Single Agent Versus Combination Chemotherapy to Treat High-risk Elderly With Non-small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Single Agent Versus Combination Chemotherapy to Treat High-risk Elderly With Non-small Cell Lung Cancer (NCT02590003), a randomized phase 2 study. The primary outcome is Treatment Failure-free Survival.

2. Clinical Question

The clinical question concerns whether the recorded treatment outcomes differ between the combination regimen and single-agent chemotherapy in this high-risk elderly population. The registry identifies the study as randomized and phase II, but its posted primary results are counts rather than a comparative effect estimate. The posted arm values for the primary outcome are 0.0 in each arm. Those values do not establish the relative effect of the regimens, and the record does not provide a confidence interval or a statistical test for their comparison.

Other stated outcomes include overall response rate, progression-free survival, overall survival, grade 3-5 adverse events, and symptom assessment using the FACT-L Symptom Assessment Scale. The registry lists symptom assessment repeatedly among the secondary outcomes. It does not report comparative results for these outcomes in the record.

3. Trial Design

This was a randomized, open-label, parallel-group phase II trial sponsored by Yale University. The registry record reports an enrollment of 3 and lists two treatment arms: platinum-based doublet chemotherapy and single-agent chemotherapy. It records the study as terminated. The start is recorded as 2015-11 and primary completion as 2016-09. The reason for termination and whether the recorded enrollment reflects the planned or final enrollment are not reported in the registry record.

Random allocation is intended to assign eligible participants to treatment groups by chance, which can reduce systematic differences between groups at assignment. In a trial with very few participants, however, randomization cannot ensure that the groups are similar on all prognostic characteristics. The registry does not report the randomization method, allocation concealment, stratification factors, or reasons for the arm sizes recorded in the results.

The study was unmasked: participants and investigators were not reported as blinded to treatment assignment. Open-label conduct can matter for outcomes involving clinical judgment, treatment decisions, symptom reports, and adverse-event assessment. The record does not describe procedures intended to reduce the influence of treatment knowledge on these measurements.

The arm records list 1.0 participant for the platinum-based doublet arm and 2.0 for the single-agent arm. The primary outcome is identified as treatment failure-free survival, with a timeframe of 90 days and a unit of participants. The registry posts a value of 0.0 for each arm. It does not provide a definition of treatment failure, follow-up detail, or a time-to-event estimate. As a result, the label alone does not establish which events or censoring rules were used to define the outcome.

The registry lists secondary outcomes covering response, survival, adverse events, and symptoms. Seven secondary outcomes are posted, including repeated entries for symptom assessment. The record does not report the planned statistical analysis, a sample-size justification, or a reason for termination. It also does not report an analysis population definition or a detailed participant flow. These omissions limit assessment of how the randomized design was implemented and how the posted results relate to the original study plan.

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 Treatment Failure-free Survival, measured over 90 days.

The primary outcome is titled treatment failure-free survival. The registry gives its unit as participants and its timeframe as 90 days, and describes the outcome parameter as a count of participants. The posted value is 0.0 in the platinum-based doublet arm and 0.0 in the single-agent arm.

In general, a failure-free endpoint is intended to capture whether a participant experiences a specified treatment failure during a defined period. Such endpoints can summarize clinically relevant events in a way that reflects more than tumor response alone. The exact events included, however, must be defined in the trial record before the endpoint can be interpreted in detail. The definition of treatment failure, including whether it includes progression, discontinuation, toxicity, or another event, is not reported in the registry record.

The word “survival” often indicates an outcome analyzed as time until an event, but the posted primary result is a participant count. The record does not provide event times, censoring information, a survival curve, or a hazard ratio. Therefore, the posted values should be read as the registered count results, not as a time-to-event comparison. A count of 0.0 in each arm does not by itself show that the underlying experience or timing of events was identical.

The endpoint could be useful for comparing treatment strategies if its event definition, follow-up, and analysis were specified and reported. With the information recorded, the outcome name and timeframe are available, but the endpoint's operational definition and comparative precision are not.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Randomization assigns participants to treatment groups by chance. In general, this supports a fair comparison by reducing the likelihood that treatment assignment is determined by prognosis or investigator choice. It does not guarantee balance in every characteristic, especially when groups are small. The registry identifies this study as randomized but does not report the sequence-generation method, allocation concealment, or any stratification procedure. Those details are needed to assess how the random allocation was carried out.

The parallel-group structure means that participants were assigned to one of the study regimens rather than being described as receiving both in sequence. The record identifies the groups as platinum-based doublet chemotherapy and single-agent chemotherapy. Because masking is listed as none, the study was open-label. In general, knowledge of treatment assignment may affect care decisions, reporting of symptoms, and assessment of outcomes. The registry does not describe blinded outcome assessment or other safeguards against such influence.

The primary result is recorded as a binary count: participants with an event, rather than a continuous measurement such as a laboratory value. For a binary outcome, a treatment comparison commonly uses the event count and the number assessed in each group to estimate an event proportion or a difference or ratio between groups. These quantities require clear denominators and a defined analysis population. Although arm sizes and outcome values are posted, the registry does not report a comparative analysis or confidence interval. No new comparison should be inferred from the arm counts alone.

