Phase 3 Study of FOLFIRINOX (mFFX) +/- SBRT in Locally Advanced Pancreatic Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase 3 Study of FOLFIRINOX (mFFX) +/- SBRT in Locally Advanced Pancreatic Cancer (NCT01926197), a randomized phase 3 study. The primary outcome is Progression-free Survival (PFS).

2. Clinical Question

3. Trial Design

This was a randomized Phase 3 study with a parallel-group design and no masking. One group received modified FOLFIRINOX alone; the other received modified FOLFIRINOX plus stereotactic body radiotherapy. The trial record identifies Stanford University as sponsor and lists an enrollment of 27. The primary outcome timeframe was 38 months.

The registry lists 13 participants in the mFFX arm and 12 in the mFFX-plus-SBRT arm for the primary outcome. The record does not explain the difference between those arm counts and the listed enrollment, so no reason for it can be inferred. The counts shown for adverse-event analyses use different denominators from the primary outcome arm counts; the record does not explain that accounting either.

The study is marked completed. Its recorded start date is 2013-08-14 and its primary completion date is 2021-09-30. These dates describe the registry record and do not, by themselves, establish when every analysis was conducted or reported.

The registry describes the trial’s objective as assessing safety and efficacy of mFFX alone or with SBRT. It lists local PFS, PFS at 1 year, metastasis-free survival, overall survival, and Grade 2 or Greater Gastrointestinal toxicity as secondary outcomes. Although those outcomes are named, numerical results are not reported in the registry record for every listed outcome. A sample-size justification, allocation ratio details beyond the parallel randomized design, and additional operational features are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Only participants that received study treatment are included.

6. Primary Endpoint

The registry lists the primary outcome as Progression-free Survival (PFS), measured over 38 months.

The primary endpoint was progression-free survival (PFS), measured in months. The registry defines PFS as the period a participant remains alive without tumor progression either locally or at a distant site in the body. This combines disease progression and death into a time-to-event outcome: a participant’s time is counted until one of those events occurs, subject to the endpoint definition and follow-up.

The reported summary is the median PFS for each treatment group, accompanied by a standard deviation. A median describes the midpoint of the observed time distribution: half of the observations fall at or below it and half at or above it, subject to how the endpoint data are handled. It is useful for describing time-to-event outcomes whose distributions may not be well represented by an arithmetic average.

The registry reports median PFS of 6.5 months for mFFX and 8.4 months for mFFX plus SBRT. The reported standard deviations are 6.6 and 9.5 months, respectively. A standard deviation describes spread; it is not a confidence interval for the median and does not, on its own, quantify uncertainty in the comparison between treatment groups.

The record does not report a confidence interval or sponsor-reported test for the primary comparison. A full interpretation therefore needs to distinguish the descriptive group summaries from an inferential estimate of the treatment-group contrast.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The record identifies a randomized, open-label, parallel-group Phase 3 trial. Randomization is intended, in general, to make treatment groups comparable on average with respect to both measured and unmeasured baseline factors. In a small trial, randomization does not guarantee close balance in every characteristic. The registry record does not report details of the randomization procedure, concealment, stratification, or baseline balance, so those features cannot be evaluated from the posted information.

How should a median time-to-event result be read?

Progression-free survival is a time-to-event outcome. In general, time-to-event methods use information about both when events occur and how long participants are observed. Participants without a recorded event during follow-up may be censored rather than treated as having an event at the end of observation. This distinction matters because the amount and pattern of follow-up can affect the estimated survival distribution.

The registry gives median PFS and standard deviation for each arm. The median is a descriptive point on the estimated or observed time distribution, while the standard deviation describes dispersion. Neither is a direct estimate of the relative treatment effect. A between-group analysis might report a hazard ratio or another time-to-event contrast with a confidence interval, but such a sponsor-reported primary comparison is not reported in the registry record. The record does not identify the specific estimation method used for the posted medians or explain censoring and follow-up handling.

What does a confidence interval convey?

In general, a confidence interval expresses the uncertainty associated with an estimate under the method used to construct it. A narrower interval indicates greater precision under that method; a wider interval indicates less precision. A confidence interval is not a range that contains a fixed proportion of individual participant outcomes. For the primary PFS medians, the registry record supplies standard deviations but no confidence intervals. The distinction is important: variability among participant times and uncertainty about a group-level estimate are related concepts, but they are not interchangeable.

How should the secondary binary comparison be read?

The registry reports a one-year PFS participant count for each group and an unadjusted risk difference, confidence interval, and p-value. In general, a risk difference is the difference between the event proportions in two groups, with the direction depending on the order of subtraction. The interval describes uncertainty in that contrast under its calculation method. The posted interval runs from -0.556 to 0.120 and includes zero; the unadjusted p-value is 0.225. These are secondary results and do not establish the result of the primary time-to-event comparison.

