Eribulin Mesylate as Second-Line Therapy for Locally Advanced, Unresectable, or Metastatic Pancreatic Cancer Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Eribulin Mesylate as Second-Line Therapy for Locally Advanced, Unresectable, or Metastatic Pancreatic Cancer Patients (NCT00383760), a na phase 2 study. The primary outcome is Objective Response (Complete and Partial) Evaluated Using RECIST Criteria.

2. Clinical Question

This is a descriptive question about response observed in a small, single-group phase II study, rather than a randomized estimate of comparative treatment effect. Objective response captures whether target lesions meet prespecified imaging criteria for disappearance or reduction. It does not by itself describe survival, symptom change, or how outcomes would have differed under another treatment.

The registry lists stable disease rate, median survival time, overall survival, median time to disease progression, time to progression, and response duration among secondary outcomes. It does not report their numerical results in the record. The study enrolled 15 participants, but the registry record does not provide the number evaluable for the primary endpoint or explain whether all enrolled participants were included in the response assessment.

3. Trial Design

This completed phase II study evaluated eribulin mesylate as second-line therapy in patients with locally advanced, unresectable, or metastatic pancreatic cancer. The National Cancer Institute (NCI) is listed as the sponsor. The study started in 2006-08 and had a primary completion date of 2011-07. Enrollment was 15 participants.

The design is recorded as single-group, with allocation listed as not applicable and masking listed as none. The intervention description specifies one treatment group receiving eribulin mesylate, or E7389, intravenously on days 1 and 8. Because there was no concurrent control group, the study does not provide a randomized comparison against another therapy or against no treatment. Outcomes observed in this group cannot, by design alone, distinguish treatment effects from the natural course of illness, other care, participant characteristics, or other influences.

The primary outcome was objective response, combining complete and partial responses under RECIST v1.0, assessed by CT for target lesions. The stated timeframe was up to 3 years. The registry also lists several secondary outcomes concerning disease stability, survival, progression, and response duration. The registry record does not report the detailed assessment schedule, treatment duration, criteria for discontinuation, or the statistical analysis plan.

Phase II describes the study phase, but the registry entry does not report a sample-size justification or a protocol-based decision threshold for interpreting response. It also does not report planned interim analyses, an independent review process, or details about how missing outcome assessments were handled. The fact that the study is marked completed indicates its recorded status; it does not establish that every planned analysis or outcome result appears 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

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 Objective Response (Complete and Partial) Evaluated Using RECIST Criteria, measured over Up to 3 years.

The primary endpoint is the number of participants with an objective response, defined as a complete or partial response using RECIST v1.0. The record says target lesions were assessed by CT. A complete response means disappearance of all target lesions. A partial response means at least a 30% decrease in the sum of the longest diameters of target lesions. The stated assessment period is up to 3 years.

Objective response is used in oncology trials to describe measurable tumor change under predefined imaging rules. A categorical response endpoint can make it possible to report whether participants crossed a consistent threshold for lesion disappearance or reduction, rather than relying only on an informal description of imaging findings. Prespecified criteria also support more reproducible classification across participants, although the record does not say who performed the assessments or whether they were independently reviewed.

This endpoint has a narrower meaning than overall benefit or comparative effectiveness. It does not include stable disease as an objective response, and it does not directly measure survival, symptoms, quality of life, or duration of response. The registry lists some of these as secondary outcomes, but their results are not reported in the registry record. Nor is the primary response result reported: the record gives the outcome definition and timeframe, but not the observed number of complete or partial responses.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical concepts can help clarify what this registry record can and cannot establish. The explanations below are general; they do not imply that a particular analysis was performed in this trial. The record does not report a statistical analysis plan or a primary outcome analysis.

What does a single-group design estimate?

In general, a single-group study describes outcomes among participants who received the study intervention. For a binary endpoint such as objective response, an analysis might report the number meeting the response definition and the number assessed. That proportion describes the observed group under the endpoint rules. It is not a randomized comparison, because no concurrent group provides the outcome that would have occurred under an alternative treatment.

