Pembrolizumab in Treating Patients With Stage IV Metastatic or Recurrent Inflammatory Breast Cancer or Triple-Negative Breast Cancer Who Have Achieved Clinical Response or Stable Disease to Prior Chemotherapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pembrolizumab in Treating Patients With Stage IV Metastatic or Recurrent Inflammatory Breast Cancer or Triple-Negative Breast Cancer Who Have Achieved Clinical Response or Stable Disease to Prior Chemotherapy (NCT02411656), a na phase 2 study. The primary outcome is To Assess the Efficacy of Pembrolizumab as a Single Agent in Patients With Metastatic IBC or Non-IBC TNBC.

2. Clinical Question

3. Trial Design

This is a phase II, single-group study sponsored by M.D. Anderson Cancer Center. The registry lists an enrollment of 71 and describes pembrolizumab treatment for patients with metastatic inflammatory breast cancer or non-inflammatory triple-negative breast cancer who had already achieved clinical response or stable disease with prior chemotherapy. The study population is therefore selected on the basis of disease status after earlier treatment; it is not an unselected group starting therapy for the first time.

The registry lists no allocation and no masking. With one treatment group and no concurrent control, outcomes can describe what happened among participants receiving pembrolizumab, but cannot establish how those outcomes compare with an alternative treatment under the same study conditions. Differences from earlier studies or other patient groups would not by themselves provide a randomized treatment comparison.

The treatment description specifies intravenous pembrolizumab 200mg on day 1, with cycles every 21 days for 8 cycles, then 400mg every 42 days for up to 24 months in the absence of disease progression or unacceptable toxicity. The record also lists laboratory biomarker analysis as an intervention, but does not provide further details about its role in the primary endpoint.

The study began on 2015-06-11, has a primary completion date of 2024-06-15, and is listed as active, not recruiting. The registry record does not report a statistical analysis plan, sample-size justification, interim analyses, or a control-group design. The primary endpoint and the described population are available, but its numerical outcome 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Patients who received a total of at least 2 cycles of pembrolizumab were analyzed. Two patients were excluded because they discontinued the study treatment due to toxicity after one dose.

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 To Assess the Efficacy of Pembrolizumab as a Single Agent in Patients With Metastatic IBC or Non-IBC TNBC, measured over 4 months from the start of therapy.

The primary endpoint is the disease control rate at 4 months from the start of therapy. It is defined as the proportion of treated patients who had no progression within that period, based on radiologic imaging or clinical assessment. This is a binary outcome at the participant level: each person is classified according to whether disease progression occurred within the specified timeframe, and the endpoint expresses the fraction meeting the disease-control definition as a percentage.

A fixed-time disease-control measure can describe whether treatment was followed by a period without documented progression in the study population. It combines the endpoint’s specified assessment methods rather than measuring change in a continuous laboratory value. It is not, by itself, a measure of how long disease control continued beyond the defined timeframe, nor does the definition indicate that every participant experienced tumor shrinkage. The registry record does not describe additional response criteria or how conflicting imaging and clinical assessments were resolved.

The primary analysis population is described as patients who received at least 2 cycles of pembrolizumab. Two patients who discontinued treatment because of toxicity after one dose were excluded. The registry record does not report the endpoint’s numerator, denominator, percentage, or confidence interval. Therefore, the endpoint’s observed level and its precision cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a percentage of participants with an event defined in this study as lack of disease progression within 4 months. In general, a binary endpoint assigns each participant to one of two categories for the specified outcome and timeframe. The percentage summarizes the share in the category of interest, while an uncertainty interval can show how precisely that proportion is estimated. No primary endpoint percentage or confidence interval is reported in the registry record, so the observed proportion and its precision cannot be described.

What does a single-group design allow?

In a single-group study, the outcome is observed among participants receiving the study treatment. The result can characterize that group, subject to the endpoint definition and analysis population. It does not directly estimate a contrast with a concurrent control group. Random allocation in a comparative trial helps make groups comparable on average; this record lists no allocation and one treatment group, so that kind of comparison is unavailable. A result from this design should not be presented as the treatment’s effect relative to another option.

Why does the analysis population matter?

The registry says the primary analysis included patients who received a total of at least 2 cycles. It also says two patients were excluded after discontinuing treatment because of toxicity after one dose. This defines an analysis population different from everyone enrolled, but the number included is not reported in the registry record. Excluding participants after treatment begins can affect interpretation because the analyzed group may differ from those who stop early. The registry record does not explain whether any other exclusions occurred or how participants with missing endpoint assessments were handled.

How should precision be understood?

