Efficacy and Safety of TD-1473 in Crohn's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of TD-1473 in Crohn's Disease (NCT03635112), a randomized phase 2 study. The primary outcome is Change From Baseline in Crohn's Disease Activity Index (CDAI) Score.

2. Clinical Question

The trial was intended to evaluate efficacy, safety, and tolerability of induction therapy with two doses. Its primary comparison is between each dose group and placebo on change in CDAI score from baseline to Week 12. The registry reports least-squares mean changes and sponsor comparisons based on a mixed model repeated measures analysis. The endpoint analysis population was narrower than everyone enrolled: participants needed to meet the registry's stated evaluability and postbaseline-score requirements.

The clinical question is therefore about the measured average change in disease activity during induction, not about every participant's individual course or longer-term outcomes. The study description includes continued treatment after induction for participants completing that period, but the primary endpoint timeframe is Baseline to Week 12. The registry record does not report a reason for the trial's terminated status.

3. Trial Design

This was a phase 2, randomized, quadruple-masked, parallel-group study of TD-1473 in Crohn's disease. The registry lists an enrollment of 167 and identifies Theravance Biopharma as the sponsor. The study started on 2018-11-19, had a primary completion date of 2021-12-30, and is listed as terminated. A reason for termination is not reported in the registry record.

Participants were assigned to placebo or one of two TD-1473 doses. During induction, each group took its assigned oral treatment daily for 12 weeks. The arm descriptions state that participants completing induction in either active-treatment group could continue their assigned dose for up to 48 additional weeks. Those completing induction in the placebo group were described as receiving TD-1473 at Dose A during the additional treatment period. These continuation details do not change the primary endpoint's stated Baseline to Week 12 timeframe.

Randomization supports a comparison of assigned groups, while masking is intended in general to limit the influence of treatment knowledge on participant reporting, clinical assessment, and study conduct. Those are general design principles, not evidence about how well masking was maintained in this trial. The registry record does not report the randomization method, allocation ratio, concealment procedures, or details of who was included in the quadruple-masking description.

The primary endpoint is reported for a modified intent-to-treat analysis set with specified evaluability and postbaseline-data requirements. Its group sizes differ from the enrollment figure, and the secondary outcomes use their own reported denominators. These are endpoint-specific analysis populations; the registry record does not provide enough information to reconcile all differences among the enrollment, safety, and endpoint counts.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified Intent-to-Treat Analysis Set: Comprised all randomized evaluable participants who received at least 1 dose of study drug, had at least one postbaseline CDAI score and had non-missing values at both baseline and postbaseline visit.

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: Intent-to-Treat Analysis Set: Comprised all randomized participants who received at least one dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Crohn's Disease Activity Index (CDAI) Score, measured over Baseline to Week 12.

The primary endpoint is change from baseline in the Crohn's Disease Activity Index score from Baseline to Week 12. CDAI combines information from eight predictors of disease activity: abdominal pain severity, general well-being, very soft or liquid stool frequency, extra-intestinal symptoms, use of antidiarrheal drugs, an abdominal mass, body weight, and hematocrit. Participants recorded symptom information in a diary. The abdominal pain, general well-being, and stool-frequency subscores were summed over the 7 days before each visit; the other predictors were also recorded and weighted to produce the total score.

The total CDAI score ranges from 0 to 600, with higher scores indicating a worse outcome. The registry gives the following activity benchmarks: scores <150 indicate clinical remission; 150 to 219 indicate mildly active disease; 220-450 indicate moderately active disease; and scores >450 indicate very severe disease. The endpoint measures change in a continuous score rather than simply whether an individual crosses one of these categories.

Change from baseline is used to express how the measured score shifts during the treatment period while retaining the scale's continuous information. Least-squares means are model-based group summaries; the endpoint description identifies the sponsor's method as a mixed model repeated measures analysis for the primary comparisons. The registry record does not report the specific covariates or model details beyond that method label. A change in the group average should not be read as the outcome of every participant, and the score alone does not describe every aspect of Crohn's disease or treatment experience.

