A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction and Maintenance Therapy in Participants With Moderately to Severely Active Ulcerative Colitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction and Maintenance Therapy in Participants With Moderately to Severely Active Ulcerative Colitis (NCT02407236), a randomized phase 3 study. The primary outcome is Induction Study - Number of Participants With Clinical Remission at Week 8 (As Per Global Definition).

2. Clinical Question

3. Trial Design

The registry describes a Phase 3, randomized, double-blind, placebo-controlled, parallel-group, multicenter study of ustekinumab induction and maintenance therapy in participants with moderately to severely active ulcerative colitis. Total enrollment was 961. The primary result discussed is from the induction study, where participants were randomized to a single Week 0 IV infusion of placebo, ustekinumab 130 mg, or ustekinumab approximating 6 mg/kg.

Clinical response at Week 8 determined eligibility for the maintenance study. Participants responding to induction ustekinumab were randomized to maintenance placebo SC, ustekinumab 90 mg SC every 12 weeks, or ustekinumab 90 mg SC every 8 weeks. The registry also describes pathways for nonresponders and delayed responders, including additional ustekinumab treatment and later maintenance participation. These later groups are selected through their prior response status and are not interchangeable with the original induction comparison.

The primary efficacy analysis set consisted of all participants randomized in the induction study. Its reported endpoint was the number with clinical remission at Week 8 under the global definition. The registry reports 319 participants in the placebo induction group, 320 in the 130 mg IV group, and 322 in the approximately 6 mg/kg IV group for this endpoint.

For the primary endpoint, the sponsor’s reported analyses used the Cochran-Mantel-Haenszel method and reported adjusted treatment differences with 95% confidence intervals. The registry record does not report the adjustment or stratification variables in the material describing those analyses. Original sample-size justification, planned interim analyses, and other details of the original statistical plan 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.

Only the first 6 of 13 groups are shown.

Induction Study (8 Weeks)

Reasons for not completing this period:

Maintenance Study (44 Weeks)

Reasons for not completing this period:

Long-term Extension Period (176 Weeks)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The primary efficacy analysis set (PEAS) consisted of all participants randomized in the induction study.

5. Baseline Characteristics

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

Population: The primary efficacy analysis set consisted of all participants randomized in the study.

Only the first 6 of 14 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Induction Study - Number of Participants With Clinical Remission at Week 8 (As Per Global Definition), measured over Week 8.

The primary endpoint counts participants who met the global definition of clinical remission at Week 8 of induction. Under the registry definition, remission requires a Mayo score less than or equal to 2 points, with no individual subscore greater than 1. The Mayo score combines stool frequency, rectal bleeding, endoscopy findings, and the physician’s global assessment. Each subscore is rated from 0, normal, to 3, severe; the total ranges from 0 to 12, with higher scores indicating worsening disease.

This endpoint turns several clinical and endoscopic measurements into a binary outcome: each participant either meets the stated remission definition or does not. That makes the result interpretable as the number meeting a defined status in each randomized group, while the between-group comparison estimates how the observed proportions differ. The use of an explicit threshold also makes clear what qualifies as remission in this analysis; it does not describe every possible aspect of disease status or change.

The endpoint uses the endoscopy subscore from central review of endoscopy video. Participants with a prohibited change in concomitant ulcerative colitis medication, an ostomy or colectomy before Week 8, or all four Mayo subscores missing at Week 8 were considered not to be in remission. This rule matters because the reported count reflects both the clinical definition and the registry’s stated handling of these situations.

7. Results Posted in the Registry

• IS: Ustekinumab 130 Milligram (mg) IV: 15.6% (50/320) versus 5.3% (17/319) on control; risk difference 10.3 percentage points (95% CI 5.6 to 15.0), risk ratio 2.93 (1.73–4.97); unadjusted two-proportion z-test p = <0.001.
• IS: Ustekinumab Approximately 6 mg/kg IV: 15.5% (50/322) versus 5.3% (17/319) on control; risk difference 10.2 percentage points (95% CI 5.5 to 14.9), risk ratio 2.91 (1.72–4.94); 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.

