Efficacy & Safety of Infliximab Monotherapy Vs Combination Therapy Vs AZA Monotherapy in Ulcerative Colitis (Part 1) Maintenance Vs Intermittent Therapy for Maintaining Remission (Part 2)(Study P04807): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy & Safety of Infliximab Monotherapy Vs Combination Therapy Vs AZA Monotherapy in Ulcerative Colitis (Part 1) Maintenance Vs Intermittent Therapy for Maintaining Remission (Part 2)(Study P04807) (NCT00537316), a randomized phase 3 study. The primary outcome is Proportion of Participants in Steroid-free Remission at Week 16.

2. Clinical Question

3. Trial Design

The registry describes a Phase 3, randomized, double-blind, parallel-group study with active treatment groups. Part 1 was designed to compare IFX alone, IFX with AZA, and AZA alone in adults with moderate to severe active ulcerative colitis. The enrollment field reports 242. The study status is terminated.

Part 1 was described as a 16-week study. Participants assigned to IFX alone or IFX/AZA were scheduled for IFX infusions at Weeks 0, 2, and 6. Depending on group, participants also received daily oral AZA or placebo. The AZA group received daily oral AZA and placebo infusions. Participants were evaluated for response at Week 8, and subsequent treatment was described as depending on response.

Part 2 concerned maintenance versus intermittent IFX treatment for participants in remission at Week 16. The brief summary describes randomization to these approaches, with stratification based on oral AZA/placebo treatment. Maintenance regimens are described as scheduled IFX infusions every 8 weeks; intermittent regimens used IFX upon relapse. Part 2 details should not be conflated with the Week 16 Part 1 endpoint, which is the primary outcome analyzed in this statistical interpretation.

The registry reports a start of 2007-07 and a primary completion of 2010-02. It does not report the reason for termination in the information available for this analysis. Further operational details, including the number of sites and the timing or conduct of any interim analyses, are not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: 2 participants were randomized but not treated. Part 1: 237 participants received ≥1 dose study medication. Part 2: Of 108 enrolled/randomized participants, 95 continued assigned treatment from Part 1 and 13 were randomized and entered into Part 2 of the study (10 participants from Part 1 and 3 participants who were enrolled directly into Part 2).

Only the first 6 of 7 groups are shown.

Part 1

Reasons for not completing this period:

Part 2

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (participants who were randomized, received at least one dose of study treatment, and had data available for baseline and at least one post baseline evaluation). Participants who dropped out prior to Week 16, had a missing Mayo score at Week 16, or were non-responders at Week 8 are considered failures (non-remission) at Week 16.

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.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Proportion of Participants in Steroid-free Remission at Week 16, measured over 16 weeks.

The primary endpoint is the proportion of participants in steroid-free remission at Week 16. The registry defines remission as a total Mayo score of 2 points or lower, with no individual sub-score exceeding 1 point, and no use of corticosteroids. The Mayo score combines stool frequency, rectal bleeding, endoscopic findings, and the physician’s global assessment. The endpoint therefore requires both a specified clinical score and absence of corticosteroid use; it is not simply a measure of symptom change or endoscopic findings alone.

A binary endpoint classifies each participant as meeting or not meeting the stated remission definition at the assessment time. Reporting the proportion meeting the definition makes the outcome interpretable as a group-level frequency, while retaining the endpoint’s specific clinical criteria. Steroid-free status is part of the definition, so remission achieved while using corticosteroids does not meet the posted endpoint definition.

Such endpoints are used in trials to assess whether participants reach a defined clinical state by a specified time. Their interpretation depends on how the state is defined, how missing assessments and withdrawals are handled, and which participants are included in the analysis. Here, the registry’s Full Analysis Set description is available, but its text is truncated. The detailed handling of missing Week 16 scores and other analysis rules is not fully reported in the registry record.

7. Results Posted in the Registry

• Azathioprine (AZA): 0.0% (0/76) versus 0.0% (0/77) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.
• IFX/AZA: 0.0% (0/78) versus 0.0% (0/77) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help clarify what the registry reports and what it leaves unresolved.

What does a proportion represent?

For a binary endpoint, each participant is classified according to whether the endpoint definition was met. The proportion summarizes the frequency of participants meeting that definition in a group. In this trial, the event is steroid-free remission at Week 16 under the posted Mayo-score and corticosteroid criteria. The registry posts proportions of 0.22 for IFX, 0.24 for AZA, and 0.4 for IFX/AZA. Those values are not interchangeable with a change in Mayo score or with a measure of remission at another time.

