A Study To Investigate The Safety And Efficacy Of CP- 690,550 In Patients With Moderate And Severe Ulcerative Colitis.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study To Investigate The Safety And Efficacy Of CP- 690,550 In Patients With Moderate And Severe Ulcerative Colitis. (NCT00787202), a randomized phase 2 study. The primary outcome is Percentage of Participants With Clinical Response.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group, placebo-controlled phase 2 study in participants with moderate to severe ulcerative colitis. The registry records triple masking and a randomized allocation. It was conducted at multiple centers, as stated in the official title, and is marked completed. The sponsor was Pfizer.

The record lists five arms: placebo and CP-690,550 at 0.5 mg, 3 mg, 10 mg, and 15 mg. Arm sizes for the primary clinical-response outcome were 40 participants for placebo, 27 for 0.5 mg, 31 for 3 mg, 30 for 10 mg, and 45 for 15 mg. These are the outcome-specific counts, not a substitute for the planned enrollment figure of 195; the registry does not explain the difference between those figures.

The study began in 2008-12 and primary completion was recorded as 2010-09. Its main efficacy assessment was at Week 8. The registry also names secondary outcomes assessed at Week 8 and at other stated time points, including changes in symptom scores and inflammatory markers, as well as plasma concentration of CP-690,550.

The registry record does not report the trial’s original sample-size justification, the planned allocation ratio, the detailed statistical analysis plan, or whether interim analyses were planned or conducted. The record identifies the masking and general design but does not give further operational details about how masking was maintained. Those omissions limit what can be inferred about the original design choices; they do not change the descriptive meaning of the posted outcome counts.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS): all participants who withdrew as treatment failure (TF) or completed \>=1 week dosing, had \>=1 valid Mayo score during active double-blind phase. Participants who withdrew as TF=non-responders. Missing data due to reason other than TF were excluded. N (number of participants analyzed)=evaluable participants for the measure.

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 Percentage of Participants With Clinical Response, measured over Week 8.

The primary endpoint was the percentage of participants with clinical response at Week 8. The registry defines response as a decrease from baseline in Mayo score of at least 3 points and at least 30 percent, together with either a decrease in the rectal-bleeding subscore of at least 1 point or an absolute rectal-bleeding subscore of 0 or 1. Thus, the endpoint combines a change in overall disease activity with a specified improvement in rectal bleeding.

The Mayo score is described as a measure of ulcerative-colitis disease activity, ranging from 0 to 12, with higher scores indicating more severe disease. It combines stool frequency, rectal bleeding, findings from flexible proctosigmoidoscopy, and physician global assessment. Each component has a range from 0 to 3.

A response endpoint is used to capture improvement relative to baseline, rather than requiring that symptoms and examination findings fall below a fixed threshold. This makes it distinct from clinical remission, which is separately listed as a secondary outcome. The registry’s hypothesis discusses inducing remission, but the registered primary measure is clinical response. Keeping that distinction clear is important when interpreting the results: the primary percentage describes participants meeting the stated response rule, not the percentage meeting the separate remission definition.

7. Results Posted in the Registry

• CP-690,550 0.5 mg: 29.6% (8/27) versus 47.5% (19/40) on control; risk difference -17.9 percentage points (95% CI -41.0 to 5.3), risk ratio 0.62 (0.32–1.21); unadjusted two-proportion z-test p = 0.144.
• CP-690,550 3 mg: 51.6% (16/31) versus 47.5% (19/40) on control; risk difference 4.1 percentage points (95% CI -19.3 to 27.5), risk ratio 1.09 (0.68–1.74); unadjusted two-proportion z-test p = 0.731.
• CP-690,550 10 mg: 63.3% (19/30) versus 47.5% (19/40) on control; risk difference 15.8 percentage points (95% CI -7.3 to 39.0), risk ratio 1.33 (0.87–2.04); unadjusted two-proportion z-test p = 0.188.
• CP-690,550 15 mg: 80.0% (36/45) versus 47.5% (19/40) on control; risk difference 32.5 percentage points (95% CI 13.1 to 51.9), risk ratio 1.68 (1.18–2.41); unadjusted two-proportion z-test p = 0.002.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percentage of Participants With Clinical Remission