Induction Study: Number of Participants With Endoscopic Healing at Week 8

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • IS: Ustekinumab 130 Milligram (mg) IV: 26.2% (84/320) versus 13.8% (44/319) on control; risk difference 12.5 percentage points (95% CI 6.3 to 18.6), risk ratio 1.90 (1.37–2.65); unadjusted two-proportion z-test p = <0.001.
  • IS: Ustekinumab Approximately 6 mg/kg IV: 27.0% (87/322) versus 13.8% (44/319) on control; risk difference 13.2 percentage points (95% CI 7.1 to 19.4), risk ratio 1.96 (1.41–2.72); 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 IS: Ustekinumab 130 Milligram (mg) IV 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 IS: Ustekinumab Approximately 6 mg/kg IV than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Induction Study: Number of Participants With Clinical Response at Week 8

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • IS: Ustekinumab 130 Milligram (mg) IV: 51.2% (164/320) versus 31.3% (100/319) on control; risk difference 19.9 percentage points (95% CI 12.4 to 27.4), risk ratio 1.63 (1.35–1.99); unadjusted two-proportion z-test p = <0.001.
  • IS: Ustekinumab Approximately 6 mg/kg IV: 61.8% (199/322) versus 31.3% (100/319) on control; risk difference 30.5 percentage points (95% CI 23.1 to 37.8), risk ratio 1.97 (1.64–2.37); 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 IS: Ustekinumab 130 Milligram (mg) IV 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 IS: Ustekinumab Approximately 6 mg/kg IV than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Induction Study - Change From Baseline in Total Inflammatory Bowel Disease Questionnaire (IBDQ) Score at Week 8

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • IS: Ustekinumab 130 Milligram (mg) IV: mean 33.40 versus 16.10 on control; difference 17.30 (95% CI 12.32 to 22.28), standardized effect d = 0.54; z-test p = <0.001.
  • IS: Ustekinumab Approximately 6 mg/kg IV: mean 35.00 versus 16.10 on control; difference 18.90 (95% CI 13.99 to 23.81), standardized effect d = 0.60; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for IS: Ustekinumab 130 Milligram (mg) IV 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 mean is statistically higher for IS: Ustekinumab Approximately 6 mg/kg IV than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes cover several connected aspects of induction and maintenance: endoscopic healing, clinical response, change in IBDQ score, later clinical response, maintenance endoscopic healing, and definitions of clinical or corticosteroid-free remission. Taken together, they broaden the description beyond the primary Week 8 remission endpoint. Induction outcomes address short-term response and endoscopic status, while maintenance outcomes concern participants who progressed through an induction-response pathway.

The induction secondary findings follow a generally favorable pattern for the ustekinumab groups compared with placebo across the outcomes included in the recalculations. This provides context alongside the primary remission result, but secondary endpoints measure different constructs and do not simply repeat the primary outcome. For example, a response threshold, an endoscopic assessment, and a questionnaire score are not interchangeable definitions of clinical remission.

The maintenance results require particular care in interpretation. Entry and assignment were linked to clinical response during induction, so the maintenance groups represent selected participants rather than the full randomized induction population. A comparison in maintenance therefore addresses a different population and question from the randomized induction comparisons.

The registry lists 71 posted secondary outcome measures, and the summarized recalculations include selected secondary endpoints. Unadjusted secondary p-values, where reported in the recalculations, carry less weight when the overall multiplicity strategy is unknown. With many outcomes, some nominal findings can occur by chance. The registry record does not report a multiplicity adjustment or a confirmatory testing hierarchy for these secondary results. They are best treated as context for the primary result rather than as independent confirmation on their own.

For the IBDQ outcome, the posted measure is mean change from baseline, with standard deviations and recalculated mean differences. This continuous measure describes average score change, while the binary outcomes count participants meeting specified categories. The distinction in measurement scale is one reason the secondary findings should be interpreted outcome by outcome rather than combined into a single claim.

