A Study Comparing the Efficacy and Safety of Etrolizumab With Adalimumab and Placebo in Participants With Moderate to Severe Ulcerative Colitis (UC) in Participants Naive to Tumor Necrosis Factor (TNF) Inhibitors: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Comparing the Efficacy and Safety of Etrolizumab With Adalimumab and Placebo in Participants With Moderate to Severe Ulcerative Colitis (UC) in Participants Naive to Tumor Necrosis Factor (TNF) Inhibitors (NCT02163759), a randomized phase 3 study. The primary outcome is Percentage of Participants in Remission at Week 10 With Etrolizumab as Compared With Placebo, as Determined by the Mayo Clinic Score (MCS), GA28948 Population.

2. Clinical Question

The central statistical question concerns the difference in the chance of meeting the specified remission definition at the assessment time. It is not a question about the speed of improvement or the duration of remission: the listed primary outcome is a binary status at Week 10. Participants with missing Week 10 assessments, or who received permitted or prohibited rescue therapy before assessment, were classified as non-remitters. That rule affects how the observed outcome represents the treatment groups and is relevant when interpreting the estimates.

3. Trial Design

This was a Phase III, randomized, double-blind, double-dummy, placebo-controlled, active-comparator, multicenter, parallel-group study. The registry record lists an enrollment of 358 and a completed study. Randomization assigned participants to placebo, adalimumab or etrolizumab. The design included placebo and active-comparator arms, permitting the primary etrolizumab–placebo comparison and additional comparisons with adalimumab.

Double-dummy masking means participants in each active arm received matching placebo for the other treatment, accommodating the different schedules while keeping treatment assignment concealed. The placebo arm received placebo matching etrolizumab up to Week 12 and placebo matching adalimumab up to Week 8. The adalimumab arm received active adalimumab up to Week 8 and matching etrolizumab placebo up to Week 12. The etrolizumab arm received active etrolizumab up to Week 12 and matching adalimumab placebo up to Week 8. The registry record does not report further details of the dosing schedules in the listed arm descriptions.

The primary outcome is reported for the GA28948 modified intent-to-treat population, defined as randomized participants who received at least one dose of study drug, with analysis according to randomized assignment. The primary analysis adjusted for the stated stratification factors: corticosteroid or immunosuppressant treatment at randomization, and screening disease activity categorized by Mayo Clinic Score. The specific allocation ratio, randomization sequence, and trial-specific sample-size justification are not reported in the registry record.

The registry describes an independently conducted study with identical design, GA28949, and lists pooled secondary outcomes combining GA28948 and GA28949. Those pooled outcomes should be distinguished from primary results for GA28948 alone: they draw on a broader, separately defined analysis population and do not replace the primary comparison in this study.

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: Participants who were on concomitant background therapy were allowed to continue receiving stable baseline doses of the following non-investigational medicinal products during the study: oral 5-aminosalicylic acid; azathioprine; 6-mercaptopurine; methotrexate; corticosteroids up to 30 mg/day of prednisone (or equivalent); and/or budesonide up to 9 mg/day.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: GA28948 Population, Modified Intent-to-Treat: including all randomized participants in study GA28948 who received at least one dose of study drug, with participants grouped according to the treatment assigned at randomization.

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 overall number of baseline participants (716) represents the total number enrolled (358) in this study, GA28948 (NCT02163759), plus the total number enrolled (358) in a second study of identical design, GA28949 (NCT02171429). Some of the pre-specified secondary outcome measure comparisons were planned to be evaluated by pooling data from both studies; all etrolizumab versus adalimumab comparisons and one family of etrolizumab versus placebo comparisons were evaluated on pooled data.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants in Remission at Week 10 With Etrolizumab as Compared With Placebo, as Determined by the Mayo Clinic Score (MCS), GA28948 Population, measured over Week 10.

The primary endpoint is the percentage of participants in remission at Week 10, measured using the Mayo Clinic Score. The score ranges from 0 to 12 and combines stool frequency, rectal bleeding, endoscopic appearance and physician’s global assessment. Each component is rated from 0 to 3, with higher scores indicating more severe disease.

