Preoperative Pemetrexed and Carboplatin for Select Stage IB, II, and III Non-Squamous Non-Small-Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Preoperative Pemetrexed and Carboplatin for Select Stage IB, II, and III Non-Squamous Non-Small-Cell Lung Cancer (NCT00906282), a na phase 2 study. The primary outcome is 3-Year Overall Survival Rate.

2. Clinical Question

The trial’s stated purpose was to examine preoperative pemetrexed and carboplatin in this selected population. Its summary describes the regimen as a subject for evaluation of safety and activity, with further investigation in a larger, prospective, randomized trial as a possible next research step if the regimen showed promise. That statement explains the study rationale; it is not a reported finding that the regimen was safe or active.

3. Trial Design

This is a completed, multicenter Phase II trial with a single-group design. Enrollment was 46. The record lists no allocation and no masking, and it contains one treatment group: Pemetrexed/Carboplatin. With no control group, results from the treated participants cannot be compared within this study against outcomes under another treatment or usual care. The design can describe outcomes among enrolled participants, but it does not by itself establish what would have happened to comparable participants without the regimen.

The study opened in 2009-06, reached primary completion in 2015-08, and is listed as completed. The intervention description specifies four preoperative cycles, each 21 days long. Pemetrexed was administered intravenously for 10 minutes at 500 mg/m2 on Day 1 of each cycle; carboplatin was administered intravenously at AUC 6.0 on Day 1 of each cycle. These are the regimen details recorded, not evidence about how consistently treatment was delivered or whether participants completed all cycles.

The primary outcome is 3-Year Overall Survival Rate, expressed as a percentage of participants, with a 95% confidence interval specified as the measure of dispersion and a timeframe of 36 months. The endpoint description identifies the Intent-to-Treat population and defines survival from the first study treatment to death from any cause. The numerical outcome and interval are not reported in the registry record.

The listed secondary outcomes include objective tumor response, pathologic response rate, residual disease as an assessment of pathological partial response, and complete resection rate. The record does not give numerical results for these outcomes. It also does not report the eligibility details beyond the selected stages and histologies summarized, the number screened or excluded, treatment completion, or reasons for discontinuation.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Patients (pts) received up to 4 cycles (12 weeks) of preoperative chemotherapy and were restaged after 6 and 12 weeks. Pts with progressive disease or intolerable toxicity came off study; pts who remained surgical candidates had resection at weeks 15-18. Following resection pts received no further planned protocol treatment.

Preoperative Chemotherapy

Reasons for not completing this period:

Surgical Resection

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 3-Year Overall Survival Rate, measured over 36 months.

The primary endpoint is the percentage of participants alive at 3 years from the date of first study treatment, with death from any cause defining the event. It is described for the Intent-to-Treat population. The registry lists a 95% confidence interval as the accompanying measure of uncertainty, but neither the percentage nor its interval is reported in the registry record.

Overall survival is a time-to-event outcome: it uses the timing of death rather than only whether a participant had a particular tumor response at a single assessment. In general, overall survival can provide a direct description of survival during a specified follow-up period. It is also affected by events after treatment begins, including factors the trial does not measure or control. A single-group result therefore describes the observed participants’ survival; without a comparator, it cannot isolate the treatment’s contribution to that survival.

The endpoint description says that time begins with first study treatment and ends at death from any cause. It does not explain how participants without a recorded death by the assessment time were handled, how incomplete follow-up was addressed, or what estimator was used. Those details are not reported in the registry record. The posted result is also needed to understand the observed survival percentage and the precision of its estimate; without it, the endpoint cannot be interpreted quantitatively.

The secondary outcomes address tumor response and surgical or pathological findings. They complement the survival question, but their definitions, assessment timing, and results are not reported in enough detail to characterize them further.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the design allow researchers to estimate?

A single-group clinical trial follows participants receiving the intervention and describes outcomes in that group. Here, the record lists one Pemetrexed/Carboplatin group, enrollment of 46, and no allocation to a separate comparator. This design can provide descriptive information about outcomes observed among the enrolled participants. It cannot, on its own, separate an intervention effect from the effects of patient characteristics, the natural course of illness, subsequent care, or other influences on outcomes.

