A Study of Tucatinib (MK-7119) in Combination With Trastuzumab and Capecitabine in Participants With Previously Treated Locally Advanced Unresectable or Metastatic Human Epidermal Growth Factor Receptor 2 Positive (HER2+) Breast Carcinoma (MK-7119-001): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Tucatinib (MK-7119) in Combination With Trastuzumab and Capecitabine in Participants With Previously Treated Locally Advanced Unresectable or Metastatic Human Epidermal Growth Factor Receptor 2 Positive (HER2+) Breast Carcinoma (MK-7119-001) (NCT04721977), a na phase 2 study. The primary outcome is Confirmed Objective Response Rate(cORR) Per Response Evaluation Criteria in Solid Tumors (RECIST) Version 1.1 Determined by Independent Central Review (ICR): Japanese Participants.

2. Clinical Question

3. Trial Design

This is a Phase 2, open-label, single-group study of tucatinib with trastuzumab and capecitabine. The registry record lists an enrollment of 66 and identifies Pfizer as the sponsor. The record gives a study start date of 2021-04-08 and a primary completion date of 2023-07-17; its status is active, not recruiting. These record details describe the study, but do not themselves provide an efficacy result.

All participants in the listed group received the combination. Because there is no separate control group, the design can describe response and safety outcomes during treatment, but it cannot directly estimate a between-group treatment effect. Open-label means treatment assignment was not masked to participants or study personnel. The primary outcome nevertheless used independent central review, which is a separate review process for response assessments; the registry record does not give further operational details about that review.

The study summary focuses on participants with unresectable locally advanced or metastatic HER2-positive breast cancer who had received prior treatment with a taxane anticancer agent, trastuzumab, pertuzumab and trastuzumab emtansine. The primary outcome is specifically defined for Japanese participants enrolled at Japanese sites who met response-evaluable criteria. Other listed outcomes include response rates and duration of response assessed by investigators or independent central review, including outcomes for Japanese participants, as well as progression-free survival assessed by independent central review for all participants.

The registry record lists 22 outcome measures, but no analysis records are reported. The protocol’s sample-size justification, detailed statistical plan, and any planned interim analyses are not reported in the registry record. The design therefore needs to be interpreted from the information actually recorded, rather than from assumptions about the protocol.

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 66 participants were enrolled in this study across Japan, Taiwan and South Korea. Results reported are based on the primary completion date of the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Response evaluable set for Japanese population included all Japanese participants enrolled in Japanese sites who met all of the following criteria: (1) had measurable disease at baseline, (2) received any amount of study treatment, and (3) had at least one post baseline disease assessment or discontinued due to clinical progression, toxicity, or death. Here, 'Number of Participants Analyzed' signifies participants evaluable for this outcome measure.

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 participants set included all participants who received any amount of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Confirmed Objective Response Rate(cORR) Per Response Evaluation Criteria in Solid Tumors (RECIST) Version 1.1 Determined by Independent Central Review (ICR): Japanese Participants, measured over From date of first dose until first documented disease progression or start of new anticancer therapy, whichever occurred first (maximum up to 17.8 months).

The primary endpoint is confirmed objective response rate, or cORR, determined by independent central review under RECIST version 1.1. It is the percentage of participants with a confirmed complete or partial response. A complete response requires disappearance of all target lesions, with pathological lymph nodes reduced below the specified short-axis threshold. A partial response requires at least a 30% decrease in the sum of target-lesion diameters from baseline. The registry record also defines progressive disease using change in target-lesion size and the appearance of new lesions, and says only assessments before documented progression or a new anticancer therapy were considered.

Response rate is used to describe the proportion of an assessed population whose disease meets prespecified response criteria. Confirmation is intended to distinguish a response supported by follow-up assessment from a single assessment. Independent central review provides an assessment separate from the investigator’s assessment, although the record does not report review procedures or agreement between reviewers.

