Pembrolizumab Plus Chemotherapy in NSCLC With Targetable Genetic Alterations After Progression on Targeted Agents: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pembrolizumab Plus Chemotherapy in NSCLC With Targetable Genetic Alterations After Progression on Targeted Agents (NCT03242915), a na phase 2 study. The primary outcome is The Percentage of Patients That Respond to Treatment.

2. Clinical Question

The clinical question concerns response to a combination regimen after disease progression on targeted treatment. It is not a question about whether the combination is better than chemotherapy alone, pembrolizumab alone, or another treatment, because no comparator arm is described. The registry record also lists progression-free survival and overall survival as secondary outcomes. Their results are not reported in the registry record.

The trial enrolled 33 participants. The safety listings distinguish EGFR+ NSCLC and ALK+ NSCLC groups, but those listings do not provide response results. Without outcome analyses, the registry record cannot establish the observed response rate or its precision, nor can it support a comparative estimate of treatment effect. The study’s stated hypothesis is that adding pembrolizumab will enhance the efficacy of carboplatin and pemetrexed; a hypothesis is not itself evidence that the proposed effect occurred.

3. Trial Design

This was a phase II, single-group study sponsored by the University of Michigan Rogel Cancer Center. The registry record lists non-small cell lung cancer as the condition, an enrollment of 33, allocation as NA, and masking as none. It describes one treatment group receiving pembrolizumab, carboplatin, and pemetrexed. There is no randomized control arm in the design described by the registry record.

The combination was specified as pembrolizumab 200 mg with carboplatin at AUC 5 and pemetrexed at 500 mg/m2, administered intravenously every 3 weeks. The intended population had targetable genetic alterations and progressive disease following appropriate targeted agents. The brief summary specifically refers to EGFR-mutation-positive NSCLC or other genetic alterations. The safety listings identify EGFR+ NSCLC and ALK+ NSCLC groups; further details about enrollment criteria or subgroup definitions are not reported in the registry record.

The study is listed as phase II and completed. Its recorded start date is 2017-10-03, and its primary completion date is 2025-02-20. The primary outcome is the percentage of patients who respond to treatment, defined as response rate comprising complete and partial response. Progression-free survival time and overall survival time are listed as secondary outcomes.

The registry record does not report outcome analyses. It therefore does not provide an observed response rate, a confidence interval for that rate, or results for the secondary outcomes. It also does not report the protocol’s sample-size justification, a statistical analysis plan, interim analyses, or details of analysis populations. These omissions limit what can be concluded from the record about how the trial was planned and how its outcomes were evaluated.

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.

6. Primary Endpoint

The registry lists the primary outcome as The Percentage of Patients That Respond to Treatment, measured over 5.5 years.

The primary endpoint is response rate, expressed as the percentage of patients with a complete or partial response. The registry definition describes complete response as disappearance of all non-target lesions, with all lymph nodes non-pathological in size (<10mm short axis). Partial response is described as persistence of one or more non-target lesions without qualifying as progressive disease. The definition also describes progressive disease as unequivocal progression of existing non-target lesions and notes that the appearance of one or more new lesions is also considered progression.

A response-rate endpoint summarizes the proportion of participants who meet the stated response criteria during the specified assessment period. It is commonly used in studies of this kind because it gives a direct measure of observed tumor response in the enrolled group. In a single-group study, however, a response rate is descriptive of that group; it does not by itself estimate how outcomes would differ under another treatment. Comparing it with results from a different population or study would require careful consideration of differences in participants, assessment, and follow-up.

The registry lists a timeframe of 5.5 years for the primary endpoint, but does not report the number of responders, the response percentage, or its 95% confidence interval. Consequently, the endpoint definition is available while its observed result and statistical precision are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study is described as phase II and single-group. A single-group design observes outcomes among participants receiving the specified regimen without a concurrent randomized control group. The design description is important because it determines the kinds of statistical claims that can be supported. An observed outcome in one group can describe that group; it does not, on its own, provide a like-for-like comparison with an alternative treatment.

What does a response percentage represent?

A response percentage is a proportion: the number of participants meeting the response definition relative to the relevant group assessed. Here, the primary outcome definition combines complete and partial response. The record specifies a 95% confidence interval as the dispersion measure, but does not report an observed percentage or interval. Without those results, neither the estimated proportion nor its precision can be assessed.

In general, a confidence interval expresses uncertainty around an estimate under the assumptions of the statistical method used. It is not a guarantee that the true value lies within a particular interval, and it does not remove limitations of the study design or outcome assessment. A narrow interval would generally indicate greater statistical precision than a wide interval, but precision is distinct from whether an effect is clinically important or attributable to a treatment. No interval is reported for this primary outcome in the registry record.

Why does the single-group structure matter?

Without random allocation to a comparator group, there is no concurrent control outcome against which to estimate a treatment contrast. A response rate can still describe the observed response experience in the enrolled group, if reported. But that rate alone cannot show what those same participants’ outcomes would have been under another treatment. Differences between separate studies may reflect differences in populations, eligibility, follow-up, assessment, or other features rather than treatment alone.

