Radiation Therapy and Cisplatin With or Without Cetuximab in Treating Patients With Stage III or Stage IV Head and Neck Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Radiation Therapy and Cisplatin With or Without Cetuximab in Treating Patients With Stage III or Stage IV Head and Neck Cancer (NCT00265941), a randomized phase 3 study. The primary outcome is Progression-free Survival (PFS) (3-year Rate Reported).

2. Clinical Question

The trial addressed whether adding cetuximab to concurrent accelerated radiation and cisplatin was associated with a different progression-free survival outcome. The registry describes the study as randomized and phase III, with a parallel-group design and no masking. Its stated purpose was to compare the combination including cetuximab with radiation and cisplatin alone in patients with stage III or stage IV head and neck cancer.

The primary outcome is a time-to-event measure, although the registry reports its three-year estimate as a percentage. The statistical question is therefore not only how many patients experienced an event, but also how the timing of progression or death contributes to the estimated rate. The posted results give Kaplan–Meier estimates for each group and a sponsor-reported log-rank comparison. Those estimates describe the trial population and follow-up available for the analysis; they do not, by themselves, establish why any difference arose.

3. Trial Design

This was a randomized phase III trial with a parallel-group structure and no masking. The registry lists enrollment as 940 and identifies the Radiation Therapy Oncology Group as sponsor. The study started in 2005-11, had primary completion in 2011-01, and is marked completed. These registry entries describe the recorded study status and timeline; they do not explain the rationale behind the original design choices.

Both groups received cisplatin and radiation therapy. Radiation was described as accelerated fractionation by concomitant boost or intensity-modulated radiation therapy. One group also received cetuximab. The official title notes surgery for selected patients, but further details about selection, timing, or use of surgery are not reported in the registry record.

The posted primary outcome estimates use 447 participants in the RT + Cisplatin group and 444 in the RT + Cisplatin + Cetuximab group. The primary outcome population is specified as randomized eligible patients with follow-up data. These analysis counts should not be treated as a substitute for the registry’s enrollment figure: the record does not provide a complete accounting of how enrollment relates to eligibility, follow-up availability, or each analysis population.

The registry describes follow-up for the primary outcome as continuing until 434 failures have occurred, approximately 5 years from the start of the study, with patients followed until death or study termination, whichever occurs first. The posted primary result is a three-year rate. Allocation concealment, randomization procedures, the original sample-size justification, and the trial’s prespecified analysis plan are 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomized eligible patients with follow-up data

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: Eligible patients with follow-up

6. Primary Endpoint

The registry lists the primary outcome as Progression-free Survival (PFS) (3-year Rate Reported), measured over From randomization until 434 failures have occurred, approximately 5 years from start of study. (Patients are followed until death or study termination, whichever occurs first.).

The primary endpoint is progression-free survival (PFS), reported as a three-year rate. It measures the time from randomization until local, regional, or distant disease progression, or death from any cause. A patient last known to be alive without progression is censored at the date of last contact. The registry states that the three-year rates were estimated using the Kaplan–Meier method.

This endpoint combines disease progression and death into one time-to-event outcome. That makes it useful for describing how long patients remain alive without a recorded progression event, rather than considering progression alone. It also means that the endpoint does not distinguish, within the single summary rate, between progression and death as the event that ended progression-free survival.

The registry gives definitions for local or regional progression based on tumor-size increase or new malignant disease, and for distant progression based on clear evidence of distant metastases. The posted outcome is therefore governed by those recorded event definitions and the follow-up process. The time frame describes follow-up until 434 failures, approximately 5 years from study start, while the reported estimate is specifically the rate at three years. The record does not report further detail about how endpoint events were independently reviewed or adjudicated.

7. Results Posted in the Registry

• RT + Cisplatin + Cetuximab: 59.0% (262/444) versus 61.3% (274/447) on control; risk difference -2.3 percentage points (95% CI -8.7 to 4.1), risk ratio 0.96 (0.87–1.07); unadjusted two-proportion z-test p = 0.485.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Overall Survival (OS) (3-year Rate Reported)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • RT + Cisplatin + Cetuximab: 75.9% (337/444) versus 72.9% (326/447) on control; risk difference 3.0 percentage points (95% CI -2.8 to 8.7), risk ratio 1.04 (0.96–1.12); unadjusted two-proportion z-test p = 0.310.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between RT + Cisplatin + Cetuximab and the control. The interval gives the range of effects that remain compatible with the data.

