Nivolumab With or Without Ipilimumab in Treating Patients With Resectable Liver Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Nivolumab With or Without Ipilimumab in Treating Patients With Resectable Liver Cancer (NCT03222076), a randomized phase 2 study. The primary outcome is Safety and Tolerability of Presurgical Nivolumab With or Without Ipilimumab, Defined as the Number of Participants With of Adverse Events..

2. Clinical Question

3. Trial Design

NCT03222076 is a randomized, open-label, parallel-group phase II pilot study sponsored by M.D. Anderson Cancer Center. Its title describes nivolumab with or without ipilimumab in patients with resectable liver cancer. The registry lists an enrollment of 30 and a completed status. The listed start date is 2017-09-28 and the primary completion date is 2022-09-14.

The active treatment groups were nivolumab alone and nivolumab plus ipilimumab. The nivolumab arm description specifies intravenous treatment over 30 minutes on day 1, repeated every 2 weeks for up to 3 cycles, followed by liver surgery on day 1 of week 7. The combination arm adds intravenous ipilimumab over 90 minutes on day 1. Both descriptions refer to treatment in the absence of disease progression or unacceptable toxicity. The registry's descriptions of treatment after surgery are cut off, so the full planned treatment course is not reported in the registry record.

The record also lists Arm C as nivolumab plus ipilimumab, but its posted enrollment is 0.0. The primary outcome population is described as Cohort C, discontinued due to non-accrual. The record does not explain the relation between that cohort label, the listed Arm C, and the posted results for Arms A and B. The listed overall enrollment and the arm-level outcome counts therefore should not be treated as interchangeable descriptions of the analysis population.

The study's stated purpose was to examine side effects and how well the treatments work in surgically resectable liver cancer. The posted primary outcome concerns safety and tolerability. Secondary outcomes include objective response rate, time to progression, and progression-free survival. The registry record does not report a detailed statistical plan, sample-size justification, or the allocation process beyond identifying the study as randomized.

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: 3 excluded: 1 did not meet inclusion criteria; 2 declined to participate The study was amended in June 2019 to remove/exclude the enrollment of Arm C .

Overall Study

Analysis population for the primary outcome, as defined in the registry: Cohort C that was discontinued due to non-accrual.

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: 2 of the patients (1 in the Nivolumab arm and 1 in the Nivolumab + Ipilimumab arm) was not included in the analysis of PFS because their surgical cancellations were unrelated to progressive disease.

6. Primary Endpoint

The registry lists the primary outcome as Safety and Tolerability of Presurgical Nivolumab With or Without Ipilimumab, Defined as the Number of Participants With of Adverse Events., measured over Up to 5 years.

The primary endpoint is described as safety and tolerability of presurgical nivolumab with or without ipilimumab, measured as the number of participants with adverse events. The outcome description also specifies the frequency of adverse events, serious adverse events (grade 3 or higher), and laboratory abnormalities, with laboratory findings assessed by worst grade. Grading is identified as using the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0. The timeframe is up to 5 years.

Safety outcomes are used to characterize harms observed during treatment and follow-up. Counts can communicate how many participants had recorded events, while serious-event and laboratory-abnormality summaries add information about event type and severity. They do not, by themselves, establish that a treatment caused an event: causal interpretation depends on the study design, event definitions, timing, assessment, and completeness of reporting.

The posted primary arm values are 13.0 for nivolumab and 14.0 for nivolumab plus ipilimumab. The comparison reports a risk difference of 0.0. Separate adverse-event summaries list serious events, other events, and deaths; those summaries are not identical in form to the primary endpoint count. The registry record does not explain how the primary endpoint count was derived from those categories, or how the Cohort C population description applies to the arm-level values. These distinctions matter when interpreting the endpoint as a safety measure.

7. Results Posted in the Registry

• Arm B: Nivolumab Plus Ipilimumab: 100.0% (14/14) versus 100.0% (13/13) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0), risk ratio 1.00 (1.00–1.00); unadjusted two-proportion z-test p = n/a.

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.

Objective Response Rate

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Arm B (Ipilimumab, Nivolumab): 0.0% (0/14) versus 23.1% (3/13) on control; risk difference -23.1 percentage points (95% CI -46.0 to -0.2); unadjusted two-proportion z-test p = 0.057.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm B (Ipilimumab, Nivolumab) and the control. The interval gives the range of effects that remain compatible with the data.

The registry identifies objective response rate, time to progression, and progression-free survival as secondary outcomes. It posts an objective-response comparison, but results for time to progression and progression-free survival are not reported in the registry record. The secondary evidence is therefore incomplete, and it does not provide a broad picture across all of the listed outcomes.