The registry lists allocation as NA and masking as none. It does not provide additional detail about outcome assessment procedures, independent review, or the handling of assessments. These features can affect the interpretation of response outcomes, but their specifics are not reported in the registry record and should not be assumed.

What can be said about the analysis plan?

The registry describes the primary endpoint and lists progression-free survival time and overall survival time as secondary outcomes. It does not report a statistical analysis plan, an interim-analysis plan, adjustment methods, or definitions for analysis populations. It also does not report a sample-size justification. Consequently, the record does not establish which statistical procedures were planned or whether the reported endpoint was analyzed according to a prespecified approach.

The general planning assumptions associated with the sample-size illustrations describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. These assumptions describe planning illustrations, not this single-group trial’s own design. The illustration convention specifies participants per arm for non-time-to-event outcomes and total events for time-to-event outcomes. Such illustrations are not observed power and do not establish that the trial was adequately powered for its actual analyses. The registry record does not report a trial-specific sample-size calculation or its rationale.

Why is over-dispersion sometimes relevant?

In general, over-dispersion means that the variability in observed data is greater than a statistical model expects. If a model assumes less variability than the data have, standard errors and confidence intervals may be too small, making results appear more precise than warranted. The importance of over-dispersion depends on the outcome and model used. The registry record does not report a model for the primary response analysis, so it is not possible to determine whether over-dispersion was assessed or relevant to the study’s analysis.

How should absent results be handled?

The record lists 3 outcome measures and 0 analyses. The primary endpoint’s definition is therefore distinct from an analyzed result. No response count, percentage, confidence interval, or secondary-outcome estimate is reported in the registry record. Statistical interpretation should not fill in those missing quantities or infer results from enrollment, completion status, or safety data. In general, missing outcome results prevent readers from evaluating the direction, magnitude, and uncertainty of an endpoint estimate.

Likewise, the registry record does not report missing-data methods, censoring rules, or the populations included in any efficacy analysis. For a binary response outcome, missing assessments can affect the denominator and the classification of participants. For time-to-event outcomes, censoring and follow-up rules affect interpretation. These are general methodological considerations; the specific procedures used in this study are not reported in the registry record.

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("EGFR+ NSCLC", "ALK+ NSCLC")
n      <- c(26, 7)
events <- c(12, 2)
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  EGFR+ NSCLC  26  12  46.2  26.6  66.6  28.8  64.5  28.7  64.7
ARM  ALK+ NSCLC  7  2  28.6  3.7  71.0  8.2  64.1  8.5  65.1

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary response-rate outcome and two secondary outcomes: progression-free survival time and overall survival time. It does not report a multiplicity strategy for interpreting several outcomes. In general, testing or estimating multiple outcomes can increase the chance of at least one apparently notable finding when results are interpreted as though only one outcome had been examined. Prespecified prioritization or adjustment can help clarify how multiple results should be read, but the applicable plan is not reported in the registry record.

The registry record also does not report missing-data methods or analysis populations. It is not stated how participants with unevaluable response assessments, incomplete follow-up, or missing outcome information were handled. For a response percentage, the treatment of participants without an evaluable assessment can influence both the numerator and denominator. For time-to-event outcomes, censoring and follow-up rules are important to the interpretation of estimates. These details are not reported in the registry record, so no particular approach should be presumed.

No outcome analyses are listed, and no response or survival estimates are reported. As a result, readers cannot assess whether missingness or multiplicity affected any reported finding, or whether planned analyses differed from the analyses conducted. The record also does not describe interim analyses or a statistical analysis plan. These omissions concern the transparency of the analysis, not evidence that a particular method was or was not used.

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: pleural effusion (42.9%), acute kidney injury (28.6%), progressive disease (28.6%), abdominal pain (14.3%), acidosis (14.3%).

The serious adverse-event listings are presented separately for EGFR+ NSCLC and ALK+ NSCLC. In the EGFR+ group, 10 of 26 participants are listed as affected by serious adverse events, and 26 of 26 are listed as affected by other adverse events. In the ALK+ group, 7 of 7 are listed as affected by serious adverse events, and 7 of 7 by other adverse events. The record lists 24 deaths in the EGFR+ group and 7 in the ALK+ group. These counts describe what is recorded in the safety listings; they do not, by themselves, establish why events occurred or whether they were caused by treatment.

The serious-event terms listed include pleural effusion (42.9%), acute kidney injury (28.6%), progressive disease (28.6%), abdominal pain (14.3%), acidosis (14.3%), and back pain (14.3%). These percentages should be read as term frequencies in the registry’s serious-event listing, not as response results or a comparison between treatments. The record does not provide enough detail to infer the clinical course, timing, severity grading, or attribution for each term.

Because the study has no concurrent comparator group, these safety counts do not estimate how event rates would differ under another regimen. They also do not replace a full safety analysis, whose methods and detailed context are not reported in the registry record.

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

The record identifies a completed phase II study of pembrolizumab with carboplatin and pemetrexed in non-small cell lung cancer after progression on appropriate targeted agents. It specifies the treatment regimen and defines the primary endpoint as the percentage of patients with complete or partial response. It also lists progression-free survival and overall survival as secondary outcomes. These features establish what the study was designed to examine, but they do not reveal the observed efficacy results.

