Niraparib + Dostarlimab + RT in Pancreatic Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Niraparib + Dostarlimab + RT in Pancreatic Cancer (NCT04409002), a na phase 2 study. The primary outcome is Disease Control Rate With RECIST 1.1 Criteria.

2. Clinical Question

The broad research question is whether disease control was observed among participants receiving the three-part study regimen. Disease control is defined through tumor assessment categories rather than through a direct comparison with another treatment. Complete response means disappearance of all target lesions, with pathological lymph nodes meeting the stated short-axis criterion. Partial response requires at least 30% decrease in the sum of longest diameters of target lesions relative to baseline. Stable disease means neither enough shrinkage for partial response nor enough increase for progressive disease. Progressive disease includes at least 20% increase in the sum of target-lesion diameters or the appearance of one or more new lesions.

Because this is a single-group study, the endpoint can describe the observed experience in the treated group, but it cannot by itself establish how that experience compares with what would have occurred under another treatment or without the study regimen. The registry record does not report an outcome count or disease-control percentage, so the primary question cannot be answered numerically from the posted results.

3. Trial Design

This is a phase II study listed as completed. The registry record gives a listed enrollment of 18 and identifies metastatic pancreatic cancer as the study population. The study started on 2020-07-23, and primary completion is recorded as 2022-01-19. These dates and the completed status describe the registry entry; they do not supply outcome results.

The design is single-group, with no masking and allocation recorded as not applicable. Participants were assigned to the regimen of niraparib, dostarlimab, and radiation; there is no separate control arm in the record. This design can document outcomes and adverse events among participants receiving the regimen, but it does not create a randomized treatment contrast. Differences between the study group and an external population could reflect differences in participants, care, or assessment as well as treatment.

The treatment description says that each cycle lasted 21 days. Niraparib was taken orally once a day at a predetermined dose, beginning on cycle 1 day 1 and continuing until the participant was taken off treatment. Dostarlimab was given by intravenous infusion once every cycle for as long as participants remained on study. The registry record does not provide a complete radiation schedule or a numerical dose for the drugs.

The primary outcome is disease control rate according to RECIST 1.1, assessed up to 17 months. Listed secondary outcomes include disease control rate using irRECIST, progression-free survival, overall survival, and treatment-related adverse events per CTCAE v5.0. The record lists five outcome measures in total. Numerical primary-outcome results, analysis results, and a detailed statistical plan are not reported in the registry record. In particular, the record does not state how the analysis population was defined or how missing assessments were handled.

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 Disease Control Rate With RECIST 1.1 Criteria, measured over up to 17 months.

The primary endpoint is disease control rate under RECIST 1.1. It classifies participants according to tumor assessment as having complete response, partial response, stable disease, or progressive disease, and counts those with complete response, partial response, or stable disease as having disease control. The outcome is specified as a count of participants, with assessment over a timeframe of up to 17 months.

RECIST provides standardized rules for interpreting changes in measurable lesions. In this record, complete response requires disappearance of all target lesions and reduction of pathological lymph nodes to a short axis <10 mm. Partial response requires at least 30% decrease in the sum of longest diameters from baseline. Stable disease is neither sufficient shrinkage for partial response nor sufficient increase for progressive disease. Progressive disease includes at least 20% increase in the sum of longest diameters or the appearance of one or more new lesions.

As a general statistical concept, a disease-control endpoint can capture a broader set of tumor assessment outcomes than an endpoint restricted to complete or partial response, because stable disease is also counted. A participant count is directly interpretable as the number meeting the endpoint definition, while a rate additionally requires a denominator and a clear account of who was assessed. The registry record does not report the primary endpoint count, rate, denominator used for the analysis, or uncertainty interval. Consequently, the endpoint definition is available, but its observed result and precision cannot be evaluated from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What can be inferred from the design?

