This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Re-challenge Pembrolizumab Study as a Second or Further Line in Patients With Advanced NSCLC (NCT03526887), a non-randomized phase 2 study. The primary outcome is Efficacy of Pembrolizumab Re-challenge Measured by Overall Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03526887 |
| Phase | Phase 2 |
| Condition | Lung Cancer |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 73 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Spanish Lung Cancer Group |
| Results posted | Yes |
2. Clinical Question
Population
Patients with advanced non-small cell lung cancer who had previously received an anti-PD-L1 drug. The record describes two groups according to the circumstances and timing of progression after earlier treatment.
Intervention
Pembrolizumab re-challenge. The study summary describes intravenous pembrolizumab at 200 mg on day 1 of each three-week cycle, administered until disease progression.
Comparator
There is no randomized control group. The study is non-randomized and has parallel cohorts defined by prior treatment and progression history; those cohorts are not like-for-like control groups.
Primary question
The stated primary outcome concerns overall survival after pembrolizumab re-challenge. However, the record gives a timeframe that includes progression or death, and its endpoint description does not consistently define the same outcome. The primary result is not reported in the registry record.
The study asks whether pembrolizumab re-challenge can be evaluated in patients whose disease progressed after earlier anti-PD-L1 treatment. It is an exploratory phase II study rather than a randomized comparison of re-challenge against another treatment. Its two cohorts reflect different prior-treatment histories, not treatment assignments intended to estimate a causal difference between groups.
Cohort 1 is described as including patients with progression while on treatment or less than 12 weeks after stopping treatment, followed by chemotherapy for at least 4 cycles and another progression before study entry. Cohort 2 is described as including patients whose progression occurred more than 12 weeks after stopping treatment, followed by re-entry after the later progression. These definitions frame the clinical question, but the registry does not report whether the cohorts had comparable prognostic characteristics.
Although the record names overall survival as the primary outcome, it does not provide an observed median or another primary result. The question can therefore be described from the study design, but its answer cannot be determined from the posted results.
3. Trial Design
This is a phase II, open-label, multicenter, exploratory study sponsored by the Spanish Lung Cancer Group. The registry lists the study as completed and reports enrollment of 73. It describes a non-randomized, parallel design with no masking. The study began on 2018-07-15, and primary completion is recorded as 2022-12-30.
Participants were assigned to cohorts by the history and timing of progression after prior treatment, rather than randomized to pembrolizumab or a control treatment. Cohort 1 involves progression during treatment or less than 12 weeks after stopping, followed by chemotherapy for at least 4 cycles and further progression. Cohort 2 involves progression more than 12 weeks after stopping treatment. Both descriptions then place patients in the study for pembrolizumab re-treatment. The registry's cohort descriptions do not provide a randomized treatment contrast.
The intervention summary describes pembrolizumab 200 mg by intravenous infusion on day 1 of each three-week cycle, until disease progression. The intervention listing names pembrolizumab for both entries; further distinctions between those entries are not reported in the registry record.
Non-randomized cohort studies can describe outcomes in the populations enrolled, but differences between cohorts may reflect their prior treatment histories or other characteristics as well as any treatment-related factors. Without random allocation and a comparable control group, a direct causal comparison is not supported by this design. The record does not report the protocol's sample-size justification, eligibility criteria in full, or a detailed statistical analysis plan.
The primary outcome is labeled overall survival, while its timeframe refers to progression or death, and its description gives a separate definition involving time alive from diagnosis or treatment start. It also refers to randomization despite the non-randomized design. These differences make the endpoint wording internally inconsistent; the record does not clarify which definition governed the analysis.
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
| Stage | Cohort 1 | Cohort 2 |
|---|---|---|
| STARTED | 55 | 18 |
| COMPLETED | 55 | 18 |
| NOT COMPLETED | 0 | 0 |
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.
