Chemotherapy, Radiation Therapy and Immunotherapy Prior to Surgery in Operable Esophageal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Chemotherapy, Radiation Therapy and Immunotherapy Prior to Surgery in Operable Esophageal Cancer (NCT00393068), a na phase 2 study. The primary outcome is Pathologic Complete Response (pCR) Rate.

2. Clinical Question

3. Trial Design

This is a phase II, single-group study of preoperative treatment for localized esophageal cancer. The registry record lists 62 participants enrolled and a completed study status. The start is recorded as 2007-02 and primary completion as 2009-09. These are registry dates; they do not by themselves describe the timing of individual participants’ treatment or follow-up.

The treatment strategy combined chemotherapy and radiation therapy with bevacizumab and erlotinib before surgery. The listed interventions also include paclitaxel, carboplatin and 5-FU. According to the arm description, study treatment was to be given over 6 weeks, Days 1-42, before surgery; continuous-infusion 5-FU was specified from Day 1 through Day 35. The available arm description does not give complete schedules for every intervention, so further timing or dose details are not reported in the registry record.

There is no separate control arm in the registry design, and allocation is NA. The study therefore describes outcomes in a treated group rather than an estimate of the difference between two assigned groups. Without a concurrent control, outcome rates cannot distinguish the contribution of the study regimen from other influences on outcomes. This is a design limitation, not evidence for or against the regimen.

The primary endpoint is pathologic complete response after resection, with overall survival and progression-free survival listed as secondary outcomes. The registry record also describes collection of toxicity information. It does not report a statistical analysis plan, the number who underwent surgery, or results for the listed efficacy outcomes. Those missing details limit how fully the study’s implementation and outcome estimates can be assessed.

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

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 (pCR) Rate, measured over 18 months.

The primary endpoint is the Pathologic Complete Response (pCR) Rate. The registry defines pCR as no residual viable cancer at the primary site or regional lymph nodes on pathologic review of the surgical specimen, for participants who went to surgical resection. It is a binary outcome at the participant level: a person either meets this definition or does not.

Pathologic review assesses the tissue removed at surgery, rather than relying only on symptoms or imaging. In a preoperative treatment study, this endpoint is used to describe whether viable cancer is found in the specified tissues after treatment and resection. It is a measure of response in the resected participants; by itself, it does not establish how the treatment compares with another strategy or what the result means for longer-term outcomes.

The registry lists an 18-month timeframe for this outcome. That registry timeframe does not establish the actual timing of each participant’s surgery or pathology assessment. The number of participants who underwent resection, the number meeting the pCR definition, and the resulting pCR rate are not reported in the registry record. Thus, the endpoint is defined, but its observed result cannot be interpreted from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study’s design and outcome definitions determine what statistical summaries would be meaningful. Because the registry describes a single treatment group, the basic efficacy summary for the binary pCR endpoint would be the number meeting the endpoint among the relevant participants, together with the denominator. The endpoint definition restricts that denominator to participants who went to surgical resection. Neither count is reported in the registry record, so a pCR proportion cannot be read from it.

What does a binary endpoint summarize?

A binary endpoint classifies each participant into one of two outcome categories, such as meeting or not meeting a specified response definition. The pCR definition is based on the absence of residual viable cancer at the primary site or regional lymph nodes on pathology review of the surgical specimen. For a group, the observed proportion is a descriptive summary. It does not alone explain why outcomes occurred, and in a single-group study it does not estimate a contrast with a control group.

The denominator matters as much as the count. Here, the pCR definition applies to participants who underwent resection, whereas the registry reports 62 enrolled. These quantities should not be treated as interchangeable. The number resected, the number with an assessable specimen, and the handling of participants who did not reach surgery are not reported in the registry record. Without that information, the completeness and scope of any pCR analysis cannot be evaluated.

How should uncertainty be understood?

In general, a confidence interval describes a range of values compatible with the observed data under a specified statistical method. It communicates sampling uncertainty around an estimate; it does not guarantee that the true value lies within the interval or account automatically for biases in study design. For a binary outcome, the interval’s interpretation depends on the numerator, denominator and method used to construct it. The registry record does not report a pCR estimate or confidence interval, so no endpoint-specific precision can be assessed.

In general, estimates based on fewer observations tend to be less precise than estimates based on more observations, all else being equal. But the enrolled count alone does not establish the effective sample size for this endpoint, because the relevant group is the resected participants. The actual number contributing to the endpoint is not reported. No additional numerical precision estimate should be inferred from enrollment alone.

What does the single-group structure allow?

A single-group analysis can describe outcomes observed in that group. It cannot, without further design or assumptions, separate treatment effects from the course of disease, participant characteristics, other care, or chance variation. Random allocation to a concurrent comparator is not described: the model is single group and allocation is NA. Therefore, statistical comparisons between randomized groups are not applicable to the recorded design.

