A Study Of Oral CP-690,550 As A Maintenance Therapy For Ulcerative Colitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Of Oral CP-690,550 As A Maintenance Therapy For Ulcerative Colitis (NCT01458574), a randomized phase 3 study. The primary outcome is Percentage of Participants In Remission at Week 52.

2. Clinical Question

The clinical question is therefore about maintenance outcomes over the trial period, not about initial induction of remission. The primary measure is a binary status at a specified visit: participants either met the remission definition at Week 52 or did not. The results describe the proportions meeting that definition in each arm and the reported between-arm differences. They do not by themselves describe how symptoms changed for every participant or how long remission lasted before the assessment.

The study summary also states that the trial proposed to assess mucosal healing, quality of life, safety, tolerability, and pharmacokinetic exposure. The posted secondary outcomes include mucosal healing and sustained outcomes. The registry record does not report the details needed to evaluate all aspects of quality of life or pharmacokinetic exposure in the results.

3. Trial Design

This was a Phase 3, randomized, double-blind, placebo-controlled, parallel-group study. Its official title describes a multicentre study of oral CP-690,550 as maintenance therapy in participants with ulcerative colitis. The registry record lists Pfizer as the sponsor and the study status as completed. Enrollment was 593.

The reported outcome arms are Tofacitinib 5 mg BID, Tofacitinib 10 mg BID, and placebo. The study title uses the compound label CP-690,550, while the result-arm labels use tofacitinib. The registry record does not provide further details about the relationship between those labels or the exact treatment regimen beyond the arm names. It also does not report the randomization ratio, concealment procedures, or how masking was maintained.

For the primary endpoint, the FAS included all randomized participants. The reported arm sizes for that endpoint were 198 for Tofacitinib 5 mg BID, 197 for Tofacitinib 10 mg BID, and 198 for placebo. These are the analysis denominators reported for the endpoint, not a description of how participants were handled after randomization or of the number with complete follow-up.

The primary endpoint was the percentage of participants in remission at Week 52. Secondary measures include remission at Week 24, mucosal healing at Week 24 and Week 52, and sustained outcomes. One sustained steroid-free remission outcome was limited to participants with remission at baseline. The study summary describes assessing participants over the age of 18 years, but full inclusion and exclusion criteria are not reported in the registry record.

The record lists start and primary-completion dates, but the enrollment duration, follow-up schedule beyond the stated outcome timeframes, and reasons for any exclusions or discontinuations are not described in the information available for these results. Details of the trial's original statistical design assumptions are also 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: FAS included all randomized participants.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Full analysis set (FAS) included all randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants In Remission at Week 52, measured over Week 52.

The primary endpoint was the percentage of participants in remission at Week 52. Remission was defined using the Mayo score: a total score of 2 points or lower, no individual subscore exceeding 1 point, and a rectal bleeding subscore of 0. The Mayo score measures ulcerative colitis disease activity using stool frequency, rectal bleeding, centrally read flexible sigmoidoscopy findings, and physician global assessment. Each subscore ranges from 0 to 3, and the subscores are summed to produce a total score ranging from 0 to 12; higher scores indicate greater disease severity.

This endpoint converts a clinical assessment at a specified follow-up visit into a binary outcome. That makes the result interpretable as a proportion meeting a defined threshold, rather than as an average change in a continuous score. The explicit threshold also clarifies what the registry calls remission, which matters because different definitions can yield different event rates.

A Week 52 assessment is relevant to a maintenance question because it records whether participants met the remission criteria at that follow-up point. It does not establish that remission was uninterrupted throughout the period, nor does it measure every dimension of disease experience. Sustained remission and mucosal healing are addressed separately among the secondary outcomes. The registry record does not report additional details about how missing Week 52 assessments were treated in the primary analysis.

7. Results Posted in the Registry

• Tofacitinib 5 mg BID: 34.3% (68/198) versus 11.1% (22/198) on control; risk difference 23.2 percentage points (95% CI 15.3 to 31.2), risk ratio 3.09 (1.99–4.79); unadjusted two-proportion z-test p = <0.001.
• Tofacitinib 10 mg BID: 40.6% (80/197) versus 11.1% (22/198) on control; risk difference 29.5 percentage points (95% CI 21.4 to 37.6), risk ratio 3.65 (2.38–5.61); unadjusted two-proportion z-test p = <0.001.

