A Study of Pembrolizumab on the Tumoral Immunoprofile of Gynecologic Cancers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Pembrolizumab on the Tumoral Immunoprofile of Gynecologic Cancers (NCT02728830), a na early/phase 1 study. The primary outcome is Change in Tumor Immune Infiltrates as Measured by PD-L1 Modified H-Score.

2. Clinical Question

The study is described as a pilot investigation of the effect of pembrolizumab on the tumoral immunoprofile of gynecologic cancers of Müllerian origin. Its stated broader goal is to identify potential biomarkers, immune gene expression signatures, and co-stimulatory pathways that may help characterize the effect of immune checkpoint inhibitors on these cancers. Those aims describe the scientific motivation, not a reported finding that any marker or pathway was identified.

The primary question is about a tissue-based biomarker measured at two time points, rather than a direct comparison of clinical outcomes between treatment groups. The primary outcome description names a change in PD-L1 modified H-score, with baseline and 14-21 Days as the timeframe. The registry record does not report the score values, the estimated change, uncertainty around the change, or a test of whether the change differs from zero. It therefore does not establish the direction or magnitude of any observed change.

3. Trial Design

The registry record describes an early phase 1, single-group study with no masking. Allocation is listed as not applicable, consistent with the record's single-group model rather than a randomized comparison. The registry lists an enrollment of 39. This is the registry enrollment figure; it is not the number reported as having matched tissue available for the primary endpoint analysis.

Participants were to receive one dose of 200mg pembrolizumab by IV 14-21 days before surgery. The arm description says that participants would undergo standard surgical cytoreductive surgery as deemed appropriate by their gynecologic oncologist, followed by standard adjuvant chemotherapy as deemed appropriate for their cancer. The description does not provide further detail about the treatment decisions or their timing.

The primary endpoint is a within-person tissue comparison: PD-L1 modified H-score at baseline and 14-21 Days. The registry record says eleven subjects had adequate matched pre- and post-treatment tissue samples for analysis of tumor immune infiltrates. Matched samples permit measurements from the same subject to be compared, but the record does not report why matched samples were unavailable for other enrolled participants or how any missing measurements were handled.

The listed conditions include gynecologic neoplasms, epithelial ovarian cancer, uterine endometrial cancer, fallopian tube cancer, and peritoneal cancer. The listed secondary outcome concerns the frequency and severity of adverse events assessed by CTCAE. The record is marked completed, with a start date of 2016-06 and a primary completion date of 2019-01-31.

Because there is no separate control group, the design does not support a randomized, between-group estimate of pembrolizumab's effect. Any before-and-after change would be observed over the treatment and pre-surgery interval; without a concurrent control, it could not by itself distinguish treatment-related change from other explanations. The registry record does not report a statistical analysis plan, a sample-size justification, or details of interim analyses.

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: Of the 39 subjects that were consented, 24 were screen failures. Nine were screen failures because exclusionary surgery, 6 because of exclusionary pathology, 3 because physician decision, 2 because of participant withdrawals, 1 was lost to follow up, 1 had exclusionary lab values, 1 had a contraindicated illness, and 1 had a language barrier.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Eleven subjects had adequate matched pre- and post- treatment tissue samples for analysis for tumor immune infiltrates.

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: Analysis was conducted only on subjects who completed the study.

6. Primary Endpoint

The registry lists the primary outcome as Change in Tumor Immune Infiltrates as Measured by PD-L1 Modified H-Score, measured over Baseline and 14-21 Days.

The primary endpoint is change in tumor immune infiltrates measured using the PD-L1 modified H-score at baseline and 14-21 Days. The H-score combines the percentage of cells staining at each intensity with the intensity level. The described intensity levels are 0 for no staining, 1+ for weak staining, 2+ for medium staining, and 3+ for strong staining. The score ranges from 0 to 300; lower scores indicate lower PD-L1 expression in the tumor sample, while higher scores indicate stronger expression.

A score of this kind turns staining information from tissue samples into a continuous measure. Comparing a person's baseline and post-treatment scores can describe whether the measured marker changed over that interval. The endpoint is relevant to the study's stated interest in tumor immune profiles and potential biomarkers. It is a tissue marker, however, and the registry record does not establish that a change in this score corresponds to a change in symptoms, disease course, or another clinical outcome.