Remission was defined as a total Mayo Clinic Score of at most 2, with each individual subscore at most 1 and a rectal bleeding subscore of 0. Participants with missing Week 10 assessments or who had received permitted or prohibited rescue therapy before the assessment were classified as non-remitters. This makes the endpoint a binary outcome: each participant is classified as meeting or not meeting the prespecified remission criteria at that assessment.

A remission endpoint brings together symptoms, endoscopic assessment and clinician assessment in a clinically interpretable status measure. It is used to compare the proportion meeting the same defined threshold across randomized groups. The endpoint does not describe every aspect of disease activity, change over time, or the durability of remission after Week 10. The Mayo Clinic Score definition and the rules for missing assessments and rescue therapy are therefore important to understanding what the percentage represents.

7. Results Posted in the Registry

• Etrolizumab: 19.4% (28/144) versus 6.9% (5/72) on control; risk difference 12.5 percentage points (95% CI 3.8 to 21.2), risk ratio 2.80 (1.13–6.95); unadjusted two-proportion z-test p = 0.016.

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 in Remission at Week 10 With Etrolizumab as Compared With Adalimumab, as Determined by the MCS, GA28948 Population

• 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:
  • Etrolizumab: 19.4% (28/144) versus 22.5% (32/142) on control; risk difference -3.1 percentage points (95% CI -12.5 to 6.3), risk ratio 0.86 (0.55–1.35); unadjusted two-proportion z-test p = 0.521.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etrolizumab and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants in Remission at Week 10 With Etrolizumab as Compared With Adalimumab, as Determined by the MCS, GA28948 & GA28949 Pooled Population

• 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:
  • Etrolizumab: 18.8% (54/287) versus 23.5% (67/285) on control; risk difference -4.7 percentage points (95% CI -11.4 to 2.0), risk ratio 0.80 (0.58–1.10); unadjusted two-proportion z-test p = 0.169.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etrolizumab and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants With Clinical Response at Week 10, as Determined by the MCS, GA28948 Population

• 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:
  • Adalimumab: 52.1% (74/142) versus 50.0% (36/72) on control; risk difference 2.1 percentage points (95% CI -12.1 to 16.3), risk ratio 1.04 (0.79–1.38); unadjusted two-proportion z-test p = 0.770.
  • Etrolizumab: 56.9% (82/144) versus 50.0% (36/72) on control; risk difference 6.9 percentage points (95% CI -7.2 to 21.0), risk ratio 1.14 (0.87–1.49); unadjusted two-proportion z-test p = 0.334.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adalimumab and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Etrolizumab and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the description of Week 10 results beyond the primary remission comparison. They include remission comparisons with adalimumab in GA28948 and in a pooled GA28948 and GA28949 population, clinical response in GA28948, and measures of endoscopic improvement and endoscopic remission. Together, these outcomes address related but distinct aspects of disease activity; they are not interchangeable with the primary remission definition.

Across the posted comparisons with adalimumab, the remission estimates do not establish a clear separation between etrolizumab and adalimumab: the reported confidence intervals include no difference. The posted clinical-response proportions are relatively close across placebo, adalimumab and etrolizumab, and the sponsor-reported intervals for the listed response contrasts include no difference. These results do not overturn the primary etrolizumab–placebo finding, because they use different comparisons and endpoints. Nor do they establish equivalence between active treatments.

The pooled remission analysis combines GA28948 with the independently conducted GA28949 study. It is a separate analysis population from the primary GA28948 population and should be interpreted as such. The registry record does not report details sufficient to assess the full pooling strategy beyond identifying the combined population.

There are many listed secondary outcomes and analyses. When numerous outcomes are tested, some may appear noteworthy by chance, even if no underlying effect exists for those outcomes. An adjustment for multiple comparisons is not reported in the registry record. That absence means nominal secondary findings deserve less weight than a prespecified primary result analyzed under the stated method. The secondary results are best treated as context and as a basis for further study, rather than as independent confirmation of the primary conclusion.

