Radiation Therapy, Pemetrexed Disodium, and Carboplatin in Treating Patients With Locally Advanced Esophageal Cancer That Can Be Removed By Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Radiation Therapy, Pemetrexed Disodium, and Carboplatin in Treating Patients With Locally Advanced Esophageal Cancer That Can Be Removed By Surgery (NCT00268437), a na phase 2 study. The primary outcome is Pathologic Complete Response Rate.

2. Clinical Question

The clinical question is about response after a combined preoperative treatment regimen, not a comparison of that regimen with another treatment. Because all participants are described within a single intervention group, the trial record does not support a direct estimate of how outcomes would differ under an alternative regimen or without the regimen.

The registry lists overall survival as a secondary outcome. It does not report an overall-survival estimate or specify a timeframe for that outcome in the record. The available entry therefore identifies what the investigators intended to assess, but it does not provide the outcome data needed to describe the observed response or survival experience.

3. Trial Design

This is a phase II trial recorded as a single-group study. The registry lists enrollment of 27 and one intervention group, Pemetrexed/Carboplatin, described as pemetrexed, carboplatin, and radiation. Conventional surgery and neoadjuvant therapy are also listed among the interventions. Allocation is recorded as not applicable, and masking as none.

The study concerns locally advanced esophageal cancer that could be removed by surgery. Its primary endpoint is pathologic complete response rate, assessed from baseline to the time of surgery, with the timing described as around 10 - 18 weeks post-baseline. Overall survival is listed as a secondary outcome. The registry record identifies two outcome measures, but it does not provide outcome results or analysis entries.

The listed start date is 2006-04, the primary completion date is 2008-01, and the current status is terminated. The reason for termination, whether recruitment ended as planned, and any changes to the study conduct are not reported in the registry record. The record also does not describe a concurrent control group, an allocation process, or masking beyond the entries noted above.

Because the design is single-group, the principal endpoint is an estimate of response within the treated group rather than a between-group treatment contrast. Enrollment is not necessarily the same as the number evaluable for the primary endpoint. The registry describes the primary response proportion as based on evaluable patients, but does not report how many were evaluable or how that population was defined.

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 Pathologic Complete Response Rate, measured over Baseline to time of surgery (around 10 - 18 weeks post-baseline).

The primary endpoint is the pathologic complete response rate. It is defined as the proportion of evaluable patients with a pathologic complete response, calculated from the response count and the total number of evaluable patients. The registry states that a 95% confidence interval for the true response rate was to be calculated. The response count, denominator, estimated rate, and confidence interval are not reported in the registry record.

Pathologic response is assessed at surgery after preoperative treatment. In general, a pathologic endpoint uses examination of tissue obtained during surgery to assess whether cancer is present in the examined tissue. It can therefore describe response at the treated site in a way that a symptom report or imaging assessment alone may not. That general rationale does not establish what this trial found.

The record also includes a description of measurable disease based on RECIST v1.0, including lesion-size criteria and the preference for CT over chest x-ray when both apply. That description does not supply a pathologic response result. The primary endpoint timeframe is baseline to surgery, around 10 - 18 weeks post-baseline; the actual timing for individual participants is not reported. Whether every enrolled participant proceeded to surgery, and how patients without an evaluable surgical specimen were handled, are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical description identifies a proportion as the primary endpoint and specifies a 95% confidence interval. It does not report the estimated proportion or interval. The methods below explain how such quantities are generally interpreted; they are not a reconstruction of the trial’s unreported analysis.

What does a response proportion describe?

In general, a response proportion is the number of evaluable participants meeting the response definition divided by the number of evaluable participants. It describes the observed frequency of the endpoint in the analysis population. The result depends on both the response definition and the denominator. If participants are excluded from evaluation, the reasons for exclusion and the handling of missing assessments can affect how representative the estimate is. In this trial, the registry gives the intended denominator concept—evaluable patients—but does not report its size, the response count, or how evaluability was determined.

How should a confidence interval be read?

In general, a confidence interval communicates uncertainty around an estimate. For a response proportion, it indicates a range of values compatible with the observed data under the method’s assumptions. A wider interval indicates less precision, while a narrower interval indicates greater precision. The interval is not a guarantee that the true value lies within its limits, and it does not measure the probability that treatment is effective. The registry specifies a 95% confidence interval for the primary rate, but the limits and the calculation method are not reported.

Precision depends in part on the amount of evaluable information and on the observed response frequency. Enrollment of 27 does not reveal the primary endpoint’s precision because the evaluable denominator and response count are not reported. It would be inappropriate to infer a rate or confidence interval from enrollment alone.