The registry also lists relative risk and odds ratio estimates with intervals for this secondary comparison. These measures express a contrast on different scales from the risk difference. They can be less intuitive to interpret, and their estimates may be imprecise when counts are limited. The record’s listed analysis is unadjusted; it does not report an adjusted model or a sponsor-reported analysis for this endpoint.

What is the role of the planning illustration?

The sample-size illustration is based on the observed secondary effect, with stated assumptions of a two-sided significance level of 0.05, equal allocation, and a normal approximation. It gives planning counts per arm for detecting specified fractions of that observed effect at 80% or 90% power. These are planning illustrations, not observed power and not evidence about the trial’s original design. The registry record does not report the trial’s sample-size justification, and the illustration should not be read as a retrospective assessment of whether the study was adequately powered.

In general, planning from an observed effect can be unstable, particularly when the underlying estimate is imprecise. A smaller effect to detect requires a larger planned sample under otherwise fixed assumptions. The values in the illustration should therefore be understood only in the context of its stated assumptions, not as a design recommendation or as an analysis of the observed trial’s probability of success.

Why can over-dispersion matter?

In general, over-dispersion occurs when variability in the data exceeds what a statistical model assumes. If a model understates variability, standard errors can be too small and intervals too narrow. For a time-to-event endpoint, the more immediate concerns include censoring, timing of events, and the choice of model or estimator. The registry record does not report enough detail to assess model assumptions or over-dispersion for the primary analysis. No specific model diagnostics are reported.

10. Multiplicity, Missing Data and Analysis Populations

Several secondary outcomes are listed alongside the primary PFS endpoint. When a study evaluates multiple outcomes, there is a general risk that interpreting each unadjusted test as if it were the only test can overstate the strength of evidence. A prespecified testing sequence or multiplicity adjustment can help control that issue. Neither a multiplicity strategy nor a hierarchy for interpreting the secondary outcomes is reported in the registry record.

The registry record lists no sponsor-reported analyses for the posted outcomes. It provides a secondary one-year PFS comparison with an unadjusted p-value, but does not report whether that analysis was planned in advance, whether additional analyses were performed, or how missing or censored observations were treated. The analysis populations, such as intention-to-treat or per-protocol populations, are not reported in the registry record. The relationship between the listed enrollment and the outcome-specific arm counts is also unexplained.

For the primary endpoint, the record supplies group medians and standard deviations but not an inferential comparison. It does not report confidence intervals for those medians, a primary hazard ratio, or details of how censoring was addressed. A missing analysis description does not show that no analysis was done; it means the methods and results cannot be assessed from the registry record.

Missingness and censoring are not the same. In general, censoring records that an event time is not observed beyond a participant’s available follow-up, while missing data may involve absent measurements or other unavailable information. Both can affect interpretation, but the record does not describe patterns of missing data, reasons for loss to follow-up, or sensitivity analyses. The registry also does not report interim analyses or a statistical analysis plan. No assumptions about these procedures can be made.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: bleed (hemorrhage), gastrointestinal upper (15.4%), other, hemoglobin decreased (8.3%), pain, abdominal (8.3%), diarrhea (8.3%), bleed (hemorrhage), gastrointestinal (8.3%).

The registry reports serious adverse events for 5 of 12 participants in the mFFX group and 7 of 13 in the mFFX-plus-SBRT group. It separately reports other adverse events for 12 of 12 and 13 of 13 participants, respectively. These are group-level counts as recorded in the registry; they do not by themselves establish that an event was caused by a treatment or indicate its timing, duration, or clinical course.

The listed serious-event terms include gastrointestinal upper bleeding, decreased hemoglobin, abdominal pain, diarrhea, gastrointestinal bleeding, and nausea. The record gives 15.4% for gastrointestinal upper bleeding and 8.3% for each of the other listed terms. It does not provide enough context to determine from these entries whether terms overlap within participants or how the percentages were calculated.

Deaths are listed for 7 participants in the mFFX group and 8 in the mFFX-plus-SBRT group. The registry information does not state that these deaths were caused by treatment, nor does it provide cause or timing in this record. Safety counts should be interpreted descriptively, alongside the incomplete reporting context, rather than as comparative evidence of causation.

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

The posted primary results describe median PFS in each treatment group: 6.5 months with mFFX and 8.4 months with mFFX plus SBRT. The associated standard deviations describe reported spread, not the precision of the difference between groups. The registry record does not provide a sponsor-reported primary comparison, confidence interval, or test, so the size and uncertainty of a primary treatment contrast cannot be evaluated from these summaries alone.