General statistical methods for a two-group comparison could be used in a different study, but they would not create a valid control group for this study. The registry does not describe a prespecified external comparator or a method for adjusting comparisons to another population. Any such comparison would need to address differences in eligibility, care, outcome assessment and follow-up; those issues are not resolved by a numerical calculation alone.

What do the planning assumptions mean?

The registry’s planning illustration is described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It defines the effect to detect as a stated fraction of the observed effect. These are assumptions for a hypothetical planning calculation, not the design of the completed single-group study. The registry record does not report the study’s own sample-size justification, and the planning illustration must not be read as observed power or as evidence that a particular effect was found.

In general, power is a design-stage probability of detecting an effect under specified assumptions if that effect is present. It is not a measure of the probability that a treatment works, and it cannot be recovered as an informative post-study result simply by inserting an observed estimate into a formula. Planning calculations based on an observed effect are especially dependent on the estimate used and the assumptions attached to it. The calculation’s illustration status does not change the study’s single-group design or supply the missing pCR result.

The normal approximation is a mathematical approach used in some planning calculations. Whether an approximation is suitable depends on the setting and assumptions; the registry does not report a separate justification for it. Equal allocation between two hypothetical arms is likewise a planning premise, not a description of allocation in this trial. No calculated sample-size value is stated in the registry record.

How should analyses be distinguished from listed outcomes?

The record lists a primary pCR outcome, secondary overall survival and progression-free survival outcomes, and three outcome measures. It reports zero analyses. This means no analysis entries are present in the registry record; it does not establish that no analyses were performed outside that record. The specific statistical methods, model specifications, confidence intervals, hypothesis tests and analysis populations are not reported in the registry record.

In general, a prespecified analysis plan helps readers distinguish planned analyses from choices made after data are available. Details such as how incomplete endpoint data are handled, whether an analysis is adjusted, and which participants are included can change interpretation. Those details are not reported in the registry record. Accordingly, the endpoint definition and study design can be described, but no unreported inferential method or result should be attributed to this trial.

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("Treatment")
n      <- c(62)
events <- c(18)
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  Treatment  62  18  29.0  18.2  41.9  19.2  41.3  19.2  41.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures: the primary pCR outcome and the secondary outcomes overall survival and progression-free survival. It reports zero analyses. The absence of analysis entries means that the registry record does not show how those outcomes were analyzed; it does not prove that no analysis occurred. The statistical analysis plan, including any hierarchy of outcomes or adjustment for multiple testing, is not reported in the registry record.

In general, multiplicity arises when several outcomes or statistical tests are considered. If each is treated as an independent opportunity to declare a result, the chance of at least one false-positive finding can increase. A prespecified outcome hierarchy or an adjustment method can address aspects of this issue, but neither is described in the registry record. Since no outcome analysis results are posted, there is no reported set of tests from which to assess multiplicity handling.

The primary endpoint definition applies to participants who went to surgical resection, but the number who underwent surgery is not reported. The registry does not state how participants without a surgical specimen were handled, whether endpoint data were incomplete, or whether missing observations were imputed or otherwise addressed. The analysis populations for the primary and secondary outcomes are also not reported in the registry record.

These omissions matter because the denominator and treatment of missing information affect how an estimate should be interpreted. It is not possible to infer that all 62 enrolled participants contributed to pCR, or that any particular missing-data approach was used. For overall survival and progression-free survival, event definitions, censoring rules and analysis populations are not reported in the registry record.

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: dehydration (16.1%), febrile neutropenia (9.7%), esophagitis (8.1%), death (4.8%), diarrhea (4.8%).

The Treatment group has 62 participants at risk for the serious adverse event summary, with 38 recorded as affected. For the other adverse event summary, 62 are at risk and 59 are recorded as affected. The registry also lists 29 deaths in the group. These are reported counts; they should not be treated as proof that the treatment caused the events or deaths.

The most frequently listed serious adverse event terms are Dehydration at 16.1%, Febrile Neutropenia at 9.7%, Esophagitis at 8.1%, Death at 4.8%, Diarrhea at 4.8%, and Leukocytes at 3.2%. These are registry-listed percentages. Detailed event definitions, timing, attribution, and the denominator used for each term are not reported in the registry record.

The count of 29 deaths and the serious adverse event term Death at 4.8% are separate entries with different forms of reporting. The registry record does not explain how they relate, so they should not be reconciled by assumption. Nor does it report whether events were considered related to treatment. Safety counts describe what is recorded in the group; they do not establish comparative risk, because no control group is described.

