Study of Tislelizumab in Participants With Locally Advanced or Metastatic Urothelial Bladder Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Tislelizumab in Participants With Locally Advanced or Metastatic Urothelial Bladder Cancer (NCT04004221), a na phase 2 study. The primary outcome is Objective Response Rate (ORR) as Assessed by Independent Review Committee (IRC).

2. Clinical Question

The study was designed to evaluate efficacy and safety in participants with advanced urothelial bladder cancer after platinum-containing treatment. Because everyone in the study received tislelizumab and no control group is listed, the primary question is descriptive: it concerns the observed response proportion in the specified analysis population. It is not a randomized estimate of how outcomes differ between treatments.

The registry lists additional outcomes, including duration of response, progression-free survival, disease control rate, and overall survival, as well as investigator-assessed response and other investigator-assessed outcomes. These broaden the clinical description beyond the primary endpoint, but their results are not reported in the registry record. The record also does not give enough information to assess how outcome assessments, follow-up, or missing observations affected those results.

3. Trial Design

This was a multicenter, single-arm Phase 2 study sponsored by BeiGene. The study record lists 113 participants enrolled and identifies tislelizumab as the intervention. The regimen was 200mg intravenously every 3 weeks. The trial status is completed; the recorded start date was 2017-06-16 and the primary completion date was 2019-09-16.

Eligibility described in the study summary centered on PD-L1+ locally advanced or metastatic urothelial bladder cancer that had progressed during or following a platinum-containing regimen. The primary endpoint population was the Efficacy Evaluable Analysis Set: participants who received any dose and had measurable disease at baseline according to the independent review committee using RECIST version 1.1. The registry record does not report the number in that analysis set or give further detail about exclusions from it.

Allocation is listed as not applicable, masking as none, and the model as a single group. Thus, the registry record does not describe random assignment, masking, or a separate comparator arm. The one-group design can describe outcomes observed in participants treated in the study, but it does not itself provide a concurrent estimate of what would have happened under another treatment or no study treatment.

The primary outcome was objective response rate assessed by an independent review committee. The stated timeframe ran from the first dose up to approximately 2 years and 2 months. The record also lists secondary outcomes assessed by the independent review committee or investigators, including duration of response, progression-free survival, disease control rate, and overall survival. It lists 10 outcome measures and one analysis, but does not report a detailed statistical analysis plan, the methods used to calculate the confidence interval, or the handling of missing assessments.

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: The Efficacy Evaluable Analysis Set includes all participants who received any dose of tislelizumab and had measurable disease per IRC according to RECIST version 1.1 at baseline

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: Safety Analysis Set includes all participants who received any dose of Tislelizumab

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate (ORR) as Assessed by Independent Review Committee (IRC), measured over From the date of first dose up to approximately 2 years and 2 months.

The primary endpoint, objective response rate, is the percentage of participants with a confirmed complete response or partial response. Responses were assessed by an independent review committee using RECIST version 1.1. The analysis population was defined as participants who received any dose of tislelizumab and had measurable disease at baseline according to that committee.

Response rate is a binary-outcome summary: each participant in the relevant analysis population is classified according to whether a qualifying response was observed. The percentage then describes the share meeting the response definition. A confidence interval expresses the statistical uncertainty around an estimated proportion; it helps distinguish a relatively precise estimate from one compatible with a wider range of underlying response proportions. The registry record specifies a 95% confidence interval for this endpoint, but does not report the endpoint estimate or its interval.

Using a prespecified response definition and an independent review committee provides a structured way to classify tumor response. The endpoint is also more directly interpretable as a proportion than a time-to-event measure. However, response rate alone does not describe how long responses lasted, how disease changed for participants without a response, or overall survival. The registry lists duration of response, progression-free survival, disease control rate, and overall survival as secondary outcomes, but their results are not reported in the registry record.

Because this is a single-arm study, the response proportion describes outcomes in the treated study population; by itself, it cannot establish the comparative effect of tislelizumab versus a control treatment.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a response percentage represent?

In general, a binary endpoint classifies each person according to whether a defined event occurred. For this study, the event was a confirmed complete or partial response under RECIST version 1.1, assessed by an independent review committee. The response percentage describes the proportion of the efficacy-evaluable population classified as having that event. It is not a measure of the average change in tumor size, and it does not by itself describe the duration of response or survival.

The analysis population is important because it sets the denominator for that proportion. The registry defines the Efficacy Evaluable Analysis Set as participants who received any dose of tislelizumab and had measurable disease per the independent review committee at baseline. The number in that set is not reported in the registry record. Although enrollment is listed as 113, enrollment and the defined endpoint population are not interchangeable unless the membership of the analysis set is known.

How should the confidence interval be understood?

In general, a confidence interval communicates uncertainty around an estimate. For a response percentage, the interval indicates a range of values compatible with the observed data under the interval method and its assumptions. A narrower interval is more precise than a wider one, while an interval can remain wide when the effective analysis population is limited or the observed response pattern is uncertain. The registry specifies a 95% confidence interval for the primary endpoint, but it does not report the interval estimate, the response estimate, or the calculation method. The exact precision therefore cannot be evaluated from the registry record.