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • CP-690,550 0.5 mg: 7.4% (2/27) versus 12.2% (5/41) on control; risk difference -4.8 percentage points (95% CI -18.9 to 9.3), risk ratio 0.61 (0.13–2.91); unadjusted two-proportion z-test p = 0.525.
  • CP-690,550 3 mg: 35.5% (11/31) versus 12.2% (5/41) on control; risk difference 23.3 percentage points (95% CI 3.7 to 42.9), risk ratio 2.91 (1.13–7.51); unadjusted two-proportion z-test p = 0.019.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CP-690,550 0.5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for CP-690,550 3 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With Endoscopic Response

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • CP-690,550 0.5 mg: 51.9% (14/27) versus 55.0% (22/40) on control; risk difference -3.1 percentage points (95% CI -27.5 to 21.2), risk ratio 0.94 (0.60–1.49); unadjusted two-proportion z-test p = 0.800.
  • CP-690,550 3 mg: 61.3% (19/31) versus 55.0% (22/40) on control; risk difference 6.3 percentage points (95% CI -16.8 to 29.3), risk ratio 1.11 (0.75–1.66); unadjusted two-proportion z-test p = 0.595.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CP-690,550 0.5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CP-690,550 3 mg and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants With Endoscopic Remission

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • CP-690,550 0.5 mg: 7.4% (2/27) versus 2.4% (1/41) on control; risk difference 5.0 percentage points (95% CI -6.0 to 15.9), risk ratio 3.04 (0.29–31.87); unadjusted two-proportion z-test p = 0.329.
  • CP-690,550 3 mg: 19.4% (6/31) versus 2.4% (1/41) on control; risk difference 16.9 percentage points (95% CI 2.2 to 31.6), risk ratio 7.94 (1.01–62.57); unadjusted two-proportion z-test p = 0.016.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CP-690,550 0.5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for CP-690,550 3 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry names eight secondary outcomes, including clinical remission, endoscopic response, endoscopic remission, changes in partial Mayo score, IBDQ score, CRP, and fecal calprotectin, and plasma concentration of CP-690,550. Numerical results are posted for three Week 8 outcomes: clinical remission, endoscopic response, and endoscopic remission. Results for the other listed secondary outcomes are not reported in the registry record.

Across the posted secondary outcomes, the arm patterns are not identical. Some dose groups have higher observed percentages than placebo for one measure, while the relative ordering of dose groups differs across measures. This is not unexpected because clinical response, clinical remission, endoscopic response, and endoscopic remission use distinct definitions and represent different aspects of disease activity. These measures can add context to the primary clinical-response result, but they are not interchangeable outcomes and should not be treated as repetitions of the same test.

The secondary findings are most useful as descriptive context for the primary endpoint. Similar directions across related measures can be informative, while differences between measures may reflect distinct definitions, outcome denominators, or uncertainty. The registry does not report sponsor analyses for the posted secondary comparisons. The recalculated secondary comparisons are unadjusted, and multiple outcomes and dose comparisons create more opportunities for chance findings.

For that reason, an isolated unadjusted secondary p-value should carry less weight than a result supported by a prespecified analysis that accounts for the full set of outcomes and comparisons. Such a plan is not reported in the registry record. The absence of posted results for the other listed secondary outcomes also means the available secondary evidence is incomplete. It is not possible to infer from the registry record whether those outcomes were analyzed without results, not measured, or omitted for another reason.