Randomized comparison is a general method for balancing measured and unmeasured characteristics between treatment groups, on average, so that outcome contrasts can more credibly estimate treatment effects. This record does not describe a randomized comparison; allocation is listed as NA, and the design is single group. Consequently, any interpretation should remain descriptive rather than comparative or causal.

How should the primary endpoint be read?

Overall survival is a time-to-event measure. The event is death from any cause, and the record defines the time origin as first study treatment. The stated assessment horizon is 36 months, also described as 3 years. The endpoint is reported as a percentage of participants in the Intent-to-Treat population, with a 95% confidence interval specified.

In general, time-to-event methods are useful when participants may have different lengths of follow-up or when some have not experienced the event by the end of observation. Those participants can contribute information up to their last known follow-up under appropriate methods. The registry record does not identify the estimator, the censoring rules, follow-up completeness, or the handling of missing survival status. It also does not report the endpoint estimate or its interval. Thus, the definition is available, but the analytic implementation and numerical result are not.

A confidence interval is a statistical expression of uncertainty around an estimated quantity, interpreted under a specified method and its assumptions. It is not a guarantee that a particular value lies within the interval, and it is not a substitute for a comparator. Here, the stated confidence level is 95%, but the interval bounds are not reported in the registry record. No assessment of the observed estimate’s precision is possible from the record alone.

What do the response outcomes add?

The registry names objective tumor response, pathologic response rate, residual disease as an assessment of pathological partial response, and complete resection rate as secondary outcomes. These outcomes concern tumor assessment or surgical and pathological findings rather than the primary survival endpoint. In general, a response measure requires an operational definition, assessment schedule, and rules for evaluability. The record does not provide those details or numerical results, so the outcomes cannot be interpreted quantitatively.

Several outcomes can answer related but distinct questions. A radiographic response, a pathological finding, and a successful resection are not interchangeable endpoints. Their meanings depend on how each is defined and assessed. No hierarchy, endpoint-specific statistical method, or adjudication process is reported in the registry record.

What is the role of planning calculations?

Sample-size calculations are design tools. They estimate how many participants may be needed under chosen assumptions about the effect to detect, variability or event frequency, significance level, power, and allocation. They do not report what happened in a completed trial. When calculations use an observed effect from the same trial, they are retrospective planning illustrations and should not be described as observed power.

The stated planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The stated effect is a fraction of the observed effect. These are assumptions for an illustration, not the design description of this single-group trial. The record does not report the numerical planning outputs, nor does it report a trial-specific sample-size justification. The justification and any other design assumptions are not reported in the registry record.

For time-to-event planning, the stated method refers to a total number of events; for other outcomes, it refers to participants per arm. That distinction is general: time-to-event precision and power often depend strongly on the number of events, while other designs may be expressed in participant counts. These hypothetical calculations cannot supply missing primary results or make the one-group design equivalent to a randomized comparison.

Why can event rates vary from their assumptions?

General statistical methods often assume a particular distribution for the outcome. For count outcomes, over-dispersion means that observed variability exceeds what a simple model assumes. If over-dispersion is ignored, standard errors can be too small and intervals too narrow. The registry record does not identify a count-outcome model for this trial, and the primary endpoint is described as a percentage of participants. This general point should not be mistaken for evidence that over-dispersion occurred in these data.

More broadly, estimates depend on outcome definitions, analysis populations, missingness, follow-up, and model choices. Those details matter when evaluating precision and reproducibility. The record identifies the primary population as Intent-to-Treat but does not describe the statistical model or the handling of missing observations. The number of analysis entries is listed as zero, and no numerical primary analysis is 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("Pemetrexed/Carboplatin/Surgery")
n      <- c(46)
events <- c(21)
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  Pemetrexed/Carboplatin/Surgery  46  21  45.7  30.9  61.0  32.2  59.8  32.1  59.9

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes or performs multiple statistical tests. In general, the more opportunities there are to declare a finding, the more important it is to understand which outcomes were primary, which were secondary, whether analyses were planned in advance, and whether any adjustment or testing hierarchy was used. This record names a primary survival endpoint and several secondary response and surgical outcomes. It does not report a multiplicity strategy, confirmatory testing hierarchy, or adjustment for multiple comparisons. Those details are not reported in the registry record.

