Switch Maintenance Pembrolizumab in Patients With NSCLC After First Line Platinum Doublet Chemotherapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Switch Maintenance Pembrolizumab in Patients With NSCLC After First Line Platinum Doublet Chemotherapy (NCT02705820), a na phase 2 study. The primary outcome is Immune Related Progression Free Survival (irPFS) at 1 Year.

2. Clinical Question

This is a single-arm question about an outcome benchmark, not a randomized test of pembrolizumab against another treatment. The study intended to assess whether the proportion of participants meeting its one-year irPFS endpoint was consistent with the target described in the registry. Because the benchmark comes from historical trials, interpretation depends on whether their participants, treatment context, outcome definitions, and follow-up are sufficiently comparable to this study. The registry does not report details that would establish that comparability.

The population was selected after first-line chemotherapy: only people who did not progress after 4-6 cycles were offered maintenance treatment. The question therefore concerns that post-chemotherapy group, not all people with metastatic non-small cell lung cancer or all people starting first-line treatment. The primary outcome result itself is not reported in the registry record, so the record does not establish whether the stated target was met.

3. Trial Design

The registry describes SWIPE as a phase II, single-group study of switch maintenance pembrolizumab for metastatic non-small cell lung cancer. The study planned enrollment of 48 participants. Participants were to receive 4-6 cycles of first-line platinum doublet chemotherapy; those without progression were offered pembrolizumab as maintenance therapy. There was no concurrent control arm, and allocation is listed as NA.

The stated primary endpoint was immune-related progression-free survival at 1 year. The study description says the aim was to show a rate of at least 25%, compared with an expected 12% one-year PFS based on Pemetrexed and Erlotinib maintenance trials. It identifies a one-stage phase II Fleming's design. The endpoint record also gives a timeframe of 4 years, while the endpoint title and description refer to one-year irPFS. The registry does not explain this difference in wording.

The study is listed as completed. The registry lists its start as 2016-04-25 and primary completion as 2023-05-02. Completion status alone does not indicate whether the endpoint met its target. The posted registry record does not report the one-year irPFS count or rate, the decision threshold for the design, or the statistical analysis details needed to reproduce the planned test.

Secondary outcomes include response rates under RECIST Version 1.1 and immune-related response criteria, radiological PFS under RECIST, immune-related PFS, overall survival, and treatment-related adverse events assessed using CTCAE v4.0. The record lists 7 outcome measures but no analyses. The absence of reported analyses means that outcome-specific results and their precision cannot be assessed from the registry information.

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:

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: Pembrolizumab is administrated at 200 mg fixed dose i.v. on day 1 of every 3 week cycle for a maximum or 2 years (expected maximum of 36 doses), or until progression of disease determined according to immune related Response Criteria (irRC) or lack of tolerability, or until the patient declines further treatment for patients that have not progressed to first line chemotherapy for advanced / metastatic NSCLC.

6. Primary Endpoint

The registry lists the primary outcome as Immune Related Progression Free Survival (irPFS) at 1 Year, measured over 4 years.

The primary endpoint is titled Immune Related Progression Free Survival (irPFS) at 1 Year. Progression-free survival generally describes the time or status of participants before disease progression or death, depending on the study's endpoint definition. An immune-related version uses immune-related response criteria to assess progression. That distinction can matter when interpreting treatment-related imaging changes, because the response criteria used determine how progression is classified. These are general features of the endpoint concept; the registry record does not provide the detailed operational definition or event-handling rules for this study.

The one-year assessment gives the study a fixed time point for judging whether participants remain progression-free under the stated immune-related criteria. A fixed-time proportion can be compared with a historical benchmark, which is consistent with the registry's stated aim of evaluating a target of at least 25% against an expected 12% one-year PFS. The endpoint is described as a count of participants, but the registry does not report the number meeting the endpoint, the denominator used for the primary analysis, or the resulting rate.