The stated primary hypothesis is that cORR is greater than 20%. The endpoint’s timeframe runs from first dose until documented progression or new anticancer therapy, whichever occurred first, with a maximum of up to 17.8 months. The actual cORR estimate and its confidence interval are not reported in the registry record, so the endpoint’s observed magnitude and precision cannot be described.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies cORR as a binary outcome expressed as a percentage of participants. For each participant in the analysis population, the response classification is whether a confirmed complete or partial response was recorded under RECIST version 1.1. The estimated response rate is based on the participants included in that analysis. The registry record says participants with missing data were considered non-responders. The number classified as responders and the resulting percentage are not reported in the registry record.

The endpoint uses independent central review rather than investigator assessment for the primary outcome. Central review can provide an assessment independent of the local investigator’s classification, which is relevant when interpreting imaging-based response. It does not eliminate all measurement uncertainty: RECIST classification depends on the assessments available, their timing, lesion selection, and the review process. Details about these procedures, adjudication, or reviewer agreement are not reported in the registry record. Other listed outcomes include investigator-assessed response, which is a distinct assessment source and should not be conflated with the primary central-review endpoint.

The record specifies a two-sided 90% exact confidence interval based on the Clopper-Pearson method. General statistical explanation: a confidence interval gives a range of values compatible with the observed data and the chosen procedure, reflecting sampling uncertainty rather than guaranteeing that the true response rate lies in that particular interval. For a binary outcome, the exact method is based on the binomial model and is often used when the number of observed events is limited. It can produce an interval that is less narrow than some approximate methods. The actual interval endpoints are not reported in the registry record.

The stated hypothesis is that the response rate is greater than 20%. This identifies a benchmark of interest, but a benchmark statement is not the same as a reported statistical test. The registry record does not provide a primary analysis result, p-value, or a decision about the hypothesis. It also does not report enough information to reconstruct the primary estimate. Therefore, no conclusion about whether the observed cORR exceeded the benchmark can be made from the registry record.

General statistical explanation: a single-group estimate and a comparative treatment effect answer different questions. A response proportion describes outcomes within the treated group. A randomized comparison estimates how outcomes differ between assigned groups under the trial’s design. Since the study is listed as single-group, there is no concurrent control estimate with which to compare response. Historical benchmarks can be useful context, but comparisons with them depend on whether populations, eligibility, endpoint definitions, assessment practices, and follow-up are sufficiently similar. The registry record does not report a historical-control analysis.

The timeframe is from first dose until documented progression or new anticancer therapy, whichever came first, with a maximum up to 17.8 months. This defines the period in which response assessments were considered for the primary outcome. It also means response ascertainment is tied to treatment and follow-up events, rather than to an unrestricted lifetime period. The record does not report the distribution of follow-up, the number of assessments, or the number of participants who reached the maximum follow-up period.

Missing-data handling matters because the record classifies participants with missing data as non-responders. General statistical explanation: such a rule fixes how missing response status enters the binary endpoint, but it does not reveal why observations are missing. If missingness is related to prognosis, treatment discontinuation, or ability to attend assessments, the resulting estimate may differ from the response rate that would have been observed with complete assessments. Treating missing participants as non-responders can affect the estimate; its direction and size cannot be evaluated without the number and reasons for missing assessments.

The registry record lists 22 outcome measures and zero analysis records. That does not show that no statistical work was done; it means analyses are not recorded in the registry data summarized by these registry entries. The planned statistical analysis, adjustment methods, sample-size justification, and interim-analysis procedures are not reported in the registry record. General statistical explanation: over-dispersion is a concern for some count models when variation exceeds what a simple model assumes, but the primary endpoint is a binary response proportion rather than a count outcome. No over-dispersion adjustment is specified for the primary analysis.

