An Open-Label, Multicenter Study Evaluating the Safety, Efficacy, and Pharmacokinetics of Mosunetuzumab in Combination With Tiragolumab With or Without Atezolizumab in Participants With B-Cell Non-Hodgkin Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Open-Label, Multicenter Study Evaluating the Safety, Efficacy, and Pharmacokinetics of Mosunetuzumab in Combination With Tiragolumab With or Without Atezolizumab in Participants With B-Cell Non-Hodgkin Lymphoma (NCT05315713), a non-randomized phase 1, phase 2 study. The primary outcome is Percentage of Participants With Adverse Events - Phase 1b.

2. Clinical Question

The central clinical question was whether the mosunetuzumab–tiragolumab combination, with or without atezolizumab, could be evaluated for safety, efficacy, and pharmacokinetics in people with relapsed or refractory B-cell non-Hodgkin lymphoma. The registry’s stated population specifically includes diffuse large B-cell lymphoma and follicular lymphoma after at least two previous systemic-treatment lines.

The posted primary outcome is a safety measure: the percentage of participants with adverse events during a defined treatment-related observation period. It describes how often events were recorded in the enrolled group; it does not, by itself, establish that treatment caused an event. Because the atezolizumab-containing arm did not enroll and the study was not randomized, the record cannot support a direct comparison between the planned combinations.

The registry lists additional intended outcomes, including response, duration of response, progression-free survival, event-free survival, overall survival, and pharmacokinetic concentration. Results for those outcomes are not reported in the registry record. The available information therefore addresses the study’s treatment setting and some safety observations, but does not answer the efficacy questions that motivated the broader study.

3. Trial Design

This was an open-label, multicenter Phase Ib/II study sponsored by Hoffmann-La Roche. Its stated purpose was to evaluate safety, efficacy, and pharmacokinetics for mosunetuzumab combined with tiragolumab, with or without atezolizumab, in participants with relapsed or refractory B-cell non-Hodgkin lymphoma. The population described in the registry included diffuse large B-cell lymphoma and follicular lymphoma after at least two previous lines of systemic therapy.

The registry characterizes the design as parallel and nonrandomized, with no masking. Those features matter when interpreting any apparent differences between treatment groups: assignment was not randomized, and treatment identity was not concealed. In addition, the planned arm receiving mosunetuzumab, tiragolumab, and atezolizumab did not enroll participants. The posted safety group is the arm receiving subcutaneous mosunetuzumab with intravenous tiragolumab.

Participants in the described arms were to receive at least 8 and up to 17 treatment cycles, with a cycle length of 21 days. The registry reports enrollment of 8. It lists a start date of 2022-05-10, a primary completion date of 2023-07-19, and a status of terminated. The reason for termination is not reported in the registry record, so the status alone does not explain why the study ended or what drove the decision.

The primary Phase 1b outcome was the percentage of participants with adverse events during treatment and through 90 days after the final dose, with the timeframe stated as up to 36 weeks. The registry also lists secondary outcomes spanning investigator-assessed response using Lugano 2014 criteria, pharmacokinetics, time-to-event outcomes, overall survival, and Phase 2 adverse events. These planned outcomes indicate a broader assessment than the posted primary safety results, but outcome results beyond the safety information described in the record are not reported in the registry record.

Although the study has Phase Ib and Phase II components, the record does not report how participants were allocated between phases, whether any planned transitions or adaptations occurred, or how phase-specific analysis populations were defined. The protocol’s operational details beyond the listed treatment schedules 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.

Overall Study

Reasons for not completing this period:

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.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Adverse Events - Phase 1b, measured over From the start of treatment until 90 days after the final dose of study treatment (up to 36 weeks).

The primary Phase 1b endpoint is the percentage of participants with adverse events. The observation period begins at the start of treatment and continues until 90 days after the final dose of study treatment, for a stated maximum timeframe of 36 weeks. The measure summarizes whether participants experienced adverse events during that period. It is a participant-level percentage outcome, rather than a measure of event severity, duration, or the total number of events.

In general, an adverse-event endpoint is used to characterize the safety experience observed during a study. Reporting the proportion of participants with an event gives a view of how common recorded events were in the studied group, while separate descriptions may distinguish serious events or particular event terms. Such reporting does not establish that treatment caused an event: an adverse event is recorded during the study, and causal attribution requires separate assessment.

