A Study of TAS-120 in Patients With Metastatic Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of TAS-120 in Patients With Metastatic Breast Cancer (NCT04024436), a non-randomized phase 2 study. The primary outcome is Objective Response Rate (ORR) - Cohorts 1, 2.

2. Clinical Question

3. Trial Design

NCT04024436 was a Phase 2, non-randomized, open-label study with a parallel model. Its planned enrollment was 64. The registry lists the study status as terminated, with a start date of 2020-01-28 and a primary completion date of 2023-05-31. The reason for termination is not reported in the registry record.

The study included four cohorts rather than randomized treatment arms intended for direct comparison. Cohort 1 involved advanced or metastatic HR+, HER2- breast cancer with FGFR2 amplification and measurable disease. Cohort 2 involved triple negative breast cancer with FGFR2 amplification and measurable disease. Cohort 3 included HR+, HER2- or triple negative disease with FGFR2 amplification and non-measurable disease. These first three cohorts received futibatinib. Cohort 4 involved HR+, HER2- disease with FGFR1 amplification and measurable disease, and received futibatinib plus fulvestrant.

The registry descriptions specify futibatinib 20 mg in each cohort. For Cohorts 2 and 3, the description states once-daily oral tablets in continuous 28-day cycles, with different maximum treatment durations. The Cohort 1 description is incomplete in the registry record, and the full Cohort 4 administration description is also incomplete. Additional protocol details should not be inferred from those partial descriptions.

Because allocation was non-randomized and the cohorts differed by tumor characteristics, disease measurability, and regimen, differences across cohorts would not isolate a treatment effect. The posted primary endpoint was planned for Cohorts 1 and 2 only. The record does not report a randomized comparison, a control group, or a reason for termination. It also does not report a sample-size justification or a detailed statistical analysis plan.

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.

Before assignment: A total of 64 participants were enrolled in either Cohorts 1, 2 or 3 to receive futibatinib or to Cohort 4 to receive futibatinib plus fulvestrant.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All treated population included all enrolled participants who received at least 1 dose of study drug. This outcome measure was planned to be reported for only Cohort 1 and 2 as pre-specified in Protocol and SAP.

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: All treated population included all enrolled participants who received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate (ORR) - Cohorts 1, 2, measured over At the end of every 2 cycles until disease progression (up to 40 months).

The primary endpoint was objective response rate (ORR) in Cohorts 1 and 2. It was defined as the percentage of participants with a confirmed complete response or partial response, based on investigator assessment. A complete response required disappearance of all target lesions; pathological lymph nodes were required to have a short axis below 10 mm. A partial response required at least a 30% decrease in the sum of target-lesion diameters relative to baseline. The best overall response from treatment start until progressive disease or the start of subsequent new anticancer treatment determined whether a participant counted as a responder.

ORR is a binary outcome at the participant level: each participant is classified as having or not having a confirmed response under the stated criteria. Reporting the proportion who respond gives a direct description of observed tumor response in the defined population. The 95% confidence interval is intended to convey uncertainty around that proportion; it is not a guarantee that the true response rate lies within the interval.

The endpoint was assessed at the end of every 2 cycles until disease progression, over a timeframe stated as up to 40 months. The analysis population was defined as all enrolled participants who received at least 1 dose of study drug. Although the registry specifies the endpoint definition and intended population, it does not report the response count, percentage, or confidence interval. The magnitude and precision of the primary outcome therefore are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a percentage of participants and specifies a 95% confidence interval. These features describe the planned reporting format, but the numerical outcome itself is not reported in the registry record. No response count, estimate, or interval can therefore be interpreted for this trial.

What does a participant-level response measure?

ORR classifies each participant according to whether the investigator recorded a confirmed complete or partial response under the outcome definitions. The complete-response criterion concerns disappearance of target lesions, with a specified criterion for pathological lymph nodes. The partial-response criterion concerns a decrease in the sum of target-lesion diameters compared with baseline. The endpoint uses the best overall response from treatment start until progressive disease or the start of subsequent new anticancer treatment.

Once each participant has been classified, the response proportion is the number classified as responders relative to the relevant analysis population. That proportion is a concise summary, but it does not describe every aspect of disease course. It does not, by itself, indicate how long a response lasted, how quickly it began, or what happened to participants who did not meet response criteria. The registry lists duration of response and other outcomes separately, but their numerical results are not reported in the registry record.

What does the confidence interval add?

A confidence interval communicates uncertainty around an estimated quantity. In general, its width depends on the amount of information in the data and the variability associated with the estimate. A more precise estimate has a narrower interval; a less precise estimate has a wider interval. The stated 95% confidence level is part of the planned reporting, not a statement that there is a 95% probability that a fixed trial result is correct.

The registry does not state the method used to calculate the ORR confidence interval. It would therefore be inappropriate to attribute a particular interval method or small-sample adjustment to the trial. Since the estimate and interval are not posted, the actual uncertainty cannot be assessed. A nominal planned confidence level cannot replace the missing numerical result.

Why do the study groups matter?