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("Tucatinib + Capecitabine + Trastuzumab")
n      <- c(48)
events <- c(17)
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  Tucatinib + Capecitabine + Trastuzumab  48  17  35.4  22.2  50.5  23.4  49.6  23.4  49.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 22 outcome measures, spanning response, duration of response, and progression-free survival, with different assessment sources and populations. It does not report a multiplicity procedure, such as a prespecified order for testing outcomes or an adjustment to account for multiple statistical tests. Nor does it report analysis records or p-values. Therefore, the record does not establish whether any secondary outcome was formally tested, whether testing was adjusted, or how a family of outcomes was interpreted.

The primary outcome is distinct from secondary outcomes that assess response in all participants or use investigator assessment. The primary endpoint concerns Japanese participants and independent central review. These populations and assessment sources should not be treated as interchangeable when interpreting estimates. The number of Japanese participants eligible for the primary analysis and the complete response-evaluable criteria are not reported in the registry record.

For cORR, participants with missing data were considered non-responders. This is the stated missing-data rule for the endpoint. The record does not report the amount or reasons for missing data, how missingness was handled for each secondary outcome, or whether sensitivity analyses assessed other assumptions. It also does not report whether participants were excluded from any analysis population or how exclusions were handled.

General statistical explanation: multiplicity can increase the chance of a false-positive finding when many hypotheses are tested, while missing-data assumptions can affect the estimated outcome. These are separate issues. The registry record does not provide enough information to evaluate the overall inferential strategy across its outcome measures, so statistical significance across outcomes should not be inferred from the outcome list alone.

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: covid-19 (3.0%), febrile neutropenia (1.5%), pneumonia bacterial (1.5%), hypercalcaemia (1.5%), hypophagia (1.5%).

For the listed tucatinib, capecitabine, and trastuzumab group, the registry record reports serious adverse events affecting 6 participants and other adverse events affecting 65 participants. It also lists 15 deaths. These are different safety summaries, and the record does not provide a causal attribution for the deaths or say in this summary whether every death was considered treatment-related. The figures should be read as registry-reported counts, not as evidence that the combination caused each event.

The serious-event terms listed most prominently are COVID-19 at 3.0%, and febrile neutropenia, bacterial pneumonia, hypercalcaemia, hypophagia, and brain oedema at 1.5% each. The registry record does not report the denominator or calculation details for these term-specific percentages in the registry record, so they should not be converted into counts. It also does not specify whether participants could contribute to more than one event category.

Safety summaries depend on definitions, observation time, and attribution practices. The record’s group counts do not indicate how long participants were followed for adverse events, the severity distribution of non-serious events, or the timing and causes of deaths. Those details are not reported in the registry record.

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

The registry record identifies the study’s clinical question and its response definition, but it does not report the primary outcome result. The stated hypothesis concerns whether cORR among Japanese participants is greater than 20%; the observed response percentage, responder count, and confidence interval are not reported in the registry record. Consequently, the posted information does not establish whether the primary result met the stated benchmark or how precisely the response rate was estimated.

Even if the response estimate were available, its interpretation would be bounded by the single-group design. It would describe outcomes in participants who received tucatinib with trastuzumab and capecitabine, not the difference between that combination and a separate control treatment. The study is open-label, and the primary endpoint relies on independent central review. Those design features help describe how the response outcome was assessed, but do not create a randomized comparison.

The primary endpoint is restricted to Japanese participants in a response-evaluable set. It should not be presented as an estimate for all enrolled participants unless the analysis population supports that interpretation. The complete analysis population size is not reported in the registry record. Several secondary outcomes concern all participants, Japanese participants, investigator assessment, or central review; their results are also not reported in the registry record.

The safety information is likewise descriptive. The listed serious-event and other-event counts, deaths, and serious-event terms do not by themselves show whether events were caused by treatment or how risks compare with another group. The registry record does not report comparative safety estimates.

In statistical terms, the record conveys definitions and selected design details, but not the efficacy estimates needed to assess magnitude or precision. It supports no conclusion that the treatment is effective or ineffective, and it does not establish a comparative benefit or harm. Interpretation should remain limited to the endpoints, populations, and safety counts actually reported.

