Phase III Study of MLN0002 (300 mg) in Treatment of Crohn's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase III Study of MLN0002 (300 mg) in Treatment of Crohn's Disease (NCT02038920), a randomized phase 3 study. The primary outcome is Induction Phase: Percentage of Participants With Crohn's Disease Activity Index (CDAI)-100 Response.

2. Clinical Question

The primary endpoint is a binary response: each participant either met the specified CDAI decrease or did not. The reported comparison therefore concerns the proportions responding in the two induction groups, rather than an average change in CDAI or a time-to-event outcome. The registry record also lists clinical remission and changes in C-reactive protein during induction, along with maintenance response and remission outcomes. Those outcomes address related but different questions and should not be treated as substitutes for the primary endpoint.

The maintenance comparisons apply to selected participants who met the Week 10 CDAI-70 response criterion and entered the corresponding maintenance groups. They do not represent a direct continuation of the full induction-population comparison. The registry record does not report a clinical question about every enrolled participant continuing through every phase.

3. Trial Design

This was a phase 3, multicenter, randomized, double-blinded, placebo-controlled, parallel-group study sponsored by Takeda. Its official title describes evaluation of efficacy, safety and pharmacokinetics of intravenous MLN0002 (300 mg) in induction and maintenance therapy in Japanese participants with moderate or severe Crohn's disease. The registry record identifies the study as completed and reports enrollment of 157.

In induction, the active-treatment group received vedolizumab 300 mg by intravenous infusion at Weeks 0, 2 and 6; the comparator group received matching placebo infusions on the same schedule. The primary outcome was assessed at Week 10. For that endpoint, the full analysis set included participants who were randomized and received at least one dose of study drug during induction. The posted endpoint groups contain 79 participants assigned vedolizumab and 78 assigned placebo.

The maintenance phase involved a response-selected population. Participants who had received vedolizumab in induction and achieved CDAI-70 response at Week 10 were randomized to maintenance vedolizumab or placebo. The registry record also describes participants who received placebo in induction and achieved the same response criterion as continuing placebo without maintenance randomization. Maintenance treatment infusions were scheduled at Weeks 14, 22, 30, 38, 46 and 54. The record lists a Week 60 maintenance CDAI-100 response outcome and other maintenance outcomes.

An open-label vedolizumab group is also described, with infusions at Weeks 0, 2 and 6 and every 8 weeks thereafter up to Week 94 as the maximum duration. The trial's masking designation is quadruple, while its brief summary describes it as double-blinded. The detailed operational meaning of the registry masking designation is not reported in the registry record. Original design assumptions, the sample-size justification and details 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: Participants with moderate to severe Crohn's disease were enrolled. 157 participants enrolled in induction phase, 41 participants entered maintenance phase and 134 participants entered open-label cohort and received placebo or vedolizumab 300 mg. Open-label cohort occurred between Week 10 and 154 through study with maximum of 94 weeks of treatment.

Induction Phase (Week 0 to 14)

Reasons for not completing this period:

Maintenance Phase (Week 14 to 60)

Reasons for not completing this period:

Open Label Cohort (Week 10 to 154)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS) in the induction phase included participants who were randomized and received at least one dose of the study drug in induction phase.

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: All participants who entered in the induction phase.

6. Primary Endpoint

The registry lists the primary outcome as Induction Phase: Percentage of Participants With Crohn's Disease Activity Index (CDAI)-100 Response, measured over Week 10.

The primary endpoint is the percentage of participants with a CDAI-100 response at Week 10 during induction. A response was defined as a decrease from baseline of at least 100 points in the Crohn's Disease Activity Index. CDAI is a scoring system for Crohn's disease activity; the registry record describes a total score ranging from 0 to approximately 600, with higher scores indicating more severe disease. It also states that index values of 150 and below are associated with quiescent disease, while values above that indicate active disease.

This endpoint converts each participant's change into a response classification based on the stated threshold. Reporting the percentage who meet that threshold makes it possible to compare the frequency of this particular response between randomized induction groups. It does not report the size of the CDAI change among responders, the average change across all participants, or whether participants met a separate remission definition.

A threshold-based response is useful because it answers a clinically framed question about how many participants crossed a prespecified change boundary. Its interpretation still depends on the definition and timing: the registry outcome is assessed at Week 10, and response means a decrease from baseline of at least 100 points. The specific definition used for clinical remission is not reported in the registry record. The endpoint's uncertainty is important as well as its point estimate, because a percentage from a finite group can be compatible with a range of underlying response rates.