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

• Induction Study: Number of Participants With Endoscopic Healing at Week 8
• Induction Study: Number of Participants With Clinical Response at Week 8
• Induction Study - Change From Baseline in Total Inflammatory Bowel Disease Questionnaire (IBDQ) Score at Week 8
• Maintenance Study: Number of Participants With Clinical Response up to Week 44
• Maintenance Study: Number of Participants With Endoscopic Healing at Week 44
• Maintenance Study: Number of Participants With Clinical Remission and Not Receiving Concomitant Corticosteroids (Corticosteroid-free Clinical Remission) at Week 44 (As Per Global Definition)
• Maintenance Study: Number of Participants With Clinical Remission and Not Receiving Concomitant Corticosteroids (Corticosteroid-free Clinical Remission) at Week 44 (As Per US Definition)
• Maintenance Study: Number of Participants With Clinical Remission up to Week 44 Among Participants Who Achieved Clinical Remission at Maintenance Study Baseline (As Per Global Definition)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: a participant was classified as in remission or not in remission at Week 8. For a binary outcome, a group’s event proportion is the number meeting the endpoint relative to the number assessed in that group. The risk difference compares two such proportions by subtraction. A positive value in the posted comparisons means the observed remission proportion was higher in the ustekinumab group than in the placebo group. The estimate is on an absolute scale, which keeps the comparison tied to the difference in the proportion of participants meeting the endpoint.

The recalculated risk differences and confidence intervals are simple, unadjusted calculations from the posted group summaries. They should be distinguished from the sponsor’s reported adjusted treatment differences, which used the Cochran-Mantel-Haenszel method. That method can combine information across strata to estimate an adjusted association; however, the registry record does not report the stratification variables in this description. The sponsor’s reported analysis is the formal analysis in the registry, while the unadjusted recalculation serves as a summary-statistic check.

A confidence interval describes the uncertainty around an estimated difference under the statistical method’s assumptions. A narrower interval indicates greater precision for that estimate; a wider interval indicates more uncertainty. The reported 95% intervals express a method-based range, not a guarantee that the true value lies inside a particular calculated interval. Their interpretation also depends on how the outcome and analysis population were defined.

A p-value assesses how unusual data at least as incompatible with the null hypothesis would be if the null and the analysis assumptions held. The reported unadjusted p-values are <0.001 for both primary comparisons. Small p-values do not establish the probability that the null is true, and they do not replace consideration of the estimated effect and its confidence interval. The sponsor’s adjusted p-values are also reported as < 0.001.

The registry reports the primary analysis set as all participants randomized in induction. The remission definition includes explicit rules for certain intercurrent events and missing endpoint components: prohibited medication change, ostomy or colectomy before Week 8, and all four Mayo subscores missing are each treated as not in remission. Those rules affect the meaning of the binary outcome. Other missing-data procedures, such as methods for individual missing subscores beyond the stated rule, are not reported in the registry record.

For general context, randomization is intended to make treatment groups comparable on average at the start of a trial, while masking can reduce the influence of knowledge of assigned treatment on care or assessment. Those design features strengthen interpretation of the induction comparison, but they do not eliminate uncertainty, missing information, measurement limitations, or the need to distinguish the original randomized comparison from later response-selected groups.

Relative measures are also available in the recalculations. A risk ratio compares the two event proportions as a ratio; an odds ratio compares the odds of meeting the endpoint. These measures answer different numerical questions from a risk difference, and their values should not be treated as interchangeable. The risk-difference scale is particularly direct for communicating the absolute contrast in remission proportions. The recalculated intervals for the relative measures are statistical estimates from the posted summaries, not a substitute for the sponsor’s adjusted analysis.

Planning illustrations use a different purpose from the trial’s observed analysis. They use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect set to a stated fraction of the observed effect. For binary outcomes, the illustrations express participants per arm. These planning values are not a record of how the trial was originally designed. The original design assumptions and sample-size justification are not reported in the registry record.

General statistical principle: the power of a planned study depends on the effect size being targeted, variability, the significance criterion, and the number of participants. If the effect to detect is smaller, more participants are generally needed to achieve the same target power, all else equal. The trial’s own intended effect and original design assumptions cannot be inferred from the posted results.

Finally, the registry includes multiple endpoints and multiple comparisons. When many hypotheses are evaluated, the chance of at least one apparently strong finding can increase if no adjustment is made. Whether the trial controlled the family-wise error rate or used another multiplicity strategy is not reported in the registry record. Consequently, unadjusted analyses of secondary outcomes should be read as supportive or exploratory evidence unless the planned testing hierarchy and adjustment are known.