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

The record establishes that a phase II, single-group study enrolled 62 participants with esophageal cancer to receive a preoperative treatment strategy that included chemotherapy, radiation therapy, bevacizumab and erlotinib, followed by surgery. It defines pCR as the absence of residual viable cancer at the primary site or regional lymph nodes on pathology review for participants who underwent resection. It also lists overall survival and progression-free survival as secondary outcomes and provides treatment-group safety counts.

The record does not report the primary outcome result. The pCR count, pCR rate, number resected and an uncertainty interval are not reported in the registry record. Overall survival and progression-free survival results are also not reported. Thus, the posted information cannot establish the frequency of pCR in this study or the precision of an efficacy estimate.

Even if a pCR result were available, a single-group result would describe outcomes in the enrolled treatment group rather than estimate a randomized difference between strategies. No concurrent comparator is identified. Separating treatment effects from other explanations would require evidence and assumptions beyond the design details recorded.

The safety counts and listed serious adverse event terms are useful descriptions of what the registry records for the treatment group. They do not, on their own, establish causation or show how risks compare with another group. The relationship between the reported deaths and the serious adverse event term Death is not reported.

The appropriate interpretation is therefore limited: the record describes the design, endpoint definition and some safety information, but it does not provide the primary efficacy result or a controlled treatment-effect estimate. Neither the completed status nor the enrollment count fills that evidentiary gap.

14. Why This Trial Matters Statistically

This trial illustrates why a defined endpoint and an available endpoint result are different things. The registry gives a specific pathology-based definition of pCR, but does not report the participants meeting it, the number who underwent surgery, or the resulting rate. An endpoint definition clarifies what investigators intended to measure; it cannot substitute for the observed numerator and denominator needed to interpret the result.

It also shows why design matters when reading a clinical outcome. A single-group study can describe the experience of participants receiving a regimen, but it does not supply a concurrent counterfactual for what might have occurred under another strategy. The fact that the study is phase II does not remove that distinction. Without a comparator, a treatment effect cannot be read as a between-group contrast.

The record further makes the difference between enrollment and endpoint population important. Although 62 participants were enrolled, pCR is defined for participants who went to resection. The number in that endpoint population is not recorded. A sound interpretation therefore requires attention to who contributed outcome data, not just the overall enrollment figure.

Finally, the registry contains safety counts while omitting efficacy results and analysis details. Statistical reporting is most informative when endpoint counts, denominators, uncertainty, analysis populations and methods can be read together. In this record, several of those elements are not reported, so conclusions must remain correspondingly narrow.

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 trial-specific limitation for interpretation is the single-group design. Allocation is recorded as NA, masking is none, and no separate control group is described. Consequently, the record does not support a randomized comparison of the treatment strategy with another approach.

The primary endpoint is defined only for participants who went to surgical resection, but the number resected is not reported. The number with pCR and the pCR rate are also not reported. The registry lists 62 enrolled, but does not establish that all enrolled participants contributed to the primary endpoint. The analysis population and handling of participants who did not undergo surgery are not reported in the registry record.

Although the study is marked completed, outcome results for pCR, overall survival and progression-free survival are not reported. The record also does not report confidence intervals, hypothesis tests, analysis methods, follow-up detail for the secondary outcomes, or a statistical analysis plan. The registry records zero analyses, but this does not establish whether analyses were conducted outside the record.

The safety information has additional limits. The registry lists affected counts and serious adverse event terms, but does not report event attribution, timing, detailed definitions or term-specific denominators. The relation between the 29 deaths and the Death term listed among serious adverse events is not reported. The data therefore do not establish treatment causality or comparative safety.

The arm description gives some treatment timing but not complete schedules for all interventions. Detailed eligibility criteria and the study’s sample-size justification are not reported in the registry record. These omissions constrain assessment of the population, treatment implementation and statistical design.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Chemotherapy, Radiation Therapy and Immunotherapy Prior to Surgery in Operable Esophageal Cancer — NCT00393068.
• Registry record last updated 2022-03-02; status last verified 2022-02.
• Results first posted 2012-11-30.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT00393068 is a completed phase II, single-group study of preoperative chemotherapy and radiation therapy with bevacizumab and erlotinib for operable esophageal cancer. The registry lists 62 participants enrolled, no masking, and allocation as NA. Treatment was followed by surgery.

The primary endpoint is pathologic complete response, defined by pathology review of the surgical specimen for participants who underwent resection; the registry lists an 18-month timeframe. Overall survival and progression-free survival are listed as secondary outcomes. The record does not report the pCR result or secondary outcome results, and reports zero analyses.

For the Treatment group, the registry records 38 of 62 affected in the serious adverse event summary, 59 of 62 affected in the other adverse event summary, and 29 deaths. It lists serious adverse event terms including Dehydration, Febrile Neutropenia, Esophagitis, Death, Diarrhea and Leukocytes. These entries do not establish causation or comparative risk.