8. Secondary Efficacy Endpoints

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

Percentage of Participants With Mucosal Healing at Week 52

• 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:
  • Tofacitinib 5 mg BID: 37.4% (74/198) versus 13.1% (26/198) on control; risk difference 24.2 percentage points (95% CI 16.0 to 32.5), risk ratio 2.85 (1.91–4.25); unadjusted two-proportion z-test p = <0.001.
  • Tofacitinib 10 mg BID: 45.7% (90/197) versus 13.1% (26/198) on control; risk difference 32.6 percentage points (95% CI 24.2 to 41.0), risk ratio 3.48 (2.36–5.13); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tofacitinib 5 mg BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tofacitinib 10 mg BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With Sustained Steroid-Free Remission (Defined as Being in Remission and Steroid-Free at Both Week 24 and 52), Among Participants With Remission at Baseline

• 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:
  • Tofacitinib 5 mg BID: 35.4% (23/65) versus 5.1% (3/59) on control; risk difference 30.3 percentage points (95% CI 17.4 to 43.2), risk ratio 6.96 (2.20–21.99); unadjusted two-proportion z-test p = <0.001.
  • Tofacitinib 10 mg BID: 47.3% (26/55) versus 5.1% (3/59) on control; risk difference 42.2 percentage points (95% CI 27.9 to 56.5), risk ratio 9.30 (2.98–28.99); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tofacitinib 5 mg BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tofacitinib 10 mg BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants in Remission at Week 24

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Tofacitinib 5 mg BID: 33.8% (67/198) versus 11.1% (22/198) on control; risk difference 22.7 percentage points (95% CI 14.8 to 30.6), risk ratio 3.05 (1.96–4.73); unadjusted two-proportion z-test p = <0.001.
  • Tofacitinib 10 mg BID: 35.5% (70/197) versus 11.1% (22/198) on control; risk difference 24.4 percentage points (95% CI 16.4 to 32.4), risk ratio 3.20 (2.07–4.95); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tofacitinib 5 mg BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tofacitinib 10 mg BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The reported secondary results follow a broadly similar direction to the primary finding in the outcomes with numerical results: the active-treatment arms have higher reported percentages than placebo for mucosal healing and the listed sustained measures. Week 24 remission also shows a higher reported proportion in each active arm than in placebo. This directional pattern is consistent with, but does not independently prove or expand, the primary Week 52 remission result.

The secondary measures capture related but distinct aspects of outcome. Mucosal healing is not the same endpoint as symptom-based remission, and a sustained outcome asks about status across more than one assessment. The steroid-free sustained remission result is specifically evaluated among participants with remission at baseline, so its analysis population differs from the full randomized population used for the primary endpoint. These differences affect how the endpoints should be compared.

The registry record lists 25 posted secondary outcomes and 70 analyses overall. Multiple outcomes and comparisons increase the chance that some apparently notable findings will arise by chance, particularly when results are interpreted individually. The record does not report a multiplicity strategy, such as a hierarchy or adjustment procedure, for the secondary analyses. Their p-values and confidence intervals should therefore be treated as less definitive than the prespecified primary comparison, whose status is identified in the record.

Secondary results can help assess whether the observed pattern is coherent across related measures, but they do not replace the primary endpoint or establish every outcome proposed in the study summary. Nor should consistency across endpoints be treated as independent confirmation when outcomes may be correlated or when several comparisons are made. Detailed interpretation is limited by the absence of a reported full analysis plan and by unreported handling of missing data for these measures.

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

• Percentage of Participants With Mucosal Healing at Week 52
• Percentage of Participants With Sustained Steroid-Free Remission (Defined as Being in Remission and Steroid-Free at Both Week 24 and 52), Among Participants With Remission at Baseline
• Percentage of Participants in Remission at Week 24
• Percentage of Participants in Sustained Remission
• Percentage of Participants With Mucosal Healing at Week 24
• Percentage of Participants With Sustained Mucosal Healing
• Percentage of Participants With Mucosal Healing at Week 24 and 52, Among Participants With Mucosal Healing at Baseline
• Percentage of Participants With Sustained Mucosal Healing, Among Participants With Mucosal Healing at Baseline

9. Statistical Interpretation

10. Statistical Methods Explained

What kind of endpoint is being compared?