| Characteristic | Unit | Cohort 1 | Cohort 2 | Total |
|---|---|---|---|---|
| Participants analysed | 55 | 18 | 73 | |
| Age, Continuous | years | 63.5 (41 to 80) | 69.8 (55 to 83) | 65 (41 to 83) |
| Sex: Female, Male: Female | Participants | 39 | 15 | 54 |
| Sex: Female, Male: Male | Participants | 16 | 3 | 19 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): White | Participants | 53 | 18 | 71 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 2 | 0 | 2 |
| Region of Enrollment: Spain | participants | 55 | 18 | 73 |
| Performance Status: ECOG 0 | Participants | 15 | 8 | 23 |
| Performance Status: ECOG 1 | Participants | 39 | 10 | 49 |
| Performance Status: ECOG 2 | Participants | 0 | 0 | 0 |
| Performance Status: ECOG 3 | Participants | 0 | 0 | 0 |
| Performance Status: ECOG4 | Participants | 0 | 0 | 0 |
| Performance Status: Not recorded | Participants | 1 | 0 | 1 |
| Cigarette Smoking History: Never smoker | Participants | 7 | 1 | 8 |
| Cigarette Smoking History: Former smoker | Participants | 39 | 15 | 54 |
| Cigarette Smoking History: Smoker | Participants | 8 | 2 | 10 |
| Cigarette Smoking History: Not recorded | Participants | 1 | 0 | 1 |
| Body Mass Index | kg/m^2 | 26 (17.1 to 35.3) | 28 (20.2 to 34.8) | 26.5 (17.1 to 35.3) |
| Histology: Adenocarcinoma | Participants | 32 | 8 | 40 |
| Histology: Squamous | Participants | 19 | 8 | 27 |
| Histology: Adeno-squamous | Participants | 2 | 0 | 2 |
| Histology: Large Cell Carcinoma | Participants | 1 | 1 | 2 |
| Histology: NOS/Undifferentiated | Participants | 1 | 0 | 1 |
| Histology: Other | Participants | 0 | 1 | 1 |
| Clinical Stage: Stage I | Participants | 0 | 0 | 0 |
| Clinical Stage: Stage II | Participants | 0 | 0 | 0 |
| Clinical Stage: Stage III | Participants | 3 | 1 | 4 |
| Clinical Stage: Stage IV | Participants | 52 | 17 | 69 |
6. Primary Endpoint
The registry lists the primary outcome as Efficacy of Pembrolizumab Re-challenge Measured by Overall Survival, measured over From date of randomization until the date of first documented progression or date of death from any cause, whichever came first, assessed up to 48 months..
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Full Range |
| Unit | months |
| Endpoint classification | Time-to-Event Analysis |
The named primary endpoint is overall survival, expressed in months as a median with a full range. Overall survival is generally used to describe time until death from any cause. It is clinically meaningful as a time-to-event measure, but its interpretation depends on a clear definition of the time origin, the event, and how follow-up and censoring are handled.
The registry wording does not provide a consistent operational definition. Its timeframe starts at “date of randomization” and ends at the first documented progression or death, whichever comes first, with assessment up to 48 months. That event definition includes progression, so it does not match the usual meaning of overall survival. The endpoint description instead defines overall survival as the length of time from diagnosis or treatment start that patients remain alive. These descriptions do not align, and the timeframe's reference to randomization is also at odds with the non-randomized design.
As a result, readers cannot confidently determine from the registry wording which time origin and event were used for the primary analysis. The median and full range are not reported in the registry record, nor are the number of events, censoring details, or a confidence interval for the endpoint. The endpoint's general statistical purpose can be explained, but the posted information does not permit an interpretation of the observed primary outcome.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record describes a completed exploratory phase II study of pembrolizumab re-challenge, but it does not report the primary efficacy result. In particular, no median overall survival, hazard ratio, confidence interval, or primary-outcome p-value is reported in the registry record. The record therefore does not establish an estimated magnitude of benefit or the precision of such an estimate.
There is also uncertainty about what the primary endpoint represents. The outcome title says overall survival, while the timeframe defines an event as progression or death, whichever comes first. The description of overall survival refers to time alive from diagnosis or treatment start. Since these are not equivalent endpoint definitions, the event and time origin cannot be resolved from the record. A time-to-progression-or-death outcome and an overall-survival outcome answer different questions.
The design limits comparisons between the cohorts. Participants were not randomized, and cohort membership reflects different prior-treatment and progression histories. Differences in outcomes, if reported elsewhere, could be associated with those histories or with other characteristics, rather than the cohort label itself. The cohorts should not be read as treatment and control arms, and the study does not provide a like-for-like randomized comparison.
The registry reports adverse-event counts for each cohort, including serious adverse events, other adverse events, and deaths. These figures are descriptive counts. The record does not establish that the deaths were caused by treatment, and the adverse-event summary does not by itself provide a comparative estimate of risk. The listed serious-event terms have percentages, but their specific denominators and attribution are not described in the record.
