Nivolumab + Cabiralizumab + Gemcitabine in Patients With Stage IV Pancreatic Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Nivolumab + Cabiralizumab + Gemcitabine in Patients With Stage IV Pancreatic Cancer (NCT03697564), a na phase 2 study. The primary outcome is Number of Participants With Progression Free Survival (PFS).

2. Clinical Question

The official title calls the study a randomized phase II trial, while the registry design fields describe a single-group model, list allocation as NA, and identify no masking. The record also lists an overall survival outcome as secondary. These design details matter to the clinical question: the stated goal mentions comparison with gemcitabine alone, but the registered group information does not document a concurrent comparator. Without numerical endpoint findings, there is no reported estimate of progression-free survival at 6 months, and without a registered control arm there is no observed between-group contrast to quantify. The record therefore documents the intended treatment and outcome, but not evidence from which to establish comparative effects.

3. Trial Design

The registry identifies NCT03697564 as a phase II study of gemcitabine, nivolumab and cabiralizumab in patients with Stage IV pancreatic cancer. Its official title describes an open-label randomized trial in patients achieving disease control after first-line chemotherapy. Other design fields, however, list allocation as NA, masking as none, and the model as single group. The recorded arm is gemcitabine plus nivolumab plus cabiralizumab. This leaves a discrepancy between the title's randomized description and the single-group design information; the record does not explain it.

The registry lists enrollment as 2 and status as completed. The recorded study start is 2019-10-31 and primary completion is 2023-04-12. These are the registry entries, not outcome findings. Enrollment is not necessarily the denominator for every outcome: the record does not report the analysis population or endpoint-specific denominators. No conclusions about participant flow, follow-up completeness, or the number assessed for the primary endpoint can be drawn from the available information.

The stated purpose is to assess whether the combination can give prolonged disease control compared with gemcitabine alone. Yet only the combination group is listed. A comparator arm, randomization procedure, allocation ratio for the actual trial, and treatment-assignment details are not reported in the registry record. The sample-size planning description refers to equal allocation between two arms, but that assumption does not establish that the registered study actually enrolled two arms. The record does not resolve this difference.

The primary endpoint is the number of participants with progression-free survival at 6 months. Overall survival is listed as a secondary outcome. The record indicates 2 outcome measures and 0 analyses, but does not provide numerical primary or secondary outcome results in the registry record. Consequently, the completed status should not be read as evidence that either endpoint showed a particular result. Completion describes the study status; it does not supply the missing estimates or their precision.

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

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 Number of Participants With Progression Free Survival (PFS), measured over 6 months.

The primary endpoint is the number of participants with progression-free survival (PFS) at 6 months. PFS generally concerns the time until disease progression or death; in this record, the outcome is expressed as a participant count at a specified follow-up point rather than as a reported time-to-event estimate. The registry description says target lesions are assessed using RECIST v1.1 and MRI. It defines complete response as disappearance of all target lesions, and partial response as a decrease of >=30% in the sum of the longest diameter of target lesions. It describes overall response as complete response plus partial response.

As a general statistical matter, a fixed-time PFS outcome can provide a clinically interpretable snapshot of how many participants remain progression-free by the specified time. Its interpretation depends on the rules for progression and death, the assessment schedule, follow-up completeness, and the population included in the analysis. Those details, beyond the definitions quoted in the record, are not reported in the registry record.

The endpoint is useful for describing disease control over follow-up, but a count alone is most informative when its denominator and uncertainty are also reported. The registry record does not report a numerical six-month PFS result, an endpoint denominator, or a confidence interval. It therefore does not support an estimate of the proportion progression-free, nor a judgment about precision. The RECIST response definitions describe lesion changes; they are not numerical findings for this trial.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical interpretation begins with identifying what kind of outcome was planned and what information is actually available. The primary outcome is a count of participants with PFS at 6 months. Although PFS is often analyzed as a time-to-event outcome, this registry endpoint is stated as a participant count at a fixed time. The record does not give a time-to-event estimate, event count for PFS, or a model-based comparison. Those quantities should not be inferred from the endpoint title.