A confidence interval is not a guarantee that the underlying value falls inside the reported range, and it is not a measure of clinical importance. Statistical uncertainty and the relevance of an effect are different questions. In this record, neither the interval nor the primary response estimate is reported, so neither precision nor the size of the observed response proportion can be characterized.

What does the single-group design permit?

In general, a single-arm study estimates outcomes within the group that received the study intervention. Without a concurrent comparison group, it cannot separate the effect of treatment from other possible influences on observed outcomes, such as participant characteristics, disease course, or assessment and follow-up practices. This is why an observed response proportion in a single group is descriptive rather than a randomized estimate of comparative benefit.

The registry lists allocation as not applicable, masking as none, and the model as single group. No control arm is listed. These design features mean that the response estimate, if reported, would characterize the study's efficacy-evaluable participants and would not directly quantify a difference between tislelizumab and another treatment.

Why do the outcome measures answer different questions?

The primary endpoint is objective response rate. The listed secondary outcomes include duration of response, progression-free survival, disease control rate, and overall survival, along with investigator assessments of response and related outcomes. In general, response rate summarizes whether a qualifying response occurred; duration of response concerns how long a response lasts; progression-free survival concerns time until progression or death under its stated definition; and overall survival concerns time until death. These outcomes are not interchangeable, and one cannot be substituted for another when interpreting the course of disease.

The registry provides the names of these secondary outcomes but does not report their results or detailed definitions. It also does not report how discordant assessments between the independent review committee and investigators were handled. The primary endpoint's committee-based RECIST definition is specified, but further assessment procedures are not reported in the registry record.

What can planning illustrations tell readers?

In general, a sample-size planning illustration asks how many participants might be needed under a stated set of assumptions to detect a chosen effect with a specified power and significance level. Such an illustration is prospective and conditional: it depends on the effect size and other assumptions selected for planning. It is not the number of participants observed in a trial, and it is not observed power.

The planning assumptions associated with this record describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They also define the target effect as a stated fraction of the observed effect. These are assumptions for illustrations, not evidence that the trial itself used a two-arm randomized design. The actual study is recorded as a single-group study. The registry does not report a sample-size justification or the trial's own statistical planning details.

When an illustration uses an observed effect as its reference, it should be read as a conditional planning exercise rather than independent confirmation of that effect. The registry record does not report the primary response estimate needed to interpret the observed effect reference in clinical terms. No planning illustration can replace the missing endpoint result or create a comparative treatment estimate where the trial had no control arm.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Tislelizumab")
n      <- c(104)
events <- c(25)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Tislelizumab  104  25  24.0  16.2  33.4  16.8  33.1  16.9  33.1

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome, objective response rate assessed by an independent review committee, and a set of secondary outcomes. The secondary list includes duration of response, progression-free survival, disease control rate, overall survival, and investigator-assessed outcomes. The record lists 10 outcome measures and one analysis. It does not report a statistical plan explaining how the primary and secondary outcomes were prioritized for inference, whether adjustment for multiple comparisons was planned, or how any secondary analyses were intended to be interpreted. Those details are not reported in the registry record.

Multiplicity arises when several outcomes or statistical comparisons are examined. In general, making many inferences can increase the chance of at least one apparently notable result arising by chance. Whether and how to address that issue depends on the analysis plan and the purpose of each outcome. Without a reported plan, readers cannot determine whether secondary results were adjusted, exploratory, or intended to support formal claims. The registry's list of outcomes alone does not resolve that question.

The record defines the primary efficacy population as participants who received any dose and had measurable disease at baseline according to the independent review committee. It does not report the size of that population, the number with missing or unevaluable response assessments, or how missing assessments were handled. It also does not report whether analysis populations were defined for secondary outcomes or how participants who discontinued follow-up were treated in time-to-event analyses.

These omissions matter because analysis population and missing-data handling can affect a reported percentage and its uncertainty. Without the endpoint numerator, denominator, estimate, and interval, the primary response result cannot be independently interpreted from this registry record. The registry does not report a detailed statistical analysis plan, an interim-analysis plan, or a sample-size justification.

12. 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: pyrexia (3.5%), pneumonia (3.5%), general physical health deterioration (2.7%), hepatic failure (2.7%), upper respiratory tract infection (2.7%).

The registry records 49 participants affected by serious adverse events among 113 at risk. It separately records 110 participants affected by other adverse events among 113 at risk, and lists 89 deaths. These figures describe the entries in the safety record; they do not, on their own, establish that an event was caused by tislelizumab. The registry record does not report the timing, relationship to treatment, event duration, or overlap among these categories in the information.

The listed serious adverse-event terms are Pyrexia and Pneumonia, each at 3.5%; General physical health deterioration, Hepatic failure, Upper respiratory tract infection, and Urinary tract infection, each at 2.7%. These percentages describe the recorded terms. The registry entry does not give additional event-level context here, such as whether a participant had more than one event or how the term-specific percentages were derived.