This was a non-randomized study, with cohorts defined by clinical and molecular characteristics. Cohorts 1 and 2 differed in breast cancer subtype, while Cohort 3 had non-measurable disease and Cohort 4 involved FGFR1 amplification and combination treatment. The primary endpoint applied to Cohorts 1 and 2 only. These design features mean that other cohorts do not act as randomized control groups for the primary endpoint.

In general, random allocation helps balance measured and unmeasured characteristics between treatment groups. Without random allocation, differences in outcomes may reflect differences between the participants or disease groups, as well as differences in treatment. That issue is especially important when cohorts are deliberately defined by different tumor characteristics. The registry record does not report an adjusted comparative analysis that would address such differences.

What should be inferred from a missing result?

A missing numerical endpoint is a reporting limitation, not a measured outcome. It does not mean the estimate is zero, that no response occurred, or that the endpoint was not assessed. The registry lists the assessment schedule and definition but does not provide the observed response data. The reason those results are absent is not reported in the registry record.

The record also does not state the number of participants with evaluable primary endpoint data, the handling of participants without an assessment, or whether any sensitivity analyses were conducted. General statistical practice recognizes that missing outcome data can affect an estimate, particularly if the chance of missingness is related to a participant's condition or outcome. Without information on missingness and analysis choices, the direction or size of any such effect cannot be determined for this trial.

The sample-size assumptions describe planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. They are not the trial's own design justification, and no numerical planning estimates are included in the registry record. Such illustrations should be understood as hypothetical planning quantities, not observed power or evidence about the results. The trial's own sample-size justification is not reported in the registry record.

The registry lists secondary outcomes including response and clinical benefit measures, six-month progression-free survival rate, progression-free survival, duration of response, overall survival, and adverse events. Listing an outcome does not establish that its result is available. The record indicates no analysis results posted. No detailed statistical analysis plan, interim analysis plan, or multiplicity procedure is reported in the registry record.

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("Futibatinib (Cohort 1)", "Futibatinib (Cohort 2)")
n      <- c(17, 21)
events <- c(0, 2)
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  Futibatinib (Cohort 1)  17  0  0.0  0.0  19.5  0.0  18.4  0.1  18.5
ARM  Futibatinib (Cohort 2)  21  2  9.5  1.2  30.4  2.7  28.9  2.9  29.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcomes, including the primary ORR for Cohorts 1 and 2 and secondary measures covering response, clinical benefit, progression-free survival, duration of response, overall survival, and adverse events. Considering multiple outcomes can raise multiplicity concerns: when many hypotheses are tested, the chance of at least one apparently unusual result can increase. In general, a prespecified analysis strategy may address this through outcome prioritization or statistical adjustment. The registry record does not report a multiplicity strategy, so it is not possible to determine how any family of statistical tests was handled. No numerical analysis results are posted for the primary endpoint.

The registry defines the primary analysis population as all enrolled participants who received at least 1 dose of study drug. This is not the same description as an intention-to-treat population, and the record does not identify a separate intention-to-treat analysis. It does not report how many participants were included in the primary endpoint analysis, how many had evaluable response data, or whether participants were excluded after treatment began.

Missing-data handling is not reported in the registry record. There is no description of how absent, incomplete, or unevaluable response assessments were classified, or whether sensitivity analyses were performed. In general, missingness can affect the interpretation of an estimate when the likelihood of a missing assessment is related to participant characteristics or outcome. The available record does not permit assessment of that possibility.

The record also does not report the trial's own sample-size justification, any interim analyses, or a detailed statistical analysis plan. These are distinct from planning illustrations based on an observed effect. No further assumptions about the study's analytic decisions can be made from the posted information.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pathological fracture (25.0%), nausea (5.9%), covid-19 (5.9%), hyperphosphataemia (5.9%), acute promyelocytic leukaemia (5.9%).

The registry reports safety counts separately for each cohort. For futibatinib in Cohort 1, 17 participants were at risk for serious adverse events and 4 were affected; 17 were at risk for other adverse events and 16 were affected. The reported death count was 10. In Cohort 2, the corresponding counts were 21 at risk and 5 affected for serious adverse events, 21 at risk and 21 affected for other adverse events, and 14 deaths. In Cohort 3, the counts were 4 at risk and 1 affected for serious adverse events, 4 at risk and 4 affected for other adverse events, and 2 deaths. For futibatinib plus fulvestrant in Cohort 4, the counts were 22 at risk and 4 affected for serious adverse events, 22 at risk and 22 affected for other adverse events, and 7 deaths.

These are cohort-specific counts, not randomized comparisons. The cohorts differ in size and clinical characteristics, and the death counts are not identified as treatment-related in the registry record. The record does not provide causes of death or further context for interpreting individual events. The most frequently listed serious adverse event term was pathological fracture at 25.0%; nausea, COVID-19, hyperphosphataemia, acute promyelocytic leukaemia, and acute kidney injury were each listed at 5.9%. The denominator for these term percentages is not reported in the registry record. These summaries describe reported events and do not establish causation.