What does a count at a fixed time tell us?

A count describes how many participants met an outcome definition. To interpret it as a rate, the count needs a suitable denominator: the number of participants assessed or otherwise included in the relevant analysis. The registry record does not report the six-month PFS count or its denominator. The enrollment entry is 2, but enrollment alone does not establish who had sufficient follow-up, who was assessed, or who was included in the endpoint analysis. A proportion or rate cannot be recovered without the needed outcome information.

As a general statistical explanation, a confidence interval gives a range of values compatible with the data under the method's assumptions. It conveys uncertainty around an estimate; it is not a guarantee that the true value lies within the particular interval. No interval for PFS or OS is reported in the registry record. Therefore, neither the size of an estimated effect nor its precision can be evaluated.

Why does the study structure matter?

A comparison of treatments requires outcomes from groups that can be contrasted. The stated purpose refers to combination therapy compared with gemcitabine alone. Yet the registry design fields identify a single-group model and list the combination arm, with allocation recorded as NA. The record does not list a gemcitabine-alone arm or report a between-group estimate. The official title's reference to a randomized trial does not resolve that mismatch. Without documented comparison data, no treatment difference, relative effect, or comparative uncertainty can be interpreted.

Random allocation, when used and properly implemented, can help make groups comparable at baseline by distributing known and unknown prognostic factors. That is a general design principle, not a claim about this study's implementation. The registry record does not report allocation procedures or explain the single-group fields. It is therefore safer to describe the registered information as it stands rather than assume a randomized comparison took place.

What does the sample-size description mean?

The planning description says its illustrations use an observed effect, a stated fraction of that effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also says time-to-event planning gives a total number of events, while other outcomes give participants per arm. These are planning assumptions, not a description of the registered single-group structure. The record does not report the numerical planning illustrations in the registry record for this analysis, and it does not say that those assumptions were the trial's own design.

As a general matter, sample-size planning asks how much information a future study would need under specified assumptions. An illustration based on an observed effect is conditional on that effect and on the assumed design and analysis. It is not observed power, and it is not evidence that the original study was adequately sized. If an observed effect is not reported, the basis for an effect-dependent planning illustration cannot be independently evaluated from the registry record. The planning assumptions should not be mistaken for evidence of two actual arms in this trial.

Over-dispersion is a general concern when count outcomes vary more than a chosen statistical model expects. If a model assumes less variability than the data have, standard errors can be too small and uncertainty can be understated. The registry record does not report a count model, a dispersion parameter, or a model-based result for the primary endpoint. Because the endpoint is presented as a participant count and numerical results are absent, there is no basis for assessing over-dispersion in this study.

How should response criteria be understood?

The endpoint description defines complete response as disappearance of all target lesions and partial response as a decrease of >=30% in the sum of the longest diameter of target lesions. It states that overall response is complete response plus partial response, with assessment by MRI under RECIST v1.1. These criteria specify how lesion response categories are defined; they do not themselves say how many participants met a category. The record does not report participant-level response counts or the procedures for handling unevaluable assessments.

Progression-free survival and tumor response are related but distinct concepts. Response criteria classify changes in measured lesions, whereas PFS concerns remaining free of progression or death through follow-up. The registry's primary endpoint description does not report a PFS analysis method, censoring rules, or the handling of death and missing assessment times. Those statistical details are not reported in the registry record. General methodological explanations cannot fill in trial-specific rules that are absent.

What can be inferred from the analysis entries?

The record lists 0 analyses. This indicates that analysis entries are not present in the registry entries used; it does not prove that no statistical work occurred. No analysis population, adjustment strategy, hypothesis test, confidence interval method, or missing-data approach is reported. There is consequently no method-specific result to reproduce or compare with the stated endpoint. A sound reading distinguishes an absent registry analysis from a conclusion that an analysis was never performed.

