Study Evaluating the Efficacy and Safety of Multiple Immunotherapy-Based Treatments and Combinations in Patients With Urothelial Carcinoma (MORPHEUS-UC): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Evaluating the Efficacy and Safety of Multiple Immunotherapy-Based Treatments and Combinations in Patients With Urothelial Carcinoma (MORPHEUS-UC) (NCT03869190), a randomized phase 1, phase 2 study. The primary outcome is Stage 1 mUC Cohorts: Objective Response Rate (ORR).

2. Clinical Question

3. Trial Design

MORPHEUS-UC was a Phase Ib/II, open-label, multicenter, randomized umbrella study. The registry lists a parallel design and no masking. The overall enrollment was 272, across the study's multiple cohorts and treatment arms; that figure is not the number of participants in the primary Stage 1 mUC comparison. The registry lists a start date of 2019-06-01, a primary completion date of 2025-10-08, and a status of terminated.

The primary endpoint is restricted to the Stage 1 mUC cohorts. Seven treatment regimens are compared with atezolizumab alone: atezolizumab combined with enfortumab vedotin, Hu5F9-G4, niraparib, RO7122290, sacituzumab govitecan, tiragolumab, or tocilizumab. The primary endpoint group sizes range from 12 to 30 participants, while the atezolizumab comparator includes 29 participants. These small groups are important context when reading the estimates and confidence intervals.

The study also included MIBC cohorts, with regimens organized by cisplatin eligibility and PD-L1 status, and Stage 2 mUC cohorts. The record describes Stage 2 treatment as an option for some mUC participants who lost clinical benefit or experienced unacceptable toxicity during Stage 1. Those other cohorts and stages broaden the study but are not part of the primary Stage 1 mUC ORR comparisons described in the main results.

The flexible umbrella design allowed treatment arms to be opened, closed, or modified as new treatments became available or as activity and toxicity were assessed. The specific reasons for closing each arm, the randomization details, and the original statistical design assumptions 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: Results include only those cohorts that enrolled participants; planned cohorts that were never opened for enrollment are not presented.

Only the first 6 of 16 groups are shown.

Stage 1

Reasons for not completing this period:

Stage 2

Reasons for not completing this period:

Long-term Survival Follow-up (LTSFU)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Efficacy-evaluable population included all participants who received at least one dose of each drug for their assigned treatment regimen.

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 randomized population included all participants randomized to each treatment arm.

Only the first 6 of 17 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Stage 1 mUC Cohorts: Objective Response Rate (ORR), measured over Up to 61.5 months.

The primary endpoint is objective response rate (ORR) in the Stage 1 mUC cohorts. It measures the percentage of efficacy-evaluable participants who had a complete response or a partial response during Stage 1, as assessed by the investigator using RECIST v1.1. A response had to be recorded on two consecutive occasions at least 4 weeks apart.

A complete response required disappearance of all target and non-target lesions and normalization of tumor marker level; pathological lymph nodes also had to shrink to less than 10 millimeters in short axis. A partial response required at least a 30% decrease in the sum of the diameters of target lesions from baseline, in the absence of a complete response. The registry notes that reported percentages were rounded.

ORR is used as a direct measure of observed tumor shrinkage. It is distinct from progression-free survival, overall survival, and duration of response: a response rate does not indicate how long a response lasts or how long participants remain alive or free of progression. The investigator-based assessment and open-label design are relevant when interpreting this endpoint. The registry does not report whether an independent review of response was used.

7. Results Posted in the Registry

• Stage 1 mUC Cohort: Atezo + EV: 26.7% (4/15) versus 27.6% (8/29) on control; risk difference -0.9 percentage points (95% CI -28.6 to 26.7), risk ratio 0.97 (0.35–2.70); unadjusted two-proportion z-test p = 0.948.
• Stage 1 mUC Cohort: Atezo + Hu5F9-G4: 26.7% (4/15) versus 27.6% (8/29) on control; risk difference -0.9 percentage points (95% CI -28.6 to 26.7), risk ratio 0.97 (0.35–2.70); unadjusted two-proportion z-test p = 0.948.
• Stage 1 mUC Cohort: Atezo + Nira: 6.7% (1/15) versus 27.6% (8/29) on control; risk difference -20.9 percentage points (95% CI -41.5 to -0.3), risk ratio 0.24 (0.03–1.76); unadjusted two-proportion z-test p = 0.103.
• Stage 1 mUC Cohort: Atezo + RO7122290: 16.7% (2/12) versus 27.6% (8/29) on control; risk difference -10.9 percentage points (95% CI -37.6 to 15.7), risk ratio 0.60 (0.15–2.44); unadjusted two-proportion z-test p = 0.459.