The available secondary comparison is not a replacement for the primary safety endpoint. It concerns a different outcome and is reported with an unadjusted p-value. The registry record reports no analyses for this outcome beyond the posted comparison. There is no reported multiplicity adjustment or analysis plan explaining how the secondary outcomes were to be interpreted together.

General statistical explanation: when multiple outcomes or comparisons are examined, the chance of seeing a seemingly notable result by chance can increase. Adjustment procedures or a prespecified hierarchy can help control how findings are interpreted, but they require a stated plan. Without one reported in the registry record, an unadjusted secondary result should be treated as exploratory rather than as confirmatory evidence.

The secondary results do not resolve the uncertainty in the primary safety comparison. They also cannot establish that a pattern applies consistently across response and time-to-event outcomes when results for the latter are not reported. A cautious reading is that the registry contains a limited secondary finding alongside unreported listed outcomes, not a complete and coordinated body of evidence about treatment effects.

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

• Objective Response Rate
• Time to Progression
• Progression Free Survival (PFS)

9. Statistical Interpretation

10. Statistical Methods Explained

General statistical explanation. A randomized comparison assigns participants to treatment groups by chance, which can support a fair comparison when allocation and follow-up are carried out as planned. Open-label means that treatment assignment was not masked. Awareness of assignment can affect treatment decisions, reporting, or assessment, particularly for outcomes that involve judgment. Randomization does not remove every possible source of bias, and it does not make a small study large or precise.

The registry identifies this study as randomized, open-label, and parallel-group. The record does not report the randomization procedure, allocation concealment, stratification factors, or a detailed statistical analysis plan. These details are useful because they explain how assignment was implemented and how the investigators intended to analyze outcomes. Their absence from the registry record does not establish that no such procedures existed; it means they are not reported in the registry record.

What does a risk difference describe?

General statistical explanation. A risk difference compares the proportion of participants with an event in one group with the proportion in another group. A value of zero denotes equal observed proportions under the stated outcome definition. The measure is expressed on the absolute scale, so it describes the difference in event occurrence rather than a relative comparison. Its interpretation depends on a clear event definition, a defined analysis population, and complete outcome ascertainment.

The registry reports a primary risk difference of 0.0. The associated interval is also reported as 0.0 to 0.0. The record does not describe the calculation method or explain why the interval has no displayed width. Without those details, the reported interval should not be treated as a typical estimate of sampling uncertainty. The record also does not report a primary-endpoint p-value.

What does a confidence interval add?

General statistical explanation. A confidence interval presents a range of estimates compatible with a statistical procedure and the observed data. A narrow interval can indicate greater precision than a wide interval, while an interval that includes values on both sides of a null difference may be compatible with more than one interpretation. An interval is not a guarantee that the true value lies within its limits, nor does it show the probability that a particular effect is true.

For the secondary objective-response comparison, the registry reports an unadjusted p-value of 0.057 and an interval. The interval endpoints are not reproduced because the posted values exceed the permitted 3-decimal display precision. That comparison is not the registered primary safety outcome. A p-value measures how unusual data would be under a specified null model and assumptions; it does not measure the size or clinical importance of an effect. The registry record does not report the full analysis method or adjusted secondary analyses.

Why does the analysis population matter?

General statistical explanation. An outcome count is interpretable only in relation to the participants included, the time during which they were observed, and the rules for handling missing information. Analyses based on participants as randomized, participants who started treatment, or participants with available outcome data can answer different questions. A clear account of exclusions and missing data is needed to understand which question a reported estimate addresses.

The primary outcome population is labelled Cohort C, discontinued due to non-accrual, while the registry posts arm-level results for Arms A and B and lists 0.0 in Arm C. The record does not reconcile these labels or state the analysis population for each result. It does not report missing-data handling, exclusions, or whether all randomized participants were included in every analysis. That limits interpretation of both the participant counts and their comparison.

How should safety counts be distinguished?

General statistical explanation. Safety reporting commonly separates participants with any event from participants with serious events, event counts, severity grades, and deaths. These summaries are not interchangeable. A participant can experience more than one event, so the number of events need not equal the number of affected participants. Severity and seriousness are also separate concepts unless the study defines them otherwise.

The registry's primary endpoint describes participants with adverse events and also mentions serious adverse events and laboratory abnormalities. A separate summary gives counts for serious events, other affected participants, and deaths. The registry does not report event-level detail, timing, attribution, or a reconciliation between the primary count and the separate categories. The values should therefore be read according to the specific category named rather than combined into a new total.