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("Gemcitabine +Nivolumab + Cabiralizumab")
n      <- c(2)
events <- c(1)
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  Gemcitabine +Nivolumab + Cabiralizumab  2  1  50.0  1.3  98.7  9.5  90.5  9.4  90.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary PFS outcome at 6 months and an OS secondary outcome. It does not report a multiplicity strategy, such as an ordering of hypotheses or an adjustment for testing more than one outcome. When several outcomes are tested, multiplicity can increase the chance of at least one apparently positive result by chance; methods to manage that issue depend on the planned analyses. This is a general statistical consideration, not a statement that the study used any particular procedure. With no analysis entries or numerical findings reported, the consequences of multiplicity cannot be assessed for this trial.

Missing-data handling is also not reported in the registry record. For the PFS outcome, relevant details could include how participants without an assessment at 6 months were treated, how incomplete follow-up was addressed, and how progression or death was timed. The record does not describe those rules. It also does not report an analysis population, the number assessed for each outcome, or participant flow. The enrollment entry alone cannot establish the denominator used for an endpoint.

As a general principle, missingness can affect both the estimate and its uncertainty, particularly when the reasons for missing observations are related to prognosis or outcome. Statistical approaches rely on assumptions, and the assumptions and sensitivity analyses need to be stated to judge their plausibility. No such methods or sensitivity analyses are reported. The absence of this information should not be taken as proof that data were missing; it means the registry record does not describe the issue or its handling.

The record lists 0 analyses and does not provide results for the primary or secondary outcome. It therefore does not support a judgment about multiplicity control, missing-data bias, or robustness to alternative analysis choices. These are reporting limitations, not evidence that a particular bias occurred.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: death (50.0%).

The safety record lists a group titled Gemcitabine +Nivolumab + Cabiralizumab. For that group, 1 participant is listed with a serious adverse event among 2 at risk, and 2 participants are listed with other adverse events among 2 at risk. The record also lists 1 death. Death is the top serious term, reported at 50.0%. These entries are descriptive counts and should be read as registry safety information, not as comparative estimates.

The counts do not establish that treatment caused an event or death. The registry entries do not report a causal attribution, event timing, event-level clinical details, or a control group with which to compare the counts. The listed at-risk numbers are 2 for the serious and other adverse-event entries, but the record does not explain the detailed analysis or reporting conventions. The death entry and serious-term listing should not be treated as proof that the death was treatment-related.

Adverse-event figures are most informative when their definitions, observation period, severity criteria, and attribution are available alongside the number at risk. Those details are not reported in the registry record used for this interpretation. The counts do not support a comparative safety conclusion or a claim about the frequency of risk beyond the recorded entries.

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

The record establishes that a phase II study was registered for a three-drug combination in Stage IV pancreatic cancer and that the primary outcome was a participant count with PFS at 6 months. It records the study as completed and lists OS as a secondary outcome. It also contains a small enrollment entry and descriptive safety counts. These registry entries characterize the registration; they do not answer whether the combination changed PFS or OS.

Numerical results for the primary and secondary outcomes are not reported in the registry record. There is no reported PFS count or denominator, no OS estimate, and no confidence interval or analysis result. Without an outcome estimate, its precision cannot be described. The absence of reported results should not be interpreted as a positive result, a negative result, or evidence of no effect.

The record's comparison language needs particular care. The brief summary describes a comparison with gemcitabine alone, and the official title calls the trial randomized. In contrast, the design fields state single group and allocation NA, and the listed arm is the combination. A gemcitabine-alone arm and a comparative estimate are not reported. Accordingly, the stated comparative goal cannot be treated as an observed treatment comparison based on the registered information.

The safety counts likewise describe recorded events, not causality or comparative risk. The entry of 1 death and the serious adverse-event listing do not establish that treatment caused the death. There is no listed control group for a relative safety assessment, and the registry entries do not report attribution details.