Serious adverse events and other adverse events are different categories in the record and should not be combined or treated as mutually exclusive. The death count also should not be interpreted as a count of deaths caused by treatment: cause and attribution are not reported in the registry record. With no control group, these figures cannot quantify a comparative safety difference.

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

The study record describes a completed, multicenter, single-arm Phase 2 evaluation of tislelizumab in participants with PD-L1+ locally advanced or metastatic urothelial bladder cancer after progression during or following a platinum-containing regimen. It defines objective response rate as the primary outcome and specifies independent review committee assessment using RECIST version 1.1. Those details explain what the primary endpoint was intended to measure and which population was eligible for its analysis.

The record does not report the primary response percentage or its 95% confidence interval. It therefore does not support a statement about the magnitude of the observed response proportion or how precisely it was estimated. The secondary outcome names are listed, but their results are not reported in the registry record. The safety entries provide selected counts and serious-event terms, but without comparison or additional context they cannot establish the causes of events or estimate a difference in risk.

Even if a response estimate were available, a one-group result would describe outcomes among participants in this study rather than a causal contrast with another treatment. There is no listed control group, and the record does not describe randomized allocation. The result would need to be interpreted in light of the eligibility criteria, the efficacy-evaluable population, the response-assessment procedures, and the follow-up period.

In general, a response endpoint is one component of an evidence base. It does not by itself answer questions about response durability, progression-free survival, overall survival, or comparative safety. This registry record lists some of those outcomes but does not report their results. The appropriate reading is therefore limited to the design and endpoint definitions that are recorded, plus the specific safety counts and terms listed.

14. Why This Trial Matters Statistically

This trial is statistically notable for the questions its design can and cannot answer. The registry identifies a clearly defined primary outcome: confirmed complete or partial response assessed by an independent review committee under RECIST version 1.1. It also defines an efficacy-evaluable population rather than treating enrollment alone as the endpoint denominator. These distinctions are central to interpreting a binary response endpoint.

At the same time, the single-arm design makes the estimand descriptive. It can characterize response in the study population, but without a concurrent control group it cannot estimate a randomized treatment contrast. That limitation is especially important when readers consider results alongside other studies: differences across studies may reflect differences in populations, assessment, follow-up, or other design features, not solely treatment effects.

The registry record also illustrates why a protocol-level endpoint definition is not a substitute for a reported result. A specified 95% confidence interval would convey uncertainty around the response estimate, but neither the estimate nor interval is reported in the registry record. The number enrolled is listed, while the number in the defined efficacy-evaluable set is not. Thus, the information needed to assess the primary estimate and its precision is absent.

Finally, the secondary outcome list points to distinct aspects of disease course, while the safety record lists selected adverse-event counts and terms. Without reported secondary results, detailed analysis methods, and comparative safety data, these entries cannot answer broader questions about comparative outcomes. The study is useful statistically as a case in separating outcome definition, analysis population, uncertainty, and causal comparison.

15. 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 main trial-specific limitation is the absence of a comparator arm. The registry records a single-group study with no allocation to a control condition. Consequently, outcomes in the treated group cannot be interpreted as a randomized difference from another treatment or from no study treatment.

The primary endpoint result is not reported in the registry record. Although the endpoint definition and 95% confidence interval are specified, the response estimate, interval, and number of participants in the efficacy-evaluable population are absent. This prevents assessment of the observed response proportion and its precision. Enrollment is listed as 113, but the defined primary analysis population requires receipt of any dose and measurable baseline disease according to the independent review committee.

The record lists secondary outcomes but does not report their results or detailed definitions beyond the outcome names. It does not report the statistical analysis plan, multiplicity procedures, missing-data handling, interim analyses, or sample-size justification. The methods used to calculate the confidence interval are also not reported in the registry record.

Safety interpretation is limited by the absence of a control group and limited context for the reported counts. The record does not provide event timing, attribution, overlap, or cause of death in the information summarized. The serious-event terms and percentages therefore cannot establish comparative risks or causal relationships.

These limitations reflect what is not reported in the registry record. They do not establish that a particular method was or was not used outside the record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of Tislelizumab in Participants With Locally Advanced or Metastatic Urothelial Bladder Cancer — NCT04004221.
• Registry record last updated 2024-10-26; status last verified 2024-10.
• Results first posted 2023-03-16.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes NCT04004221, a completed, multicenter, single-arm Phase 2 study sponsored by BeiGene. It lists 113 enrolled participants and tislelizumab at 200mg intravenously every 3 weeks. The study concerned PD-L1+ locally advanced or metastatic urothelial bladder cancer after progression during or following a platinum-containing regimen.

The primary endpoint was objective response rate, defined as the percentage with confirmed complete or partial response assessed by an independent review committee using RECIST version 1.1. The record specifies a 95% confidence interval and an efficacy-evaluable population, but the response estimate, interval, and analysis-set size are not reported in the registry record. Secondary outcomes include duration of response, progression-free survival, disease control rate, and overall survival.

The safety record lists serious adverse events affecting 49 participants among 113 at risk, other adverse events affecting 110 among 113 at risk, and 89 deaths. These data come from a single-arm record and do not establish comparative effects or event causation.