Phase I-Ib/II Study of MBG453 as Single Agent and in Combination With PDR001 in Patients With Advanced Malignancies: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase I-Ib/II Study of MBG453 as Single Agent and in Combination With PDR001 in Patients With Advanced Malignancies (NCT02608268), a non-randomized phase 1, phase 2 study. The primary outcome is Phase I-Ib and Dose Ranging Part: Number of Participants With Adverse Events (AEs) and Serious Adverse Events (SAEs) During the On-treatment Period.

2. Clinical Question

This study was designed to characterize safety, tolerability, pharmacokinetics, pharmacodynamics, and anti-tumor activity of sabatolimab as a single agent or with another study drug in adults with advanced solid tumors. The posted primary endpoint concerns adverse events, not a randomized estimate of benefit relative to a control treatment.

Because treatment allocation was not randomized, differences between groups may reflect who entered each group, the disease groups represented, the doses and schedules used, or other factors, as well as any differences in treatment. The registry record does not report a like-for-like control comparison. Thus, the central question for reading the posted primary results is descriptive: what events were recorded in the treated groups, and how much detail is available to interpret them?

3. Trial Design

This was a Phase I-Ib/II, open-label, multi-center study of sabatolimab in adults with advanced malignancies. The registry lists non-randomized allocation and a parallel model. Enrollment was 252. The study began on 2015-11-23, had a primary completion date of 2022-08-30, and is listed as terminated.

The study included Phase I dose escalation, Phase Ib dose escalation, a dose-ranging part, and Phase II groups. The Phase I groups tested sabatolimab alone at different doses and dosing intervals, including groups in Japan and the rest of the world. Phase Ib groups combined sabatolimab with spartalizumab at differing doses and schedules. The dose-ranging part included sabatolimab alone. Phase II included a combination group for non-small cell lung carcinoma and a combination group for melanoma. The listed Phase Ib sabatolimab plus decitabine arm and Phase II sabatolimab-alone arm were not opened for enrollment.

Open-label means that masking was not used; the registry lists masking as none. Non-randomized allocation means that participants were not assigned to groups at random, so the groups should not be treated as exchangeable arms of a controlled trial. The many dose and disease groups reflect the study's staged, exploratory structure rather than a single direct comparison.

The registry lists a broad set of secondary outcomes, including response, progression-free survival, duration of response, overall survival, and serum concentration measures. It records 35 outcome measures, but no analyses in the record. The statistical plan, dose-selection decision rules, sample-size justification, interim analyses, and reasons for termination 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: The screening period began once patients had signed the study informed consent. Screening evaluations were performed within 28 days prior to the first dose of study medication. After screening, the treatment period started on Cycle 1 Day 1.

Only the first 6 of 30 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All patients in Phase I-Ib and Dose ranging Part who received at least one dose of assigned single agent sabatolimab, or at least one full or partial dose of assigned combination sabatolimab+spartalizumab

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.

Only the first 6 of 31 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Phase I-Ib and Dose Ranging Part: Number of Participants With Adverse Events (AEs) and Serious Adverse Events (SAEs) During the On-treatment Period, measured over From first dose of study medication up to 30 days after last dose, with a maximum duration of 2 years for sabatolimab and 5 years for sabatolimab in combination with spartalizumab.

The primary endpoint counts participants who experienced adverse events (AEs) and serious adverse events (SAEs) during the on-treatment period in the Phase I-Ib and Dose Ranging parts. The specified population includes patients who received at least one dose of assigned single-agent sabatolimab or at least one full or partial dose of assigned sabatolimab plus spartalizumab.

The on-treatment period begins on the day of the first study-treatment administration and continues through 30 days after the last administration. The registry describes a maximum duration of 2 years for sabatolimab and 5 years for sabatolimab in combination with spartalizumab. Reported AEs include events of any seriousness and qualifying changes from baseline in vital signs, electrocardiograms, and laboratory results that were reported as AEs. Severity was graded using CTCAE version 4.03: Grade 1 is mild, Grade 2 moderate, Grade 3 severe, Grade 4 life-threatening, and Grade 5 death related to an AE.

Counting participants with events is a useful descriptive safety measure in an early-phase study: it records whether events occurred during exposure and can help characterize the observed safety experience. It does not, by itself, establish that treatment caused an event. Nor does a count describe the timing, duration, recurrence, or clinical course of each event. Those details, and any formal inferential analysis of safety, are not reported in the registry record. The endpoint also does not measure comparative efficacy.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a count of participants with events. This is a binary participant-level outcome for each event definition: a participant either had at least one qualifying event during the defined period or did not. The registry also reports counts for serious AEs and other AEs. These categories describe event seriousness, and they are not a measure of event severity alone. Severity grading is described separately using CTCAE version 4.03.

In general, an event count is often presented alongside the number of participants who could have experienced the event. That denominator matters: the same count can convey different information in groups of different sizes. The posted tables and plots make the group sizes and counts available for inspection. They do not turn the groups into randomized comparisons or adjust for differences in participant characteristics, disease groups, dose, schedule, or time under observation.