What can a single-group estimate establish?

In general, a single-group estimate describes outcomes observed in the group receiving the study regimen. It does not establish what would have happened to comparable participants under another regimen or no regimen. Without a concurrent comparator, differences attributable to treatment cannot be separated from differences in patient characteristics, selection, clinical practice, or other influences. The registry records this study as single-group and lists no comparator arm.

The regimen includes radiation, pemetrexed, and carboplatin before surgery. A response observed after that sequence would reflect the overall treatment context; a single-group result would not identify the contribution of each component. The registry does not report component-specific analyses.

What do planning assumptions mean?

In general, sample-size calculations are planning tools: they describe the number of participants or events that a hypothetical design would require under specified assumptions. They are not observed power, and they do not show that a study had a particular probability of detecting an effect after its results are known. The registry’s planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. For time-to-event calculations, the assumption describes a total number of events; for other calculations, it describes participants per arm.

Those assumptions do not describe this trial’s recorded single-group design. They should be interpreted as planning illustrations, not as the study’s actual allocation or as evidence of a prespecified sample-size justification. The protocol’s own sample-size justification is not reported in the registry record. Since the observed primary effect is not reported, the planning illustration cannot be independently interpreted from the registry entry alone.

Why does over-dispersion matter in general?

In general, a statistical model can be too confident if the observed variation exceeds what its assumptions allow. Over-dispersion is such a mismatch, most often discussed for count outcomes. It can make standard errors too small and confidence intervals too narrow if it is ignored. The primary endpoint in this trial is a binary response proportion, not a count rate with a reported over-dispersion model. The registry does not report an over-dispersion assessment or a model-based analysis, so this concept does not supply additional evidence about the trial’s result.

The registry identifies no analysis entries. It does not specify the estimator for the confidence interval, covariate adjustment, interim monitoring, or analysis populations. These details cannot be inferred from the stated endpoint definition or from the phase designation.

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")
n      <- c(26)
events <- c(6)
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  26  6  23.1  9.0  43.6  11.0  42.1  11.1  42.3

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity concerns the interpretation of multiple statistical questions, such as several endpoints, comparisons, or repeated analyses. In general, making many significance tests without an appropriate plan can increase the chance of at least one apparently notable result arising by chance. Whether and how to address multiplicity depends on the study’s prespecified objectives and analysis plan.

The registry lists pathologic complete response as the primary outcome and overall survival as a secondary outcome. It records two outcome measures and no analysis entries. A multiplicity strategy, including any adjustment or testing hierarchy, is not reported in the registry record. The absence of analysis entries does not establish whether analyses were conducted elsewhere, nor does the listing of outcomes establish which results were formally tested.

The record describes the primary rate as based on evaluable patients, but does not report the evaluable denominator, the response count, or the rules for inclusion in that population. Missing-data handling, reasons for missing or unevaluable assessments, and any sensitivity analyses are not reported. The record also does not say whether the primary analysis followed an intention-to-treat, per-protocol, or other population definition. It is therefore not possible to assess how exclusions or incomplete assessments may have affected the endpoint estimate.

The overall-survival outcome lacks a reported analysis population, follow-up description, estimate, and uncertainty measure. The registry does not report how censoring or missing follow-up was handled. These omissions limit interpretation of both outcome completeness and the relationship between the intended analysis and the results available in the registry.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hemoglobin decreased (3.7%), atrial fibrillation (3.7%), dysphagia (3.7%), esophageal fistula (3.7%), nausea (3.7%).

The registry records 27 at risk and 6 affected for serious adverse events in the Pemetrexed/Carboplatin group. For other adverse events, it records 27 at risk and 27 affected. These are separate summaries and should not be added together: the record does not explain whether participants could appear in both categories or how the categories relate.

The listed serious adverse event terms are hemoglobin decreased, atrial fibrillation, dysphagia, esophageal fistula, nausea, and small intestinal obstruction. Each is reported as 3.7%. The term-level denominator and whether these percentages refer to participants or events are not described in the registry entry. The list is not a complete account of every possible event, and the percentage entries should not be interpreted as comparative risks because no control group is listed.

Deaths are not reported in the registry record. The event observation period, seriousness criteria, collection procedures, and attribution to treatment are also not reported. These limitations mean that the entries offer a descriptive account of recorded adverse events, not a full safety profile or a causal estimate of harm.

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

The registry record establishes that this phase II study examined preoperative radiation with pemetrexed and carboplatin for locally advanced esophageal cancer that could be removed by surgery. It identifies pathologic complete response at surgery as the primary endpoint and overall survival as a secondary outcome. It also provides limited adverse-event summaries.