The registry describes a phase II, single-group study with no masking and no allocation to a separate control group. These design features are important when interpreting any outcome estimate. A single-group study can describe outcomes in the enrolled participants, but it lacks a concurrent randomized comparison that would help distinguish treatment-associated differences from other explanations. No treatment-effect contrast between study arms is available in the registry record.

No detailed statistical analysis plan is reported in the registry record. The record does not state a primary hypothesis test, a model for the primary endpoint, an adjustment strategy, or a prespecified approach to confidence intervals. It is therefore not possible to identify the statistical procedure used for a primary-outcome analysis from the posted registry information. A binomial proportion is a general way to describe a binary outcome such as whether a participant did or did not meet a disease-control definition, but that general description should not be mistaken for a method documented as used in this study.

How should a binary endpoint be understood?

The primary endpoint is expressed as a count of participants meeting a composite disease-control definition. If a count and an analysis denominator were available, a rate could describe the fraction of the defined analysis population meeting that endpoint. The denominator matters: it could differ from the listed enrollment if participants did not receive treatment, lacked an assessment, or were excluded under an analysis rule. Which denominator was used is not reported in the registry record.

For a binary endpoint, uncertainty arises because the observed count is one realization from the underlying outcome process. A confidence interval is a general statistical summary of uncertainty around an estimated proportion. Its width depends on the amount of information and the method used. It does not account automatically for every possible source of bias, such as incomplete follow-up, outcome misclassification, or lack of a comparison group. No primary-outcome estimate or interval is reported in the registry record, so the endpoint's precision cannot be assessed.

What does the outcome definition contribute?

RECIST 1.1 supplies operational rules for categorizing tumor change. It establishes criteria for complete response, partial response, stable disease, and progressive disease, including thresholds based on target-lesion diameters and the appearance of new lesions. Standardized criteria can improve consistency in how imaging-based outcomes are classified. They do not, on their own, reveal who performed assessments, whether assessments were reviewed independently, or how disagreements were handled. Those procedures are not reported in the registry record.

The composite endpoint counts complete response, partial response, and stable disease together. General interpretation of a composite endpoint should consider whether its components are clinically and statistically similar and whether one component accounts for most of the combined result. Separate component counts would help with that interpretation, but they are not reported in the registry record. The registry also lists disease control using irRECIST as a secondary outcome; the numerical result and its analysis are not reported.

What are the sample-size illustrations?

The sample-size assumptions describe planning illustrations based on a fraction of an observed effect, with two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. The general purpose of such an illustration is to show how a hypothetical study size might change under specified assumptions about the effect to detect and the desired power. These assumptions are not evidence that the actual trial used that design: the registry describes this trial as single-group, not a two-arm comparison.

Planning illustrations are not observed power. They are also not the trial's sample-size justification unless the registry record explicitly identifies them as the planned calculation; it does not. The observed primary outcome needed to anchor an illustration is not reported in the registry record. Accordingly, the assumptions should be understood as hypothetical inputs, not as an account of the trial's original design rationale or a statement about the informativeness of its observed results. Any sample-size values should be interpreted as planning examples under the stated assumptions, not as participants actually enrolled or a guarantee of detecting an effect.

How do over-dispersion and model choice matter in general?

Over-dispersion is a general concern when observed variability is greater than a model assumes. If a model assumes less variability than is present, its standard errors may be too small and its uncertainty intervals too narrow. Whether over-dispersion is relevant depends on the outcome and the analysis model. The registry record does not report a model for the primary endpoint or an assessment of over-dispersion, so it cannot be said that such a check was performed or required for the reported analysis.

More broadly, model choice should reflect the outcome structure, design, and available data. A single-group binary count differs from a time-to-event endpoint, and the listed secondary outcomes include both disease-control and survival measures. The record does not describe how these outcomes were analyzed, how censoring was handled, or whether estimates were adjusted for participant characteristics. Those details are not reported in the registry record and should not be inferred from the outcome names alone.

What information is missing for statistical appraisal?