The registry identifies the outcome as a mean with standard deviation, but does not report the baseline or post-treatment means, the standard deviations, or a mean change. It also does not report a confidence interval or a hypothesis test for change. Eleven subjects are described as having adequate matched tissue samples for this analysis. Consequently, the endpoint's definition and scoring direction are available, but the observed result and its precision are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes the primary outcome as a continuous measure summarized by a mean and standard deviation. It does not report the endpoint's numerical summaries or the statistical method used to compare baseline with follow-up. The concepts below explain how such data are commonly interpreted; they are general statistical explanations, not a description of methods confirmed to have been used in this study.

What does a mean change describe?

For a continuous outcome, a mean summarizes the central value across measured subjects. With paired baseline and follow-up measurements, a mean change would summarize the individual within-subject differences. This is often more directly relevant than comparing two group means as if the measurements came from different people, because paired measurements are linked within each person. The registry record defines a change endpoint but does not report the estimated mean change or confirm how it was calculated.

The standard deviation describes variation in observed measurements around a mean. It is not the same as the standard error, which describes the precision of an estimated mean, and neither is the same as a confidence interval. A confidence interval gives a range of values compatible with the data and the assumptions of the statistical method. Its width reflects uncertainty; a narrower interval indicates more precision, while a wider interval indicates less. No confidence interval for the primary outcome is reported in the registry record.

Why does matching matter?

When the same subject contributes a baseline sample and a later sample, the two observations are paired. A paired analysis uses the within-subject relationship, rather than treating each measurement as independent. In general, accounting for this relationship can improve estimation of change when measurements from the same subject are correlated. The analysis population named in the record consists of subjects with adequate matched pre- and post-treatment tissue. The record does not say whether a paired test, a model, or another approach was used.

A paired analysis also depends on having both measurements for a subject and on a clear definition of the difference being analyzed. In this study, the record identifies baseline and 14-21 Days as the endpoint timeframe, but does not report exact sample collection details or a statistical method. The subset with matched tissue is relevant because subjects without both adequate samples cannot contribute to a simple paired change calculation. The record does not describe how the subset was selected or whether availability of matched tissue could be related to characteristics that also affect the measured score.

What does the H-score represent?

The H-score combines the percentage of cells at different staining intensities with the intensity of staining. The registry describes a 0 to 300 range, with higher scores representing stronger PD-L1 expression in the tumor samples. The score therefore offers a quantitative summary of staining, but it remains a biomarker measure. Interpretation depends on the scoring procedure and the tissue samples available; the registry record does not report information about measurement reproducibility, assessor agreement, or blinding of the scoring process.

For any continuous measure, the mean and standard deviation can conceal features of the distribution. For example, observations may be skewed or contain unusually high or low values. General statistical practice is to examine the distribution and consider whether a mean-based summary and its analysis are appropriate. The registry record does not report distributional checks or any alternative summaries, so it is not possible to assess whether such issues were evaluated for this endpoint.

What can a single-group comparison establish?

A within-subject before-and-after comparison describes the change observed over time among those measured. It does not, on its own, isolate the effect of the study drug. Without a concurrent comparator, other explanations for a change cannot be separated from a treatment effect using randomization between groups. This limitation applies even when the same person serves as their own baseline reference: the baseline measurement is not a contemporaneous untreated control.

A statistical test, if performed, would address a particular question under specified assumptions. A p-value would measure how compatible the observed data are with a null model; it would not measure the probability that the null hypothesis is true, nor would it establish clinical importance. A confidence interval would add information about precision and the range of effects compatible with the data. Neither a p-value nor an interval is reported for the primary endpoint in the registry record. The record also does not describe a pre-specified analysis method, adjustment for covariates, or an approach to outliers.

Why can small analysis sets be imprecise?

In general, analyses based on a limited number of paired observations can yield imprecise estimates, particularly when individual changes vary substantially. The eleven matched samples identified in the record define the available tissue-analysis population, but no numerical endpoint estimates or interval are reported. The enrollment count should not be substituted for the number with matched tissue in this particular analysis. Nor can an unreported effect or variance be reconstructed from the enrollment figure.

