Neoadjuvant Chemotherapy Versus Radiochemotherapy for Cancer of the Esophagus or Cardia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Neoadjuvant Chemotherapy Versus Radiochemotherapy for Cancer of the Esophagus or Cardia (NCT01362127), a randomized phase 2 study. The primary outcome is Pathological Complete Histological Response (pCR) After Resection Than Chemotherapy Alone in Patients With Resectable Carcinoma of the Esophagus and Cardia..

2. Clinical Question

The clinical question concerns response in the resected tumour, rather than a time-to-event outcome. It compares two preoperative treatment approaches within a randomized, parallel-group study. The registry’s brief description frames the comparison as whether radiochemotherapy gives a higher degree of complete histological response than chemotherapy alone. The outcome record reports counts for each arm; it does not provide individual participant data or additional detail about how the tumour regression grade was applied.

The trial also lists safety, health-related quality of life and swallowing function as secondary outcome topics. Those are relevant to characterizing treatment experience, but the posted primary comparison focuses on pathological response. The registry record does not report posted analyses for the secondary outcomes, so their results cannot be characterized from the information recorded for this comparison.

3. Trial Design

NCT01362127 is recorded as a completed phase 2 randomized trial with a parallel-group model and no masking. Its stated enrollment is 181. The sponsor is Karolinska University Hospital. The registry records a start date of 2006-10 and a primary completion date of 2013-06.

Participants were assigned to radiochemotherapy followed by surgery or chemotherapy followed by surgery. The interventions are listed as radiation and drug treatment, respectively. Both arms include surgery; the question is about the preoperative treatment strategy and the subsequent pathological response after resection. The registry describes the treatment period as 14-16 weeks before surgery and follow-up after surgery until 60 weeks after completed therapy.

The primary outcome is recorded as a count of participants with pathological complete histological response after resection. Its description names the Chireac tumour regression grade. The outcome record does not give a separate analysis population description or explain how participants without an assessable resection specimen would be handled. Those details are not reported in the registry record.

Safety of the respective neoadjuvant therapies, health-related quality of life and swallowing function are also listed as secondary outcomes. The registry identifies three outcome measures and two posted secondary outcomes. It records no posted analyses. This does not establish whether analyses were or were not conducted outside the registry record; it means that the registry entry does not supply those analysis results or a detailed statistical analysis plan.

The trial is unmasked, so participants and study personnel were not described as blinded. Whether pathology assessment was blinded is not reported in the registry record. Details about sequence generation, allocation concealment, eligibility criteria beyond the stated cancer types and resectability, and the conduct of surgery are also not reported in the registry record.

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.

Population: see Klevebro et al Annals of Oncology, primary outcome

6. Primary Endpoint

The registry lists the primary outcome as Pathological Complete Histological Response (pCR) After Resection Than Chemotherapy Alone in Patients With Resectable Carcinoma of the Esophagus and Cardia., measured over Therapy followed in 14-16 weeks before surgery. After surgery the patients will be followed until 60 weeks after completed therapy..

The primary endpoint counts participants with a pathological complete histological response after resection. In general, a pathological response endpoint is based on examination of tissue removed during surgery. A complete response indicates that the specified pathology grading system classified the resected tumour as having a complete histological response. The registry identifies the Chireac tumour regression grade, but does not give further operational details about its scoring or thresholds.

This endpoint is used to describe the tumour’s response to treatment before surgery. Because it is assessed in resected tissue, it provides a direct pathological measure rather than relying only on symptoms or imaging. It can therefore address the study’s stated question about whether one neoadjuvant strategy is associated with a higher degree of complete histological response than the other.

A response endpoint is not the same as a measure of long-term disease control or survival. The registry gives a follow-up timeframe after completed therapy, but the posted primary result is the count of participants meeting the pathological response definition. Whether pathological response predicts later outcomes in this trial is not reported in the registry record. The endpoint also does not, on its own, summarize safety, quality of life or swallowing function; those are separate outcome topics in the record.

7. Results Posted in the Registry

• Chemotherapy: 9.0% (7/78) versus 28.2% (22/78) on control; risk difference -19.2 percentage points (95% CI -31.1 to -7.4), risk ratio 0.32 (0.14–0.70); unadjusted two-proportion z-test p = 0.002.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: the registry records whether a participant had a pathological complete histological response. A count is reported for each arm, with an outcome denominator of 78 in each. For a binary endpoint, a group’s response proportion is the count with the event relative to the group denominator. The registry reports proportions of 0.282 for radiochemotherapy and 0.090 for chemotherapy. These are posted values and are not newly calculated.