What does a risk difference represent?

A risk difference compares two group proportions by subtracting one from the other. It is expressed on the same proportion scale as the endpoint, and its direction depends on the order in which groups are contrasted. This is a general explanation of the statistic, not a new calculation for this study. Although the registry lists risk differences for two comparisons, those entries are 0.0 with limits of 0.0 to 0.0 and event counts of zero. Because those entries do not align with the posted arm-level proportions, they should not be used as meaningful estimates of the contrasts.

How should the chi-squared result be read?

The registry records a chi-squared test for AZA and IFX/AZA and reports p = 0.032. In general, a chi-squared test evaluates whether observed categorical outcome frequencies are compatible with a null model of no association or difference, given the test’s setup. The p-value is not the probability that the null hypothesis is true, and it does not measure the magnitude of an effect. The registry does not provide the test statistic, an effect estimate, or a confidence interval for this analysis. The specific construction of the test and any adjustments are not reported in the registry record.

What does a confidence interval add?

In general, a confidence interval conveys uncertainty around an estimated quantity under the method used to construct it. A narrow interval indicates greater statistical precision than a wide interval, while an interval spanning values on both sides of a null value may be compatible with a range of possible effects. These general points cannot be applied to a valid interval for the principal group comparison because the reported analysis gives no interval and the comparison fields contain zero limits that are not interpretable alongside the arm-level results.

Why do the analysis population and missing data matter?

The endpoint is reported for a Full Analysis Set described as participants who were randomized, received at least one dose, and had baseline data plus at least one post-baseline evaluation. The endpoint description also refers to participants who dropped out before Week 16, had a missing Mayo score at Week 16, or were non-responders at Week 8, but the description is truncated. Missing outcome handling can affect a binary endpoint because the analysis must determine how participants without an observed Week 16 result are classified or otherwise handled. The complete rule is not reported in the registry record.

What does the planning illustration mean?

In general, a sample-size calculation estimates how many participants per arm a future study would need under stated assumptions. The record describes a planning illustration based on the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is a stated fraction of the observed effect. Such calculations are planning illustrations, not observed power and not evidence about the probability that a finding is true. Numerical planning results are not reported in the information available for this analysis.

The record does not report the full statistical analysis plan, whether the reported test was prespecified, or the handling of multiple comparisons. Those omissions limit how fully the posted inferential result can be evaluated.

10. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimate. For a group proportion, precision depends in part on the number of observations and the frequency of the event; for a contrast, it also depends on the information in both groups and on the method used to estimate uncertainty. A confidence interval is generally used to express uncertainty around an estimate, but the registry does not report a usable confidence interval for the primary group comparison.

The arm-level endpoint values are 0.22 for IFX, 0.24 for AZA, and 0.4 for IFX/AZA, with posted counts of 77.0, 76.0, and 78.0, respectively. Those values describe the recorded endpoint results, but they do not by themselves show the sampling uncertainty around each proportion. The registry comparison fields contain limits of 0.0 to 0.0 for the listed risk differences. Since the fields also show zero event counts while arm-level proportions are nonzero, those limits do not offer a credible description of precision for the observed outcomes.

The chi-squared analysis reports p = 0.032 for AZA and IFX/AZA, but no confidence interval or effect estimate is attached. A p-value alone cannot show the range of effects compatible with the data. It also does not distinguish a precisely estimated modest contrast from a less precise larger contrast. Without a coherent effect estimate and interval, the precision of the comparison cannot be assessed from the registry’s posted analysis.

The registry does not report further details needed to assess precision, including the exact event counts underlying the endpoint proportions, the complete missing-data method, or the statistical plan for the primary comparison. Thus, the posted arm values and p-value should be read as limited summaries rather than a full account of uncertainty.

11. Analyses Reported by the Sponsor

The registry’s sponsor-reported analysis is a chi-squared test for AZA and IFX/AZA, with p = 0.032. It does not include an effect estimate or confidence interval. This is distinct from the listed recalculated comparison fields, which show risk differences of 0.0 and limits of 0.0 to 0.0, alongside zero event counts. Those fields are not interpretable as a valid reconstruction of the contrasts given the arm-level proportions of 0.24 for AZA and 0.4 for IFX/AZA.