7. Results Posted in the Registry

• TD-1473 80 mg: mean -105.62 versus -104.86 on control; difference -0.76 (95% CI -40.04 to 38.52), standardized effect d = -0.01; z-test p = 0.970.
• TD-1473 200 mg: mean -117.99 versus -104.86 on control; difference -13.13 (95% CI -52.06 to 25.80), standardized effect d = -0.15; z-test p = 0.509.

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.

Number of Participants Who Demonstrated a Clinical Response as Measured by CDAI

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • TD-1473 80 mg: 51.9% (28/54) versus 54.3% (19/35) on control; risk difference -2.4 percentage points (95% CI -23.6 to 18.8), risk ratio 0.96 (0.64–1.42); unadjusted two-proportion z-test p = 0.822.
  • TD-1473 200 mg: 60.7% (34/56) versus 54.3% (19/35) on control; risk difference 6.4 percentage points (95% CI -14.5 to 27.3), risk ratio 1.12 (0.77–1.62); unadjusted two-proportion z-test p = 0.545.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between TD-1473 80 mg 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 TD-1473 200 mg and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants Who Demonstrated CDAI Clinical Remission

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • TD-1473 80 mg: 24.1% (13/54) versus 37.1% (13/35) on control; risk difference -13.1 percentage points (95% CI -32.7 to 6.6), risk ratio 0.65 (0.34–1.23); unadjusted two-proportion z-test p = 0.185.
  • TD-1473 200 mg: 39.3% (22/56) versus 37.1% (13/35) on control; risk difference 2.1 percentage points (95% CI -18.3 to 22.6), risk ratio 1.06 (0.62–1.82); unadjusted two-proportion z-test p = 0.838.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between TD-1473 80 mg 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 TD-1473 200 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Simple Endoscopic Score for Crohn's Disease (SES-CD) at Week 12

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • TD-1473 80 mg: mean -0.20 versus -1.90 on control; difference 1.70 (95% CI -1.42 to 4.82), standardized effect d = 0.30; z-test p = 0.286.
  • TD-1473 200 mg: mean -1.90 versus -1.90 on control; difference 0.00 (95% CI -3.11 to 3.11), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between TD-1473 80 mg 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 TD-1473 200 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several aspects of disease activity: CDAI clinical response, CDAI clinical remission, endoscopic change and response, and stool frequency and abdominal pain remission. Together, they provide a broader set of measurements than the primary CDAI change endpoint, but they do not all measure the same construct. Symptom-based criteria and endoscopic measures can therefore show different patterns without being direct substitutes for one another.

Across the reported secondary comparisons, the pattern is mixed rather than a uniform separation of both TD-1473 groups from placebo across all measures. The point estimates and confidence intervals vary by outcome and dose. The registry lists endoscopic response and SFAP clinical remission among the secondary outcomes, but their numerical comparisons are not reported in the registry record summarized in these results. Interpretation should also account for the different denominators used for these outcomes.

The secondary findings do not overturn or independently settle the primary result. The primary CDAI comparison remains the prespecified central efficacy measure identified in the record, while secondary outcomes can offer context about related domains. A pattern that is not consistent across outcomes is a reason to avoid selecting only favorable comparisons as if they were conclusive.

Unadjusted secondary p-values do not account for the fact that several outcomes and dose comparisons are examined. In general, examining many comparisons increases the chance that some may appear notable by chance, even when there is no underlying effect. The registry record does not report a multiplicity adjustment strategy. Accordingly, the secondary analyses carry less evidential weight than a single prespecified primary analysis and should be treated as supportive or exploratory, not as independent confirmation.