The endpoint description identifies the Intent-to-Treat population for overall survival. The record does not give further analysis-population definitions for the secondary outcomes, such as which participants were eligible for each assessment or how participants with unavailable assessments were treated. It also does not describe protocol deviations or exclusions from any analysis. The number of analysis entries is listed as zero, and numerical outcome analyses are not reported in the registry record.

Missing data can affect estimates when outcome status or follow-up is unavailable. General approaches depend on why data are missing and on the endpoint’s structure. For survival outcomes, incomplete follow-up is commonly handled through censoring methods, but the method and assumptions must be specified to assess their implications. For response outcomes, missing assessments can affect which participants are evaluable and how response percentages are interpreted. This record does not report missing-data amounts, reasons for missingness, imputation methods, censoring details, or sensitivity analyses.

Because the primary estimate and confidence interval are absent, and because methods for multiplicity and missing data are not stated, the registry record does not support a detailed evaluation of the statistical analysis. These gaps should be distinguished from evidence of a particular analytic problem: the methods are not reported in the registry record, so their adequacy cannot be assessed from this information.

12. Safety Results

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

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: thrombocytopenia (4.3%), neutropenia (4.3%), anemia (4.3%), atrial fibrillation (2.2%), leukopenia (2.2%).

The adverse-event record lists 46 participants at risk for serious events and 18 participants affected in the Pemetrexed/Carboplatin group. It also lists 46 participants at risk for other events and 46 affected. These are counts in the registry record; they are not a comparison with an untreated or alternative-treatment group. The record gives no deaths value for this group.

The listed serious-event terms and percentages are thrombocytopenia, 4.3%; neutropenia, 4.3%; anemia, 4.3%; atrial fibrillation, 2.2%; leukopenia, 2.2%; and anorectal cellulitis, 2.2%. These are the terms identified as the top serious events in the record. They should be read as reported safety entries, not as proof that the regimen caused each event. The record does not provide event timing, severity grades, duration, relatedness assessments, or participant-level overlap among event terms.

In general, the number of participants with at least one event is different from the total number of event episodes. A participant may experience more than one event or more than one event type. The registry record does not give episode counts or enough detail to determine overlap. The absence of a reported deaths value is not evidence that no deaths occurred; a death result is not reported in the registry record.

These figures describe safety reporting in a single group. Without a comparator and without fuller event definitions and follow-up details, they cannot establish a comparative safety difference or provide a complete account of treatment risk.

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

The record documents a completed Phase II, single-group study of preoperative pemetrexed and carboplatin in select stage IB, II, and III non-squamous non-small-cell lung cancer. It specifies the regimen, enrollment of 46, a primary 3-year overall-survival endpoint, and several secondary outcomes. It also records serious and other adverse-event counts and selected serious-event terms.

Those design and reporting details describe what the trial set out to examine and what safety entries were recorded. They do not provide the primary survival percentage or its 95% confidence interval. Numerical results for the secondary outcomes are also not reported in the registry record. Without these results, the magnitude and precision of the primary outcome and the measured response outcomes cannot be evaluated.

Even with complete outcome estimates, a single-group study would not establish a treatment effect relative to another option. There is no concurrent control group against which to compare survival, response, resection, or adverse events. In general, observed outcomes can reflect the intervention as well as patient selection, disease characteristics, subsequent care, follow-up, and other factors. The record therefore supports description of the trial design and its listed safety data, not a causal conclusion about benefit or comparative risk.