The registry does not provide enough information to reproduce the sponsor’s chi-squared analysis from the posted summary. It does not report the event counts used in that test, a full test statistic, or a confidence interval. The sponsor’s p-value should therefore be understood as the reported inferential result, while the posted arm proportions describe the outcomes by group. Neither should be substituted for the other: a p-value is not an effect size, and an arm-level proportion does not by itself supply an uncertainty interval for a between-group contrast.

The discrepancy flag is false. Even so, the zero-valued recalculation entries do not explain the arm-level results, and the record does not give a reconciliation. No unadjusted re-calculation is therefore a useful alternative estimate of the treatment contrast.

12. Multiplicity, Missing Data and Analysis Populations

The study has multiple outcomes and treatment groups. The registry lists six outcome measures, with four secondary outcomes posted by count. These include response at Weeks 8 and 16, mucosal healing at Week 16, steroid-free remission during Part 2, and mucosal healing during Part 2. The primary endpoint is steroid-free remission at Week 16. When multiple outcomes or comparisons are examined, the chance of obtaining at least one small p-value under a collection of null hypotheses can increase. Statistical plans may address this through prespecified testing sequences or adjustments, but the record does not report whether such methods were planned or used.

The reported chi-squared result applies to AZA and IFX/AZA, but the registry does not state how that comparison relates to a prespecified testing strategy or to the other treatment comparisons. It does not report an adjustment for multiple comparisons. This absence of detail does not establish that no adjustment was made; it means the registry record does not allow that question to be answered.

The population description identifies a Full Analysis Set and specifies randomization, treatment receipt, and availability of baseline and post-baseline data. Its text is truncated. The primary endpoint description says that participants who dropped out before Week 16, had a missing Week 16 Mayo score, or were non-responders at Week 8 are relevant to the analysis, but the registry does not provide the complete classification or imputation rules. Missing data matter because the observed Week 16 remission proportion can depend on how participants without an observed assessment are handled.

The registry also does not report a complete statistical analysis plan, the prespecification status of the analysis, or details of any interim analyses. The handling of missing data, multiplicity, and analysis populations is therefore not fully assessable from this record. These limitations do not change the values that are posted, but they constrain the strength of conclusions that can be drawn from them.

13. 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: humerus fracture (10.0%), aspergillosis (5.0%), colitis ulcerative (4.1%), intestinal haemorrhage (2.7%), amoebiasis (2.7%).

The safety record separates serious adverse events and other adverse events by treatment period and group. Through Week 8, the serious-event entries are 5 among 79 at risk for AZA, 4 among 80 for IFX/AZA, and 3 among 78 for IFX. The corresponding other-event entries are 27 among 79, 17 among 80, and 17 among 78. These are registry counts, not estimates of comparative treatment effects.

After Week 8, participants appear under further treatment categories, including treatment changes and Part 2 regimens. The record lists serious and other adverse-event counts for these categories, but some groups have small at-risk counts. For example, the Part 2 maintenance and intermittent groups have at-risk counts of 5, 2, 4, and 2, depending on the listed category. The entries should be interpreted within their stated treatment periods and not combined across categories, because participants may have moved between regimens.

Death counts are not reported in the adverse-event entries. The record lists serious-event terms including Humerus Fracture at 10.0%, Aspergillosis at 5.0%, and Colitis Ulcerative at 4.1%, among others. The information does not establish how these terms relate to individual treatment periods or provide comparative risk estimates. Differences in follow-up and group composition also limit direct comparisons. These safety figures describe what the registry records; they do not establish causation or provide medical advice.

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

The posted Week 16 values describe the proportion recorded as being in steroid-free remission under a defined Mayo-score and corticosteroid-use criterion. The results are 0.22 for IFX, 0.24 for AZA, and 0.4 for IFX/AZA. The registry separately reports a chi-squared p-value of 0.032 for AZA and IFX/AZA. These values address different aspects of the result: the proportions describe group outcomes, while the p-value is an inferential statistic from a test.

The p-value is not the probability that one treatment is better, nor does it quantify the size or importance of a contrast. The record does not supply a usable effect estimate or confidence interval for the reported comparison. Its zero-valued risk-difference entries are not compatible with the posted arm-level proportions as a meaningful summary of the contrast. The direction and magnitude of the comparison therefore cannot be characterized with confidence from those recalculated fields.