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

• Percentage of Participants With Clinical Remission
• Percentage of Participants With Endoscopic Response
• Percentage of Participants With Endoscopic Remission
• Change From Baseline in Partial Mayo Score at Week 2, 4, 8 and 12
• Change From Baseline in Total Inflammatory Bowel Disease Questionnaire (IBDQ) Score at Week 8
• Change From Baseline in Level of C-Reactive Protein (CRP) at Week 4 and 8
• Change From Baseline in Level of Fecal Calprotectin at Week 2, 4, 8 and 12
• Plasma Concentration of CP-690,550

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: each participant either met the registry’s clinical-response definition at Week 8 or did not. A percentage summarizes the proportion meeting that definition in each arm. For an arm percentage, the numerator is the number of participants classified as responders and the denominator is the number evaluated for that outcome. The outcome counts and arm sizes describe observed results; they do not by themselves explain why a participant was or was not included in a particular denominator.

How are the dose comparisons expressed?

The posted comparisons use risk differences. A risk difference subtracts the response proportion in placebo from the response proportion in a dose group. A positive value means that the observed response proportion was higher in the dose group; a negative value means it was lower. The registry reports the differences on the proportion scale, for example 0.325 for 15 mg versus placebo. This estimate is unadjusted, meaning it does not incorporate a reported covariate-adjustment model.

The risk difference is useful because it expresses the contrast in absolute terms on the probability scale. Other measures, such as a risk ratio or odds ratio, express the same two-arm comparison on relative scales. The registry includes recalculated risk-ratio and odds-ratio estimates as well, but those measures do not answer a different clinical question and can appear numerically quite different because their scales differ. The registry record does not report a sponsor statistical model for the comparisons.

What does a p-value describe?

A p-value assesses how compatible the observed data are with a specified null model, commonly a model of no difference between groups. It is not the probability that the null model is true, nor does it measure the size or clinical importance of a difference. The posted p-values are unadjusted. Because several dose groups are compared with placebo, interpreting each p-value without accounting for the family of comparisons can overstate the strength of evidence. A planned approach to multiplicity is not reported in the registry record.

The p-values should be read alongside the estimates and confidence intervals. A small p-value can occur with an estimate that is still imprecise, while a non-small p-value does not establish that the groups are equivalent. The confidence interval conveys a range of effect estimates compatible with the data under the method used; it is not a range containing a known fixed treatment effect with a probability attached after the study.

What does randomization contribute?

Randomization is intended to make group assignment independent of participants’ baseline characteristics on average. That helps support a causal comparison when the trial is properly conducted and analyzed. In a trial of this size, however, randomization does not guarantee that every baseline feature is balanced, and the registry material does not give baseline characteristics or details of the assignment procedure. The parallel-group design means each participant was assigned to one group rather than being compared with their own result under another assigned treatment.

Triple masking indicates that the registry categorized the study as having three masked parties, but the record does not identify them or describe the methods used to maintain masking. Masking can reduce the risk that knowledge of assignment affects participant reporting, clinical assessment, or trial conduct. It does not remove uncertainty due to the number of participants, missing data, or multiple comparisons.

How was the primary population defined?

The primary outcome population is called the full analysis set in the registry. It includes participants who withdrew as treatment failures or completed at least one week of dosing and had at least one valid Mayo score during the active double-blind phase. Participants who withdrew as treatment failures were counted as non-responders. Missing data for reasons other than treatment failure were excluded. This rule means that some types of missing outcome information were handled differently from treatment-failure withdrawal. Further details about other exclusions and the timing of assessments are not reported in the registry record.

The posted primary percentages use outcome-specific arm counts of 40 for placebo, 27 for 0.5 mg, 31 for 3 mg, 30 for 10 mg, and 45 for 15 mg. Those counts should be understood in relation to the stated analysis population, not assumed to represent every randomized participant. The record does not provide a full participant flow that would explain the disposition of all enrolled participants.

What can be inferred from the posted analyses?