The registry lists a 4 years timeframe for the primary outcome even though its title specifies one-year irPFS. It does not clarify how these labels relate to follow-up or endpoint ascertainment. The result and its precision are not reported in the registry record, so it is not possible to judge from that record whether the target was attained.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical plan described in the registry is a one-stage phase II Fleming's design. In general, this design assesses a single-group response or event proportion against a prespecified reference value, using a decision rule defined before the study. A one-stage design bases its planned decision on the study's accumulated observations rather than dividing enrollment into multiple planned stages. This general explanation does not establish what exact decision rule was used in SWIPE. The registry record does not report the cutoff or the full statistical plan.

What a historical benchmark can do

A historical benchmark gives a single-group study a reference for judging whether an observed outcome is sufficiently promising under its design. In this study, the registry states a target of at least 25% for one-year irPFS and an expected 12% one-year PFS based on Pemetrexed and Erlotinib maintenance trials. Those values explain the stated rationale; they are not results from the pembrolizumab group. A historical comparison is not equivalent to randomization. Differences between the current participants and historical trial populations, as well as differences in follow-up and outcome assessment, can influence a comparison. The registry record does not provide enough detail to determine how the historical evidence was selected or made comparable.

Why outcome definitions matter

The primary endpoint uses immune-related progression-free survival, while some listed secondary outcomes use RECIST Version 1.1 and others use immune-related response criteria. In general, changing the criteria for classifying progression or response can change which participants are counted as having an event or response. The same endpoint label is meaningful only alongside a clear definition of assessment timing, event rules, censoring, and the analysis population. Those operational details are not reported in the registry record. The listed 4 years timeframe for the primary outcome is also not reconciled with its one-year title.

Interpreting estimates and precision

A point estimate, if reported, would describe the observed endpoint in the analyzed participants. A confidence interval would express uncertainty around that estimate under the assumptions of the analysis; a narrower interval generally indicates greater precision, while a wider interval indicates more uncertainty. Neither a primary endpoint estimate nor a confidence interval is reported in the registry record. Therefore, neither the position of the result relative to the target nor the precision of that result can be assessed.

The enrollment of 48 is the planned or recorded enrollment count in the registry, not necessarily the denominator for the primary endpoint analysis. Participants may be excluded from a particular analysis or have incomplete endpoint follow-up, depending on study rules. The registry record does not report how many participants were included in the primary analysis, how missing endpoint observations were handled, or whether an intention-to-treat or other analysis population was used.

Statistical significance and design decisions

The registry identifies a one-stage Fleming's design but does not report a significance level, power target, sample-size justification, or decision threshold. It is therefore not possible to reconstruct the planned operating characteristics or determine what result would have qualified as meeting the design criterion. The historical target and benchmark alone do not supply those missing details. Interim analyses or monitoring plans are not reported in the registry record.

In general, statistical significance is not the same as clinical importance, and a design decision against a benchmark does not demonstrate a treatment's causal effect. Interpretation also depends on effect size, uncertainty, endpoint validity, population, and potential bias. In this single-group study, a reported comparison with historical outcomes would remain vulnerable to differences between study settings. As no primary result is posted, even a statistical comparison with the stated benchmark cannot be assessed.

Safety summaries and event counts

Safety counts describe participants or events according to the registry's categories, but they need a defined observation period and clear rules for attribution and repeated events. A count of affected participants is not necessarily a count of all event episodes. The registry's serious-event summary reports participants at risk and affected, while the separate other-adverse-event field also gives participants at risk and affected. These categories should be read as the registry reports them and not combined into a single measure. The record does not provide enough detail about event collection, exposure time, or analysis methods to calculate rates or compare safety between treatments.

In general, over-dispersion is relevant when count data vary more than a simple statistical model expects; ignoring it can make uncertainty estimates too small. The primary outcome is recorded as a count of participants, but the registry does not report a count model or a dispersion estimate. There is no basis in the record for claiming that an over-dispersion adjustment was used or needed.