The record does not report the primary result, a confidence interval, a p-value, the analysis population, missing-assessment handling, or a detailed statistical analysis plan. It also does not report whether any interim analyses were conducted, how the target enrollment was justified, or how outcome assessors were organized. These omissions limit appraisal of the numerical evidence and the risk of bias. They do not change the endpoint definition or the single-group design that are recorded, but they prevent a fuller assessment of how the study's data were converted into a statistical conclusion.

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("Niraparib+Dostarlimab + Radiation")
n      <- c(15)
events <- c(0)
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  Niraparib+Dostarlimab + Radiation  15  0  0.0  0.0  21.8  0.0  20.4  0.2  20.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures, including the primary RECIST 1.1 disease-control endpoint and secondary outcomes for disease control under irRECIST, progression-free survival, overall survival, and treatment-related adverse events per CTCAE v5.0. When a study examines multiple outcomes, multiplicity is a general statistical concern: the more inferential comparisons made, the greater the opportunity for apparently notable findings to arise by chance. Possible responses include defining a primary endpoint in advance, distinguishing confirmatory from exploratory analyses, or adjusting inference. The registry record does not report a multiplicity strategy, a family of hypotheses, or whether any adjustment was planned.

The numerical results and statistical analyses for the outcomes are not reported in the registry record. The record lists zero analyses. It therefore does not permit evaluation of whether the reported outcome hierarchy was followed or whether any secondary findings were adjusted for multiple testing.

Missing data can affect an endpoint when participants lack tumor assessments or leave follow-up before the endpoint timeframe ends. In general, the handling approach can change both the analysis denominator and the estimate. The registry record does not report the number of missing assessments, reasons for missingness, rules for classifying participants without an assessment, or any imputation method. The analysis population is also not described. The listed enrollment of 18 and the safety group of 15 at risk are not enough to identify the population used for the primary endpoint.

These omissions mean that the completeness of the outcome data and the robustness of any estimate cannot be assessed from the registry record. No assumptions about interim analyses, missing-data methods, or analysis populations should be made in the absence of a report.

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: fever (6.7%), stroke (6.7%).

The safety record lists one group, Niraparib+Dostarlimab + Radiation. It records 15 participants at risk for serious adverse events and 2 with serious adverse events. It also records 15 participants at risk for other adverse events and 15 with other adverse events. The deaths field lists 14. The registry record does not explain the timing or cause of these deaths, and the count alone does not establish that they were related to treatment.

The two most frequent serious-event terms listed are Fever at 6.7% and Stroke at 6.7%. These are registry entries for serious adverse-event terms; the record does not provide further event-level detail or explain the denominator used for those percentages. They should not be combined or interpreted as a complete account of all serious events without that information.

These safety fields are descriptive, not comparative. With no control group, they cannot establish whether event frequency differs from another treatment or from the background risk in a comparable population. The listed safety denominator is also distinct from the listed enrollment of 18. The registry record does not explain the difference between those populations, so the safety counts should be read using their stated at-risk group rather than treated as rates for all enrolled participants.

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

The registry record establishes that this completed phase II study evaluated a single group receiving niraparib, dostarlimab, and radiation for metastatic pancreatic cancer. It specifies a primary disease-control endpoint based on RECIST 1.1 and lists several secondary outcomes. It also gives limited safety counts. These registry entries describe the design and the types of outcomes collected; they do not provide a numerical primary efficacy result.

The primary outcome count, disease-control rate, and uncertainty interval are not reported in the registry record. It is therefore not possible to state how many participants met the endpoint, how precise an estimate was, or whether the observed result was compatible with a range of clinically different possibilities. The listed enrollment of 18 is not a substitute for those outcome data, and it does not identify the endpoint analysis denominator.

Even with numerical results, the single-group design would support a description of outcomes in the participants studied rather than a randomized comparison of treatments. No control group is listed, and the registry record does not provide a comparison that could isolate the regimen's effect from other factors. A disease-control result would also combine complete response, partial response, and stable disease; component results would be needed to describe each category separately.