Even when a response proportion is available, its interpretation depends on the denominator, outcome ascertainment, and uncertainty. The registry record states enrollment of 15, but does not report how many participants had sufficient assessment for the primary endpoint. It also does not report a response count or an interval estimate. Thus, neither an observed response proportion nor its precision can be derived from this record.

What do RECIST response categories mean?

As a general matter, an endpoint based on RECIST turns imaging measurements into prespecified categories. In this trial, complete response is defined as disappearance of all target lesions, while partial response requires at least a 30% decrease in the sum of the longest diameters of target lesions. Objective response combines those two categories. The registry says CT was used and identifies RECIST v1.0. It does not report details such as the schedule of scans, handling of unevaluable lesions, confirmation requirements, or whether image reviewers were blinded to other information.

A binary response endpoint is easy to describe but compresses information. Participants just meeting a threshold and those with more substantial reduction both qualify as responders, while a change that does not meet the threshold does not count as an objective response. This does not make the endpoint invalid; it means the category should be interpreted according to its definition and alongside other outcomes when those are available.

What does a confidence interval communicate?

In general, a confidence interval expresses the uncertainty around an estimate under a specified statistical procedure and its assumptions. For a response proportion, an interval would communicate how imprecisely the observed proportion is estimated from the participants assessed. An interval is not a range containing a known treatment effect with certainty, and it does not remove limitations such as selection, measurement, or lack of a comparator. No confidence interval for the primary endpoint is reported in the registry record.

For time-to-event outcomes, general methods account for the fact that some participants may not experience the event during follow-up or may have incomplete follow-up. These methods use the timing of events and censoring information. The registry lists survival and progression outcomes but does not report event counts, follow-up summaries, censoring details, or estimates with uncertainty. It is therefore not possible to assess the precision or assumptions of those analyses from the registry record.

Why does the comparator matter?

In a randomized comparative trial, random allocation is intended to make treatment groups comparable on average, allowing a contrast to estimate the effect of assignment under the trial conditions. This study is not randomized and has no control arm. An outcome observed after treatment cannot be directly contrasted with an outcome under another intervention within this study. Comparing its results with external populations would introduce additional questions about differences in eligibility, disease status, assessment, and care; the registry record does not provide such a comparison.

What do the planning assumptions mean?

General sample-size planning for a two-arm comparison depends on assumptions about the outcome, the effect size to detect, the significance level, allocation, and the statistical method. The planning assumptions recorded for the associated illustrations specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Those assumptions are not a description of the trial's single-group design. The registry record does not report that these assumptions were used to design the study, and it does not report a protocol-based sample-size justification.

Planning illustrations are not observed power. They describe hypothetical sample requirements under stated assumptions; they do not show the probability that the completed trial detected an effect, and they cannot compensate for absent outcome results. The record's stated sample-size assumptions also say that the effect to detect is expressed as a fraction of an observed effect. Since the primary response result is not reported in the registry record, no trial-specific conclusion about such planning assumptions can be drawn from the clinical information recorded. More generally, post hoc calculations based on an observed effect should not be treated as design justification.

Why can over-dispersion matter?

As a general statistical concept, over-dispersion occurs when variability in data is greater than a model assumes. If a model understates variability, standard errors may be too small and intervals may appear too narrow. This issue is relevant to some count and rate models, but the primary endpoint in this study is recorded as a participant count for a binary response outcome. The registry record does not describe a model that was fitted or any assessment of over-dispersion, so there is no basis to say whether it affected an analysis of this trial.

In short, statistical methods can describe response and uncertainty, but their interpretation requires the actual endpoint data, analysis population, and method details. For this study, the registry record documents the endpoint definition and design while leaving the primary result and analysis details unreported.