The primary outcome is binary: each participant is classified according to whether the remission definition was met at Week 52. For a binary outcome, an arm's percentage is the proportion of its analyzed participants who had the event—in this case, remission. It is a group summary, not a prediction for a particular participant. The analysis denominators and observed percentages appear in the posted results.

A risk difference expresses the contrast between two event proportions by subtracting one proportion from the other. Here the sponsor reports the contrast in percentage points, with each Tofacitinib arm compared with placebo. A positive difference means the event percentage was higher in the named active-treatment arm than in placebo. The risk difference is an absolute measure: it retains the scale of percentage points and does not express the contrast as a relative multiple.

What does the reported test assess?

The sponsor reports a Cochran-Mantel-Haenszel chi-square test for each primary comparison. In general, this is a method for comparing categorical outcomes while combining information across strata. The registry record does not report the stratification variables, stratum-specific results, or enough analysis detail to determine precisely how the method was implemented. The reported p-values are less than < 0.001 for both comparisons.

A p-value summarizes how inconsistent the observed data are with a specified null hypothesis, under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. In this study, the sponsor's test concerns the difference in remission percentage between each active arm and placebo. The p-value should be read alongside the estimated difference and its confidence interval.

How should confidence intervals be read?

A confidence interval gives a range of estimates that reflects sampling uncertainty under the method used. The sponsor reports 95% confidence intervals for the risk differences. A narrower interval indicates greater precision on the measured scale, while a wider interval indicates more uncertainty about the size of the difference. The intervals do not account automatically for every potential source of uncertainty, such as bias from missing data or limitations in the outcome definition.

The sponsor's primary estimates are 23.2 and 29.5 percentage points. The corresponding 95% intervals are 15.3 to 31.2 and 21.4 to 37.6 percentage points. Since each reported interval excludes zero, the interval-based reading is consistent with the sponsor's reported small p-values. The ranges remain important: the point estimate alone would give an incomplete account of precision.

What do relative measures add?

Relative risk compares the event proportion in one arm with the event proportion in another. An odds ratio compares the odds of the event rather than the proportions themselves. These measures can describe the same binary comparison on different scales. The registry record includes recalculated relative measures, but the sponsor's reported primary analysis is the difference in percentage. Absolute differences are often easier to relate to event frequencies, while relative measures can appear larger or smaller depending on the baseline event proportion. None of these scales should be treated as a direct statement about an individual participant.

The recalculated comparisons are simple unadjusted calculations from the posted summary statistics. The sponsor reports CMH-based analyses, which are the analyses reported by the trial record. They need not be numerically identical because the methods and inputs can differ. The registry record does not indicate a discrepancy between the posted summary and the sponsor's reported results.

Why does randomization matter?

Random allocation is intended to make treatment groups comparable at the point of assignment, on average, including with respect to factors that were not measured. Double masking is intended to reduce the influence of treatment knowledge on participant behavior, care, and outcome assessment. Those design features strengthen the interpretation of between-arm differences, but they do not eliminate all possible bias. The registry record does not provide enough detail to assess allocation concealment, masking success, or the balance of individual baseline characteristics.

The primary population is the FAS, which the record defines as all randomized participants. That definition indicates the intended analysis population, but it does not explain how missing outcomes, protocol deviations, or treatment discontinuation were handled. Those details matter for a Week 52 endpoint, because participants without an observed assessment may require an analysis rule. The missing-data method is not reported in the registry record.

What is not established by these statistics?

The reported analyses establish what the observed group summaries and specified tests show; they do not establish the original design assumptions, a causal mechanism, or the appropriateness of every modeling choice. The trial's original sample-size justification, planned interim analyses, and complete statistical analysis plan are not reported in the registry record. No adjustment for multiple comparisons is described. These omissions limit assessment of how the analysis decisions relate to the full set of measured outcomes and tests.