Local-regional Failure (LRF) (3-year Rate Reported)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • RT + Cisplatin + Cetuximab: 25.9% (115/444) versus 19.9% (89/447) on control; risk difference 6.0 percentage points (95% CI 0.5 to 11.5), risk ratio 1.30 (1.02–1.66); unadjusted two-proportion z-test p = 0.033.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for RT + Cisplatin + Cetuximab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Rate of Mucositis Toxicity ≥ Grade 3

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • RT + Cisplatin + Cetuximab: 43.2% (192/444) versus 33.3% (149/447) on control; risk difference 9.9 percentage points (95% CI 3.6 to 16.3), risk ratio 1.30 (1.10–1.54); unadjusted two-proportion z-test p = 0.002.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for RT + Cisplatin + Cetuximab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes add several kinds of context: survival, local-regional disease control, treatment-related toxicity, treatment tolerance, early deaths after discontinuation, and quality of life. The posted numerical results cover some of these outcomes, while the registry record does not report numerical results for every outcome it lists. The overall pattern across the reported outcomes is mixed rather than a uniform shift in one direction across efficacy and toxicity measures.

That pattern should be considered alongside, not instead of, the primary PFS analysis. Overall survival and local-regional failure describe distinct outcomes, and toxicity outcomes describe harms or tolerability rather than disease control. Their estimates do not redefine PFS, and a secondary endpoint cannot resolve uncertainty in the primary result simply because its point estimate appears more or less favorable.

Unadjusted secondary comparisons carry less weight when many outcomes are examined. In general, testing multiple outcomes increases the chance of obtaining apparently notable findings by chance. A prespecified hierarchy or multiplicity adjustment can help control interpretation, but the registry record does not report whether either approach was planned or used. The posted unadjusted results should therefore be read as exploratory or descriptive unless further analysis-plan information is available.

The secondary outcome list includes measures of mucositis and other toxicity, treatment tolerance, quality of life, and deaths within a specified period after stopping protocol treatment. Numerical results for several listed measures are not reported in the registry record. The absence of posted estimates for those outcomes should not be interpreted as evidence that the outcomes did not occur or that the groups were alike.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Overall Survival (OS) (3-year Rate Reported)
• Local-regional Failure (LRF) (3-year Rate Reported)
• Rate of Mucositis Toxicity ≥ Grade 3
• Rate of Other Toxicity ≥ Grade 3 (Not Mucositis)
• Rate of Patients Who Tolerated Treatment
• Rate of Deaths ≤ 30 Days After Discontinuation of Protocol Treatment
• Quality of Life as Measured by European Quality of Life Questionnaire (EQ-5D)
• Quality of Life as Measured by Proportion of Patients With Performance Status Scale for Head and Neck Cancer (PSS-HN) Scores ≤ 50

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is time to an event, rather than a simple count of events by a fixed date. In general, time-to-event analysis uses information about both whether an event occurred and when it occurred. It can also accommodate censoring: a participant may be known to be event-free up to the last observation without having a complete event time. The registry says that patients last known to be alive without progression were censored at the date of last contact.

What does the Kaplan–Meier estimate describe?

In general, the Kaplan–Meier method estimates the proportion remaining event-free over time while accounting for participants whose follow-up ends before an event is observed. The three-year PFS percentages are estimates at a specified time point, not simple proportions obtained by dividing the number of participants without an event by the number enrolled. The registry identifies the outcome population as randomized eligible patients with follow-up data and states that the three-year rates were estimated by this method.

The result depends on recorded event times, censoring times, and the endpoint definition. Here, the event is local, regional, or distant progression, or death from any cause. The record gives definitions for progression and says follow-up continued until the specified number of failures, approximately 5 years from study start, or until death or study termination. Details such as censoring patterns by group and the completeness of follow-up are not reported in the registry record.

How should the log-rank result be read?

The sponsor’s reported analysis uses the log-rank method and gives a hazard ratio with a confidence interval and p-value. In general, the log-rank test compares the event-time experience between groups over follow-up, rather than comparing only the Kaplan–Meier estimates at one selected time. A hazard ratio summarizes the relative event rate as represented by the analysis. Its interpretation depends on group ordering and the assumptions used for the time-to-event analysis; the registry does not provide enough detail to assess those assumptions fully.

The sponsor reports a hazard ratio of 1.08, a 95% confidence interval from 0.88 to 1.32, and p = 0.76. The interval includes 1, the value corresponding to equal hazards on the ratio scale. The p-value describes how compatible the test result is with the null hypothesis under the test’s assumptions; it is not the probability that the null hypothesis is true. A p-value does not measure the size or practical importance of an effect.

Why is there also a recalculated risk difference?