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

• Number of Participants Who Demonstrated a Clinical Response as Measured by CDAI
• Number of Participants Who Demonstrated CDAI Clinical Remission
• Change From Baseline in Simple Endoscopic Score for Crohn's Disease (SES-CD) at Week 12
• Number of Participants With Endoscopic Response at Week 12
• Number of Participants With Stool Frequency and Abdominal Pain (SFAP) Clinical Remission

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is a continuous outcome: each participant's CDAI score can take values across a numeric scale, and the analysis summarizes change rather than dividing the endpoint into event and non-event categories. The reported arm summaries are least-squares means with standard errors. A standard error describes the uncertainty of an estimated mean; it is not the spread of individual participant scores. The registry reports sponsor comparisons using a mixed model repeated measures analysis and labels the contrast as a least-squares mean difference.

What does a least-squares mean difference represent?

In general, a least-squares mean is a model-based estimate of a group mean, and a mean difference subtracts one group's estimated mean from another's. Here, the registry reports the placebo comparison for each TD-1473 dose. The sign of the contrast depends on the order of subtraction used in the analysis; the posted results give the estimates directly. Since lower CDAI scores are more favorable on the stated scale, a negative change from baseline describes a decrease, but the contrast and its uncertainty are the basis for comparing groups.

A mixed model repeated measures analysis is generally used to analyze repeated measurements while representing correlation among measurements from the same participant. It can use available repeated outcome observations under assumptions about the model and missingness. The registry record does not report the detailed model specification, covariance structure, covariates, or missing-data assumptions. Those details should not be inferred from the method name alone.

Why consider a confidence interval as well as a p-value?

In general, a confidence interval expresses a range of values compatible with the estimate under the analysis and its assumptions. Its width conveys precision, and whether it includes zero is relevant for a difference where zero represents no difference. A p-value measures how unusual data at least as incompatible with the null hypothesis would be under that hypothesis and the model. It does not give the probability that the null hypothesis is true, and it does not measure clinical importance. For the primary contrasts, both reported 95% confidence intervals include zero, and the reported p-values are 0.970 and 0.510.

The recalculated continuous comparisons are simple unadjusted estimates based on posted group summaries. The pooled standard deviations used in those calculations are approximate because the arm-level dispersion is reported as a standard error, not as a standard deviation. The calculation reconstructs an approximate standard deviation using the reported standard errors and group sizes. It does not reproduce the sponsor's repeated-measures model, its treatment of repeated observations, or any model adjustments. The recalculated results should therefore be read as a check using summary information, not as a replacement analysis.

How are the binary outcomes different?

Clinical response, clinical remission, and other participant-count outcomes are binary in their underlying form: each participant either meets the stated criterion or does not. The registry reports counts and analysis denominators, while the recalculated comparisons express differences in proportions. A risk difference is the estimated proportion in one group minus the proportion in the comparator group, using the calculation's stated ordering. Its confidence interval represents uncertainty in that difference. The unadjusted recalculation does not necessarily match a sponsor analysis that uses a stratified or otherwise adjusted method.

For example, the registry identifies Cochran-Mantel-Haenszel chi-square tests for the sponsor's clinical response and CDAI remission comparisons, and ANCOVA for the sponsor's SES-CD comparisons. These method labels indicate that the sponsor's tests are not identical to a simple comparison reconstructed from unadjusted group summaries. No sponsor comparison method or numerical comparison for each listed secondary outcome is detailed in the record summarized in these results. No secondary multiplicity adjustment is reported in the registry record.

Why can over-dispersion matter?

In general, over-dispersion occurs when the variability in count data is greater than a statistical model expects. If a model understates variability, standard errors can be too small and confidence intervals too narrow. This is a general consideration for count outcomes and is not a claim that over-dispersion occurred in this trial. The primary outcome is analyzed as a continuous CDAI change, and the registry does not report an over-dispersion assessment for the secondary outcomes.

For the primary and secondary recalculations, reported p-values are unadjusted. They do not account for the number of outcomes or comparisons considered. The sponsor's reported analyses should be distinguished from these simple recalculations, particularly where methods differ. Details of a broader statistical analysis plan, multiplicity strategy, interim analyses, and the trial's original sample-size justification are not reported in the registry record.