Overall, the posted information supports a description of the study's design, cohort definitions, intervention, and selected safety counts. It does not provide the primary efficacy estimate or enough endpoint detail to resolve what was analyzed. The study's exploratory, non-randomized structure further means that the cohort outcomes, even if available elsewhere, would not constitute a randomized test of pembrolizumab against a control treatment.
Planning illustrations based on an observed effect are not trial results or observed power. The record does not state that those illustrative assumptions were the study's actual design assumptions. Any estimate generated under those assumptions would describe a hypothetical design exercise, not the evidence collected in this trial.
9. Statistical Methods Explained
The statistical interpretation of this study begins with the distinction between design and analysis. The registry describes a non-randomized parallel study with two cohorts. Cohort membership is determined by prior treatment and timing of progression, not by random allocation. Consequently, a comparison between cohort outcomes is descriptive and potentially confounded by the clinical histories that define the cohorts. It is not equivalent to a randomized treatment comparison.
What does a median time-to-event result describe?
For a time-to-event endpoint, the median is the time by which half of the analyzed population is estimated to have experienced the event, under the analysis method used. The registry names a median in months and a full range for the primary outcome, but it does not report the median result. The range, even if reported, would describe the observed spread and would not on its own show the precision of the estimate.
Time-to-event analyses often include participants who have not experienced the event by the end of their available follow-up. Such observations are generally treated as censored rather than as events at an assumed later time. Interpretation depends on the event definition, time origin, follow-up process, and censoring approach. These analysis details are not reported in the registry record. The endpoint wording itself is inconsistent: the title identifies overall survival, while the timeframe uses progression or death and refers to randomization. The record's non-randomized design makes that time-origin reference unclear.
What would a hazard ratio mean?
A hazard ratio is a relative comparison of event rates over follow-up under a statistical model. It is not the ratio of median times, and it does not directly state how many months a participant gains or loses. A hazard ratio also needs a clear event definition and a suitable comparison group. Since the registry does not report a hazard ratio for this study, no relative event-rate estimate can be interpreted from the posted record. The methodological note indicates that only sponsor-reported hazard ratios are used, but no such value is included in the record.
A confidence interval communicates uncertainty around an estimate: a narrower interval indicates greater statistical precision than a wider interval, while its interpretation depends on the model and assumptions. Confidence intervals for the primary outcome are not reported in the registry record. Without an estimate and its uncertainty, it is not possible to assess how precisely the primary endpoint was measured or whether the data are compatible with a broad range of effects.
Why does the design matter for inference?
Randomization is used to make assigned treatment groups comparable, on average, with respect to measured and unmeasured characteristics. This study did not randomize participants. Its cohorts encode different prior treatment and progression histories, which may also relate to prognosis. Statistical adjustment can sometimes address measured differences, but the registry does not report an adjusted analysis or the covariates used. It is therefore not possible to determine whether any cohort comparison accounted for such differences.
Open-label status means masking is listed as none. Knowledge of treatment can influence reporting or assessment in some settings, particularly for outcomes that involve judgment. The record does not describe blinded outcome assessment. This does not establish that assessments were biased, but it is relevant context when interpreting an exploratory study.
How should planning illustrations be read?
The planning illustration is described as a calculation based on the observed effect, not as the trial's own design. Its stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For time-to-event calculations, it gives a total number of events using the Schoenfeld approach; for other outcomes, it gives participants per arm. These are hypothetical planning quantities, not observed power and not evidence that the trial was designed with those assumptions. Specific planning totals are not reported in the registry record, and the protocol's sample-size justification is not reported in the registry record.
Over-dispersion is a general statistical issue in count outcomes: it occurs when variability exceeds what a simple count model assumes. A model that ignores substantial over-dispersion may understate uncertainty. The registry entries do not establish that an over-dispersed outcome model was used in this study, and no such analysis is reported in the registry record. More generally, the appropriate method depends on the endpoint and data structure, and those details cannot be inferred from the study title or outcome count alone.
The registry lists three outcome measures and one analysis. That count does not describe the complete analysis plan, define how endpoints were prioritized beyond the named primary outcome, or explain the statistical models. The methods used for the reported safety counts are also not described in the registry record. Interpretations should therefore remain limited to the endpoints and descriptive values actually stated.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists three outcome measures, including the primary overall-survival outcome and secondary outcomes of progression-free survival per RECIST v1.1 and best global response. Testing or estimating multiple outcomes can create multiplicity concerns, particularly when several results are interpreted as confirmatory. The registry record does not report a multiplicity adjustment, a testing hierarchy, or how secondary outcomes were intended to support conclusions. It also does not give the results for those secondary outcomes.