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("Single Arm Study")
n      <- c(48)
events <- c(16)
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  Single Arm Study  48  16  33.3  20.4  48.4  21.7  47.5  21.7  47.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists several secondary outcomes in addition to the primary one-year irPFS endpoint: response rates using RECIST Version 1.1 and immune-related response criteria, radiological PFS using RECIST, immune-related PFS, overall survival, and treatment-related adverse events assessed by CTCAE v4.0. It lists 7 outcome measures but reports no analyses. The record does not describe a multiplicity adjustment, a hierarchy among secondary outcomes, or which comparisons, if any, were designated confirmatory. Without those details and the outcome results, the chance of false-positive findings across multiple tests cannot be evaluated.

The record does not report the primary endpoint count or rate, the analysis denominator, confidence intervals, or the handling of missing observations. These details matter because participants may not all have complete follow-up at the endpoint assessment, and the chosen analysis population can affect the estimate. The registry does not state whether all enrolled participants were included in the primary analysis or how any incomplete data were treated.

Analysis populations such as intention-to-treat, per-protocol, or safety populations are not specified in the registry record. Interim analyses and any related adjustments are also not reported. These omissions prevent a full assessment of the study's analysis strategy. They should not be filled in by assuming standard practice; the available record supports only the stated design and listed outcomes.

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: dyspnoea (6.2%), infections and infestations (4.2%), hypotension (2.1%), myocardial infarction (2.1%), pericardial effusion (2.1%).

The serious adverse event summary lists 48 participants at risk and 13 affected in the single-arm group. It also lists 48 at risk and 48 affected for other adverse events, and 42 deaths. These are counts in separate registry fields. The record does not explain their observation periods, whether deaths were treatment-related, or how adverse events were attributed to treatment. They should not be interpreted as a comparison between treatments, because there was no concurrent control arm.

The registry lists serious-event terms with percentages: Dyspnoea, 6.2%; Infections and Infestations, 4.2%; Hypotension, 2.1%; Myocardial Infarction, 2.1%; Pericardial Effusion, 2.1%; and Adrenal Insufficiency, 2.1%. These entries describe selected serious-event terms in the registry summary. They do not provide a complete account of all adverse events, the number of episodes for each term, timing, severity details, or the relationship to pembrolizumab. The adverse-event method is listed as CTCAE v4.0 among the secondary outcomes, but further safety-analysis details are not reported.

These counts should be read descriptively. They do not establish a comparative safety profile or explain the causes of deaths.

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

The registry record establishes that SWIPE was a completed phase II, single-group study with planned enrollment of 48 participants and maintenance pembrolizumab after first-line platinum doublet chemotherapy for non-progressors. It states a primary question about one-year immune-related progression-free survival and gives a target of at least 25%, with 12% as the expected historical one-year PFS benchmark. These are descriptions of study intent, not evidence that the target was achieved.

The primary endpoint result is not reported in the registry record. Consequently, the record does not show the observed irPFS rate, the number of participants meeting the endpoint, or the precision of the estimate. It also does not report the secondary outcome results. No conclusion about the study's efficacy findings can be drawn from its completion status or from its planned benchmark.

Because the study had no concurrent comparator, it cannot directly estimate a randomized treatment contrast. Even if a one-year outcome were reported, comparing it with historical trial outcomes would require caution: populations, eligibility, follow-up, and assessment methods may differ. The registry does not give enough information to evaluate those differences. The study population was also selected after chemotherapy, so its findings would concern non-progressors after 4-6 cycles rather than all people beginning first-line treatment.

The registry does report safety counts and selected serious-event terms, but these counts lack details about observation time, attribution, and causes. They are descriptive and do not show how safety compares with another treatment. The death count is not accompanied by causes or a treatment relationship in the record.