7. Results Posted in the Registry

• Induction Phase: Vedolizumab, 300 mg: 26.6% (21/79) versus 16.7% (13/78) on control; risk difference 9.9 percentage points (95% CI -2.9 to 22.7), risk ratio 1.59 (0.86–2.96); unadjusted two-proportion z-test p = 0.132.

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 Phase: Percentage of Participants With Clinical Remission

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Induction Phase: Vedolizumab, 300 mg: 17.7% (14/79) versus 10.3% (8/78) on control; risk difference 7.5 percentage points (95% CI -3.3 to 18.2), risk ratio 1.73 (0.77–3.89); unadjusted two-proportion z-test p = 0.178.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Induction Phase: Vedolizumab, 300 mg and the control. The interval gives the range of effects that remain compatible with the data.

Maintenance Phase: Percentage of Participants With Crohn's Disease Activity Index (CDAI)-100 Response

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Maintenance Phase: Vedolizumab 300 mg: 58.3% (7/12) versus 8.3% (1/12) on control; risk difference 50.0 percentage points (95% CI 18.0 to 82.0), risk ratio 7.00 (1.01–48.54); unadjusted two-proportion z-test p = 0.009.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Maintenance Phase: Vedolizumab 300 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Maintenance Phase: Percentage of Participants With Durable Clinical Remission

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Maintenance Phase: Vedolizumab 300 mg: 33.3% (4/12) versus 25.0% (3/12) on control; risk difference 8.3 percentage points (95% CI -27.9 to 44.5), risk ratio 1.33 (0.38–4.72); unadjusted two-proportion z-test p = 0.653.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Maintenance Phase: Vedolizumab 300 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes address several related aspects of the study: induction clinical remission and C-reactive protein, maintenance response and remission, and vedolizumab concentration and antibody outcomes. The posted binary secondary results do not form a uniformly precise pattern. In induction, the clinical-remission comparison's reported uncertainty includes no difference. In maintenance, the CDAI-100 result is more favorable to the vedolizumab group in the posted comparison, whereas durable clinical remission has an interval that includes no difference. These outcomes use different definitions, assessment periods and populations, so they should not be combined into a single summary effect.

The induction secondary result is relevant context for the primary Week 10 CDAI-100 finding, but it does not replace it. Clinical remission and CDAI-100 response are distinct endpoints; the registry record does not report the remission definition. Likewise, the maintenance findings relate to a selected group that met the Week 10 CDAI-70 response criterion, rather than the full induction population. Any apparent contrast between primary and secondary estimates therefore has to be understood in light of the endpoint and population differences.

The registry lists 13 posted secondary outcomes, while only some binary outcomes have detailed comparison results in the record summarized for this analysis. Statistical testing across multiple outcomes raises a multiplicity concern: the more hypotheses considered, the greater the chance that at least one result appears unusual under a collection of null hypotheses. Whether the trial used a prespecified hierarchy or adjustment is not reported in the registry record. Consequently, unadjusted secondary p-values carry less weight than a result interpreted within a clearly stated, prespecified testing strategy.

Secondary outcomes can help describe consistency, explore different dimensions of outcome, and identify questions for further study. They should not be treated as independent confirmation of the primary endpoint when the statistical plan and multiplicity handling are unavailable. Nor should an isolated favorable comparison outweigh broad confidence intervals or the distinction between induction and response-selected maintenance populations.

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

• Induction Phase: Percentage of Participants With Clinical Remission
• Induction Phase: Change From Baseline in C-reactive Protein (CRP) Values
• Maintenance Phase: Percentage of Participants With Crohn's Disease Activity Index (CDAI)-100 Response
• Maintenance Phase: Percentage of Participants With Durable Clinical Remission
• Maintenance Phase: Percentage of Participants With Corticosteroid-free Clinical Remission
• Serum Vedolizumab Concentration in Induction Phase
• Serum Vedolizumab Concentration in Maintenance Phase
• Number of Participants With Anti-vedolizumab Antibodies (AVA) in Induction Phase

9. Statistical Interpretation

10. Statistical Methods Explained

For a binary endpoint, the basic descriptive result is the proportion of participants meeting the event definition in each group. Here, the event is meeting the CDAI-100 response definition at the stated assessment time. The posted percentages and their confidence intervals describe estimated response frequencies and uncertainty around those estimates. They do not describe continuous CDAI change, and they do not by themselves adjust for other variables.