11. Confidence Intervals and Precision

Precision is conveyed most directly by the width and placement of the confidence intervals around the primary contrasts. For TD-1473 80 mg compared with placebo, the sponsor reported a least-squares mean difference of -0.76 and a 95% confidence interval from -40.62 to 39.11. For TD-1473 200 mg compared with placebo, the reported difference was -13.13 and the 95% confidence interval ranged from -52.46 to 26.19. Each interval includes zero, so the posted estimates do not rule out no difference under the reported analysis.

The intervals also include values on both sides of zero. That means the results are compatible with a difference favoring either group, as well as with a small difference. Their width limits how precisely the treatment contrasts can be estimated from these results. The point estimate is only one value within a broader range of uncertainty; it should not be interpreted on its own.

The approximate recalculations produce intervals of -40.04 to 38.52 for 80 mg and -52.06 to 25.8 for 200 mg. Their proximity to the sponsor's intervals reflects similar point estimates, but the recalculations rely on approximate dispersion and do not reproduce the repeated-measures model. For formal interpretation, the sponsor's model-based intervals are the relevant registry analyses. The discrepancy flag is false.

In general, a confidence interval is not a guarantee that the true effect lies in the interval, and its interpretation relies on the analysis and its assumptions. It is also not a range of individual participant changes. The registry record does not report a prespecified clinically important difference for the CDAI contrast, so these intervals alone cannot establish whether any particular value within them would be clinically important.

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 illustrations are not the trial's original design calculation. For these continuous and binary comparisons, the displayed figures are participants per arm, not total participants. The registry's 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 set as the stated fraction of the observed effect. The sample-size illustrations are not observed power.

The 80% power column describes the number of participants per arm that a hypothetical study would require, 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 forward-looking design calculations, not a measure of how likely this completed trial was to detect an effect after observing its results.

For the primary outcome, the illustration based on 100% of the observed effect gives 210,103 participants per arm for 80% power for the 80 mg comparison, and 711 participants per arm for the 200 mg comparison. These values differ because the observed contrasts differ in size, and the 80 mg observed estimate is close to zero. They are not recommendations for a future study and do not establish that the observed effects are stable or reproducible.

The rows for 75% and 50% of the observed effect represent progressively smaller effects to detect. Smaller effects require more participants under the same planning assumptions, so their corresponding planning figures are larger. The sample-size illustrations do not include information about the trial's own original design rationale, which is not reported in the registry record. They also do not incorporate any design features beyond the listed assumptions.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.760; pooled standard deviation σ = 87.925.
2. n = 2 × (1.96 + 0.842)² × 87.925² / -0.760² = 210,102.2, rounded up to 210,103 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
delta <- -0.760000000000005                      # observed difference in means
sigma <- 87.9245240630732                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3635112)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 210,103 (80% power), 281,268 (90% power)
 75% of observed effect: n per arm = 373,515 (80% power), 500,031 (90% power)
 50% of observed effect: n per arm = 840,409 (80% power), 1,125,069 (90% power)
power.t.test: 210103.1 
Simulated power at n = 210,103 per arm: 0.804 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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 used a mixed model repeated measures analysis and reported least-squares mean differences for each dose compared with placebo. The reported results were -0.76 with a 95% confidence interval of -40.62 to 39.11 and p = 0.970 for TD-1473 80 mg, and -13.13 with a 95% confidence interval of -52.46 to 26.19 and p = 0.510 for TD-1473 200 mg.

The simple recalculations based on the posted summary statistics produce the same point estimates to the reported precision, with approximate intervals of -40.04 to 38.52 and -52.06 to 25.8, and unadjusted p-values of 0.97 and 0.509. Small differences in intervals and p-values are expected because the recalculation is not the sponsor's repeated-measures model and uses approximate dispersion reconstructed from arm-level standard errors. It does not reproduce all features of the sponsor analysis.