The safety counts identify adverse events recorded in the registry, including serious adverse-event terms and a deaths count. They do not by themselves determine causation or establish comparative safety. The record does not report the details necessary to attribute the deaths or explain the percentages attached to serious-event terms.

In short, the registry record makes the study question and endpoint definition clear, but it does not contain the outcome analyses needed to judge the observed disease-control result. It neither demonstrates a treatment effect nor establishes a lack of effect. Any such conclusion would require results and comparative evidence that are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how design determines what can be concluded from an outcome. A single-group study can document a treated cohort's observed disease-control classifications, but it does not create the randomized comparison needed to estimate a treatment contrast directly. This distinction remains important regardless of how an endpoint is defined or how clearly the assessment criteria are stated.

The primary endpoint is a composite of complete response, partial response, and stable disease under RECIST 1.1. The standard definitions make the intended classification rules explicit, including thresholds for changes in target-lesion diameters. At the same time, a combined disease-control count does not reveal the distribution across its component categories. Statistical interpretation therefore depends not only on the endpoint label but also on component results, the analysis denominator, and uncertainty estimates.

The record also illustrates the difference between registering outcomes and posting analyses. It names primary and secondary outcomes, but numerical outcome results and analysis details are not reported in the registry record. Without those, readers cannot evaluate precision, missingness, multiplicity, or the statistical methods used. A completed status does not itself supply those missing results.

Finally, planning illustrations based on hypothetical assumptions should be kept separate from a study's observed evidence. The assumptions described include a two-arm equal-allocation setup, whereas this study is registered as single-group. Such illustrations may aid general planning, but they are not a record of the trial's own sample-size rationale and are not observed power. These distinctions help prevent design assumptions, outcome definitions, and actual evidence from being conflated.

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.

Important trial-specific limitations arise from the single-group design. There is no concurrent control group, so any observed disease-control experience cannot be interpreted as a randomized comparison with another treatment. The registry record itself flags the absence of a like-for-like control comparison. Differences from external results would not remove the possibility of differences in participant characteristics, assessment, follow-up, or other care.

The primary outcome is defined, but its numerical result is not reported in the registry record. The record also does not report the denominator used for the primary analysis, the component counts for complete response, partial response, and stable disease, or a confidence interval. Thus, neither the observed disease-control frequency nor its precision can be assessed. The listed enrollment of 18 should not be assumed to equal the primary analysis population.

The safety summary uses 15 participants at risk, while the listed enrollment is 18. The registry record does not explain the population difference. It lists 14 deaths but does not report their causes or timing in the information available, so attribution cannot be made from the count. Event-level detail and the denominator for the serious-event percentages are also not reported in the registry record.

Further limitations include the absence of a detailed statistical analysis plan, missing-data rules, analysis-population definitions, multiplicity procedures, and information about interim analyses. The radiation description is incomplete, and the record does not give numerical drug doses. These omissions constrain interpretation of both treatment exposure and the methods used to analyze outcomes. No additional protocol details or statistical decisions can be inferred from the registry entry.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Niraparib + Dostarlimab + RT in Pancreatic Cancer — NCT04409002.
• Registry record last updated 2025-08-15; status last verified 2025-08.
• Results first posted 2023-07-03.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes NCT04409002, a completed phase II, single-group study of niraparib, dostarlimab, and radiation in metastatic pancreatic cancer. Listed enrollment is 18. The primary endpoint is disease control by RECIST 1.1, assessed up to 17 months. Secondary outcomes include disease control by irRECIST, progression-free survival, overall survival, and treatment-related adverse events.

The treatment description specifies 21-day cycles, daily oral niraparib at a predetermined dose, and intravenous dostarlimab once per cycle. The radiation description is incomplete. The safety group lists 15 participants at risk, 2 with serious adverse events, 15 with other adverse events, and 14 deaths; Fever and Stroke are each listed at 6.7% among the top serious-event terms.

Numerical primary-outcome results, analysis results, and detailed statistical methods are not reported in the registry record. The single-group design does not provide a randomized control comparison.