The registry lists the primary endpoint for Phase 1b, while a Phase 2 adverse-event percentage appears among the secondary outcomes. The available record does not describe the detailed rules for event coding, severity grading, attribution, or handling of recurrent events. It also does not report the precise analysis population or denominator used for the primary endpoint. Those details are important for interpreting a participant percentage and are not reported in the registry record.

The primary safety endpoint should be read separately from efficacy outcomes such as response or survival. A safety percentage cannot establish tumor response, and an efficacy outcome would not by itself characterize tolerability. Although the registry lists efficacy-related outcomes, their results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted primary outcome is a percentage: the fraction of the analysis group that experienced an adverse event during the stated observation period. In general, a binary percentage outcome classifies each participant according to whether the event occurred. It is distinct from an event count, because a participant may have repeated events while still contributing as an affected participant to a participant-level percentage. The record does not report the detailed rules for recurrent events or the analysis denominator.

What does a confidence interval describe?

In general, a confidence interval expresses the uncertainty around an estimated quantity under a specified statistical procedure. A narrow interval indicates greater precision under that procedure; a wide interval indicates greater uncertainty. For a participant percentage, precision depends in part on how many participants contribute usable outcome information and on the observed event frequency. No confidence interval for the primary endpoint is reported in the registry record, so the precision of the posted safety estimate cannot be assessed from an interval.

A confidence interval is not a guarantee that the true value lies within its bounds, nor does it repair problems in the study design. It quantifies sampling uncertainty under assumptions, not uncertainty due to selection, missing observations, outcome classification, or lack of a concurrent randomized control. In this study, the nonrandomized design and absence of enrollment in the atezolizumab-containing arm remain relevant even if an interval were available.

Why does the design affect statistical interpretation?

Random allocation is generally used to make treatment groups comparable, on average, with respect to measured and unmeasured baseline characteristics. This registry describes a nonrandomized study, so treatment assignment cannot be assumed to have created comparable groups. More fundamentally, the planned arm that included atezolizumab enrolled no participants. The posted record therefore does not provide an enrolled comparison between the treatment combinations.

Open-label means that masking was absent. In general, knowledge of treatment can influence event reporting, participant experience, clinical management, or investigator assessment. That does not show that such influence occurred in this study; it identifies a possible source of measurement or conduct bias that should be considered alongside the reported results. The registry record does not describe any blinded review process for the safety outcome.

How should percentages and event terms be read?

A participant percentage is most interpretable when its denominator, event definition, observation window, and analysis population are clear. The registry gives a primary outcome timeframe from treatment start until 90 days after the final dose, up to 36 weeks. It also reports group counts for serious adverse events, other adverse events, and deaths. However, the detailed analysis population and data-handling rules are not reported in the registry record. The available counts should be treated as descriptive record entries rather than as a complete statistical analysis.

The listed serious-event terms each carry a reported percentage of 12.5%. Those entries should not be added together to make an overall percentage: the record does not describe whether event terms overlap within participants, and a person may have multiple terms recorded. Nor should the percentages be treated as an estimate of treatment-attributable risk. In general, an adverse event occurring during a trial is not automatically an adverse reaction caused by the intervention.

Over-dispersion is a general issue for count data when variability exceeds what a simple statistical model expects. It can matter when analyzing repeated event counts, because treating observations as less variable than they are can make uncertainty appear too small. The primary endpoint is described as a participant percentage, not a repeated-event count model. The registry record does not report a model for recurrent events, so over-dispersion cannot be evaluated as an implemented method in this analysis.

What do the planning illustrations represent?

The sample-size assumptions are planning illustrations computed from an observed effect, not a description of the trial’s own design. The stated framework uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is expressed as a fraction of the observed effect. In general, such calculations estimate the number of participants per arm that a hypothetical future comparison might require under the chosen assumptions. They are not observed power, and they do not show that the current study was adequately powered.

These assumptions are especially consequential for this record. The study was nonrandomized, the intended atezolizumab arm did not enroll, and the observed group is not a randomized comparison. A hypothetical equal-allocation comparison therefore does not describe the actual enrollment pattern. In addition, a calculation anchored to an observed effect can be unstable when the observed estimate is imprecise. The registry record does not report a protocol sample-size justification, so the illustrations should not be attributed to the study’s original planning.

Are analyses or model details reported?