11. Confidence Intervals and Precision

The sponsor's primary comparisons are reported as differences in percentage. For Tofacitinib 5 mg BID versus placebo, the estimated difference is 23.2 percentage points, with a 95% confidence interval from 15.3 to 31.2 percentage points. For Tofacitinib 10 mg BID versus placebo, the estimate is 29.5 percentage points, with a 95% confidence interval from 21.4 to 37.6 percentage points. These ranges communicate uncertainty around the estimated differences, not uncertainty about the recorded percentages themselves.

In general, a 95% confidence interval is generated by a method that, across repeated applications under its assumptions, would cover the target quantity at the stated rate. It does not mean there is a 95% probability that this particular interval contains the true effect. Nor does it guarantee that the method's assumptions are correct or account for unmeasured bias. It is best read as a range of effect sizes reasonably compatible with the observed data and analysis method.

Both intervals exclude zero, which accords with the sponsor's reported p-values of less than < 0.001. The interval for the 10 mg BID comparison has endpoints that are higher than those for the 5 mg BID comparison, but the registry record does not report a direct comparison of the active arms. Comparing separate intervals is not a formal test of the difference between active treatments.

Confidence intervals for the recalculated comparisons are based on unadjusted methods applied to posted summary statistics, whereas the sponsor reports CMH analyses. The sponsor's reported analysis is the primary reference for interpreting the trial's own inferential results. Neither interval captures all limitations of the record, including unreported handling of missing assessments, the absence of reported multiplicity procedures, and incomplete information about analysis implementation.

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 participants per arm for 80% power and participants per arm for 90% power. They are illustrations based on the observed effect, not the trial's original sample-size design. The registry record does not report the trial's own original design assumptions or sample-size justification.

The planning calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers effects equal to the stated fraction of the observed effect: 100%, 75%, or 50%. The figures are for the two-arm comparison specified for each outcome. They are not total enrollment targets for a multi-arm study.

In plain terms, the 80% power column gives the estimated number of participants per arm for a design with an 80% chance of detecting the assumed effect under the planning assumptions, if that effect is present. The 90% power column uses a higher target chance of detection and therefore generally calls for more participants per arm. These are prospective planning quantities, not a retrospective measure of how likely the completed trial was to find its result.

Smaller effects are harder to distinguish from random variation, so the planning figures rise as the assumed effect is reduced from the observed effect to 75% or 50% of it. This is not observed power. The calculations are illustrations whose results depend on their assumptions, including the event rates and effect size used; they do not establish whether the trial was adequately planned or whether another study would reproduce the observed effect.

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 = 0.111; treatment proportion p₁ = 0.343; average p̄ = 0.227.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5927 = 1.1616.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5694 = 0.4792.
4. n = (1.1616 + 0.4792)² / (0.2323)² = 49.9, rounded up to 50 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 <- 0.111111111111111                          # control proportion
p1 <- 0.343434343434343                          # 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(1458574)
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 = 50 (80% power), 67 (90% power)
 75% of observed effect: n per arm = 81 (80% power), 108 (90% power)
 50% of observed effect: n per arm = 163 (80% power), 217 (90% power)
power.prop.test: 49.9 
Simulated power at n = 50 per arm: 0.818 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The sponsor's primary results use a Cochran-Mantel-Haenszel chi-square test and report differences in percentage. The reported estimates are 23.2 percentage points for Tofacitinib 5 mg BID versus placebo and 29.5 percentage points for Tofacitinib 10 mg BID versus placebo. Both comparisons have reported 95% confidence intervals and p-values of less than < 0.001.

The unadjusted recalculations from posted arm summaries produce estimates close to the sponsor's reported differences, with small differences in the displayed rounded values. The sponsor's own analysis takes precedence for describing the trial's inferential result because it is the analysis reported by the study record. The recalculated figures are useful as a transparent check based on the summary data, but they do not reproduce all features of the sponsor's method.

The registry record does not flag a discrepancy between the reported and recalculated results. The calculation methods differ: an unadjusted comparison of summary proportions is not identical to a CMH analysis. The record does not report enough detail about stratification or other analysis choices to fully reconcile any minor rounding or method-related differences. Accordingly, the sponsor's estimates and intervals are the appropriate primary figures to quote, while recalculations should be understood as supplementary unadjusted summaries.