In general, a trial result is most informative when its outcome definition, analysis population, denominator, estimate, uncertainty, and analysis method can be considered together. Several of those elements are absent from the record. The appropriate conclusion is limited: the registry documents the intended clinical question and some design and safety information, but does not report numerical efficacy outcomes or a documented concurrent comparison from which to determine treatment effects.

14. Why This Trial Matters Statistically

This trial record is statistically notable because the clinical aim and the registered design fields do not align cleanly. The brief summary describes a comparison of the combination with gemcitabine alone, and the official title describes a randomized trial. The design fields instead identify a single-group model and allocation as NA, with the combination as the listed arm. This contrast illustrates why the design information, not only the title or stated aim, matters when deciding what treatment effect can be estimated.

The endpoint also emphasizes the difference between specifying an outcome and reporting its result. The primary measure is a count of participants with PFS at 6 months, with response assessment described using RECIST v1.1 and MRI. But the registry record does not report the count, its denominator, or uncertainty. A defined endpoint makes the intended question clearer; it does not by itself provide evidence about the answer.

The enrollment entry of 2 is relevant context for the scale of the study, while the endpoint-specific analysis population is not reported. This distinction is important in small studies: the recorded enrollment need not equal the population contributing to an outcome, and uncertainty cannot be judged from enrollment alone. The absent numerical results and analysis methods prevent further assessment.

Finally, the record separates descriptive safety reporting from causal interpretation. Adverse-event counts and a death entry can be useful signals to document, but without attribution details or a comparative group they do not quantify a treatment-related excess. Taken together, these features make the record a useful example of how design consistency, outcome reporting, denominators, and uncertainty shape the limits of statistical interpretation.

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 most direct limitation is that numerical results for the primary PFS outcome and secondary OS outcome are not reported in the registry record. The primary outcome is specified as a count at 6 months, but its count, denominator, estimated rate, and uncertainty are absent. The record therefore does not permit assessment of the size or precision of either outcome.

Design information is internally difficult to reconcile. The official title describes a randomized trial, while the design fields say single group and allocation NA. Only the combination arm is listed, even though the brief summary describes a comparison with gemcitabine alone. The registry record does not explain this discrepancy or document a concurrent gemcitabine-alone arm. Any interpretation that treats the registration as a completed randomized comparison would go beyond the recorded design details.

Further trial-specific limitations include the lack of a reported analysis population, endpoint denominators, participant flow, analysis methods, missing-data rules, and multiplicity strategy. The record lists 0 analyses. This is an absence of analysis entries, not proof that analysis was never conducted, but it prevents evaluation of how outcomes were handled and estimated. The primary endpoint description references RECIST v1.1 and MRI, yet details such as assessment timing and procedures for unevaluable assessments are not reported in the registry record.

The enrollment entry is 2, but the record does not say how many participants contributed to each outcome. The safety entries are likewise limited: attribution, event timing, and a comparative safety group are not reported. These limitations constrain both efficacy and safety interpretation and should be kept separate from claims about what happened clinically.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Nivolumab + Cabiralizumab + Gemcitabine in Patients With Stage IV Pancreatic Cancer — NCT03697564.
• Registry record last updated 2025-08-06; status last verified 2025-07.
• Results first posted 2025-08-06.
• 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.

19. Record Summary

NCT03697564 is recorded as a completed phase II study of gemcitabine, nivolumab and cabiralizumab in Stage IV pancreatic cancer. The primary outcome is the number of participants with PFS at 6 months, assessed under the described RECIST v1.1 and MRI framework; OS is listed as a secondary outcome. The registry records enrollment as 2 and lists a single combination-treatment group.

The record's stated comparison with gemcitabine alone is not matched by a listed control arm. Its official title describes randomization, while design fields state single group and allocation NA. Numerical PFS and OS results, analysis methods, and endpoint-specific denominators are not reported in the registry record. Safety entries include 1 participant with a serious adverse event, 2 with other adverse events, and 1 death; the record does not establish causality or comparative risk. These limitations prevent an estimate of treatment effect from the registry information.