Ponatinib Hydrochloride in Treating Patients With Advanced Biliary Cancer With FGFR2 Fusions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Ponatinib Hydrochloride in Treating Patients With Advanced Biliary Cancer With FGFR2 Fusions (NCT02265341), a na phase 2 study. The primary outcome is Clinical Benefit Rate (Percentage), Which Includes Confirmed Tumor Response (Complete Response [CR] or Partial Response [PR]) or Stable Disease (SD).

2. Clinical Question

3. Trial Design

This was a phase II pilot study of ponatinib hydrochloride in patients with biliary cancer and FGFR2 fusions. The registry identifies the design as single-group, with no masking and no randomized allocation. Enrollment was 12. The sponsor was Mayo Clinic, and the study status is completed. The recorded start was 2014-12, and primary completion was 2018-05-01.

Participants in the treatment group received ponatinib hydrochloride by mouth once daily on days 1-28. Courses were repeated every 28 days unless disease progression or unacceptable toxicity occurred. The registry also lists laboratory biomarker analysis and quality-of-life assessment. Secondary outcomes included CA 19-9 response, overall toxicity rate assessed using NCI CTCAE v4, progression-free survival, and survival time.

A single-group design describes outcomes among participants receiving the study treatment; it does not directly estimate how those outcomes differ from a concurrent control group. Without random allocation or a comparator, changes over time or outcomes observed in this group cannot be separated from other influences using a between-group contrast. The registry does not report a control group, masking, or a randomized comparison.

The study is described as a pilot phase II trial. The registry record does not report a sample-size justification, the intended precision of the primary endpoint, or a statistical plan for comparing outcomes with an external benchmark. It lists 8 outcome measures and 0 analyses. The primary endpoint is described as clinical benefit, but its observed result is 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

Analysis population for the primary outcome, as defined in the registry: Only patients who received treatment for at least 8 weeks are evaluable for this outcome (i.e. one patient refused further treatment and was therefore unevaluable for response at 8 weeks)

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 Clinical Benefit Rate (Percentage), Which Includes Confirmed Tumor Response (Complete Response [CR] or Partial Response [PR]) or Stable Disease (SD), measured over Up to 10 months of treatment.

The primary endpoint is the clinical-benefit rate: the proportion of evaluable patients with a confirmed complete response, a confirmed partial response, or stable disease lasting at least 4 cycles. Its timeframe is up to 10 months of treatment. A confirmed response required the response status to be present on 2 consecutive evaluations at least 8 weeks apart. Complete response required disappearance of all target lesions and reduction of each target lymph node's short axis to <1.0 cm. Partial response was defined as at least a 30% decrease in the specified sum of target lesion and lymph-node measurements. Stable disease meant neither enough shrinkage to qualify as partial response nor enough increase to qualify as progressive disease.

This composite endpoint includes both tumor shrinkage and sustained disease stability. It can therefore describe a broader category of clinical benefit than response alone. The registry restricts evaluation to patients treated for at least 8 weeks; it notes that one patient refused further treatment and was unevaluable at 8 weeks. The observed numerator, denominator, clinical-benefit percentage, and 95% confidence interval are not reported in the registry record. Without those results, the endpoint definition alone does not establish how often clinical benefit occurred.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study's primary endpoint is a proportion: the number of evaluable participants meeting the clinical-benefit definition divided by the total number of evaluable participants. This is a binary-outcome framework because each eligible participant either meets the composite definition or does not. The definition combines confirmed complete response, confirmed partial response, or stable disease lasting at least 4 cycles. The registry sets the assessment timeframe as up to 10 months of treatment.

What does a confidence interval describe?

General statistical explanation: a confidence interval describes uncertainty around an estimate. For a proportion, it indicates a range of values compatible with the observed data under the method used to construct the interval. Wider intervals generally indicate less precision, while narrower intervals indicate greater precision. The registry specifies a 95% confidence interval for the primary endpoint, but does not report the interval or the underlying estimate. It is therefore not possible to assess the reported precision from the registry record.

A confidence interval is not a guarantee that the true value lies within a particular interval, and it does not replace the need to understand how the endpoint and analysis population were defined. Here, the endpoint definition includes a response-confirmation rule and a minimum treatment duration for stable disease. The population rule also excludes patients who did not receive at least 8 weeks of treatment from evaluation. Those features determine which observations contribute to the estimate.

Why does the endpoint definition matter?

The response categories have distinct criteria. A complete response required disappearance of all target lesions and reduction in each target lymph node's short axis to <1.0 cm. A partial response required at least a 30% decrease in the specified sum of measurements. Either response needed to be observed on 2 consecutive evaluations at least 8 weeks apart to count as confirmed. Stable disease was defined by not meeting the criteria for partial response or progressive disease, and it counted toward clinical benefit only when sustained for at least 4 cycles.