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.

Stage 1 mUC Cohorts: Progression-free Survival (PFS) After Randomization

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + EV: hazard ratio 1.39 (95% CI 0.72–2.69) (as reported in the registry).
  • Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + Hu5F9-G4: hazard ratio 1.16 (95% CI 0.57–2.34) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + EV and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + Hu5F9-G4 and the control. The interval gives the range of effects that remain compatible with the data.

Stage 1 mUC Cohorts: Overall Survival (OS) After Randomization

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + EV: hazard ratio 2.52 (95% CI 1.23–5.17) (as reported in the registry).
  • Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + Hu5F9-G4: hazard ratio 1.01 (95% CI 0.46–2.24) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically higher for Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + EV than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Stage 1 mUC Cohort: Atezo (Control) vs Stage 1 mUC Cohort: Atezo + Hu5F9-G4 and the control. The interval gives the range of effects that remain compatible with the data.

Stage 1 mUC Cohorts: OS Rate at Month 12

• 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:
  • Stage 1 mUC Cohort: Atezo + EV: 26.7% (4/15) versus 65.5% (19/29) on control; risk difference -38.9 percentage points (95% CI -67.1 to -10.6), risk ratio 0.41 (0.17–0.98); unadjusted two-proportion z-test p = 0.014.
  • Stage 1 mUC Cohort: Atezo + Hu5F9-G4: 60.0% (9/15) versus 65.5% (19/29) on control; risk difference -5.5 percentage points (95% CI -35.7 to 24.7), risk ratio 0.92 (0.56–1.50); unadjusted two-proportion z-test p = 0.718.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Stage 1 mUC Cohort: Atezo + EV than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Stage 1 mUC Cohort: Atezo + Hu5F9-G4 and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes add context to the primary response-rate results by describing progression-free survival, overall survival, survival status at a specified time, and other measures. The Stage 1 mUC progression-free survival medians are numerically lower in the combination groups than in the atezolizumab comparator group. Overall survival medians vary across the groups rather than following the same pattern. These endpoint summaries do not give a uniform direction across outcomes, and their intervals are often broad.

The reported hazard-ratio comparisons are limited to the atezolizumab versus enfortumab vedotin and atezolizumab versus Hu5F9-G4 comparisons for progression-free and overall survival. The registry does not report p-values or named analysis methods for these comparisons. For the month-12 overall survival rate, estimates also vary among the groups; the reported comparisons cover only some of the treatment groups. These findings should be considered alongside the primary ORR rather than as independent confirmation of a single treatment ranking.

The secondary endpoints relate to the primary result by addressing different aspects of outcome. ORR records whether a confirmed response occurred, while progression-free survival concerns time to progression or death, overall survival concerns time to death, and a month-specific rate describes status at a particular time point. Their differing patterns are not inherently contradictory because they measure different things and use different analyses.

Many outcomes and treatment comparisons create opportunities for chance findings. The registry does not report whether secondary analyses were adjusted for multiplicity or whether they were prespecified as confirmatory. Unadjusted secondary results therefore carry less weight as evidence of treatment differences than a clearly specified primary analysis with an appropriate inferential plan. Results for duration of response and disease control rate are not reported in the registry record.

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

• Stage 1 mUC Cohorts: Progression-free Survival (PFS) After Randomization
• Stage 1 mUC Cohorts: Overall Survival (OS) After Randomization
• Stage 1 mUC Cohorts: OS Rate at Month 12
• Stage 1 mUC Cohorts: Duration of Response (DOR)
• Stage 1 mUC Cohorts: Disease Control Rate (DCR)
• Stages 1 and 2 mUC Cohorts: Number of Participants With Adverse Events (AEs)
• Stage 2 mUC [Atezo + EV Cohort]: Serum Concentration of Atezolizumab at Specified Timepoints
• Stage 2 mUC [Atezo + EV Cohort]: Number of Participants With Anti-drug Antibodies (ADAs) to Atezolizumab

9. Statistical Interpretation

10. Statistical Methods Explained

The results involve several statistical quantities that answer different questions. Understanding what each one describes helps avoid treating a percentage, a confidence interval, a test result, and a planning calculation as interchangeable evidence.

