Neoadjuvant Pembrolizumab: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Neoadjuvant Pembrolizumab (NCT02818920), a na phase 2 study. The primary outcome is Surgical Feasibility Rate as Measured by the Number of Subjects Who Undergo Surgery Following Neoadjuvant Pembrozulimab.

2. Clinical Question

The registry frames the study as an investigation of pembrolizumab around surgery in this disease setting. It also describes interest in immune effects and tumor-infiltrating lymphocytes. Those topics do not replace the listed primary endpoint: the endpoint concerns whether surgery took place within the stated window under the stated eligibility and treatment conditions.

Because no randomized comparison is recorded, the study cannot by itself distinguish an outcome associated with the treatment strategy from outcomes that might have occurred under another strategy. The endpoint definition is operational: it classifies specified circumstances as feasible and all other situations as infeasible. The registry record does not report a primary endpoint count, proportion, confidence interval, or test result. Therefore, the clinical question is identifiable from the registration, but its observed answer cannot be assessed from the recorded primary outcome information.

3. Trial Design

The registry describes a phase 2 study with a single-group model and no masking. Allocation is listed as NA, consistent with the absence of a randomized comparison. Enrollment is recorded as 35. The study status is completed, and the registry gives a start of 2017-01 and a primary completion date of 2019-03-19.

The intervention is pembrolizumab before and after surgery. The brief summary describes two doses before surgery and 4 doses after surgery. The primary outcome concerns whether participants underwent surgery during the specified window after initiating neoadjuvant pembrolizumab. The intervention description also places treatment after surgery, but the primary endpoint is specifically about surgery occurring within the window.

A single-group design can describe outcomes among participants receiving the study strategy, but it does not create a concurrent untreated or alternative-treatment comparison. Thus, differences between this group and an external group would not be randomized treatment effects, and no such comparison is recorded as the trial's design. The registry also describes the study as multi-institutional; it does not list the institutions in the information recorded.

The registry lists 15 outcome measures and 0 analyses. This does not establish whether analyses were or were not performed outside the registry record. The statistical analysis plan, the handling of missing outcome data, the planned analysis population, and any sample-size justification for the actual single-group study are not reported in the registry record. The adverse-event summary uses 30 participants at risk, while enrollment is 35; the reason for that difference is not reported in the registry record. These design and reporting details matter when judging how far the recorded summaries can support conclusions about the intended population.

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: Thirty-five subjects consented to the study. Two subjects withdrew consent before receiving protocol treatment and three were deemed ineligible.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Subjects who completed surgery.

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: Subjects who completed surgery.

6. Primary Endpoint

The registry lists the primary outcome as Surgical Feasibility Rate as Measured by the Number of Subjects Who Undergo Surgery Following Neoadjuvant Pembrozulimab, measured over 29-56 days after initiation of pembrolizumab.

The primary endpoint is the count of participants classified as surgically feasible. The registry defines feasibility as meeting eligibility criteria, receiving at least 1 dose of pembrolizumab, and undergoing surgery within 29-56 days after initiation of pembrolizumab. All other situations are classified as infeasible. The endpoint's recorded unit is participants, and its population is described as subjects who completed surgery.

This endpoint is used to describe whether the planned preoperative treatment strategy was followed by surgery in the specified time window. In a study conducted before an operation, the ability to proceed to surgery within a prespecified interval is a practical outcome of interest. That is a general explanation of the endpoint's role; it is not evidence about the results of this study.

The definition makes the classification rule explicit, but the registry record does not report how many participants met it. It also does not report a proportion, uncertainty interval, or hypothesis test for surgical feasibility. The listed population wording and definition are not fully interchangeable: the definition requires surgery in the time window, whereas “subjects who completed surgery” does not itself state the timing or all eligibility requirements. How the analysis population was operationalized is not reported in the registry record. Without the observed count and denominator, the primary endpoint's result cannot be quantified from the registry information.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What kind of outcome is the primary endpoint?

The primary endpoint is recorded as a count of participants, with each participant classified according to a binary rule: feasible or infeasible. A binary outcome has two possible classifications for the endpoint, even when the registry's displayed measure is a count. The registry definition combines several conditions: eligibility, receipt of at least 1 dose, and surgery in the interval 29-56 days after treatment initiation. All other situations count as infeasible under that definition.

General explanation: a binary endpoint may be summarized by the number meeting the definition and by the number or proportion in the analysis population. Those summaries answer different questions: a count gives the observed number, while a proportion relates that count to a specified denominator. Neither primary endpoint summary is reported in the registry record. The denominator and analysis population used for a primary endpoint analysis are also not reported in the registry record.

What can a single-group design estimate?

The recorded model is single group. General explanation: a single-group study can summarize outcomes observed in the enrolled group. It does not, by design, provide a randomized estimate of the difference between the treatment strategy and a concurrent alternative. Without such a comparator, external comparisons would involve other sources of variation and would not have the same interpretation as a randomized contrast.

The registry records allocation as NA and masking as none. It does not record a control arm. Consequently, analyses framed as treatment-versus-control comparisons do not describe the registered design. The statistical analysis plan is not reported in the registry record, so the specific estimand, inferential procedure, or adjustment strategy for the primary outcome cannot be established.