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

• Percentage of Participants in Remission at Week 10 With Etrolizumab as Compared With Adalimumab, as Determined by the MCS, GA28948 Population
• Percentage of Participants in Remission at Week 10 With Etrolizumab as Compared With Adalimumab, as Determined by the MCS, GA28948 & GA28949 Pooled Population
• Percentage of Participants With Clinical Response at Week 10, as Determined by the MCS, GA28948 Population
• Percentage of Participants With Clinical Response at Week 10 With Etrolizumab as Compared With Adalimumab, as Determined by the MCS, GA28948 & GA28949 Pooled Population
• Percentage of Participants With Improvement in Endoscopic Appearance of the Mucosa at Week 10, as Determined by the MCS Endoscopy Subscore, GA28948 Population
• Percentage of Participants With Improvement in Endoscopic Appearance of the Mucosa at Week 10 With Etrolizumab as Compared With Adalimumab, as Determined by the MCS Endoscopy Subscore, GA28948 & GA28949 Pooled Population
• Percentage of Participants in Endoscopic Remission at Week 10, as Determined by the MCS Endoscopy Subscore, GA28948 Population
• Percentage of Participants in Endoscopic Remission at Week 10 With Etrolizumab as Compared With Adalimumab, as Determined by the MCS Endoscopy Subscore, GA28948 & GA28949 Pooled Population

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is binary: participants either met the remission definition at Week 10 or did not. For a binary outcome, a group percentage expresses the frequency of participants meeting the event definition in that group. The registry reports percentages with confidence intervals for the treatment groups, as well as a between-group difference. The difference is easier to read as an absolute contrast in event frequency than as a relative measure, although the sponsor-reported adjusted estimate is the principal inferential result.

What does the adjusted analysis do?

The sponsor used a Cochran-Mantel-Haenszel analysis to estimate the difference in remission rates while adjusting for the stratification factors used in the trial: concomitant corticosteroid or immunosuppressant treatment at randomization, and screening disease activity categorized by Mayo Clinic Score. In general, stratification adjustment compares treatment groups within relevant categories and combines information across those categories. This can account for differences in the distribution of those factors across the observed groups. The registry record does not report the detailed implementation beyond identifying the method and the factors.

The sponsor’s adjusted result is not numerically identical to a simple comparison calculated from the displayed group percentages. That is expected to be possible because the simple recalculation does not account for the stratification factors. In this record, the adjusted and unadjusted results point in the same direction and are close. The discrepancy flag is false, so there is no recorded material discrepancy between them. The sponsor’s own analysis remains the key reported test of the primary comparison.

How should the confidence interval be read?

A confidence interval expresses the uncertainty associated with an estimated effect under the assumptions of the analysis. It gives a range of values compatible with the data and method, rather than identifying a single known treatment effect. For the sponsor’s primary analysis, the reported 95% confidence interval for the difference in remission rates extends from 1.59 to 20.60. Because the interval is entirely on the positive side of no difference, it is consistent with a positive etrolizumab–placebo contrast under that analysis. The interval’s width also indicates that the estimate is not precise enough to pin down a narrow value for the difference.

A confidence interval is not a guarantee that the true effect lies within its bounds, and it does not describe the range of individual participant outcomes. It is also not interchangeable with a prediction interval. Its interpretation depends on the chosen analysis and the endpoint definition, including how missing assessments and rescue therapy were handled.

What does a p-value contribute?

A p-value evaluates how compatible the observed data are with a specified null hypothesis, under the statistical model and test. It does not measure the probability that the null hypothesis is true, the size of an effect, or its importance. The registry reports a superiority analysis for the primary comparison. The direction and interval estimate add essential context beyond a significance test: they show the estimated contrast and its uncertainty. The registry record does not provide enough detail to infer a broader testing strategy from the primary test alone.

Why can binary outcomes need careful interpretation?