The recalculated comparison uses posted summary statistics to give an unadjusted risk difference and related measures. It is not the same analysis as the sponsor’s log-rank test. In general, a risk difference compares event proportions on an absolute scale, while a hazard ratio compares event rates over time. The recalculated primary risk difference is -0.023, with a 95% confidence interval from -0.087 to 0.041 and p = 0.485. It should be understood as a straightforward summary-based recalculation, not a replacement for the trial’s time-to-event analysis. The registry record does not flag a discrepancy.

Confidence intervals express uncertainty around estimates under the analysis method. In general, a 95% confidence interval is generated by a procedure that would cover the underlying parameter in 95% of repeated studies under its assumptions. It does not mean there is a 95% probability that the fixed parameter lies in the particular interval. The interval’s width also matters: a wide interval allows a broader range of effects than a narrow one.

What can the secondary analyses add?

The registry lists 12 posted secondary outcomes, including overall survival, local-regional failure, and toxicity outcomes. The two survival outcomes with reported analyses also use log-rank methods and hazard ratios. Other outcomes are represented as participant percentages. These measures have different definitions and analysis scales, so a result from one should not be treated as a substitute for the primary PFS result.

In general, multiple outcome comparisons create more opportunities for apparently unusual results, even when there is no underlying treatment difference. Adjustment methods or a prespecified hierarchy can address some of this concern, but the multiplicity plan is not reported in the registry record. The unadjusted recalculations are descriptive comparisons and do not account for multiple testing. The record also does not report the trial’s complete statistical analysis plan, so details such as covariate adjustment, subgroup analyses, or sensitivity analyses cannot be assumed.

The registry reports adverse-event counts for serious events and other events, but counts alone do not describe timing, duration, severity distribution, or causality. Those issues require additional definitions and summaries. The record also does not report death counts in the adverse-event summary. These boundaries matter when interpreting safety information alongside the efficacy endpoints.

11. Confidence Intervals and Precision

The primary PFS estimates are 61.2% for RT + Cisplatin and 58.9% for RT + Cisplatin + Cetuximab. Their 95% confidence intervals are 56.7% to 65.8% and 54.2% to 63.6%, respectively. These intervals communicate uncertainty around each group’s Kaplan–Meier estimate. They are not a direct interval for the difference between groups; the recalculated risk-difference interval addresses that comparison on a different scale.

The recalculated risk difference is -0.023, with a 95% confidence interval from -0.087 to 0.041. Because this interval includes zero, the summary-statistic recalculation is compatible with no difference and with effects on either side of zero within the interval. The sponsor’s hazard-ratio confidence interval, 0.88 to 1.32, includes 1. That interval likewise does not establish a clear difference on the hazard-ratio scale.

Precision is not the same as importance. An estimate can be clinically important but imprecise, or precisely estimated but of limited practical importance; judgments about clinical importance require context not established by the statistical interval alone. In general, confidence intervals also depend on the method and assumptions used to construct them. The registry gives the confidence level and estimates but does not report all analysis details needed to independently assess those assumptions.

The intervals overlap, but overlap between separate group intervals is not itself the formal test of the treatment contrast. The direct recalculated risk-difference interval and the sponsor-reported hazard-ratio interval are more appropriate summaries for their respective comparisons. Their different scales and methods should be kept in view when reading the degree of uncertainty.

12. 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 illustrations based on the observed effect, not the trial’s original design calculation. The registry record does not report the original sample-size justification or its assumptions. The stated illustration assumptions are 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 the stated fraction of the observed effect.

For time-to-event outcomes, the planning figures represent the total number of events, using the Schoenfeld method. For other outcomes, they represent participants per arm. The PFS outcome is defined by time to progression or death, so its planning interpretation is in total events rather than participants per arm. These units are not interchangeable: an event target is not a group enrollment target.

The 80% power column describes the number of events or participants indicated by the illustration for an 80% chance of detecting the specified effect under its assumptions, if that effect were present. The 90% power column makes the analogous illustration for a 90% chance. Higher power generally requires more information, and the posted figures show that planning for a smaller fraction of the observed effect requires more participants or events.

These values are not observed power. They are planning illustrations under stated assumptions, not evidence that the trial had a particular probability of detecting its actual result. The observed effect is itself uncertain, and using it as a planning anchor does not establish the effect that a future study should expect.