How should uncertainty be understood?

General explanation: an estimate is a summary of observed data, while a confidence interval describes the range of values compatible with the data under a specified statistical procedure and its assumptions. A narrower interval indicates greater precision under those assumptions; a wider interval indicates more uncertainty. An interval does not remove design limitations, such as the lack of a randomized comparator, and it does not by itself establish a causal effect.

No primary endpoint estimate, confidence interval, or p-value is reported in the registry record. Thus, the precision of the surgical-feasibility result cannot be evaluated from the recorded endpoint information. No significance level or primary-outcome testing procedure is reported in the registry record.

What does the sample-size illustration represent?

The stated planning assumptions describe an illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The illustration is not the trial's recorded single-group design. It describes hypothetical planning under a two-arm setup and should not be read as the design actually used in this study.

General explanation: sample-size planning asks how many participants might be needed under chosen assumptions to detect a specified effect using a chosen statistical procedure. Such a calculation is conditional on those assumptions. It is not an observed study outcome, and it does not indicate the probability that the trial's result is correct. The stated illustration refers to the effect as a fraction of the observed effect, but numerical planning results are not reported in the registry record. The actual study's sample-size justification is not reported in the registry record.

Why can count data need special care?

General explanation: counts can be analyzed using methods that reflect their distribution and the way participants enter the denominator. For binary participant outcomes, the number meeting the endpoint is tied to the number assessed. If participants are not assessed or the analysis population differs from enrollment, the denominator and missing-data handling affect interpretation. The registry record does not report those details for the primary endpoint.

Over-dispersion is a general concern when count outcomes vary more than a chosen statistical model expects; ignoring it can make uncertainty estimates too optimistic. That concept does not establish that over-dispersion occurred in this trial. The primary outcome is recorded as a participant count with a binary classification, and the record does not specify a model or report diagnostics.

What is available for safety interpretation?

The safety summary gives counts of participants at risk and affected, along with deaths and selected serious adverse event terms. General explanation: adverse-event counts are descriptive and depend on the definitions, observation period, and population used for reporting. A count of participants with at least one event is not the same as a count of all individual events; the registry record does not give a full event-level analysis in the listed summary.

The listed summary reports 30 at risk for both serious and other adverse events. The primary endpoint denominator, follow-up details for safety, and the relationship between the adverse-event population and enrollment are not reported in the registry record. Accordingly, the safety data can be described as recorded, but should not be treated as a comparative estimate of risk.

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("Pembrolizumab Prior to and After Surgery")
n      <- c(25)
events <- c(24)
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  Pembrolizumab Prior to and After Surgery  25  24  96.0  79.6  99.9  80.5  99.3  80.4  99.1

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 15 outcome measures and 0 analyses. A list of multiple outcome measures raises a general statistical issue: when many hypotheses are tested, the chance of at least one apparently notable result can increase unless the analysis plan addresses multiplicity. This is a general explanation and does not imply that all listed outcomes were tested or that any particular adjustment was required in this study.

The registry record does not report a multiplicity strategy, a hierarchy of outcomes, or which outcomes were treated as confirmatory versus exploratory. It also does not report a primary endpoint p-value or other inferential result. Therefore, it is not possible to determine from the record whether multiplicity was addressed in analyses conducted for the study. The listed 0 analyses should be read as a registry entry, not as proof that no analyses took place.

Missing-data handling is also not reported in the registry record. The primary endpoint definition classifies all situations other than the specified conditions as infeasible, but the record does not explain how missing eligibility information, treatment records, or surgery timing information were handled. Nor does it state whether participants without an observed endpoint were excluded, classified in a particular way, or addressed through a sensitivity analysis. These choices can affect the denominator and the meaning of an estimated rate.

The analysis population is not reported in the registry record beyond the endpoint's description as subjects who completed surgery. That wording leaves unclear how participants who did not complete surgery relate to the endpoint denominator, despite the endpoint definition classifying other situations as infeasible. The registry does not report a protocol-deviation approach, an intention-to-treat population, or another named analysis set. These gaps prevent a detailed assessment of multiplicity control, missing-data bias, and whether the analyzed population matches the intended population.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: abdominal pain (5.9%), lung infection (5.9%), respiratory, thoracic and mediastinal disorders - other (5.9%), supraventricular tachycardia (4.0%), thromboembolic event (4.0%).

The registry safety summary lists 30 participants at risk for serious adverse events and 6 participants affected. It lists 30 participants at risk for other adverse events and 30 affected. The summary also records 0 deaths. These are participant counts as recorded, not comparisons against another treatment group.

The listed serious adverse event terms include abdominal pain, lung infection, and respiratory, thoracic and mediastinal disorders - Other, each reported as 5.9%; supraventricular tachycardia and thromboembolic event, each 4.0%; and infusion related reaction, 3.3%. The record does not provide event-level details or explain the calculation basis for each displayed percentage in the safety summary. The listed terms should not be interpreted as a complete account of all events.