What does an objective response rate measure?

ORR is a binary outcome summarized as a percentage: each participant either met the response definition or did not. In this study, a response required a complete or partial response assessed by the investigator under RECIST v1.1 and confirmed on two consecutive occasions at least 4 weeks apart. The rate is calculated within each reported treatment group. It is not a measure of the duration of response or of time to progression.

A percentage is an estimate based on the participants included in the relevant analysis population. With small groups, a change in the outcome of a small number of participants can noticeably affect the rate. The record defines the primary endpoint population as participants who received at least one dose of each drug in their assigned regimen. It does not give further detail about handling participants with missing response assessments.

What does a risk difference tell the reader?

A risk difference compares two event probabilities by subtracting one group rate from the other. It is expressed either as a proportion or, after scaling, as percentage points. A value of zero represents no difference in the estimated rates. The sign depends on the order in which the groups are subtracted, so the comparison's stated direction matters. The recalculated risk differences in the posted comparisons use the atezolizumab comparator and the named combination group; the estimated differences are negative for each listed combination. The associated intervals are wide, reflecting substantial uncertainty in these small samples.

The registry also reports differences in overall response rates. Its values are expressed as percentage-point differences, and the estimates and intervals should be read with the specified comparator in mind. The analysis method is not identified in the registry record. A difference estimate alone does not establish that the groups differ; its interval and the analysis plan are also relevant.

How should confidence intervals be read?

A confidence interval gives a range of values compatible with the data under the method used to calculate it. It is not a guarantee that the true value falls inside the interval, nor does it list the possible outcomes for individual participants. Wider intervals indicate less precision. For a difference, an interval that includes zero leaves no difference among the values compatible with the interval. For a ratio such as a hazard ratio, the corresponding no-association value is one. These are general interpretations of confidence intervals; the particular interval method used for each registry analysis is not reported.

The widths of the ORR intervals are important because the treatment groups are small. For example, the estimates for two groups can be numerically different while their confidence intervals remain broad. That pattern makes the precise size and direction of a group difference uncertain. Overlapping intervals are not a formal test of a treatment difference; the comparison-specific estimate and interval are more directly relevant.

What are p-values and what do they not say?

A p-value, when reported, describes how unusual data at least as incompatible with a specified null hypothesis would be under that hypothesis and the test's assumptions. It is not the probability that the null hypothesis is true, and it does not quantify the size or clinical importance of a difference. The sponsor's posted ORR comparison analyses do not include p-values. The simple unadjusted recalculations include p-values, but those values are not substitutes for a prespecified, multiplicity-aware analysis.

Testing several treatment groups against a comparator raises a multiplicity issue: with repeated tests, the chance of at least one apparently unusual result can increase. Whether and how the trial addressed this issue is not reported in the registry record. This matters when interpreting individual unadjusted p-values, especially alongside multiple secondary outcomes and a flexible umbrella structure.

What do medians and hazard ratios describe?

For a time-to-event endpoint, the median is the time by which half of the analyzed group has experienced the event, under the endpoint's definition and analysis. It is not an average survival time. A confidence interval around a median shows uncertainty in its estimated location. The posted secondary results include median progression-free survival and median overall survival for Stage 1 mUC groups.

A hazard ratio compares the event rates over follow-up between groups under a time-to-event model. A hazard ratio of one is the no-difference reference value. The registry labels the posted hazard ratios as reported, but does not name the method or provide p-values for those comparisons. A hazard ratio should not be read as a ratio of median times; these are different summaries and need not convey identical information.

What does a planning calculation mean?