No response count, response percentage, or confidence interval is reported in the registry record. No analyses are listed for the outcome measures, and no survival estimates are reported. Thus, the record does not show the size of any observed response or how precisely it was estimated. It also does not permit an assessment of whether outcomes differed between treatment groups, because the study is described as single-group and no comparator is listed.

The study’s stated hypothesis—that adding pembrolizumab will enhance the efficacy of carboplatin and pemetrexed—is an aim to be tested, not a finding. The treatment regimen and the fact that the study is marked completed should not be interpreted as evidence that a response occurred or that the hypothesis was supported. The safety listings provide separate information about adverse events and deaths; they do not answer the primary efficacy question.

Even if a single-group response estimate were available, it would describe outcomes among participants in that group and would not by itself establish a comparative treatment effect. In general, interpretation also depends on uncertainty, outcome definitions, analysis populations, missing-data handling, and follow-up. Several of these details are not reported in the registry record.

The most defensible conclusion is limited: the registry documents the design, planned outcomes, regimen, and selected safety counts, but does not report the outcome analyses needed to evaluate response or survival. No conclusion about whether the combination improves outcomes or how it compares with another approach can be drawn from these registry entries.

14. Why This Trial Matters Statistically

This trial is statistically notable because its design and outcome question can be described, while the registry record does not report the corresponding efficacy analyses. That distinction is central to evidence interpretation: a prespecified endpoint tells readers what investigators intended to measure, whereas an analyzed result is needed to assess what was observed and with what precision. Here, the response-rate definition is available, but the response estimate and its confidence interval are not.

The single-group structure also clarifies the scope of any potential conclusion. A single-group response estimate could characterize the enrolled participants, but without a concurrent comparator it would not directly quantify a treatment contrast. Recognizing this design boundary helps avoid interpreting an uncontrolled outcome as evidence of comparative benefit. It also emphasizes why outcome estimates, uncertainty, and design context need to be considered together.

The registry’s safety information has a separate role. Counts of participants with serious and other adverse events, deaths, and listed serious-event terms describe recorded safety occurrences in the listed groups. They cannot serve as a substitute for the unreported response analysis, nor do they establish treatment causation or comparative safety in the absence of a control group.

The record lists secondary survival outcomes, but does not report their analyses, missing-data procedures, analysis populations, or a statistical analysis plan. These gaps matter because they affect how a reader would interpret estimates if they were available. The study therefore illustrates a broader statistical lesson: trial registration details can provide useful context, but design information and endpoint definitions are not equivalent to results. Transparent reporting of estimates, uncertainty, and analysis methods is needed to evaluate the evidence.

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 of the registry information is that it does not report outcome analyses. The primary response-rate result and the secondary progression-free survival and overall survival results are not available in the record described. The response definition and listed timeframe do not supply an observed estimate, its confidence interval, or the number of participants evaluated for response.

The single-group design has no concurrent randomized comparator. Accordingly, even a response result would be descriptive of the treated group rather than a direct randomized comparison. The record lists allocation as NA and masking as none; further information about assessment procedures is not reported. This limits assessment of potential influences on outcome measurement.

The registry does not report the statistical analysis plan, sample-size justification, interim analyses, analysis populations, missing-data methods, or detailed follow-up and censoring procedures. These omissions make it difficult to assess how outcomes were intended to be analyzed and how incomplete assessments may have been handled. The planning assumptions attached to sample-size illustrations are not a trial-specific design justification and are not evidence of observed power.

The safety listings distinguish EGFR+ NSCLC and ALK+ NSCLC, but the record does not provide further details about subgroup definitions or the context needed to interpret the listed events. The listed counts and percentages do not establish causal attribution or a comparison with another regimen. These limitations are specific to the information reported in the registry record and should not be taken as evidence that unreported procedures or analyses were not performed.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pembrolizumab Plus Chemotherapy in NSCLC With Targetable Genetic Alterations After Progression on Targeted Agents — NCT03242915.
• Registry record last updated 2025-05-30; status last verified 2025-05.
• Results first posted 2025-05-30.
• 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

NCT03242915 is a completed phase II, single-group study sponsored by the University of Michigan Rogel Cancer Center. It enrolled 33 participants with non-small cell lung cancer and evaluated pembrolizumab with carboplatin and pemetrexed after progression on appropriate targeted agents. The regimen was specified as pembrolizumab 200 mg, carboplatin at AUC 5, and pemetrexed at 500 mg/m2 intravenously every 3 weeks.

The primary endpoint was the percentage of patients with complete or partial response; progression-free survival time and overall survival time were listed as secondary outcomes. The registry record does not report outcome analyses or efficacy estimates. It lists serious and other adverse-event counts and deaths for EGFR+ NSCLC and ALK+ NSCLC groups, but those records do not provide a comparative safety analysis or establish event causation.

The study began on 2017-10-03 and has a primary completion date of 2025-02-20. The record provides design and endpoint information, but not the results needed to evaluate response or survival.