A confidence interval is a statistical range expressing uncertainty around an estimated effect under the assumptions of the analysis. It helps distinguish a relatively precise estimate from one compatible with a broad range of possible effects. The record contains no confidence interval for the primary outcome. A p-value, if reported, would address compatibility of the data with a specified null hypothesis under an analysis model; it would not measure clinical importance or prove that treatments are equivalent. No p-value or null hypothesis test for the primary comparison is reported.

The endpoint name includes “survival,” but the result is posted as a participant count. Time-to-event analyses generally use event times and account for participants whose event has not been observed by the end of follow-up, a process called censoring. A time-to-event result may be summarized with a survival curve or a hazard ratio, depending on the analysis. The registry does not report event times, censoring rules, a survival curve, or a hazard ratio for treatment failure-free survival. It also does not define treatment failure. Consequently, those time-to-event methods cannot be attributed to this trial based on the record.

For small samples, standard statistical approximations can be unreliable, and estimates can be highly sensitive to individual observations. Exact or other small-sample methods may be considered in some settings, but the registry does not report which method, if any, was planned or used. It would be inappropriate to select a method after seeing the posted counts and present its result as the trial's own analysis. The registry lists no analyses for the trial and indicates that values are missing or not estimable for a two-arm comparison of the binary endpoint.

The registry includes planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It describes the effect to detect as a stated fraction of the observed effect and distinguishes participant-per-arm planning for non-time-to-event outcomes from total-event planning for time-to-event outcomes. These are planning assumptions, not the trial's reported design justification, and not observed power. The registry does not report that these assumptions formed the original protocol's sample-size plan. The reason for the trial's enrollment and termination is not reported in the registry record.

Over-dispersion is a general issue when observed variation exceeds what a chosen statistical model expects. It can make standard errors too small and confidence intervals too narrow if the model does not account for the extra variability. The posted primary outcome is a count with very limited arm sizes, but the registry does not report a model or dispersion estimate. There is no basis for asserting that over-dispersion was assessed or that a particular correction was used.

Finally, prespecified analysis details matter because endpoint definitions, censoring, analysis populations, and handling of missing outcomes affect the interpretation of results. The registry does not report a statistical analysis plan, a protocol-based primary analysis, or methods for missing data. Without those details and without comparative estimates, the posted outcome counts remain descriptive registry results rather than a complete inferential analysis.

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("Platinum-based Doublet Chemotherapy", "Single Agent Chemotherapy")
n      <- c(1, 2)
events <- c(0, 0)
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  Platinum-based Doublet Chemotherapy  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2
ARM  Single Agent Chemotherapy  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists secondary outcomes for overall response rate, progression-free survival, overall survival, grade 3-5 adverse events, and symptom assessment using the FACT-L Symptom Assessment Scale. Symptom assessment appears repeatedly in the posted list. The record reports seven secondary outcomes and eight outcome measures, but does not explain the repeated entries or identify which outcomes were formally prespecified for testing.

Multiplicity arises when several outcomes or comparisons are tested. In general, more testing creates more opportunities for a nominally notable result to occur by chance. Researchers may address this through a testing hierarchy, adjustment procedures, or by treating secondary findings as exploratory. The registry does not report a multiplicity strategy, a testing hierarchy, or adjustment for multiple comparisons. Since it reports no analyses, there are no posted secondary comparison results to interpret as confirmatory evidence.

Missing-data handling is also not reported in the registry record. The record does not describe whether outcome information was missing, how missingness was defined, or whether imputation, censoring, or another approach was used. It does not identify an intention-to-treat, per-protocol, safety, or other analysis population. These details matter because exclusions or incomplete follow-up can change the groups represented in an analysis.

The posted arm counts and outcome values do not establish which participants were included in each outcome analysis or whether all randomized participants contributed data. Participant flow, reasons for missing measurements, and the relationship between the recorded enrollment and analysis sets are not reported. These omissions prevent assessment of whether the reported counts reflect the intended analysis population or whether missingness could have influenced the results.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hyponatremia (50.0%), pleural effusion (50.0%), edema limbs (50.0%), dyspnea (50.0%), non-cardiac chest pain (50.0%).

The registry records serious adverse events for the single-agent group as 2 participants at risk and 1 affected, and for the platinum-based doublet group as 1 at risk and 0 affected. Other adverse events are recorded as 2 at risk and 2 affected in the single-agent group, and 1 at risk and 0 affected in the doublet group. The deaths field records 2 for the single-agent group and 1 for the doublet group. These are counts as entered in the registry record; they are not adjusted comparisons.

The serious-event terms listed include Hyponatremia, Pleural effusion, Edema limbs, Dyspnea, Non-cardiac chest pain, and Thromboembolic event, each shown as 50.0%. The registry does not clarify in this record whether these percentages refer to participants, events, or another denominator. They should therefore be read as reported term percentages, not as directly comparable group-specific risks.

Safety counts do not establish that treatment caused an event. The record does not report event timing, duration of treatment exposure, seriousness criteria, attribution, or whether the deaths were treatment-related. The small recorded groups also make descriptive counts highly sensitive to individual cases. These entries support a description of what was recorded, but not a reliable comparative estimate of safety between regimens.