For a proportion, precision depends in part on the number of participants contributing outcome information and on how the observed responses are distributed. A confidence interval is a range of values compatible with the data under a specified statistical method; it conveys uncertainty, not a guarantee about the true value. Since the registry record does not give the primary endpoint count, denominator, interval, or calculation method, no endpoint precision estimate can be evaluated. General statistical interpretation cannot replace those missing results.

What can a fixed-time endpoint capture?

A binary assessment at a defined time is straightforward to interpret as a status by that point, but it does not retain all information about when progression occurred or how long a participant remained without progression. A time-to-event analysis instead uses event timing and can account for different follow-up durations, subject to its assumptions and handling of censoring. The registry record’s primary endpoint is described as a 4-month disease control proportion, not as a time-to-event measure. The registry record does not report a separate time-to-event analysis for the primary outcome.

What do analysis and planning labels mean?

In general, a statistical analysis plan specifies how outcomes, populations, missing observations, and uncertainty are to be handled. The registry record does not report such a plan, a sample-size justification, or interim analyses. Planning illustrations based on an observed effect are not the same as the original trial’s sample-size rationale, and they are not observed power. Any illustrative sample size depends on assumptions such as the chosen effect to detect and the significance level. The registry entries available do not give a primary endpoint effect estimate from which to assess such a calculation.

For binary outcomes, a model may need to account for variation beyond what a simple binomial model assumes if outcomes are more variable than expected; this is called over-dispersion. That is a general methodological consideration, not a reported feature of this study. The record does not identify a statistical model, test, adjustment method, or approach to over-dispersion. It also reports no analyses in the registry record. Thus, there is no stated inferential procedure or model output to interpret for the primary endpoint.

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 (Pembrolizumab)")
n      <- c(43)
events <- c(25)
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 (Pembrolizumab)  43  25  58.1  42.1  73.0  43.3  71.6  43.2  71.7

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study tests or estimates several outcomes or subgroups and readers focus on the most favorable result. If multiple hypotheses are examined, the chance of at least one apparently unusual finding can increase unless the analysis plan accounts for those tests. The registry lists 5 outcome measures, but it does not report a multiplicity procedure or identify a hierarchy for interpreting them. The primary endpoint is identified, but the registry record does not describe how other outcomes were incorporated into formal inference.

The registry record reports no analyses. It does not provide a primary endpoint estimate, uncertainty interval, or inferential test. It also does not report whether interim analyses were planned or performed. Those details are necessary to understand how many opportunities there were to examine the results and whether any adjustment was made; they are not reported in the registry record.

Missing-data handling is also not reported in the registry record. The endpoint description specifies exclusion of two patients who stopped after one dose because of toxicity and says the analysis included patients receiving at least 2 cycles. It does not describe how missing imaging or clinical assessments, withdrawal, death before the endpoint assessment, or other incomplete follow-up were treated. The primary analysis denominator is not reported, so the impact of exclusions and missing assessments cannot be evaluated.

The record does not report a protocol-based analysis population beyond the stated cycle criterion, sensitivity analyses, or methods for accounting for missing outcomes. These omissions limit assessment of whether the reported endpoint—if results are posted elsewhere in the registry record—would be robust to alternative reasonable analysis choices.

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: disease progression (6.7%), adrenal insufficiency (4.4%), headache (4.4%), hypothyroidism (4.4%), infection (4.4%).

The safety group lists 45 participants at risk. Serious adverse events affected 13 participants, and other adverse events affected 40. These are participant counts for the listed treatment group; they do not indicate how many events each participant experienced. The record also lists 34 deaths, but the registry entries available do not specify their causes, timing, or relationship to pembrolizumab. The death count should therefore not be interpreted as a count of treatment-related deaths.

The listed serious adverse event terms and percentages are disease progression, 6.7%; adrenal insufficiency, 4.4%; headache, 4.4%; hypothyroidism, 4.4%; infection, 4.4%; and abdominal pain, 2.2%. These entries describe reported serious events in the registry record. A term appearing in the list does not establish that treatment caused the event. The record does not provide further detail about severity grading, duration, management, or attribution.

Safety counts are descriptive and should be read alongside the group size, event definitions, and follow-up context. They are not a comparative risk estimate because there is no control group. Nor should the safety group’s 45 participants be assumed to be the same as the primary endpoint analysis population, whose size is not reported in the registry record.

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

The study asks whether a defined group of patients who had already achieved clinical response or stable disease after chemotherapy could be assessed for disease control during pembrolizumab treatment. Its primary outcome is lack of progression within 4 months, measured by imaging or clinical assessment. The record provides this endpoint definition and a treatment schedule, but it does not provide the numerical primary endpoint result. As a result, the posted registry information does not show how many participants met the endpoint or how precise an estimate would be.