The sample-size illustrations use the observed effect or a stated fraction of it under assumptions of two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. For binary endpoints, the figures are participants per arm. For time-to-event outcomes, the figures are total events, using the Schoenfeld approach. These are planning illustrations derived from the observed estimates, not the original trial's design calculations. The original sample-size justification and its assumptions are not reported in the registry record.

11. Confidence Intervals and Precision

The primary ORR estimates are imprecise because the groups are small and their confidence intervals cover broad ranges. The comparator estimate of 27.6% has a 95% interval from 12.73% to 47.24%. The combination estimates likewise have broad intervals: for example, 26.7% with enfortumab vedotin has an interval from 7.79% to 55.1%, and 6.7% with niraparib has an interval from 0.17% to 31.95%. These ranges illustrate that the observed percentages do not pin down the underlying response rates closely.

Comparison intervals are also broad. For enfortumab vedotin, the recalculated risk difference is -0.009 with an interval from -0.286 to 0.268; the interval includes zero and spans differences in either direction. For niraparib, the recalculated interval runs from -0.415 to -0.003. Although that interval does not include zero under this simple calculation, the result comes from small groups and is unadjusted across the multiple comparisons. The sponsor-reported ORR difference interval for niraparib, from -46.57 to 4.73 percentage points, includes zero. The sponsor's method is not reported, so the two analyses should not be conflated.

Confidence intervals describe statistical precision, not all sources of uncertainty. They do not resolve potential differences in participant characteristics, assessment practices, missing data, or the implications of the open-label design. The record's response assessments were investigator-determined, and an independent review process is not reported. The width of an interval can therefore be read as a feature of the estimate's sampling uncertainty, but not as a complete account of study limitations.

For time-to-event secondary outcomes, the reported intervals also vary in width, and some upper bounds are not reported. An absent bound should not be interpreted as a particular numerical value. Overall, the posted intervals support cautious, descriptive interpretation rather than a precise ranking of the regimens.

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 translate an observed treatment contrast into an illustrative number of participants or events that a future study might need under stated assumptions. For binary outcomes such as ORR, the entries are participants per arm, not a total across both groups. For time-to-event outcomes, the entries are total events, using the Schoenfeld approach. The time-to-event figures therefore do not represent participants per arm.

The illustrations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They use the observed effect or a stated fraction of that effect. A row describing planning for 75% or 50% of the observed effect means that the planning calculation uses a smaller effect than the one observed; smaller effects need more participants or events to distinguish from the null under the same assumptions. The registry's original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the calculated number of participants per arm, or total events for time-to-event outcomes, for a design with an 80% probability of detecting the assumed effect if that effect is truly present and the planning assumptions hold. The 90% column gives the corresponding quantity for a design with a 90% probability under those same conditions. These are design targets, not estimates of how likely the completed trial was to detect an effect. This is not observed power.

Because the calculations are anchored to effects observed in small groups, the apparent effect can be unstable. A planning figure based on a large observed contrast may be much smaller than one based on a modest contrast, and neither establishes what a future trial should enroll. The figures are illustrations for interpreting assumptions, not a substitute for a prospective design justified for its own population, endpoint, and analysis plan.

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 = 0.276; treatment proportion p₁ = 0.267; average p̄ = 0.271.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6288 = 1.2324.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6287 = 0.5292.
4. n = (1.2324 + 0.5292)² / (-0.0092)² = 36,698.2, rounded up to 36,699 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 <- 0.275862068965517                          # control proportion
p1 <- 0.266666666666667                          # 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(3869190)
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 = 36,699 (80% power), 49,129 (90% power)
 75% of observed effect: n per arm = 65,416 (80% power), 87,573 (90% power)
 50% of observed effect: n per arm = 147,574 (80% power), 197,559 (90% power)
power.prop.test: 36698.2 
Simulated power at n = 36,699 per arm: 0.799 (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 includes sponsor-reported differences in overall response rates for the Stage 1 mUC comparisons. For the enfortumab vedotin and Hu5F9-G4 groups, the reported difference is -0.92 percentage points, with a 95% confidence interval from -33.64 to 31.80. The corresponding simple recalculated risk difference is -0.009, with an interval from -0.286 to 0.268. Those estimates express the same small observed difference on different scales, while the intervals are not identical.