13. 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 = 0.613; treatment proportion p₁ = 0.590; average p̄ = 0.602.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6924 = 1.3570.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6922 = 0.5826.
4. n = (1.3570 + 0.5826)² / (-0.0229)² = 7,183.0, rounded up to 7,183 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 <- 0.612975391498881                          # control proportion
p1 <- 0.59009009009009                          # 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(265941)
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 = 7,183 (80% power), 9,616 (90% power)
 75% of observed effect: n per arm = 12,740 (80% power), 17,054 (90% power)
 50% of observed effect: n per arm = 28,593 (80% power), 38,277 (90% power)
power.prop.test: 7183 
Simulated power at n = 7,183 per arm: 0.798 (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.

14. Analyses Reported by the Sponsor

The sponsor’s own primary analysis is a log-rank comparison with a hazard ratio of 1.08, a 95% confidence interval from 0.88 to 1.32, and p = 0.76. This analysis uses event-time information over follow-up. The recalculated comparison is an unadjusted summary-statistic risk difference of -0.023, with a 95% confidence interval from -0.087 to 0.041 and p = 0.485. The two results use different measures and methods, so their numerical estimates and p-values should not be compared as if they were the same calculation.

The registry record does not flag a discrepancy between the sponsor analysis and the recalculation. The sponsor’s reported log-rank result is the trial’s own reported analysis; the recalculation is an additional, simple comparison based on the posted summaries. The primary interpretation should keep the sponsor’s time-to-event analysis distinct from that recalculated absolute-scale comparison.

Neither result establishes a clear difference in PFS: both intervals include their respective no-difference values. This does not establish equivalence, since the registry does not report an equivalence or noninferiority design. The record does not give further details of the sponsor’s model or analysis plan, so no additional adjustment or assumption should be inferred.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary PFS endpoint and multiple secondary outcomes, with 13 outcome measures and 18 analyses recorded. When a study evaluates multiple outcomes or conducts multiple analyses, the chance of observing at least one apparently noteworthy result by chance can increase. In general, researchers may address this through a prespecified testing hierarchy, adjustment for multiple comparisons, or a clearly stated exploratory interpretation. The multiplicity approach is not reported in the registry record, so it is not possible to determine whether the posted secondary p-values were adjusted or how their testing was prioritized.

The primary outcome population is described as randomized eligible patients with follow-up data. The record does not say that all randomized participants were included in every analysis, nor does it describe an intention-to-treat analysis population. The relation between the enrollment figure and the analysis populations, including exclusions or unavailable follow-up, is not fully reported in the registry record.

For PFS, the record states that patients last known alive without progression were censored at last contact. This is a standard feature of time-to-event analysis in general, but the interpretation depends on follow-up and censoring patterns. The registry does not report the amount or pattern of missing outcome data, reasons for loss to follow-up, or sensitivity analyses for missingness. It also does not describe imputation procedures, if any.

The original sample-size assumptions, allocation concealment, randomization procedures, and complete statistical analysis plan are not reported in the registry record. Consequently, the available information does not establish whether the analyses followed a prespecified plan, how potential protocol deviations were handled, or whether additional subgroup or sensitivity analyses were performed. These unreported details limit assessment of potential analytical flexibility and the robustness of the posted findings.

16. 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: dehydration (18.0%), nausea (13.3%), radiation mucositis (11.7%), dysphagia (11.4%), vomiting nos (10.1%).

The serious adverse-event summary reports 208 affected participants among 447 at risk in the RT + Cisplatin group, and 237 among 444 at risk in the RT + Cisplatin + Cetuximab group. For other adverse events, the summary reports 442 among 447 at risk and 433 among 444 at risk, respectively. These are counts of affected participants and participants at risk as recorded; they are not measures of event duration, severity distribution, or causal attribution.

The listed serious adverse-event terms include dehydration, nausea, radiation mucositis, dysphagia, vomiting NOS, and leukopenia NOS, with percentages reported for those terms. The registry does not attach those listed percentages to a particular treatment group in the record, so they should not be assigned to one group without further information. Death counts in the adverse-event summary are not reported in the registry record.

In general, adverse-event counts require context about definitions, ascertainment, exposure time, and whether events are considered treatment-related. Those details are not all available in the registry record. The counts therefore describe the posted safety summaries but should not be read alone as a complete comparative assessment of safety.

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

The posted results describe three-year PFS estimates for two randomized treatment groups and a sponsor-reported comparison over follow-up. The estimated rates are close, and both the recalculated risk-difference interval and sponsor-reported hazard-ratio interval include their respective no-difference values. The data do not establish a clear separation on the primary endpoint. They also do not demonstrate that the regimens are equivalent or that any possible difference is clinically unimportant.