What does the risk difference express?

The risk difference subtracts one group’s event proportion from the other’s. The comparison is framed as chemotherapy relative to radiochemotherapy, so the reported risk difference is -0.192. A negative value means the observed event proportion is lower in the first-named group under that comparison direction. The interval, -0.311 to -0.074, represents uncertainty around the estimated absolute contrast. In general, an interval gives a range of effect sizes reasonably compatible with the data and the method’s assumptions; it is not a range containing a known fixed truth with certainty.

How do relative measures differ?

The relative risk compares event proportions as a ratio, while the odds ratio compares event odds. The registry reports a relative risk of 0.318, with an interval from 0.144 to 0.701, and an odds ratio of 0.251, with an interval from 0.1 to 0.630. Both are reported for chemotherapy relative to radiochemotherapy. These measures can sound more or less dramatic than an absolute difference because they use different scales. The risk difference is often useful for communicating the absolute separation in observed outcome proportions; relative measures describe the contrast proportionally. None should be interpreted without its group direction and endpoint context.

What does the p-value say?

The reported unadjusted p-value is 0.002. In general, a p-value measures how compatible the observed data, or more extreme data, are with a specified null model under the test assumptions. It does not give the probability that the null hypothesis is true, nor does it measure the clinical importance of the difference. The registry does not report a more detailed test specification, such as whether any covariate adjustment was used; the comparison is described as unadjusted because that is the reported p-value characterization.

Why are the endpoint counts and safety counts different?

The primary endpoint reports an outcome denominator of 78 for each arm. The serious-adverse-event record instead lists 90 at risk in the radiochemotherapy group and 91 at risk in the chemotherapy group. These are distinct record entries for different outcome reporting. The registry does not explain the relationship between these denominators, so they should not be reconciled by assumption or treated as interchangeable. The enrollment figure is also not a substitute for an endpoint-specific denominator.

What does randomization contribute?

In general, random assignment is intended to make treatment groups comparable on average at baseline, supporting a causal interpretation when conduct and analysis preserve that assignment. It does not guarantee balance in every factor, and it does not prevent bias from missing outcomes, deviations from assigned treatment or outcome assessment. The trial is recorded as randomized and unmasked. The registry record does not report details of sequence generation, allocation concealment, the analysis population or handling of missing primary outcomes, so the extent to which those safeguards were implemented cannot be assessed from this record.

What remains unspecified?

The registry records no posted analyses and does not provide the full statistical analysis plan, confidence level for the reported intervals, adjustment strategy, or method used to test the comparison. It also does not report whether any interim analyses occurred or how multiplicity was handled. The reported estimates can be described on their stated scales, but these undocumented methods should not be inferred from the numerical results. No sample-size justification from the trial’s original design is reported; the separate planning illustrations are based on the observed effect and are not a substitute for that information.

10. Confidence Intervals and Precision

The risk-difference estimate is -0.192 for chemotherapy relative to radiochemotherapy, with a reported interval from -0.311 to -0.074. The endpoints indicate uncertainty about the magnitude of the absolute contrast. They are both below zero, which is consistent with a lower observed response proportion in the chemotherapy arm under the stated direction of comparison. The registry does not state the interval’s confidence level, so its nominal coverage is not reported.

Relative measures also have intervals: the relative risk is 0.318, with an interval from 0.144 to 0.701, and the odds ratio is 0.251, with an interval from 0.1 to 0.630. These intervals describe uncertainty on ratio scales, rather than the absolute scale of the risk difference. Because each uses a different scale, their numerical widths should not be compared directly with the risk-difference interval.

In general, a confidence interval conveys both the direction and the precision of an estimate. A narrower interval indicates greater precision on its own scale, while a wider interval leaves more uncertainty about the effect size. Precision depends on the amount of information and the variability of the outcome, among other factors. Here, the group counts and intervals make clear that the point estimate is not an exact measure of the underlying contrast. The registry record does not report a confidence level or additional interval-method details, limiting how specifically the interval procedure can be described.

The p-value of 0.002 is a separate summary of compatibility with a no-difference model; it does not replace the interval or describe precision by itself. The estimates apply to the recorded pathological response outcome, not automatically to unposted secondary outcomes or to other populations.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total event counts. They use a two-sided significance level of 0.05, equal allocation between the two arms and a normal approximation. The effect to detect is set as a stated fraction of the observed effect. These are planning illustrations based on the observed effect, not the original trial design assumptions; those original assumptions are not reported in the registry record. These figures are not observed power.