11. Confidence Intervals and Precision

The primary unadjusted recalculated risk-difference estimate for ustekinumab 130 mg IV versus placebo is 0.103, with a 95% confidence interval from 0.056 to 0.150. For approximately 6 mg/kg IV versus placebo, the estimate is 0.102, with a 95% confidence interval from 0.055 to 0.149. Both intervals exclude zero, corresponding to the reported unadjusted p-values of <0.001.

The endpoints of an interval express the range of effects compatible with the observed data under the interval method and its assumptions. They are not a range of individual participant outcomes, and they do not indicate that every value inside the interval is equally likely. The intervals quantify sampling uncertainty for the comparisons, but do not cover every source of uncertainty, such as measurement choices, departures from assumptions, or selective reporting.

The sponsor’s adjusted confidence intervals are 5.7 to 14.9 for the 130 mg IV comparison and 5.6 to 14.8 for the approximately 6 mg/kg IV comparison, both at 95% confidence. Their scale is reported as adjusted treatment difference. The numerical presentation differs from the unadjusted risk-difference notation, while the point estimates are closely aligned as described in the sponsor reading discussion.

Because the two comparisons share the placebo group, their estimates are not independent of one another. Overlapping intervals do not provide a formal test of whether the doses differ. A direct dose-to-dose comparison and its uncertainty are not reported in the registry record. Precision should therefore be considered separately for each ustekinumab-versus-placebo contrast, rather than used to claim that the two doses are equivalent or different.

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 sample-size illustrations are planning calculations based on the observed effect, not the trial’s original design calculation. For these binary-outcome comparisons, the stated figures are participants per arm, not total participants across both groups. The calculations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They consider effects set at 100%, 75%, and 50% of the observed effect.

The 80% power column gives the estimated number of participants per arm for a hypothetical study designed to have an 80% chance of detecting the specified effect under the planning assumptions, if that effect and the assumptions were correct. The 90% power column makes the analogous illustration for a 90% chance. Higher target power generally requires more participants when the other assumptions remain fixed.

Planning for 75% of the observed effect means the calculation targets a smaller effect than planning for 100% of the observed effect. Planning for 50% targets a still smaller effect. The listed requirements rise as the target effect becomes smaller, consistent with the general statistical relationship that smaller effects are harder to distinguish from random variation without more information.

These are illustrative planning values and not observed power. They do not say how much power the completed trial had after its results were known, and they do not establish the assumptions used by the trial investigators. The original sample-size justification and design assumptions are not reported in the registry record. The curve and figures should be read as conditional planning aids, not as a retrospective verdict on trial adequacy.

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 = 0.053; treatment proportion p₁ = 0.156; average p̄ = 0.105.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4331 = 0.8489.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4270 = 0.3593.
4. n = (0.8489 + 0.3593)² / (0.1030)² = 137.7, rounded up to 138 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 <- 0.0532915360501567                          # control proportion
p1 <- 0.15625                          # 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(2407236)
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 = 138 (80% power), 184 (90% power)
 75% of observed effect: n per arm = 219 (80% power), 293 (90% power)
 50% of observed effect: n per arm = 430 (80% power), 576 (90% power)
power.prop.test: 137.7 
Simulated power at n = 138 per arm: 0.829 (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 registry reports sponsor analyses using the Cochran-Mantel-Haenszel method. For ustekinumab 130 mg IV versus placebo, the adjusted treatment difference is 10.3, with a 95% confidence interval from 5.7 to 14.9 and p < 0.001. For approximately 6 mg/kg IV versus placebo, the adjusted treatment difference is 10.2, with a 95% confidence interval from 5.6 to 14.8 and p < 0.001.

The simple unadjusted recalculations from the posted counts give risk differences of 0.103 and 0.102, respectively, with confidence intervals that are close to the sponsor-reported intervals when expressed on the same percentage-point scale. The registry data do not flag a discrepancy. The sponsor analysis should therefore be treated as the formal reported result, while the recalculation is a check based on group summaries rather than a replacement for the sponsor’s analysis.

Adjustment can account for stratification or other analysis features, but the specific variables used are not reported in the registry record. The registry reports the method and adjusted results, not enough detail to reproduce that adjusted analysis from the summarized counts alone. The two analyses are also not identical in method: the recalculation is unadjusted, and the reported sponsor result is adjusted.