General statistical explanation: combining several outcomes into a composite can make a single summary useful when the components all represent outcomes of interest. However, a composite estimate does not show the distribution across its components. If the result were available, component counts would help clarify whether the overall proportion was driven by confirmed responses, stable disease, or both. The registry record does not report those component counts.

How does the single-group design affect interpretation?

General statistical explanation: a single-group study estimates outcomes in the participants who received the study treatment. Without a concurrent comparator, it does not estimate a randomized difference between treatment groups. Other factors, including the natural course of disease and differences in participant characteristics, cannot be ruled out as explanations for observed outcomes using a direct between-group contrast. The design recorded is single-group, with no masking and no randomized allocation. The registry does not report an external comparison as part of the primary analysis.

The endpoint's evaluability rule creates a distinction between enrollment and the analysis population. The record reports enrollment of 12, says that participants needed at least 8 weeks of treatment to be evaluable, and notes that one patient refused further treatment and was unevaluable at 8 weeks. It does not report the final evaluable denominator, the endpoint count, or how other forms of missing outcome information were handled. Those omissions prevent reconstruction or checking of the primary proportion.

What can the planning assumptions tell us?

The registry's sample-size assumptions describe a planning illustration, not a randomized design for this study. That illustration specifies a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. For time-to-event outcomes it uses the total number of events; for other outcomes it uses participants per arm. The actual study is single-group, so the equal-allocation, two-arm assumptions do not describe its enrolled design. The registry record does not report the study's own sample-size justification or a planned power calculation for the primary endpoint.

General statistical explanation: planning calculations use assumptions to explore what a future study might require; they do not describe the precision achieved by the completed study. They are not observed power. A planning illustration based on an observed effect is also distinct from an estimate of that effect's uncertainty. No planning values or primary endpoint estimate are reported in the registry record, so the assumptions alone cannot establish whether the study was adequately sized or how precise its primary result would be.

Finally, the registry lists 8 outcome measures and 0 analyses. The availability of outcome definitions does not establish that numerical results were posted for each one. For this primary endpoint, the registry record does not report the estimate or confidence interval. Statistical interpretation must therefore stop short of claims about the observed clinical-benefit rate, its precision, or a treatment effect.

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 (Ponatinib Hydrochloride)")
n      <- c(11)
events <- c(5)
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 (Ponatinib Hydrochloride)  11  5  45.5  16.7  76.6  21.3  72.0  21.1  72.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary clinical-benefit endpoint and secondary outcomes that include CA 19-9 response, overall toxicity rate, progression-free survival, and survival time. It does not report a multiplicity strategy, such as an adjustment across multiple outcome tests or a stated hierarchy for interpreting them. Without that information, the registry record does not establish how the risk of false-positive findings across outcomes was addressed. The record lists 8 outcome measures and 0 analyses, and does not give numerical results for the primary endpoint.

Missing outcome information is especially relevant because primary-endpoint evaluation required at least 8 weeks of treatment. The record states that one patient refused further treatment and was unevaluable for response at 8 weeks. It does not report the final number of evaluable participants, how other missing assessments were handled, or whether any participants were excluded for reasons besides the stated refusal. No imputation method or missing-data sensitivity analysis is reported in the registry record.

The analysis population is not described beyond the evaluability rule. The record does not state whether all enrolled participants were included in any other outcome analyses or how safety outcomes were analyzed. It also does not report an interim-analysis plan or other details of the statistical analysis plan. These omissions limit the ability to assess the handling of incomplete data, the consistency between planned and reported analyses, or the effect of multiple outcome assessments on interpretation.

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: pancreatitis (8.3%), fatigue (8.3%), sepsis (8.3%), cough (8.3%), pleural effusion (8.3%).

The registry reports safety figures for the treatment group, with 12 participants at risk. It records 5 participants affected by serious adverse events and 11 affected by other adverse events. It also lists 9 deaths. The serious-event terms with the largest reported percentages are pancreatitis, fatigue, sepsis, cough, and pleural effusion, each reported as 8.3%.

These are descriptive registry figures. They do not, by themselves, show that ponatinib caused an event or death. The record does not state the attribution of the deaths, and the listed serious-event percentages do not provide event timing or individual clinical context. A participant count affected is also distinct from a count of separate events; the registry does not provide enough information to infer how many episodes occurred or whether participants experienced more than one event.

The figures should not be treated as a comparison with another treatment because the study had no control group. The record also does not describe a causal analysis relating adverse events to treatment exposure. The percentages and counts therefore characterize what was recorded in this group, rather than establishing comparative risk or causation.

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

The study addresses a defined clinical question: whether patients with advanced biliary cancer and FGFR2 fusions experienced confirmed tumor response or sustained stable disease while receiving ponatinib. Its primary endpoint combines those categories into a clinical-benefit proportion, and the registry sets an assessment timeframe of up to 10 months of treatment. The record gives detailed response definitions and specifies that patients needed at least 8 weeks of treatment to be evaluable.