For planning at 100% of the observed effect, the illustration gives 64 participants per arm for 80% power and 84 participants per arm for 90% power. At 75% of the observed effect, it gives 124 participants per arm for 80% power and 166 participants per arm for 90% power. At 50% of the observed effect, it gives 304 participants per arm for 80% power and 406 participants per arm for 90% power.

Power is the probability, under a specified effect and statistical assumptions, of detecting an effect at the chosen significance level. The 80% column represents planning for a design with an 80% chance of detecting its assumed effect under those assumptions; the 90% column represents a higher target chance. A smaller effect is harder to distinguish from random variation, so planning for a smaller fraction of the observed effect requires more participants per arm. The figures do not show what power the completed trial actually had, and they do not establish that the assumed effect will occur in a future study.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.282; treatment proportion p₁ = 0.090; average p̄ = 0.186.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5502 = 1.0783.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5331 = 0.4487.
4. n = (1.0783 + 0.4487)² / (-0.1923)² = 63.0, rounded up to 64 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.282051282051282                          # control proportion
p1 <- 0.0897435897435897                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1362127)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 64 (80% power), 84 (90% power)
 75% of observed effect: n per arm = 124 (80% power), 166 (90% power)
 50% of observed effect: n per arm = 304 (80% power), 406 (90% power)
power.prop.test: 63 
Simulated power at n = 64 per arm: 0.811 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary pathological complete histological response outcome, as well as secondary topics covering safety, health-related quality of life and swallowing function. It records three outcome measures and two posted secondary outcomes, but no posted analyses. The record does not state how the primary endpoint was prioritized analytically beyond identifying it as primary, or how multiple secondary outcomes were handled in statistical testing.

Multiplicity arises when several outcomes, comparisons or analyses are examined. In general, making many statistical tests without an adjustment can increase the chance of at least one apparently unusual result. Adjustments or a prespecified hierarchy can address that concern, but the registry record does not report a multiplicity strategy. It also does not report whether additional subgroup analyses or interim analyses were planned or conducted. Those details should be treated as not reported in the registry record rather than inferred from the posted p-value.

Missing-data handling and analysis populations are also not reported in the registry record. It does not specify whether the primary comparison followed an intention-to-treat principle, whether participants were analyzed as randomized, or what happened when a primary outcome was unavailable. The reported endpoint denominator is 78 in each arm, while the safety record has separate at-risk counts. The relationship between those sets is not explained and should not be guessed.

In general, missing outcomes can affect a treatment comparison when availability is related to treatment assignment or outcome status. A transparent analysis would describe how many outcomes were available, why any were missing, and the assumptions or methods used to address them. Those details are not reported in the registry record. Consequently, the displayed primary comparison can be described as posted, but the robustness of that comparison to missing data and the scope of its analysis population cannot be evaluated from the record alone.

14. 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

Serious events with the highest arm-level rate: sae (63.3%), cardiovascular events (15.6%), nutritional deficiency (14.4%), renal failure (7.7%), nausea and vomiting (6.7%).

The serious-adverse-event record lists 57 affected participants among 90 at risk in the radiochemotherapy group and 41 among 91 at risk in the chemotherapy group. These are the safety record’s own risk sets and should not be substituted for the primary endpoint denominators. The registry also lists SAE at 63.3%, cardiovascular events at 15.6%, nutritional deficiency at 14.4%, renal failure at 7.7%, nausea and vomiting at 6.7%, and infection at 5.6% among the leading serious terms.

These terms are registry summaries; the record does not clarify whether categories overlap or how events were attributed. The figures should therefore not be added together or interpreted as mutually exclusive categories. Other adverse events affected 0 participants in each listed group. Death counts are not reported in the registry record. The safety figures describe recorded events and risk sets; without further definitions, follow-up details and analysis methods, they do not by themselves establish comparative safety or explain the severity, duration or treatment relationship of individual events.

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

The posted data show a higher recorded count of pathological complete histological response in the radiochemotherapy arm than in the chemotherapy arm: 22 versus 7, with an outcome denominator of 78 in each arm. The registry comparison, expressed as chemotherapy relative to radiochemotherapy, reports a risk difference of -0.192 and an unadjusted p-value of 0.002. The interval around the risk difference ranges from -0.311 to -0.074. Together, these summaries describe a difference in the trial’s recorded pathological response outcome and uncertainty about its size.