What does a risk difference express?

A risk difference is the event proportion in one group minus the event proportion in the comparator group, expressed on the probability scale. A value of zero corresponds to equal proportions. Positive values indicate a larger observed proportion in the treatment group when the comparison is defined in that direction; negative values indicate a smaller observed proportion. The primary endpoint's unadjusted recalculation is reported as 0.099, with a 95% confidence interval from -0.029 to 0.227. The interval crossing zero indicates that the posted data remain compatible with effects on either side of no difference under this calculation.

A risk difference is often straightforward to interpret, but it is sensitive to the event frequency in the groups and can be uncertain when the groups are limited in size. Its confidence interval is therefore central to interpretation. A point estimate alone does not indicate how precisely the underlying difference has been estimated.

What do the odds ratio and the reported analysis add?

An odds ratio compares the odds of an event between groups. An odds ratio of 1 corresponds to equal odds; values above or below 1 describe higher or lower odds in the numerator group, respectively. The sponsor reports a Cochran-Mantel-Haenszel odds ratio for the primary comparison. This method combines information across strata, but the registry record does not report the stratification details. The recalculated odds ratio is a separate unadjusted calculation from the posted counts. Odds ratios and risk differences use different scales, so their numerical values are not directly interchangeable.

The sponsor's reported analysis and the recalculation also use different methods. The sponsor reports an odds ratio of 1.80, with a 95% confidence interval from 0.816 to 3.958 and p = 0.145. The unadjusted recalculation reports a risk difference and an unadjusted p-value. Differences in method and in the scale being estimated mean the analyses should be read as complementary descriptions, not as two copies of the same calculation. The registry discrepancy flag is false.

How should a p-value be read?

A p-value describes how unusual data at least as discrepant from a null hypothesis would be under the assumptions of the specified statistical test, if that null hypothesis were true. It is not the probability that the null hypothesis is true, and it does not measure the probability that the observed result occurred by chance in an everyday sense. Nor does a p-value give the size or clinical importance of an effect. The sponsor's p-value and the recalculated unadjusted p-value belong to their respective analyses and should not be conflated.

A confidence interval provides an estimated range of values compatible with the data and the method used, with a stated coverage level. In repeated use under the method's assumptions, intervals constructed in this way have the stated coverage property. For an individual trial result, it is more useful to read the interval as a display of estimation uncertainty than as a probability statement about the particular interval. Intervals that include the null value indicate that the data do not rule out that value at the corresponding confidence level under the analysis.

Why can over-dispersion matter?

General statistical explanation: in a model for count or binary outcomes, over-dispersion means the observed variability is greater than the model assumes. If it is ignored, standard errors can be too small and intervals too narrow, making evidence look more precise than it is. The registry record does not report a model that identifies or adjusts for over-dispersion for this primary binary endpoint, so it is not possible to say that it was present or how it was handled. The concept is relevant as a general caution, not as a claim about this trial.

What is different about the maintenance comparisons?

The maintenance groups are not simply the original randomized induction groups carried forward unchanged. The record says the maintenance randomization applied to participants who had received vedolizumab in induction and achieved CDAI-70 response at Week 10. Participants in the placebo-continuation group had received placebo in induction and achieved that response but were not randomized in maintenance. The Week 60 maintenance endpoint is therefore about the listed maintenance groups and their selected participants. It should not be interpreted as a randomized comparison of all participants enrolled at the start of induction.

For the maintenance CDAI-100 comparison, the posted groups each contain 12 participants and the reported intervals are wide. This illustrates a general feature of binary outcomes: with limited observations, a small change in the event count can appreciably alter the estimated percentage and the uncertainty. The reported maintenance odds ratio and risk difference are distinct effect measures. Neither should be generalized beyond the population and comparison actually reported.

What is not described in the record?

The trial's full statistical analysis plan, original design assumptions, handling of missing endpoint data, and multiplicity strategy are not reported in the registry record. Those omissions limit assessment of whether analyses followed a prespecified hierarchy, how participants without an observed Week 10 measurement were treated, or whether the reported inferential results account for the number of endpoints examined. No method should be inferred from the endpoint labels alone.