However, the registry record does not report the observed clinical-benefit proportion, its numerator or denominator, or its 95% confidence interval. The endpoint's presence in the registry is not a result. Without a posted estimate, the clinical-benefit rate and its precision cannot be described. The study's completion status does not fill that information gap.

The single-group design also limits what any observed outcome could establish. It can describe outcomes among the participants who received ponatinib, but it cannot provide a randomized comparison with another group. The record does not report a comparator, randomized allocation, or masking. A finding in this group alone would not isolate a treatment effect from other possible influences.

The safety record gives counts of participants affected by serious and other adverse events and lists deaths and serious-event terms. Those figures are descriptive; the record does not establish causation or provide a comparative safety estimate. In particular, the listed deaths should not be attributed to treatment based on these figures alone.

Accordingly, the registry supports a description of the study design, population, dosing schedule, endpoint definitions, and posted safety information. It does not support a numerical conclusion about the primary clinical-benefit result or a comparative conclusion about treatment effects. The distinction is between what the protocol intended to measure and what numerical outcomes the registry actually reports.

14. Why This Trial Matters Statistically

This study illustrates why a registered endpoint and a posted result are different kinds of evidence. The registry defines a clinical-benefit proportion in detail, including confirmation requirements, response criteria, a stable-disease duration threshold, and an evaluability rule. Those definitions make clear what would count as an event. Yet the registry record does not report the event count, evaluable denominator, estimate, or confidence interval. Without these, readers cannot assess either the observed proportion or its precision.

The example also highlights the importance of analysis populations. Enrollment was 12, but primary-endpoint evaluation required treatment for at least 8 weeks. The record notes that one patient refused further treatment and was unevaluable at 8 weeks, while not reporting the final evaluable denominator. Enrollment and the endpoint analysis population are therefore not interchangeable.

Design is equally important. A single-group phase II study can provide descriptive evidence about outcomes in its treated participants, but it does not create a randomized contrast. The lack of a comparator must shape how the endpoint and safety information are interpreted. The clinical-benefit definition may combine multiple outcomes, while the listed safety counts do not establish causation. These are distinct interpretive limits, not details that can be resolved by the endpoint definition alone.

Finally, the record lists secondary outcomes and describes planning assumptions, but does not report a primary-outcome analysis or a statistical plan for multiplicity and missing data. The case therefore shows why readers need both clear endpoint definitions and complete numerical reporting, including the analysis population and uncertainty. When those results are absent, careful interpretation means identifying what the registry describes without treating the intended measurement as an observed finding.

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 main trial-specific limitation for interpreting the primary endpoint is that its result is not reported in the registry record. The clinical-benefit estimate, confidence interval, event count, and evaluable denominator are unavailable. Although the record defines the endpoint and its timeframe, those details cannot substitute for the observed outcome.

The primary endpoint excludes patients who did not receive at least 8 weeks of treatment. One patient refused further treatment and was unevaluable for response at 8 weeks, but the record does not report the final evaluable population or how other missing assessments were handled. This leaves uncertainty about the precise analysis denominator and the treatment of incomplete outcome data.

The single-group design has no randomized comparator, and the registry reports no masking. Therefore, outcomes in the treated group cannot be interpreted as a controlled difference between treatments. The record also does not report a sample-size justification, the statistical plan for the primary endpoint, a multiplicity strategy, or an interim-analysis plan.

Safety interpretation is limited by the descriptive nature of the posted figures. The record lists participants affected by adverse events and deaths but does not establish event causality or provide a comparative safety analysis. The listed serious-event terms have percentages, but the record does not give further detail on timing or individual event histories. These limitations restrict interpretation to the information recorded rather than supporting causal or comparative conclusions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Ponatinib Hydrochloride in Treating Patients With Advanced Biliary Cancer With FGFR2 Fusions — NCT02265341.
• Registry record last updated 2020-11-25; status last verified 2019-04.
• Results first posted 2019-08-06.
• 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

NCT02265341 is a completed phase II pilot study sponsored by Mayo Clinic. It enrolled 12 patients with advanced biliary cancer and FGFR2 fusions in a single-group, unmasked design without randomized allocation. Participants received ponatinib hydrochloride by mouth once daily on days 1-28, with courses repeating every 28 days unless disease progression or unacceptable toxicity occurred.

The primary endpoint was the clinical-benefit rate, defined by confirmed complete or partial response, or stable disease lasting at least 4 cycles, with assessment up to 10 months of treatment. Patients had to receive treatment for at least 8 weeks to be evaluable; one patient refused further treatment and was unevaluable at 8 weeks. The registry does not report the primary endpoint estimate or its 95% confidence interval.

The safety record lists 5 participants affected by serious adverse events, 11 by other adverse events, and 9 deaths. Serious-event terms each reported at 8.3% include pancreatitis, fatigue, sepsis, cough, and pleural effusion. The registry does not establish causation or provide a comparator-based treatment effect.