These results do not establish that a treatment works or does not work. A registry result is interpreted in light of its design, endpoint definition, population, missing-data procedures, analysis plan, and completeness of reporting. The available record is limited on several of those points: the population description is truncated, missing-data rules are incomplete, and multiplicity handling is not reported. Secondary outcome results are also not numerically available in the information.

The registry lists the study as terminated, but the reason is not reported in the registry record. That status alone does not explain the results or indicate why the study ended. The appropriate reading is consequently bounded: the record reports endpoint proportions and a sponsor-reported test, while leaving important details about precision and analysis choices unresolved.

15. Why This Trial Matters Statistically

This trial illustrates why a reported p-value should not be read in isolation. The record includes arm-level proportions and a chi-squared p-value, but no effect estimate or confidence interval for the sponsor-reported comparison. The recalculated comparison fields contain zero estimates and zero limits despite nonzero arm-level proportions. That contrast between descriptive outcome summaries and an unusable comparison entry makes clear that a numerical p-value cannot substitute for a coherent estimate of effect.

It also illustrates how endpoint definitions shape interpretation. Steroid-free remission requires meeting the Mayo-score threshold and having no corticosteroid use. The result is therefore specific to that combined definition at Week 16, rather than a general measure of disease activity or remission at any point in follow-up.

The record further shows why population and missing-data methods matter for binary outcomes. Participants without a Week 16 assessment need explicit handling, and the registry’s endpoint description is incomplete on this point. In addition, the study lists several outcomes and comparisons, but multiplicity handling is not reported. These are not merely technical details: they determine how results are assembled and how confidently a reader can interpret inferential claims.

Finally, the safety entries are arranged across changing treatment periods and Part 2 groups, with differing at-risk counts. Reading these as if they were directly comparable parallel-group estimates would overlook the structure of follow-up. The statistical lesson is to keep the endpoint, analysis population, contrast, uncertainty, and safety observation period distinct when evaluating a clinical-trial record.

16. 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 record has several trial-specific reporting limitations. The primary endpoint population description is truncated, leaving the complete Full Analysis Set definition unavailable. The endpoint text mentions early dropout, missing Week 16 scores, and Week 8 non-response, but the full rules for classifying or handling those situations are not reported in the registry record.

The sponsor-reported chi-squared analysis gives p = 0.032 for AZA and IFX/AZA but no effect estimate or confidence interval. The recalculated risk-difference entries are all 0.0 with limits of 0.0 to 0.0 and zero event counts, while the arm-level proportions are nonzero. The registry does not explain this mismatch in the posted summaries. Accordingly, the comparison’s magnitude and precision cannot be established from these entries.

The registry lists four secondary outcomes, but their numerical results are not reported in the information available for this analysis. The multiplicity strategy, prespecification of the reported comparison, full statistical analysis plan, and interim-analysis details are also not reported in the registry record. The reason for the terminated status is not reported.

Safety reporting spans treatment periods and groups that include treatment changes and Part 2 regimens. These categories should not be treated as directly comparable randomized groups without additional design and follow-up detail. Death counts are not reported. Overall, the available record supports description of the posted values but does not provide a complete basis for assessing analysis robustness, precision, or causal interpretation.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Efficacy & Safety of Infliximab Monotherapy Vs Combination Therapy Vs AZA Monotherapy in Ulcerative Colitis (Part 1) Maintenance Vs Intermittent Therapy for Maintaining Remission (Part 2)(Study P04807) — NCT00537316.
• Registry record last updated 2017-04-12; status last verified 2017-03.
• Results first posted 2011-12-16.
• 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.

20. Record Summary

The registry record describes a Phase 3, randomized, double-blind, parallel-group study of IFX, AZA, and IFX/AZA for moderate to severe active ulcerative colitis. The enrollment field is 242, and the study status is terminated. The primary endpoint is steroid-free remission at Week 16, defined by a total Mayo score of 2 points or lower, no sub-score above 1 point, and no corticosteroid use.

Posted remission proportions are 0.22 for IFX, 0.24 for AZA, and 0.4 for IFX/AZA. A chi-squared analysis for AZA and IFX/AZA reports p = 0.032, without an effect estimate or confidence interval. The recalculated comparison entries are not interpretable alongside the arm-level proportions. Missing-data handling, multiplicity, and other analysis details are not fully reported in the registry record.