The registry records 10 outcome measures and 0 analysis entries. This means that the record does not supply analysis-level results for the listed outcomes. It does not establish that no analyses were conducted outside the registry, and it does not explain why analysis entries are absent. Statistical models, adjustment variables, hypothesis tests, and confidence intervals are not reported in the registry record.

The outcome list includes binary response measures and time-to-event measures such as duration of response, progression-free survival, event-free survival, and overall survival. In general, time-to-event methods account for the time until an event and can accommodate participants whose event has not occurred by the end of observation through censoring. The registry lists these outcomes but does not report their estimates, censoring rules, follow-up summaries, or analysis methods. No time-to-event conclusion can therefore be drawn from the outcome names alone.

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("Subcutaneous (SC) Mosunetuzumab in Combination Wit")
n      <- c(8)
events <- c(8)
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  Subcutaneous (SC) Mosunetuzumab in Combination Wit  8  8  100.0  63.1  100.0  67.6  100.0  66.4  99.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcomes across Phase 1b and Phase 2, including response, complete response, duration of response, progression-free survival, event-free survival, overall survival, pharmacokinetic concentration, and adverse-event measures. When several outcomes or comparisons are tested, the chance of at least one apparently notable result can increase. In general, multiplicity procedures may control that risk, or investigators may specify a hierarchy of confirmatory and exploratory outcomes. The registry record does not report a multiplicity strategy, a testing hierarchy, or an analysis plan.

The record identifies the Phase 1b primary endpoint as the percentage of participants with adverse events, but does not report the statistical test or estimation procedure used for it. It also does not state which outcomes were intended to support confirmatory claims. Since 0 analysis entries are recorded, there are no posted analysis-level results from which to assess how multiple outcomes were handled.

Missing-data methods are not reported in the registry record. This matters because an adverse-event percentage depends on who is included and how incomplete observation is treated. The registry specifies an observation period from treatment start through 90 days after the final dose, up to 36 weeks, but does not describe loss to follow-up, data cutoffs, missing event information, or rules for excluding participants from an analysis.

Analysis populations are also not described in detail. The record does not specify whether the primary safety analysis used all enrolled participants, participants who received treatment, or another defined group. It does not report separate population definitions for the listed efficacy and pharmacokinetic outcomes. These omissions limit reproducibility and make it difficult to connect the reported group counts to a prespecified analysis population.

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: abdominal pain (12.5%), diarrhoea (12.5%), covid-19 (12.5%), pneumonia aspiration (12.5%), septic shock (12.5%).

The registry lists 8 participants at risk for serious adverse events, with 2 affected. For other adverse events, it lists 8 at risk and 8 affected. It also reports 5 deaths. These are distinct entries in the record; the available information does not explain the relationship among the categories or whether the deaths were considered treatment-related. No causal conclusion should be drawn from the counts alone.

Seriousness and intensity are different concepts in general safety reporting. “Serious” describes an event’s regulatory or clinical consequences, while severity describes how intense the event was. The registry does not provide the grading or seriousness criteria used for these entries. Nor does it report participant-level event histories that would clarify overlap across categories.

The listed top serious-event terms are abdominal pain, diarrhoea, COVID-19, pneumonia aspiration, septic shock, and fall, each reported as 12.5%. The record does not state the denominator or provide enough detail to determine overlap among terms. These percentages should be read individually as registry entries, not summed or interpreted as a total serious-event rate.

The record does not report whether an independent committee reviewed safety events, how attribution was assigned, or how follow-up was handled for participants who stopped treatment. These details are important context for interpreting the observed safety experience.

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

The posted information documents safety observations in the enrolled mosunetuzumab and tiragolumab group and describes the planned study population and treatment schedule. It reports serious and other adverse-event counts, a death count, and several serious-event terms with their registry-listed percentages. Those observations are useful as a record of what was reported for this study group.

They do not establish that the combination caused the reported events. The record does not provide causal attribution for the listed counts, a randomized control group, or a like-for-like comparison. The atezolizumab-containing arm did not enroll, so no observed comparison of the planned combinations is available. The findings also cannot be generalized confidently from this enrollment alone to all people with relapsed or refractory B-cell non-Hodgkin lymphoma.

The safety figures do not answer the efficacy questions. Response, duration of response, progression-free survival, event-free survival, overall survival, and pharmacokinetic outcomes are listed, but their results are not reported in the registry record. The presence of planned outcomes is not evidence that a particular response or survival result was observed.