The record lists one analysis, but that count does not establish whether the analysis was prespecified, whether it was the only analysis conducted, or which populations and methods were used. A detailed statistical analysis plan is not reported in the registry record. Interim analyses and their handling are not reported in the registry record.
Missing-data methods and the extent of missing outcome data are not reported in the registry record. For time-to-event outcomes, incomplete follow-up and censoring can affect estimates; for response outcomes, missing assessments can affect the analyzed population and response proportions. The record does not describe how such observations were handled, whether participants were excluded from particular analyses, or whether sensitivity analyses were performed.
The analyzed populations are not defined in the registry record. It is therefore not possible to determine whether the primary analysis included all enrolled participants, only those who received treatment, or another defined group. These omissions limit assessment of potential bias and reproducibility. They do not imply that the study did or did not use particular procedures; they mean the procedures are not reported in the registry record.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Cohort 1 | 2 | 55 | 3.6% |
| Cohort 2 | 1 | 18 | 5.6% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Cohort 1 | 44 of 55 |
| Cohort 2 | 5 of 18 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Cohort 1 | 29 of 55 |
| Cohort 2 | 11 of 18 |
Most frequent serious adverse event terms
| Event term | Organ system | Cohort 1 | Cohort 2 |
|---|---|---|---|
| Gastrointestinal disorder | Gastrointestinal disorders | 0 of 55 | 1 of 18 |
| Alkaline phosphatase increased | Blood and lymphatic system disorders | 1 of 55 | 0 of 18 |
| Blood bilirubin increased | Blood and lymphatic system disorders | 1 of 55 | 0 of 18 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Cohort 1 | Cohort 2 |
|---|---|---|---|
| Fatigue | General disorders | 7 of 55 | 4 of 18 |
| Diarrhea | Gastrointestinal disorders | 8 of 55 | 1 of 18 |
| Pruritus | Skin and subcutaneous tissue disorders | 5 of 55 | 3 of 18 |
| Nausea | Gastrointestinal disorders | 4 of 55 | 1 of 18 |
| Aspartate aminotransferase increased | Investigations | 3 of 55 | 2 of 18 |
| Alanine aminotransferase increased | Blood and lymphatic system disorders | 3 of 55 | 1 of 18 |
| Blood bilirubin increased | Investigations | 3 of 55 | 1 of 18 |
| Arthralgia | Musculoskeletal and connective tissue disorders | 3 of 55 | 1 of 18 |
| Rash | Skin and subcutaneous tissue disorders | 3 of 55 | 1 of 18 |
| Stomach pain | Gastrointestinal disorders | 3 of 55 | 0 of 18 |
Serious events with the highest arm-level rate: gastrointestinal disorder (5.6%), alkaline phosphatase increased (1.8%), blood bilirubin increased (1.8%).
The registry reports serious adverse-event figures separately for Cohort 1 and Cohort 2. Cohort 1 has 55 listed as at risk for serious adverse events and 2 affected; Cohort 2 has 18 listed as at risk and 1 affected. For other adverse events, the corresponding at-risk and affected counts are 55 and 29 for Cohort 1, and 18 and 11 for Cohort 2. The record also lists deaths of 44 in Cohort 1 and 5 in Cohort 2.
These are descriptive registry counts, not proof that pembrolizumab caused the events or deaths. The record does not report attribution, event timing, severity details beyond the serious-event grouping, or a comparative safety analysis. It does not clarify the relationship between the deaths listed in the adverse-event data and the overall-survival endpoint.
The serious-event terms listed include gastrointestinal disorder at 5.6%, alkaline phosphatase increased at 1.8%, and blood bilirubin increased at 1.8%. Their denominators and attribution are not reported in the registry record, so these percentages should not be interpreted as cohort-specific risks or as estimates of treatment-caused events. Differences in cohort size and clinical history also make unadjusted comparisons difficult to interpret.
12. What the Result Does, and Does Not, Mean
The result that can be taken from the registry is primarily a description of the study: a completed, exploratory phase II evaluation of pembrolizumab re-challenge in advanced non-small cell lung cancer after prior anti-PD-L1 treatment. The record specifies two cohorts based on treatment and progression history and reports selected adverse-event counts. It does not report the primary efficacy estimate.