The registry marks no discrepancy between the sponsor results and the recalculations. The sponsor's analysis is the formal registry result; the unadjusted recalculation is a transparent comparison based on the available summaries, not a competing analysis. Where methods differ, estimates and uncertainty should be interpreted in light of the method used rather than treating the numerical outputs as interchangeable.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity concerns arise when several hypotheses are tested, such as comparisons across multiple doses and outcomes. Without an adjustment strategy or a clearly defined testing hierarchy, unadjusted p-values do not control the overall chance of a false-positive conclusion across the full set of tests. The registry reports several secondary outcomes and comparisons, but a multiplicity procedure or hierarchy is not reported in the registry record. The secondary p-values should therefore be read as unadjusted results rather than as multiplicity-controlled evidence.

The registry identifies the primary analysis population as a modified intent-to-treat analysis set. Participants had to be randomized and evaluable, receive at least 1 dose, have at least one postbaseline CDAI score, and have non-missing values at baseline and a postbaseline visit. This definition gives readers important information about who contributed to the primary endpoint analysis, but it is not a full account of how missing observations were handled within the repeated-measures model.

Specific missing-data assumptions, imputation methods, sensitivity analyses, and reasons for missing measurements are not reported in the registry record. The record also does not report how participants with incomplete follow-up were handled in each secondary endpoint analysis. Because the endpoint denominators vary, results should be interpreted with attention to the population contributing to each outcome rather than assuming a single analysis set was used everywhere.

The registry's primary analysis method is named as mixed model repeated measures, but the detailed model specification is not reported. The randomization procedure, allocation concealment details, interim analyses, and the original statistical analysis plan are not reported in the registry record. The original sample-size justification and its design assumptions are also not reported. The planning illustrations based on observed effects are separate calculations and do not fill those gaps.

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 (12.7%), intussusception (3.0%), drug hypersensitivity (3.0%), pancreatitis acute (2.6%), covid-19 pneumonia (2.6%).

The registry reports serious adverse events by safety group, with the group described as at risk and the number affected. The figures are: placebo, 3 of 38; TD-1473 80 mg, 9 of 58; TD-1473 80mg Post-Placebo, 3 of 33; and TD-1473 200 mg, 10 of 63. Deaths are reported as 1 in the placebo group and 0 in each of the listed TD-1473 groups. The post-placebo group is a separate registry category and should not be merged with the induction 80 mg group when reading these entries.

These counts describe the registry's safety reporting; they are not adjusted comparisons and do not by themselves establish causation. The groups differ in their reported at-risk denominators, and the record does not provide a randomized comparative analysis of serious adverse events in these results. The most frequently listed serious-event terms include Crohn's Disease at 12.7%, Intussusception at 3.0%, Drug hypersensitivity at 3.0%, Pancreatitis acute at 2.6%, COVID-19 pneumonia at 2.6%, and Granuloma at 2.6%. The registry record does not specify the denominator for those term percentages in this summary.

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

The primary results describe average CDAI changes during induction and the estimated differences between each TD-1473 dose and placebo at Week 12. The sponsor's point estimates are close to zero for the 80 mg comparison and more negative for the 200 mg comparison, but both 95% confidence intervals include zero and span values on both sides of it. The data therefore do not provide a precise estimate that clearly separates either dose from placebo on this endpoint.

This does not establish that the treatment groups are identical. A result with a confidence interval including zero can reflect uncertainty as well as a small underlying difference, and the intervals allow effects in either direction. Conversely, the more negative point estimate for the 200 mg group is not, by itself, proof of a dependable treatment difference. The registry does not report a clinically important threshold for interpreting the size of the CDAI contrast.

The endpoint is a score based on symptoms and other clinical measures, not a complete summary of every potential benefit or harm. The primary timeframe is Baseline to Week 12. Although the protocol description includes longer continuation treatment for participants completing induction, the primary comparison does not establish longer-term outcomes.