Precision is another limit. Confidence intervals and analysis details are not reported, and the record does not give the analysis population or missing-data rules. The percentages and counts therefore should not be treated as precise estimates of future event rates. They are descriptive observations in the context of a study with enrollment of 8 and status terminated; the reason for termination is not reported in the registry record.

In general, an early-phase, nonrandomized study can provide descriptive safety and pharmacokinetic information that may inform later research, but it is not designed by those features alone to establish comparative effectiveness. Any sample-size planning illustrations based on an observed effect are hypothetical calculations under stated assumptions, not evidence of observed power or confirmation of a treatment effect.

14. Why This Trial Matters Statistically

Statistically, this study illustrates why design and enrollment context must accompany outcome percentages. The registry describes a nonrandomized, open-label study and reports that the atezolizumab-containing arm did not enroll. The available safety results therefore describe the enrolled combination group rather than a comparative experiment. A percentage without an appropriate comparator cannot show how the observed experience differs from another treatment or from the course of illness without that treatment.

The study also shows the distinction between a planned outcome and a reported result. The registry lists a broad set of response, survival, and pharmacokinetic outcomes, but records 0 analysis entries. Those outcome names convey what investigators intended to assess; they do not substitute for estimates, uncertainty intervals, or analysis details. For readers, the absence of posted analyses is itself a key limit on what can be concluded from the record.

Safety summaries require careful attention to denominators, event definitions, observation time, and overlap. The registry gives counts and term-specific percentages, but does not report all details needed to reconstruct those summaries. In general, this is why event terms should not be summed and why an observed adverse event should not automatically be attributed to treatment.

Finally, the planning illustrations should be separated from the actual study design. Their assumptions describe a hypothetical two-arm framework, whereas the record documents a nonrandomized study with an unfilled arm. This distinction helps prevent a planning calculation from being misread as evidence that the trial had adequate power or produced a comparative result.

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 trial-specific limitations begin with its nonrandomized, open-label design. Treatment groups cannot be assumed comparable as they might be under random assignment, and masking was absent. The planned arm combining mosunetuzumab, tiragolumab, and atezolizumab enrolled no participants, so the record offers no enrolled comparison of regimens with and without atezolizumab.

Enrollment was 8, and the registry reports status as terminated. The reason for termination is not reported in the registry record. The available record also does not explain whether termination affected the planned follow-up, the timing of analyses, or completion of outcome assessments.

The primary endpoint is described, including its timeframe, but important operational details are not reported: the analysis population, event coding and grading rules, attribution process, treatment of recurrent events, and missing-data handling. The record does not provide confidence intervals or a statistical model for the posted safety figures. This limits assessment of precision and reproducibility.

Although efficacy and pharmacokinetic outcomes are listed, their results and methods are not reported in the registry record. The record does not describe analysis populations or censoring rules for the listed time-to-event outcomes. It also does not state a multiplicity strategy or protocol sample-size justification. The hypothetical planning assumptions should therefore be kept distinct from the trial’s actual design and from any claim about its ability to detect an effect.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: An Open-Label, Multicenter Study Evaluating the Safety, Efficacy, and Pharmacokinetics of Mosunetuzumab in Combination With Tiragolumab With or Without Atezolizumab in Participants With B-Cell Non-Hodgkin Lymphoma — NCT05315713.
• Registry record last updated 2024-10-04; status last verified 2024-09.
• Results first posted 2024-10-04.
• 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

NCT05315713 is an open-label, multicenter Phase Ib/II study sponsored by Hoffmann-La Roche. It evaluated mosunetuzumab with tiragolumab, with or without atezolizumab, for relapsed or refractory B-cell non-Hodgkin lymphoma, including diffuse large B-cell lymphoma and follicular lymphoma after at least two previous systemic-treatment lines.

The design was parallel and nonrandomized, with no masking. Enrollment was 8. The arm receiving subcutaneous mosunetuzumab with intravenous tiragolumab enrolled participants; the planned arm adding intravenous atezolizumab did not enroll. The listed treatment schedule was at least 8 and up to 17 cycles, each 21 days long. The study started on 2022-05-10, had primary completion on 2023-07-19, and is marked terminated.

The primary Phase 1b outcome was the percentage of participants with adverse events through 90 days after the final dose, up to 36 weeks. The record lists safety counts and serious-event terms, but no analysis entries. Results for the listed efficacy, survival, and pharmacokinetic outcomes are not reported in the registry record.