11. Confidence Intervals and Precision

The primary endpoint estimates have appreciable uncertainty. The posted 95% confidence interval for the vedolizumab response percentage extends from 17.27 to 37.72, while the placebo interval extends from 9.184 to 26.81. These ranges communicate that the estimated percentages are not precise point values. Confidence intervals are not boundaries on every possible true value, but they indicate the degree of uncertainty under the method used to construct them.

The unadjusted risk-difference interval runs from -0.029 to 0.227. It includes zero and also includes positive values. Thus the estimate is consistent with no difference as well as with a range of positive differences under the calculation. The interval is more informative than describing the point estimate alone, because it shows why the primary comparison is not conclusive despite the higher observed percentage in the vedolizumab group.

For a binary endpoint, precision depends on the event counts and group sizes, as well as on the statistical method. More information generally narrows uncertainty, but a relatively uncommon event or a selected population can leave wide intervals. The registry record reports 21 primary events among 79 participants in the vedolizumab group and 13 among 78 in the placebo group through the posted summary counts. The method-based sponsor odds-ratio interval, from 0.816 to 3.958, is also broad and includes 1.

Maintenance precision is particularly limited in the posted comparisons. The Week 60 CDAI-100 result has 12 participants in each group and a broad confidence interval for the odds ratio, from 1.473 to 160.972. The durable clinical-remission comparison also has broad group intervals. These features call for caution when interpreting the magnitude of maintenance estimates. A wide interval does not prove the absence of an effect; it means the estimate leaves substantial uncertainty about its size and direction.

12. Planning a Similar Trial: Sample Size

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

The planning figures are illustrations based on observed effects, not the trial's original design assumptions. For the binary comparisons, the figures are participants per arm. The stated 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 trial's own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the illustrative participants per arm associated with an 80% chance of detecting the specified effect under those planning assumptions, if that effect and the assumptions were correct. The 90% power column expresses the corresponding planning target at 90% power. These are design calculations, not guarantees that a future study will produce a particular result. The displayed scenarios include planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. Smaller effects need more participants to detect with the same power under the stated assumptions.

The calculations use the observed effect as their reference, which is uncertain in its own right. A planning result based on an observed estimate can therefore be sensitive to the estimate's imprecision and should not be read as a definitive requirement for another trial. The figures do not account for aspects not included in the stated assumptions, such as a different analysis model or missing data. The record's sample-size assumptions specify equal allocation; they do not establish that those were the assumptions used to design this completed trial.

This is not observed power. Observed power is not a useful substitute for reading the effect estimate and its confidence interval: after results are observed, the interval and the uncertainty in the estimate provide the relevant information about precision. The planning figures instead illustrate how hypothetical study size changes across effect-size assumptions and target power levels.

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 = 0.167; treatment proportion p₁ = 0.266; average p̄ = 0.216.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5822 = 1.1411.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5780 = 0.4864.
4. n = (1.1411 + 0.4864)² / (0.0992)² = 269.4, rounded up to 270 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 <- 0.166666666666667                          # control proportion
p1 <- 0.265822784810127                          # 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(2038920)
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 = 270 (80% power), 361 (90% power)
 75% of observed effect: n per arm = 460 (80% power), 615 (90% power)
 50% of observed effect: n per arm = 988 (80% power), 1,322 (90% power)
power.prop.test: 269.4 
Simulated power at n = 270 per arm: 0.805 (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 the sponsor's primary comparison as a Cochran-Mantel-Haenszel analysis with an odds ratio of 1.80, a 95% confidence interval from 0.816 to 3.958, and p = 0.145. The simple recalculation from the posted summary counts gives an unadjusted risk difference of 0.099, with a 95% confidence interval from -0.029 to 0.227 and an unadjusted p-value of 0.132. These are different effect measures and methods, so their estimates and p-values need not match exactly.

The sponsor result is the trial's own reported analysis. The recalculation is an unadjusted check based on the posted group summaries; it does not replace the sponsor analysis or reproduce any stratification used by the Cochran-Mantel-Haenszel method. The sponsor odds-ratio interval includes 1, and the unadjusted risk-difference interval includes zero. Both analyses therefore show uncertainty around the primary comparison rather than a precise estimate that excludes the null value on their respective scales.