15. Multiplicity, Missing Data and Analysis Populations

The registry record describes one primary endpoint and numerous secondary outcome measures. It reports two primary induction comparisons against placebo and a broad set of secondary measures across induction and maintenance. When multiple outcomes or comparisons are tested, the probability of at least one nominally small p-value can increase. Statistical plans often address this through a prespecified hierarchy, adjustment, or designation of analyses as exploratory. A multiplicity strategy for the reported results is not reported in the registry record.

Without that information, the reported p-values should be interpreted as applying to their individual analyses, not as proof that the probability of any false-positive result across all outcomes is controlled. This matters especially for secondary outcomes, where the number of posted measures is large relative to the number of primary questions. The registry lists 71 secondary outcome measures; the multiplicity implications cannot be resolved from the outcome list alone.

The primary analysis population is defined as all participants randomized in induction. The endpoint description also specifies that participants with prohibited changes in concomitant ulcerative colitis medication, ostomy or colectomy before Week 8, or all four Mayo subscores missing at Week 8 were considered not to be in clinical remission. These rules are part of the estimand as operationalized in the record: they determine how certain events are reflected in the endpoint classification.

The registry does not report a comprehensive approach for other missing data, such as how partial missingness among the Mayo subscores was handled, or whether any sensitivity analyses assessed alternative assumptions. Details of missing-data handling for the secondary endpoints are also not reported in the registry record. The stated primary-endpoint rules are therefore informative but do not establish every data-handling decision.

The registry describes the sponsor’s primary method as Cochran-Mantel-Haenszel and gives adjusted results, but does not report all details needed to reconstruct the analysis, including the adjustment variables in this summary. Planned interim analyses, the original statistical analysis plan, and the original sample-size justification are not reported in the registry record. Their absence limits assessment of whether the reported analyses followed a prespecified testing plan.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 3% 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 (5.5%), cytomegalovirus infection (3.6%), jaw fracture (2.7%), suicidal ideation (2.7%), cardiac arrest (1.8%).

The registry reports serious adverse event counts separately by study group and study phase. During induction, serious adverse events affected 22 of 319 participants in the placebo IV group, 12 of 321 in the 130 mg IV group, and 11 of 320 in the approximately 6 mg/kg IV group. The safety denominators are not identical to the primary efficacy denominators in every group, so the safety counts should be read with their own reported denominators.

These are descriptive counts, not adjusted comparisons. The registry also reports additional groups for induction nonresponders, maintenance, and long-term extension. Those groups have different selection histories and follow-up contexts; comparing their counts directly with the original randomized induction groups would not be a like-for-like randomized safety comparison. The registry lists one death in the induction approximately 6 mg/kg group, one among maintenance delayed responders, and one in a long-term extension group that switched from placebo SC to ustekinumab SC 90 mg every 8 weeks. Counts alone do not establish cause or treatment attribution.

The listed serious-event terms and percentages describe reported event categories; they do not imply that the treatment caused an event. Safety interpretation depends on definitions, time at risk, clinical context, and comparative analyses, which are not fully conveyed by counts alone.

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

The Week 8 results describe the proportion of randomized induction participants classified as being in clinical remission under the registry’s global definition. The posted counts and the recalculated comparisons show higher observed remission proportions in each ustekinumab IV group than in the placebo IV group. The sponsor’s adjusted analyses also report positive treatment differences with p-values below the stated threshold.

These results do not establish that every participant benefited, that the outcome applies to all people with ulcerative colitis, or that the effect is the same in settings outside this trial. They concern a particular population, a specified remission definition, and an assessment at Week 8. The data do not by themselves answer questions about longer-term outcomes, all dimensions of health, or an individual participant’s likely outcome.

The induction result also does not demonstrate the comparative effect of the maintenance regimens. Maintenance participation depended on response to induction, and responders were then assigned among maintenance approaches. Findings in those selected groups answer a different question from the induction comparison across all randomized participants.

The reported p-values do not measure the chance that a treatment is effective, and statistical significance is not the same as clinical importance. The confidence intervals communicate the precision of the estimated group differences under the analysis assumptions, but do not capture every uncertainty about measurement, missing data, multiplicity, or generalizability. The registry record does not report enough detail to resolve all of those considerations.