The result does not, by itself, establish that radiochemotherapy improves survival, prevents recurrence, changes quality of life or improves swallowing function. Those are different outcomes, and the registry does not include posted analyses for the listed secondary topics. Nor does a pathological response count summarize all aspects of treatment, including the type, duration or consequences of adverse events. The separate safety record lists serious events, but the available summaries do not provide enough detail to characterize all safety trade-offs.

General interpretation of a randomized trial depends not only on the point estimate, but also on the study population, implementation of randomization, outcome ascertainment, missing data and analysis choices. This record states that the trial was randomized and unmasked, but does not report allocation concealment, the analysis population, missing-outcome handling or the detailed statistical plan. The interval confidence level is also not stated. These omissions limit how fully the statistical result can be evaluated.

The finding pertains to participants represented in this trial and the endpoint as recorded. It should not be treated as a guarantee of an individual outcome or as a conclusion about all patients with esophageal or cardia cancer. The registry result is evidence about a specific preoperative treatment comparison and pathological endpoint; other claims require outcome data and methods that are not reported in the registry record.

16. Why This Trial Matters Statistically

This trial illustrates why defining the endpoint and comparison direction matters. Its primary measure is a participant count for pathological complete histological response, and the posted relative measures are framed as chemotherapy compared with radiochemotherapy. With the direction made explicit, a negative risk difference can be read as a lower recorded response proportion in the first-named group, rather than misread as a result favoring chemotherapy.

It also demonstrates the value of reporting both absolute and relative effect measures. The risk difference describes the absolute separation in response proportions, while the relative risk and odds ratio describe the contrast on ratio scales. Confidence intervals add information about precision that a p-value alone cannot provide. Readers can therefore distinguish the direction of the estimate, its scale and the uncertainty around it.

The registry record also shows why outcome-specific denominators matter. The primary endpoint and safety outcomes use different at-risk counts, and those counts cannot be silently substituted for one another. Separate reporting of response and adverse events is essential because an efficacy endpoint does not describe treatment burden or safety.

Finally, the record identifies important areas where statistical transparency would help: the analysis population, missing-data methods, multiplicity handling, interval confidence level and original design assumptions are not reported. Recognizing what is documented and what is not is part of interpreting clinical trial evidence. The planning illustrations can support future study planning, but they are not evidence about the completed trial’s observed power or a guarantee of future results.

17. 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 is limited in its description of methods. Although the study is recorded as randomized, the method of sequence generation and allocation concealment is not reported. The trial is listed as having no masking, but whether outcome assessors, including pathology assessors, were blinded is not reported. The record also does not specify the analysis population, the handling of participants without an evaluable resection outcome, or any missing-data approach.

The primary outcome is named and linked to the Chireac tumour regression grade, but scoring procedures and operational criteria are not described. The reported group counts therefore cannot be independently evaluated against detailed grading rules from the registry entry. The confidence level and calculation method for the reported intervals, along with fuller test details, are not reported. No posted analyses are recorded, and results for the listed secondary outcome topics are not available in this entry.

Safety reporting uses at-risk groups that differ from the primary endpoint denominators, and the registry does not explain why. It also does not report death counts, event-level detail, exposure duration, attribution rules or whether the listed serious-event categories overlap. These gaps restrict interpretation of safety summaries and comparisons.

The sample-size illustrations are based on the observed effect and stated statistical assumptions, but the original trial design assumptions and sample-size justification are not reported. The record gives an enrollment figure, but that does not resolve the endpoint-specific analysis population. These limitations do not alter the posted numbers; they constrain how completely the trial’s conduct and statistical robustness can be assessed.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Neoadjuvant Chemotherapy Versus Radiochemotherapy for Cancer of the Esophagus or Cardia — NCT01362127.
• Registry record last updated 2020-02-26; status last verified 2020-02.
• Results first posted 2020-02-26.
• 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.

21. Record Summary

NCT01362127, titled “Neoadjuvant Chemotherapy Versus Radiochemotherapy for Cancer of the Esophagus or Cardia,” is recorded as a completed phase 2 randomized, parallel-group trial with no masking. Karolinska University Hospital is the sponsor. The registry records a start date of 2006-10, a primary completion date of 2013-06 and enrollment of 181.

The comparison was radiochemotherapy plus surgery versus chemotherapy plus surgery. The primary outcome was the participant count with pathological complete histological response after resection, described using the Chireac tumour regression grade. The posted counts were 22 in the radiochemotherapy arm and 7 in the chemotherapy arm, with an outcome denominator of 78 in each arm. The registry lists safety, health-related quality of life and swallowing function as secondary topics, but does not record posted analyses. Important details of the statistical plan, missing-data handling and analysis populations are not reported in the registry record.