The secondary outcomes provide additional but mixed information and have unadjusted comparisons. The safety counts describe reported serious adverse events and deaths but do not establish whether treatment caused an event. The study is listed as terminated, and the reason is not reported in the registry record. Taken together, the posted data describe uncertain Week 12 comparisons in this phase 2 study; they do not by themselves establish a definitive overall treatment conclusion or replace a full assessment of the trial record.

18. Why This Trial Matters Statistically

This trial illustrates why a group mean change should not be interpreted without its comparison and uncertainty. The reported CDAI changes in the active groups and placebo group are all decreases, but the decision-relevant statistical question is how the change differs between groups. The sponsor's estimates and confidence intervals show that the primary contrasts are uncertain and compatible with values on either side of zero.

It also illustrates the distinction between a model-based sponsor analysis and a simple recalculation from summary statistics. The recalculated estimates are close to the sponsor's estimates, but they do not reproduce the mixed model repeated measures analysis. Recognizing this distinction helps prevent a transparent arithmetic check from being mistaken for a second formal trial analysis.

The endpoint populations and safety denominators are not identical, and secondary analyses use outcome-specific denominators. This is a practical reminder that enrollment is not necessarily the analysis denominator for every endpoint. Reading each estimate alongside its population definition and denominator is essential for understanding what data contribute to it.

Finally, the planning illustrations emphasize how strongly sample-size requirements depend on the assumed effect. Calculations based on a small observed effect can produce very large participant targets, but those calculations are not the study's original design rationale and are not observed power. The registry record does not report its original sample-size assumptions, multiplicity strategy, or detailed missing-data plan. Those absences limit how fully the design and analysis can be evaluated from the registry record alone.

19. Limitations

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

The primary comparison is reported for a modified intent-to-treat analysis set rather than every enrolled participant. The registry provides the inclusion requirements for that set but does not report a complete participant flow or reconcile the endpoint, safety, and enrollment denominators. This limits evaluation of how exclusions or unavailable measurements may have affected each result.

The sponsor identifies a mixed model repeated measures analysis, but detailed model specifications, covariates, covariance assumptions, and missing-data assumptions are not reported in the registry record. The unadjusted continuous recalculations use approximate pooled dispersion reconstructed from reported standard errors; they are not substitutes for the sponsor analysis. The registry's discrepancy flag is false, but agreement in point estimates does not make the methods identical.

The registry does not report a multiplicity adjustment for the secondary outcomes, a prespecified clinically important difference for the CDAI contrast, or a detailed statistical analysis plan. Numerical comparisons for some listed secondary outcomes, including endoscopic response and SFAP clinical remission, are not reported in the registry record summarized in these results. Those limitations constrain interpretation of the broader efficacy pattern.

The trial is listed as terminated, but the reason is not reported in the registry record. The record also does not provide the original sample-size justification, detailed information about interim analyses, or a full explanation of the safety term percentages' denominators. The reported safety counts are descriptive and do not establish causal attribution. These limitations are specific to what is documented in the registry record and should not be replaced with assumptions about the study's conduct or analysis.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of TD-1473 in Crohn's Disease — NCT03635112.
• Registry record last updated 2023-03-13; status last verified 2023-02.
• Results first posted 2023-03-13.
• 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

NCT03635112 is a phase 2 randomized, quadruple-masked, parallel-group study of TD-1473 in moderately-to-severely active Crohn's disease, sponsored by Theravance Biopharma. The registry lists enrollment of 167, a start date of 2018-11-19, a primary completion date of 2021-12-30, and terminated status. A reason for termination is not reported in the registry record.

The primary endpoint is change from baseline in CDAI score from Baseline to Week 12. The sponsor's mixed model repeated measures comparisons with placebo have 95% confidence intervals that include zero for both TD-1473 80 mg and TD-1473 200 mg. Secondary results are mixed and unadjusted; safety counts are descriptive. The registry does not report the original sample-size justification, multiplicity strategy, or detailed missing-data assumptions.