For niraparib and sacituzumab govitecan, the sponsor-reported difference is -20.92 percentage points, with a 95% confidence interval from -46.57 to 4.73. The simple recalculation gives -0.209, with an interval from -0.415 to -0.003. The other sponsor-reported comparisons also have their own estimates and intervals in the registry results. No p-values are reported for the sponsor's primary comparisons, and the specific methods used to calculate the sponsor estimates are not identified in the record.

The sponsor's analysis is the registry's reported analysis. The recalculations are unadjusted comparisons made from posted summary statistics, so differences between the intervals should not be treated as proof that one analysis is wrong. The registry's discrepancy indicator is false. No further explanation of the interval-method differences is reported, and the recalculations do not replace the sponsor-reported results.

15. Multiplicity, Missing Data and Analysis Populations

The study evaluates multiple treatment regimens across mUC and MIBC cohorts, and the registry record lists numerous outcomes and analyses. Multiple comparisons can increase the likelihood that at least one result appears noteworthy by chance. The record does not report a multiplicity strategy, such as a hierarchy of hypotheses or an adjustment to control the overall error rate. The sponsor's primary ORR comparison p-values are not reported, and the methods underlying the sponsor's confidence intervals are not identified.

The primary endpoint description defines the efficacy-evaluable population as participants who received at least one dose of each drug in their assigned treatment regimen. The registry does not report additional details about the analysis population, the handling of participants without evaluable response assessments, or whether any participants were excluded after randomization. It also does not describe imputation or other missing-data procedures. These omissions limit assessment of how incomplete outcome information may have affected the reported rates.

For progression-free and overall survival, the endpoint definitions specify the event and the posted results give median estimates and confidence intervals. The censoring rules, follow-up procedures for participants without an event, and details of the time-to-event analysis are not reported in the registry record. The hazard-ratio analysis methods are likewise not named. Those details can matter to the interpretation of time-to-event estimates.

The open-label design and investigator assessment of response are relevant because participants and investigators were not masked. The record does not report whether response assessments were reviewed independently. The flexible umbrella design, with arms that could be opened, closed, or modified, also makes it important to know how comparisons and populations were prespecified. The reasons for arm closures and the full statistical plan are not reported in the registry record. These are uncertainties about the evidence record, not proof that a particular analysis was biased.

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: urinary tract infection (36.4%), acute kidney injury (33.3%), sepsis (27.3%), pyrexia (20.0%), acute coronary syndrome (16.7%).

The adverse-event summaries report the number of participants at risk and the number with serious adverse events or other adverse events in each group. For example, the Stage 1 mUC atezolizumab group lists 11 participants with serious adverse events among 29 at risk, while the atezolizumab plus enfortumab vedotin group lists 9 among 15. The other-event counts are also reported by group. These are descriptive counts; the groups differ in size and should not be compared by raw counts alone.

The registry's listed serious adverse event terms include urinary tract infection at 36.4%, acute kidney injury at 33.3%, sepsis at 27.3%, pyrexia at 20.0%, acute coronary syndrome at 16.7%, and enterovesical fistula at 16.7%. The record excerpt does not identify the specific group or denominator for these percentages, so they should not be assigned to an individual treatment group. Serious adverse events and other adverse events are distinct summaries, and the counts do not by themselves establish causation.

The record also reports deaths by group, but the registry record does not establish that each death was caused by treatment. More detailed event timing, severity, relatedness, and exposure-adjusted comparisons are not reported in the registry record. Safety counts should therefore be read as observed descriptive data rather than as a comparative causal analysis.

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

The posted primary results describe the proportion of efficacy-evaluable participants in each Stage 1 mUC group who had a confirmed investigator-assessed response. They show a range of observed response rates across the treatment groups, with the atezolizumab-alone group at 27.6% and combination-group estimates between 6.7% and 26.7%. The small group sizes and broad confidence intervals mean that these point estimates are uncertain.

The figures do not establish that a particular regimen is better or worse. Some intervals include no difference, and simple unadjusted recalculations do not account for the many comparisons. Sponsor-reported analyses and the recalculations also use different intervals; the sponsor's method is not reported. The registry does not give the original statistical plan, multiplicity approach, or detailed missing-data procedures needed to fully assess the inferential framework.