The registry discrepancy flag is false, so the record does not identify a discrepancy between the sponsor result and the simple recalculation. The sponsor's analysis should be distinguished from the recalculated estimate because it is the reported trial analysis, while the unadjusted calculation is supplementary context.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, time points, groups or statistical hypotheses are examined. Even if each individual test is conducted at its stated significance level, considering many tests can increase the chance of at least one apparently notable result when there is no underlying effect. Common approaches include prioritizing a primary endpoint, using a prespecified testing sequence, or adjusting inference across a family of tests. These are general methods; the registry record does not report which, if any, were used for this trial's secondary outcomes.

The registry identifies a primary endpoint and lists secondary outcomes, but the full statistical analysis plan and the multiplicity strategy are not reported in the registry record. It is therefore not possible to determine from the posted record whether secondary testing was confirmatory, exploratory, adjusted, or controlled through a hierarchy. Unadjusted secondary p-values should be treated cautiously in that setting, especially when several outcomes are considered together.

Missing-data handling is also not reported in the registry record. The primary population is defined as randomized participants who received at least one induction dose, but that definition alone does not explain what happened if a participant lacked a Week 10 assessment. The record does not state whether missing outcomes were imputed, treated as nonresponse, excluded from a particular analysis, or handled in another way. No one approach should be assumed.

The distinction between analysis populations matters across phases. The primary endpoint uses the induction full analysis set as described in the registry. Maintenance outcomes apply to the response-selected participants listed for maintenance, including the groups randomized after vedolizumab induction and the placebo-continuation group. The registry does not report a complete statistical analysis plan that would clarify all population rules or the handling of participants moving between phases.

These gaps do not invalidate the posted descriptive results, but they constrain how confidently inferential findings can be interpreted. A complete statistical plan would clarify endpoint hierarchy, analysis populations, missing-data rules and any adjustment for repeated testing. Without those details, the primary reported analysis remains the principal trial result, while secondary findings are best read as supportive or exploratory descriptions rather than definitive tests.

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: crohn's disease (26.1%), subileus (8.3%), abdominal adhesions (8.3%), erythema nodosum (8.3%), cellulitis (5.9%).

The registry reports serious adverse event counts by treatment group and phase. During induction, 8 participants in the vedolizumab group and 10 in the placebo group had a serious adverse event; the corresponding groups at risk were 79 and 78. The record reports zero deaths in both induction groups. In maintenance, serious adverse events were reported for 2 participants in the vedolizumab group, 4 in the placebo group, and 2 in the placebo-continuation group; zero deaths were reported in each group. In the open-label vedolizumab group, 70 participants had a serious adverse event and zero deaths were reported.

These counts should be read with their group sizes and phase context. The maintenance groups are smaller and selected after induction, while the open-label group has a different treatment context. They are not interchangeable comparisons, and the counts alone do not establish a causal difference in risk between treatments. The registry also reports other adverse event counts separately from serious events; those are not the same category.

The record lists serious-event terms with percentages, including Crohn's disease, subileus, abdominal adhesions, erythema nodosum, cellulitis and intestinal obstruction. The denominator and detailed phase context for those term percentages are not reported in the registry record. The adverse-event summaries are descriptive and do not, by themselves, establish attribution to study treatment or provide a complete assessment of safety.

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

The primary result describes the proportion of participants in the induction full analysis set who met the registry's CDAI-100 response definition at Week 10. The observed percentage was higher in the vedolizumab group than in the placebo group, but the unadjusted confidence interval for the risk difference includes zero, and the sponsor's reported odds-ratio interval includes 1. The estimate is therefore uncertain. The data should not be reduced either to a claim that the groups are definitively different or to a claim that they are identical.

The endpoint is specific: it is a decrease from baseline of at least 100 CDAI points. It does not establish how much participants improved on average, whether they reached clinical remission, or how their symptoms changed beyond the defined score. The registry lists remission as a secondary outcome but does not report its definition. Nor does the primary outcome answer questions about long-term outcomes for all participants.

Maintenance findings address a different population and period. Participants eligible for the randomized maintenance comparison had received vedolizumab in induction and achieved CDAI-70 response at Week 10. The maintenance estimates therefore cannot be interpreted as if they were results from an unchanged randomization of everyone enrolled at the start. The placebo-continuation group is also described as not randomized in maintenance.

The trial record concerns Japanese participants with moderate or severe Crohn's disease in this study setting. The results alone do not settle what would occur in other populations, under other treatment schedules, or for outcomes not measured in the posted record. The registry does not report the full plan for missing data or multiplicity, which limits interpretation of all analyses considered together.