15. Multiplicity, Missing Data and Analysis Populations

The registry record identifies a primary endpoint and lists multiple secondary outcomes. It reports 26 outcome measures and 70 analyses overall, with 25 secondary outcomes posted. When many endpoints or treatment comparisons are tested, the chance of at least one false-positive result can rise if each test is interpreted without accounting for the wider set of tests. This is the general multiplicity problem.

The record does not report whether the sponsor used a multiplicity adjustment, a prespecified testing hierarchy, or another procedure to control the interpretation of multiple comparisons. It also does not report which secondary analyses, if any, were considered confirmatory. As a result, the primary comparisons can be distinguished from the secondary findings by their stated endpoint status, but the strength of evidence across the full family of tests cannot be assessed from the registry record alone.

The FAS is defined as all randomized participants, which gives a clear stated population for the primary analysis. However, the record does not explain how participants with missing Week 52 outcomes were handled, whether imputation was used, or how treatment discontinuation and protocol deviations affected inclusion. It does not report the amount or pattern of missing outcome data. Those details are important because different missing-data approaches can lead to different estimated proportions and treatment contrasts.

The record also does not report interim analyses, stopping rules, the trial's original statistical plan, or the sample-size justification. These omissions do not negate the reported results, but they limit independent assessment of whether the analysis followed a prespecified plan and how the multiple outcomes were managed. The reported p-values and intervals describe the stated analyses; they do not resolve uncertainties about multiplicity or missing-data assumptions.

16. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: colitis ulcerative (4.0%), invasive ductal breast carcinoma (1.2%), abortion spontaneous (1.1%), abdominal pain (0.5%), diarrhoea (0.5%).

The registry record reports serious adverse events affecting 10 of 198 participants in the Tofacitinib 5 mg BID group, 11 of 196 in the Tofacitinib 10 mg BID group, and 13 of 198 in the placebo group. These are counts and denominators for participants at risk and affected, not a statistical comparison. The record also reports other adverse events affecting 87 of 198, 107 of 196, and 108 of 198 participants in those respective groups.

Among the listed serious adverse-event terms, the record gives colitis ulcerative as 4.0%, invasive ductal breast carcinoma as 1.2%, abortion spontaneous as 1.1%, abdominal pain as 0.5%, diarrhoea as 0.5%, and cholecystitis acute as 0.5%. These terms and percentages should be read as registry listings; the record does not provide the corresponding event counts, analysis details, or a basis for interpreting them as comparative risks.

The deaths field is not reported in the registry record. No inference about comparative safety should be made from raw counts alone: the groups have different reported denominators for the serious adverse-event measure, and the record does not provide adjusted comparisons or detailed follow-up information for safety outcomes.

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

The primary result describes a higher observed percentage meeting the Week 52 remission definition in each reported Tofacitinib arm than in the placebo arm. The sponsor-reported differences and confidence intervals indicate that the estimated between-arm contrasts are positive and are measured with finite uncertainty. The p-values summarize the results of the sponsor's specified comparisons under the test assumptions.

That statistical pattern is not a guarantee that every participant will experience remission, and it is not a forecast for an individual. Group percentages describe outcomes in the analyzed trial populations. The endpoint is a threshold-based assessment at Week 52; it does not establish continuous remission throughout the maintenance period, symptom improvement for every participant, or an effect on every aspect of ulcerative colitis.

The trial's randomized, double-blind design supports comparisons between assigned groups, but the registry record does not report all information needed to assess risk of bias, missing outcomes, or analysis decisions. The results also do not establish a direct difference between the two active arms, because a direct active-arm comparison is not reported. The fact that one observed percentage is higher than another is not, by itself, evidence that the underlying treatments differ.

Secondary results address related outcomes and appear directionally consistent for the reported comparisons, but multiple testing and the lack of a reported multiplicity plan mean they should be interpreted cautiously. Safety counts are descriptive in the available record and should not be read as definitive evidence of comparative safety. The record does not report the details needed to assess all outcomes proposed in the study summary.