14. Why This Trial Matters Statistically

This trial raises a useful statistical distinction between a response benchmark and a comparative treatment effect. The stated primary hypothesis asks whether cORR is greater than 20% in a defined population. That is a single-group question about whether an observed response proportion is above a benchmark, not a randomized estimate of how outcomes differ between treatments. The distinction matters because a benchmark comparison relies on how well the benchmark represents the study population and assessment context.

The endpoint is also population-specific. The primary outcome concerns Japanese participants enrolled at Japanese sites who met response-evaluable criteria, while some secondary outcomes concern all participants. Separate populations can yield estimates with different precision and applicability. The registry record does not report the primary analysis population size or cORR result, so readers cannot assess those features from the posted primary outcome information.

Using independent central review for the primary endpoint, alongside investigator-assessed outcomes among the secondary measures, highlights that outcome assessment source is part of the estimand. The two sources may classify response differently; the registry record does not report their agreement or comparative results. Identifying the assessment source is therefore essential when reading any eventual response estimate.

The record’s missing-data rule for cORR is explicit: missing participants were treated as non-responders. This makes the handling rule visible, but not the extent or causes of missingness. In addition, the listed outcome measures raise a multiplicity question, while no multiplicity method or analysis records are reported. These gaps illustrate why a defined endpoint alone is not enough to assess statistical evidence: observed estimates, uncertainty, populations, missingness, and analysis procedures all contribute to interpretation.

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 immediate limitation of the registry information is that the primary cORR result is not reported. The observed percentage, number of responders, and 90% confidence interval are absent, as is a primary p-value. Without these, the magnitude and precision of the primary finding cannot be assessed, and the stated hypothesis cannot be evaluated against the posted result.

The study is single-group and open-label. There is no concurrent randomized comparator, so any response or safety description cannot estimate a between-group effect. The record does not provide a historical-control analysis that might offer a separate comparison, and it does not report enough protocol detail to assess how the stated 20% benchmark was selected.

The primary endpoint applies to a Japanese response-evaluable population, but the complete eligibility criteria and analysis population size are not reported in the registry record. Enrollment of 66 is a study-level figure and should not be substituted for the primary endpoint denominator. The record also does not report response counts, missing-assessment counts, or reasons for missingness, despite specifying that missing data were classified as non-response.

The registry record lists multiple outcomes but does not report a multiplicity strategy or analysis records. It does not report the sample-size justification, planned interim analyses, or a detailed statistical analysis plan. The capecitabine dosing details are incomplete in the record summary, and the safety summary does not give event timing, attribution, or detail on the deaths. These omissions limit interpretation of both efficacy and safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Tucatinib (MK-7119) in Combination With Trastuzumab and Capecitabine in Participants With Previously Treated Locally Advanced Unresectable or Metastatic Human Epidermal Growth Factor Receptor 2 Positive (HER2+) Breast Carcinoma (MK-7119-001) — NCT04721977.
• Registry record last updated 2026-09-10; status last verified 2026-08.
• Results first posted 2024-08-09.
• 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

The registry record describes a Phase 2, open-label, single-group study sponsored by Pfizer, with an enrollment of 66 and status active, not recruiting. The study evaluates tucatinib with trastuzumab and capecitabine in previously treated HER2-positive breast cancer.

The primary outcome is confirmed objective response rate by independent central review under RECIST version 1.1 among Japanese participants in a response-evaluable set. The stated hypothesis is that cORR is greater than 20%, and the planned interval is a two-sided 90% exact Clopper-Pearson confidence interval. The observed primary result is not reported in the registry record.

The record lists 22 outcome measures and no analysis records. It reports serious adverse events affecting 6 participants, other adverse events affecting 65 participants, and 15 deaths in the listed group; causal attribution is not established by these figures.