ORR is one dimension of outcome. It does not show how long participants remained in response, how long they lived without progression, or whether they experienced adverse events. Secondary outcomes provide additional kinds of information, but they are not interchangeable with the primary endpoint. The broad secondary patterns and limited number of reported comparisons do not resolve the uncertainty in the primary ORR comparisons.

The registry record describes a randomized, open-label umbrella study with multiple cohorts and stages, and reports an overall enrollment of 272. That overall enrollment should not be mistaken for the number of participants contributing to the primary comparisons. The study status is terminated, but the reason is not reported in the registry record. Neither the status nor the response rates alone support conclusions about why the study ended or whether a treatment should be used.

18. Why This Trial Matters Statistically

Statistically, MORPHEUS-UC illustrates the challenge of interpreting multiple small treatment groups within an umbrella study. The primary Stage 1 mUC endpoint is a clearly defined binary response outcome, and the registry reports group-specific rates with confidence intervals. Those intervals make the uncertainty visible: a single percentage can look definitive when read alone, but estimates from small groups may be compatible with a broad range of underlying response rates.

The study also demonstrates why the analysis population and comparison structure matter. Although the overall enrollment is 272, the primary endpoint groups contain between 12 and 30 participants. The sample size relevant to a particular treatment comparison is not the total study enrollment. The flexible design, multiple arms, and multiple outcomes further raise questions about multiplicity and how analyses were prespecified, which the registry does not answer.

The secondary results emphasize that different endpoints can present different perspectives. Response, progression-free survival, overall survival, and adverse-event summaries do not measure the same thing. Their interpretation depends on their definitions, analysis populations, and uncertainty, not just on whether point estimates differ.

Finally, the record permits a useful distinction between the sponsor's reported estimates and simple re-calculations from posted summary data. Keeping those sources separate helps readers see what is directly reported and what is an unadjusted statistical comparison, without treating the latter as a replacement for the trial's own analysis.

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.

Several limitations are specific to the available record. The study is open label, and the primary ORR was assessed by investigators. Whether an independent response review was conducted is not reported in the registry record. The response endpoint is based on a defined confirmation rule, but the registry does not provide further detail about assessment scheduling or missing response evaluations.

The Stage 1 mUC groups are small, and the umbrella study includes numerous treatment regimens, cohorts, and outcomes. This makes the primary estimates imprecise and creates a multiplicity concern. The statistical plan, randomization procedure, allocation ratio, and original sample-size justification are not reported. The sponsor's analysis methods are also not specified, including the approach used for confidence intervals. Consequently, the differences between sponsor-reported intervals and simple unadjusted recalculations cannot be attributed to a particular method from the registry record.

The record describes the ability to open or close arms and modify the population, but it does not report the reasons for specific arm decisions or their timing. The reason for the terminated status is also not reported. Details of missing-data handling, censoring, and analysis populations for secondary outcomes are unavailable. These omissions make it difficult to assess whether the posted analyses reflect all randomized participants or how incomplete follow-up was addressed.

Finally, the primary endpoint addresses response rather than duration of benefit or survival. Secondary measures may add context, but the posted data and methods are not sufficient to turn the observed group rates into a definitive ranking of regimens.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study Evaluating the Efficacy and Safety of Multiple Immunotherapy-Based Treatments and Combinations in Patients With Urothelial Carcinoma (MORPHEUS-UC) — NCT03869190.
• Registry record last updated 2026-08-03; status last verified 2026-07.
• Results first posted 2026-08-03.
• 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

MORPHEUS-UC is a Phase Ib/II, randomized, open-label, multicenter umbrella study with a parallel design. The registry reports overall enrollment of 272 and a terminated status. The primary endpoint is Stage 1 mUC objective response rate, based on investigator-assessed complete or partial responses confirmed on consecutive occasions at least 4 weeks apart.

The primary results come from small treatment groups, with broad confidence intervals around response rates and comparisons. The registry reports sponsor analyses without p-values or named methods; unadjusted recalculations provide additional descriptive comparisons but do not replace those analyses. Multiple arms and outcomes, missing-data procedures, multiplicity handling, and the reason for termination are not fully described in the registry record. The results are therefore best read as uncertain, descriptive estimates rather than a definitive ranking of treatments.