These are statistical descriptions, not individual predictions or treatment recommendations. A group percentage does not determine an individual's outcome, and a confidence interval represents uncertainty in an estimated group effect rather than a range of outcomes for each participant. Clinical decisions require considerations beyond the posted trial statistics.

18. Why This Trial Matters Statistically

This trial is statistically informative because it places a randomized placebo comparison around a defined response endpoint and reports both group estimates and inferential analyses. The Week 10 CDAI-100 outcome gives a clear binary event definition, while the sponsor analysis and the unadjusted recalculation illustrate how a result can be expressed on different scales and through different methods.

The primary comparison also demonstrates why point estimates need uncertainty intervals. The observed response percentage is higher in the vedolizumab group, yet the risk-difference interval includes zero and the sponsor odds-ratio interval includes 1. A directional difference in the observed data is not the same as a precise estimate that excludes the null. Reading the result this way avoids both overstatement and the opposite error of treating a non-significant result as proof of equivalence.

The study's induction and maintenance structure highlights an important design distinction. Maintenance analyses are based on participants selected by a response criterion after induction, and the record describes a separate placebo-continuation group. Those results answer narrower questions than the primary randomized induction comparison. Recognizing that population shift is essential to avoid treating the phases as one continuous, directly comparable experiment.

Finally, the registry record shows the value of knowing what is not reported. Missing-data rules, multiplicity handling and original sample-size assumptions affect how the inferential record can be assessed. Their absence does not erase the posted results, but it places limits on claims about robustness and confirmatory status. For clinical researchers, this is a practical example of reading design, population, endpoint, estimate and uncertainty together rather than relying on a single p-value.

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 primary comparison is based on the induction full analysis set, not necessarily every participant enrolled. The registry defines that set as randomized participants who received at least one dose in induction. The record does not provide enough detail to determine how all deviations from follow-up or treatment were handled for the Week 10 outcome.

The analysis is limited by uncertainty in the primary estimate. The unadjusted interval for the risk difference includes zero, and the sponsor's reported odds-ratio interval includes 1. The group confidence intervals also span a range of possible response percentages. The posted data therefore do not establish a narrowly estimated treatment difference.

Maintenance analyses involve response-selected groups rather than the original induction population. The registry describes maintenance randomization among participants who received vedolizumab in induction and achieved CDAI-70 response at Week 10, and a nonrandomized placebo-continuation group for placebo induction responders. Small listed maintenance groups and broad intervals limit precision. The phase-specific outcomes should not be generalized to all induction participants.

The registry record does not report the clinical-remission definition, original sample-size justification, detailed statistical analysis plan, missing-data handling or multiplicity strategy. It also does not state the stratification details for the sponsor's Cochran-Mantel-Haenszel analysis. These omissions restrict independent assessment of the full inferential plan and the relation among secondary tests.

The trial record describes Japanese participants with moderate or severe Crohn's disease and a specified intravenous treatment schedule. The posted result does not itself establish applicability outside that population and setting. Safety counts are reported across induction, selected maintenance groups and open-label treatment, which have different contexts; they should not be combined or interpreted as a single randomized comparison.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Phase III Study of MLN0002 (300 mg) in Treatment of Crohn's Disease — NCT02038920.
• Registry record last updated 2019-12-09; status last verified 2019-11.
• Results first posted 2019-12-09.
• 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

The registry describes a completed phase 3, multicenter, randomized, double-blinded, placebo-controlled, parallel-group study sponsored by Takeda, with 157 participants enrolled. It evaluated intravenous vedolizumab (MLN0002) in induction and maintenance therapy for Japanese participants with moderately or severely active Crohn's disease.

The primary endpoint was CDAI-100 response at Week 10 in the induction full analysis set. The posted response percentages were 26.6% with vedolizumab and 16.7% with placebo. The sponsor's reported odds-ratio analysis and the simple unadjusted recalculation both leave uncertainty around the comparison. Maintenance outcomes concern selected participants who met the Week 10 CDAI-70 response criterion, and should be read separately from the induction result.

The record includes secondary outcomes and adverse-event summaries, but does not report the full statistical analysis plan, missing-data handling, multiplicity strategy or original sample-size justification. These limitations are important when interpreting the full set of registry findings.