For binary outcomes, the observed percentage depends on both the event definition and the analysis population. Here, missing Week 10 assessments and rescue therapy before assessment were classified as non-remission. This rule avoids treating those participants as if their remission status were simply unobserved in the reported outcome, but it also means that the resulting percentage incorporates those classifications. The modified intent-to-treat population excludes randomized participants who did not receive at least one dose, while retaining the randomized assignment for those included. That population definition should be considered when comparing the result with analyses using different inclusion rules.

General statistical considerations also apply to the secondary outcomes. Repeated testing across multiple endpoints and treatment contrasts increases the chance of obtaining a nominally notable result by chance if every result is interpreted separately without adjustment. The registry lists many outcome measures and analyses, but a multiplicity adjustment for the secondary results is not reported in the registry record. Accordingly, secondary estimates are best read as supportive descriptions rather than as a collection of independent confirmatory findings.

11. Confidence Intervals and Precision

The primary sponsor-reported 95% confidence interval for the difference in remission rates runs from 1.59 to 20.60. Its location above no difference is consistent with the positive estimated contrast, while its width signals meaningful uncertainty about the size of that contrast. The group percentages, 19.4% with etrolizumab and 6.9% with placebo, describe the observed proportions; they do not remove uncertainty about how well those observations estimate the underlying group outcomes.

Confidence intervals for the group-specific percentages also show uncertainty in each estimate. The posted interval for placebo runs from 3.0 to 15.25, and the interval for etrolizumab runs from 13.81 to 26.67. These intervals overlap. Overlap or non-overlap between separate group intervals is not a substitute for the formal interval for the treatment difference, particularly when the primary analysis adjusts for stratification factors. The sponsor-reported interval for the difference is the relevant interval for that adjusted comparison.

Precision is affected by the number of observed outcomes and the size of the analysis groups, as well as the variability inherent in a binary endpoint. A confidence interval can be compatible with a range of effects even when the point estimate is positive. Precision does not by itself determine whether an effect is important; that judgment requires a relevant clinical context, which the registry results do not define. The confidence interval describes uncertainty for the stated endpoint and analysis, not the range of individual responses or the likely result in another population.

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 table and curve are planning illustrations derived from the observed effect, not the trial’s original design assumptions. For these binary outcomes, the entries are participants per arm, not total participants and not event counts. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The original trial’s sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm that would be needed, under those assumptions, for an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column expresses the corresponding planning target for a 90% chance. These are design calculations, not descriptions of the probability that the completed trial’s result is correct.

For the primary etrolizumab–placebo contrast, planning for 100% of the observed effect gives 114 participants per arm for 80% power and 152 participants per arm for 90% power. Planning for 75% of the observed effect gives 183 participants per arm for 80% power and 244 participants per arm for 90% power. Planning for 50% of the observed effect gives 363 participants per arm for 80% power and 486 participants per arm for 90% power.

The smaller the effect a future study aims to detect, the more participants are generally needed to distinguish it from chance variation under otherwise similar assumptions. The examples are illustrations based on the observed effect, which itself is uncertain; they are not observed power and should not be treated as a definitive recommendation for a future study.

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 = 0.069; treatment proportion p₁ = 0.194; average p̄ = 0.132.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4786 = 0.9381.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4704 = 0.3959.
4. n = (0.9381 + 0.3959)² / (0.1250)² = 113.9, rounded up to 114 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 <- 0.0694444444444444                          # control proportion
p1 <- 0.194444444444444                          # 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(2163759)
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 = 114 (80% power), 152 (90% power)
 75% of observed effect: n per arm = 183 (80% power), 244 (90% power)
 50% of observed effect: n per arm = 363 (80% power), 486 (90% power)
power.prop.test: 113.9 
Simulated power at n = 114 per arm: 0.814 (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 analysis is a Cochran-Mantel-Haenszel comparison adjusted for the stratification factors specified in the primary outcome description. It reports a difference in remission rates of 12.3, with a 95% confidence interval from 1.59 to 20.60. The associated group percentages are 19.4% for etrolizumab and 6.9% for placebo.