That absence matters. Without a reported median, hazard ratio, confidence interval, or primary-outcome p-value, the magnitude and precision of the primary outcome cannot be assessed. The outcome wording also leaves uncertainty about whether the intended event was death alone or progression or death. The registry does not resolve this discrepancy.
The non-randomized design does not support treating the cohorts as equivalent treatment and control groups. Their differing progression histories are part of how they were defined and may be related to outcomes. Even if cohort-level outcomes were available, a difference would not by itself demonstrate that the cohort assignment or re-challenge caused the difference.
The safety counts describe recorded events, not causal effects. In particular, the listed deaths should not be attributed to treatment on the basis of these figures. The record does not provide enough information about attribution, follow-up, or analysis to make a comparative estimate of safety.
These limitations do not negate the study's descriptive value. The record identifies the clinical setting and an exploratory re-challenge approach that can be examined in subsequent research. But neither the study title nor its completed status supplies missing outcome results. The posted record cannot establish whether re-challenge improves outcomes, whether it is ineffective, or how its effects compare with another treatment.
13. Why This Trial Matters Statistically
This study raises statistical questions that commonly arise in exploratory oncology research: how to interpret outcomes after prior treatment, how to distinguish descriptive cohort patterns from causal comparisons, and how endpoint definitions shape the meaning of an estimate. Its cohorts reflect prior progression histories, so the central design feature is also a source of potential differences between groups.
The primary endpoint wording illustrates why a clear event definition matters. Overall survival conventionally concerns time to death, while the timeframe in the registry includes progression as an event. Those endpoints have different interpretations. The record also refers to randomization in a study listed as non-randomized. Resolving such wording is necessary before a reported median or model estimate can be understood.
The study's open-label, non-randomized, exploratory design means its role is not the same as a randomized comparative trial. It can describe outcomes in the selected cohorts, but the design does not ensure that differences between them arise from pembrolizumab re-challenge. The absence of the primary outcome result in the registry record further limits what can be inferred from this entry.
Finally, safety counts need careful interpretation. Affected counts and listed serious-event terms are useful descriptive information, but without attribution, clear denominators for listed percentages, and comparative methods, they do not quantify a causal treatment effect. Statistical reporting is most informative when the endpoint, analysis population, missing-data handling, and uncertainty are clearly described alongside observed results.
14. 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 registry entry has several limitations specific to interpretation of this trial. First, it names overall survival as the primary outcome but gives a timeframe that treats progression or death as the event. Its separate endpoint description refers to time alive from diagnosis or treatment start. These definitions are not consistent, and the record does not explain which one was used in analysis.
Second, the timeframe refers to the date of randomization even though the study is listed as non-randomized. This makes the stated time origin unclear. The registry does not report a primary outcome estimate, uncertainty interval, event count, or detailed censoring approach, so the primary result cannot be reconstructed or evaluated from the entry.
Third, the cohorts are based on different prior-treatment and progression histories. They are not like-for-like control groups, and the record does not report adjustment for prognostic differences or provide baseline characteristics for assessing comparability. No randomized control group is described.
The registry does not report the sample-size justification, full statistical analysis plan, analysis populations, missing-data methods, or multiplicity procedures. It lists secondary outcomes, but their results are not reported in the registry record. The adverse-event figures provide counts and selected percentages, but attribution and the denominators for the listed serious-event percentages are not reported in the registry record. These limitations restrict conclusions to the descriptive information recorded and prevent a complete assessment of efficacy or comparative safety.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Re-challenge Pembrolizumab Study as a Second or Further Line in Patients With Advanced NSCLC — NCT03526887.
- Registry record last updated 2025-03-24; status last verified 2025-03.
- Results first posted 2025-03-24.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry record describes NCT03526887, a completed, open-label, multicenter, exploratory phase II study sponsored by the Spanish Lung Cancer Group. It lists enrollment of 73 and a non-randomized parallel design with two cohorts defined by prior treatment and progression history. The intervention summary describes intravenous pembrolizumab 200 mg on day 1 of each three-week cycle until disease progression.
The primary outcome is titled overall survival, but the timeframe and endpoint description are inconsistent. No primary efficacy result is reported in the registry record. Secondary outcomes are progression-free survival per RECIST v1.1 and best global response; their results are also not reported in the registry record. Selected adverse-event counts and serious-event terms are listed, without enough detail for causal or comparative conclusions.