The record contains outcome counts and unadjusted comparisons, but detailed sponsor analyses are not reported in the registry record. Therefore, the displayed comparisons are best understood as simple unadjusted recalculations from posted summary data, not as a substitute for an unavailable prespecified model. They do not account for baseline covariates, center effects, or multiplicity unless those features are explicitly stated; they are not stated for these comparisons.

General statistical caution applies to every estimate: sampling variation can produce differences even when underlying group outcomes are similar, and modest sample sizes can yield wide intervals. The combination of randomized allocation, a defined response rule, and posted arm counts makes the comparison interpretable, while the missing analysis details limit how far it can be taken as a definitive test of a planned hypothesis.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data constrain an estimated effect. The confidence intervals for the primary risk differences are -0.410 to 0.053 for 0.5 mg, -0.193 to 0.275 for 3 mg, -0.073 to 0.39 for 10 mg, and 0.131 to 0.519 for 15 mg, each compared with placebo. These intervals are wide enough to show substantial uncertainty, especially for comparisons whose ranges extend across zero.

For 0.5 mg, the interval includes both negative and positive differences. For 3 mg and 10 mg, it likewise includes zero, so the posted data do not sharply distinguish those contrasts from no difference. For 15 mg, the interval lies above zero, but its width still allows a range of possible positive differences. The estimate is not known exactly simply because its interval does not include zero.

The intervals are recalculated unadjusted comparisons, and the registry does not report sponsor confidence intervals or a sponsor statistical model for these results. An unadjusted interval addresses uncertainty for its individual contrast; it does not account for the fact that multiple doses are compared with placebo. If an inference is intended to cover the collection of dose comparisons, a multiplicity approach may affect the uncertainty assessment, but the registry record does not report one.

Confidence intervals for the individual arm percentages are not reported in the registry record. The arm percentages are therefore useful descriptions of observed proportions, while the risk-difference intervals convey uncertainty for the pairwise contrasts. General interpretation: greater precision would usually be reflected by a narrower interval, whereas a wider interval leaves more plausible values for the underlying contrast. The relatively small arm counts make it especially important not to treat a single point estimate as exact.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect, not the trial’s original design assumptions. For these binary outcomes, the listed sample sizes are participants per arm, not a total across both arms and not total events. The planning calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the stated fraction of the observed effect as the effect to detect. The trial’s own original sample-size justification and assumptions are not reported in the registry record.

The column for 80% power describes the participants per arm estimated to give an 80% chance of detecting the specified effect under those planning assumptions, if that effect is the true one and the assumptions hold. The 90% power column uses a higher target chance of detection and generally calls for a larger planned sample. These are design-planning quantities, not guarantees about what a future study would observe.

Each planning row represents a different target effect size: the full observed effect, 75% of the observed effect, or 50% of the observed effect. A smaller target effect is harder to distinguish from random variation, so it requires more participants per arm. When the observed difference is small, the resulting planning figures can be very large. That reflects the difficulty of detecting a small contrast under the stated calculation, not evidence that the trial actually had that many participants or that its observed findings have a particular probability of being true.

These figures are not observed power. They do not tell whether the completed trial was adequately powered according to its original plan, because that plan is not reported. They also do not replace interpretation of the confidence intervals, which describe the precision of the posted estimates.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.475; treatment proportion p₁ = 0.296; average p̄ = 0.386.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6884 = 1.3492.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6767 = 0.5695.
4. n = (1.3492 + 0.5695)² / (-0.1787)² = 115.3, rounded up to 116 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 <- 0.475                          # control proportion
p1 <- 0.296296296296296                          # 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(787202)
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 = 116 (80% power), 154 (90% power)
 75% of observed effect: n per arm = 210 (80% power), 281 (90% power)
 50% of observed effect: n per arm = 481 (80% power), 644 (90% power)
power.prop.test: 115.3 
Simulated power at n = 116 per arm: 0.811 (20000 replicates)

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

14. Multiplicity, Missing Data and Analysis Populations

Four dose groups are compared with placebo for the primary endpoint. That creates a family of related comparisons, rather than a single isolated test. When several hypotheses are tested, the chance of at least one apparently unusual result can increase if each is assessed separately at the same threshold. General statistical methods address this issue by defining a testing hierarchy or adjusting the error rate across comparisons. The registry record does not report whether a multiplicity method was planned or used, nor does it report a prespecified ordering of the dose comparisons.