In short, the registry provides useful design and safety-summary information but does not contain the outcome reporting needed to judge the primary endpoint. It neither demonstrates that the stated target was reached nor establishes a causal treatment effect. Those questions remain unresolved by the information in the registry record.

14. Why This Trial Matters Statistically

This study illustrates how a phase II single-group design can frame an early clinical question around a prespecified historical benchmark. The registry states an intended one-year irPFS target of at least 25% and an expected 12% one-year PFS benchmark from prior maintenance trials. A design of this kind can be used to assess whether an observed proportion merits further study, but it does not provide the same protection against confounding as a concurrent randomized comparison.

The trial also highlights the importance of defining the endpoint precisely. Its primary outcome is immune-related PFS, while listed secondary outcomes include response and PFS measures under both RECIST and immune-related criteria. Those distinctions can affect classification of progression and response. The registry's primary outcome timeframe is listed as 4 years even though the endpoint title specifies one year, and the record does not explain the relationship between those descriptions.

For statistical interpretation, a target and a historical benchmark are not substitutes for an observed estimate and its uncertainty. Since the primary endpoint result and analysis details are not reported in the registry record, readers cannot evaluate the result against the target or determine its precision. The absence of analysis reporting also prevents assessment of missing data, analysis populations, or multiplicity across secondary outcomes.

Finally, the study demonstrates why completion status and safety summaries must be kept distinct from efficacy conclusions. The record identifies the study as completed and lists adverse-event counts, but neither establishes whether the primary objective was met. A registry entry can support a careful description of design and reporting gaps without supporting a treatment-effect conclusion.

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 principal limitation for interpreting this study from the registry record is that the primary outcome result is not reported. The record gives the one-year irPFS objective and historical benchmark but no observed count, rate, confidence interval, or statistical decision. Secondary outcome results and analyses are also not reported, despite the list of outcomes.

The single-group design has no concurrent comparator. Historical control values can offer context, but comparisons across separate studies may be affected by differences in participant eligibility, prior treatment, follow-up, and endpoint assessment. The registry record does not report enough information to judge whether the historical studies were comparable to this study's population or methods.

The treatment population was selected after 4-6 cycles of first-line platinum doublet chemotherapy and included non-progressors. This limits the scope of any inference to that post-chemotherapy group. The registry does not report the number of participants who received the initial chemotherapy, were assessed for eligibility, or proceeded to pembrolizumab beyond the overall enrollment figure.

The primary endpoint timeframe is listed as 4 years, while its title and description refer to one-year irPFS. The record does not clarify this difference. It also does not report the endpoint's detailed event definition, assessment schedule, censoring rules, missing-data approach, analysis population, or the planned decision threshold for the one-stage Fleming's design.

Safety interpretation is limited by the lack of detail on observation duration, event attribution, causes of death, and event-level data. The registry provides counts and selected serious-event terms but no comparative safety analysis. Multiplicity handling and interim analysis plans are not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Switch Maintenance Pembrolizumab in Patients With NSCLC After First Line Platinum Doublet Chemotherapy — NCT02705820.
• Registry record last updated 2025-06-17; status last verified 2025-06.
• Results first posted 2025-01-24.
• 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

SWIPE is listed as a completed phase II, single-group study of maintenance pembrolizumab for metastatic non-small cell lung cancer after 4-6 cycles of first-line platinum doublet chemotherapy in participants without progression. The registry lists enrollment of 48 and describes a one-stage phase II Fleming's design.

The primary question concerns one-year immune-related progression-free survival, with a stated target of at least 25% and an expected historical benchmark of 12%. The primary endpoint result is not reported in the registry record. Secondary outcomes include response, PFS, overall survival, and treatment-related adverse events.

The safety summary lists 13 participants affected by serious adverse events, 48 affected by other adverse events, and 42 deaths. The registry does not report the primary analysis details, outcome estimates, or information needed to interpret those safety counts comparatively.