Statistical uncertainty is central to that reading. The confidence intervals allow a range of effects, not just the point estimates. A result that does not show a clear difference is not proof that the groups have identical outcomes. Conversely, the point estimates alone do not show that one treatment caused a meaningful improvement or worsening. Such conclusions require careful attention to the outcome definition, analysis method, uncertainty, and clinical context.

The trial’s endpoint combines progression and death, and its estimates concern randomized eligible patients with follow-up data. Generalizing beyond that analyzed population requires consideration of eligibility, follow-up, treatment delivery, and the extent to which other patients resemble those represented in the trial. The registry does not report all details needed to evaluate those considerations.

The secondary outcomes provide additional information about survival, local-regional failure, and toxicity, but they are not substitutes for the primary endpoint. Their interpretation is also limited by the unreported multiplicity approach and missing results for some listed outcomes. Safety counts are descriptive and do not, by themselves, establish causation or fully characterize comparative risk.

These statistics describe a clinical trial record; they are not individual medical advice. They do not determine which treatment is appropriate for a particular patient, and no treatment recommendation follows from the estimates alone.

18. Why This Trial Matters Statistically

This trial illustrates why endpoint definitions and analysis methods matter in clinical research. PFS is defined using time to progression or death and is estimated at a specified time with Kaplan–Meier methods. The sponsor’s log-rank analysis uses the event-time experience over follow-up, while the recalculated risk difference compares summaries on an absolute scale. These are related but distinct ways to describe treatment-group results.

It also demonstrates why estimates should be read together with their uncertainty. The point estimates alone might draw attention to a direction, but confidence intervals show that a range of effects remains compatible with the data. A nonsignificant comparison is not a proof of equivalence, and a secondary result should not be elevated above a primary endpoint without considering the analysis plan and multiplicity.

The study record connects efficacy and safety questions in a randomized comparison: both groups received radiation and cisplatin, with cetuximab added in one group. The outcomes span progression-free and overall survival, local-regional failure, and toxicity. Their different meanings require separate interpretation rather than a single summary claim about benefit or harm.

Finally, the registry’s missing details are statistically relevant. The original sample-size rationale, full analysis plan, multiplicity approach, and several aspects of missing-data handling are not reported. Recognizing what is known and what remains undocumented helps prevent overinterpretation of a result, and clarifies why transparent reporting is important for readers assessing trial evidence.

19. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several trial-specific limitations of interpretation arise from information not reported in the registry record. The original sample-size justification and design assumptions are unavailable, so the planning illustrations based on the observed effect cannot establish what effect the trial was originally intended to detect. The full statistical analysis plan, including any prespecified hierarchy for outcomes or multiplicity adjustment, is also not reported.

The registry identifies the primary outcome population as randomized eligible patients with follow-up data, but does not provide a complete accounting of how that population relates to enrollment or how missing follow-up was handled. For PFS, censoring at last contact is described for patients last known to be alive without progression, but censoring patterns and sensitivity analyses are not reported. These gaps limit assessment of the potential influence of incomplete follow-up.

The study was unmasked. The record does not provide details about allocation concealment, endpoint assessment procedures, or whether progression events were independently reviewed. Those details should not be inferred from the randomized design alone. The official title mentions surgery for selected patients, but the registry record does not describe the selection criteria or how surgery was incorporated into treatment delivery or analysis.

Both groups could receive radiation using accelerated fractionation by concomitant boost or intensity-modulated radiation therapy. The record does not report how use of these approaches was distributed across groups or how variation in radiation delivery was addressed analytically. It also does not report numerical results for every listed secondary outcome, limiting assessment of the complete outcome profile.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Radiation Therapy and Cisplatin With or Without Cetuximab in Treating Patients With Stage III or Stage IV Head and Neck Cancer — NCT00265941.
• Registry record last updated 2022-06-14; status last verified 2022-05.
• Results first posted 2017-12-21.
• 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.

23. Record Summary

NCT00265941 is a completed randomized phase III, parallel-group trial sponsored by the Radiation Therapy Oncology Group. It compared accelerated radiation and cisplatin with the same treatment plus cetuximab in patients with stage III or stage IV head and neck cancer. The study was unmasked; enrollment is listed as 940.

The primary endpoint was three-year PFS, defined as time from randomization to progression or death and estimated by Kaplan–Meier methods. The registry reports group estimates and a sponsor-reported log-rank hazard ratio. The posted primary analyses do not establish a clear difference, and they do not establish equivalence.

Secondary outcomes include survival, local-regional failure, toxicity, treatment tolerance, and quality of life. The registry also reports serious and other adverse-event counts. The original sample-size rationale, multiplicity plan, and several details about missing data and analysis populations are not reported in the registry record.