Safety counts should likewise be interpreted descriptively. They report participants with serious adverse events in groups with different denominators and, in later study phases, different eligibility and exposure histories. The counts do not establish causation or provide a complete comparative safety assessment. No medical recommendation follows from these statistical summaries.

18. Why This Trial Matters Statistically

Statistically, the study illustrates the distinction between a clearly defined binary endpoint and the broader clinical questions that surround it. The primary outcome applies a specified threshold to the Mayo score and uses a central review of endoscopy video for the endoscopy component. That definition makes the endpoint reproducible in principle, while the stated rules for medication changes, surgery, and fully missing scores show how analysis decisions shape what the count represents.

The randomized induction comparison provides a direct placebo reference for two ustekinumab regimens. Reporting both adjusted sponsor estimates and unadjusted recalculations makes it possible to see how the summary-statistic contrast relates to the formal reported analysis. Their close alignment supports consistency of the numerical account, while the adjusted method and the unadjusted calculation remain distinct analyses.

The study also makes clear why later treatment stages need separate interpretation. Maintenance assignment follows induction response, which changes the population represented by those comparisons. The maintenance results cannot simply be read as an extension of the full randomized induction contrast.

Finally, the registry record contains many outcome measures alongside the primary result. This creates a practical lesson in multiplicity: a coherent pattern across different outcomes may offer context, but unadjusted secondary findings deserve less evidential weight when the planned hierarchy or adjustment is unavailable. Statistical evidence is strongest when endpoint definitions, analysis populations, uncertainty, missing-data rules, and the testing plan are considered together.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record provides a clear primary endpoint definition and reported group counts, but does not provide all information needed to independently reconstruct the sponsor’s adjusted analysis. The adjustment or stratification variables for the Cochran-Mantel-Haenszel analyses are not reported in this description. Accordingly, the recalculated risk differences are useful unadjusted summaries, not substitutes for a fully reproducible adjusted analysis.

Missing-data handling is only partly specified. The record states that all four Mayo subscores missing at Week 8 leads to classification as not in remission, and it identifies other conditions also classified as not in remission. It does not report how other patterns of incomplete endpoint data were handled or provide sensitivity analyses for alternative assumptions. The impact of those unreported decisions cannot be assessed from the posted summary.

The registry lists numerous secondary outcomes, but a multiplicity procedure or confirmatory testing order is not reported. This limits the strength of inference for secondary findings, particularly when considered individually. The original sample-size justification, design assumptions, and planned interim analyses are also not reported, so the planning illustrations based on observed effects should not be mistaken for the trial’s original design.

The primary result is an induction comparison at Week 8. The maintenance groups are selected based on induction response, and the safety listing spans induction, maintenance, and long-term extension groups with differing histories. These features make cross-phase comparisons less direct. The safety counts are descriptive and do not establish causal attribution. Finally, one endpoint and one time point cannot capture every relevant outcome or establish how results generalize beyond the studied population and trial setting.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction and Maintenance Therapy in Participants With Moderately to Severely Active Ulcerative Colitis — NCT02407236.
• Registry record last updated 2025-04-29; status last verified 2025-04.
• Results first posted 2019-12-23.
• 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

NCT02407236 is a completed Phase 3 randomized, double-blind, placebo-controlled, parallel-group, multicenter study of ustekinumab induction and maintenance therapy in moderately to severely active ulcerative colitis. The primary induction endpoint was clinical remission at Week 8 under the global Mayo score definition. The analysis set included all participants randomized in induction.

The registry reports 17 participants in remission among 319 assigned to placebo IV, 50 among 320 assigned to ustekinumab 130 mg IV, and 50 among 322 assigned to ustekinumab approximately 6 mg/kg IV. Sponsor-reported adjusted comparisons and unadjusted recalculations both indicate positive differences versus placebo, with the estimates described in the statistical interpretation.

Interpretation remains bounded by the endpoint definition, the induction time point, uncertainty, unreported multiplicity and design details, and the response-selected nature of maintenance groups. Safety counts are descriptive and span distinct study groups and phases.