It does not report the primary response count, evaluable denominator, response rate, or confidence interval. It therefore does not support a conclusion about the magnitude or precision of observed pathologic response. The enrollment entry cannot replace the endpoint denominator, because the record defines the rate using evaluable patients and does not state how many met that definition.

The single-group design has no listed concurrent comparator. Even if an outcome estimate were available, it would describe the treated group rather than establish a difference attributable to the regimen. The design cannot isolate the effects of radiation, pemetrexed, carboplatin, or surgery from one another. No overall-survival result is reported either.

The safety entries should be read as recorded event counts and percentages with important interpretive gaps. They do not provide a comparison with another treatment, and the record does not specify the observation period or term-level denominator. The terminated status does not explain why the trial ended.

In general, a trial registry entry can help identify study design, intended endpoints, and reported results, but incomplete outcome reporting constrains statistical interpretation. For this record, the study question and selected safety information are available, while the central efficacy result and key analysis details are not reported. Neither a claim of benefit nor a claim of no benefit can be drawn from the registry information alone.

14. Why This Trial Matters Statistically

This trial illustrates why study design and outcome reporting must be considered together. The registry identifies a single-group phase II design and a pathologic response endpoint. Those details clarify the intended question, but they do not substitute for the response count, denominator, estimate, and confidence interval needed to judge the result.

The distinction between enrollment and evaluability is especially important for a proportion endpoint. The record lists enrollment of 27, while the endpoint is defined among evaluable patients. Because the evaluable population is not reported, the reader cannot determine whether the endpoint denominator matches enrollment or how incomplete assessments were handled.

The design also matters for causal interpretation. A single-group study can describe outcomes after a combined regimen, but without a comparator it cannot quantify a treatment contrast. The regimen includes several components and surgery, so a response in this setting would not by itself identify which component accounts for the observed outcome.

The record provides a useful example of the difference between an endpoint definition and an endpoint result. It also shows why confidence intervals, analysis populations, missing-data methods, and prespecified testing plans matter: these elements explain the uncertainty and scope of an estimate. Their absence from the registry record leaves major questions unanswered.

Finally, the adverse-event summaries demonstrate that safety counts require context, including event definitions, observation time, denominators, and a comparison group. The entries can be described, but they cannot establish comparative safety. These statistical distinctions apply broadly to clinical research and are particularly important when registry outcome reporting is incomplete.

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 main limitation for statistical interpretation is that the registry record does not report the primary endpoint result. The pathologic complete response count, evaluable denominator, rate, and 95% confidence interval are all absent. The record therefore does not permit an assessment of response magnitude or precision.

Although enrollment is listed as 27, the number included in the primary analysis is not reported. The definition of evaluability, handling of missing pathology assessments, and reasons for any exclusions are also not reported. It is not possible to establish whether all enrolled participants proceeded to surgery or contributed to the endpoint.

The single-group design has no listed concurrent comparator, limiting comparative and causal interpretation. The record does not report the treatment-delivery details needed to assess completion of radiation, pemetrexed, or carboplatin. It also does not report the reason for termination, the protocol’s sample-size justification, interim analyses, or a prespecified statistical analysis plan.

Overall survival is listed as a secondary outcome, but no estimate, follow-up information, or analysis details are reported. The adverse-event record lacks an observation period, term-level denominator information, and details on event ascertainment and attribution. These gaps limit both efficacy and safety interpretation; they do not establish that unreported procedures or analyses did not occur.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Radiation Therapy, Pemetrexed Disodium, and Carboplatin in Treating Patients With Locally Advanced Esophageal Cancer That Can Be Removed By Surgery — NCT00268437.
• Registry record last updated 2016-07-06; status last verified 2016-07.
• Results first posted 2015-03-27.
• 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

The registry describes NCT00268437 as a phase II, single-group trial of preoperative radiation with pemetrexed and carboplatin for locally advanced esophageal cancer that could be removed by surgery. Enrollment is listed as 27, masking as none, and status as terminated. The primary endpoint is pathologic complete response rate by surgery, around 10 - 18 weeks post-baseline; overall survival is listed as a secondary outcome.

The registry record does not report the primary response count, evaluable denominator, rate, or confidence interval, and it lists no analysis entries. It records 6 affected by serious adverse events and 27 affected by other adverse events, with listed serious-event terms reported at 3.7% each. The reason for termination and key analysis details are not reported. The single-group design does not provide a concurrent treatment comparison.