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("Treatment (Eribulin Mesylate)")
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  Treatment (Eribulin Mesylate)  15  0  0.0  0.0  21.8  0.0  20.4  0.2  20.6

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes or statistical comparisons are examined. In general, testing many hypotheses can increase the chance of at least one apparently positive finding if no adjustment or prespecified hierarchy is used. This study's registry record lists an objective-response primary outcome and several secondary outcomes, including stable disease rate, survival measures, progression measures, and response duration. It does not report a multiplicity strategy, a hierarchy for interpreting outcomes, or a statistical analysis plan. The number of registered outcomes alone does not show how many tests were conducted.

The registry record reports 0 analyses, and does not provide numerical primary or secondary outcome results. It therefore does not permit assessment of which analyses, if any, were performed, how they were specified, or whether multiplicity was addressed. A missing analysis record should not be interpreted as evidence that no analysis occurred; it means that analyses are not reported in the registry record.

Missing-data handling is also not reported in the registry record. For the response endpoint, the enrollment count is 15, but the number with evaluable imaging, the handling of missing scans, and the analysis population are not stated. In general, excluding participants with missing assessments can affect an estimated response proportion if missingness is related to outcome. Other approaches may rely on assumptions that also need explanation. Without a stated approach and participant flow, readers cannot assess the influence of missing primary endpoint data.

The record does not report interim analyses, stopping rules, or whether outcome assessments were reviewed independently. These details should not be inferred from the phase designation or the completed status. In general, transparent reporting of outcome definitions, analysis populations, missingness, and multiplicity helps readers judge how much confidence to place in an estimate. Here, the absent outcome results and analysis details substantially limit that assessment.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: death nos (13.3%), dry skin (6.7%), urinary tract obstruction (6.7%), creatinine increased (6.7%), constipation (6.7%).

The safety record describes one treatment group with 15 participants at risk. It reports that 4 participants were affected by serious adverse events and that 15 were affected by other adverse events. These are participant-level counts: they indicate how many people were recorded as affected, not how many separate event episodes occurred. A participant may have experienced more than one event, so the listed terms should not be added together to infer a total number of affected participants.

The registry lists Death NOS at 13.3%, Dry skin at 6.7%, Urinary tract obstruction at 6.7%, Creatinine increased at 6.7%, and Constipation at 6.7% among the top serious-event terms. These entries describe the terms and percentages recorded; they do not provide full case narratives or establish that the intervention caused the events. The record does not report timing, duration, seriousness criteria, relatedness assessment, or outcomes for the listed events. It also does not provide a comparison group, so these figures cannot be used to estimate a difference in risk versus another treatment.

Safety counts should be read alongside the size and design of the study. The cohort was small and uncontrolled, and the registry entry does not give detailed follow-up or ascertainment methods. The figures are useful as a record of reported events in the treated group, but they do not by themselves establish a causal safety profile.

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

The registry record establishes that a phase II study of eribulin mesylate was completed in a single group of 15 participants with locally advanced, unresectable, or metastatic pancreatic cancer. It specifies objective response under RECIST v1.0 as the primary outcome and describes how complete and partial responses were defined. It also records participant counts for serious and other adverse events.

What the record does not establish is the observed objective-response result. The primary outcome count is not reported in the registry record, and neither are numerical results for the listed secondary outcomes. Without those results, readers cannot determine the response frequency, survival estimates, progression results, or the precision of any such estimates. The outcome's stated timeframe of up to 3 years is not itself a report of how long participants were followed or what outcomes occurred.

The study also does not establish a comparative treatment effect. With no concurrent comparator and no random allocation, outcomes in the treated group cannot show what would have happened to comparable participants under an alternative. Observed events may reflect the intervention, the disease, other care, participant factors, or multiple influences; the registry information does not separate these explanations.

The safety entries likewise have a limited interpretation. They record participants affected by serious and other adverse events and list selected serious-event terms, but do not establish causal attribution. The absence of detailed event descriptions or comparative data prevents a broader inference about relative safety.

Thus, the record is informative about the study's design, intended endpoint, treatment schedule, enrollment, and selected safety reporting. It is not a complete account of efficacy or comparative safety outcomes. Any conclusion about benefit, lack of benefit, or relative performance would go beyond the results reported in the registry record.