General statistical explanation. Over-dispersion matters for outcomes analyzed as event counts when their variability is greater than a model expects. A model that assumes too little variability can produce intervals that are too narrow and tests that are too optimistic. This issue is relevant to count-data methods in general, but the posted primary outcome is a count of participants with an event, not a reported event-rate model. The registry record does not report a model choice or an over-dispersion assessment.

The record lists an enrollment of 30, but the posted arm-level counts differ from that overall enrollment figure and Arm C has no posted participants. The registry does not explain whether the enrollment field refers to a planned target, an actual enrollment, or another registry convention. The record also does not report a sample-size justification. Consequently, the listed enrollment should not be treated as proof that the study had adequate precision for a comparative safety or efficacy conclusion.

11. Confidence Intervals and Precision

Precision is limited by the scale and completeness of the posted results. The primary comparison is reported as a risk difference of 0.0 with a confidence interval from 0.0 to 0.0. Because the interval has no displayed width, it does not convey the uncertainty readers would ordinarily use to judge how precisely a difference was estimated. The registry record does not describe the interval method or explain this result.

The posted primary arm counts are 13.0 and 14.0. The overall enrollment field is 30, and the listed third arm has enrollment 0.0. These entries describe different elements of the record; the registry does not explain their relationship to the analysis population. The primary outcome description refers to Cohort C, discontinued due to non-accrual, adding another population-label uncertainty. Without a clear denominator and analysis-population description, precision cannot be assessed fully from the displayed primary comparison.

The objective-response comparison has a reported unadjusted p-value of 0.057 and an interval whose limits are not repeated because they exceed the required display precision. A p-value does not show how precisely an effect has been estimated on its own, and this secondary comparison is not a substitute for a primary safety analysis. The registry record does not report a confidence interval method, adjusted analyses, or a sample-size justification.

General statistical explanation: interval width reflects both the amount of information in the data and the method used to estimate uncertainty. Small studies often leave substantial uncertainty, especially for uncommon outcomes, but no new precision estimate can be inferred from the record without making additional assumptions. The posted numbers should therefore be treated as the registry's reported summaries, not as a complete account of uncertainty.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary safety endpoint and secondary outcomes concerning objective response, time to progression, and progression-free survival. It does not report a statistical analysis plan describing how these outcomes were prioritized, whether particular comparisons were prespecified, or whether any adjustment for multiple testing was planned. The objective-response p-value is explicitly unadjusted. Without a reported multiplicity strategy, secondary findings carry less confirmatory weight than a single prespecified primary analysis would ordinarily carry.

General statistical explanation: multiplicity arises when investigators make several statistical comparisons. Even if each comparison uses the same significance threshold, considering many results can increase the chance that at least one appears notable by chance. A prespecified hierarchy, adjustment, or clear exploratory designation helps readers understand the intended interpretation. The registry record does not state which approach was used.

Missing-data handling is also not reported in the registry record. The record does not say how participants without an observed endpoint, withdrawn participants, or participants with incomplete follow-up were handled. It does not report the analysis population for each endpoint or provide a participant-flow account that reconciles the overall enrollment field, the posted arm counts, the empty Arm C, and the Cohort C population description.

The registry record does not report whether analyses followed randomized assignment, included only treated participants, or used another analysis population. This matters because exclusions or incomplete observation can affect both the estimated outcome and its precision. The absence of these details should not be taken as evidence that no procedures were used; it means readers cannot assess them from the registry record.

Sample-size justification, interim analyses, and stopping rules are not reported in the registry record. The study is described as a pilot, but that label alone does not specify its intended precision or inferential role. The available information is insufficient to determine how the investigators planned to address missingness or multiplicity, so secondary results should be interpreted as limited and exploratory.

13. 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 1% reporting threshold.

Most frequent non-serious adverse event terms

The separate safety summary reports no serious adverse events in either active treatment group. It lists other adverse events affecting 10 participants receiving nivolumab and 12 receiving nivolumab plus ipilimumab. Death counts are 0 and 4, respectively. These are separate descriptive categories in the registry record; they should not be added together or treated as if they measure the same outcome.

The record does not state that the deaths were treatment-related, and the counts alone cannot establish causation. It also does not provide event descriptions, timing, attribution, or a detailed account of follow-up. The primary endpoint count and the separate safety categories are not reconciled in the registry record. Readers should therefore distinguish the posted primary outcome from the separate serious-event, other-event, and death summaries, and avoid drawing causal conclusions from these counts alone.