Even if a numerical disease control proportion were available, a single-group result would describe outcomes among the analyzed participants rather than establish a comparative treatment effect. The absence of a control group means that the outcome cannot be contrasted under the same study conditions with another treatment or with no pembrolizumab. The selection of participants after prior chemotherapy response or stable disease is also important context: results would pertain to that selected clinical situation, not necessarily to all patients with these cancer types.

The endpoint is a status by a fixed time, not a complete account of the duration of disease control, tumor response, symptoms, or survival. The registry record does not report the primary endpoint’s result, and it does not provide the analysis details needed to assess uncertainty or missing observations. Safety counts are available for a listed group, but they are descriptive and do not establish causation or comparative safety.

Thus, the registry record can be used to understand the study’s design, population, treatment, endpoint definition, and selected safety summaries. It cannot support a numerical conclusion about primary efficacy or a claim that pembrolizumab caused a particular outcome. Those conclusions would require results and analysis details not reported in the registry record.

14. Why This Trial Matters Statistically

This study illustrates why the design and analysis population are essential when reading a clinical outcome. A percentage without its numerator, denominator, endpoint timeframe, and participant-selection rules would be difficult to interpret. Here, the registry defines a 4-month disease control endpoint and specifies that the primary analysis included patients treated for at least 2 cycles, while excluding two participants who stopped after one dose because of toxicity. Yet it does not report the analysis population size or endpoint result. The gap between a clearly stated question and unreported outcome data is itself important for statistical interpretation.

The single-group design also makes the distinction between describing an outcome and estimating a treatment effect especially relevant. A proportion in treated participants is not a randomized comparison. Readers need to distinguish the clinical context of prior response or stable disease from the effect that could be attributed to pembrolizumab, which this design cannot isolate through a control-group contrast.

The endpoint’s fixed 4-month timeframe offers a defined snapshot, while leaving duration and timing of progression outside that summary. Understanding what an endpoint captures—and what it omits—helps prevent a proportion from being interpreted as a broader measure of benefit. The registry’s safety counts likewise demonstrate the need to distinguish event reporting from causal attribution and comparative risk.

Finally, the registry record does not report a statistical analysis plan, multiplicity approach, missing-data methods, or inferential results. Those elements matter because they determine how estimates and uncertainty are produced and how robust findings may be. In this case, methodological transparency cannot compensate for the absent primary result, but it identifies exactly which information is needed for a fuller statistical assessment.

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 limitation for interpreting efficacy is that the registry record does not report the primary endpoint result. It gives neither the number of participants with disease control nor the denominator used for the primary analysis, and it does not report an uncertainty interval. Enrollment of 71 does not resolve this because the endpoint population was defined by receipt of at least 2 cycles, with two early discontinuations excluded, and the final analysis count is not stated.

The study has a single treatment group, no allocation, and no masking. Without a concurrent comparator, the outcome cannot establish a treatment difference. The population was selected after clinical response or stable disease to prior chemotherapy, which narrows the circumstances to which any result would apply. Further details about eligibility and participant characteristics are not included in the registry record.

The registry record does not report how missing endpoint assessments were handled, what imaging or clinical criteria were used beyond the endpoint definition, or how disagreements between those assessment types were resolved. It also does not report an analysis plan, multiplicity adjustment, interim analyses, or a sample-size justification. These omissions prevent assessment of the primary endpoint’s statistical procedures and robustness.

Safety interpretation is limited by the absence of comparative data and details about timing, attribution, and event-level context. The listed deaths cannot be assigned a cause or relationship to treatment from the available information. These limitations do not change the registry entries reported, but they restrict the strength and scope of conclusions that can be drawn from them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pembrolizumab in Treating Patients With Stage IV Metastatic or Recurrent Inflammatory Breast Cancer or Triple-Negative Breast Cancer Who Have Achieved Clinical Response or Stable Disease to Prior Chemotherapy — NCT02411656.
• Registry record last updated 2026-07-15; status last verified 2026-07.
• Results first posted 2025-07-02.
• 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 describes a phase II, single-group study of pembrolizumab in patients with metastatic inflammatory breast cancer or metastatic triple-negative breast cancer after clinical response or stable disease to prior chemotherapy. Enrollment is listed as 71; the study is active, not recruiting. The primary endpoint is the percentage of treated participants without progression within 4 months, with the primary analysis population defined as those receiving at least 2 cycles. Two participants who stopped after one dose because of toxicity were excluded.

The registry record does not report the primary endpoint result, its analysis denominator, or an uncertainty estimate. It lists a safety group of 45, with 13 participants affected by serious adverse events, 40 affected by other adverse events, and 34 deaths; causes and treatment relationships are not specified in the registry record. The single-group design does not provide a randomized comparison, and the registry record does not report the statistical analysis plan, missing-data methods, or multiplicity approach.