14. Why This Trial Matters Statistically

This study illustrates an important distinction in clinical-trial interpretation: a clearly defined outcome is not the same as a reported outcome result. The registry record gives a recognizable imaging-based primary endpoint, including definitions for complete and partial response, but does not report the number of participants who met those criteria. Statistical interpretation requires both a sound definition and the observed data to which it applies.

Its single-group design also makes the scope of inference especially important. A response count, if reported, would describe outcomes among participants in the study. Without a concurrent comparator, it would not provide a randomized estimate of how outcomes differ from another treatment. This distinction is central when interpreting early-phase clinical research and when considering whether a result answers a descriptive or comparative question.

The small enrollment of 15 participants adds another reason to focus on uncertainty and reporting completeness. Even when results are available, limited data can leave estimates imprecise. In this registry record, however, the primary outcome result and its uncertainty are not reported, so precision cannot be assessed at all. It would be inappropriate to replace missing trial results with assumptions about what the observed response might have been.

The safety information demonstrates a related lesson: participant counts and listed event terms are not a causal comparison. They provide a record of reported events in the treatment group, but attribution, context, and comparative risk are not established by those counts. Together, these features make the trial statistically instructive as an example of how design, endpoint definition, analysis reporting, and causal interpretation must be considered separately.

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.

Several limitations specific to this study constrain interpretation. First, the registry record does not report the primary objective-response result. The number of participants with complete or partial response, the number assessed, and any uncertainty estimate are absent. This prevents a quantitative description of the primary outcome and makes it impossible to judge its precision.

Second, the study has a single-group design with no randomized comparator. Its results, even if fully reported, would describe outcomes in the treated group rather than directly estimate a difference against another therapy. The lack of masking is also recorded. The registry does not report whether outcome assessors were independent or whether CT interpretation was blinded, so the potential influence of assessment procedures cannot be evaluated.

Third, the enrollment was 15 participants. The registry does not describe the eligibility criteria in detail, participant characteristics at baseline, the number who completed treatment, or the number included in each outcome analysis. Those omissions limit assessment of who the findings might represent and whether all enrolled participants contributed to each endpoint.

Fourth, the registry lists multiple secondary outcomes but does not report their results, analysis methods, missing-data handling, multiplicity approach, interim analyses, or sample-size justification. It also does not report detailed follow-up or assessment schedules. These missing elements make it difficult to evaluate completeness, analytic choices, or the potential for selective outcome reporting.

Finally, the safety record offers counts and selected terms without detailed event descriptions or attribution. The record does not allow causal assessment of the listed events, and the lack of a control group precludes comparative safety interpretation. These limitations are specific to the information recorded and should not be filled by assumptions about the protocol or analyses.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Eribulin Mesylate as Second-Line Therapy for Locally Advanced, Unresectable, or Metastatic Pancreatic Cancer Patients — NCT00383760.
• Registry record last updated 2017-10-20; status last verified 2017-09.
• Results first posted 2016-12-09.
• 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

This completed phase II study evaluated eribulin mesylate as second-line therapy for locally advanced, unresectable, or metastatic pancreatic cancer. The National Cancer Institute (NCI) is listed as sponsor. The study started in 2006-08, had a primary completion date of 2011-07, and enrolled 15 participants.

The design was single-group, with allocation listed as not applicable and masking as none. Participants received eribulin mesylate intravenously on days 1 and 8. The primary outcome was the number of participants with complete or partial objective response under RECIST v1.0, assessed by CT, with a timeframe of up to 3 years. The primary result is not reported in the registry record.

The registry lists secondary outcomes involving stable disease, survival, progression, and response duration, but does not report their numerical results. The safety record reports serious adverse events affecting 4 of 15 participants and other adverse events affecting 15 of 15. The study has no concurrent comparator, and the record does not report an analysis plan, missing-data approach, or multiplicity strategy.