The posted p-values are unadjusted recalculations. They should not be treated as if they establish a family-wise conclusion across every dose. The same caution applies to the secondary results, where multiple outcome definitions are listed and some have several dose comparisons. A full statistical analysis plan and sponsor analyses are not reported in the registry record, so the intended inferential status of each outcome cannot be established from the posted information alone.

The primary full analysis set has a stated missing-data rule. Participants who withdrew as treatment failures were counted as non-responders, while missing data for reasons other than treatment failure were excluded. This is a specific approach to missingness rather than a universal rule that all missing outcomes count as failures. The registry record does not report how many participants had each type of missing outcome, why they were missing, or whether sensitivity analyses examined alternative assumptions.

Outcome denominators vary across measures and arm groups. The registry gives the counts used for each posted result, but does not supply a complete participant flow that explains those variations. Different denominator sizes can affect both observed percentages and precision. The primary analysis population and its rules should therefore be kept in mind when comparing the result with secondary outcomes or with the enrollment target.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: colitis ulcerative (6.2%), general physical health deterioration (3.2%), drug hypersensitivity (3.0%), anal abscess (3.0%), postoperative abscess (3.0%).

The registry reports serious adverse event counts by arm, along with the number at risk for that summary. Serious adverse events affected 4 of 48 participants in the placebo group, 1 of 31 in the 0.5 mg group, 1 of 33 in the 3 mg group, 2 of 33 in the 10 mg group, and 2 of 49 in the 15 mg group. These are counts within the registry’s reported safety groups; they should not be confused with the smaller outcome-specific denominators used for clinical response.

The record also lists serious-event terms and reported percentages, including Colitis ulcerative at 6.2%, General physical health deterioration at 3.2%, Drug hypersensitivity at 3.0%, Anal abscess at 3.0%, Postoperative abscess at 3.0%, and Ischaemic stroke at 2.1%. The registry does not give enough detail in the posted summary to assign those listed terms to a particular treatment arm or to establish whether a person could contribute more than one event term.

Deaths are not reported in the registry record. Safety counts are descriptive and do not by themselves establish comparative risk, especially when event numbers are small. The record’s serious adverse event summary should be read separately from efficacy outcomes.

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

The posted primary result describes the proportion of participants meeting a specific clinical-response definition at Week 8 in each randomized arm. It shows a higher observed response percentage in the 15 mg group than in placebo, while the lower dose comparisons have estimates with intervals that include zero. These findings summarize the registry results and the uncertainty attached to the unadjusted comparisons.

The result does not establish that every participant in a dose group experienced improvement, nor does it indicate how an individual participant will respond. A group percentage is an aggregate measure. The response rule uses a decrease in Mayo score and a rectal-bleeding criterion; it is not equivalent to clinical remission. Remission and endoscopic outcomes are registered separately and have their own posted results.

The results also do not establish a definitive dose-response relationship. The point estimates vary across doses, but the registry record does not report a formal dose-trend analysis or a full sponsor analysis plan. Nor does the posted unadjusted evidence resolve the issue of multiple comparisons. Interpretation should distinguish the observed pattern from a prespecified confirmatory conclusion.

The trial’s randomized, masked, parallel-group design supports a comparative reading of the arms, but available registry information has limits. The original sample-size justification, detailed analysis plan, participant flow, and multiplicity methods are not reported in the registry record. The primary analysis population applies a particular rule to treatment-failure withdrawals and other missing data, and the record does not describe sensitivity analyses for those choices.