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

The posted record establishes the design and intended outcomes of a Phase 2 study of futibatinib, alone or with fulvestrant, in people with metastatic breast cancer and specified FGFR amplifications. It gives cohort descriptions, defines the primary endpoint, and reports selected safety counts. It does not report the primary ORR estimate or its 95% confidence interval. The central response result therefore cannot be quantified from the registry record.

This absence does not show that the treatment had no activity, nor does it establish that a response occurred. It simply leaves the primary outcome unanswered in the posted results. The same caution applies to other listed outcomes: their presence in the registry does not mean their estimates or analyses are available.

The study was non-randomized and had no like-for-like control group. Its cohorts differed in cancer subtype, FGFR amplification, measurability, and regimen. Even if cohort outcome estimates were available, comparisons across them would not have the interpretation of a randomized treatment comparison. They could reflect cohort composition as well as treatment.

Safety counts provide a description of reported events in the listed cohorts, but they do not establish that an event was caused by study treatment. The record does not identify the reported deaths as treatment-related or provide the context needed to attribute them. Counts and event-term percentages should therefore be read descriptively, not as evidence of comparative risk.

Overall, the registry record supports conclusions about what the study intended to evaluate and what safety counts it posted. It does not support a numerical conclusion about primary response, an estimate of comparative benefit, or an inference about treatment causality. Additional outcome and analysis information is not reported in the registry record.

14. Why This Trial Matters Statistically

This study raises important statistical questions because it evaluated a response endpoint in molecularly defined groups without random allocation. FGFR amplification and breast cancer subtype were used to define cohorts, while the treatment regimen also differed for Cohort 4. Those design choices make clear why the analysis population and the cohort-specific outcome results matter: a response estimate for one group cannot automatically be generalized to another group with different disease characteristics.

ORR is a clinically interpretable summary, but its meaning depends on a transparent denominator, response assessment, and uncertainty interval. The registry supplies the endpoint definition and intended assessment schedule, yet the response estimate and confidence interval are not reported. Without those figures, readers cannot assess the size or precision of the observed response proportion. That gap illustrates the difference between a prespecified outcome and a reportable result.

Safety summaries also require statistical context. Counts across cohorts with different numbers of participants at risk are not direct comparative evidence. Serious event terms and death counts are useful descriptive information, but causal attribution and relative risk cannot be established from these summaries alone. The record's lack of denominator detail for the listed serious-event term percentages further limits their interpretation.

The record also shows the value of complete reporting about analysis populations, missing outcomes, and multiplicity. These details help readers judge how faithfully the posted estimate represents the intended population and how uncertainty should be understood. For this trial, several of those analytic details are not reported in the registry record, and no analysis results are listed.

The statistical significance of the study's record is therefore not a claim about efficacy. It is an example of how design, endpoint definitions, outcome availability, and transparent analysis reporting jointly determine what can be learned from a clinical trial.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most consequential limitation for statistical interpretation is that the registry does not report the primary ORR result. The endpoint definition and intended 95% confidence interval are available, but the response count, estimate, and interval are not. The reason for this missing result is not reported in the registry record, and it should not be interpreted as a particular outcome.

The trial was non-randomized and did not include a like-for-like control group. Its four cohorts differed in receptor status, FGFR amplification, measurable disease status, and regimen. Consequently, cohort contrasts would not provide a clean treatment comparison. The record does not report adjusted comparative estimates that might address differences in cohort characteristics.

The registry identifies 64 as planned enrollment and gives a terminated status, but the reason for termination is not reported. It does not give the number contributing to the primary endpoint, the number with evaluable response data, or a flow of participants through treatment and assessment. Safety populations should not be assumed to match the primary analysis population.

Missing-data rules, sensitivity analyses, the trial's sample-size justification, interim analyses, and a detailed statistical analysis plan are not reported in the registry record. The record also does not report a multiplicity approach for its several outcomes. These absences limit assessment of the analysis choices and the robustness of any potential estimates.

Some arm descriptions are incomplete. The Cohort 1 treatment description is truncated, and the Cohort 4 description does not give full administration details. The serious-event term percentages are listed without a stated denominator. Death counts are not identified as treatment-related, and causes are not reported. These limitations constrain interpretation beyond the available descriptive information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of TAS-120 in Patients With Metastatic Breast Cancer — NCT04024436.
• Registry record last updated 2025-11-13; status last verified 2025-10.
• Results first posted 2025-11-13.
• 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

NCT04024436 was a Phase 2, non-randomized, open-label study of futibatinib in metastatic breast cancer with specified FGFR amplifications, used alone or with fulvestrant. Planned enrollment was 64, and the registry status is terminated. The primary endpoint was investigator-assessed ORR for Cohorts 1 and 2 among participants who received at least 1 dose. It was defined using confirmed complete or partial response and was intended to be reported with a 95% confidence interval.

The registry record does not report the primary response count, estimate, or confidence interval, and indicates no analysis results posted. It lists secondary outcomes and provides cohort-level adverse-event counts, but those counts are descriptive and the cohorts are not randomized controls for one another. The termination reason, missing-data methods, multiplicity approach, and detailed analysis plan are not reported in the registry record.