The available secondary one-year PFS result is also uncertain: its unadjusted interval includes zero. Since it is a secondary comparison and no multiplicity approach is reported, it should not be treated as definitive evidence about the primary endpoint. Other outcomes are named, but their numerical results are not reported in the registry record.

These results do not establish whether adding SBRT changes PFS, overall survival, metastasis-free survival, local progression, or toxicity. They also do not establish that the treatments are equivalent. A failure to report a primary comparison or complete secondary results is not evidence that the groups are the same; it limits what can be concluded from this record.

The trial’s randomized design is relevant because random allocation is intended to support a fair comparison, but interpretation also depends on follow-up, endpoint handling, analysis populations, missing data, and the prespecified statistical plan. Those details are not reported in the registry record. The posted figures should therefore be read as a limited account of group summaries and selected event counts, not as a complete trial analysis or treatment recommendation.

13. Why This Trial Matters Statistically

This trial addresses a clear comparative question: whether adding SBRT to modified FOLFIRINOX changes outcomes relative to mFFX alone for locally advanced pancreatic cancer. From a statistical perspective, the randomized parallel-group design is a useful framework for comparing treatment strategies, while the open-label designation and the available group counts set important context for interpreting the record.

The primary endpoint is a time-to-event measure, which captures both whether progression or death occurs and when. Reporting only group medians and standard deviations leaves unanswered questions about the comparative effect and its precision. A sponsor-reported estimate with an uncertainty interval, together with details of follow-up and censoring, would be needed to assess the primary comparison more fully. Those items are not reported in the registry record.

The record also illustrates why endpoint labels and counts are not a substitute for complete statistical reporting. Several clinically relevant secondary outcomes are listed, but numerical findings and methods are incomplete. One available unadjusted secondary comparison has an interval that includes zero, and multiple outcomes create interpretive concerns when no adjustment or testing hierarchy is reported.

Finally, the listed enrollment, outcome-specific participant counts, adverse-event denominators, and death counts do not all share a single explained accounting framework in this record. Clarifying analysis populations and the disposition of participants would help readers understand which observations contribute to each result. These reporting limits do not invalidate the study design, but they constrain the strength and scope of conclusions that can be drawn from the posted registry information.

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.

Several limitations are specific to the available record. The principal one is that the primary result is presented as group medians and standard deviations without a reported comparative estimate, confidence interval, or sponsor-reported test. This leaves the precision of the primary contrast unclear. The record also does not describe the estimator or model used for the medians, how censoring was handled, or the distribution of follow-up.

The number of participants recorded for the primary endpoint differs from the listed enrollment, and the record does not explain the difference. The adverse-event denominators also differ from the primary endpoint arm counts. Without participant disposition and analysis-population details, the reasons for these counts cannot be established. It would be inappropriate to infer the amount or causes of missing data from the numbers alone.

Only one secondary outcome has a numerical comparison in the posted information. Other listed secondary outcomes lack numerical results in the registry record, and no multiplicity strategy is reported. The available secondary comparison is unadjusted and imprecise, so it cannot resolve the missing primary inference.

The safety information is also limited. Serious and other adverse-event counts are reported, but causality, timing, event overlap, and the calculation basis for the listed percentages are not explained. Death counts are present without cause or treatment attribution in this record.

Finally, the registry does not report eligibility details beyond the condition, a statistical analysis plan, the sample-size justification, interim analyses, or analysis populations. These gaps prevent a complete assessment of design assumptions, analysis decisions, and robustness. The sample-size illustration is explicitly a planning exercise based on an observed effect, not a record of the original design.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Phase 3 Study of FOLFIRINOX (mFFX) +/- SBRT in Locally Advanced Pancreatic Cancer — NCT01926197.
• Registry record last updated 2022-10-28; status last verified 2022-10.
• Results first posted 2022-10-28.
• 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

NCT01926197 is a completed randomized Phase 3 parallel-group study sponsored by Stanford University. It compared modified FOLFIRINOX alone with modified FOLFIRINOX plus stereotactic body radiotherapy in locally advanced pancreatic cancer. The registry lists enrollment of 27 and identifies the primary outcome as median PFS over a 38-month timeframe.

Median PFS was reported as 6.5 months for mFFX and 8.4 months for mFFX plus SBRT, with standard deviations of 6.6 and 9.5 months. A sponsor-reported primary comparative estimate and its uncertainty are not reported in the registry record. The record lists several secondary outcomes, but numerical results are incomplete. It also reports adverse-event and death counts that should be read descriptively, without inferring treatment causation.