The recalculation uses the posted summary statistics in a simple unadjusted comparison. It is not the same analysis as the sponsor’s stratified method, so small differences between the estimates or intervals are possible. In this case, the recalculated result is close to the sponsor’s adjusted estimate and has the same direction. The registry discrepancy flag is false; the record does not identify a material conflict between the reported analysis and the recalculation.

The adjusted sponsor analysis should therefore anchor interpretation of the primary endpoint. The recalculation provides a transparent, complementary check based on the posted group summaries, but it does not reproduce the stratification adjustment. Neither estimate should be read without its confidence interval or the endpoint rules, especially the handling of missing Week 10 assessments and rescue therapy.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 27 outcome measures and 35 analyses, including multiple secondary remission, response and endoscopic outcomes. Evaluating many endpoints or treatment contrasts creates a multiplicity problem: the more separate statistical tests considered, the greater the chance that at least one result may appear unusual by chance alone. This general issue is relevant even when each individual comparison is calculated correctly.

A strategy for handling multiplicity might specify which outcomes are confirmatory, how tests are ordered, or how the threshold for interpreting several tests is controlled. Such a strategy is not reported in the registry record. It is also not reported whether the secondary analyses were intended as confirmatory or exploratory, or whether any adjustment was applied outside the information listed for these outcomes. Consequently, the secondary p-values and confidence intervals should not be treated as if each arose from an isolated, prespecified test with no other comparisons.

The primary endpoint is clearly identified, and its analysis is reported as a stratified Cochran-Mantel-Haenszel comparison. However, the registry record does not report a complete statistical analysis plan, the timing of any interim analyses, or the original sample-size justification. The absence of those details does not establish that such planning or procedures did not occur; it means they cannot be assessed from the registry record.

The primary outcome description gives the modified intent-to-treat population and specifies how missing Week 10 assessments and rescue therapy were classified. That is useful information, but details of other analysis populations, sensitivity analyses for missing data, and the handling of protocol deviations are not reported in the registry record. The primary estimate should therefore be understood within the population and outcome rules that are documented, without assuming additional analyses or safeguards not described there.

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 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 (2.1%), pneumonia bacterial (1.4%), pyelonephritis acute (1.4%), prostatitis (1.4%), anaemia (0.7%).

The registry reports participants with serious adverse events, participants with other adverse events, and deaths by treatment group. Serious adverse events were reported for 2 among 72 placebo participants, 3 among 142 adalimumab participants, and 8 among 144 etrolizumab participants. Other adverse events were reported for 7 among 72 placebo participants, 17 among 142 adalimumab participants, and 6 among 144 etrolizumab participants. Deaths were reported as 0 in the placebo group, 0 in the adalimumab group, and 1 in the etrolizumab group.

These are descriptive counts from the registry record. They do not, by themselves, establish that a treatment caused an event or that the groups differ reliably in safety. The groups differ in size, and the record excerpt does not provide event timing, exposure-adjusted rates, detailed event-level context, or a statistical comparison of these safety figures. The listed top serious-event terms include ulcerative colitis, bacterial pneumonia, acute pyelonephritis, prostatitis, anaemia and cardiac arrest; the displayed term percentages should be read as reported categories, not as a complete safety profile.

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

The primary result describes a higher observed Week 10 remission percentage in the etrolizumab group than in the placebo group, with a sponsor-reported adjusted difference whose 95% confidence interval is entirely above no difference. This is evidence about the defined group comparison in the GA28948 modified intent-to-treat population and under the documented remission and analysis rules.

It does not mean every participant assigned to etrolizumab met the remission criteria, or that participants assigned to placebo could not do so. It does not predict an individual participant’s outcome. The endpoint is a status at Week 10, so it does not establish how quickly remission occurred or whether it persisted after the assessment. The registry result also does not establish effects on every symptom, complication, or outcome that is not included in the endpoint.