Overall, the posted data support a statistical description of the Week 52 endpoint and its uncertainty in this completed study. They do not, without additional design, analysis, and clinical context, settle questions about individual outcomes, durability beyond the reported assessment times, or the balance of benefits and harms.

18. Why This Trial Matters Statistically

This trial illustrates why a clinical result should be read through its estimand, analysis population, and uncertainty rather than through a p-value alone. The primary outcome has a defined clinical threshold, and the sponsor reports absolute differences in percentage with confidence intervals. That combination conveys both the direction and estimated scale of the group contrast, together with its sampling precision.

The record also shows how an analysis method matters. The sponsor used a CMH chi-square test, while simple recalculations from the posted summaries are unadjusted. Similar estimates can provide a useful check, but agreement in the broad pattern does not make the methods interchangeable. The sponsor's reported analysis remains the reference for the trial's own inferential claim.

Several design and reporting features shape interpretation: randomized allocation, double masking, an FAS defined as all randomized participants, multiple secondary outcomes, and limited reporting about missing data and multiplicity. Each feature affects a different part of the evidence. A strong primary comparison does not remove the need to understand how incomplete outcomes were handled, whether multiple testing was addressed, or how secondary findings were selected and interpreted.

The results further demonstrate that precision and statistical significance are not the same as certainty about clinical meaning. The confidence intervals narrow the range of compatible differences under the analysis, while the endpoint definition and trial record constrain what can be concluded. For biostatisticians and clinical researchers, this is a useful example of separating the numerical result from the assumptions and reporting details that govern its interpretation.

19. Limitations

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

The registry record does not report the full eligibility criteria, baseline characteristics, allocation ratio, allocation concealment procedures, masking assessment, or details of treatment adherence and discontinuation. These omissions limit assessment of how well the randomized groups remained comparable and how the analyzed population relates to all enrolled participants.

Although the primary population is defined as all randomized participants, the missing-data method for the Week 52 endpoint is not reported. The amount and pattern of missing outcome data, imputation procedures, and sensitivity analyses are also not reported in the registry record. Without those details, it is not possible to assess how incomplete follow-up may have affected the estimated remission percentages.

The sponsor's reported primary method is a CMH chi-square test, but the registry record does not describe the strata or provide a full account of model implementation. The original sample-size justification and statistical analysis plan are not reported. Neither planned interim analyses nor a multiplicity strategy for the many listed outcomes and analyses is described.

The primary endpoint is a binary assessment at Week 52, so it does not capture the timing or duration of remission between visits. Secondary endpoints address some related questions, but the record does not provide a full account of all proposed outcomes, including detailed quality-of-life and pharmacokinetic results. The safety information is descriptive, with deaths not reported in the registry record and no comparative statistical analysis specified.

These limitations do not change the recorded arm percentages or sponsor estimates. They constrain the broader interpretation of those statistics and the extent to which results can be independently evaluated from the registry record alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study Of Oral CP-690,550 As A Maintenance Therapy For Ulcerative Colitis — NCT01458574.
• Registry record last updated 2017-05-18; status last verified 2017-04.
• Results first posted 2017-05-18.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

This completed Phase 3 study evaluated CP-690,550 as maintenance therapy for ulcerative colitis in a randomized, double-blind, placebo-controlled, parallel-group design. Enrollment was 593, and the primary endpoint was the percentage of participants in remission at Week 52, using a Mayo-score definition. The primary analysis population was the FAS, defined as all randomized participants.

The posted results report Week 52 remission percentages of 34.3% for Tofacitinib 5 mg BID, 40.6% for Tofacitinib 10 mg BID, and 11.1% for placebo. The sponsor-reported differences versus placebo were 23.2 and 29.5 percentage points, respectively, with 95% confidence intervals excluding zero and p < 0.001 for both comparisons. Secondary results with numerical reports generally follow the same direction, while multiplicity and missing-data details are not reported.

The registry record provides descriptive serious adverse-event counts but does not report deaths. The results support interpretation of the specified endpoint comparisons, while important details of the original statistical plan, missing-data handling, and multiplicity procedures remain unreported.