Planning calculations are different from observed results. A sample-size illustration based on an assumed effect and statistical conditions is a planning tool, not an outcome of the trial and not observed power. The registry record does not report the study's own sample-size justification or analysis plan. It therefore does not support claims about whether the study was designed with a particular power or whether an analysis was planned at an interim point.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Pembrolizumab")
n   <- c(11)
m   <- c(13.2)
sp  <- c(23.2)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Pembrolizumab  11  13.200  23.200  -2.386  28.786

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists four outcome measures and no analyses. It names the primary outcome and describes a secondary outcome concerning the frequency and severity of adverse events assessed by CTCAE, but does not provide a full statistical analysis plan or results for the listed outcomes. It is not reported in the registry record whether hypotheses were prioritized beyond the primary designation, whether multiple comparisons were adjusted, or whether any outcome analyses were planned as exploratory.

Multiplicity arises when several outcomes, time points, subgroups, or statistical tests are examined. In general, testing many questions can increase the chance of at least one apparently notable result arising by chance. Adjustment methods or a pre-specified hierarchy can address that issue, but the record does not say whether either was used. The broader scientific goal mentions potential biomarkers, immune gene expression signatures, and co-stimulatory pathways; the record does not report analyses or findings for those aims.

For the primary endpoint, the record says eleven subjects had adequate matched pre- and post-treatment tissue samples. It does not report the number with baseline tissue alone, follow-up tissue alone, or other reasons that subjects may not have contributed matched samples. It also does not report a missing-data method, imputation, sensitivity analysis, or a comparison of subjects with and without matched samples. These omissions prevent assessment of whether missing tissue could have affected the estimated change.

The record does not define an analysis population beyond the matched-sample statement, nor does it report whether all enrolled subjects were included in safety summaries or how exposure was handled in those summaries. The denominator figures in the adverse-event record are described separately below. No assumption about the handling of missing outcome or safety information should be made in the absence of further reporting.

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: pleural effusion (6.7%), allergic reaction (6.7%), febrile neutropenia (6.7%).

The registry lists a pembrolizumab group with 15 at risk for serious adverse events and 3 affected by serious adverse events. It also lists 15 at risk for other adverse events and 12 affected by other adverse events. The record lists 3 deaths. These are registry safety counts; the record does not provide a randomized control group for comparison.

The listed serious adverse event terms are pleural effusion, allergic reaction, and febrile neutropenia, each reported as 6.7%. These are term-level frequencies, not an estimate of comparative risk. The registry record does not report whether the terms were considered related to pembrolizumab, whether a person could contribute more than one event term, or how the timing of events related to treatment.

Seriousness and severity are different concepts: seriousness generally refers to regulatory criteria such as outcomes or the need for substantial medical intervention, while severity describes intensity. The record lists a secondary outcome concerning frequency and severity assessed by CTCAE, but the figures above do not provide a complete severity breakdown. The listed deaths should not be assumed to have been caused by treatment; attribution is not reported in the registry record. With no control group and limited contextual detail, these figures describe reported events but do not establish comparative safety.

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

The record establishes that this was a completed, single-group pilot study of pembrolizumab before surgery in gynecologic cancers of Müllerian origin. It defines a primary tissue endpoint based on PD-L1 modified H-score and states that eleven subjects had adequate matched tissue samples for analysis. It also records selected adverse-event counts and terms. These details describe the study's design, endpoint, analysis subset, and safety reporting.

The record does not report baseline and follow-up H-scores, a mean change, standard deviations, a confidence interval, or a statistical test for the primary endpoint. Thus, it does not establish whether the measured score increased, decreased, or remained similar. Without an effect estimate and a measure of uncertainty, the magnitude and precision of any change cannot be evaluated from the registry record.

The single-group structure also limits causal interpretation. A before-and-after difference, if reported, would not be equivalent to a randomized comparison with a control group. It could describe a change over the observed interval, but the design alone would not separate a treatment effect from other changes over time or influences associated with the clinical care described in the record.

The study's stated goal of identifying possible biomarkers and immune pathways is a research aim, not a finding. The registry record does not report that a biomarker signature was discovered or validated, or that the H-score predicts a clinical outcome. The primary endpoint is a tumor tissue measure, and no link to patient-level clinical benefit is reported. Likewise, the adverse-event counts do not, without attribution and a control group, quantify treatment-related risk relative to another option.

In short, the registry entry is informative about what the investigators set out to measure and which matched samples were available, but it does not contain the primary numerical results needed to judge the observed tissue change. It supports description of the study, not a conclusion about efficacy, biomarker performance, or comparative safety.