General explanation: serious adverse events and other adverse events are separate reporting categories, and the count of participants affected is not necessarily the count of events. A participant may experience more than one event, but participant-level totals do not reveal that event pattern. The registry's safety summary is descriptive; without a comparator, it does not establish whether the recorded event frequency differs from another treatment strategy. The reason the safety population of 30 differs from the enrollment of 35 is not reported in the registry record.

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

The registry establishes that this was a completed, phase 2, single-group study of pembrolizumab given before and after surgery, with enrollment of 35. It also defines the primary endpoint as surgical feasibility under a stated eligibility, treatment, and timing rule. These details clarify what the study was designed to examine, but they do not show how many participants met the endpoint.

The registry record does not report the primary outcome result. It therefore cannot support a numerical estimate of surgical feasibility, a statement about the precision of such an estimate, or an inferential claim about the endpoint. This is a limitation of the available registry record, not evidence that the outcome was favorable or unfavorable.

The single-group design does not compare the strategy with a randomized control group. General explanation: outcomes observed in one group may reflect multiple influences, and without a concurrent randomized comparator the study cannot isolate a treatment effect in the same way a randomized contrast can. This design feature should guide interpretation even if outcome results are available from another source.

The safety summary records 6 participants with serious adverse events among 30 at risk, 30 participants with other adverse events among 30 at risk, and 0 deaths. Those figures describe the registry's safety summary; they do not answer the primary surgical-feasibility question or establish comparative safety. The relationship between the safety population and the enrolled population is not reported in the registry record.

Overall, the record supports a description of the study question, design, endpoint definition, and selected safety counts. It does not report the primary endpoint result, analysis methods, or the details needed to assess uncertainty and missingness. Any broader interpretation should distinguish those documented features from conclusions that the registry information cannot establish.

14. Why This Trial Matters Statistically

This trial is statistically notable for the distinction between a well-defined endpoint and an available endpoint result. The registry gives a specific rule for classifying surgical feasibility: eligibility, receipt of at least 1 dose, and surgery within 29-56 days after pembrolizumab initiation. Yet it does not report the number meeting that rule. A precise definition is useful, but it cannot substitute for the observed count, denominator, and uncertainty estimate needed to interpret the outcome.

The single-group design also makes the target of inference important. A description of outcomes among participants receiving the study strategy is different from an estimate of how that strategy compares with another approach. Since no randomized comparator is recorded, a treatment contrast is not available from the registered design. This is a general statistical distinction that remains relevant regardless of the direction of any unreported outcome.

The registry's 15 outcome measures and 0 listed analyses highlight the importance of transparent reporting about analysis plans, outcome hierarchy, multiplicity, and missing data. Those details are not reported in the registry record. Without them, readers cannot determine which analyses were intended to carry confirmatory weight or how incomplete endpoint information was handled.

The safety summary further illustrates why denominators matter: enrollment is 35, while the adverse-event summary lists 30 at risk. The record does not explain the relationship. Making that relationship clear is important for understanding which participants contribute to a safety count and how broadly it applies. These issues do not invalidate the recorded design or counts, but they limit the precision and scope of statistical 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 central limitation in interpreting this trial from the registry record is that the primary endpoint result is not reported. Although the surgical-feasibility definition is stated, the number classified as feasible or infeasible, the analysis denominator, and any uncertainty estimate are absent. The endpoint population is described as subjects who completed surgery, while the definition refers to eligibility, treatment receipt, and surgery in a specified interval. How these descriptions were reconciled for analysis is not reported in the registry record.

The study is single group, with no randomized comparator recorded. This limits what can be learned about differences between the study strategy and an alternative. The registry also records no masking. The potential implications depend on outcome measurement and conduct details that are not reported in the registry record.

The registry lists enrollment of 35, but the safety summary lists 30 participants at risk. The reason for this difference, including whether it reflects eligibility, exposure, follow-up, or another factor, is not reported in the registry record. The safety summary provides selected terms and percentages, but does not supply the event-level details or observation information needed for a fuller assessment.

The statistical analysis plan, sample-size justification for the actual design, missing-data methods, analysis populations, multiplicity approach, and inferential results are not reported in the registry record. These omissions limit evaluation of the primary outcome's precision and of how representative the analyzed participants are of enrollment. The registry status of completed does not fill those reporting gaps, and the listed 0 analyses does not establish whether analyses were conducted elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Neoadjuvant Pembrolizumab — NCT02818920.
• Registry record last updated 2026-03-17; status last verified 2025-07.
• Results first posted 2020-02-18.
• 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 NCT02818920, titled “Neoadjuvant Pembrolizumab,” as a phase 2, single-group study for stage IB, II or IIIA non-small cell lung cancer. Enrollment is 35, and the status is completed. The intervention is pembrolizumab before and after surgery; the brief summary describes two doses before surgery and 4 doses afterward.

The primary endpoint is a participant count based on a surgical-feasibility definition involving eligibility, at least 1 dose, and surgery within 29-56 days after initiation. The primary endpoint result and its statistical analysis are not reported in the registry record. The safety summary lists 6 serious adverse events among 30 at risk, 30 participants with other adverse events among 30 at risk, and 0 deaths. The reason for the safety denominator is not reported in the registry record.