoint but does not explain how participants lacking qualifying data were handled, whether missing endpoint values were imputed, or whether sensitivity analyses were performed.

In general, missing outcomes can affect estimates when the chance of missingness is related to the unobserved outcome or differs between groups. The registry record does not provide information to assess those possibilities. It also does not report further details about adherence handling, analysis covariates, or protocol deviations for this endpoint. Interpretation should therefore remain specific to the reported Modified Intent-to-Treat population and the analysis details that are stated, rather than assuming an unreported missing-data or multiplicity strategy.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: prostate cancer (2.4%), jaw fracture (2.2%).

The serious adverse event figures are counts of affected participants in each reported group. The registry lists 1 participant with a serious adverse event in the placebo group, 1 in the R115866 0.5 mg group, and 0 in each of the R115866 1.0 mg and 2.0 mg groups. Deaths are reported as 0 in every group. These counts should be read with the corresponding at-risk counts: 41 for placebo and 45 for each R115866 dose.

The record also lists prostate cancer at 2.4% and jaw fracture at 2.2% among the top serious adverse event terms. It does not identify the treatment-group attribution for those terms in the information, so they should not be assigned to a particular arm. The listed figures do not establish whether an event was caused by treatment.

Other adverse events affected 20 participants in the placebo group, 34 in the 0.5 mg group, 38 in the 1.0 mg group, and 37 in the 2.0 mg group. These are counts, not a comparison adjusted for exposure or other factors. The registry record does not provide enough detail in these figures to characterize event duration, severity beyond the serious/non-serious grouping, or causal assessment. Sparse serious-event counts also limit comparisons across groups.

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

The posted primary endpoint results show PASI75 success counts of 4 with placebo, 7 with R115866 0.5 mg, 4 with R115866 1.0 mg, and 5 with R115866 2.0 mg at Week 12. The sponsor's reported Cochran-Armitage Trend Test has p = 0.884. The pattern of counts does not show a clear increase with dose, and the sponsor test does not provide evidence of an ordered trend in these data.

The unadjusted recalculations compare each dose with placebo. All three confidence intervals include no difference, and the p-values are 0.421, 0.89, and 0.838 for the 0.5 mg, 1.0 mg, and 2.0 mg comparisons, respectively. These findings mean the posted data do not precisely establish a difference from placebo in any of those recalculated contrasts. They do not prove that there is no effect, and they do not demonstrate equivalence. The intervals allow for uncertainty in both directions.

The endpoint has a specific scope: PASI75 at Week 12 among participants in the Modified Intent-to-Treat population who took medication and had post-Visit 2 efficacy data. It is not a complete description of PASI change over time, outcomes after the follow-up period, or the secondary outcomes. The registry's listed 8-week post-treatment follow-up is part of the study description, but the primary endpoint result is tied to Visit 6 at Week 12.

The safety counts provide a separate descriptive view. Serious adverse events were reported in small counts, and deaths were reported as 0 in all groups. These data do not by themselves establish comparative safety or causation. Overall, the result is best understood as an imprecise dose-ranging result on a binary Week 12 endpoint, rather than a definitive answer about every dose, outcome, or participant.

17. Why This Trial Matters Statistically

This trial illustrates the difference between testing for a dose trend and comparing each dose with placebo. The sponsor's Cochran-Armitage Trend Test assesses an ordered pattern across placebo and three doses, while the recalculated risk differences address separate pairwise contrasts. Those methods answer related but non-identical questions, so their results should not be collapsed into a single claim.

It also shows why event counts and estimates need to be read together. The arm counts are straightforward, but the uncertainty around the recalculated differences is substantial enough that each interval includes no difference. The counts alone can suggest a pattern, but they do not establish that pattern as a reliable dose response. Here, the largest success count is in the lowest active-dose group, not the highest, underscoring why the trend analysis matters.

The endpoint further demonstrates the trade-off involved in dichotomizing a continuous clinical measure. PASI75 offers an interpretable threshold, but it reduces the information in participants' actual PASI scores to whether they reached a cut point. It cannot show how close participants were to the threshold or describe the size of score changes among those who did not cross it.

Finally, interpretation depends on details beyond the p-value: the population definition, handling of missing data, multiplicity plan, and precision of the estimates. The registry reports some of these elements but does not report others. A statistically careful reading therefore recognizes the reported outcome and analysis without extending them into unsupported claims about equivalence, clinical importance, or longer-term effects.

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.

The primary outcome analysis applies to the Modified Intent-to-Treat population, not necessarily every randomized participant. The registry defines this population by medication use and post-Visit 2 efficacy data, but does not report the number excluded for failing to meet those criteria or explain how missing outcomes were handled. This leaves uncertainty about the consequences of missing post-baseline data.

The sponsor-reported analysis is a trend test. The registry does not report a sponsor-reported pairwise comparison for each dose against placebo, nor does it provide a confidence interval for the trend result. The recalculated pairwise contrasts are unadjusted, and the record does not describe a multiplicity adjustment or hierarchy despite listing multiple outcomes and analyses. Those limitations constrain claims based on isolated contrasts.

The endpoint is a binary threshold at Week 12. The record does not provide the full distribution of PASI changes for this primary result, which would give more detail than whether participants met PASI75. It also does not establish the clinical-importance threshold for a difference in PASI75 success. The primary endpoint result does not describe persistence after treatment or replace the separately listed secondary outcomes.

The trial is described as double-masked in the registry, while its official title says evaluator-blind; details reconciling those descriptions are not reported. The record does not report the original sample-size justification or the full statistical analysis plan. The listed serious adverse event terms are not attributed to treatment groups in the reported information. These record-level gaps limit assessment of implementation, precision, and safety comparisons.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Dose Ranging Study of the Safety and Efficacy of R115966 in Plaque Psoriasis — NCT00716144.
• Registry record last updated 2018-01-29; status last verified 2017-05.
• Results first posted 2018-01-29.
• 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

NCT00716144 is a completed phase 2 randomized, parallel-group, placebo-controlled dose-ranging trial of oral R115866 for plaque psoriasis. The registry lists double masking, while the official title describes evaluator blinding. The listed enrollment was 176, and the planned treatment period was 12 weeks, followed by an 8-week post-treatment follow-up.

The primary endpoint was PASI75 success at Week 12, defined as at least 75% reduction from baseline. The analysis used a Modified Intent-to-Treat population. The sponsor-reported Cochran-Armitage Trend Test had p = 0.884. Recalculated unadjusted dose-versus-placebo comparisons have confidence intervals that include no difference. The registry also reports serious adverse event counts and 0 deaths in each group. Original sample-size assumptions, detailed missing-data handling, and multiplicity procedures are not reported in the registry record.