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

The posted results establish that the registry contains a primary outcome value of 0.0 for each treatment arm, with the outcome named treatment failure-free survival and a timeframe of 90 days. They also establish the recorded arm sizes and the adverse-event counts. Those details are useful for understanding the limited results available for this randomized phase II trial.

They do not establish that the two regimens have the same effect. A matching count is not an equivalence analysis: equivalence requires a prespecified margin and an analysis that estimates uncertainty relative to that margin. The registry record does not report such a margin, a confidence interval, or a comparative test. Similarly, the posted values do not establish that either regimen is better or worse.

The endpoint definition is not reported in enough detail to determine which events qualified as treatment failure. The record also does not report event times, censoring, follow-up completeness, or the analysis population. Accordingly, the endpoint title should not be treated as a complete description of how the result was measured. The counts cannot be converted into a time-to-event conclusion from the information in the registry.

The safety entries are likewise descriptive. They do not show whether adverse events were caused by treatment, whether reported deaths were related to treatment, or how event experience compares after accounting for duration of exposure and follow-up. The serious-event term percentages have no clear denominator in the record, which further limits interpretation.

In general, a randomized trial can provide evidence about treatment effects when the allocation, outcome definitions, follow-up, analysis, and uncertainty are sufficiently reported. Here, the registry record provides a trial design and a small set of arm-level counts, but no comparative analysis and limited methodological detail. The responsible conclusion is that the posted information is insufficient to determine the relative efficacy or safety of the regimens. It is not evidence that a treatment works, fails, or is equivalent to its comparator.

14. Why This Trial Matters Statistically

This trial raises statistical issues that are common in studies of treatment strategies for clinically vulnerable populations. The registry describes an elderly, high-risk population and compares a combination regimen with a single agent. Such a comparison makes both efficacy and tolerability relevant, while the design must distinguish treatment effects from chance variation and differences in follow-up.

The recorded enrollment is 3, with small arm records. This makes precision central to interpretation: counts can change substantially in meaning when there are few observations, and a single participant's experience can have considerable influence on an estimate. Reporting an effect estimate with an interval would help readers understand uncertainty, but the registry does not report a comparative estimate or interval for the primary outcome.

The trial also illustrates why endpoint labels need operational definitions. “Treatment failure-free survival” suggests a time-to-event concept, while the posted primary result is a participant count over a stated timeframe. Without the failure definition and analysis details, readers cannot know exactly what the count represents or how it connects to time-to-event methods.

Finally, safety counts and efficacy outcomes are informative only when their denominators, follow-up, and analysis populations are clear. The registry's limited reporting does not resolve these issues. Statistically, the record is a reminder that randomization alone is not enough to support a treatment comparison: transparent outcome definitions, analysis methods, uncertainty, and participant flow are also needed.

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 registry record reports the study as terminated but does not give the reason. It does not describe whether termination affected recruitment, follow-up duration, the ability to assess outcomes, or the analyses. Without that context, the status cannot be used to infer why the trial ended or what conclusions the investigators intended to draw.

Enrollment is recorded as 3, and the arm records are very small. The registry does not report the intended enrollment target, a sample-size justification, or whether the posted enrollment represents the final number of participants randomized. It also does not report the allocation process or a participant-flow account. These gaps limit assessment of the conduct and completeness of the trial.

The primary outcome lacks an operational definition, and its reported participant-count format does not include event times or censoring information. The record does not provide a comparative estimate, confidence interval, or test. As a result, neither the meaning of treatment failure-free survival nor the precision of a between-arm comparison can be evaluated from the posted results.

Safety information is limited to counts and listed event terms. The registry does not report event attribution, timing, exposure duration, follow-up completeness, or the denominator behind the listed term percentages. The record also does not explain the relationship between deaths and adverse events.

Other limitations include unreported missing-data procedures, analysis populations, multiplicity handling, and statistical methods. The record lists secondary outcomes but provides no analyses. These limitations are specific to what is documented for this trial and do not establish that a particular procedure was or was not performed outside the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Single Agent Versus Combination Chemotherapy to Treat High-risk Elderly With Non-small Cell Lung Cancer — NCT02590003.
• Registry record last updated 2018-02-28; status last verified 2018-01.
• Results first posted 2018-02-28.
• 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

NCT02590003 is a randomized, open-label, parallel-group phase II trial sponsored by Yale University. It compared platinum-based doublet chemotherapy with single-agent nab-paclitaxel in elderly patients with advanced non-small cell lung cancer who were described as at high risk of chemotherapy toxicity. The registry records a start of 2015-11, primary completion of 2016-09, enrollment of 3, and status of terminated.

The primary outcome is treatment failure-free survival, recorded as a participant count over 90 days. The posted arm values are 0.0 for the doublet and 0.0 for single-agent chemotherapy. The record lists safety counts and secondary outcomes, but reports no trial analyses or comparative estimate for the primary outcome. The treatment-failure definition, analysis methods, missing-data approach, analysis populations, sample-size justification, and reason for termination are not reported in the registry record.