In general, an adverse event recorded after treatment starts is not necessarily caused by treatment. A safety summary can describe events observed during a specified period, while causal attribution requires more than temporal sequence. This distinction is especially relevant in a study of advanced malignancies, where the registry describes a broad patient population and multiple study groups. The record does not report a formal causal analysis for the posted counts.

The observation window matters when interpreting an event count. Here, the on-treatment period starts at first administration and runs through 30 days after last administration. The registry gives maximum durations of 2 years for sabatolimab and 5 years for the combination with spartalizumab. Participants with different treatment duration may have different opportunities for events to be observed. A simple count does not account for differing time at risk. The record does not report a time-adjusted safety analysis for these data.

In general, a confidence interval expresses uncertainty around an estimated quantity under the assumptions of its statistical method. No confidence intervals or inferential analyses are reported for the primary outcome in the registry record. The available counts therefore describe the observed groups but do not quantify statistical uncertainty in a formal way. A lack of a reported interval should not be mistaken for certainty.

The study's non-randomized design is central to the statistical interpretation. Random assignment, when used, can help balance measured and unmeasured prognostic factors on average across treatment groups. Here, allocation was non-randomized and masking was absent. The groups may differ in ways related to disease, dose selection, study phase, or other characteristics. Consequently, comparing their raw event counts cannot isolate the effect of a treatment or dose.

The registry identifies many secondary outcome types, including response assessed under RECIST v1.1 and irRC, progression-free survival, duration of response, overall survival, and pharmacokinetic measures. These outcomes answer different questions and have different statistical requirements. In general, response proportions summarize categorical outcomes, while time-to-event analyses account for event timing and censoring. The registry lists no analyses, so the posted record does not show which methods were used, how censoring was handled, or what results were obtained for these secondary outcomes.

The sample-size section is a planning illustration based on an observed effect, with two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. For those illustrations, the effect to detect is specified as a fraction of the observed effect; time-to-event illustrations use a total event count, while other illustrations give participants per arm. These figures are planning scenarios, not observed power and not the study's original design calculation. They do not remedy the absence of a randomized comparison, and they should not be interpreted as evidence that the actual trial had a particular probability of detecting an effect. The trial's own sample-size justification is not reported in the registry record.

In general, statistical interpretation also depends on prespecification: which outcomes and comparisons were planned, how missing data would be handled, and whether analyses were adjusted for multiple testing. Those details are not reported in the registry record. It is therefore not possible to verify from the record whether the posted summaries correspond to a prespecified analysis plan.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 35 outcome measures, spanning safety, response, progression, survival, and pharmacokinetic outcomes, but records no analyses. The statistical plan, including whether outcomes were prioritized or how multiple comparisons were handled, is not reported in the registry record.

In general, testing or estimating many outcomes creates more opportunities for apparently notable results to arise by chance, particularly when several comparisons are examined. A prespecified analysis strategy can clarify which results are confirmatory and how uncertainty should be interpreted. Here, the registry record does not describe such a strategy, and no analysis results are posted for the listed secondary outcomes. Therefore, multiplicity adjustments or a hierarchy of outcomes cannot be assessed.

Missing-data procedures are also not reported in the registry record. The primary endpoint population is defined by receipt of at least one dose, but this definition does not explain how missing AE information, incomplete follow-up, or discontinuation was handled. The record does not state whether participants with incomplete observation were included in a particular way, or whether the extent and patterns of missing data were assessed.

Analysis populations are described for the primary endpoint as patients in Phase I-Ib and the Dose Ranging part who received the specified treatment exposure. The posted adverse-event groups also include Phase II entries and survival-period entries. The record does not explain how these listings relate to a single analysis population or provide a detailed accounting of participant flow across all outcomes. These gaps limit assessment of whether denominators and follow-up are consistent across summaries.

11. 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: small intestinal obstruction (33.3%), clostridium difficile colitis (33.3%), escherichia urinary tract infection (33.3%), pleural effusion (33.3%), pyrexia (22.2%).

The serious adverse-event figures are group-level counts of participants with at least one SAE, alongside the number at risk in each listed group. The tables and plots should be read within each group's stated dose, schedule, and study part. The study was not randomized, so comparisons between these figures are descriptive and not like-for-like control comparisons.

The term-level summary lists small intestinal obstruction, Clostridium difficile colitis, Escherichia urinary tract infection, and pleural effusion at 33.3%; pyrexia and back pain are listed at 22.2%. The record does not report the denominator or further calculation details for these percentages. They are not necessarily percentages of all participants in the study, and they should not be treated as directly comparable group-specific risks.

SAE status and severity grade are distinct concepts. The registry describes AE severity using CTCAE version 4.03, whereas its primary outcome counts participants with AEs and SAEs. A participant may have more than one event, but the participant count does not convey the total number of episodes. Detailed event-level timing and attribution are not reported in the registry record.