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

The posted data describe safety and secondary outcomes recorded for a randomized, open-label phase II pilot trial of presurgical nivolumab alone or in combination with ipilimumab. The primary safety comparison is reported as a risk difference of 0.0, with an interval from 0.0 to 0.0. Separate safety summaries report no serious adverse events in either active group, other adverse events in both groups, and deaths in the combination group. These are registry summaries, not a complete event-level account.

The primary result does not establish that the treatments have equivalent safety profiles. The interval's reported zero width, the lack of a stated method for its calculation, and uncertainty about the analysis population limit what can be inferred. Equal reported comparison estimates do not rule out differences in event type, timing, severity, or outcomes not captured by the primary count.

The secondary results do not establish a treatment benefit or lack of benefit. The registry posts an unadjusted objective-response comparison, while results for time to progression and progression-free survival are not reported in the registry record. The record also does not describe a multiplicity plan or provide a complete statistical analysis plan. These limitations make the secondary evidence exploratory.

The record does not report enough information to attribute deaths to treatment, reconcile the primary count with other safety categories, or determine how missing observations were handled. It also does not resolve the Cohort C population label against the posted arm results. The findings should be understood as the specific summaries entered in the registry, with important uncertainties about definitions and analysis populations, rather than as a definitive comparison of the two treatment strategies.

15. Why This Trial Matters Statistically

This trial illustrates why a registry result needs to be read alongside its endpoint definition, population description, and reporting structure. The primary outcome is framed as a participant count for adverse events, while the registry also gives separate summaries for serious events, other events, and deaths. Those categories answer different questions and cannot be collapsed into a single measure of safety.

The record also demonstrates the importance of interpreting an estimate together with its uncertainty. The primary risk difference is reported as 0.0, and its confidence interval is also 0.0 to 0.0. Without a reported method explaining that interval, it cannot provide the usual reassurance about precision. A numerical point estimate alone does not show how much uncertainty remains.

Population labels matter as well. The record lists results in Arms A and B, no enrollment in Arm C, and a primary outcome population described as Cohort C discontinued due to non-accrual. Because that relationship is not explained, readers cannot confidently map every result to the same analysis population.

Finally, the study is described as a pilot and includes secondary outcomes, but the registry does not report a sample-size justification, a full analysis plan, or a multiplicity strategy. These omissions limit confirmatory interpretation. The statistical lesson is to separate descriptive registry counts from causal or treatment-effect conclusions and to give weight to the design and reporting details that determine how much the estimates can support.

16. 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.

Several limitations specific to this record constrain interpretation. First, the primary outcome population is labelled Cohort C, discontinued due to non-accrual, while the posted primary results are associated with the active nivolumab and combination arms. The registry does not explain how those descriptions relate. It also lists Arm C with enrollment 0.0, leaving the cohort and arm structure unclear.

Second, the primary comparison has a reported risk difference of 0.0 and an interval from 0.0 to 0.0. The registry does not identify the method producing that interval or explain why it has no displayed width. No primary p-value or detailed analysis is reported. This limits assessment of the uncertainty behind the primary comparison.

Third, the separate safety summary does not reconcile with the primary endpoint count. It distinguishes serious events, other affected participants, and deaths, but the record provides no event-level detail, timing, attribution, or explanation of how categories overlap. The death counts cannot be interpreted as treatment-caused deaths from the registry information alone.

Finally, the record does not report a detailed analysis plan, missing-data procedures, sample-size justification, interim analyses, or multiplicity handling. The listed secondary outcomes do not all have posted results. The arm descriptions are also incomplete after surgery. These are reporting limitations; the registry record does not establish whether additional information exists elsewhere.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Nivolumab With or Without Ipilimumab in Treating Patients With Resectable Liver Cancer — NCT03222076.
• Registry record last updated 2023-07-10; status last verified 2023-06.
• Results first posted 2023-07-10.
• 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.

20. Record Summary

NCT03222076 is a completed, randomized, open-label, parallel-group phase II pilot study of nivolumab alone versus nivolumab plus ipilimumab in resectable hepatocellular carcinoma. The registry lists enrollment of 30. The primary endpoint concerns safety and tolerability, measured as participants with adverse events, with a timeframe of up to 5 years. The posted primary comparison is a risk difference of 0.0, with a confidence interval from 0.0 to 0.0.

The registry also lists objective response rate, time to progression, and progression-free survival as secondary outcomes. It posts an objective-response comparison, but results for the time-to-event outcomes are not reported in the registry record. Interpretation is limited by the unexplained Cohort C population label, the empty listed Arm C, the lack of a reported analysis plan and missing-data details, and the absence of a reconciliation between the primary count and separate safety summaries.