The primary comparison is etrolizumab versus placebo. It should not be recast as proof that etrolizumab is superior to adalimumab: the listed remission comparisons with adalimumab are secondary and have confidence intervals that include no difference. Those findings also do not prove the treatments are equivalent. A failure to establish a difference is not the same as a well-designed equivalence or noninferiority conclusion.

Safety counts need similar restraint. The serious adverse event and death figures are descriptive, and the registry information presented does not establish causation or a comparative safety effect. Broader judgments about benefits and harms require context beyond the primary endpoint and the listed counts. No medical advice follows from these statistical results; they describe trial-level evidence, not an individual treatment decision.

18. Why This Trial Matters Statistically

This trial is statistically informative because it uses a randomized, double-blind, double-dummy design with both placebo and active-comparator arms. Randomization supports a fair group comparison in expectation, while masking and matching placebos are design features intended to reduce differences in expectations and assessment associated with knowing treatment assignment. The primary endpoint is defined in advance in the registry record as a Week 10 remission measure, and the analysis accounts for stated stratification factors.

The result also illustrates why an effect estimate and its uncertainty should be read together. The observed remission percentages differ, and the adjusted confidence interval for their contrast excludes no difference, while still spanning a broad range of possible positive differences. The distinction between adjusted sponsor analysis and unadjusted recalculation matters: both point in the same direction here, but only the former reflects the documented stratification adjustment.

The presence of an active comparator and pooled secondary analyses adds context but does not turn every comparison into a confirmatory test. Different outcomes, populations and comparisons answer different questions. The many listed secondary analyses make multiplicity an important interpretive issue, particularly because an adjustment is not reported in the registry record.

Finally, sample-size planning illustrations show how planning requirements depend on the effect a future study seeks to detect. They are not the study’s original design rationale and cannot explain the choices made for this completed trial. Distinguishing observed results, adjusted inference, secondary evidence and illustrative planning calculations is essential for a statistically careful reading.

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 original sample-size justification or enough detail to assess how the planned enrollment was chosen. The planning figures based on the observed effect are retrospective illustrations, not evidence of the design assumptions used for this trial. Randomization ratio, sequence generation and concealment details are not reported in the record excerpt.

The primary analysis population includes randomized participants who received at least one dose, rather than every randomized participant. The outcome description classifies missing Week 10 assessments and rescue therapy before assessment as non-remission; this is important to the estimand represented by the reported percentage. The registry record does not report sensitivity analyses that would show how alternative missing-data assumptions might affect the result.

Several secondary outcomes and analyses are listed, but multiplicity handling is not reported. The pooled outcomes incorporate an independent study, and the available information does not detail all aspects of that pooling. Secondary comparisons should therefore be distinguished from the primary result and treated with appropriate caution.

The registry excerpt also does not provide detailed exposure duration, event timing, or adjusted comparative safety analyses. The adverse event counts cannot by themselves establish causality or provide a full account of relative safety. Generalizability is limited to the population and eligibility context described in the record; further participant characteristics and subgroup details are not reported in the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study Comparing the Efficacy and Safety of Etrolizumab With Adalimumab and Placebo in Participants With Moderate to Severe Ulcerative Colitis (UC) in Participants Naive to Tumor Necrosis Factor (TNF) Inhibitors — NCT02163759.
• Registry record last updated 2021-07-23; status last verified 2021-07.
• Results first posted 2021-04-05.
• 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

NCT02163759 is a completed randomized, double-blind, double-dummy Phase III parallel-group study of etrolizumab, adalimumab and placebo in participants with moderate to severe ulcerative colitis who were naive to TNF inhibitors. The primary outcome was Week 10 remission by Mayo Clinic Score in the GA28948 modified intent-to-treat population.

The sponsor reported an adjusted difference in remission rates of 12.3 for etrolizumab versus placebo, with a 95% confidence interval from 1.59 to 20.60. The posted group percentages were 19.4% and 6.9%, respectively. The adjusted and unadjusted analyses are close and point in the same direction. Secondary comparisons with adalimumab and safety counts provide additional context but do not change the definition or interpretation of the primary comparison.