Finally, the results do not provide individual medical advice or establish the overall benefit-risk profile. Safety counts are descriptive, and some listed outcomes have no posted results. The most defensible interpretation is therefore bounded: the registry records arm-specific response proportions and unadjusted contrasts at the specified time point, with varying degrees of statistical precision.

17. Why This Trial Matters Statistically

This trial is statistically informative because it compares several active doses with a placebo group under randomized, masked, parallel-group conditions. That structure makes it possible to describe dose-specific response proportions against a common comparator. It also creates a multiple-comparison problem: a set of dose contrasts cannot automatically be interpreted as one isolated test.

The primary endpoint is a defined binary response measure, combining a change in Mayo score with a rectal-bleeding condition. The definition is clinically interpretable as a threshold-based change, while its binary form discards some detail in the underlying score. Participants just meeting the rule and those with larger changes both count as responders. This distinction matters when comparing the binary endpoint with continuous secondary measures that were listed in the registry.

The results illustrate why point estimates, intervals, and analysis plans must be read together. The displayed response percentages rise across some of the dose groups, but the estimates have different levels of precision, and the registry does not report a formal dose-trend test or multiplicity strategy. The strongest-looking individual comparison should not substitute for an account of how the full set of comparisons was planned and interpreted.

The registry also makes clear that sample-size planning illustrations based on observed effects are separate from the study’s original design rationale. Without the original assumptions, those illustrations cannot establish whether the completed study was designed for a specific target effect. More broadly, absent details on missing-data sensitivity analyses and full participant flow constrain the interpretation of the posted results.

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.

Several limitations are specific to the available record. The primary analysis population is defined, but the registry does not report the complete participant flow or explain the differences between the planned enrollment of 195 and the outcome-specific arm counts. The reasons for missing outcomes other than treatment-failure withdrawal are not described, and sensitivity analyses addressing missing data are not reported.

The record does not include the original sample-size justification, detailed statistical analysis plan, or the sponsor’s full comparative analyses. In particular, it does not state whether multiplicity was addressed across the four dose comparisons or across the listed secondary outcomes. The posted comparisons are unadjusted recalculations and may not reproduce the sponsor’s prespecified analysis.

The primary result is reported as a binary response percentage. The definition is detailed, but the registry does not provide the distribution of score changes that underlies the threshold outcome. Arm-level confidence intervals for the percentages are not reported. Secondary results are posted for only some of the listed secondary measures; results for the remaining listed outcomes are not reported in the registry record.

The safety summary provides serious adverse event counts and selected event terms, but the record does not give enough detail to connect every listed term with an arm or to reconstruct all event-level information. Deaths are not reported in the registry record. These limitations do not invalidate the descriptive counts, but they restrict broader inference about comparative safety and the complete pattern of outcomes.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study To Investigate The Safety And Efficacy Of CP- 690,550 In Patients With Moderate And Severe Ulcerative Colitis. — NCT00787202.
• Registry record last updated 2013-04-16; status last verified 2013-03.
• Results first posted 2013-04-16.
• 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

NCT00787202 is a completed randomized, triple-masked, parallel-group phase 2 study sponsored by Pfizer. It enrolled participants with moderate to severe ulcerative colitis and compared placebo with CP-690,550 at 0.5 mg, 3 mg, 10 mg, and 15 mg. The registry records the study start as 2008-12 and primary completion as 2010-09.

The primary outcome was clinical response at Week 8, defined using change in Mayo score and rectal bleeding. The posted response percentages were 47.5% for placebo, 29.6% for 0.5 mg, 51.6% for 3 mg, 63.3% for 10 mg, and 80.0% for 15 mg. The unadjusted recalculated risk-difference interval excludes zero for the 15 mg comparison and includes zero for the other dose comparisons.

The registry also posts selected Week 8 secondary outcomes and serious adverse event counts. Detailed sponsor analyses, the original sample-size justification, and a multiplicity plan are not reported in the registry record.