14. Why This Trial Matters Statistically

This study is statistically notable for the distinction between enrollment and the endpoint-specific analysis population. Although the registry lists enrollment of 39, it identifies eleven subjects with adequate matched tissue samples for the primary tumor immune infiltrate analysis. For a paired biomarker endpoint, availability of both samples is central: the relevant analysis is based on matched observations, not simply on everyone who enrolled. The record does not explain why the samples were unavailable for others, making the analysis subset's representativeness difficult to assess.

The endpoint also illustrates the difference between a biological measurement and a clinical outcome. A tissue score can help characterize a tumor marker, and the H-score offers a structured way to quantify staining intensity and proportion. But a change in a biomarker does not automatically demonstrate a change in a patient's clinical outcome. The registry's broader biomarker aims should therefore be distinguished from evidence of clinical benefit or a validated predictive marker, neither of which is reported.

The single-group design is important when interpreting any temporal change. Each participant's baseline measurement provides a reference, but it does not create an untreated concurrent comparison. The observed change cannot be attributed to pembrolizumab alone on the basis of the design as recorded. This is especially relevant when interpreting a pilot study as a source of hypotheses rather than a controlled estimate of comparative effects.

Finally, the registry record demonstrates why transparent reporting of results and analysis methods matters. It specifies the endpoint and the sample availability but does not report the numerical endpoint results, uncertainty, or analysis approach. Without those details, readers cannot assess the size and precision of the observed change, evaluate assumptions, or determine how missing tissue affected inference. The record supports understanding the research question and its constraints, but not a quantitative conclusion about the primary endpoint.

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 this specific registry entry is that the primary outcome is defined but its numerical findings are not reported in the registry record. No baseline or follow-up score summaries, mean change, standard deviation, confidence interval, or test result are available. Consequently, the direction, magnitude, and precision of the primary endpoint cannot be evaluated.

The endpoint analysis is described for eleven subjects with adequate matched tissue, while registry enrollment is 39. The record does not explain the reasons that matched samples were unavailable for others, characterize the analysis subset, or report a missing-data strategy. If sample availability were related to patient or tumor characteristics, results among those with matched tissue might not represent the broader enrolled group; whether that occurred is not reported.

The study has a single-group design and no masking. There is no concurrent control group to distinguish an observed pre-post change from other changes over time. The record describes surgery and subsequent adjuvant chemotherapy as part of planned care, but does not provide enough detail to separate their timing or influence on the tissue outcome. The record also does not report how tissue scoring was performed, whether assessors were blinded, or the reproducibility of the H-score measurements.

The statistical analysis plan, sample-size justification, and any interim analyses are not reported in the registry record. The record does not provide a statistical method for the primary comparison, distributional checks, handling of outliers, or multiplicity procedures. These absences limit assessment of how the endpoint was analyzed and how robust an estimate would be.

Safety reporting is also limited for comparative interpretation. The record lists event counts and selected serious event terms but does not report attribution, timing, or a control-group comparison. The death count is listed without a reported cause or treatment relationship. These limitations mean the safety figures should be read as descriptive registry information rather than evidence of comparative risk.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Pembrolizumab on the Tumoral Immunoprofile of Gynecologic Cancers — NCT02728830.
• Registry record last updated 2024-02-21; status last verified 2024-02.
• Results first posted 2020-07-09.
• 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

NCT02728830 is a completed early phase 1, single-group study of pembrolizumab in gynecologic cancers of Müllerian origin. The registry lists enrollment of 39 and describes one 200mg IV dose 14-21 days before surgery, followed by standard surgical cytoreductive surgery and standard adjuvant chemotherapy as deemed appropriate by the gynecologic oncologist.

The primary outcome is change in tumor immune infiltrates measured by PD-L1 modified H-score at baseline and 14-21 Days. Eleven subjects had adequate matched tissue samples for this analysis. Although the endpoint is defined and the H-score range and direction are described, the primary numerical results and analysis are not reported in the registry record.

The registry lists serious adverse events in 3 of 15 at risk, other adverse events in 12 of 15 at risk, and 3 deaths. It also lists pleural effusion, allergic reaction, and febrile neutropenia, each at 6.7%, among the serious event terms. The record does not establish causation or provide a control-group comparison.