Deaths shown in group summaries, including survival-period rows, are not automatically deaths caused by an AE. The posted counts alone do not establish cause or treatment attribution. Interpret the figures as the registry's descriptive safety reporting, not a causal comparison.

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

The posted data describe adverse events observed among treated participants in a complex, non-randomized study. They show how many participants were recorded with serious and other AEs in the listed groups, and the registry identifies several serious-event terms with associated percentages. This is useful as a descriptive account of the study's reported safety experience.

These results do not establish that sabatolimab, spartalizumab, or their combination caused a particular event. They also do not establish that one dose or schedule is safer than another. Participants were not randomized to groups, groups differed by study part and disease setting, and the record does not provide an adjusted comparative analysis. A numerical difference between group summaries would therefore not isolate a treatment effect.

The primary outcome does not answer whether treatment improves response, delays progression, or extends survival. Although the registry lists these as secondary outcomes, it reports no analyses. No treatment-effect estimate, confidence interval, or statistical test for those outcomes is available in the record. The absence of posted analyses is not evidence that an outcome was favorable, unfavorable, or unchanged.

The results also do not show how often events occurred over a standardized amount of follow-up. The endpoint covers a defined on-treatment period, but participant exposure and observation may vary; the registry does not report a time-adjusted safety analysis. Nor do participant counts fully describe event severity, duration, recurrence, or clinical course.

In general, early-phase studies can provide descriptive information about tolerability and help characterize outcomes in the groups studied. That general purpose does not make the observed event counts causal or broadly generalizable. The registry record supports careful description of its enrolled participants and reported events, with substantial limits on comparative interpretation.

13. Why This Trial Matters Statistically

This trial illustrates several statistical issues that are common in early-phase clinical research. Its multiple stages and dose schedules produce a collection of groups rather than a single randomized comparison. The structure is relevant when interpreting safety tables: each count belongs to a particular treatment context, and comparisons across those contexts can be confounded by differences in participants, disease groups, and treatment exposure.

It also demonstrates why denominator and follow-up information matter. A participant count provides a direct description of who experienced an event, but counts alone do not account for group size or time at risk. The registry defines the AE observation period and identifies participant counts by group, yet does not report a time-adjusted analysis or detailed accounting of follow-up for the summaries.

The record further shows the distinction between collecting outcomes and reporting analyses. It lists numerous secondary outcomes, including response and survival measures, but records no analyses. An outcome name in a registry is not itself an estimate or a result. Readers need the analysis population, statistical method, uncertainty, and handling of missing information to assess such results; these details are not reported.

Finally, planning illustrations based on an observed effect can be useful for showing how assumptions influence hypothetical sample sizes. They are not observed power and do not reconstruct the original trial's design rationale. The trial's sample-size justification is not reported in the registry record. Keeping these distinctions clear helps prevent descriptive safety summaries and planning exercises from being mistaken for controlled evidence of treatment effects.

14. 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 registry describes a non-randomized, open-label study with many treatment groups, doses, schedules, and study parts. These features make direct comparisons between groups difficult. There is no randomized control group in the record, so the reported group counts cannot be interpreted as a controlled estimate of the effect of treatment.

The primary endpoint is a participant count over an on-treatment period. It does not communicate event rates per unit of time, event recurrence, or detailed event trajectories. The registry gives maximum treatment durations but does not report an analysis that accounts for differing observation time. The exact denominator and calculation details for the percentages attached to the listed serious-event terms are not reported in the registry record.

Although deaths are listed for treatment groups and separate survival-period rows, those counts do not establish that deaths were caused by an AE. Event attribution and detailed cause-of-death information are not reported. The registry record also does not report analyses for the listed response, progression, survival, or pharmacokinetic outcomes.

Further limitations include the absence of a reported statistical analysis plan, sample-size justification, interim-analysis details, multiplicity strategy, and missing-data methods. The record does not explain the reasons for termination. These omissions restrict assessment of prespecification, completeness, and the intended interpretation of the outcomes. The enrollment figure of 252 does not by itself establish the number included in each endpoint analysis.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Phase I-Ib/II Study of MBG453 as Single Agent and in Combination With PDR001 in Patients With Advanced Malignancies — NCT02608268.
• Registry record last updated 2023-12-06; status last verified 2023-12.
• Results first posted 2023-12-06.
• 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.

18. Record Summary

This is a Phase I-Ib/II, open-label, non-randomized, parallel study of sabatolimab alone and in combination with spartalizumab in adults with advanced malignancies. The registry lists enrollment of 252 and a status of terminated. It includes dose-escalation, dose-ranging, and Phase II groups; the listed decitabine combination and Phase II sabatolimab-alone arms were not opened for enrollment.

The primary endpoint counts participants with AEs and SAEs during the on-treatment period. The posted safety summaries are descriptive and cover groups with different doses, schedules, and disease settings. The registry lists 35 outcome measures but no analyses. Thus, the record supports a limited reading of reported safety counts and does not provide an inferential or randomized comparison of treatment effects.

Details on the statistical plan, sample-size justification, missing-data handling, multiplicity, and reasons for termination are not reported in the registry record.