The trial summary states that the work was intended to examine safety and activity and to inform whether further investigation in a larger, prospective, randomized trial might be justified. That rationale should not be read as a finding that the regimen proved safe or active. The registry record does not report such a conclusion or the numerical outcomes needed to evaluate it.

The safety figures likewise need a limited reading. They report affected-participant counts and selected serious-event terms, but do not provide detailed event-level context or a control group. They cannot show whether the events were caused by treatment or how the safety profile compares with another approach.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how design determines the scope of a conclusion. A Phase II study can describe outcomes and safety observations in a defined group, but a single-group structure does not provide the direct contrast required for a randomized treatment comparison. That distinction remains important even when the clinical question concerns a consequential endpoint such as overall survival.

The registry gives a clear endpoint concept: the percentage alive at 3 years from first study treatment, in the Intent-to-Treat population, with a 95% confidence interval specified. Yet the numerical estimate and interval are not reported in the registry record. This makes the case a useful reminder that a stated endpoint is not itself a result. To interpret an estimate, readers need both its value and an account of its uncertainty and analysis.

The trial also lists response, pathological, residual-disease, and resection outcomes. These can address different aspects of treatment evaluation, but they require operational definitions and suitable analysis populations. The record’s outcome names alone do not reveal how the measures were assessed or what the results were. Statistical interpretation depends on those details, not merely on an outcome label.

Finally, planning calculations and observed analyses serve different purposes. A planning illustration based on an observed effect is not observed power and does not repair the limitations of a single-group design. It can only show what participant counts might follow under chosen assumptions. Keeping descriptive outcomes, comparative effects, uncertainty intervals, and design calculations conceptually separate is essential when reading clinical trial records.

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 most direct limitation is that the primary endpoint’s numerical result and 95% confidence interval are not reported in the registry record. Numerical results for the named secondary outcomes are also not reported. The record therefore does not permit a quantitative assessment of survival, tumor response, pathological response, residual disease, or complete resection.

The trial has one treatment group and no like-for-like control group. This limits interpretation to descriptive outcomes among participants receiving the regimen and prevents an internal estimate of comparative effect. The record lists enrollment of 46, but does not report screening, exclusions, the number completing treatment, or detailed follow-up. It also does not report the number of deaths or survival censoring information.

Important methods are not described in the registry record, including the statistical estimator for overall survival, the handling of missing observations, analysis methods for secondary outcomes, multiplicity procedures, and sensitivity analyses. The endpoint definition and specified confidence interval are not enough to reconstruct those methods. The record also does not report a trial-specific sample-size justification or numerical planning outputs.

The safety information is limited to affected-participant counts and selected serious-event terms. Event-level detail, timing, severity grading, attribution, participant overlap, and comparative safety data are not reported. The death field has no value recorded, which should not be interpreted as a zero count.

These limitations are limitations of what is reported in the registry record. They do not establish that unreported procedures or analyses were not performed; those details are not reported in the registry record and cannot be assessed from it.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Preoperative Pemetrexed and Carboplatin for Select Stage IB, II, and III Non-Squamous Non-Small-Cell Lung Cancer — NCT00906282.
• Registry record last updated 2017-02-24; status last verified 2017-01.
• Results first posted 2017-02-24.
• 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

NCT00906282 is a completed, multicenter Phase II trial of preoperative pemetrexed and carboplatin for patients with select stage IB, II, and III non-squamous non-small-cell lung cancer. It enrolled 46 participants and used a single-group design with no masking. The intervention description specifies four 21-day cycles, with pemetrexed and carboplatin given on Day 1 of each cycle.

The primary outcome is the percentage alive at 3 years from first study treatment, with a 95% confidence interval specified for the Intent-to-Treat population. The primary numerical result and interval are not reported in the registry record. Secondary outcomes include objective tumor response, pathologic response, residual disease, and complete resection; their numerical results are not reported either.

The registry lists 18 participants affected by serious adverse events and 46 affected by other adverse events. It names selected serious-event terms and percentages but does not provide detailed event context or a comparator. The record describes the study’s rationale as examining safety and activity; it does not report evidence sufficient to determine comparative treatment effects.